Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
sequencelengths 1
91
| keywords
sequencelengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
Metapopulation processes are important determinants of epidemiological and evolutionary dynamics in host-pathogen systems , and are therefore central to explaining observed patterns of disease or genetic diversity . In particular , the spatial scale of interactions between pathogens and their hosts is of primary importance because migration rates of one species can affect both spatial and temporal heterogeneity of selection on the other . In this study we developed a stochastic and discrete time simulation model to specifically examine the joint effects of host and pathogen dispersal on the evolution of pathogen specialisation in a spatially explicit metapopulation . We consider a plant-pathogen system in which the host metapopulation is composed of two plant genotypes . The pathogen is dispersed by air-borne spores on the host metapopulation . The pathogen population is characterised by a single life-history trait under selection , the infection efficacy . We found that restricted host dispersal can lead to high amount of pathogen diversity and that the extent of pathogen specialisation varied according to the spatial scale of host-pathogen dispersal . We also discuss the role of population asynchrony in determining pathogen evolutionary outcomes .
In spatially structured populations , habitat spatial heterogeneity plays a crucial role in determining the potential for species and genotypes to coexist [1] , [2] and in shaping the evolution of populations and species [3] , [4] . However , environments are not static and temporal fluctuations in habitat quality and distribution can also impose strong selection pressure [5] . Populations can meet this challenge by evolving or migrating [6] , [7] to track high-quality environments over time [8] and in host-pathogen systems such metapopulation processes are important determinants of observed patterns of disease or genetic diversity [9] , [10] , [11] , [12] . In particular , the spatial scale of interactions between a pathogen and its host is seen to be of prime importance in determining evolutionary trajectories of host-pathogen metapopulations [13] . Indeed , migration rates of one of the species affect the spatial and temporal heterogeneity of selection on the other [14] , [15] . Some of the first models investigating the role of dispersal on host-pathogen coevolving patterns assumed a qualitative type of interaction ( single locus population genetics model ) . The work of Gandon and colleagues [16] , [17] put emphasis on local adaptation by developing a metapopulation model composed of a finite number of patches each of which could exchange propagules with its neighbouring populations . They demonstrated that asymmetry in host-pathogen dispersal can have strong effect on patterns of local adaptation . Thus when the parasite disperses more than the host , it is more likely to become locally adapted ( and vice-versa ) . This prediction has been verified experimentally [18] , [19] , [20] , [21] and formally qualified [17] . In a complementary way , Thrall and Burdon [22] examined the maintenance of host and pathogen genotypic diversity as a function of dispersal . They found that local dispersal for both the host and the pathogen favoured evolution of the highest number of resistance and infectivity genotypes across the metapopulation ( diversity was highest when there was still some degree of among-population asynchrony ) . Host-pathogen interactions are however not limited to qualitative relationships but are also largely determined by quantitative traits [23] . The role of spatial variation in host and pathogen life-history traits ( components of quantitative interactions ) in determining the evolutionary potential of parasitic organisms and their demographic and evolutionary histories is still poorly understood [24] . Best et al . [25] and Débarre et al . [26] modelled the evolution of host life-history traits in spatially structured host-pathogen populations . Both models assumed a lattice structure with interactions ( reproduction and transmission ) occurring either locally ( to the nearest neighbours ) or globally ( randomly across the entire population ) . These studies underline the importance of spatial structure in affecting evolutionary outcomes . In particular , spatial structure can promote the evolution of decreased disease transmissibility but its effects on disease-related mortality depend on other aspects of life history such as the extent to which infected hosts contribute to population growth ( the latter also impacts the potential for disease persistence [27] ) . In addition , Best et al . [25] studied the conditions for branching , i . e . when hosts undergo disruptive selection and branch into two coexisting types . They found that branching was possible in a spatial model but requires higher virulence ( i . e . disease-related mortality ) and stronger trade-offs than in a non-spatial model ( i . e . where dispersal is homogeneous in space ) . However , they did not characterise the coexisting genotypes . In contrast , in a similar model but focusing on pathogen evolution , Kamo et al . [28] did not observe branching points for pathogen transmission and virulence . Coevolution between host and pathogen for quantitative traits was only studied by Best et al . [25] whose key result was that the globalisation of interactions selects for low host defence and high pathogen transmission and virulence . Another crucial question that arises in coevolving systems is how the geographical structure of coevolution may shape spatial patterns of variation in the coevolving species [29] . Nuismer et al . [30] used a spatially explicit genetic model to study polymorphic clines in a one-dimensional environment . In particular , considering the more general framework of purely antagonistic interactions ( including host-pathogen interactions ) , they found that in the absence of spatial heterogeneity in environmental conditions ( which is our focus in this work ) clines can only evolve when there is initial heterogeneity in allele frequencies . In addition , increases in gene flow among populations will eventually lead to the loss of spatially structured adaptation . Gavrilets and Michalakis [31] developed and analysed an island model of antagonist coevolution . Here also , when selection was homogeneous in space , the maintenance of genetic variation across time required initial differences in allele frequencies between populations and low migration rates . With increased migration , stronger selection was required . However , these two studies only considered coevolution between two alleles and neglected the important issue of genetic drift and mutation . When some spatial variations in environmental conditions are considered , one would expect that spatially heterogeneous selection together with some restrictions on migration should favour the stable maintenance of polymorphism [30] , [31] . Here we develop a simulation model to specifically examine the joint effects of host and pathogen dispersal on the evolution of pathogen specialisation in a spatially explicit metapopulation . We consider a plant-pathogen system in which the host metapopulation is composed of two plant genotypes and the pathogen is dispersed by air-borne spores . We assumed that the pathogen population is characterised by a single life-history trait under selection , the infection efficacy of the pathogen on the host genotypes . We did not consider environmental spatial heterogeneity . In particular we addressed the following questions: How do the scale of dispersal and the strength of evolutionary trade-offs affect the potential for multiple pathogen genotypes to coexist ? Does the level of pathogen specialisation depend on host and pathogen dispersal scales ? Is there spatial heterogeneity in patterns of diversity ? We first present the model and the simulation experiment . Then we study the extent of synchrony among local populations , the effect of dispersal on pathogen diversity and level of specialisation . We also analyse the sensitivity of our results to the pathogen life-history traits , to the shape of the dispersal function and to the metapopulation structure . Finally we discuss our results with an emphasis on the role of population asynchrony in determining evolutionary outcomes .
Ten mean dispersal distances ( in proportion of the region size ) were considered for both host and pathogen by varying and in {1 . 25% , 2 . 5% , 5% , 7 . 5% , 10% , 17 . 5% , 25% , 37 . 5% , 50% , 75%} . We fixed the infection efficacy of the generalist to 0 . 2 . In addition to the generalist we defined 10 specialists on each host ( thus 20 specialists ) by increasing by 10% , 20% , … , 90% and 100% on their susceptible host . Finer steps for the discretisation of did not change the results . Two different scenarios were explored with regard to the cost that specialists suffered on their resistant host: ( the specialisation gain is greater than the specialisation cost ) and ( the specialisation gain is equal to the specialisation cost ) . The case is not reported in the present study because no qualitative differences were observed relative to the case when : only stabilised the generalist population even more . The probability that a spore was of the same genotype as its parental lesion was set as , then we set . Exceptions were the two full specialist pathogen genotypes ( gain = cost = 100% ) – these mutated toward less specialised genotypes ( gain equal to 90% ) with probability 0 . 004 to keep their overall mutation rate equal to that of other genotypes . Sensitivity to the other parameters ( spore production , ; infectious period , ; latency period ; shape of the density dependence function , ; shape of the dispersal function , and total proportion covered by the metapopulation , ) was assessed by studying the 7 case-studies detailed in Table 1 . The convergence was checked on a subset of simulations by computing the descriptors of the global pathogen evolutionary trajectory ( see Section 2 . 3 . 2 ) at different times until they stabilised . Simulations were thus performed over 20 , 000 time steps . The system started with the two host genotypes present in all patches and with the pathogen population composed of the generalist only . For each case-study , five different metapopulations were drawn and four model replicates were performed on each of them leading to 20 replicates for each of the 1400 scenarios ( 10 by 10 by 2 by 7 case-studies ) . Table 2 summarises the terms , parameters and values that we used .
Here we focus on local populations and the extent to which they were synchronised to characterise the spatial structure of the metapopulation . These results were used in Section 3 . 2 and 3 . 3 to explain patterns of coexistence and specialisation . When host and pathogen dispersed very locally , populations were largely asynchronous ( Figs . S2 and S3 in Text S1 ) . Increases in both host and pathogen mean dispersal distances resulted in an increase in the level of correlation between local and global dynamics , i . e . an increase in synchrony . Finally when both host and pathogen dispersed at large distances , the metapopulation was totally synchronised . The spatial spline correlograms estimated how the between-populations correlation was a function of spatial distance . Fig . 2 shows the correlograms for host genotype when . The profiles for the population dynamics exhibited a characteristic decrease of similarity ( spatial autocorrelation ) with distance with significantly positive autocorrelation at short spatial distances . Interestingly , when host and pathogen dispersed locally , the correlation dropped more quickly with distance than for intermediate dispersal distances ( Fig . 2a , b ) . Thus , at intermediate mean dispersal distances , the metapopulation formed aggregates of local populations which displayed asynchronous dynamics - populations belonging to the same aggregate showed synchronised behaviour , and populations belonging to distinctly different aggregates followed different dynamics ( Fig . 3 ) . For large host and/or pathogen dispersal scales no spatial structure in correlation was observed ( Fig . 2c ) again indicating full synchrony of the metapopulation . Sasaki et al . [37] developed an island metapopulation model with migration occurring either globally or locally and found a similar pattern of population synchrony according to dispersal ability with a gene-for-gene epidemiological model . Thus , the increase in the size of population aggregates that behave similarly when dispersal increases is certainly not restricted to the case presented here . Note also the existence of negative correlations for intermediate distance , even for relatively high ( ) host dispersal . In this section we study the situations in which either the two moderately specialised ( when ) or the generalist ( when ) genetic clusters occurred ( Fig . 4 ) , and characterise their infection efficacy ( ) and homogeneity ( efficacy range ) . Overall , the 7 case-studies ( Table 1 ) did not differ qualitatively and the patterns of pathogen diversity and evolution presented above were robust to changes in: the latency and infectious periods , pathogen reproduction , shape of the density dependence function , the value , and the shape of the dispersal function .
This study assesses the demographic-genetic dynamics of a spatially explicit host-pathogen metapopulation in the absence of environmental heterogeneity . The host population was composed of two genotypes whereas the pathogen was able to evolve toward more or less specialised genotypes . We assessed the spatial structure of genotypic variation in pathogen diversity and the level of synchrony among populations . We then studied the influence of host and pathogen dispersal abilities on the evolution of pathogen specialisation and found the number of coexisting pathogen genetic clusters and their trait values under several strength of evolutionary trade-off and different dispersal scenarios of both organisms . Finally , we also assessed the sensitivity of these results to variations in pathogen life-history traits , in metapopulation structure and in dispersal function shape . Our key results are that restricted host dispersal can lead to a high level of pathogen diversity and that the degree of pathogen specialisation is strongly influenced by the spatial scale of host-pathogen interactions . Pathogen metapopulation diversity evolved to its highest level when host and pathogen dispersal were both very local . Under those conditions , most populations were highly asynchronous with respect to disease dynamics leading to spatial coexistence among up to four pathogen genetic clusters . An increase in dispersal ability for both host and pathogen first resulted in the loss of the two fully specialised genetic clusters and selection for moderately specialised ( or a generalist ) genetic cluster ( s ) with low genotypic variability ( efficacy range ) . Here also among-population asynchrony was sufficient to allow spatial coexistence . As the scale of dispersal increased , the system became increasingly dominated by boom-and-bust dynamics . Local populations became increasingly synchronised and spatial coexistence was only transitory . For large pathogen and host dispersal scales , severe oscillations appeared leading to frequent local extinction-recolonisation of hosts . When moderately specialised genetic clusters were selected , these oscillations also resulted in the global extinction of one of the hosts and the subsequent full specialisation of the pathogen population on the remaining host . In this case the metapopulation acted as a single population where host and pathogen frequencies experienced growing oscillations , resulting in a single genotype of each species being fixed . The effect of host and pathogen dispersal on the maintenance of diversity was specifically studied by Thrall and Burdon [22] in a multi locus model . Their results confirmed that spatial structure is a crucial factor for explaining the levels of host and pathogen genetic diversity that are maintained in a metapopulation . In particular they found that restricted dispersal led to the highest diversity in terms of host resistance and pathogen infectivity genotypes . In addition , increases in the spatial scale of pathogen dispersal favoured more generalist pathogens carrying many infectivity genes . Those results are in line with the ones presented here showing that qualitative as well as quantitative interactions could lead to similar patterns of pathogen evolution . The conditions for the maintenance of genetic variation in a victim-exploiter system was also studied by Nuismer et al . [30] , with a one-dimensional environment , and by Gavrilets and Michalakis [31] , using an island model structure for the metapopulation . Both studies showed that when the environment was homogeneous , synchronisation among populations led to rapid loss of variation unless significant variation in allele frequencies was initially imposed . In addition , maintenance of variation required low migration rates , intermediate selection strength and the presence of a large number of populations . Asynchrony among populations and thus coexistence among genetic clusters was also found in our model to be favoured by low migration rates . However , the maintenance of polymorphism did not require that genotype frequencies varied among populations at initialisation , probably due to mutation , drift and the higher number of populations [31] , [42] . For instance , we found that local drift can lead to the extinction of one of the host genotypes in a few populations potentially enhancing the maintenance of diversity in the host and pathogen metapopulation and contributing to asynchrony in dynamics [39] . The competitive exclusion principle suggests that in an environment composed of two resources , competition can either lead to selection for a generalist or for two specialists . We found here that up to four pathogen genetic clusters can coexist in a two-resource ( host ) environment . Using a one-patch model for studying adaptive evolution , Abrams [38] found that evolution can lead to such an output when the system undergoes sustained fluctuations ( see also [43] ) . In addition , polymorphisms can be stabilised if direct frequency-dependent selection is negative , so that the net contribution of a given allele to fitness declines with increasing frequency [44] . Here we found that asynchrony among populations produces spatial heterogeneity in selection which implies negative direct frequency-dependent selection [42] and due to a restricted dispersal , favours stable coexistence among specialists and generalist morphs [2] . However , asynchrony in gene frequency fluctuation can develop even in the case of metapopulations where migration is occurring globally . In that case , a single population with asynchronous dynamics leads to a stable polymorphism at the metapopulation level [37] . This situation was not observed here . A particular feature of the current study was to characterise the level of specialisation , as measured by , that evolves and persists in pathogen populations . When hosts disperse very locally , evolution favours the most specialised pathogen genotypes which can attack only one of the hosts . Indeed , local interactions due to local dispersal made easier the evolution of highly specialized pathogens as they adapted faster than less specialized pathogens ( local growth rate was higher because of a higher infection efficacy ) [45] . On the other hand , when host and/or pathogen dispersal scales are large , oscillations can lead to the extinction of one of the hosts resulting in the fixation of one of the full specialist pathogens . Otherwise , the specialisation level of the pathogen metapopulation depends on the spatial scale of host-pathogen interaction . When we fixed the host mean dispersal distance , the pathogen specialisation level was found to increase with pathogen dispersal capability , reaching its maximum when the pathogen dispersed more than the host . The impact of gene flow on levels of pathogen specialisation has only rarely been discussed in the literature . Most studies deal with spatially unstructured populations [17] , [46] , [47] , [48] , assume qualitative interactions [16] , [17] or focus on host life history traits [25] , [26] . Best et al . [25] studied the conditions required for two specialised host genotypes to coexist but did not characterise the coexisting genotypes . Gandon [17] analysed a deterministic host-parasite coevolutionary model based on a modified matching allele model for genetic interactions . He found that the species with the higher migration rate evolved faster and was more likely to be locally adapted which is consistent with greater specialisation . Even if the properties of the two host genotypes do not change through time in our model , our results are consistent with Gandon's predictions since we found that the pathogen specialisation level was generally greater when it disperses more than the host . A contrasting pattern of specialisation was observed when pathogen dispersal was fixed to be spatially local ( mean dispersal distance ≤5% of the region size ) and host dispersal scale was varied . Under these conditions , the pathogen generally dispersed less than the host and it was expected that this differential would decrease the level of pathogen specialisation . However , contrary to this prediction , increasing host mean dispersal distance initially resulted in greater pathogen specialisation , which reached its maximum for intermediate host mean dispersal distances ( Fig . 7a ) . For larger host dispersal capabilities , the level of pathogen specialisation decreased . Recent results on the evolution of specialisation in spatially heterogeneous environments can explain this pattern of specialisation . Papaïx et al . [4] showed that the spatial distribution of habitats ( hosts ) impacts the mean phenotype of specialist morphs: specialist phenotypes are better adapted when habitat aggregation is high ( see also [49] ) . In our case , for very local host dispersal , most local populations behaved asynchronously leading to a low aggregation of host genotypes in space which favoured low pathogen specialisation level . When host dispersal increased , we first observed increased aggregation among host populations with respect to both their dynamics and the distribution of host genotypes . This resulted in an increase in the specialisation level of the pathogen . For more global host dispersal the metapopulation as a whole became synchronised and the pathogen simply tracked fluctuations in the frequencies of the two host genotypes . In this last case , pathogen specialisation decreased because oscillations between host genotypes were too rapid to make it possible for the pathogen to fully adapt to its host – clearly this situation would favour generality . In our study , although the host population was diversified , we assumed that the properties of the two host genotypes did not change through time and thus that the pathogen evolves faster than the host . This is particularly the case for crops , for which the same host genotypes are used for several years and that integrate relatively low genetic diversity for disease resistance . In agricultural landscapes host dispersal and aggregation is restricted to spatial and temporal patterns of cropping . Although there have been a few attempts to produce advice on optimal crop spatial organisation for restricting pathogen evolution ( e . g . [50]; but see [51] ) , this question clearly still requires more investigation . Papaïx et al . [52] analysed the pathogenicity structure of leaf rust populations ( Puccinia triticina ) on wheat ( Triticum aestivum ) at the scale of France and found coexistence among qualitative specialists ( very restricted host range ) , quantitative specialists ( large host range but transmitted efficiently only by a few of them ) and generalists ( large host range with roughly equal preference among them ) . This high diversity of pathogenicity patterns is consistent with the high diversity found here when the host disperses essentially locally . In addition , we found , in the present study , that more generalist , and thus less damaging , pathogen genotypes were favoured when the host population fluctuates in time because , under these conditions , the pathogen population is forced to track host oscillations . Such fluctuations in variety frequencies could be impose in agricultural landscapes to prevent the evolution of pathogen populations toward highly specialised and adapted morphs . Moreover , local spatial aggregates of crops or varieties that behave similarly with regard to a given disease should be avoided in order to prevent the local emergence of specialised ( and damaging ) pathogen genotypes . Another example is provided by invasive plants which are likely to exhibit low diversity with respect to disease resistance [53] , [54] . In that context , the use of fungal pathogens as biocontrol agents has often been used against invasive weeds with several successful examples , but in a high number of cases bio-control has failed [55] , [56] . Metapopulation processes and dispersal features are known to be important determinants in the success of biological control by parasites [57] , [58] . By considering pathogen evolution , our results emphasise the importance of the spatial scale of interaction between host and pathogen as this may influence the level of pathogen adaptation to host genotypes and the nature of disease dynamics ( whether or not boom-and-bust cycles occur ) , and thus the success of controlling a recently introduced host . However , the resistance structure of host populations can respond rapidly to selection pressures imposed by a pathogen [54] and further work must consider the coevolution of host and their parasite . Relatively little is known about actual patterns of generalisation and specialisation in natural systems largely because many fewer studies have focused on the pathogenicity structure of pathogen populations [52] , [59] than on the resistance structure of their host populations [12] . In general though , assessments of pathogen population structure have found multiple outcomes at the level of individual populations ( monomorphic to highly polymorphic ) – an observation likely reflecting a number of factors including local adaptation [22] , fitness costs associated with pathogenicity [60] and stage in a frequency-dependent cycle of interaction between pathogen and host [61] , as well as dispersal capability ( isolation-by-distance ) and life history features associated with effective local survival . The present approach gives insights into the role of host and pathogen dispersal in driving pathogen diversity and adaptation and encourages further characterisation of pathogenicity structure of crop and natural pathogen populations . | Relatively little is known about actual patterns of generalisation and specialisation in natural plant-pathogen systems largely because many fewer studies have focused on the pathogenicity structure of pathogen populations than on the resistance structure of their host populations . The spatial scale of interactions between a pathogen and its host is seen to be of prime importance in determining evolutionary trajectories of host-pathogen metapopulations because migration rates of one of the species affect the spatial and temporal heterogeneity of selection on the other . Here we develop a simulation model to specifically examine the joint effects of host and pathogen dispersal on the evolution of pathogen specialisation in a spatially explicit metapopulation . The present approach gives insights into the role of host and pathogen dispersal in driving pathogen diversity and adaptation and encourages further characterisation of the pathogenicity structure of crop and natural pathogen populations . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"evolutionary",
"ecology",
"medicine",
"and",
"health",
"sciences",
"ecology",
"population",
"modeling",
"epidemiology",
"biology",
"and",
"life",
"sciences",
"population",
"biology",
"spatial",
"epidemiology",
"computational",
"biology",
"evolutionary",
"biology"
] | 2014 | Evolution of Pathogen Specialisation in a Host Metapopulation: Joint Effects of Host and Pathogen Dispersal |
Food borne trematodes ( FBTs ) are an assemblage of platyhelminth parasites transmitted through the food chain , four of which are recognized as neglected tropical diseases ( NTDs ) . Fascioliasis stands out among the other NTDs due to its broad and significant impact on both human and animal health , as Fasciola sp . , are also considered major pathogens of domesticated ruminants . Here we present a reference genome sequence of the common liver fluke , Fasciola hepatica isolated from sheep , complementing previously reported isolate from cattle . A total of 14 , 642 genes were predicted from the 1 . 14 GB genome of the liver fluke . Comparative genomics indicated that F . hepatica Oregon and related food-borne trematodes are metabolically less constrained than schistosomes and cestodes , taking advantage of the richer millieux offered by the hepatobiliary organs . Protease families differentially expanded between diverse trematodes may facilitate migration and survival within the heterogeneous environments and niches within the mammalian host . Surprisingly , the sequencing of Oregon and Uruguay F . hepatica isolates led to the first discovery of an endobacteria in this species . Two contigs from the F . hepatica Oregon assembly were joined to complete the 859 , 205 bp genome of a novel Neorickettsia endobacterium ( nFh ) closely related to the etiological agents of human Sennetsu and Potomac horse fevers . Immunohistochemical studies targeting a Neorickettsia surface protein found nFh in specific organs and tissues of the adult trematode including the female reproductive tract , eggs , the Mehlis’ gland , seminal vesicle , and oral suckers , suggesting putative routes for fluke-to-fluke and fluke-to-host transmission . The genomes of F . hepatica and nFh will serve as a resource for further exploration of the biology of F . hepatica , and specifically its newly discovered trans-kingdom interaction with nFh and the impact of both species on disease in ruminants and humans .
Food borne trematodes ( FBTs ) are an assemblage of platyhelminth parasites that are transmitted through the food chain [1] . Among the four major groups of FBT infections recognized as neglected tropical diseases ( NTDs ) by the World Health Organization [2] , fascioliasis stands out due to its zoonotic impact on both human and animal health [3] . Fasciola species are major pathogens of domesticated ruminants , but they infect numerous other species of mammals , including people [4] . Due to the significant burden to livestock globally , with annual losses exceeding US $3 . 2 billion [5] and public health with ~50 million infected people [4] , these parasites are among the most-extensively studied FBTs . Like other digenetic trematodes , Fasciola hepatica has a complex developmental cycle [1] . The hermaphroditic adult stage resides in the host bile ducts and reproduces sexually , releasing thousands of eggs each day that pass with the bile into the intestines and exit in the fecal stream . Eggs that reach fresh water embryonate over a couple of weeks , hatching a free-swimming miracidium that seeks out and infects a snail of the family Lymnaeidae . Within the snail , the parasite progresses through sporocyst , redia , and daughter redia stages by asexual replication and development , resulting in the release thousands of the cercariae [6] . The free-living , aquatic cercaria encysts as the metacercarial stage on solid substrates , including vegetation at the margins of the watercourse . When infected vegetation ( for example , uncooked watercress ) are ingested by a suitable host , the metacercaria excysts in the duodenum , transverses the wall of the small intestine , migrates through the peritoneal cavity , and penetrates the Glisson's capsule of the liver [7] . The migration of the juvenile fluke though the liver parenchyma into the biliary ducts damages the liver and provokes reactions associated with the acute phase of the infection . This phase is accompanied by systemic disease including fever , nausea and abdominal pain . Once the adult is established in the bile ducts , anemia , inflammation , fibrosis , cholangitis and biliary stasis may ensue . In this chronic phase adult worms can survive several years in the absence of intervention [8 , 9] . Despite its potent and broad action against other human parasitic flatworms the anthelmintic drug praziquantel has no effect on F . hepatica [10] . Triclabendazole ( TCBZ ) is the drug of choice since its effective against juveniles and adult liver flukes , but resistance to this benzimidazole has emerged in livestock in different countries [11] . There have been recent reports of human fascioliasis refractive to TCBZ treatment in Peru and Chile [12 , 13] , highlighting a need for alternative drugs and treatments . In addition to being important pathogens themselves , some digeneans serve as vectors of bacterial pathogens . Neorickettsia ( family Anaplasmataceae ) belongs to a poorly characterized assemblage of obligate , intracellular α-Proteobacteria associated with serious , even fatal disease in mammals [14] . These bacteria can be horizontally transmitted from the fluke to host tissue invading and multiplying within mammalian cells such as macrophages , monocytes and other cells types , e . g . intestinal epithelium , eventually leading to severe disease . Neorickettsia can be detected by PCR in trematodes spanning the major lineages of the Digenea [15 , 16] , but it has never been reported from a trematode that is itself a prevalent human and livestock pathogen . Furthermore , the fact that Neorickettsia is not found among all fluke species ( or all members of infected species ) suggests that these endobacteria are not essential to fluke survival . Indeed , the exact nature of their relationship is remains unclear . Here we describe the second reported reference genome of the common liver fluke , F . hepatica and the first discovery and genome sequences of the Neorickettsia endobacteria of F . hepatica . In contrast to the previously sequenced isolate from cattle from the UK , the presently described strain , taken from a sheep in Oregon , US , was infected with a Neorickettsia species closely related to the etiological agents of Potomac horse and Sennetsu fevers . Histological , PCR , and gene sequence analyses revealed its presence in tissues of liver fluke isolates from Oregon , and in one of several liver fluke isolates from Uruguay that were screened . Taken together , these genomes represent a benchmark resource for studies of trematode and Neorickettsia biology , pathogenesis and evolution .
The nuclear genome of F . hepatica Oregon was sequenced and assembled with a total length of 1 . 14 Gb , N50 number of 2 , 036 and N50 length of 161 kb ( S1 Table ) . Completeness was estimated at 90 . 6% using the CEGMA method [17] . GC content was similar to other Food Borne Trematodes ( FBTs ) including Clonorchis sinensis [18] and Opisthorchis viverrini [19] , but differed from blood flukes . Intriguingly , the genome of F . hepatica Oregon had a markedly higher repeat content ( 55 . 29% , S1 Table ) than other FBTs , including the recently published genome of F . hepatica United Kingdom ( 32 . 0% ) isolated from cattle . We detected > 92 Mb corresponding to LTR elements , 268 Mb corresponding to LINEs , and 235 Mb of unclassified repetitive sequences , values all higher than other trematodes . Functional RNAs including rRNA , tRNA and miRNA ( S2 Table ) were identified , representing 0 . 002% of the coding genome , most supported by RNAseq data . Consistent with other FBTs [19 , 20] , a very small percentage of the genome assembly was predicted to encode proteins ( 1 . 08% , considering only exonic regions ) . A total of 14 , 642 protein-coding genes were identified using a combination of de novo and evidence-based methods . Predicted genes had an average of 3 . 3 exons and 2 . 3 introns , average footprint of 3 , 078 bp , and average coding length of 837 bp ( S1 Table ) . Comparisons of protein-coding genes between F . hepatica Oregon and F . hepatica UK ( S1 Fig ) revealed that most functional elements ( e . g . KEGG orthologous groups ) were shared , despite the fact that the gene models showed relatively poor overlap . In general , both genome annotations showed a predominance of short genes compared with other trematodes . This could be an indication of incomplete gene models in both assemblies , as the size , complexity , and incompleteness of both hindered gene prediction . Accordingly , long reads from third generation sequencing and additional RNAseq data will be needed to improve gene predictions , as demonstrated for S . mansoni [21] and C . sinensis [20] . Comprehensive functional annotation of the deduced proteins of F . hepatica Oregon is provided in S3 Table , including ( a ) 3 , 907 unique InterPro protein domains predicted from 8 , 609 proteins , associated with 1 , 147 unique gene ontology ( GO ) terms , ( b ) 3 , 175 proteins associated with 2 , 685 KEGG orthologous groups , ( c ) 339 proteins classified as putative proteases , ( d ) 65 proteins classified as protease inhibitors , and ( e ) 855 of proteins predicted to be secreted . Majority of the genes ( 94% of the predicted 13 , 740/14 , 642 ) were supported by RNAseq data from the developmental stages sampled for this study ( eggs , metacercariae , and adult flukes; S2 Fig ) . Of the >6 , 000 genes expressed in these stages , ~2 , 500 showed no differential expression , with GO terms related to core cellular functions such as translation , RNA processing , and vesicular transport ( S3 Table ) . Among the differentially expressed gene sets resulting from the DESeq analysis , stage-specific overexpressed gene sets were identified ( e . g . , for metacercariae , genes significantly overexpressed in the metacercarial stage relative to the adult and to the egg , but not differentially expressed between adult and egg ) . Using these criteria , four of the top five most significantly metacercaria-overexpressed genes ( 2 , 076 total ) were cysteine proteases , including four papain-like family proteases and one C13-family protease ( P < 10−15 for all comparisons ) , while the most significantly adult-overexpressed was also a papain-family cysteine protease ( P < 10−38 ) . Among the other 1 , 169 adult-overexpressed genes , four of the top 11 were tubulin genes ( P < 10−20 ) . Fewer genes ( 259 ) were egg-overexpressed since adult females contain eggs expressing transcripts , but the most significantly differentially expressed gene in this set ( P < 10−13 ) was a glucose-6-phosphate dehydrogenase . Approximately 88 . 5% of the 14 , 642 inferred proteins from the F . hepatica Oregon isolate found at least one BLAST hit ( E < 1e-05 ) to non-Fasciola proteins in the non-redundant database ( NR ) , with most matching sequences from other FBTs ( particularly the liver flukes C . sinensis and O . viverrini ) . 11 . 5% of genes are Fasciola-specific with respect to NR ( S3 Table ) , 4 . 9% of which were assigned additional functional annotations ( Interpro domains , GO terms or KEGG orthologous groups; compared to 67 . 7% for genes with non-Fasciola NR matches ) . The putative Fasciola-specific genes were enriched for GO terms related to cysteine-type endopeptidase inhibitor activity and neurotransmitter secretion ( S4 Table ) . Protein conservation among flatworm parasites and their hosts was analyzed by clustering predicted proteins from 10 genomes into orthologous protein families ( OPFs; Fig 1A ) , and 7 , 624 F . hepatica proteins were included in 5 , 721 unique OPFs with proteins from other flatworms including the free-living planarian Schmidtea mediterranea , trematodes ( including schistosomes , the liver flukes C . sinensis and O . viverrini ) , and mammalian hosts ( human , cow and sheep ) . Some 2 , 875 F . hepatica proteins ( 1 , 451 OPFs ) were conserved across the 10 species , and these were more likely than other genes to be differentially expressed across developmental stages of F . hepatica ( P = 7 x 10−5 , binomial distribution test; S3 Table ) . In contrast , 393 F . hepatica genes ( 359 OPFs ) were conserved between F . hepatica and at least one other FBT ( C . sinensis and/or O . viverrini ) ; these were significantly enriched for GO terms related to microtubule-based processes , cysteine endopeptidase activity and pH regulation ( S4 Table ) . An additional 29 genes ( 29 OPFs ) were conserved with at least one FBT and at least one host species; these were significantly enriched for the GO terms related to phagocytosis and L-ascorbic acid binding . OPF analysis also enabled the identification of gene sets specific to schistosomes; 1 , 365 OPFs conserved among at least two of the three species of Schistosoma were analyzed . Like the FBTs , these were significantly enriched for GO terms related to microtubule activity and cysteine endopeptidases ( see below ) . This suggests that certain functions are conserved across the platyhelminth clades despite clear divergence at the sequences level , a possible indication of rapid evolution [22] . Excreted and secreted proteins ( ESPs ) play a crucial role in parasitism . We identified 855 ( 5 . 8% ) proteins with computationally predicted signal peptides but no transmembrane domains ( S3 Table ) , indicating they may be secreted/ excreted . These proteins were significantly enriched for GO terms related to proteolysis ( S4 Table ) . Again , the most significantly enriched molecular process GO term was “cysteine-type endopeptidase activity” ( GO:0004197 ) . Secreted cysteine proteases have a well-defined role in the biology of F . hepatica and liver fluke disease [23] . Cathepsin L’s are predominant in adult ESPs , where they participate in feeding , immune evasion and immune modulation . Distinct suites of cathepsin L’s and cathepsin B’s are abundant in the juvenile fluke , participating in excystment , migration through gut wall and liver capsule , and immune evasion [24 , 25] . Although it was known that liver fluke cathepsins constitute a multigene family [26] , the complexity and diversity within the family is now apparent . In addition to the six known cathepsin L’s , other isoforms were detected consistently in both the Oregon and the UK isolates , raising the total count to 14; most of these overexpressed in the adult stage ( Fig 1B ) . Independent amplifications of cathepsin L’s occurred in schistosomes and the opisthorchiids , but the resulting gene copy number is less than in F . hepatica . In contrast , Cathepsin F’s showed a divergent pattern in these lineages , with single enzymes in Fasciola and schistosomes , and an amplified family in the carcinogenic fish-borne liver flukes ( Fig 1B ) . A similar pattern of independent amplifications among trematodes was observed for cathepsin B’s ( Fig 1C ) , again with a distinct expansion in F . hepatica . In contrast to cathepsin Ls , cathepsin Bs were overexpressed in metacercariae ( MC ) , confirming biochemical , genetic and proteomic evidence of differential expression along the life cycle [27] . Interestingly , within both the cathepsins L and B , a clade comprising a single enzyme from each trematode species and vertebrates was identified , which might be basal to all the lineage-specific expansions . F . hepatica enzymes of this clade have not been described yet , and , notably , they are expressed in eggs ( Fig 1B and 1C ) . The remarkable amplification and diversity of secreted cysteine proteases in trematode lineages suggests key roles during parasite adaptation . Diverse trematodes express different ( and amplified ) subfamilies of cathepsins , reflecting their host parasite relationships , including host niche , organ sites , and transmission strategies . For example , cathepsins B and an L3 ( CL3 ) participate in transit of the juvenile liver fluke through the gut wall with collagenolysis [28 , 29] , whereas juvenile of the fish-borne liver flukes ascend into the biliary tree through the ampulla of Vater [30] . In turn , cathepsins F and the aspartic protease cathepsin D are characteristically overrepresented in these carcinogenic liver flukes [19 , 31] . The blood flukes , on the other hand , invade the skin of their hosts , with conserved serine proteases essential to this process , with the cysteine protease cathepsin B providing critical activities in some species [32] . Asparaginyl endopeptidases , Class C13 ( also known as legumain ) [33] were expanded in F . hepatica with ≥10 members , and were also differentially expanded among the trematodes; 3 copies in S . mansoni , 5 in S . japonicum , 4 in C . sinensis and ~ 100 in O . viverrini . ( Fig 1D ) . These proteases might participate in the activation of cathepsins and the digestion of infected host tissues , liberating essential amino acids . A recent proteomic study of ESPs from juvenile and adult F . hepatica provide further support for the differential expression of these gene families [34] , confirming our transcriptomic data ( S3 Table ) . For example , within the cathepsin B family , members of 10 out of 13 clusters are detected by LC-MS/MS in ESPs; while 3 isoforms are exclusively expressed by adults , 2 are characteristic of juveniles ( also detected by RNAseq in metacercariae ) , and 5 are expressed in both stages but clearly predominant in juveniles ( Fig 1C ) . Similarly , a predominance of expression of cathepsin Ls variants in adults is observed at proteomic level consistent with our transcriptomic data ( Fig 1B ) . Within these some of the novel clades here described were detected as being expressed . Metabolic pathways predicted in the F . hepatica Oregon strain were compared to those of other sequenced flatworms . All parasitic flatworms showed a significant reduction in metabolic capabilities compared to free living platyhelminth species , including planaria ( Fig 2 ) . As shown previously [35] , parasitic flatworms depend on the hosts for provision of fatty acids . Unlike the blood flukes , however , F . hepatica and the other liver flukes possess enzymatic pathways for fatty acid elongation by reversal of beta-oxidation ( S3A Fig ) and fatty acid catabolism ( S3B Fig ) , allowing them to take advantage of the fatty acid rich environment of bile . Additional differences between blood and liver flukes were evident in amino acid metabolism . Inabilities to synthesize several amino acids were generally observed in neodermatan flatworms , including Fasciola . However , the liver flukes operate a complete catabolic pathway of aliphatic amino acids , and enzymes of these pathways ( e . g . branched chain amino acid aminotransferase [BCAT , EC . 2 . 6 . 1 . 42] ) are missing in schistosomes ( S3C Fig ) . Aliphatic amino acids are more abundant in the bile than in blood , which may have been exploited since it facilitates access to protein synthesis precursors and alternative energy sources . While expanded families of secreted proteases are involved in protein digestion , conserved oligopeptide transporters that mediate the uptake of di- and tri-peptides in metazoans were not identified [36] , suggesting the lack of these specific transporters in trematodes at large . Consequently , protein digestion up to individual amino acids may occur extracellularly , explaining the rare presence of usually cytoplasmic enzymes as leucine aminopeptidase in the secreted products and vesicles released by the parasite [37] . Further metabolic differences may have evolved in liver flukes compared to blood flukes in relation to an environment characterized by low oxygen tension . Flukes switch from aerobic to anaerobic metabolism in the low oxygen environment of the bile duct , but instead of fermenting carbohydrates to lactate , the parasite exploits the more energy-efficient malate dismutation pathway [38] , where phosphoenolpyruvate from glycolysis is converted to oxaloacetate via the phosphoenolpyruvate kinase ( PEPCK ) , and further reduced to malate . After entering the mitochondria , some malate is oxidized to acetate , and some is reduced to succinate and transformed to propionate , in a series of reactions that reverts the Krebs cycle ( Fig 3A ) , providing a source of electrons for the respiratory chain finally yielding five ATP molecules per glucose molecule . While the whole pathway was precisely described biochemically , we now for the first time identify the cognate enzymes ( Fig 3A ) . To provide further insight into the metabolism of F . hepatica , we compared the metabolic pathway modules present and complete in F . hepatica , the carcinogenic liver flukes C . sinensis and O . viverrini , the blood fluke S . mansoni and mammalian hosts of F . hepatica ( sheep , cow and human ) were incorporated in the analysis ( S4A Fig ) . Twenty-five KEGG pathway modules were complete in our assembly ( i . e . , contain the complement of KO’s necessary to convert the initial substrate to the final product based on strict completion [39] ) . These values are similar in other trematodes , but far lower than in mammals . Use of a lenient completion criterion of ≤2 missing steps in a module extended these values , with similar trends ( S4B Fig ) . The analysis also identified modules that differ between the liver flukes and S . mansoni . Module M00020 ( serine biosynthesis ) also showed differences consistent with those observed in amino acid metabolism . Further differences in inositol phosphate metabolism were detected with two steps missing in S . mansoni ( R03427 and R04372 , INPP1 and INPP4 phosphatases ) . Module M00087 ( beta-oxidation ) occurs in liver flukes but is absent from schistosomes . Notably , this module revealed differences with the host , since in step 2 ( R04738 ) , F . hepatica shares two KOs with mammals , but there is an additional , putative platyhelminth-specific ortholog ( K01692 , EC 4 . 2 . 1 . 17 ) corresponding to enoyl coA-hydratase , which warrants investigation in flatworms . Differences between liver fluke and mammal were evident in module M00009m , corresponding to the tricarboxylic acid cycle ( Fig 3B ) . The fumarate forming reaction ( R01082 ) is dependent on fumarate hydratase class II enzyme ( EC 4 . 2 . 1 . 2B , K01679 ) by the hosts , while a second fumarate hydratase , class I enzyme for this step ( EC 4 . 2 . 1 . 2A , K01676 ) , was detected in trematodes , an enzyme that might participate in the reverse step of the malate dismutation pathway ( above ) . The most striking feature of the F . hepatica Oregon isolate was an apparent infection with Neorickettsia endobacteria ( nFh ) . Alpha-protobacterial sequences were first identified among “contaminating” sequences in the F . hepatica genome and the presence of Neorickettsia was confirmed and validated by both PCR and 16S rRNA sequencing ( S5 Fig ) . The genome of nFh was assembled from 241 , 957 2x100bp read pairs that were identified during the sequencing of F . hepatica Oregon . A single 859 , 205 bp scaffold with average 56 . 3x sequence coverage was constructed from two contigs joined by 189 bp of inferred gaps ( Fig 4 ) . This novel Neorickettsia genome was similar in size and GC content to those of previously sequenced Neorickettsia species [40 , 41] . Full genome alignments indicated that , with the exception of a small inversion , it shared nearly complete synteny with the genomes of Neorickettsia risticii and N . sennetsu ( Fig 5A ) . Synteny among Neorickettsia species may reflect the lethality of large genome rearrangements due to a reduced set of DNA repair genes , but this may have increased the genetic variation in a stable intra-trematode environment by accumulation of mutations in non-essential genes [41] . Table 1 outlines the inferred features of nFh . Similar to related species , nFh encodes 33 tRNA genes and one copy each of 5S , 16S , and 23S rRNA genes . A total of 744 protein-coding genes were predicted , slightly fewer than N . risticii and N . sennetsu ( Table 1 ) . Gene conservation analysis among representative bacterial species of the Anaplasmataceae identified three orthologous protein families ( OPFs ) that were conserved in all analyzed species except nFh ( S5 Table ) . Closer inspection revealed that these genes may be present and intact , although absent from the gene calls . Two additional OPFs were conserved in all sequenced Neorickettsia except nFh ( S5 Table ) , but in both of these cases the corresponding sequences were identified but stop codons appeared to disrupt them . S5 Table presents a functional annotation of the 744 predicted proteins of nFh , including ( a ) 1 , 453 unique InterPro protein domains predicted from 620 proteins and associated with 596 unique gene ontology ( GO ) terms , ( b ) 720 proteins associated with 509 KEGG orthologous groups , further binned into 120 enzymatic pathways and 101 pathway modules , ( c ) 25 proteins classified as putative proteases , ( d ) two protease inhibitors , and ( e ) 25 proteins with secretion signals ( which were enriched for biological process GO terms related to proteolysis and protein transport; S4 Table ) . Protein transporters such as porins , identified in previous proteomic studies might transport nutrients from the host cytoplasm [42] . Whether Neorickettsia enzymes interact with those of the fluke is a fascinating but unresolved question . Revealingly , however , N . risticii synthesizes nucleotides , vitamins , and cofactors that the fluke cannot , raising the possibility that they may be harvested by the trematode for their mutual advantage [41] . While sequencing reads from the previously reported UK strains ( SRA Project ID: ERP006249 ) did not map to our nFh genome ( Fig 6 ) , suggesting that no Neorickettsia DNA was present in the samples , one of our Uruguay isolate ( out of five that were screened ) tested positive for the presence of Neorickettsia by 16S rDNA PCR ( S5 Fig ) . Whole genome sequencing of this sample recovered the genome of nFh ( 99 . 9% breadth of genome coverage; S6 Table ) , allowing a comparative analysis of sequence variation in both the Neorickettsia and the fluke genomes . In total , 15 single nucleotide variants were identified between the two nFh genomes , 11 of which occurred within the coding regions ( 7 non-synonymous and 4 synonymous variants; S7 Table ) . Notably , 2-C-methyl-D-erythritol 4-phosphate cytidylyltransferase ( EC 2 . 7 . 7 . 60; AS219_00645 ) , a key enzyme in the MEP pathway of isoprenoid biosynthesis [43] , was found among the genes that harbored non-synonymous SNPs . It has been hypothesized that the Wolbachia endosymbionts of Brugia malayi and Dirofilaria immitis rely on their helminth host for the completion of the MEP pathway [44 , 45] . Additionally , in many pathogenic and opportunistic bacteria , the MEP pathway intermediate ( HMB-PP ) has the capacity to modulate vertebrate host immune response [46] , suggesting an interesting possibility of the involvement of the isoprenoid biosynthesis pathway in the host-parasite-endosymbiont interactions . Overall , the genetic distance ( 1-ibs , identity by state ) between the Oregon ( US ) and Uruguay ( UY ) nFh genomes ( 1 . 75 × 10−5 ) was three orders of magnitude lower than that estimated between the respective nuclear genomes of F . hepatica ( 1 . 08 × 10−2; S8 Table ) . The observed level of genetic divergence between the US and UY isolates indicated that these flukes are not substantially more closely related to each other than either is to the five published UK isolates ( S8 Table ) although their Neorickettsia endosymbionts are genetically close to each other . A phylogenetic analysis was undertaken using the 16S rRNA sequences from nFh and 16 other species and isolates; the findings were similar to those previous reports [15] . Based on the 16S rRNA locus , nFh is closely related to the agent of Sennetsu fever ( a strain of N . sennetsu ) and a Neorickettsia isolate isolated from species of Metagonimoides ( Heterophyidae ) ( Fig 5B ) . A complementary phylogenetic analysis was undertaken with conserved , single-copy homologues from sequenced species of Neorickettsia and representatives of the Anaplasmataceae ( Fig 5C ) ; in regard to this collection of 473 gene families , nFh appeared to be closer to N . risticii and N . sennetsu than to N . helminthoeca , the agent of salmon poisoning in dogs , consistent with synteny-based observations . Approximately 97% of the predicted proteins of nFh showed BLASTP matches in NR ( e-value ≤ 1e-05 , S5 Table ) . The top hits were to N . risticii or other Neorickettsia species . Indeed , most nFh genes ( 721 of 744 ) belonged to 719 OPFs shared with other Neorickettsia species ( S5 Table ) . Of the 22 genes that were excluded from OPFs , 17 failed to find a match in NR and most lacked other functional annotations; the other five matched to hypothetical proteins from N . risticii and N . sennetsu . A single OPF was identified with members from nFh , Wolbachia species , Anaplasma phagocytophilum and Ehrlichia chaffeensis . The nFh gene assigned to this OPF was annotated as a replicative DNA helicase . Further assessment will be needed to validate these genes and explore their roles . A set of 625 OPFs contained members common to all four sequenced Neorickettsia genomes; 83 of these OPFs were specific to Neorickettsia . The nFh proteins included in the Neorickettsia-specific OPFs were enriched for cellular component GO terms related to the outer membrane and biological process GO terms related to transport ( S4 Table ) . The association with the cell surface suggested a role in endobacterial-digenean host interactions . Although reports of Neorickettsia-infected trematodes have emerged , they relied on detection by PCR; localization within the trematode was poorly established . Gibson et al . [47] employed Ig from the serum of a horse infected with N . risticii to detect Neorickettsia in eggs from the bat-infecting trematode Acanthatrium oregonense to support the hypothesis of vertical transmission . The same serum was used to localize Neorickettsia in discrete developmental stages of Plagiorchis elegans , a trematode of rodents and birds [14] . We attempted to use the horse serum to localize nFh in adult F . hepatica Oregon , but background staining interfered with interpretation of the signals . However , polyclonal antibodies raised against recombinant surface protein-3 of P . elegans Neorickettsia ( PeNsp-3 ) provided useful to support localization studies ( Fig 6 ) . The PeNsp-3 protein ( Genbank KX082665 ) and of nFh ( AS219_03540; S5 Table ) share 98% identity . Minimal background signal occurred with the PeNsp-3 antisera , and nFh were sensitively detected as a ‘donut’-shaped structure surrounding the blue DAPI-stained nucleus , consistent with the staining pattern expected for surface proteins ( Fig 6C ) [48] . Whereas staining of Neorickettsia surface protein was not observed in Neorickettsia-negative ( as confirmed by PCR; Fig 6A and 6H ) F . hepatica from Uruguay endobacteria were detected in six of six individual adult F . hepatica worms from Oregon . Because of the size of F . hepatica ( ~2 cm; F ) and because Neorickettsia may be transmitted vertically , analysis focused on intra-uterine eggs and reproductive tissues . Endobacteria were frequently detected in varying numbers in the ovary , ootype , Mehlis’ gland , vitelline glands and in intrauterine eggs ( Fig 6D–6G ) as well as mature eggs isolated from liver tissue ( Fig 7 ) . The presence of nFh in female reproductive tissue is highly suggestive of vertical transmission . Furthermore , we analyzed by PCR adult flukes obtained after an experimental infection with Oregon metacercariae , detecting a few individual worms positive for the presence of nFh ( S7 Fig ) . More interestingly , eggs collected from this assay were both PCR positive and presented the characteristic images of Neorickettsia supports the notion of vertical transmission . Surprisingly , we also observed nFh in the testis and other parts of the male reproductive organs ( Fig 6I and S8 Fig ) . Although F . hepatica is a hermaphrodite , cross-fertilization is assumed to be the usual reproductive strategy [49] . The presence of Neorickettsia in spermatozoa and seminal fluid could provide an alternative route for fluke-to-fluke transmission , as it was described for tick-borne pathogens [50] , though further studies will be needed to explore this possibility . The somatic tissues of F . hepatica Oregon were mostly Neorickettsia-free . Clusters of nFh were occasionally seen in the tegument adjacent to some syncytial nuclei ( Fig 6B ) and in intestinal tissue , particularly near the oral suckers . Liver flukes use the oral suckers to penetrate the host tissues and anchor themselves to the bile ducts , thus providing a potential mechanism for fluke-to-host transmission of nFh . Several infectious diseases described in the medical and veterinary literature are attributable to Neorickettsia carried by digenean parasites; among the more relevant are the ‘Salmon Poisoning Disease’ ( SPD ) of dogs and the Elokomin fluke fever ( EFF ) of fish-eating mammals in the west coast of North America , Sennetsu fever described in humans mainly in Japan and southeast Asia , and Potomac Horse Fever ( PHF ) in the east coast of North America [15 , 16] . Notably , PHF , also known as ‘churrido equino’ , has been described in horses in the Lake Merin region of Uruguay and Brazil ( reviewed in [51] ) . Several species of Neorickettsia based on pathology , serology , antigen profile and/or genomic sequence , are considered the causative agents for these diseases; in particular , Neorickettsia ( Ehrlichia ) sennetsu causes acute , debilitating , mononucleosis-like disease [52] , and has been implicated as a significant cause of human fevers of unknown etiology in southeastern Asia [53 , 54] . Whereas the disease potential of Neorickettsia found in F . hepatica remains to be established , the F . hepatica-nFh association should be explored as a cryptic rickettsial pathogen of humans and ruminants in regions endemic for fasciolosis [55] . Additionally , more thorough studies of both the vertical transmission of nFh among the developmental stages of the liver fluke , and the potential horizontal transmission to the mammalian host might shine a light on the mechanisms behind the pathology induced by Neorickettsia endosymbionts of digenean parasites .
Two isolates of Fasciola hepatica were analyzed: adult worms collected from livers of naturally infected sheep from a commercial slaughterhouse in Oregon ( provided by Baldwin Aquatics Inc . , Monmouth , Oregon ) , i . e . Oregon isolate; and worms isolated from livers of naturally infected sheep obtained from a commercial slaughterhouse in Montevideo , Uruguay , i . e . Uruguay isolate . For transcriptomic analysis , total RNAs were obtained from the egg , metacercarial and adult developmental stages ( in duplicate ) . Eggs were collected from gall bladder of naturally infected sheep . Metacercariae were purchased from Baldwin Aquatics Inc . ( Monmouth , Oregon ) . Tissue sections for the histological analysis were prepared from adult worms of the Oregon and Uruguay isolates . Fresh or ethanol-preserved adult worms were fragmented using a scalpel blade , and genomic DNA ( gDNA ) was extracted and purified using the kit E . Z . N . A . SQ Tissue DNA Kit ( Omega Bio-tek ) , and the yield and purified assessed by Bio-Analyzer . Whole genome shotgun fragment and paired-end sequencing libraries ( 3 kb and 8 kb ) were constructed from the gDNAs , as described [39 , 56] , and sequenced on the Illumina HiSeq2000 platform . Linker and adapter sequences were trimmed , and cleaned reads were assembled using ALLPATHS-LG [57] . Pygap , an in-house assembly improvement tool , was used to join and extend contigs using unassembled reads when possible . Annotation of different features present in the assembly was done as previously described [58] and outlined in S1 Text . Total RNA was extracted from eggs and adults from the gall bladder of naturally infected sheep and metacercariae ( Baldwin Aquatics Inc . , Monmouth , Oregon ) using TRIzol reagent ( Invitrogen/Life Technologies , Carlsbad , CA ) according to the manufacturer’s instructions , and treated with Ambion Turbo DNase ( Ambion/Applied Biosystems , Austin , TX ) . As previously described [59] , RNA quality and yield were assessed , the purified RNA was poly ( A ) selected , reverse transcribed , paired-end cDNA libraries were generated , sequenced on the Illumina HiSeq 2000 platform and reads were analytically processed . Remaining , high-quality RNAseq reads ( from one egg , two metacercariae and two adult biological replicates ) were aligned to the genome assembly and constitutively expressed and differentially expressed genes were identified using standard protocols outlined in S1 Text . A total of 126 contigs were identified as being from bacterial origin in the F . hepatica genome assembly , and BLAST analyses indicated significant homology to Neorickettsia species . The total complement of raw reads were re-mapped to the 126 Neorickettsia contigs using BWA-MEM version 7 . 10 with default parameters [60] , and matching reads were assembled and assembly improved using standard protocols ( see S1 Text ) . The genome assembly was annotated via the NCBI prokaryotic genome annotation pipeline [61] . Deduced protein sequences were subjected to BLASTP against informative databases , including NCBI NR , InterPro , gene ontology ( GO ) , KEGG , MEROPS using default cutoffs and release versions as specified in the S1 Text . Module completion was assessed as described [39] , and transmembrane domains and classical secretion peptides were predicted using standard protocols ( see S1 Text ) . Inferred protein sequences of F . hepatica were compared to proteins from other trematodes & cognate mammalian hosts and from Neorickettsia were compared to proteins from representative species from the Anaphasmataceae , included all four fully sequenced Neorickettsia ( accession numbers are provided in S1 Text ) . Orthologous protein families ( OPFs ) were constructed from pairwise InParanoid comparisons using MultiParanoid [62] . More detailed phylogenetic analyses at a level of rRNA sequences or single copy genes were performed as outlined in the S1 Text . F . hepatica Oregon and F . hepatica UK gene sets were compared using orthologs identified by Orthofinder v . 0 . 7 . 1 [63] . We sequenced the genomic DNA of an F . hepatica isolate from Uruguay that was PCR-positive for Neorickettsia using the Illumina platform ( 2 × 100bp paired-end sequencing ) , as previously described [56] . We included the published genomes of the United Kingdom isolates ( SRA Project ID: ERP006249 ) in the variant analysis to help contextualize our data . Genomic reads were mapped against the combined Oregon reference assembly of Neorickettsia and F . hepatica using bwa v0 . 7 . 15 [60] , followed by removal of PCR and optical duplicates using picard tools v2 . 6 . 0 [64] . Single-nucleotide variants were called via local de-novo assembly of haplotypes using the GATK pipeline v3 . 6 [65] . The following set of quality filters were applied to obtain high-confidence SNP calls: DP ( maximum depth ) > median depth+ ( median absolute deviation×1 . 4826 ) ×2; QD ( variant confidence divided by the unfiltered depth of non-reference samples ) < 2 . 0; FS ( Phred-scaled p-value using Fisher’s Exact Test to detect strand bias in the reads ) > 60 . 0; MQ ( Root Mean Square of the mapping quality of the reads across all samples ) < 40 . 0; MQRankSum ( Mann-Whitney Rank Sum Test for mapping qualities ) < -12 . 5; ReadPosRankSum ( Mann-Whitney Rank Sum Test for the distance from the end of the read for reads with the alternate allele ) < -8 . 0 . Using SnpEff [66] , variants were annotated based on their genomic locations and predicted coding effects . The genetic distance between isolates ( 1-ibs , identity by state ) were computed using PLINK v1 . 90 after excluding loci with missing genotypes in any of the isolates . To investigate vertical transmission of nFh , genomic DNA was extracted from individual worms obtained after two experimental infections in bovines ( Bos taurus ) performed at the Experimental Farm of the Institute of Hygiene , Montevideo , Uruguay , following international standards for care of research animals , and approved by the National Committee of Experimental Animal Health ( CHEA ) . Polled Hereford calves negative to F . hepatica by fecal egg count and ELISA ( Piacenza et al . , 1999 ) and treated orally with ivemectin 1% ( Mexiver , Laboratorios Santa Elena ) , were used in immunization studies that included challenge infection by mouth with 400 metacercariae ( MCs ) . The MCs were obtained from Baldwin Aquatics , Oregon ( assay 1 ) or DILAVE , Uruguay ( assay 2 ) . The cattle were euthanized at a commercial abattoir on week 20 , and adult flukes were recovered from the liver of each of the calves , and flukes from each calf stored separately in >70% ethanol . Liver fluke eggs from the gall bladder of the calves also were recovered , and stored in the immunization group . DNA was extracted from individual worms of each assay , and from the pooled eggs , as described above . The presence of Neorickettsia within the F . hepatica adult flukes was investigated by nested PCR directed to the 16s rRNA gene , as described [67] . Oregon strain flukes from sheep ( and Neorickettsia-negative flukes from Uruguay ) were fixed first in 70% ethanol and then in 10% buffered formalin overnight , tissue processed ( Shandon 1000 Tissue Processor , Thermo Scientific , Waltham , MA , USA ) , embedded in paraffin , sectioned at 5 μm . Serial sections were used for immunohistochemical studies and hematoxylin & eosin staining ( S6 Fig ) . The sections stained with hematoxylin & eosin ( according to standard technique ) were used to assess morphology and determine the anatomical structures to be expected on adjacent slides used for immunohistochemical localization of nFh . Unstained tissue sections were rehydrated and blocked with 5% bovine serum albumin ( Sigma , St . Louis MO , USA ) for 30 min to prevent non-specific antibody binding . Polyclonal mouse antisera raised against a recombinant Neorickettsia surface protein from Plagiorchis elegans ( Genbank Accession KX082665 , PeNsp-3 ) diluted 1:250 in phosphate buffered saline containing 0 . 1% Triton-X and 1% bovine serum albumin was used as the primary antibody . Anti-mouse IgG Alexa Fluor 488 ( Invitrogen ) was used as a secondary antibody for fluorescence microscopy . Wheat germ agglutinin 633 ( 200 μg/ml , Invitrogen , Carlsbad , CA , USA ) and DAPI ( Prolong Antifade with DAPI , Molecular Probes by Life Technologies , Carlsbad , CA , USA ) were used to label membranes and double-stranded DNA , respectively . Sections were examined using a wide field fluorescence microscope ( WFFM , Zeiss Axios Imager Upright Fluorescence Microscope ) with plan-apochromat 100X oil , 63X or 40X objectives . Fluorescence microscopy was performed at the Washington University Molecular Microbiology Imaging Facility ( http://micro . imaging . wustl . edu/ ) . | This report presents novel findings revealing ( a ) the genome sequence of the food-borne trematode Fasciola hepatica ( the liver fluke ) isolated from sheep , which stands out among neglected tropical diseases due to its zoonotic impact on both human and animal health and ( b ) the first instance ( and the genome ) of the rickettsial endobacterium of the genus Neorickettsia in F . hepatica . Using stage-specific gene expression data , we identified liver fluke proteins likely involved in host-parasite interactions , and using immunolocalization , we confirmed Neorickettsia in organs and tissues of the adult trematode . The presence of the bacteria in fluke reproductive tissues and eggs suggests a possible mechanism for vertical transmission , and the presence of bacteria in the oral sucker used to anchor flukes to the lining of the biliary tract suggests a potential mechanism for horizontal transmission to the mammalian host . This is of interest because related Neorickettsia cause severe , even deadly , illness in a variety of species , including humans . This is the first report to localize Neorickettsia endobacteria within the tissues of adult F . hepatica . The discoveries in our manuscript have wide impact for the fields of both the pathophysiology and evolution of Fasciola and related FBTs , and the transmission strategies of Neorickettsia . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"helminths",
"computational",
"biology",
"animals",
"trematodes",
"genome",
"analysis",
"research",
"and",
"analysis",
"methods",
"fasciola",
"hepatica",
"proteins",
"flatworms",
"fasciola",
"biological",
"databases",
"comparative",
"genomics",
"pathogenesis",
"biochemistry",
"host-pathogen",
"interactions",
"database",
"and",
"informatics",
"methods",
"genetics",
"protein",
"domains",
"biology",
"and",
"life",
"sciences",
"genomics",
"genomic",
"databases",
"organisms"
] | 2017 | Genomes of Fasciola hepatica from the Americas Reveal Colonization with Neorickettsia Endobacteria Related to the Agents of Potomac Horse and Human Sennetsu Fevers |
Large-scale codon re-encoding represents a powerful method of attenuating viruses to generate safe and cost-effective vaccines . In contrast to specific approaches of codon re-encoding which modify genome-scale properties , we evaluated the effects of random codon re-encoding on the re-emerging human pathogen Chikungunya virus ( CHIKV ) , and assessed the stability of the resultant viruses during serial in cellulo passage . Using different combinations of three 1 . 4 kb randomly re-encoded regions located throughout the CHIKV genome six codon re-encoded viruses were obtained . Introducing a large number of slightly deleterious synonymous mutations reduced the replicative fitness of CHIKV in both primate and arthropod cells , demonstrating the impact of synonymous mutations on fitness . Decrease of replicative fitness correlated with the extent of re-encoding , an observation that may assist in the modulation of viral attenuation . The wild-type and two re-encoded viruses were passaged 50 times either in primate or insect cells , or in each cell line alternately . These viruses were analyzed using detailed fitness assays , complete genome sequences and the analysis of intra-population genetic diversity . The response to codon re-encoding and adaptation to culture conditions occurred simultaneously , resulting in significant replicative fitness increases for both re-encoded and wild type viruses . Importantly , however , the most re-encoded virus failed to recover its replicative fitness . Evolution of these viruses in response to codon re-encoding was largely characterized by the emergence of both synonymous and non-synonymous mutations , sometimes located in genomic regions other than those involving re-encoding , and multiple convergent and compensatory mutations . However , there was a striking absence of codon reversion ( <0 . 4% ) . Finally , multiple mutations were rapidly fixed in primate cells , whereas mosquito cells acted as a brake on evolution . In conclusion , random codon re-encoding provides important information on the evolution and genetic stability of CHIKV viruses and could be exploited to develop a safe , live attenuated CHIKV vaccine .
Many emerging infectious diseases are caused by arthropod-borne viruses ( arboviruses ) , almost all of which are single strand RNA viruses . The major outbreaks of dengue fever [1] , West Nile encephalitis [2] , Chikungunya fever [3] , and Rift Valley fever [4] that have occurred in recent decades , each with a significant impact on human health , highlight the urgent need to understand the factors that allow these viruses to invade new territories or adapt to new host or vector species [5]–[6] . Understanding the factors that shape the adaptability of these rapidly evolving infectious agents may provide new opportunities for their eventual control . Codon usage bias is an important indicator of the evolutionary forces shaping genomes and could arise either through neutral mutational pressure or because specific synonymous codons are selectively advantageous; for example , by increasing the efficiency and/or accuracy of protein expression by maximizing the match to cellular tRNA abundance . Thus , determining the underlying causes of codon bias has become a key topic in evolutionary genetics [7] . RNA viruses often exhibit codon biases that match the nucleotide biases across viral genomes as a whole , suggesting that background mutational pressure is the dominant factor shaping codon choice . However , natural selection may still act at the scale of overall nucleotide composition [8] . Direct selection for specific codon biases has also been documented . For example , in hepatitis A virus rare codons that utilize non-abundant tRNAs are preferred , slowing down the translation process to ensure proper protein folding [9] . The large-scale re-encoding of codon usage in poliovirus , influenza A virus and bacterial virus T7 by reverse genetic methods resulted in virus attenuation [10]–[15] , demonstrating that mutations at synonymous sites can indeed have a major impact on viral fitness . To date , all studies of codon re-coding have employed a specific approach such as codon de-optimisation [12] , [14] , codon pair de-optimisation [10] , [13] and increase of CpG/UpA dinucleotide frequency [11]; but all result in a reduction of viral fitness . Currently , there are no studies of codon re-coding in arboviruses . However , studies of these viruses are of special interest because their evolution is strongly constrained by host alternation , such that mutations that may be advantageous in one host type ( e . g . mosquitoes or mammals ) are deleterious in another [16] . Chikungunya virus ( CHIKV; Togaviridae; Alphavirus ) is a small ( 60–70 nm ) , enveloped , single-strand positive-sense RNA virus . Its genome of approximately 12 kb contains two open reading frames ( ORFs ) encoding non-structural and structural proteins , respectively [17] . In Swahili , “Chikungunya” means “bent walker” , reflecting the severe arthralgia associated with CHIKV infections . First isolated in Tanzania in 1952 [18] , CHIKV is transmitted by mosquito vectors of the Aedes ( Stegomyia ) subgenus and has caused a number of outbreaks in Africa and Asia during the last 50 years [19] . It is believed that the original natural history of viral transmission relies on virus maintenance in a yellow fever-like zoonotic sylvatic cycle involving non-peridomestic mosquitoes and nonhuman primates , as previously described in Africa . However , explosive urban outbreaks were associated with a dengue-like direct “human-mosquito-human” transmission cycle implicating A . aegypti or more recently A . albopictus mosquitoes [20] . Particularly large CHIKV outbreaks have occurred in Indian Ocean islands , in India , and in Southeast Asia since 2005 [6] . All these epidemics originated from east Africa and were associated with the East-Central-South African genotype [21] . Significantly , these recent epidemics are also associated with viral transmission by A . albopictus , although this was previously considered a secondary vector , and convergent adaptation to this mosquito has been observed in different geographical regions [22]–[23] . Approximately 40% of the population of Reunion Island was infected ( around 300 , 000 people ) [24] , including a relatively small proportion of severe disease in adults and newborns , as well as some vertical transmission [25]–[26] . Worryingly , the increasingly widespread distribution of A . albopictus may result in CHIKV epidemics in more temperate regions . Indeed , in 2007 , a CHIKV outbreak occurred in Italy , and the detection of two autochthonous cases in France in 2010 confirmed fears of the possible expansion of this important viral disease [27]–[28] , particularly as there is currently no commercialized antiviral or vaccine for this virus . Codon re-encoding may represent an important tool for the design of effective vaccines for CHIKV and other arboviruses . However , for such a strategy to succeed it is essential to determine how the virus might respond to this profound change in selection pressure , particularly given its reliance on both mosquitoes and mammals for transmission . Key questions include: to what extent will codon re-encoding reduce viral fitness , how rapidly will CHIKV recover fitness following re-encoding , and will this fitness recovery involve direct reversion at synonymous sites ? To address these questions , we studied the in cellulo replicative fitness of codon re-encoded CHIKV and the in cellulo evolution of re-encoded viruses by combinations of either alternate or continuous passage of each virus in primate or insect cells .
The WT virus and the six re-encoded viruses were derived following transfection of the corresponding infectious DNA clones into Vero cells . Viruses were then passaged once in Vero cells and their replicative fitness was studied . All the viruses produced a cytopathic effect ( CPE ) which was delayed proportionally with the degree of re-encoding ( from 2 days for the WT virus to 5–6 days for the Φnsp1 Φnsp4 Φenv virus ) . After their recovery by transfection , two re-encoded viruses ( Φnsp4 and Φnsp1 Φnsp4 Φenv ) and the WT virus were passaged using three different protocols: 50 serial passages in non-human primate ( Vero ) or mosquito ( C6/36 ) cells , and 50 alternate passages in Vero and C6/36 cells ( i . e . , 25 double passages ( Vero/C6/36 cells ) ) . At the time of each passage , the estimated MOI was bottlenecked at approximately 0 . 1 to minimize the generation of defective interfering particles without generating a major population bottleneck . Each passage was terminated after 48 hours . It is therefore estimated that each virus completed ∼300 replication cycles after fifty passages on the basis of 8 hours per replication cycle [17] .
We have evaluated the effect on replicative fitness and cytopathogenicity of large-scale re-encoding of CHIKV , a re-emerging Old World pathogenic arbovirus . The generation of attenuated viruses by large-scale re-encoding represents an exciting and potentially important route to vaccine development , and also to understanding the basis of the evolution of viral pathogenicity . Site-directed re-encoding , associated with no modification of amino acid sequences , alleviates the likelihood of novel phenotypic properties , allows us to modulate fitness by altering the length of the codon replacement interval , but additionally provides benefits to the generic development of live attenuated vaccines , including reduced costs and single dose induction of long-term immunity [42] . A key result was the observation that our random re-encoding method decreased the replicative fitness of CHIKV in both primate and arthropod cells . The diminution of CHIKV replicative fitness correlated directly with the degree of re-encoding . As reported in previous studies with unrelated viruses , we found that during one replicative cycle in mosquito cells , codon re-encoding profoundly reduced the infectious titre of released virus whilst the number of viral particles remained stable [12] , [14] . This implies that the maturation process ( i . e . the formation of ribonucleoproteins and their insertion into plasma membranes that contain HA ) could be at fault when viruses are re-encoded . In contrast , in primate cells , this decline in infectivity of the viral particles was associated with the reduced generation of viral RNA and proteins probably due to a compromised replication complex . Because they can be identified at different stages of the CHIKV replication cycle , these results imply that the observed decrease in replicative fitness is probably the consequence of several independent re-encoding induced events . However , it is important to note that ( i ) alphaviruses produce very different kinds of infection in mosquito and primate cells ( i . e . persistent infection in mosquito cells and cytolytic infection in vertebrate cells ) , and ( ii ) there are differences in host cell response and innate immunity between mosquito and primate cells , and which could also explain these observed differences in the different cell lines [17] , [43]–[45] . There is mounting evidence that synonymous mutations in viral genomes may have major fitness effects and not only in the small number of cis-acting elements described previously [46] . In the current study , six re-encoded viruses were produced of which the most re-encoded virus modified in three regions that encode different proteins ( together , 882 synonymous mutations were introduced spanning 4 , 212 nt ) . In support of previous studies which demonstrated that re-encoded poliovirus and influenza A viruses are attenuated [10]–[14] , our observations of a reduction in replicative fitness strongly suggest that a proportion of synonymous mutations are not neutral in RNA viruses . Indeed , it is likely that some synonymous mutations were positively selected during the passaging process , reinforcing the idea that synonymous sites are central to viral fitness . In conclusion , it is likely that synonymous mutations can be either neutral , beneficial or deleterious as is the case for non-synonymous mutations . Evolutionary patterns at synonymous sites could be shaped by genome-wide mutational processes , such as G+C% , codon usage bias and dinucleotide frequency [8] , [47]–[49] . These global constraints , which theoretically produce a subset of viable genomes , were assessed by previous studies of codon re-encoding in poliovirus , influenza A virus and bacterial virus T7 which applied specific modification of codon usage bias , codon pair bias or CpG/UpA frequencies [10]–[15] . Using a large-scale random re-encoding method , which only slightly modified these global properties , we still observed replicative fitness reductions in both primate and arthropod cells . Our results suggest that local constraints may also provide significant selection pressure on synonymous sites in RNA viruses , for example by disrupting RNA secondary structures . Since numerous functional secondary structures are present in coding regions of RNA viruses , and hence include synonymous sites ( with notable examples in poliovirus [50] , tick-borne encephalitis virus [51] , alphaviruses [17] , [52] and HIV-1 [53] ) , it is likely that similar structures are common in CHIKV . Recently , it was demonstrated that a similar re-encoding strategy applied to the noncapsid regions of the poliovirus resulted in the identification of two novel functional RNA elements [54] . The concept of large-scale random re-encoding , as described here , is also supported by the report of the negative impact of random single synonymous mutations ( which did not modify the genetic characteristics of the genome ) on viral replicative fitness [46] . Finally , our results and those of previous studies of re-encoded viruses [10]–[14] suggest that the reduction of viral replicative fitness is driven by a variety of factors . First , the nature of the virus studied is an important parameter: we found that introducing up to 882 random synonymous mutations clearly affected the replicative fitness of the CHIKV , whilst two previous studies demonstrated that comparable random re-encoding methods applied to the capsid precursor ( P1 ) region of the poliovirus did not significantly affect replicative fitness ( 934 [14] and 153 [11] synonymous substitutions were introduced , respectively ) . The location of the re-encoded region constitutes the second factor of importance: re-encoding in the E2/E1 region resulted in a greater loss of fitness than in other genomic regions . The analysis of complete wild type CHIKV genomes revealed naturally low levels of synonymous diversity in this re-encoded region ( Figure 10 ) indicating that this region is subject to specific local evolutionary constraints which in part explain the significant impact of re-encoding in this region . The re-encoding method applied is obviously an additional important parameter: previous studies with polioviruses showed that reduction of replicative fitness was strongly dependent on the method used to re-encode the genome [11]–[14] . The average impact of one mutation is clearly likely to be less important in random re-encoding than in specific approaches [11]–[14] . This suggests that random large-scale re-encoding could be advantageous in several aspects when designing future vaccine candidates , namely: ( i ) reversion to wild-type should be intrinsically more difficult , given the high number of mutations produced; ( ii ) since in our experiments the reduction of replicative fitness decreased with the degree of re-encoding , the method opens the door to finely tuning fitness reduction through modulation of the length of re-encoded regions and the number of synonymous mutations introduced; ( iii ) the use of a combination of several re-encoded regions located throughout the viral genome may prevent complete phenotypic reversion due to recombination between WT and re-encoded viruses: large scale sequence modification may render recombination intrinsically more difficult , and in the case of recombination , the part of the genome representing the re-encoded strain would likely still carry some mutations associated with fitness reduction . Taken together , these observations suggest that , following large-scale random re-encoding , recovery of the original replicative fitness should require a large number of reversion mutations . Consequently these re-encoded viruses should be very stable [12] , [14] . To test this hypothesis and to study the constraints that shape CHIKV codon usage , we passaged the wild type and two re-encoded CHIKVs in cellulo . It is commonly stated that arboviruses are subject to strong evolutionary trade-offs , such that mutations that are favoured in one host are deleterious in another , and that this imposes constraints on viral evolution [16] , [37] , [39] . This implies that a strain that has been adapted to a mammalian cell line should have a reduced replicative fitness in mosquito cells , and vice versa . Consequently , we could have initiated our experiments ( i ) with a strain previously adapted to a given cell line , with the disadvantage of introducing an obvious bias in other cell lines ( i . e . , follow a strategy inspired by Bull et al . who used a pre-adapted bacterial virus T7 [12] , [15] ) , or ( ii ) with a clinical strain , i . e . isolated from previous alternate passages in mosquitoes and humans , with the disadvantage of possibly observing a simultaneous response to codon re-encoding and adaptation to culture conditions ( as performed by Burns et al . who used a non-adapted poliovirus [12] , [15] ) . We chose the latter , based on the observation that no criteria existed for defining the adaptation period in the case of arboviruses . This choice was retrospectively justified by the observation that the follow-up of replicative fitness and molecular evolution could not distinguish the criteria necessary for differentiating both phenomena . A key observation of our study was that few reversion mutations occurred , despite specific replicative fitness enhancements in response to codon re-encoding . This suggests that the effectiveness of large-scale re-encoding methods results from the accumulation of slightly deleterious mutations that push the virus into a fitness valley , and that there are multiple opportunities through diverse mutational pathways , sometimes in genomic regions other than those we re-encoded , in which these viruses can partially restore their fitness . For example , amongst the mutation candidates emerging in response to codon re-encoding , two were in the 5′ UTR . That these mutations were fixed so rapidly is strongly suggestive of their selectively beneficial effect [55] . Moreover , it was notable that although we only made modifications to synonymous sites , some of the mutations observed were non-synonymous . Therefore , the evolution of these viruses in response to codon re-encoding was largely compensatory in nature and very few mutations were the result of reversion . Moreover , even with the specific fitness enhancements in response to codon re-encoding observed , the most re-encoded virus failed to reach fitness levels ( infectious titres ) equivalent to those observed at the first passage of the WT virus . Interestingly , these fitness improvements were not always accompanied by greater numbers of mutations or specific molecular modifications . For example , during alternate passage , the most re-encoded virus succeeded in increasing its fitness without accumulating more mutations than other viruses ( i . e . , WT and the less re-encoded viruses ) . During serial passage of the re-encoded viruses , we observed that the response to codon re-encoding and adaptation to culture conditions occurred simultaneously . However , the high levels of observed convergent evolution between the WT virus and the re-encoded viruses indicates that selection arising from codon re-encoding was likely weaker than that for adaptation to culture conditions , and/or that the beneficial mutations to restore the cost of re-encoding were less likely to arise . Therefore , this indirect insight into the difficulty of reversing the effects of re-encoding further highlights the stability of these re-encoded viruses . Our experiments also confirm that mutations acquired in one host can be deleterious in a different host type ( serial passages in primate cells increased viral replicative fitness in primate cells , whilst serial passages in mosquito cells decreased viral fitness in primate cells ) and , with the exception of the most re-encoded virus , that alternate passages seriously ( i ) limit replicative fitness enhancement , and ( ii ) delay the appearance of the mutations . It is noteworthy that replicative fitness in mosquito cells remained globally unchanged following serial passage in mosquito cells , in Vero cells or in both . Moreover , as described previously with dengue virus [38] , replication in mosquito cells appears to act as a brake on viral evolution; in our case , very few mutations were detected after 50 passages and only one was fixed , suggesting that the majority of the emerging mutants have a deleterious effect on viral fitness . In addition , we observed weaker selection pressure in these cells during competition experiments . The fact that the C6/36 cell line was selected more than 30 years ago for its capacity to replicate CHIKVs and dengue viruses [56] could explain this weaker selection pressure observed . Conversely , the rapid adaptation of CHIKV to A . albopictus was accompanied by multiple appearances of the E1-A226V mutation [19] , [22]–[23] and the appearance of amino-acid deletions in the nsP3 and E2 genes . In conclusion , this study demonstrates that random codon re-encoding significantly decreases the replicative fitness of CHIKV . Although all these results are important and encouraging , they cannot be easily extended to RNA viruses producing chronic infections . Thus , studies in animal models are obviously needed to evaluate the potential of these new generation attenuation methods for producing vaccine candidates . However , this approach could assist in the development of future RNA virus vaccines , including those for arboviruses . Introducing a large number of slightly deleterious synonymous mutations reduced the replicative fitness of CHIKV by orders of magnitude in both primate and arthropod cells . This strategy resulted in limited reversion and recovery of fitness after intensive serial subculture of the viruses , and is likely to reduce the risk of complete phenotypic reversion if recombination with wild type virus occurs . Our results encourage us that such modified viruses would find it difficult to return to their natural arboviral cycle in the real world . Furthermore , the decrease of the replicative fitness correlated with the extent of re-encoding , an observation that may be advantageous in the development of future strategies to modulate viral attenuation .
Three regions of the CHIKV genome were re-encoded using a computer program that randomly attributed nucleotide codons based on their corresponding amino acid sequence: for example , the amino acid valine was randomly replaced by GTT , GTC , GTA or GTG . To minimize the influence of rare codons in primate cell lines , the number and the position of such rare codons in primate genomes [57] ( i . e . CGU , CGC , CGA , CGG , UCG , CCG , GCG , ACG ) were not modified . In addition , unique restriction sites were conserved by correcting synonymous mutations at some sites . The location of the re-encoded cassettes , first based on the availability of unique restriction sites was adjusted to avoid overlap with known RNA secondary structures [17] , [52] . Finally , three cassettes of 1302 , 1410 and 1500 bases and located in the nsP1 , nsP4 and E2/E1 regions , respectively , were designed using this method ( Text S2 ) . We modified a previously described IC of the LR2006 strain of CHIKV [58] ( GenBank accession EU224268 ) and all the re-encoded regions were synthesized ( GenScript ) and then inserted into ICs as described in Protocol S1 . Using a combination of re-encoded regions , six re-encoded ICs were generated: Φnsp1 , Φnsp4 and Φenv with one re-encoded region; Φnsp1 Φnsp4 , and Φnsp4 Φenv with two re-encoded regions and Φnsp1 Φnsp4 Φenv with three re-encoded regions ( Figure S1 in Text S1 ) . A fragment of 179 nt located in the nsP2 region ( nucleotide position 2631 to 2809 ) was used to detect the genomic RNA ( plus strand ) of all the CHIKVs ( universal assay ) , re-encoded or not . Another fragment of 168 nt located in the nsP4 region ( nucleotide position 6804 to 6971 ) was used to analyze cell supernatants from competition experiments: two sets of primers and probes allowed us to specifically detect either the viruses re-encoded in the nsP4 region or the viruses without modification in the same region . Primer and probe sequences and the real time PCR protocol are detailed in Table S6 in Text S1 and Protocol S1 , respectively . The replicative fitness of each virus was determined using the results of replication kinetics studies , performed in triplicate in Vero , HEK293 or C6/36 cells . For comparison of the seven viruses from the seven ICs ( the WT virus and the 6 re-encoded viruses ) , one experiment was performed with all the viruses . Virus stock ( see Protocol S1 ) or ICs were used to infect or transfect cells respectively . For the evaluation of replicative fitness of the passaged viruses , we performed one experiment for each virus ( WT , Φnsp4 and Φnsp1 Φnsp4 Φenv viruses ) with the first passage in Vero and the 12th , 25th , 37th and 50th passages for each passage regimen ( 13 supernatants tested in triplicate ) . For the single cycle replication kinetics ( Figure 1 ) , an estimated MOI of 5 was used to infect a 75 cm2 culture flask of confluent Vero , C6/36 or HEK293 cells . Cells were washed twice ( HBSS ) 30 minutes after the infection and 20 ml of medium was added . 1 ml of cell supernatant was sampled just before the washes and at 2 , 8 , 14 , 20 and 28 hours pi . For the replication kinetics with low estimated MOI ( Figure 3 ) and the evaluation of the replicative fitness of the passaged viruses ( Figure S3 , S3 , S5 in Text S1 ) , an estimated MOI of 0 . 01 was used to infect a 25 cm2 culture flask of confluent Vero or C6/36 cells . Cells were washed twice ( HBSS ) 2 hours after infection and 8 ml of medium was added . 1 ml of cell supernatant was sampled after the washes ( T0 ) and at 24 , 48 and 72 hours pi . For the replication kinetics using infectious DNA clones ( Figure 4 ) , a 75 cm2 culture flask of subconfluent HEK293 cells was transfected with the ICs using Lipofectamine 2000 ( Invitrogen ) according to the manufacturer's instructions . Cells were washed twice ( HBSS ) 4 hours after the transfection and 20 ml of medium was added . 1 ml of cell supernatant was sampled after the washes ( T0 ) and at 16 , 24 and 48 hours pi . All the sampled cell supernatants were clarified by centrifugation , aliquoted and stored at −80°C . They were then analysed using a TCID50 assay and a real time RT-PCR assay ( not performed systematically , see figure legends ) . Nucleic acids were extracted from clarified cell supernatants using the EZ1 Virus Mini Kit v2 on the EZ1 Biorobot ( both from Qiagen ) . WT virus was grown in competition with one of four re-encoded viruses ( Φnsp4 , Φnsp1 Φnsp4 , Φnsp1 Φenv or Φnsp1 Φnsp4 Φenv ) using five different PFU ratios ( WT/re-encoded virus 1/99 , 20/80 , 50/50 , 80/20 , 99/1 ) . A global estimated MOI of 0 . 5 was used for the first inoculation . For each experiment , a 25 cm2 flask culture of confluent cells was infected for 2 hours , washed ( HBSS ) and then incubated for 48 h after the addition of 7 ml of medium . Viruses from each experiment were then passaged nine times as follows: a 25 cm2 flask culture of confluent cells was infected for 2 hours with the purified culture supernatant ( centrifugation ) , washed ( HBSS ) and then incubated for 48 h after the addition of 7 ml of medium . At each passage , the estimated MOI was bottlenecked at approximately 0 . 5 . After each infection , nucleic acids were extracted from the clarified culture supernatant using the EZ1 Virus Mini Kit v2 on the EZ1 Biorobot ( both from Qiagen ) . Using two specific real time RT-PCR assays targeting the ΦnsP4 region ( see above ) , the amount of each virus was assessed and the ratio of the two values ( WT/re-encoded ) was calculated . A global estimated MOI of 5 was used to infect confluent 12 well-plates of HEK293 cells with virus stock ( see Protocol S1 ) . Cells were washed once ( HBSS ) 30 minutes after the infection and 2 ml of media was added . At 8 hours pi , the absence of cytopathic effect was checked , culture supernatants were discarded , and cells were washed once ( HBSS ) . All experiments were performed in triplicate . For Western blot analysis and intracellular viral RNA quantification , total RNA and protein isolation was performed using the same well with the Nucleospin RNA/protein kit according to the manufacturer's instructions ( Macherey-Nagel ) . Protein extracts were resolved on 10% polyacrylamide gels containing SDS and transferred to PVDF membrane . Anti-Nsp1/2 rabbit pAb ( see Protocol S1 ) , anti-actin C-2 mAb ( Santa Cruz Biotechnology ) and the corresponding HRP-conjugated secondary antibody were used . Protein bands were revealed using Immobilon ( Millipore ) followed by exposure of blot to radiographic film . Real time RT-PCR assay ( see above ) was performed to assess viral intracellular RNA ( mRNA actin was used as a normalizer to account for differences in cells number and/or quality of extracted RNA as described previously [59] ) . For the quantification of viral proteins by ELISA , cells were mechanically harvested using a cell scraper , resuspended in 800 µL of PBS , vortexed and disrupted by sonication ( 30 seconds at 20 KHz , Misonix Sonicator XL ) . Pre-treated CHIKV-specific immune human serum was used to detect viral proteins . ELISA protocol is detailed in Protocol S1 . The WT and two re-encoded viruses ( Φnsp4 and Φnsp1 Φnsp4 Φenv ) were passaged 50 times following three regimens: serial passages in Vero or C6/36 cells and alternate passages between Vero and C6/36 . For each passage , a 25 cm2 culture flask of confluent cells was infected for 2 hours with the diluted clarified cell supernatant , washed ( HBSS ) and incubated for 48 hours after the addition of 7 ml of medium . Cell supernatant was then harvested , clarified by centrifugation , aliquoted and stored at −80°C . For each passage , the estimated MOI was bottlenecked at approximately 0 . 1 . To avoid contamination , virus passages were performed in three phases: serial passages of WT and Φnsp4 viruses , alternate passages of the same viruses and passages of the Φnsp1 Φnsp4 Φenv virus . All the viruses passaged at the same time were manipulated sequentially and in different laminar flow cabinets . Whole genome nucleotide sequences ( excluding the first 18 nucleotides of the 5′UTR and the 22 nucleotides upstream of the polyA tail ) were determined for all the 50th passage viruses ( nine viruses in total ) . The timing of emergence of each mutation found in the 50th passage was then determined by sequencing with appropriate primer pairs for the 6th , 12th , 18th , 25th , 31st , 37th and 43rd passages . To avoid contamination by PCR products and plasmids , we utilized a molecular biology laboratory that is specifically designed for clinical diagnosis using molecular techniques , and which includes separate laboratories dedicated to perform nucleic acid extraction , PCR/RT mix , RNA/cDNA manipulations and PCR products/plasmids manipulations . In addition , each step from extraction to sequencing or cloning of PCR products ( see below ) was performed in separate experiments for each passaged virus . Nucleic acids were extracted from the purified culture supernatant using the EZ1 Virus Mini Kit v2 on the EZ1 Biorobot ( both from Qiagen ) . A set of specific primer pairs ( Table S5 in Text S1 ) was used to generate amplicons with the Access RT-PCR System ( Promega ) according to the manufacturer's instructions . PCR products were then purified and sequenced using forward and reverse primers with the BigDye Terminator v3 . 1 Cycle Sequencing Kit on an ABI Prism 31310X Genetic Analyser sequencer ( both from Applied Biosystems ) . Analysis of sequencing chromatogram and combination of sequences was performed using the Sequencher 4 . 9 software ( Gene Codes Corporation ) . Three genomic regions were chosen for analysis: one of 619 nt overlapping the nsP2 and nsP3 regions ( positions 3601 to 4220 ) , one of 619 nt overlapping the E3 and E2 regions ( positions 8351 to 8970 ) , and one of 740 nt located in the E1 region ( positions 10140 to 10880 ) . The 1st , 12th , 25th , 37th , 43rd and 50th passages of each passaged virus were chosen to assess the intra-population genetic diversity in the E3/E2 region , while the nsP2/nsP3 and E1 regions were analyzed in the passaged Φnsp1 Φnsp4 Φenv viruses ( using the same passage numbers as above ) . In addition , the 12th , 25th , 37th , 43rd and 50th passages of the Φnsp4 virus serially passaged in C6/36 were also analyzed in the nsP2/nsP3 region to study the emergence of the 4139_4147del mutation . The AccuScript PfuUltra II RT-PCR Kit ( Agilent ) was used according to the manufacturer's instructions to generate amplicons from extracted nucleic acids ( see above ) . PCR products were cloned after purification into the StrataClone PCR Cloning Vector and transformed into competent cells ( StrataClone PCR Cloning Kit; Agilent ) . A plasmid extraction was performed from bacterial colonies with the correct insert which had been previously cultured and plasmid DNA was automatically sequenced with a T7 primer ( GATC Biotech ) . Finally , approximately 20 clones were sequenced for each PCR product analyzed ( mean: 19 . 2+/−4 . 2; range from 14 to 44 ) , resulting in a total of 315 , 933 and 227 sequenced clones for the nsP2/nsP3 , E3/E2 and E1 regions , respectively . Complete viral genome consensus sequences were manually constructed and aligned . Base ambiguity symbols were used to represent all the mixed viral populations when double peaks were observed on the sequencing chromatograms ( as represented ≈ , > and < in Table S4 in Text S1 ) . The 132 CHIKV complete genome nucleotide sequences already available in GenBank ( Text S2 ) , along with the outgroup O'Nyong-Nyong virus ( ONNV ) strain SG650 ( GenBank accession AF079456 ) sequence , were extracted from GenBank/NCBI . The two ORFs of these viruses were manually extracted and concatenated using the Bioedit v7 . 0 . 9 program [60] and aligned with ClustalW [61] according to the amino acid sequence . Ambiguously aligned regions were removed manually . This alignment was used to estimate the variability at each nt position using the Mega5 software [62]–[63] . All the sequences obtained to assess the intra-population genetic diversity were edited using the Sequencher 4 . 9 software ( Gene Codes Corporation ) . Using the Mega5 software [62] , all the sequences from the same region were aligned and each mutation which was found only once in the comparative alignment ( i . e . singletons ) was removed to ensure that no mutations introduced during the RT-PCR were included in the analysis: 22 , 49 and 15 singletons were removed from the nsP2/nsP3 , E3/E2 and E1 alignments , respectively . Minimum spanning trees of clonal CHIKV data were constructed using the TCS1 . 21 software [64] with a 99% connection limit . For each virus and primer pair , we used the alignment of all the sequences ( see above ) after removing singletons . Each deletion ( 6 or 9 nt ) was considered to be a unique event irrespective of length . When a mutation occurred in a deleted region ( detected in another clone ) , it was assigned the position of the corresponding deletion . To study the evolution of replicative fitness during the serial subculture of viruses we performed replicative kinetics studies at the 1st , 12th , 25th , 37th and 50th passage of each virus in Vero and C6/36 cells ( Figure S3 , S3 , S5 in Text S1 ) . Data from these kinetic studies were analyzed using two different methods . We first estimated the relative fitness effect values ( Figure 7 ) : we measured the viral growth rate at 24 hours pi , based on TCID50 values , to calculate the relative fitness effect as described previously [65] . This measure has to be performed during the viral exponential growth phase . Because most of the infectious titre values obtained with the Φnsp1 Φnsp4 Φenv virus at 24 hours pi were close to the detection threshold of our TCID50 assay , we chose the 24 hours pi values for all the viruses , even though some of the values of the WT and Φnsp4 viruses reached a plateau before 24 hours ( Figure S3 , S4 , S5 in Text S1 ) . Titres at time t0 ( all the values were under the detection threshold of our TCID50 assay and then considered as zero ) and at t1 ( 24 hours pi ) were used to calculate the growth rate ( r ) as the increase in log-titre per 24 hours . Relative fitness ( W ) was defined as the growth rate ratio and the relative fitness effect as s = W−1 . The relative fitness for each experiment i was calculated as where is the average of three determinations for the first passage of the virus in the same cells . We then performed a global analysis of TCID50 values at 24 , 48 and 72 hours pi by performing two-way repeated-measures ANOVA [66] and tukey's HDS post-hoc comparisons ( Table S3 in Text S1 ) . This method was used before to analyze similar results [38] . A tukey's HDS post-hoc test was used to compare , once in Vero cells and once in C6/36 cells , the replicative fitness of the first passage for each virus ( WT , Φnsp4 or Φnsp1 Φnsp4 Φenv virus ) with that of the corresponding passaged virus . Comparison of the significance of the results from both methods showed that they gave very similar results ( 67/72 were concordant; Table S3 in Text S1 ) . Infectious titres , viral RNA yields and relative fitness effect values were compared using a Student's t test . For all statistical tests used , all p values below 0 . 05 were considered significant . | Emerging arthropod-borne viruses ( arboviruses ) are a major cause of human and animal morbidity and mortality . Climatic and anthropological activities are responsible for the dispersal of arbovirus transmission vectors into new territories . Chikungunya virus ( CHIKV ) is an important example of a re-emerging pathogen for which no licensed vaccine exists . One of the vectors of CHIKV , the mosquito Aedes albopictus , has dispersed into new temperate regions resulting in outbreaks where they had not been previously observed . Here , we demonstrate that random codon re-encoding , a method that modifies the nucleic acid composition of large coding regions without modifying the encoded proteins , can significantly decrease the replicative fitness of CHIKV . This powerful method of attenuating viruses has several potential advantages for vaccine development , including the possibility to modulate precisely the degree of replicative fitness loss and to generate safe , live-attenuated vaccines that confer long-term protection , in a cost effective manner . Our studies also demonstrate that these re-encoded viruses exhibit a stable phenotype , and that the response to codon re-encoding was largely compensatory in nature , with little reversion of mutations . Finally , we provide further evidence that many synonymous sites in RNA viruses are not neutral and clearly impact viral fitness . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"infectious",
"diseases",
"viral",
"vaccines",
"emerging",
"infectious",
"diseases",
"virology",
"neglected",
"tropical",
"diseases",
"biology",
"microbiology",
"arboviral",
"infections",
"infectious",
"disease",
"control",
"viral",
"evolution"
] | 2013 | Random Codon Re-encoding Induces Stable Reduction of Replicative Fitness of Chikungunya Virus in Primate and Mosquito Cells |
Beckwith-Wiedemann syndrome ( BWS ) is a fetal overgrowth and human imprinting disorder resulting from the deregulation of a number of genes , including IGF2 and CDKN1C , in the imprinted gene cluster on chromosome 11p15 . 5 . Most cases are sporadic and result from epimutations at either of the two 11p15 . 5 imprinting centres ( IC1 and IC2 ) . However , rare familial cases may be associated with germline 11p15 . 5 deletions causing abnormal imprinting in cis . We report a family with BWS and an IC2 epimutation in which affected siblings had inherited different parental 11p15 . 5 alleles excluding an in cis mechanism . Using a positional-candidate gene approach , we found that the mother was homozygous for a frameshift mutation in exon 6 of NLRP2 . While germline mutations in NLRP7 have previously been associated with familial hydatidiform mole , this is the first description of NLRP2 mutation in human disease and the first report of a trans mechanism for disordered imprinting in BWS . These observations are consistent with the hypothesis that NLRP2 has a previously unrecognised role in establishing or maintaining genomic imprinting in humans .
Genomic imprinting is an epigenetic modification that causes genes to be expressed according to their parent of origin . Although less than 100 imprinted genes have been identified in human and mice , many imprinted genes appear to have a critical role in prenatal growth and development [1] . Molecular genetic analysis of rare human imprinting disorders has played a critical role in elucidating the mechanisms of genomic imprinting . In particular , studies of the imprinting disorder Beckwith-Wiedemann syndrome ( BWS MIM 130650 ) have provided important insights into the structure and function of imprinting centres [2] . BWS is a congenital overgrowth syndrome , characterised by prenatal and postnatal overgrowth , macroglossia and anterior abdominal wall defects . Additionally , variable features include organomegaly , neonatal hypoglycaemia , hemihypertrophy , urogenital abnormalities and in about 5% of children , embryonal tumours ( most frequently Wilms' tumour ) . The genetics of BWS are complex , but involve mutation or altered expression of several closely linked genes associated with cell cycle and growth control in the imprinted 11p15 . 5 chromosomal region . Imprinted genes most frequently implicated in the aetiology of BWS include the paternally expressed IGF2 , KCNQ1OT1 ( LIT1 ) genes and the maternally expressed H19 and CDKN1C ( P57KIP2 ) genes . KCNQ1OT1 and H19 transcripts are not translated but the IGF2 gene product is an important prenatal growth factor and the CDKN1C protein is a candidate tumour suppressor that negatively regulates the cell cycle [3] . The majority of BWS cases are sporadic and result from epimutations of the distal ( IC1 ) or proximal ( IC2 ) 11p15 . 5 imprinting centres ( see [4] and references within ) . IC1 is a differentially methylated region ( DMR ) about 5kb upstream of H19 that has an “insulator function” regulated by the zinc finger transcription factor , CTCF . The insulator is methylation sensitive , such that when CTCF binds to the unmethylated maternal allele , the IGF2 promoters do not have access to ( are insulated from ) enhancers downstream of H19 . Methylation on the paternal allele prevents CTCF from binding , thus permitting interaction between the IGF2 promoters and the enhancers [5] . About 5–10% of sporadic BWS cases have hypermethylation of the H19 DMR and in these cases IGF2 shows loss of imprinting ( LOI ) and biallelic expression [6] . The second imprinting centre , IC2 is a DMR located in intron 10 of the KCNQ1 gene and is known as KvDMR1 . The unmethylated paternal allele permits transcription of the antisense transcript KCNQ1OT1 ( also known as LIT1 ) and silencing of genes including KCNQ1 and CDKN1C . Maternal methylation at the KvDMR1 is thought to prevent transcription of the KCNQ1OT1 gene and enable expression of CDKN1C . Loss of methylation ( LOM ) at the KvDMR1 is seen in up to 50% of sporadic BWS and is associated with biallelic expression ( loss of imprinting ) of KCNQ1OT1 and silencing of maternal CDKN1C expression [7]–[9] . Apparent hypomethylation of IC2 may , in rare cases , result from a germline IC2 deletion [10] . However , most BWS patients with IC2 methylation defects appear to have an epimutation of unknown cause ( although there is an increased risk of BWS with IC2 epimutation in children conceived by assisted reproductive technologies ) [11]–[13] . In order to gain insights into the factors responsible for IC2 imprinting defects , we studied a family with BWS that displayed evidence of IC2 epimutations through a trans mechanism .
This study was conducted according to the principles expressed in the Declaration of Helsinki . The study was approved by the South Birmingham Research Ethics Committee ( equivalent to the Institutional Review Board of Birmingham Women's Hospital ) reference number CA/5175 . All patients provided written informed consent for the collection of samples and subsequent analysis . A consanguineous family of Pakistani origin with two affected children with BWS due to loss of methylation at KvDMR1 were investigated in the first instance . Following the identification of NLRP2 mutations in this family , a further 11 BWS families , each with a single case of BWS ( mean age 10 . 8 years ) with KvDMR1 loss of methylation were analysed for NLRP2 mutations . This cohort included 10 patients who also had loss of methylation at other imprinted loci . Ethnically matched laboratory control samples were analysed to evaluate the significance of novel sequence variants . Genomic DNA was extracted from peripheral lymphocytes by standard techniques . In preliminary examinations chromosomal abnormalities were excluded and the methylation status of IC1 and IC2 of imprinting region 11p15 . 5 were determined . Methylation analysis of KvDMR1 was performed as described previously , with PCR amplification of bisulphite modified DNA and digestion with restriction enzyme BstU1 yielding different sized fragments which is separated using ABI377 or 3730 [4] . In addition , methylation status at 3 additional DMRs was evaluated at the Transient Neonatal Diabetes Mellitus ( TND ) locus at 6q24 , 7q32 ( PEG1 ) and the Angelman/Prader-Willi locus at 15q13 ( SNRPN ) as described previously [14] . Primers and methods for the analysis of the methylation status of PEG1 DMR by methylation specific PCR ( MS-PCR ) were obtained from previously published report by Mackay et al [15] . For linkage studies a genome wide linkage scan was undertaken using the Affymetrix 250k SNP microarray . Mutation analysis of JMJD2D , ZFP57 , NLRP7 and NLRP2 was carried out by direct sequencing . The genomic DNA sequence of these genes was taken from Ensembl ( http://www . ensembl . org/index . html ) and primer pairs for the translated exons were designed using primer3 software ( http://fokker . wi . mit . edu/primer3/input . htm ) . Amplification was performed according to standard protocols with Bio Mix Red provided by Bioline . PCR products were directly sequenced by the Big Dye Terminator Cycle Sequencing System with the use of an ABI PRISM 3730 DNA Analyzer ( Applied Biosystem ) . DNA sequences were analyzed using Chromas software .
A family with complex consanguinity ( Figure 1 ) was ascertained after the diagnosis of two children with BWS . Both pregnancies were complicated by polyhydramnios and raised hHCG levels . Both affected children were born by Caesarean section . At birth Child 1 ( V-1 in Figure 1 ) was macrosomic ( 4 . 55 kg at 38 weeks gestation ) and was noted to have macroglossia , an omphalocele , ear creases , right inguinal hernia , undescended testis and neonatal hypoglycaemia in the first two days after birth . Similarly Child 2 ( V-2 ) was macrosomic ( 3 . 633 kg at 35 weeks gestation ) with macroglossia , ear lobe creases and neonatal hypoglycaemia that was difficult to control . Subsequently a third child was born without features of BWS ( between the second and third children a probable hydatidiform mole was diagnosed ) . Molecular studies demonstrated loss of maternal allele KvDMR1 ( IC2 ) methylation in both affected children , but the unaffected sibling had normal methylation . H19 methylation status was normal in both affected children and MLPA analysis demonstrated no evidence of an IC1 or IC2 deletion . Genotyping revealed no evidence of paternal uniparental disomy and linkage analysis with microsatellite markers flanking IC2 ( TH and D11S4088 ) demonstrated that the two children had inherited opposite maternal and paternal 11p15 . 5 alleles . These findings were consistent with IC2 epimutation resulting from a trans imprinting defect . In view of the history of consanguinity , an autosomal recessive disorder was suspected ( either affecting both children or affecting the mother ) . Genetic linkage studies were undertaken by genotyping the two children and both parents on an Affymetrix 250k SNP array platform . Five regions of homozygosity ( >2Mbases ) were shared by the two children but these did not contain a gene known to be implicated in the establishment or maintenance of genomic imprinting . However inspection of the maternal genotypes revealed an ∼8 Mbase homozygous region containing NLRP2 and NLRP7 at 19q13 . 4 . NLRP7 is a homologue of the mouse NLRP2 gene ( NLRP7 is not present in the mouse ) and the human NLRP2 gene . Sequencing of NLRP2 in the mother identified a homozygous frameshift mutation ( ( c . 1479delAG , NM_017852; Figure 2 ) that was predicted ( in the absence of nonsense-mediated RNA decay ) to result in a truncated protein ( p . Arg493SerfsX32 ) lacking 539 amino acids from the C-terminal that includes the LRR domain . The mutation was not detected in 542 ethnically matched control chromosomes but the father was heterozygous for the mutation , child 1 was homozygous for the mutation and the other two children were heterozygous . Mutation analysis of 11 additional families with BWS did not reveal any evidence of pathogenic NLRP2 mutations . To determine if the trans imprinting defect extended beyond KvDMR1 , we analysed methylation levels at the TND ( 6q24 ) , SNRPN ( 15q13 ) and PEG1 ( 7q32 ) DMRs . Both affected siblings ( and all controls ) had normal methylation levels at the TND and SNRPN DMRs but Child 2 demonstrated partial loss of methylation at the PEG1 DMR ( Figure 3 ) .
We identified a homozygous frameshift mutation in the mother of two children with BWS caused by epimutations at IC2 ( KvDMR1 ) . Most cases of BWS due to loss of methylation of KvDMR1 are sporadic , but a handful of familial cases have been described with maternally inherited germline IC2 deletions [10] . However , in our family there was no evidence of a germline deletion by MLPA analysis [16] and the two affected children were shown to have inherited opposite maternal and paternal 11p15 . 5 alleles . This suggested that in this family , KvDMR1 LOM resulted from a trans and not a in cis effect . Germline NLRP2 mutations have not been reported previously , but mutations in ZFP57 and NLRP7 can cause imprinting disorders in which epimutations at imprinted loci result from a trans effect . Thus individuals homozygous for ZFP57 mutations presented with transient neonatal diabetes mellitus ( TNDM ) . The major cause of TNDM is aberrant expression of imprinted genes at chromosome 6q24 and about 20% of cases have LOM at the TND differentially methylated region ( DMR ) . Patients with homozygous ZFP57 mutations have LOM at the TND DMR , but also at other imprinted loci including KvDMR1 [15] . However , there was no evidence of germline ZFP57 mutations in our family . Germline NLRP7 mutations are associated with familial recurrent biparental complete hydatidiform mole ( FHM ) in which there is epigenetic abnormalities at DMRs in multiple imprinting regions [17]–[19] . Although FHM associated with NLRP7 mutations is inherited in an autosomal recessive manner , in contrast to ZFP57 mutation , homozygotes have normal genomic methylation but in female homozygotes there is a failure to establish methylation imprints in their germ cells leading to hydatidiform moles and reproductive wastage ( male homozygotes do not have imprinting defects in their sperm ) . The methylation defects in FHM are specific for imprinted loci , and DNA methylation at non-imprinted genes and genes subject to X-inactivation is unaffected [18] . The human NLRP2 and NLRP7 genes are highly homologous and the two proteins consist of 1 , 062 and 1 , 009 amino acids respectively and have identical structure and about 64% amino acid identity . Only one of the two affected children with BWS was homozygous for a NLRP2 mutation and , by analogy with FHM caused by NLRP7 mutations , the BWS phenotype most likely results from homozygosity in the mother , such that familial BWS associated with NLRP2 mutations is inherited in a similar manner to NLRP7 associated FHM and not in a conventional autosomal recessive manner . Both FHM and ZFP57–TNDM are associated with imprinting aberrations at multiple loci , and we identified partial loss of methylation at the PEG1 DMR in one of the affected children . Nevertheless it seems that NLRP2 mutations have a less severe effect on imprinting than NLRP7 or ZFP57 inactivation . A subset of children with BWS and an IC2 epimutation display hypomethylation at multiple imprinting centres ( DMRs ) ( in our series these children are more likely to have been conceived by assisted reproductive technologies ) [14] , [20] . To our knowledge , none of these cases have been familial and we did not identify NLRP2 mutation in the sporadic cases we studied . It appears that the establishment ( or maintenance ) of methylation at KvDMR1 is particularly sensitive to genetic and/or environmental insults . We note that one of the affected children demonstrated a partial loss of methylation at PEG1 DMR1 both by bisulphite sequencing and MS-PCR suggesting that NLRP2 mutations may be associated with an incomplete failure of imprinting establishment and/or a partial failure of maintenance methylation at this DMR . Interestingly , investigation of a mouse knockout of ZFP57 has suggested a role in both the establishment of germline methylation imprints and in the postfertilisation maintenance of methylation imprints [21] . NLRP2 and NLRP7 encode members of the NLRP ( Nucleotide-binding oligomerization domain , Leucine rich Repeat and Pyrin domain ) family of CATERPILLER proteins . NLRP family of cytoplasmic proteins comprises 14 members of similar structure that are principally encoded by two gene clusters on chromosome 11p15 ( NLRP6 , 10 and 14 ) and 19q13 . 4 ( NLRP2 , 4 , 5 , 7 , 8 , 9 , 11 , 12 and 13 ) . Most of the family members are well conserved from C . elegans , D . melanogaster , rat , and mouse to human but there is no rodent homologue for NLRP7 and the gene is found in only a few genomes ( human , primate and cow ) . Some NLRP proteins are components of the inflammasome that is implicated in the sensing of , and inflammatory reaction to , extracellular pathogens and intracellular noxious compounds [22] . Germline mutations in NLRP3 and NLRP12 are associated with familial cold autoinflammatory syndrome [23] , [24] . NLRP2 was suggested to function as a modulator of macrophage NFKB activation and procaspase 1 [25] , however we found that the two family member homozygous for a NLRP2 truncating mutation did not show any evidence of an immune or autoinflammatory disorder . Nevertheless most NLRP family proteins are widely expressed and not restricted to the immune system . In addition , many are expressed in human oocytes and embryos at an early stage of development . Thus Zhang et al . have reported that NLRP4 , 5 , 8 , 9 , 11 , 12 , 13 , and 14 were highly expressed in oocytes and then gradually decreased in embryos with a very low level in day 5 embryos , whilst NLRP2 and NLRP7 progressively decreased from oocytes to day 3 embryos then showed a sharp increase on day 5 [26] . These observations are consistent with NLRP2 and NLRP7 having a similar role in early development/imprinting establishment . Although it has been suggested that FHM might result from an immune-related defect in oogenesis or early embryo development ( with methylation changes being a secondary phenomenon ) the specific association of the methylation defects with imprinted DMRs suggests a more direct role in the establishment or maintenance of imprinting marks . Such a view is supported by the identification of germline NLRP2 mutations in BWS and should prompt further investigation of the role of NLRP2 and NLRP7 in genomic imprinting . The apparent involvement of NLRP proteins in genome methylation and the sensing and inflammatory response to extracellular pathogens and intracellular noxious compounds is intriguing given the suggestion that cytosine methylation may have evolved as a host response to transposons [27] . It is interesting that the third child in the family we report was unaffected . Most mothers with NLRP7 mutations have recurrent molar pregnancies only , but at least three families have been reported in which affected women had liveborn offspring ( see [28] and references within ) . In one family , three affected members had , in addition to the molar phenotype , several miscarriages and three term pregnancies [29] . In one of the term pregnancies the baby was born with severe intrauterine growth retardation , but grew into a healthy adult with normal methylation levels [30] . In another term pregnancy the baby was born with unilateral cleft lip and palate and later manifested idiopathic delayed mental and motor development [31] . Murdoch et al . detected a homozygous splice site mutation in NLRP7 in all three affected women of this family [32] . Thus homozygous NLRP7 mutations may be associated with clinical heterogeneity/incomplete penetrance , possibly resulting from genetic modifier or environmental effects . In the light of these observations and the apparently milder phenotypic effects of maternal NLRP2 inactivation than NLRP7 inactivation ( Beckwith-Wiedemann syndrome and molar pregnancy respectively ) it might be predicted that clinical heterogeneity/incomplete penetrance would be a feature of maternal NLRP2 inactivation . Although maternal NLRP2 mutations appear to be a rare cause of familial BWS , the identification of these cases is important , as the inheritance pattern differs from the autosomal dominant inheritance ( with parent of origin effects ) associated with other inherited forms of BWS . The inheritance of NLRP2-associated BWS has similarities to other autosomal recessive disorders in which homozygous mothers are well , but there is a high risk to their offspring ( e . g . FHM and treated maternal phenylketonuria ) . | A small set of genes ( imprinted genes ) are expressed in a “parent-of-origin” manner , a phenomenon known as genomic imprinting . Research in human disorders associated with aberrant genomic imprinting provided insights into the molecular mechanisms of genomic imprinting and the role of imprinted genes in normal growth and development . Beckwith-Wiedemann syndrome ( BWS ) is a congenital overgrowth syndrome associated with developmental abnormalities and a predisposition to embryonic tumours . BWS results from alterations in expression or function of imprinted genes in the imprinted gene cluster at chromosome 11p15 . Although BWS may be caused by a variety of molecular mechanisms , to date , all the genetic and epigenetic defects associated with BWS have been limited to 11p15 . 5 . We report a family with two children affected with BWS and an epigenetic defect at 11p15 . 5 in which the primary genetic defect mapped outside the imprinted gene cluster . Using autozygosity mapping , we found an extended homozygous region on chromosome 19q13 . 4 ( containing NLRP2 and NLRP7 genes ) in the mother . Homozygous inactivating mutations in NLRP7 in women have been associated previously with abnormal imprinting and recurrent hydatidiform moles . We identified a homozygous frameshift mutation in NLRP2 in the mother of the two children with BWS implicating NLRP2 in the establishment and/or maintenance of genomic imprinting/methylation . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"genetics",
"and",
"genomics/epigenetics",
"genetics",
"and",
"genomics/gene",
"discovery",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"genetics",
"and",
"genomics/medical",
"genetics"
] | 2009 | Germline Mutation in NLRP2 (NALP2) in a Familial Imprinting Disorder (Beckwith-Wiedemann Syndrome) |
Legionella pneumophila is an intracellular pathogen responsible for Legionnaires' disease . This bacterium uses the Dot/Icm type IV secretion system to inject a large number of bacterial proteins into host cells to facilitate the biogenesis of a phagosome permissive for its intracellular growth . Like many highly adapted intravacuolar pathogens , L . pneumophila is able to maintain a neutral pH in the lumen of its phagosome , particularly in the early phase of infection . However , in all cases , the molecular mechanisms underlying this observation remain unknown . In this report , we describe the identification and characterization of a Legionella protein termed SidK that specifically targets host v-ATPase , the multi-subunit machinery primarily responsible for organelle acidification in eukaryotic cells . Our results indicate that after being injected into infected cells by the Dot/Icm secretion system , SidK interacts with VatA , a key component of the proton pump . Such binding leads to the inhibition of ATP hydrolysis and proton translocation . When delivered into macrophages , SidK inhibits vacuole acidification and impairs the ability of the cells to digest non-pathogenic E . coli . We also show that a domain located in the N-terminal portion of SidK is responsible for its interactions with VatA . Furthermore , expression of sidK is highly induced when bacteria begin to enter new growth cycle , correlating well with the potential temporal requirement of its activity during infection . Our results indicate that direct targeting of v-ATPase by secreted proteins constitutes a virulence strategy for L . pneumophila , a vacuolar pathogen of macrophages and amoebae .
The delivery of newly formed phagosomes to the lysosomal system by the endocytic pathway is essential for the digestion of phagocytosed materials . To evade such destruction , successful intracellular bacterial pathogens have evolved various mechanisms , including inhibition of phagolysosomal fusion , resistance to lysosomal digestion or the escape to the host cell cytosol . For intravacuolar pathogens , active modification of lipid and protein composition of phagosomal membrane is critical for their survival and replication . Moreover , since lysosomal enzymes often are active only in an acidic environment , regulation of pH in the phagosomal lumen is one common strategy employed by pathogens to avoid lysosomal killing [1] , [2] . Legionella pneumophila is a facultative intracellular pathogen responsible for Legionnaires' disease . Upon being phagocytosed , this bacterium orchestrates various cellular processes to initiate a unique trafficking pathway that eventually leads to the formation of a phagosome permissive for its multiplication [3] . The biogenesis and maintenance of the bacterial replicative vacuole is mediated by protein substrates of the Dot/Icm type IV secretion system [4] , [5] . For example , RalF activates and recruits the small GTPase Arf1 to the bacterial vacuole [6] . Similarly , another small GTPase Rab1 is recruited to the bacterial vacuole by SidM/DrrA , which with LepB [7] , completely hijacks the activity of this important regulatory molecule in membrane trafficking [8] , [9] . Whereas SidM/DrrA functions to release Rab1 from its GDI and activates the protein by loading it with GTP , LepB promotes the GTPase activity [10] . These proteins , along with other effectors such as SidJ that is involved in the recruitment of endoplasmic reticulum ( ER ) proteins to the bacterial vacuole [11] , are thought to be responsible for the transformation of the nascent phagosome into a vacuole derived from the ER that resembles an immature autophagosome [12] , [13] , [14] . L . pneumophila also actively modulates cell death pathways of infected macrophages , presumably to ensure the well being of the host cell for a complete infection cycle . Inhibition of cell death is mediated through the activation of an NF-κB-dependent induction of antiapoptotic genes and by effectors such as SidF that directly antagonize proapoptotic BNIP3 and Bcl-rambo [15] , [16] , [17] , and SdhA , an effector of unknown mechanism of action [18] . Effectors that modulate other cellular processes , including protein synthesis , ubiquitination and lipid metabolism have also been identified , but how the bacterium benefits from the functions of these virulence factors is less clear [19] , [20] , [21] , [22] . Finally , a recent study indicated that the effector AnkB contributes significantly to bacterial intracellular growth but does not affect any of the above host cellular processes , suggesting the targeting of yet unidentified host pathways by L . pneumophila [23] . The yeast Saccharomyces cerevisiae has been widely used to study bacterial effectors , largely due to its genetic manipulability and the conservation of many eukaryotic cellular processes [24] . A large number of L . pneumophila effectors have been identified by their ability to kill yeast cells [20] , [25] or to interfere with its vesicle trafficking processes [26] , [27] . In eukaryotic cells , the pH of intracellular compartments is an intricately regulated parameter that is crucial for many biological processes , including membrane trafficking , protein degradation and coupled transport of small molecules [28] . Organellar acidification primarily is mediated by ATP-dependent proton transporters known as the vacuolar H+-ATPases or v-ATPases , which is a large multisubunit complex with an approximate molecular mass of 103 kDa [28] . The structure of v-ATPases can be divided into two major functional domains: a 570-kDa peripheral subcomplex , known as V1 , that binds and hydrolyzes ATP , and an integral membrane subcomplex , termed V0 , that serves as the pore through which protons traverse the membrane bilayer [28] . L . pneumophila is able to maintain a neutral luminal pH during infection , particularly within the first 6 hrs after uptake [29] , [30] , a hallmark shared by many intravacuolar pathogens . One such example is Mycobacterium avium whose vacuoles fail to acidify below pH 6 . 3 , probably by selectively inhibiting fusion with v-ATPase-containing vesicles or by rapidly removing the complex from its phagosomes [31] . Interestingly , a recent organelle proteomic study reveals that , in the soil amoebae host Dictyostelium discoideum , v-ATPase is associated with the Legionella containing vacuole ( LCV ) even in early phase of infection [32] . This finding is contradictory to the well-established notion that in macrophages the bacterial phagosome maintains a neutral luminal pH for several hours , suggesting that the pathogen may initially antagonize the activity of the proton transporter . Proton transporters from different orders of eukaryotes are highly conserved in structure and function , and some genes of mammalian or plant v-ATPase components can complement the corresponding yeast mutants [33] , [34] . However , whereas in mammals mutations eliminating subunits of v-ATPase generate various phenotypes , ranging from the absence of any severe phenotype to lethality to embryonic development [35] , [36] , yeast v-ATPase mutants are viable but only in acidic medium [37] . In this study , we have taken the advantage of this conditional phenotype of yeast vma mutants to identify L . pneumophila proteins that may target the host v-ATPases . Here we report one such protein that inhibits v-ATPase by directly interacting with one component of the proton transporter .
One of the prominent phenotypes associated with yeast v-ATPase mutants is their inability to grow in neutral pH medium [37] . We reasoned that if L . pneumophila codes for proteins that inhibit v-ATPase activity , expression of such proteins in a yeast strain would impair its ability to grow in neutral pH medium . To this end , we cloned individual L . pneumophila hypothetical genes into pGBKT7 ( Clontech ) [20] . Yeast strains harboring each of the plasmids were tested for their ability to grow in medium with a pH of 7 . 5 . Of the first 97 genes screened ( Table S1 ) , one gene that consistently interferes with yeast growth under this condition was obtained ( Fig . 1A and B ) . This gene ( lpg0968 ) , designated SidK is predicted to code for a protein of approximately 65 kDa . It is present in the genomes of all sequenced strains of L . pneumophila but has no detectable homology to proteins in the database , nor does it contain predictable domains or motifs suggestive of known biochemical activities . Interestingly , this gene is divergently transcribed from lpg0969 , a gene that appears to inhibit yeast growth by interfering with unknown host functions [27] . Since the original construct was made by fusing the testing gene to the DNA binding domain of the Gal4 protein on pGBKT7 ( Clontech ) , we attempted to eliminate the potential effect of protein fusion by expressing untagged sidK in yeast strain BY4741 [38] . To this end , we used a series of vectors that differ in copy number and promoter strength ( Table S3 and ref . [39] ) . The expression of sidK on these vectors is proportional to the strength of the promoter , with the GPD ( glyceraldehyde-3-phosphate dehydrogenase ) promoter giving the highest protein level ( Fig . 1C ) . Consistent with the protein levels , in 18 hrs after the establishment of subcultures of identical cell density , strains in which SidK was expressed from the GPD promoter almost completely lost the ability to grow in neutral pH medium ( Fig . 1D ) . On the other hand , only a marginal growth defect was observed when the gene was expressed from the moderate ADH promoter ( Fig . 1C and D , strain 1 ) , indicating that the effect of SidK on yeast growth under this condition is dose-dependent . Taken together , these data indicate that we have identified a L . pneumophila gene that affects yeast growth in neutral pH medium , probably by interfering with its v-ATPase activity directly or with activities relevant to the proton transporter . To exert an effect on its cellular targets , a bacterial virulence factor must first reach the host cytosol via specialized secretion systems . We thus examined whether SidK is a substrate of the Dot/Icm secretion system . We first employed the Cya assay [40] by fusing SidK to the carboxyl end of the catalytic domain of the Bordetella pertusis cyclic AMP synthetase . Infection of macrophages with a L . pneumophila strain expressing Cya fused to the known effector SidJ led to production of high-level cAMP in a Dot/Icm-dependent manner ( Fig . 2A ) . Importantly , although the Cya-SidK fusion expressed similarly in the wild type and the dotA mutant , high levels of cAMP were only detected in infections using the wild type strain ( Fig . 2A ) , indicating that SidK contains signals recognizable by the Dot/Icm system . Similar results were obtained with the SidC staining assay [41] , in which fusion to SidK restores the translocation of the transfer deficient SidCΔC100 mutant to wild type levels ( Fig . 2B–D ) . These results indicate that SidK is a substrate of the Dot/Icm system . To determine whether SidK is injected into host cells by L . pneumophila during infection , we attempted to detect SidK in lysates of infected cells generated by saponin fractionation [41] . Despite considerable effort , we were unable to detect this protein in the soluble fraction of lysates of cells infected by L . pneumophila for up to 3 hrs ( data not shown ) . Considering the possibility that the amount of translocated SidK is beyond detection by this method; we used a SidK specific antibody to enrich the protein . After immunoprecipitation , SidK protein was detected in lysates of cells infected with wild-type strain but not with the Dot/Icm deficient mutant or a sidK deletion mutant ( Fig . 2E , lanes 2–3 ) . Expression of SidK from a plasmid in the sidK deletion mutant restored the delivery of this protein into infected cells ( Fig . 2E , lane 4 ) . Collectively , these results indicate that SidK is a substrate of the Dot/Icm system and is injected into infected cells by L . pneumophila during infection . Furthermore , we cannot readily detect SidK in concentrated lysates of infected cells ( data not shown ) , suggesting that the amount of translocated protein is low . Next , we examined the potential role of sidK in L . pneumophila infection by constructing an in-frame deletion mutant and tested its intracellular growth . The mutant did not exhibit detectable growth defect in either mouse bone marrow-derived macrophages or D . discoideum ( Fig . S1 ) . These observations extend the list of L . pneumophila effectors not essential for its intracellular growth in standard infection models , thus adding another layer to the remarkable potential functional redundancy among substrates of the Dot/Icm system [5] . Since wild type L . pneumophila maintains a neutral pH in its vacuole and SidK appears to interfere with the functions of v-ATPase , we analyzed whether deletion of sidK affects luminal pH of LCVs in mouse macrophages . Relevant L . pneumophila strains labeled with 5 ( 6 ) -carboxyfluorescein-N-hydroxysuccinamide ester ( FLUOS , Fluka ) were used to infect macrophages and images obtained from individual phagosomes were used to calculate intravacuolar pH against a standard curve as described [30] . As expected , vacuoles containing heat-killed bacteria quickly acidified to pH values of about 4 , whereas phagosomes harboring wild type L . pneumophila maintain a neutral pH at the time points examined ( Fig . S2-A ) . Furthermore , although the dotA mutant was not lysed by the macrophages in the experimental duration ( Fig . S2-B ) , its vacuoles were also acidified , indicating that the Dot/Icm system is required for the biogenesis of a bacterial phagosome of neutral pH . Interestingly , vacuoles containing the sidK deletion mutant still are able to block their acidification , thus maintaining a neutral luminal pH in the experimental duration ( Fig . S2-A ) . Given the proficient intracellular growth displayed by the mutant ( Fig . S1 ) , this result was not unexpected . L . pneumophila mutants lacking a single effector gene rarely exhibit detectable intracellular growth defect , possibly due to functional redundancy among bacterial and/or host factors [5] , [42] . Consistent with the observation that L . pneumophila grown at post-exponential phase are more infectious , many substrates of the Dot/Icm system are highly induced when bacterial cultures enter this growth phase [5] . That the luminal pH of LCVs is neutral in the early phase of infection points to the requirement of bacterial factors that target v-ATPase in this period of infection . We thus examined the protein level of SidK at different time points throughout the L . pneumophila growth cycle in broth . In contrast to many substrates of the Dot/Icm transporter whose expression is induced at post-exponential phase , very little SidK was present in L . pneumophila grown at this stage ( Fig . 3 ) . Rather , accumulation of SidK was apparent within 1 hr after diluting saturated cultures into fresh medium , and reached the peak 2 hrs after dilution ( Fig . 3B ) . When the bacterium begins to replicate ( approximately 4–5 hrs after dilution ) , protein level of SidK begins to decrease and became difficult to detect throughout the rest of the growth cycle ( Fig . 3A–B ) , a pattern consistent with the slow progression of LCVs to lysosomal organelles [30] . We also determined the kinetics of SidK translocation during infection by saponin fractionation . Translocated SidK was not detectable until 3 hrs after infection and the protein is present in the soluble fraction of infected cells in the first 12 hrs of infection ( Fig . 3C ) . Since mutations affecting various yeast genes not directly involved in v-ATPase function can result in mutants sensitive to neutral pH medium [43] , we furthered our study on the mechanism of action of SidK by identifying its cellular targets with the unbiased affinity chromatograph method . We incubated Affigel beads ( Bio-Rad ) coated with purified SidK ( Fig . S3 ) with PBS soluble fraction of U937 cell lysates . After removing unbound proteins by washing with PBS , proteins retained on the beads were separated by SDS-PAGE and were visualized by silver staining . At least three proteins with molecular weights ranging between 20 kDa and 75 kDa were retained by beads coated with SidK , but not by uncoated beads ( Fig . 4A , lane 3 ) . By mass spectrometry analysis , we identified these proteins as three components of the mammalian vacuolar H+-ATPase: VatA , the ubiquitous VatB2 subunit and VatE ( Fig . 4A , lane 3 ) . These results indicate v-ATPase is the potential target of SidK . To further investigate the interactions between v-ATPase and SidK , we transfected mammalian cells with combinations of plasmids that direct the expression of GFP-SidK or Flag-VatA , one essential component of the proton transporter [28] . Lysates of transfected cells were subjected to co-immunoprecipitation ( co-IP ) with an anti-Flag antibody to detect possible SidK/VatA complexes . GFP-SidK was detected in immunoprecipitates only from cells coexpressing Flag-VatA ( Fig . 4B ) . No signals were detected in untransfected samples or in samples transfected with plasmids expressing only Flag-VatA or GFP-SidK , indicating that the interactions were specific . Similar results were obtained in reciprocal experiments using the anti-GFP antibody ( Fig . 4C ) . Furthermore , VatA was detected in precipitates from cells that was transfected only with the plasmid expressing SidK , indicating that this protein interacts with endogenous VatA ( Fig . 4C , lane 3 ) . The structure and function of v-ATPase from mammals and yeast are highly similar [28] . We thus examined whether SidK interacts with yeast v-ATPase . When lysates of yeast cells expressing GFP or GFP-SidK were immunoprecipitated with a GFP-specific antibody , Vma1 ( equivalent of VatA ) was detected in precipitates , again only in samples expressing SidK ( Fig . 4D , lane 1 in left panel ) . Similar results were obtained in reciprocal immumoprecipitation using a Vma1 specific antibody ( data not shown ) . Taken together , these data establish that v-ATPase is the cellular target of SidK . Under normal physiological condition , components of the V1 domain of v-ATPases form a stable complex [28] . In agreement with this notion , in immunoprecipitation experiments aiming at detecting interactions between SidK and components of the V1 complex , positive interactions were observed in many if not all components ( data not shown ) . Thus , we further investigated which subunit of the V1 domain is directly targeted by SidK . Because some V1 components are recalcitrant to purification in their soluble form , we used yeast mutants that lack individual V1 component genes [38] to identify the subunit that directly interacts with SidK . We first examined the formation of protein complexes between SidK and two V1 subunits , Vma1 and Vma2 in these mutants . Vma1 and Vma2 can be coimmunoprecipitated by the SidK antibody in mutants lacking vma4 , 5 , 7 , 8 , 10 or 13 , indicating that none of these subunits is required for the formation of protein complexes between SidK and Vma1 or Vma2 ( Fig . 5A ) . However , in the absence of Vma1 , no interactions between SidK and Vma2 or any other V1 components were detected ( Fig . 5A , lane 7 and data not shown ) . Importantly , although at a low level , Vma1 was detected in precipitates obtained by the SidK antibody in the vma2 mutant ( Fig . 5A , lane 3 ) . Furthermore , when beads coated with SidK were incubated with lysates of different vma mutants , Vma1 from the lysates of the vam2 mutant was retained ( Fig . 5B , lane 3 ) . Under the same condition , SidK coated beads did not retain Vma2 or other V1 components expressed in the vma1 mutant ( Fig . 5B , lane 2 and data not shown ) . Collectively , these results point to Vma1 as the direct target of SidK . To confirm this conclusion , we purified recombinant mammalian VatA as a GST tagged protein ( GST-VatA ) and tested its interaction with SidK . As expected , formation of SidK/GST-VatA complexes can be captured by GST beads ( Fig . 5C ) . From these results , we conclude that SidK targets the v-ATPase by directly interacting with the VatA ( Vma1 ) subunit . We extended our analysis of the interactions between SidK and VatA by determining the region on SidK important for target binding . To this end , we constructed a series of SidK deletion mutants ( Fig . 6A ) . To eliminate the potential discrepancy that may arise from the loss of epitopes in these mutants when the polyclonal anti-SidK antibody is used in subsequent experiments , we expressed GFP fusions of these mutants in mammalian cells . Deletion of 30 amino acids from the N-terminus of SidK did not detectably affect its binding with VatA ( Fig . 6B , lane 2 ) . However , although it was expressed at a high level , a mutant missing the first 94 amino acids no longer detectably bound VatA ( Fig . 6B , lane 3 ) . On the other hand , the VatA binding activity of SidK is far more tolerant to deletions in its carboxyl end . A mutant lacking as many as 382 amino acids from this end of SidK still co-precipitated with VatA at a level similar to that of the full-length protein ( Fig . 6B , lane 8 ) . Similar results were obtained in yeast , with the exception of sidKΔC382 , which did not bind Vma1 , probably due to the different binding affinity of SidK to v-ATPases from these two organisms ( Fig . S4 ) . Collectively , these results indicate that a domain that lies within residue 30 to 191 of SidK is important for interacting with VatA . To determine whether any of the SidK deletion mutants are still active in inhibiting v-ATPase functions , we tested their ability to inhibit yeast growth in neutral pH medium . Our results indicate that under this condition , SidKΔN30 consistently inhibits yeast growth , whereas the binding inactive mutant SidKΔN94 exhibits very little effect ( Fig . 6C ) . Similarly , the binding of SidKΔC98 and SidKΔ286 to VatA is comparable to that of full-length protein and both mutants exhibit detectable inhibition in yeast growth ( Fig . 6C ) . Although these mutants can be stably expressed in yeast from the GPD promoter , their expression levels are considerably lower than that of full-length SidK ( Fig . S5 ) . Since high protein level is required for SidK to exert full inhibitory effect on yeast growth ( Fig . 1 ) , the low activity exhibited by these mutants may result from the their low protein levels in yeast . The V1 domain of v-ATPase provides the energy required for proton translocation across membranes by binding and hydrolyzing ATP via subunit Vma2 and Vma1 , respectively [44] . Our observation that SidK directly binds Vma1 prompted us to examine the effect of such binding on ATP hydrolysis . Thus , we followed a standard procedure [45] to prepare yeast membrane and examined the effect of SidK on its ATPase activity . In this assay , exogenous ATP was added to purified yeast membranes and the release of free phosphate was measured by malachite green [46] . In samples receiving 2 µM BSA , the level of free phosphate steadily increased throughout the experimental duration ( Fig . 7A , open triangles ) . On the other hand , the addition of 1 µM bafilomycin A1 ( Baf A1 ) , a commonly used inhibitor of v-ATPase [47] led to strong inhibition of ATP hydrolysis , thus low level of free phosphate ( Fig . 7A , diamonds ) . Importantly , we found that recombinant SidK inhibited v-ATPase activity in a dose-dependent manner . Significant inhibition was observed by 0 . 1 µM SidK , and a higher amount of protein caused more severe inhibition ( Fig . 7A , closed triangles ) ; 1 µM of SidK exerted inhibition at a level similar to that of Baf A1 ( Fig . 7A , squares ) . Next , we probed the mechanism of action of SidK by adding its antibody to the reactions . The antibody did not detectably affect SidK activity even when it was two-fold in excess ( Fig . S6 ) . To confirm that the observed inhibition of ATP hydrolysis was indeed a result of blocking v-ATPase activity , we did similar experiments with membranes prepared from the vma1 mutant . For the same amount of membranes , the overall ATP hydrolysis activity of the mutant markedly decreased ( compare 3rd bar of mutant and 1st bar of WT in Fig . 7B ) . Moreover , the ATP hydrolysis activity of membranes from the mutant is resistant to Baf A1 or SidK , but not to EDTA and vanadate , two general inhibitors for ATPases [48] ( Fig . 7B , mutant 4th & 5th bar ) . On the other hand , ATPase activity in membranes prepared from wild type yeast consistently exhibited sensitivity to SidK , again in a dose-dependent manner ( Fig . 7B , wild type 2nd-5th bar ) . We also tested the effect of SidK on the ATP hydrolysis activity of mammalian Hsp70 , an ATPase structurally distant from the v-ATPase [49] . EDTA but not SidK abolished ATP hydrolysis by this heat shock protein; vanadate also exhibited inhibitory effect , but to a lesser extent ( Fig . S7 ) . Taken together , these results indicate that binding of SidK to Vma1 leads to specific inhibition of the ATP hydrolysis activity of the proton transporter . Because v-ATPase-mediated ATP hydrolysis is coupled with proton translocation , and thus the acidification of vesicles [50] , inhibition of v-ATPase activity by SidK would block vesicle acidification . To test this hypothesis , we determined the effect of SidK on the sequestration of the lipophilic amine acridine orange ( AO ) by yeast vesicles . Nonprotonated AO permeates membranes , and , if the pH of the vesicles drops as a result of v-ATPase-mediated proton translocation , it becomes protonated and sequestered , leading to quenching of its fluorescence [51] . In samples receiving the solvent DMSO that does not affect v-ATPase activity , more AO was trapped in the vesicles as proton translocation was initiated by adding ATP , leading to quenching of fluorescence at 525 nm ( Fig . 7C , diamonds ) . On the other hand , inclusion of 1 µM SidK to the reaction blocked such quenching during the entire experimental duration ( Fig . 7C , squares ) . The effect of SidK at this concentration is comparable to that of Baf A1 , which almost completely blocked AO fluorescence quenching ( Fig . 7C , triangles ) . From these observations , we conclude that inhibition of ATP hydrolysis activity of v-ATPase by SidK prevents proton translocation . To determine whether SidK affects the functions of v-ATPase in vivo , we delivered His6-SidK into mouse bone marrow-derived macrophages by syringe loading [52] and examined the acidification of phagosomes by using the dextran-coupled pH sensitive fluorescein , whose fluorescence drops very sharply at pH values below 5 . 5 [53] . The pH insensitive Cascade Blue dextran was included in the feeding mixture as a loading control . Macrophages in all samples emitted blue fluorescence signals at similar intensity , indicating that the dyes were equally loaded into the cells ( Fig . 8A , lower panel ) . Importantly , compared to macrophages receiving BSA , cells loaded with SidK gave significantly stronger green fluorescence signals ( Fig . 8A , upper panel , the first two images ) . Similarly , compared to cells loaded with BSA , cells treated with the v-ATPase inhibitor Baf A1 also emitted stronger fluorescence signals ( Fig . 8A , upper panel , the right image ) . These results indicate that SidK interferes with efficient acidification of phagosomes , thus inhibiting the decrease of their luminal pH values . To substantiate this observation , we used LysoRed , which accumulates in acidified organelles to stain protein-loaded macrophages that have been fed with E . coli cells expressing GFP for 10 hrs . Strong red fluorescence signals were readily detected in cells loaded with BSA , but not in cells receiving SidK ( Fig . 8B , left panel ) , further supporting the notion that SidK inhibits phagosomal acidification . No effect was detected in additional controls with two other Legionella effector proteins ( data not shown ) . Moreover , at this time point , we observed that the number of E . coli cells in macrophages loaded with SidK was significantly higher than that of cells containing BSA ( Fig . 8B , middle panel ) , suggesting that inhibition of phagosomal acidification by SidK impaired the lysosomal digestion of internalized bacteria . Since macrophages from mice lacking a functional a3 subunit exhibit delayed digestion of bacteria [54] , we set to more thoroughly examine the effect of SidK on macrophage-mediated lysis of E . coli cells . We first determined the survival of E . coli in macrophages loaded with different proteins . Macrophages loaded with His6-SidK or BSA are capable of killing phagocytosed bacteria , but cells containing SidK were less efficient in the clearance of the bacteria , and such differences became significant 6 hrs after adding the bacteria ( Fig . 8C ) . Similar results were obtained when macrophages harboring one or more intact E . coli cells were scored . In cells receiving His6-SidK , more than 90% of the cells harbor intact bacterial cells throughout the 24 hrs experiment duration ( Fig . 8D ) . However , 8 hrs after adding the bacteria , less than 40% of the macrophages loaded with BSA contained intact bacterial cells and the ratio of such cell population dropped to less than 10% at the12-hr time point ( Fig . 8D and E ) . Taken together , these results indicate that SidK can inhibit v-ATPase activity in vivo and such inhibition leads to defects in phagosomal acidification and impairment in lysosomal digestion of bacteria by macrophages .
Many intracellular bacterial pathogens reside and replicate in phagosomes of unique physiological and biochemical properties . One such property is an actively regulated pH homeostasis important for successful infection of these pathogens . Since the vacuolar ATPase is the primary cellular machinery involved in controlling vacuolar pH [50] , it is believed that pathogens capable of maintaining a neutral phagosomal pH encode specific traits to inhibit the accumulation of the proton transporters on their vacuolar membranes . However , although the importance of maintaining proper phagosomal pH , presumably by actively regulating the activity of v-ATPase , is generally recognized , almost nothing is known concerning the molecular mechanisms responsible for such regulation . Using a screening strategy based on the sensitivity of yeast v-ATPase mutants to neutral pH medium , we have identified SidK , a L . pneumophila protein that targets the proton transporter . A pathogen can employ at least two mechanisms to maintain a neutral luminal pH in its phagosomes: By preventing the accumulation of v-ATPases on the phagosomal membranes or by inhibiting the activity of acquired v-ATPases . Although we have not been able to consistently detect SidK on LCVs , probably because of low protein level and/or poor antibody quality ( Fig . 2 and data not shown ) , the presence of v-ATPases on LCVs [32] strongly suggests that SidK targets the proton pumps on the bacterial phagosomes . This feature differs from vacuoles of Mycobacterium ovium that do not contain detectable v-ATPases [31] . However , these two mechanisms are not mutually exclusive , because in addition to blocking its acquisition , the pathogen may need to antagonize v-ATPases that accidentally associate with its phagosomes . It is worth noting that detecting the association of v-ATPase with specific organelles can be complicated by low abundance of this protein complex on the membranes . For example , only a few v-ATPases were detected on a phagosome containing a latex bead [55] . Similarly , v-ATPases associated with LCVs can be detected by the sensitive mass spectrometry but not by standard immunostaining ( ref . [32] and data not shown ) . Thus , direct targeting of v-ATPase by specific virulence factors could be a mechanism shared by many intravacuolar pathogens . Our data showed that SidK interacts with v-ATPases by directly binding to the VatA subunit ( Fig . 5 ) . Moreover , this protein appears to have a higher affinity for VatA in the presence of VatB ( Fig . 5A ) . Such differences may result from the conformation assumed by VatA when it is associated with VatB [28] . Preferably binding to the VatA/VatB and/or the fully assembled v-ATPase complex clearly will result in higher inhibitory efficiency for SidK , as its effect can be diluted by free VatA if these two proteins interact similarly regardless of their statuses . Reversible assembly of the V1 and V0 domain is important in regulating v-ATPase activity under different physiological conditions [28] . Whether binding of SidK to v-ATPase causes disassembly of the transporter , block of its rotary movement or other functional mechanisms of the proton transporter remains to be determined . Our biochemical studies indicate that SidK blocks organelle acidification during infection . Two lines of evidence indicate that SidK inhibits v-ATPase in vivo . First , macrophages harboring physically delivered recombinant SidK failed to block emission of fluorescence signals by the pH sensitive fluorescein ( Fig . 8A ) . Second , cells loaded with SidK sequester significantly lower levels of LysoRed , a fluorescence dye that prefers to accumulate in acidic environments ( Fig . 8B ) . Moreover , similar to macrophages from mice lacking a subunit of the v-ATPase [54] , cells loaded with SidK displayed a significant delay in the digestion of phagocytosed bacteria ( Fig . 8C–E ) . Thus , it is clear that by binding to VatA , SidK is able to block phagosomal acidification , thus contributing to the protection of internalized L . pneumophila during infection . Bacterial effectors often enzymatically modify their targets to divert the cellular processes in ways beneficial to the survival of the pathogens [42] . However , despite considerable effort , we were unable to detect novel post-translational modifications on VatA co-purified with SidK from yeast ( data not shown ) . Moreover , our attempt to determine the mechanism of action of SidK by its antibody is not conclusive because the antibody is able to immunoprecipitate Vma1 or VatA , indicating that it can form a stable complex with these two proteins ( Fig . 5A and data not shown ) . Although SidK-mediated modifications of VatA or other v-ATPase subunits could substantiate its effect , two lines of evidence indicate the importance of physical binding in the activity of SidK . First , in contrast to other highly effective L . pneumophila effectors , such as those involved in inhibiting host protein synthesis or membrane trafficking [5] , a much higher level of SidK is required to significantly inhibit yeast growth in neutral pH medium , a condition that is completely unable to support growth of yeast vma mutants ( Fig . 1 ) . For instance , if the effect of SidK was mediated by a highly catalytic mechanism , one would expect more severe inhibition when expressed from the ADH ( alcohol dehydrogenase ) promoter ( Fig . 1C ) . Similarly , a considerable amount of SidK is needed to inhibit v-ATPase activity in yeast membranes ( Fig . 7 ) . Second , some deletion mutants capable of binding VatA still are able to exert inhibitory effect on yeast growth in neutral pH medium ( Fig . 6C ) . Given the requirement of high-level SidK for full growth inhibition , the loss of inhibitory effect by deletion mutants still competent for binding VatA very likely is a result of lower protein levels ( Fig . S5 ) . Alternatively , SidK may need other L . pneumophila proteins to exert its full activity or our experimental conditions are not optimal for its activity . In A/J mouse macrophages , vacuoles containing L . pneumophila become acidified during the replication phase of infection . Delayed maturation of the LCV promotes intracellular growth , since Baf A1 treatment blocks acidification and acquisition of lysosomal markers and also reduces the bacterial yield [30] . Consistent with the observation that expression of sidK peaks early in the lag phase before declining to almost undetectable levels as the bacteria enter the post-exponential phase , translocated SidK did not become detectable until 3 hrs after bacterial uptake . That translocated SidK is still detectable 12 hrs after infection suggests a delayed inhibition of SidK expression during infection or that SidK expresses differently in intracellular bacteria . In D . discoideum , the association of v-ATPases with the LCV is detectable from 15 min to 14 hrs after bacterial uptake [32] . The presence of this transporter in the early phase of infection suggests that the undetected SidK and/or other effectors antagonize its activity . Given the important roles of v-ATPase in vesicular trafficking , particularly in the endocytic pathway [56] that was recently shown to be involved in remodeling the membranes of the LCVs in D . discoideum [32] , prolonged biochemical modifications of the v-ATPase may interfere with the ability of the bacteria within phagosomes to efficiently acquire nutrients and materials from certain membrane trafficking pathways . Thus , L . pneumophila appears to use a combination of strategies to modulate the activities of v-ATPase at different phases of infection for its benefit . L . pneumophila appears to acquire many of its genes important for its interactions with host by horizontal gene transfer , which may account for at least in part the high plasticity of its genomes [57] , [58] . Although the distinct biochemical activities of these genes can interfere with host cellular processes , a single gene often plays only a small incremental role in its evolution to parasitize its hosts , which may explain the remarkable functional redundancy among effectors of the Dot/Icm system [5] , [42] . For example , at least four proteins are involved in inhibiting host protein synthesis by targeting the elongation factor eEF1A [20] , [59] . Consistent with this notion , with a few exceptions , deletion of one or more Dot/Icm substrate genes did not cause detectable defect in intracellular growth [5] , [42] , [60] . Thus , our observation that deletion of sidK did not impair intracellular growth of L . pneumophila or its phagosomal pH is not completely unexpected . It is very likely that multiple Dot/Icm substrates are involved in the modulation of v-ATPase activity . Identification and elucidation of activities of such proteins should pave the way toward further understanding of the mechanisms underlying organelle acidification and of the means whereby it can be disrupted by pathogens .
Bacterial strains used in this study are listed in Table S2 . Strains of E . coli were grown in LB and the medium was supplemented with the appropriate antibiotics when necessary . The L . pneumophila strain Philadelphia-1 strain Lp02 [61] was the parent of all derivatives used in this study . L . pneumophila was grown and maintained on CYE medium as previously described [62] . When necessary antibiotics were included as described [62] . To construct the sidK in-frame deletion mutant ZL114 , we constructed plasmid pZL886 by cloning two DNA fragments generated by primers PL192/PL193 and PL194/PL195 ( Table S4 ) into SacI/SalI digested pSR47s [61] . The primers were designed so that after deletion , only the first and last 15 amino acids are left in the mutant . pZL886 was introduced into Lp02 and the deletion mutant was obtained by following the standard allelic exchange method [63] . To complement the mutation , we inserted the coding region of sidK into pJB908 [40] . In complementation experiments , the vector used for expressing the gene of interest was introduced into the wild type strain or mutants and the bacterial cultures grown to the post-exponential phase as determined by optical density of the cultures ( OD600 = 3 . 3-3 . 8 ) as well as an increase of bacterial motility . The plasmids used in this study are listed in Table S3 and the sequences of all primers are in Table S4 . Plasmids harboring individual full-length L . pneumophila hypothetical genes were in Table S1 . The open reading frame of sidK and its derivatives were cloned into pEGFPC1 ( Clontech ) for expression in mammalian cells . A number of vectors , including pGBKT7 ( Clontech ) , p415ADH , p415TEF , p425TEF and p425GPD [39] were used to express sidK in yeast either as an untagged form or as GFP fusions ( see text for details ) . To express His6-SidK in L . pneumophila , we first amplified the multiple cloning site region of pQE30 ( Qiagen ) with primers QE5′EcoRV/QE3′XbaI and inserted it into Ecl136II/XbaI digested pJB908 [40] . to generate pZL507 . The sidK gene was then inserted into pZL507 as a BamHI/XhoI fragment to give pZL1333 . cDNA clones coding for relevant subunits of the v-ATPase were amplified from a human kidney cDNA library ( Clontech ) or from clones purchased from the ATCC . For expression in mammalian cells , sidK or each of these genes was inserted into pEGFPC1 ( Clontech ) or pFlag-CMV ( Sigma ) . The vatH gene ( pEF-HA-NBP1 ) was a gift from Dr . M . Peterlin of University of California , San Francisco . The integrity of all genes was verified by sequencing analysis . Yeast strains used were PJ69-4A [64] , BY4741 [38] and their derivatives ( Table S2 ) . Yeast was grown in YPD medium or in appropriate amino acid dropout minimal media at 30°C . Using a standard protocol [65] , we transformed plasmids carrying full-length hypothetical L . pneumophila genes [20] into yeast strain PJ69-4A [64] and grew the resulting strains over night in minimal medium of pH 5 . 5 . After diluting at 1∶40 into medium of pH 7 . 5 buffered with 50 mM MES and 50 mM MOP , the cultures were incubated with vigorous shaking for 24 hrs . Cultures that did not grow to high density were retained for further analysis . For quantitative study of growth , yeast subcultures of 2×106 cells/ml were made in appropriate Dropout medium and cell growth was monitored by measuring the OD600 at indicated time points . To prepare cell lysates for protein analysis , cells from 5 ml overnight cultures were first lysed with a cracking buffer ( 40 mM Tris-Cl [pH 6 . 8] , 5% SDS , 0 . 1 mM EDTA , 8 M urea , bromothymol Blue 0 . 4 mg/ml ) with glass beads . Samples were resolved by SDS-PAGE after adding Laemmli buffer . Mouse macrophages were prepared from bone marrow of female A/J mice of 6–10 weeks of age following published protocols [12] . U937 cells were cultured in RPMI medium supplemented with 10% fetal bovine/calf serum ( FBS ) and 5 mM glutamate , and if needed , the cells were differentiated into macrophages with 50 ng/ml phorbol myristic acid ( PMA ) as described [12] . 293T cells were cultured in Dulbecco's modified minimum Eagle's medium ( DMEM ) supplemented with 10% FBS . For transfection , we grew cells to about 80% confluence and transfected them with Lipofectamine 2000 ( Invitrogen ) following manufacturer's instructions . For growth curve experiments , macrophages were plated into 24-well plates at 2×105 cell per well . For immunoprecipitation and fractionation , about 2×107 cells plated on standard petri dishes were used . Infection was performed at the indicated MOIs as required by the particular experiments . To purify GST-SidK , we inserted the predicted sidK orf into pGEX-4T-1 ( Qiagen ) to generate pZL797 . E . coli strain XL1Blue containing pZL797 was grown in 1 liter LB to OD600 of 0 . 7 . After inducing with 0 . 2 mM IPTG for 6 hrs at 25°C , harvested cells were lysed by passing through a French press twice at 1 , 500 psi . Cleared supernatant was incubated with glutathione beads ( Qiagen ) for 2 hrs at 4°C and the beads were washed with 40X bed volume of PBS buffer containing 0 . 5% Triton X-100 . GST-SidK was eluted with PBS containing 10 mM reduced glutathione . When needed , GST tag was removed by Thrombin , a protease that was subsequently removed by benzamidine-Sepharose beads ( GE ) . GST-VatA was purified with a similar procedure . To purify His6-SidK from L . pneumophila , we introduced pZL1333 into the non-virulent strain Lp03 [61] . A 50 ml of saturated culture was diluted into 1 liter AYE broth , when the culture reached exponential growth phase ( OD600 = 0 . 5 ) , expression of the gene was induced with 0 . 2 mM IPTG for 16 hrs . Cleared cell lysates were incubated with Ni2+-Agrose beads for 2 hrs at 4°C and the beads were washed with 40 times of the bed volume of TBS buffer ( 50 mM Tris-HCl , 150 mM NaCl , pH 7 . 4 ) containing 10 mM imidazole . The protein was eluted with 200 mM imidazole . After dialyzing against TBS to remove imidazole , the protein was further purified by gel filtration with an FPLC system using a Superdex 200 10/300 GL column ( GE Healthcare ) . TBST buffer ( 50 mM Tris-Cl , 150 mM NaCl , 0 . 1% Triton-X100 , pH 7 . 4 ) was used as eluent and the flow rate was set at 0 . 4 ml/min . The single peak corresponding to the protein was collected , dialyzed in the appropriate buffer for subsequent use . His6-Hsp70 was similarly purified from E . coli . Protein concentrations were determined by the Bradford assay; the purity of all proteins was more than 95% as assessed by SDS-PAGE followed by Coomassie bright blue staining ( Figs . S3 and S7 ) . The procedure for affinity pulldown was described elsewhere [20] . Briefly , U937 cells collected from 500 ml culture suspended in 3 . 0 ml PBS containing 5 mM DTT and protease inhibitors ( Roche ) were lysed with a glass homogenizer ( Wheaton ) . The lysates were subjected to centrifugation at 10 , 000×g for 10 min at 4°C to remove unbroken cells and nuclei , the post-nuclear supernatant was added to Affigel beads coated with SidK and incubated for 14 hrs at 4°C . We then washed the beads five times with PBS and dissolved bound proteins with SDS sample buffer . After SDS-PAGE , proteins were visualized by silver staining ( Bio-Rad ) . Individual protein bands retained by beads coated by SidK but not by GST were excised , digested with trypsin , and analyzed by matrix-assisted laser desorption/ionization/mass spectrometry ( MALDI/MS ) ( Taplin Biological Mass Spectrometry Facility , Harvard Medical School ) . For GST pull down experiments , 10 µg purified GST or GST-VatA was mixed with 2 µg His6-SidK in PBS for 4 hrs at 4°C . After adding 40 µl of 50% pre-washed glutathione beads slurry , binding was allowed to proceed for 1 hr . The beads were then washed 5 times with PBS containing 500 mM NaCl . Retained proteins were detected by immunoblot after SDS-PAGE . Twenty-four hrs after transfection , cells were collected and lysed in a lysis buffer ( 0 . 2% of NP-40 , 50 mM Tris-HCl pH = 7 . 5 , 150 mM NaCl , 1 mM EDTA , 15% glycerol , and protease inhibitors ( 1 mM Na3VO4 , 1 mM PMSF , 10 µg/ml Aprotinin , 2 µg/ml Leupeptin , 0 . 7 µg/ml Pepstatin ) ) . After removing debris by centrifugation at 10 , 000 g for 10 min at 4°C , 2 mg protein ( approximately 1 ml ) was used for immunoprecipitation by adding the appropriate antibody and 30 µl of 40% protein G-sepharose beads ( GE Healthcare ) . After incubating at 4°C on a rotary shaker for 4 hrs , the beads were washed 5 times with the lysis buffer before being dissolved in Laemmli buffer . For immunoprecipitation with yeast lysates , cells harvested from 50 ml mid-log phase cultures were digested with Zymolyase , and the resulting spheroplasts were lysed with the lysis buffer used for mammalian cells . The lysates containing approximately 2 mg proteins were incubated with appropriate antibody and protein G sepharose for 16 hrs at 4°C . The beads were removed by washing 5 times with the lysis buffer . In both cases , protein associated with beads were dissolved in Laemmli buffer and resolved by SDS-PAGE . Proteins transferred to nitrocellulose membranes were detected by immunoblot . Antisera against Legionella , ICDH ( isocitrate dehydrogenase ) were described in an early study [4] . SidK cleaved from GST-SidK were used as an antigen to produce a specific antibody following a standard protocol ( Pocono Rabbit Farm and Laboratory , Canadensis , PA ) . GFP antibody was prepared similarly with purified His6-GFP . When necessary , antibodies were affinity-purified against the antigens covalently coupled to an Affigel matrix ( Bio-Rad ) using standard protocols [66] . Monoclonal or polyclonal antibodies against Flag , Vma1 , Vma2 , PGK ( 3-phosphoglycerate kinase ) , VatA and Hsp70 were purchased from Sigma , Invitrogen , Abcam and Santa Cruz Biotechnology ( sc-65521 ) , respectively . For Western blots , samples resolved by SDS-PAGE were transferred onto nitrocellulose membranes . After blocking with 4% milk in PBS buffer containing 0 . 2% Tween 20 , membranes were incubated with the appropriate primary antibody: anti-SidK , 1∶2 , 500; anti-VatA , 1∶10 , 000; anti-Vma1 , 1∶1 , 000; anti-Vma2 , 1∶1 , 000; anti-GFP , 1∶50 , 000; anti-PGK , 1∶ 2000; anti-ICDH , 1∶5 , 000; anti-Hsp70 , 1∶2000 . Horseradish peroxidase conjugated secondary antibodies and enhanced bioluminescence reagents were used to detect the signals ( Pierce , Rockford , IL ) . Alternatively , membranes were incubated with an appropriate IRDye infrared secondary antibody ( Li-Cor's Biosciences , Lincoln , Nebraska , USA ) and the signals were detected , and if necessary , the intensity of the bands are quantitiated by using the Odyssey infrared imaging system . Yeast vacuolar membrane vesicles were prepared according to the standard method [45] with some modification . Briefly , exponentially growing yeast cells ( O . D . = 0 . 6 ) were harvested , washed twice with distilled water and digested with zymolyase at 30°C for 90 min in 1 M sorbitol . The spheroplasts were resuspended in 7 volumes of Buffer A ( 10 mM MES/Tris ( pH 6 . 9 ) , 12% Ficoll 400 , and 0 . 1 mM MgCl2 ) , homogenized in a loosely fitting Dounce homogenizer ( Wheaton ) with 20 strokes , and centrifuged in a swinging bucket rotor at 4 , 500 g for 10 min . The supernatants were transferred to new centrifuge tubes , and buffer A was layered on the top . After centrifugation at 51 , 900×g for 40 min , the white layer on the top was collected and resuspended in Buffer A with a homogenizer , and Buffer B ( 10 mM MES/Tris ( pH 6 . 9 ) , 8% Ficoll 400 , and 0 . 5 mM MgCl2 ) was layered on the top . After similar centrifugation , vacuoles free from lipid granules or other membranous organelles were collected from the top of the tube . Vacuolar membrane vesicles were prepared by diluting the purified vacuoles in a vesicle buffer ( 10 mM MES/Tris ( pH 6 . 9 ) , 5 mM MgCl2 , and 25 mM KC1 ) . Equal amount of vacuolar membranes in ATPase buffer ( 10 mM HEPES , 5 mM MgCl2 , 125 mM KCl , pH = 7 . 0 ) were preincubated for 40 min at room temperature with or without the testing chemicals or proteins . To test the effect of antibody , purified antibody was added to the reactions for 20 min before the addition of ATP . BSA dissolved in the same buffer as that of SidK was used as a negative control . The reaction was initiated by adding 1 mM of ATP , and samples were withdrawn at indicated time points to measure the production of inorganic phosphate using the malachite green method [46] . Briefly , the malachite green reagent was made of 2 volumes of 0 . 0812% malachite green , 1 volume of 5 . 72% ammonium molybdate dissolved in 6 M HCl , 1 volume of 2 . 32% polyvinyl alcohol and 2 volumes of distilled water . 90 µl of the malachite green reagent was added to 10 µl samples withdrawn at indicated time points . The reactions were allowed to proceed for 2 min and were terminated with 1/10 volume of 34% sodium citrate . After incubation for another 20 min , absorbance at OD620 nm was measured . A standard curve simultaneously obtained with a series of phosphate solutions of known concentrations was used to determine the amount of phosphate released by the membranes . The ATP-driven proton transport activity was assayed by measuring the uptake of proton in the yeast membrane vesicles using acridine orange ( AO ) quenching assay [67] . Purified vacuolar membrane vesicles were diluted in AO buffer ( 5 mM HEPES , pH = 7 . 0 , 5 mM MgCl2 , 150 mM KCl , 6 µM AO ) , and preincubated for 40 min at room temperature with or without the testing chemicals or proteins . The reaction was initiated by adding 2 mM of ATP and quickly mixed . The quenching of acridine orange was monitored by the Spex FluoroMax 3 spectrofluorometer ( Jobin Yvon ) with excitation at 493 nm and emission at 525 nm . The pH of L . pneumophila-containing phagosomes was determined by fluorescence ratio imaging using 5 ( 6 ) -carboxyfluorescein-N-hydroxysuccinamide ester ( FLUOS , Fluka ) stained L . pneumophila as previously described [30] with the modifications detailed below . For labeling , L . pneumophila were cultured to the post-exponential phase , defined by motility and OD600 = 3 . 6 - 4 . 6 , washed once with 100 mM potassium phosphate buffer , pH 8 . 0 , and then incubated for 20 min at room temperature with 0 . 8 mg/ml FLUOS in 4% DMSO in 100 mM potassium phosphate , pH 8 . 0 . This treatment did not affect viability of bacteria as determined by quantifying colony formation . We infected macrophages plated on 24 mm coverslips with an MOI of ∼10 . Infections were synchronized by washing infected cells 4 times with RPMI/FBS 60 minutes after uptake . Following an additional 70–180 minutes of incubation , we washed the monolayers 3 times with 37°C Ringers Buffer ( RB; 55 mM NaCl , 5 mM KCl , 2 mM CaCl2 , 1 mM MgCl2 , 2 mM NaH2PO4 , 10 mM HEPES , and 10 mM glucose , pH 7 . 2 ) , and placed the samples at a 37°C chamber with 1 ml RB and visualized on an Olympus IX70 inverted microscope . Images were acquired from the attached CoolSNAP HQ2 14-bit CCD camera ( Photometrics ) and subsequently analyzed using Metamorph Premier v6 . 3 software ( Molecular Devices ) . Fluorescence images were obtained at excitation wavelengths of 492 nm and 436 nm and corrected for bias , shading , and background . Individual bacteria were masked by manual thresholding using an addition image of both wavelengths . The mask was then applied to each original corrected image , and the average fluorescence intensity over the masked area of each particle was determined at each wavelength . The pH of each L . pneumophila-containing phagosome was calculated from the ratio of the fluorescence intensity at an excitation wavelength of 492 nm to the intensity at excitation 436 nm . The fluorescence intensity ratios from two independent experiments were converted to pH using a single standard curve of quartic function . The standard curve was established using FLUOS-labeled bacteria immobilized on a coverslip coated with poly- ( L ) -lysine . Samples were processed as above , with greater than 130 bacteria analyzed per pH at 10 incremental values ranging from pH 3 . 5 to 8 . 5 in clamping buffer ( 130 mM KCl , 1 mM MgCl2 , 15 mM HEPES , 15 mM MES ) . Only intact rod-shaped bacteria were evaluated , with greater than 50 bacteria analyzed per coverslip in each of two independent experiments . To verify that sidK did not affect lysosomal degradation , the fraction of wild-type and sidK mutant bacteria that were degraded was quantified 1 h post-infection by fixed immunofluorescence microscopy as previously described [68] ( Fig . S2 ) . Heat killed ( 80°C 20 minutes ) wild-type bacteria served as a control for particles that trafficked to an acidic compartment [69] . We delivered recombinant proteins into macrophages by the syringe loading method [52] with some modifications: Briefly , cells were washed and collected in ice cold Dulbecco's PBS ( DPBS ) ( Cellgro ) by centrifugation ( 200 g 5 min ) . We then washed the cells twice with 37°C DPBS containing 1 . 2 mM CaCl2 before adding 200 µl warm loading solutions ( DPBS containing 1 . 2 mM CaCl2 and the protein to be loaded at 0 . 4 mg/ml ) to cell pellet containing 5×106 cells . A P-200 micropipettor ( Rainin ) set at 100 µl was used to pipett the cell suspension for 100 times at 37°C . After pipetting , the mixture was incubated at 37°C for 2 min . After washing twice with warm DPBS containing 1 . 2 mM CaCl2 , cells were seeded onto glass coverslips in 24-well plates with a density of 4×105 per well and were incubated at 37°C for 12–16 hrs . The bactericidal activity of macrophages loaded with different proteins was measured according to a published method [54] with minor modifications . Cells of E . coli strain XL1-Blue expressing the mCherry RFP or GFP were added to macrophages at an MOI of 10 for 1 hr at 37°C . The culture supernatant was replaced with fresh tissue culture medium containing 100 µg/ml gentamicin to kill extracellular bacteria . After 1 hr of incubation , the medium was replaced with fresh medium containing 10 µg/ml gentamicin . At indicated time points , cells were washed extensively ( 5x ) with warm PBS and lysed with 0 . 02% saponin . The lysates were plated on LB plates and colonies were counted after overnight incubation at 37°C . For the staining with fluorescein dextran , 10 kD fluorescent dextran ( Invitrogen ) was added to the cells to 0 . 2 mg/ml; the pH insensitive 10 kD Cascade Blue dextran ( Invitrogen ) was added at 0 . 2 mg/ml as a loading control . After incubating at 37°C for1 hr , the cells were washed 5 times and incubated at 37°C for an additional 4 hrs before being imaged . For the staining with LysoRed ( LysoTracker Red DND-99 , Invitrogen ) , cells of an E . coli strain expressing GFP were added to macrophage monolayer at an MOI of 20 for 1 hr at 37°C . After incubating with a medium containing 100 µg/ml gentamicin for 1 hr and then a medium containing 10 µg/ml gentamicin for an additional 8 hrs , a medium containing 50 nM LysoRed was added to the samples for 15 min . Cells were washed 3 times with fresh medium and subjected to imaging analysis immediately under an Olympus X-81 fluorescence microscope . All images were acquired with identical digital imaging parameters ( objectives , exposure duration , contrast ratios , etc . ) and were similarly processed using the IPlab software package ( BD Biosciences ) . The gene described in this manuscript is lpg0968 with an accession number of YP_095002 in the Genebank . | One hallmark of the lysosome is a low luminal pH that is important for its maturation as well as the activity of many hydrolyzing enzymes responsible for efficient digestion of phagocytosed contents . To survive and replicate in phagocytes , successful intracellular pathogens have evolved various mechanisms to circumvent the challenges posed by lysosomal killing . One salient feature associated with infection of the intracellular bacterial pathogen Legionella pneumophila is the maintenance of a neutral pH of the Legionella containing vacuoles ( LCVs ) that supports its intracellular growth in the early phase of infection , while the nonpathogenic mutants are believed to be immediately trafficked to an acidic compartment . In eukaryotic cells , organelle acidification is mediated by the vacuolar H+-ATPase that translocates protons into target compartments in a process energized by ATP hydrolysis . The recent discovery of the association of v-ATPase with LCVs points to the necessity for active modulation of v-ATPase activity by the bacterium . By screening L . pneumophila proteins that cause a yeast phenotype similar to its v-ATPase mutants , we have identified a substrate of the L . pneumophila Dot/Icm type IV secretion system that specifically inhibits the activity of the proton transporter . This protein , termed SidK , inhibits the activity of v-ATPase by directly interacting with the VatA subunit that is responsible for hydrolyzing ATP . Moreover , macrophages harboring SidK display defects in phagosomal acidification and lysosomal killing of non-pathogenic bacteria . We also found that expression of sidK is highly induced right after stationary bacteria are diluted into fresh medium , suggesting that SidK plays an important role in the early phase of infection . Our results reveal a mechanism by which an intravacuolar pathogen engages the v-ATPase protein and inhibits its activity , rather than actively avoiding its association with the pathogen's vacuolar membrane . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"cell",
"biology",
"microbiology",
"infectious",
"diseases/bacterial",
"infections",
"molecular",
"biology"
] | 2010 | Inhibition of Host Vacuolar H+-ATPase Activity by a Legionella pneumophila Effector |
Macrophages are a heterogeneous cell population strongly influenced by differentiation stimuli that become susceptible to HIV-1 infection after inactivation of the restriction factor SAMHD1 by cyclin-dependent kinases ( CDK ) . Here , we have used primary human monocyte-derived macrophages differentiated through different stimuli to evaluate macrophage heterogeneity on cell activation and proliferation and susceptibility to HIV-1 infection . Stimulation of monocytes with GM-CSF induces a non-proliferating macrophage population highly restrictive to HIV-1 infection , characterized by the upregulation of the G1/S-specific cyclin D2 , known to control early steps of cell cycle progression . Knockdown of cyclin D2 , enhances HIV-1 replication in GM-CSF macrophages through inactivation of SAMHD1 restriction factor by phosphorylation . Co-immunoprecipitation experiments show that cyclin D2 forms a complex with CDK4 and p21 , a factor known to restrict HIV-1 replication by affecting the function of the downstream cascade that leads to SAMHD1 deactivation . Thus , we demonstrate that cyclin D2 acts as regulator of cell cycle proteins affecting SAMHD1-mediated HIV-1 restriction in non-proliferating macrophages .
Macrophages are a highly heterogeneous cell population that plays a prominent role in innate immune system as key effector cells for the elimination of pathogens , infected cells and cancer cells [1 , 2] . Macrophages also play an essential role in maintaining tissue homeostasis by supporting tissue development and repairing damaged tissue architecture [1 , 3] . Macrophage differentiation from monocytes occurs in the tissue in concomitance with the acquisition of a functional phenotype that depends on microenvironmental signals , accounting for the wide and apparently opposed variety of macrophage functions [4 , 5] . Macrophages , as well as other myeloid lineage cells , become susceptible to HIV-1 infection after degradation or inactivation of the restriction factor SAMHD1 , a triphosphohydrolase enzyme that controls the intracellular level of dNTPs [6–9] . Phosphorylation of SAMHD1 by cyclin dependent kinases ( CDK ) has been strongly associated with inactivation of the virus restriction mechanism , providing an association between virus replication and cell proliferation [10–12] . The activity of CDK is regulated by the binding of cyclins , a family of proteins characterized by a periodic , cell-cycle dependent pattern of expression [13 , 14] . Cyclin-CDK complexes govern cell cycle progression and proliferation of mammalian cells and thus , pinpoint the specific time in which an event occurs during the cell cycle [13 , 14] . We and others have shown that the complex cyclin D3-CDK6 acting upstream of CDK2 controls SAMHD1 phosphorylation and function in primary lymphocytes and macrophages [11 , 15–17] . Cyclin-CDK function is also controlled by cyclin dependent kinase inhibitors ( CDKIs ) that generally act as negative regulators of the cell cycle by binding to CDKs and inhibiting their kinase activity [18] . Of particular importance is p21/waf1 , a G1/S phase CDKI , that may also control HIV-1 replication through SAMHD1 [19 , 20] . D-type cyclins ( cyclins D1 , D2 and D3 ) are regarded as essential links between cell environment and the core cell cycle machinery . D-type cyclins drive cells through the G1 restriction point and into the S phase , after which growth factor stimulation is no longer essential to complete cell division [21] . D-type cyclins share the capacity to activate both CDK4 and CDK6 [14] . Studies on single , double and triple cyclin D knockout mice revealed that D-type cyclin complexes have redundant functions . However , different D-type cyclins exhibit distinct expression patterns depending on the cell type , indicating that each D-type cyclin has essential functions in particular settings , as suggested by the narrow and tissue-specific phenotypes of the knockout mice ( reviewed in [21] ) . Here , we have used primary human monocyte-derived macrophages ( MDMs ) differentiated through different stimuli to evaluate macrophage heterogeneity on cell activation and proliferation , characteristics that influence gene and protein expression patterns and determine susceptibility to HIV-1 infection . The comparative study has led to the identification and characterization of a cell cycle dependent pathway that restricts HIV-1 infection in primary macrophages . These non-proliferating macrophage population is characterized by a high expression of the G1/S-specific cyclin D2 . Cyclin D2 acts through the binding to CDK4 and p21 in GM-CSF macrophages , a complex which is responsible for the lack of the active CDK that phosphorylates SAMHD1 . Data from mouse peritoneal macrophages confirmed the existence of cyclin D2 expressing macrophages in vivo , further supporting the key role of cyclin D2 .
Primary monocyte-derived macrophages were differentiated either with M-CSF or GM-CSF , with the aim to characterize differences in cell activation and proliferation patterns and susceptibility to HIV-1 infection . Differentiated macrophages displayed different morphological characteristics dependent on the differentiation stimuli , but no significant differences in cell surface antigen expression or HIV receptor and co-receptors were observed ( S1A Fig ) . Interestingly , cell proliferation and cell cycle patterns were significantly different between macrophage types , i . e . , M-CSF macrophages proliferated at higher rates than GM-CSF measured by intracellular Ki67 staining ( Fig 1A , 7% vs . 0 . 5% of Ki67+ cells in M-CSF and GM-CSF macrophages respectively , p = 0 . 02 ) . Similar results were obtained analyzing cell cycle profile by DNA and RNA content staining ( Fig 1B ) , showing higher percentage of cells in S/G2M stage in M-CSF than GM-CSF macrophages . Susceptibility to HIV-1 infection was also significantly different , being M-CSF macrophages in average roughly 10-fold more susceptible to HIV-1 infection ( Fig 1C , left panels and 1D , p = 0 . 0072 ) , which correlated with higher dNTP levels in cycling M-CSF macrophages compared to GM-CSF as previously reported ( S1B Fig and [22 , 23] ) . These results point towards the established link between cell cycle progression and susceptibility to infection as the determinant of the differences observed between macrophage types , a process where SAMHD1 restriction is central [11] . Indeed , degradation of SAMHD1 by HIV-2 Vpx increased HIV replication in both macrophage types , minimizing the initial differences in infection ( Fig 1C , right panels , and 1D , p = 0 . 12 ) . No differences were found in antiviral activity of drugs either targeting viral reverse transcription ( AZT or nevirapine ) or the cell cycle inhibitor palbociblib [24] ( PD , specifically targeting CDK4/6 and inhibiting SAMHD1 phosphorylation ) between M-CSF and GM-CSF differentiated macrophages ( Fig 1E and 1F ) . Moreover , after SAMHD1 degradation by HIV-2 Vpx , decreased antiviral potency of AZT and a complete loss of PD antiviral activity were observed , suggesting that cellular events leading to SAMHD1-mediated viral restriction were similar in both types of macrophages [11 , 15] . Similar results were obtained when culturing macrophages in the presence of human serum , indicating that the differentiation stimuli induced by the cytokine determines the macrophage phenotype ( S2 Fig ) . To further investigate the molecular determinants of the observed differences between macrophages types , expression of cell cycle genes implicated in SAMHD1 control were evaluated ( Fig 2A ) . No major gene expression differences were observed , except for a significant upregulation of D-type cyclins ( CCND1 , 2-fold , p = 0 . 0006; CCND2 , 40-fold , p = 0 . 0004; and CCND3 , 3-fold p = 0 . 0002 ) and the CDK inhibitor p21 ( CDKN1A , 4-fold , p = 0 . 038 ) in GM-CSF macrophages ( Fig 2A ) . Analysis of protein expression confirmed the upregulation of cyclin D2 , cyclin D3 and p21 in GM-CSF macrophages , and revealed a clear downregulation of CDK protein levels ( CDK1 , CDK2 and CDK6 but not of CDK4 ) and the negative cell cycle regulator p27 ( Fig 2B ) . As expected , no differences were found in SAMHD1 expression but a change in SAMHD1 activation was observed , being only phosphorylated and partially inactivated in M-CSF macrophages ( Fig 2B , upper panels ) . Importantly , mouse peritoneal macrophages showed a similar expression pattern than that observed in GM-CSF macrophages , suggesting the existence of cyclin D2 expressing macrophages in vivo ( S3 Fig ) . From a functional point of view , stimulation with LPS resulted in the upregulation of IFNB1 , IL-10 and CCL-2 production in both types of macrophages , albeit basal expression levels were different in M-CSF and GM-CSF as reported elsewhere [25 , 26] ( Fig 2C ) . These results demonstrate that differentiation stimuli strongly impact the cell cycle profile of primary macrophages , and consequently their capacity to support HIV-1 replication , that may be in part determined by differences in the restriction factor SAMHD1 activation . Differential expression of D-type cyclins , and specially cyclin D2 may represent key regulatory proteins that shape the distinct cell cycle profile and affect HIV-1 restriction . The role of D-type cyclins in macrophage differentiation and HIV-1 susceptibility was further evaluated by RNA interference . Effective and specific downregulation of cyclin D2 and cyclin D3 was achieved at both mRNA and protein level in M-CSF and GM-CSF macrophages ( Fig 3A and 3B ) . Higher expression of CCND2 in the GM-CSF macrophage population compared to M-CSF , was again clearly observed in all conditions tested ( Fig 3A , left panel , and 3B ) . SAMHD1 expression was not affected by D-type cyclin inhibition , as measured by mRNA level ( S4A Fig ) or protein expression ( Fig 3B ) . However , different effects on SAMHD1 phosphorylation were observed depending on the targeted cyclin and the differentiation stimuli . As previously reported [16] , cyclin D3 knockdown resulted in the abolishment of SAMHD1 phosphorylation in M-CSF macrophages compared to a non-targeting siRNA ( Fig 3B , lanes 1–3 ) , an effect that could not be evaluated in GM-CSF macrophages due to low expression levels of phosphorylated SAMHD1 . On the contrary , cyclin D2 knockdown resulted in increased SAMHD1 phosphorylation in the GM-CSF macrophage population ( Fig 3B , lanes 4–6 ) . Similarly , cyclin D3 but not cyclin D2 knockdown was associated with lower cell proliferation measured as percentage of Ki67 positive macrophages , an effect that was evident in proliferating M-CSF macrophages , but difficult to evaluate in non-proliferating GM-CSF macrophages ( Fig 3C ) . siRNA-treated macrophages were tested for their capacity to support HIV-1 replication . HIV-1 replication was inhibited after cyclin D3 knockdown in M-CSF macrophages ( roughly 60% inhibition , p = 0 . 0095 ) , whereas inhibition of cyclin D2 expression did not have any effect ( Fig 3D , white bars ) . Conversely , in GM-CSF macrophages cyclin D2 downregulation led to a significant increase of HIV-1 replication ( 3-fold increase , p = 0 . 03 ) , but no effect was observed in cyclin D3 knockdown macrophages ( Fig 3D , black bars ) . These results demonstrated different roles for cyclin D2 and D3 at controlling cell proliferation and HIV-1 infection . To determine the molecular bases of cyclin D2 effect on HIV-1 replication in GM-CSF macrophages , most common protein interactions for cyclin D2 were identified by database searches [27 , 28] . p21 , CDK4 and CDK6 were found as the most common proteins bound to cyclin D2 . Cyclin D2 is known to form a complex with CDK4 and CDK6 where it functions as a regulatory subunit of both CDK at G1/S transition [18] . However , CDK6 expression could not be detected in GM-CSF macrophages ( Fig 2B ) , therefore , further characterization was centered in D-type cyclins and their interactors p21 and CDK4 . RNA interference was used to effectively and specifically downregulate cyclin D2 , cyclin D3 , p21 and CDK4 expression in GM-CSF macrophages ( Fig 4A and S4B Fig ) . Evaluation of gene expression in siRNA-treated macrophages suggested that CCND2 ( cyclin D2 ) and CDKN1A ( p21 ) expression is cross-regulated , as interference of CCND2 expression lead to a significant downregulation of CDKN1A ( Fig 4A , third panel , 40% inhibition compared to mock macrophages , p = 0 . 007 ) and a similar trend was observed when CCND2 expression was evaluated in siCDKN1A macrophages , although it did not reach statistically significance ( Fig 4A , first panel , 30% inhibition compared to mock , p = 0 . 06 ) . Protein expression confirmed the effective downregulation of all target proteins ( Fig 4B ) as well as the mutual regulation of cyclin D2 and p21 , as knockdown of cyclin D2 or p21 induced a downregulation of p21 or cyclin D2 expression , respectively ( Fig 4B , lane 3 and lane 5 ) . As above , SAMHD1 expression was not altered in siRNA-treated macrophages , but significant changes were observed in its phosphorylated form , showing an increased phophorylation in cyclin D2 and p21 knockdown macrophages ( Fig 4B , two upper panels ) . Cell proliferation status showed a small percentage of proliferating cells in all cases , albeit a slight increase in Ki67 positive cells was observed in cyclin D2 and p21 knockdown GM-CSF macrophages ( Fig 4C , upper panels ) . Cell cycle analysis did not show any significant differences between the different macrophages ( Fig 4C , lower panels ) . Altogether , these results reinforce the idea of cyclin D2 and p21 sharing a common regulatory pathway in GM-CSF macrophages . siRNA-treated GM-CSF macrophages were also tested for their capacity to support HIV-1 replication . Knockdown of cyclin D2 and p21 significantly increased HIV-1 replication in GM-CSF macrophages infected with a VSV-pseudotyped NL4-3 GFP expressing virus ( roughly 4-fold increase , p = 0 . 0007 for siCCND2 and p = 0 . 0003 for siCDKN1A , respectively , Fig 4D ) . On the contrary , no effect was seen when cyclin D3 or CDK4 expression were inhibited . Proviral DNA formation was also enhanced in cyclin D2 and p21 knockdown GM-CSF macrophages in short-term infections with the fully replicative HIV-1 R5-tropic strain BaL ( roughly 3-fold increase , p = 0 . 007 for siCCND2 and p = 0 . 03 for siCDKN1A respectively , Fig 4E ) . As expected , the HIV-1 reverse transcriptase inhibitor AZT completely blocked proviral DNA formation , while the HIV-1 integrase inhibitor Raltegravir ( RAL ) did not have an effect on viral DNA formation ( Fig 4E ) . Importantly , confirmatory siRNA sequences targeting Cyclin D2 showed similar effects on infection and proviral DNA formation after HIV-1 BaL infection ( S4C and S4D Fig ) , indicating that cyclin D2 acts as part of a viral restriction mechanism in GM-CSF macrophages . No significant differences in basal cytokine expression and the capacity to induce cytokine expression after LPS stimulation was also preserved in cyclin D2 or p21 knockdown macrophages , suggesting no major functional abnormalities as a result of inhibition of cyclin D2 or p21 ( S5 Fig ) . To investigate the interaction between cyclin D2 and p21 , a plasmid expressing a fusion protein Flag-p21 [29] was transfected into HEK293T cells and p21 was immunoprecipitated using Flag-specific agarose beads ( Fig 5A ) . Cyclin D2 co-immunoprecipitated with Flag-p21 ( Fig 5A , last lane ) and it was not identified when using lysates from mock-transfected cells ( M ) , demonstrating the existence of a protein complex with cyclin D2 and p21 . The presence of a CDK in the complex was also investigated and found that CDK1 , but not CDK4 or CDK6 , immunoprecipitated together with cyclin D2 and p21 in HEK293T cells . The interaction of cyclin D2 and p21 was confirmed by overexpression of a cyclin D2–HA fusion protein [30] in HEK293T cells followed by immunoprecipitation using HA-specific agarose beads . As expected , p21 co-immunoprecipitated with cyclin D2-HA and CDK1 but not CDK4 ( Fig 5B ) , demonstrating the existence of a protein complex between cyclin D2 , p21 and a relevant CDK associated to SAMHD1 function . Deregulation of the cell cycle is a hallmark of most laboratory adapted cell lines in which CDK1 may play a preponderant role in controlling cell proliferation , while other CDKs may have a tissue specific role in driving the cell cycle and cell differentiation [13] . Taking this into account , co-immunoprecipitation experiments of endogenous cyclin D2 and p21 in GM-CSF macrophages were also performed . Importantly , immunoprecipitation of endogenous cyclin D2 or p21 resulted in the identification of p21 or cyclin D2 , respectively , demonstrating the coexistence of both proteins in a complex also in GM-CSF primary macrophages ( Fig 5C and 5D ) . The presence of a CDK in the same protein complex was also investigated and CDK4 , but not CDK1 , was found to co-immunoprecipitate with either cyclin D2 or p21 in primary macrophages ( Fig 5C and 5D ) . When immunoprecipitating p21 higher amounts of cyclin D2 and CDK4 were found than when immunoprecipitating cyclin D2 , which might indicate that most p21 is found complexed with cyclin D2 and CDK4 , whereas in the case of cyclin D2 , only a fraction of the total protein is bound to p21 and CDK4 ( Fig 5D ) . These results further demonstrate the existence of a protein complex formed by cyclin D2 , p21 and CDK4 that may govern cell cycle progression and HIV-1 susceptibility in GM-CSF primary macrophages . CDK1 and CDK2 have been identified as the kinases responsible for SAMHD1 phosphorylation in cycling cells and macrophages , respectively [10–12] . Thus , to delineate the molecular pathway regulated by cyclin D2 , expression and activation of CDK1 and CDK2 were analyzed in siRNA-treated GM-CSF macrophages ( Fig 6 ) . Knockdown of cyclin D2 enhanced significantly CDK1 mRNA ( roughly 3-fold , p = 0 . 0021 , Fig 6A , left panel ) and protein expression ( Fig 6B ) . Although no significant upregulation of CDK2 expression was observed ( Fig 6A , right panel ) , CDK2 CDK2 regulatory phosphorylation at Thr130 ( pCDK2 , Fig 6B ) was significantly increased suggesting a higher activity of CDK2 . According to the classical model of cell cycle control , CDK4 or CDK6 regulate events in early G0 to G1 phase , CDK2 triggers S phase , CDK2/CDK1 regulate the completion of the S phase and CDK1 is responsible for mitosis [13] . Thus , the complex formed by cyclin D2-CDK4-p21 might be responsible for the lack of active CDK2 and CDK1 expression in GM-CSF macrophages . This situation is reversed in the absence of cyclin D2 , leading to the activation of CDK2 and CDK1 with the subsequent phosphorylation of its substrates , including SAMHD1 . As a consequence , SAMHD1 phosphorylation results in inactivation of virus restriction and enhancement of HIV-1 replication ( Fig 6C ) .
Macrophages are key components of the innate immune system that reside in tissues , where they function as immune sentinels . Although macrophage heterogeneity has classically been organized around two polarized endpoints known as M1 or classical and M2 or alternative activation [31] , recent evidence have revealed an unrecognized greater diversity in the ontogeny and functional diversity of tissue-resident macrophages . It is now established that macrophages from embryonic progenitors can persist in tissues into adulthood and self-maintain by local proliferation ( reviewed in [4 , 5] ) . However , monocytes also contribute to the resident macrophage population , on which the local environment can impose tissue-specific macrophage functions ( reviewed in [4 , 5] ) . The myeloid colony-stimulating factors , M-CSF and GM-CSF are known to modulate macrophage phenotype and many studies illustrate their importance in the magnitude , duration and character of inflammatory responses [32 , 33] . Thus , the description of the distinct molecular phenotypes consequence of M-CSF or GM-CSF stimulation may be important for both acute and chronic inflammatory pathology . Here , we show that differentiation stimuli determine distinct cell proliferation and cell cycle progression in primary macrophages , characteristics that are mostly dependent on the differential expression patterns of cell cycle proteins , especially D-type cyclins , their catalytic partners , CDKs , and the CDK inhibitor p21 . Consistent with our observations , upregulation of cyclin D2 in response to GM-CSF treatment has already been reported in different hematopoietic cells [34–36] , suggesting that cyclin D2 role might be relevant also for other cell types . The molecular interplay between the cell cycle , cyclins , and cell function is far from being fully understood . Conceptual advances in the field continue to uncover novel and interesting roles for cyclins in cellular processes that contribute to diseases , such as cancer or pathogenic infections [37] . It is well accepted that cell cycle control plays a major role in determining susceptibility to HIV-1 infection [38 , 39] . It has been previously shown that limiting dNTP synthesis , for example through ribonucleotide reductase inhibition by hydroxyurea , limits HIV-1 replication [40] . Thus , differences on cell cycle regulation between M-CSF and GM-CSF macrophages also determined the susceptibility to HIV-1 infection and this may be a direct consequence of the control of the restriction factor SAMHD1 by CDK-mediated phosphorylation . Previous observations in M-CSF macrophages , showed that SAMHD1 phosphorylation was directly phosphorylated by CDK2 , whose kinase activity was upstream regulated by the cyclin D3-CDK6 complex [11 , 16] . However , the molecular mechanism might be different in GM-CSF macrophages because CDK6 was barely expressed and D-type cyclins expression was significantly upregulated , especially that of cyclin D2 . Accordingly , knockdown of cyclin D2 and cyclin D3 resulted in opposite effects depending on the macrophage type , in terms of susceptibility to HIV-1 infection: cyclin D2 restricts infection in non-cycling GM-CSF macrophages and cyclin D3 enables infection in proliferating M-CSF macrophages . All D-type cyclins ( D1 , D2 and D3 ) are closely associated to the G1 phase , whose expression is induced by mitogenic signals and therefore play a significant role in cell cycle entry , by assembling together with CDK4 or CDK6 [18 , 21 , 37] . However , although apparently redundant , single knockout mice exhibit several cell type specific abnormalities [21] , suggestive of essential functions on particular settings for each cyclin , explaining , in part , the apparently opposed effects observed here for cyclin D2 and D3 when assessing susceptibility to HIV infection . On the other hand , D-type cyclins are linked to many human malignancies . Cyclin D1 is a well-established oncogene ( review in [18] ) but cyclin D2 or cyclin D3 overexpression are rarely reported [41 , 42] . However , CCND2 is frequently methylated , with loss of cyclin D2 expression in pancreatic , breast and prostate cancer [43–45] , pointing to a potential role as a tumor suppressor rather than an oncogene , indicating that overexpression of cyclin D2 might be limiting cell proliferation , as described here in GM-CSF macrophages . To uncover the molecular mechanism underlying cyclin D2 function in GM-CSF macrophages and taking into account that cyclins alone do not have catalytic activity per se , the study of putative partners of cyclin D2 was addressed . In concordance with the differential gene expression patterns , the inhibition of CDK4 which is similarly expressed in both macrophage types did not change protein profile or HIV-1 susceptibility . Conversely , knockdown of the CDK inhibitor p21 , whose expression was also upregulated in GM-CSF macrophages , showed the same effect as that of Cyclin D2 , i . e . , upregulation of SAMHD1 phosphorylation and enhancement of HIV-1 replication , similar to our previous observation in M-CSF macrophages [20] . p21 belongs to the Cip/Kip family of CDKIs that have historically been considered negative regulators of the cyclin-CDKs and therefore controlling cell cycle progression , especially when referred to cyclin-CDK1 or -CDK2 complexes [20 , 46] . However , Cip/Kip family of CDKIs interaction with Cyclin D-CDK4/6 appear much more complex , being involved in both the stabilization of the cyclin D/CDK complex but also acting as inhibitors of its kinase activity [47 , 48] . The observation that a complex formed by Cyclin D2/CDK4/p21 was identified in non-cycling GM-CSF macrophages argues in favor of the inhibitory hypothesis at least in this specific cell type . Moreover , the fact that p27 , another Cip/Kip family member , is barely expressed in GM-CSF macrophages indicates the specificity to the effect of Cyclin D2/CDK4/p21 complex . A direct inhibition of CDK2 function , similar to that observed in M-CSF [20] cannot be completely rule out , but seems improbable due to the low expression of CDK2 and lack of cell proliferation markers seemed in GM-CSF macrophages . The present works also highlights the interplay between cell cycle control and viral replication , with important implications that might be broader than simply affecting susceptibility to HIV-1 infection . Interestingly , both p21 and Cyclin D2 were identified as potential markers for viral latency , as both genes showed increased histone modification levels in HIV latently infected cells [49] . These observation suggests that the maintenance or not of HIV-1 latency may also be controlled by cell cycle related proteins such as Cyclin D2 and p21 and thus , open the opportunity for new therapeutic interventions . In addition , p21 and Cyclin D2 were also found upregulated after HTLV-I infection , a process that is mediated by the viral protein Tax , indicating that deregulation of G1/S checkpoint is also relevant for other retroviruses [48] . In summary , the identification of a novel cell cycle-mediated viral restriction pathway in primary non cycling macrophages have provided new evidences of the tight interplay between viral replication and cell cycle control , pointing towards the concerted action of Cyclin D2 and p21 in HIV-1 replication . The demonstration and characterization of specific D-type Cyclin roles in certain cell types , such as that reported for Cyclin D2 here , may also offer a window of opportunity for targeting D Cyclins in viral infections and in human cancers as well , where D-type Cyclin expression is frequently deregulated .
PBMC were obtained from buffy coats of blood of healthy donors using a Ficoll-Paque density gradient centrifugation and monocytes were purified using negative selection antibody cocktails ( StemCell Technologies ) as described before [22] . Monocytes were cultured in complete culture medium ( RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum ( FBS; Gibco ) or human serum ( HS; Sigma ) and penicillin/streptomycin ( Gibco ) and differentiated to monocyte derived macrophages ( MDM ) for 4 days in the presence of monocyte-colony stimulating factor ( M-CSF , Peprotech ) or granulocyte-macrophage colony-stimulating factor ( GM-CSF , Peprotech ) both at 100 ng/ml . The protocol was approved by the scientific committee of Fundació IrsiCaixa . Buffy coats were purchased from the Catalan Banc de Sang i Teixits ( http://www . bancsang . net/en/index . html ) . The buffy coats received were totally anonymous and untraceable and the only information given was whether or not they have been tested for disease . When appropriate , differentiated macrophages were incubated with 100 ng/ml of lipopolisaccaride ( LPS , Sigma-Aldrich ) overnight at 37°C . TZM cells were received from the National Institutes of Health , AIDS Research and Reference Reagent Program . HEK293T cells were purchased from Dharmacon ( Madrid , Spain ) . Wild-type C57BL/6 inbred mice were purchased from Harlan Laboratories ( Sant Feliu de Codines , Barcelona , Spain ) and housed at the Animal House facility at the Research Institute Germans Trias i Pujol . Mice were housed under specific pathogen free conditions in a temperature and humidity-controlled room with 12-h light/12-h dark cycle . Only adult males were used in this study . In vivo experiments were performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Generalitat de Catalunya , Catalan Government and the Principles of laboratory animal care ( NIH pub . 85–23 revised 1985; http://grants1 . nih . gov/grants/olaw/references/phspol . htm ) . Murine peritoneal macrophages were isolated as described [50] . Briefly , C57Bl/6 mice were sacrificed by cervical dislocation . Immediately after , peritoneal wall was exposed and 10 ml of a cold solution of PBS with 3% FBS per mouse was injected into the peritoneal cavity . Using the same syringe and needle , the fluid from the peritoneum was aspirated , and 8 ml of fluid was recovered . Peritoneal fluid was centrifuged at 4°C and cell pellet was resuspended to adjust cell concentration at 1–3 x 106 cells / ml . To obtain monolayers of peritoneal macrophages , total cells were plated at 2–3 × 106 total nucleated cells/ml in DMEM/F12-10 medium . Cells were allowed to adhere for 1–2 hr at 37°C . Non-adherent cells were removed by gently washing three times with warm PBS . At this time , cells were greater than 90% macrophages and were processed . Isolated monocytes were transfected as previously described [11 , 16] . Briefly , 50 pmol of the corresponding siRNA ( siGENOME SMARTpool from Dharmacon , Thermo-Scientific , Waltham , USA and ThermoFisher Scientific ) , were transfected using a Monocyte Amaxa Nucleofection kit ( Lonza , Basel , Switzerland ) following manufacturer instructions . Monocytes were left untreated overnight and then differentiated to macrophages as described above . For intracellular Ki67 staining , cells were fixed for 3 min with Fixation Buffer ( Fix & Perm , Life Technologies ) before adding pre-cooled 50% methanol for 10 min at 4°C . Cells were then washed in PBS with 5% FBS and incubated for 30 min with the Ki-67 FITC antibody diluted in permeabilitzation buffer ( 1:10; clone B56 , BD Biosciences ) . For cell cycle analysis , cell were suspended in 0 . 03% saponin ( Sigma-Aldrich ) in PBS and then incubated in 20 mM 7-aminoactinomycin D ( 7AAD; Sigma-Aldrich ) for 30 min at room temperature in the dark , followed by 5 min at 4°C . Then , Pyronin Y ( Sigma-Aldrich ) was added at a final concentration of 1 . 5 μg/ml and cells were further incubated at 4°C for 15 min . Flow cytometry was performed in a LSRII flow cytometer ( BD Biosciences ) . The data were analyzed using the FlowJo software ( BD Biosciences ) . To correct the overestimation of G2/M population by miss discrimination of cellular doublets , FL2W versus FL2A of the 7AAD dye was plotted before gating for the distinct cell cycle phases [51] . 3-Azido-3-deoxythymidine ( zidovudine , AZT ) was purchased from Sigma-Aldrich ( Madrid , Spain ) . nevirapine ( NVP ) and raltegravir ( RAL ) were obtained from the NIH AIDS Research and Reference Reagent Program . PD-0332991 ( palbociclib ) was purchased from Selleckchem . For relative mRNA quantification , RNA was extracted using the NucleoSpin RNA II kit ( Magerey-Nagel ) , as recommended by the manufacturer , including the DNase I treatment step . Reverse transcriptase was performed using the High Capacity cDNA Reverse Transcription Kit ( Life Technologies ) . mRNA relative levels of all genes were measured by two-step quantitative RT-PCR and normalized to GAPDH mRNA expression using the DDCt method . Primers and DNA probes were purchased from Life Technologies ( TaqMan gene expression assays ) . Cytokine expression was evaluated by using the commercial TaqMan Human Cytokine Network array ( 4414255 , Life Technologies ) , which included primers and probes for 28 different cytokine genes . mRNA relative levels of all cytokine genes were measured by two-step quantitative RT-PCR and normalized to GAPDH mRNA expression using the DDCt method . Intracellular dNTP content was determined using a polymerase-based method [52] as previously described [23] . Envelope-deficient HIV-1 NL4-3 clone encoding IRES-GFP ( NL4-3-GFP ) was pseudotyped with VSV-G by cotransfection of HEK293T cells using polyethylenimine ( Polysciences ) as previously described [11 , 23] . For the production of viral-like particles carrying Vpx ( VLPVpx ) , HEK293T cells were cotransfected with pSIV3+ and a VSV-G expressing plasmid . Three days after transfection , supernatants were harvested , filtered and stored at -80°C . Viral stocks were concentrated using Lenti-X concentrator ( Clontech ) . Viruses were titrated by infection of TZM cells followed by GFP quantification by flow cytometry . R5-tropic HIV-1 strain BaL was grown in stimulated PBMC and specifically titrated for its use in assays of total viral DNA formation in MDM . M-CSF or GM-CSF differentiated MDM were infected with VSV-pseudotyped NL4-3-GFP and antiviral drugs were added at the time of infection . When necessary , differentiated MDM were pretreated with VLPVpx for 4h before infection or left with fresh media as a control . Viral replication was measured in all cases two days later by flow cytometry ( LSRII , BD Biosciences ) . Measurement of cell cytotoxicity was performed by flow cytometry , i . e . , cells were gated as living or dead , according to flow cytometry FSC and SSC parameters . BaL infections were stopped at 16h to measure only early events of viral infection ( reverse transcription ) . For quantification of proviral DNA , a primer and probe set that is able to amplify both unintegrated and integrated viral DNA was used as described before [11 , 16] . DNA was extracted using a DNA extraction kit ( Qiagen ) and proviral DNA quantifications were performed . Ct values for proviral DNA were normalized using RNaseP as housekeeping gene by the ΔΔCt method . Infections were normalized to an untreated control . To ensure that measured proviral DNA was the product of infection and not result from DNA contamination of the viral stocks samples treated with RT inhibitor AZT ( 1 μM ) were run in parallel . raltegravir ( 2 μM ) was used to ensure that no post-RT steps were being quantified by the assay . Cells were rinsed in ice-cold phosphate-buffered saline ( PBS ) and extracts prepared in lysis buffer ( 50 mM Tris HCl pH 7 . 5 , 1 mM EDTA , 1 mM EGTA , 1 mM Na3VO4 , 10 mM Na β-glycerophosphate , 50 mM NaF , 5 mM Na Pyrophosphate , 270 mM sucrose and 1% Triton X-100 ) supplemented with protease inhibitor ( Roche ) and 1 mM phenylmethylsulfonyl fluoride . Lysates were subjected to SDS-PAGE and transferred to a PVDF membrane ( ImmunolonP , Thermo ) . The following antibodies were used for immunoblotting: anti-rabbit and anti-mouse horseradish peroxidase-conjugated secondary antibodies ( 1:5000; Pierce ) ; anti-human Hsp90 ( 1:1000; 610418 , BD Biosciences ) , anti-SAMHD1 ( 1:1000; ab67820 , Abcam ) , anti-CDK1 ( 9116 ) anti-CDK2 ( 2546 ) , anti-phosphoCDK2 ( Thr160; 2561 ) , anti-CDK4 ( D9G3E ) , anti-CDK6 ( 3136 ) , anti-cyclin A2 ( BF683 ) , anti-cyclin D2 ( D52F9 ) , anti-cyclin D3 ( DCS22 ) , anti-p21 ( 2947 ) and anti-p27 ( 2552 ) all 1:1000 from Cell Signaling . Anti-phospho-SAMHD1 Thr592 was obtained by immunization of rabbit using a phosphorylated peptide as described before [53] . HEK293T cells were transfected with Flag-tagged p21 ( Addgene plasmid # 16240 , gift from Mien-Chie Hung ) [29] or HA-tagged Cyclin D2 ( Addgene plasmid # 8950 , gift from Philip Hinds ) [30] expression vectors using lipofectamine 2000 ( Invitrogen ) . 48 h later , cells were chilled to 4°C and cell extracts prepared with lysis buffer as described above . Lysates were cleared by centrifugation at 10500 rpm for 10 min and incubated with anti-FLAG ( anti-Flag M2 Affinity Gel , Sigma ) or anti-HA ( monoclonal anti-HA-agarose , Sigma ) antibodies covalently attached to agarose overnight at 4°C on a rocking platform . Beads were then collected by centrifugation at 3000 rpm for 5 min at 4°C , extensively washed in lysis buffer and resuspended in SDS gel loading buffer . The proteins were separated on a 10% SDS-polyacrylamide gel , transferred to a PVDF membrane , and analyzed by immunoblotting with the corresponding antibodies . Co-immunoprecipitation of endogenously expressed proteins was performed using GM-CSF differentiated macrophages . Cell extracts were prepared as above and lysates were incubated with anti-Cyclin D2 antibody ( D52F9 , Cell Signaling ) , anti-p21 antibody ( 2947 , Cell Signalling ) or rabbit IgG overnight at 4°C and further incubated with Fast flow Sepharose ( Sigma-Aldrich ) for 1-2h . Beads were then collected by centrifugation at 3000 rpm for 5 min at 4°C , extensively washed in lysis buffer and resuspended in SDS gel loading buffer . The proteins were separated on a 10% SDS-polyacrylamide gel , transferred to a PVDF membrane , and analyzed by immunoblotting with the corresponding antibodies . Data were analyzed with the PRISM statistical package . If not stated otherwise , all data were normally distributed and expressed as mean ± SD . p-values were calculated using an unpaired , two-tailed , t-student test . | Macrophages are a heterogeneous population of immune cells that provide crucial innate immune defense against pathogens , including HIV-1 . The molecular biology of HIV-1 infection in macrophages is influenced by the presence of the host cell restriction factor SAMHD1 , which is regulated by phosphorylation by cyclin dependent kinases , the catalytic proteins responsible for cell cycle progression . This study shows that differentiation stimuli strongly influence macrophage cell cycle and proliferation characteristics as well as susceptibility to HIV-1 infection through modulation of SAMHD1 activation . We have identified cyclin D2 as the key step controlling susceptibility to HIV-1 infection by modulation of the signaling pathway leading to SAMHD1 phosphorylation . We show that a complex formed by cyclin D2-CDK4-p21 in GM-CSF macrophages is responsible for the lack of the active CDK , which phosphorylates SAMHD1 . This situation is reversed in the absence of cyclin D2 , leading to the activation of CDKs and subsequent phosphorylation of its substrates , including SAMHD1 . Thus , we propose that the differential expression of the G1/S-specific cyclin D2 controls the HIV-1 restriction pathway in primary macrophages . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"blood",
"cells",
"phosphorylation",
"medicine",
"and",
"health",
"sciences",
"cell",
"cycle",
"inhibitors",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"cell",
"cycle",
"and",
"cell",
"division",
"pathogens",
"immunology",
"cell",
"processes",
"microbiology",
"cell",
"differentiation",
"retroviruses",
"viruses",
"immunodeficiency",
"viruses",
"developmental",
"biology",
"rna",
"viruses",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"medical",
"microbiology",
"hiv",
"gene",
"expression",
"microbial",
"pathogens",
"hiv-1",
"biochemistry",
"cell",
"biology",
"post-translational",
"modification",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"lentivirus",
"organisms",
"cyclins"
] | 2016 | The G1/S Specific Cyclin D2 Is a Regulator of HIV-1 Restriction in Non-proliferating Cells |
Pseudomonas syringae pv . phaseolicola is the causative agent of halo blight in the common bean , Phaseolus vulgaris . P . syringae pv . phaseolicola race 4 strain 1302A contains the avirulence gene avrPphB ( syn . hopAR1 ) , which resides on PPHGI-1 , a 106 kb genomic island . Loss of PPHGI-1 from P . syringae pv . phaseolicola 1302A following exposure to the hypersensitive resistance response ( HR ) leads to the evolution of strains with altered virulence . Here we have used fluorescent protein reporter systems to gain insight into the mobility of PPHGI-1 . Confocal imaging of dual-labelled P . syringae pv . phaseolicola 1302A strain , F532 ( dsRFP in chromosome and eGFP in PPHGI-1 ) , revealed loss of PPHGI-1::eGFP encoded fluorescence during plant infection and when grown in vitro on extracted leaf apoplastic fluids . Fluorescence-activated cell sorting ( FACS ) of fluorescent and non-fluorescent PPHGI-1::eGFP F532 populations showed that cells lost fluorescence not only when the GI was deleted , but also when it had excised and was present as a circular episome . In addition to reduced expression of eGFP , quantitative PCR on sub-populations separated by FACS showed that transcription of other genes on PPHGI-1 ( avrPphB and xerC ) was also greatly reduced in F532 cells harbouring the excised PPHGI-1::eGFP episome . Our results show how virulence determinants located on mobile pathogenicity islands may be hidden from detection by host surveillance systems through the suppression of gene expression in the episomal state .
The exchange of genomic pathogenicity islands by horizontal gene transfer is recognised as a key factor in the development of new more virulent bacterial pathogens of both animals and plants [1] , [2] , [3] , [4] . Pseudomonas syringae is a bacterial species sub-divided into pathovars based on host range . Infection of a wide variety of plants results in necrotic symptoms in leaves , stems , and fruit [5] . P . syringae pv . phaseolicola causes halo-blight of common bean and has emerged as a model bacterial pathogen for the analysis of the evolution of pathogenicity [6] , [7] , [8] , [9] . The primary mechanism of plant resistance against P . syringae is a basal defense response that is induced upon detection of conserved microbe associated molecular patterns ( MAMPS ) [10] . Pathovars of P . syringae are able to overcome MAMP triggered defense by delivering into the plant cytoplasm an array of effector proteins that inactivate surveillance mechanisms and signal transduction pathways , thereby allowing bacterial multiplication [11] , [12] , [13] . Many plants have evolved resistance proteins that recognise a subset of these effector proteins , termed avirulence ( Avr ) proteins , which trigger a hypersensitive resistance response ( HR ) leading to programmed cell death at infection sites and the restriction of colonisation [14] , [10] . Molecular analysis of genetic interactions between P . syringae pv . phaseolicola and its host bean has led to the identification and characterization of a number of bacterial avr genes and plant R genes [7] . One of these avr genes , avrPphB ( syn . hopAR1 ) , encodes an effector protein that induces an HR in bean cultivar ( cv . ) Tendergreen ( TG ) which carries the R3 resistance gene . In P . syringae pv . phaseolicola strain 1302A , avrPphB is located on a 106 kb genomic island ( GI ) designated PPHGI-1 [6] . The similarity of PPHGI-1 to other integrative and conjugative elements ( ICElands ) suggests that the island integrates and excises at att sites within the tRNA locus found at its borders , via an episomal circular intermediate [15] , [16] . PPHGI-1 can be acquired by transformation between strains of P . syringae pv . phaseolicola [17] and is physically lost from cells of P . syringae pv . phaseolicola during infection of TG undergoing the R3-AvrPphB mediated HR [18] , [6] . Following excision from the chromosome , a critical step in the mobilization of PPHGI-1 is formation of the circular episome that is capable of limited replication [17] , [19] . The plant apoplast comprises the intercellular space surrounding plant cells where metabolic and physiological processes relating to cell wall biosynthesis , nutrient transport , and stress responses occur [20] . It is also the site of colonisation for P . syringae pv . phaseolicola which obtains nutrients directly from apoplastic fluid for in planta survival and multiplication . We have previously shown that complete PPHGI-1 transfer can occur between P . syringae pv . phaseolicola 1302A and P . syringae pv . phaseolicola 1448A , both in the leaf apoplast and in extracted P . vulgaris TG and Canadian Wonder ( CW ) apoplastic fluids [17] . We hypothesized that such a transfer requires four distinct processes: excision of the island from the chromosome , release of the circular episome from the bacterium , relocation into competent bacterial cells and finally integration into a specific att site within the genome . As GI movement is responsible for major evolution in P . syringae pv . phaseolicola [21] , [8] , understanding the stages involved in GI transfer between bacterial strains will greatly aid our understanding of the evolutionary process . Although we know that exposure to the plant's immune response activates excision and transfer of the PPHGI-1 , there is no information available on the dynamics of island loss from colonies of bacteria within infected tissues . In order to address the spatial dynamics of excision in the context of the emergence of new strains of P . syringae pv . phaseolicola at microsites within infected tissues , we developed fluorescent protein-based systems to monitor the movement of PPHGI-1 in and out of the genome . In addition , using a combination of fluorescence-activated cell sorting and quantitative PCR , we found that gene expression from the excised PPHGI-1 episome was reduced so that absence of fluorescence did not reflect loss of the GI , but its excision and circularization . Switching off gene expression after excision of pathogenicity islands may be a widespread phenomenon allowing bacterial products to be hidden from host defenses and thereby facilitating “stealthy” transfer of genes encoding virulence factors that may be of benefit under new infection conditions .
We aimed to visualise PPHGI-1 dynamics in the plant and in vitro using constitutively expressed fluorescent markers . Strains of P . syringae pv . phaseolicola 1302A were generated expressing fluorescent proteins encoded from PPHGI-1 ( e . g . PPHGI-1::eGFP ) and also from other regions of the chromosome as illustrated in Figure 1A . P . syringae pv . phaseolicola F341 contained chromosomal eYFP and PPHGI-1::eCFP ( Figure 1B ) and P . syringae pv . phaseolicola F532 chromosomal dsRFP and PPHGI-1::eGFP ( Figure 1C ) . Loss of PPHGI-1 was expected to lead to loss of the GI encoded fluorescence but retention of emission from the other fluorophore ( Figure 1D , E ) . P . syringae pv . phaseolicola F341 and P . syringae pv . phaseolicola F532 also displayed wildtype P . syringae pv . phaseolicola 1302A phenotypes; for example in vitro and in planta growth rates , plasmid profiles , and equal rates of PPHGI-1 loss when passaged through TG ( data not shown ) . We have previously described the use of bacteria separately labeled with fluorophores for confocal microscopy of infected leaves [22] . With the strains constructed to express two fluorophores , we planned to follow loss of PPHGI-1 occurring at microsites within challenged leaves and in particular during the HR in TG . Colonies of P . syringae pv . phaseolicola develop in the intercellular spaces , typically attached to leaf mesophyll cells . A feature of tissue undergoing the HR in TG was the dispersal of colonies to reveal individual bacterial cells as illustrated in Fig . 2A . When P . syringae pv . phaseolicola F532 was infiltrated into TG or CW leaves , colonies frequently showed peripheral loss of PPHGI-1::eGFP fluorescence ( Figure 2B ) . Dispersed bacterial cells of F532 in TG and around the cut edges of leaf samples in both cultivars also displayed mixed fluorescence ( Fig . 2C ) . The loss of fluorescence from colonies was more easily detected in the susceptible variety CW , rather than resistant TG , because imaging whole colonies after the HR was partly compromised by the accumulation of autofluorescent plant-derived metabolites ( Figure 2D , [22] ) . Limiting initial P . syringae pv . phaseolicola inoculum concentrations in TG was effective in minimizing the spread of the HR and was necessary to avoid spectral overlap of dsRFP and HR autofluorescence . Our observations with F532 highlighted that the outer layer ( s ) of bacteria within established colonies ( in both TG and CW leaves ) preferentially exhibit loss of PPHGI-1::eGFP fluorescence . Such colony differentiation may be due to the activities of diffusion mediated gradients similar to those proposed in biofilm models [23] , where the colony periphery receives higher concentrations of diffused substrates ( e . g . nutrients , oxygen , environmental stimuli , and/or antimicrobial compounds ) and therefore exhibits differing metabolic activity compared with the colony centre [24] . Higher rates of in planta loss of PPHGI-1::eGFP fluorescence from the periphery of F532 colonies may be due to higher levels of environmental cue ( s ) from the P . vulgaris apoplast and/or due to a higher metabolic activity due to enhanced nutrient acquisition . Significantly , loss of PPHGI-1 derived fluorescence appeared to be as frequent in the susceptible CW leaves as in the resistant TG which underwent the HR . By contrast , previous work had shown that cells that have lost PPHGI-1 are more readily selected during the HR than during the development of infection in susceptible tissues such as CW leaves [6] , [19] . Lovell et al . [17] demonstrated that apoplastic fluids extracted from leaves provide a medium that promotes the mobility of PPHGI-1 . To determine if PPHGI-1::eGFP fluorescence was lost from F532 in vitro , the strain was grown on apoplastic fluids extracted from leaves of CW and TG , mixed with agarose and coated on microscope slides to facilitate microscopy . Fluorescence was lost rapidly from colonies grown on apoplastic fluids but not from M9-based media ( Figure 3 ) . Increasing the concentration of either CW or TG apoplastic fluid in the agarose slide medium caused a corresponding increase in the frequency of loss of fluorescence due to PPHGI-1::eGFP ( Figure 3A ) . To confirm that loss of PPHGI-1 encoded fluorescence was not specific to strain F532 , P . syringae pv . phaseolicola F341 was examined in the same way and rates of loss of PPHGI-1::eCFP were comparable to those found for PPHGI::eGFP in F532 ( data not shown ) . Clearly , some component ( s ) of apoplastic fluid from either cultivar promoted loss of fluorescence of respectively labeled PPHGI-1 constructs , suggesting that the occurrence of the HR is not required for the release of inducing metabolites from plant cells . To determine whether or not changes in PPHGI-1::eGFP fluorescence were due to cellular loss of PPHGI::eGFP , F532 cells were first grown on M9/TG agarose slides . Following confocal imaging to confirm loss of PPHGI-1::eGFP fluorescence from some cells but positive dsRFP fluorescence , bacteria from the imaged agarose slide were resuspended in ¼ Ringers and cultured to individual colonies on KB agar . All resulting colonies were found to express eGFP . Furthermore , 200 colonies were analysed for the presence of avrPphB by the TG pod-stab pathogenicity assay and all colonies were found to cause the HR . This suggested that , despite the temporary loss of expression of the PPHGI-1::eGFP construct after growth on apoplastic fluids , PPHGI-1 was retained and functional . To investigate the apparent silencing of eGFP , fluorescence-activated cell sorting ( FACS ) was employed to obtain pure populations of F532 with or without eGFP fluorescence; F532/GFP+ and F532/GFP- , respectively ( Figure 4 ) . We first tested the suitability of FACs to differentiate the fluorescent Pseudomonads . Low background was recorded in 1302A without added fluorophores ( Fig . 4A ) but this did not compromise the strong signal from eGFP ( Fig . 4B ) . To collect eGFP fluorescing and non-fluorescent bacteria , F532 cultures were grown ( 48 h ) in stationary M9/TG apoplastic fluid ( 1∶1 ) broths and 200 , 000 particles ( events ) were collected for both F532/GFP+ and F532/GFP- cells . Sub-populations were serially diluted and grown on KB+Gm and the viable cells recovered were: F532/GFP+ ( 2 . 64×104 CFU/ml ) and F532/GFP- ( 1 . 06×104 CFU/ml ) . The decrease in viable CFU/ml from F532/GFP- events compared to F532/GFP+ was likely due to detection of non-fluorescent particles , from either non-viable F532 cells and/or plant debris from apoplastic fluid . The separated sub-populations of F532/GFP- and F532/GFP+ cells were monitored for further changes in PPHGI-1::eGFP fluorescence over a 48 h period ( in LB , static at 25°C , Figs . 4C–F ) . If PPHGI-1::eGFP excision and re-insertion into the chromosome was a dynamic process , we expected to see a reversion of the F532/GFP- ( Figure 4 C , M1 ) sample to form a major population of F532/GFP+ cells , with a minor population of F532/GFP- cells . As predicted , F532/GFP- cells were observed to revert to a population of predominantly F532/GFP+ cells ( Figure 4D , E ) over a 48 h period . Also , F532/GFP+ cells from gated population M2 ( Figure 4C ) formed a sub-population of F532/GFP- cells ( M1-like , Figure 4F ) indicating the continued mobility of PPHGI-1::eGFP in F532 . It is interesting that after 48 h both FACS sub-populations reverted back to the same general population structure ( as in Figures 4E and 4F ) suggesting that there may be an equilibrium between F532 cells containing PPHGI-1 in the chromosome ( expressing eGFP ) and F532 cells containing the excised PPHGI-1 ( not expressing eGFP ) . Figure 4G illustrates the changes in patterns of sorted fluorescent particles comparing F532 grown in apoplastic fluid which produces a broader GFP detection spectrum than F532 grown in LB broth . This result correlates with different levels of PPHGI-1::eGFP fluorescence ( whilst dsRFP remained constant ) observed in F532 cells examined by confocal microscopy ( e . g . Figure 3C ) . We had shown that loss of PPHGI-1::eGFP fluorescence was often not due to the deletion of PPHGI-1 from F532 . Therefore , we investigated whether down-regulation of gene expression occurs after GI excision and circularization , thereby accounting for loss of PPHGI-1::eGFP fluorescence . qPCR was performed on FACS separated F532 populations to determine differences between F532/GFP- and F532/GFP+ cells by comparison of: a ) the amount of PPHGI-1 forming a circular PPHGI-1 episome ( CE ) ; and b ) the expression levels of two genes contained within PPHGI-1 ( avrPphB and xerC ) . Results reveal that there is increased formation of CE in F532/GFP- cells , but greatly reduced expression of avrPphB and xerC ( Figure 5 ) . We conclude that loss of fluorescence from F532 does not , therefore , usually indicate loss of PPHGI-1 , but is in fact reporting creation of the excised CE from which gene expression is greatly reduced . Such reduced expression of GI genes could be explained by transcriptional and translational down-regulation due to supercoiling of DNA ( Figures 1D & E ) [25] , [26] and/or possible association of histone-like proteins influencing gene expression [27] , [28] . It is well established that PPHGI-1 is lost from 1302A cells during the HR in bean leaves . The antimicrobial conditions generated by the HR appear to select for bacteria lacking the avirulence gene avrPphB [6] , [21] , [19] . The first stage in deletion of PPHGI-1 is its excision from the chromosome and the formation of the CE that was first detected by PCR in bacteria grown in LB [6] , [21] , [19] . Here we applied the more stringent method of qPCR to confirm the presence of CE in F532/GFP- cells ( Fig . 5 ) . The high frequency of occurrence of bacteria down-regulating gene expression on PPHGI-1 detected in vitro in apoplastic fluids suggested that bacteria harbouring the excised episome might be far more common in certain environments . We therefore used qPCR to examine the levels of PPHGI-1 CE in bacteria grown under various conditions for 5 h ( Fig . 6 ) . The experiment revealed strikingly higher levels of CE in bacteria inoculated into bean leaves than grown in LB or minimal media . At the 5 h time point , however , there was no difference in CE concentration in resistant TG , or susceptible CW leaves . As expected , CE was present at high levels in bacteria incubated in apoplastic fluids . The highest amounts were found after growth in fluids recovered from leaves undergoing the HR and the lowest from susceptible , diseased tissue . The addition of whole leaf extracts to M9 also created conditions favouring formation of the CE . Taken together these results suggest that excision of PPHGI-1 is promoted by a plant factor ( s ) but there is little differentiation between susceptible and resistant plant tissues before infection . The occurrence of the HR does appear to enhance excision and formation of the circular intermediate , but not to such an extent that would explain the high frequency of loss of PPHGI-1 during the HR [6] . We conclude that P . syringae pv . phaseolicola PPHGI-1 excision , circularisation and re-insertion is a very dynamic process necessary to maintain PPHGI-1 replication within P . syringae pv . phaseolicola populations . This would explain the variance in detectable levels of PPHGI-1::eGFP fluorescence from adjacent cells during confocal imaging ( e . g . Figure 3C ) where images obtained are most likely ‘snapshots’ of F532 cells at different stages of PPHGI-1::eGFP excision . It is highly likely that the excision of PPHGI-1 may interfere with the timeline of eGFP transcription , translation , protein folding , maturation and degradation – all ultimately affecting the levels of PPHGI-1::eGFP fluorescence . The reduced expression of genes on the CE , notably avrPphB , would provide a selective advantage for the bacterial cells allowing their increased multiplication in resistant TG tissues and favouring subsequent selection of bacteria that had lost PPHGI-1 . Alternatively , the multiplication of bacteria maintaining the silenced PPHGI-1 may prove advantageous should they be spread to other plants lacking the R3 resistance gene . The re-integration of PPHGI-1 into the chromosome would lead to a return to normal levels of expression of genes encoded on the GI . In our previous work on the mobility of PPHGI-1 we used failure to induce the HR as an indication of loss of the island and confirmed this result by PCR-based diagnostics [6] . Our results on the suppression of gene expression from the circularised episome show that there is a complex pattern of selection occurring within the infected plant . Because we have demonstrated movement of the island both out of and back into the chromosome , there must be a resultant fluctuation in levels of expression of avrPphB . It would only be those bacteria that have physically lost PPHGI-1 that would retain virulence during colonisation and presumably break out from sites undergoing the effector-triggered HR . In order to dissect the dynamics further we need to investigate the mechanisms controlling deletion of the island in more detail . The initial aim of our work was to use confocal microscopy to examine PPHGI-1 deletion at micro-sites within infected leaves . We were only partially successful because of the suppression of gene expression from the excised episome . However , our findings may be of more general significance to our understanding of microbial pathogenicity . Dynamic excision and reinsertion of GIs has been described in other Pseudomonas species [29] , [30] . We suggest that the silencing of genes carried on GIs following excision from their chromosomal location may be an important strategy utilized not only by P . syringae pv . phaseolicola but also by other bacterial pathogens of plants and animals . Switching off gene expression after GI excision is a novel mechanism with an enormous biological potential . It may , for example , represent a new way to modulate gene expression to the pathogen's advantage and to facilitate GI transfer . The microbes may have evolved a means to facilitate the “stealthy” transfer of genes encoding virulence factors that may be of benefit under new infection conditions
Bacterial strains and plasmids used in this study are listed in Table 1 . Escherichia coli strains were grown at 37°C in Luria Bertani ( LB , Difco ) media and Pseudomonas strains were grown at 25°C on Kings medium B ( KB ) [31] , in LB broth or M9 minimal medium ( M9 , [32] ) . Antibiotics were used at the following concentrations ( µg/ml ) : gentamicin ( Gm ) 10 , kanamycin ( Km ) 50 , ampicillin ( Ap ) 100 , nitrofurantoin ( NF ) 100 , and rifampicin ( Rif ) 100 . Plasmid DNA was isolated from a pure bacterial culture using the QIAprep Spin Miniprep Kit ( Qiagen ) and restriction enzymatic digests performed as per manufacturers protocols ( NEB biolabs ) . Unless stated otherwise , standard PCR reactions were performed using Phusion TM polymerase ( NEB biolabs ) using recommended cycling parameters with oligonucleotide primers ( Table S1 ) . Automated DNA sequencing was performed with an ABI 3130xl genetic analyser . P . syringae pv . phaseolicola plasmid profiles were determined by extracting total uncut plasmid DNA from overnight cultures as described previously [33] . Strand overlapping extension PCR [34] was used to introduce the rare cutting restriction enzyme sites PmeI and PacI into a non-coding-region ( NCR; 55667 bp – 56577 bp , Genbank accession AJ870974 ) of DNA within PPHGI-1 for introduction of eGFP or eCFP ( Figure 1A ) . Three transcriptional terminators ( TTs ) in each translational frame were introduced upstream of PmeI and downstream of PacI ( Figure 1A ) . Overlapped PCR products were cloned into pK18mobsacB [35] to create pSG028-1 . Both eGFP and eCFP were PCR amplified from AKN100 and AKN033 respectively using primers ( F-pEXFP-PmeI and R-pEXP-PacI , Table S1 ) that were designed to introduced PmeI and PacI restriction sites respectively at amplicon terminal ends . PCR products were cloned into Zero-Blunt-Topo ( Invitrogen ) to create pSG097 ( eGFP ) and pSG103 ( eCFP ) . PmeI and PacI digests of pSG097 and pSG103 were respectively cloned into similarly digested pSG028-1 to create pSG111-eGFP and pSG114-eCFP . Introduction of the eGFP ( pSG111-eGFP ) and eCFP ( pSG114-eCFP ) into the NCR of PPHGI-1 was achieved by allelic exchange [36] with P . syringae pv . phaseolicola 1302A to create KmR first cross-over construct strains , P . syringae pv . phaseolicola SG120 ( eGFP ) and P . syringae pv . phaseolicola SG126 ( eCFP ) . P . syringae pv . phaseolicola SG120 and P . syringae pv . phaseolicola SG126 were confirmed to have P . syringae pv . phaseolicola 1302A wild type phenotypes as described previously [6] . dsRFP ( AKN132 , Table 1 ) and eYFP ( AKN069 ) were introduced respectively into P . syringae pv . phaseolicola SG120 and P . syringae pv . phaseolicola SG126 via the Tn7 transposon delivery system kindly provided by Lambertsen et al . [36] as described previously [22] to create P . syringae pv . phaseolicola F341 ( P . syringae pv . phaseolicola 1302A with chromosomal eYFP and PPHGI-1::eCFP , Figure 1B ) and P . syringae pv . phaseolicola F532 ( P . syringae pv . phaseolicola 1302A with chromosomal dsRFP and PPHGI-1::eGFP , Figure 1C ) . P . syringae pv . phaseolicola inoculations and in planta confocal imaging of P . syringae pv . phaseolicola fluorescence in P . vulgaris was performed essentially as described previously [22] . Confocal visualisation of in planta infiltration of dual-labelled strains ( F532 and F341 ) was performed on the Leica TCS-SP2-DM IRE2 confocal laser scanning microscope ( Leica Microsystems Wetzlar GmbH ) at 25× or 40× ( objective magnification ) for colony morphology and 63× or 100× for visualisation of individual cell dispersal . Variable AOTF filters were used for the following fluorophores ( excitation/emission ) : eYFP ( 514 nm/525–600 nm ) ; eCFP ( 440 nm/465–495 nm ) ; eGFP ( 488 nm/516–539 nm ) ; dsRFP ( 568 nm/600–644 nm ) ; and for plant autofluorescence ( 440 nm , 650–785 nm ) . Z-series imaging was performed at intervals of 0 . 3 µm ( individual cells ) and 1 µm ( colonies ) . All confocal images were assigned false colour in images ( eCFP cyan , eGFP green , eYFP yellow , dsRFP red and plant tissue blue ) . 1 µm z-section scan intervals were used to ensure all bacterial fluorescence was analysed . Three 5 mm2 leaf samples were analysed from each variable and each sample had at least 12 random areas confocal imaged to ensure representative data . Glass slides were layered with approximately 0 . 17 mm of agarose ( 2 . 2% final volume ) containing desired liquid ( ddH2O , M9 , or apoplastic fluid ) and used immediately . Apoplastic fluid was extracted from 10 day old TG or CW bean leaves , infiltrated CW leaves undergoing disease or TG leaves undergoing the HR as described previously [17] . All apoplastic fluids were used immediately after preparation following filter-sterilization ( 0 . 2 µm ) . Overnight cultures were washed and serially diluted in ¼ Ringers solution to OD600 equivalent of: 1×10−2 , 10−4 , 10−6 , or 10−8 . Inoculum droplets ( 10 µl ) were placed onto respective agarose slides and incubated within a sterile sealed container ( containing high humidity ) at 25°C for 24 h–144 h as required . The optimal micro-colony development was usually observed at 1×10−4 OD600 after 24 h on M9/TG ( 1:1 ) apoplastic fluid . Cover slips were added immediately prior to confocal imaging using the Zeiss Axiovert 200 in conjunction with the Ultraview FRET H rapid confocal imaging system ( Perkin Elmer Instruments Ltd ) as described previously [22] . Slide preparations of P . syringae pv . phaseolicola F532 were inoculated ( 1×10−4 OD600 ) onto M9/TG slides for 48 h . Agarose slides were imaged using both dsRFP and eGFP channels to ensure P . syringae pv . phaseolicola F532 cells had loss of PPHGI-1::eGFP fluorescence ( such as Fig . 3Av ) . Agarose from the imaged slide was immediately re-suspended into ¼ Ringers by Eppendorf-pestle homogenization and vortex mixing . Extracts were serially diluted onto KB agar and incubated ( 25°C , 48 h ) . Resulting GmR colonies were replica plated onto KB vs . KB+Gm ( to select for Tn7-dsRFP ) and KB+Gm+Km ( to select for Tn7-dsRFP and PPHGI-1::eGFP ) . After 48 h at 25°C , all colonies were visualised for eGFP fluorescence and subjected to TG pod-stab assay ( to determine the presence of avrPphB , and therefore PPHGI-1 ) by analysis of HR/disease phenotype as described previously [6] P . syringae pv . phaseolicola F532 was grown in broth cultures consisting of: M9/TG apoplastic fluid ( 1:1 ) . 200 , 000 particles ( events ) were collected for both F532/GFP+ and F532/GFP- and samples were serially diluted and plated onto KB+Gm to correlate FACS events with viable P . syringae pv . phaseolicola cell recovery . FACS was performed on either ( 1 ) a FACSVantage cell sorter ( Becton Dickinson ) , equipped with a 488 nm argon laser; or ( 2 ) an Influx cell sorter ( Becton Dickinson ) . P . syringae pv . phaseolicola F532 cultures were separated based on GFP fluorescence ( F532/GFP+ ) and non-GFP fluorescence ( F532/GFP- ) . Both F532/GFP+ and F532/GFP- FACS sub-populations were analysed immediately using confocal microscopy to confirm status of eGFP and dsRFP fluorescence . qPCR was used to quantify xerC and avrPphB expression and PPHGI-1 CE production in vitro and in planta . P . syringae pv . phaseolicola strains were added to respective growth medium ( either LB , M9 , apoplastic fluid and/or M9/apoplastic fluid mixes ) and after 5 h , cells were harvested and gene expression stopped using RNA protect reagent ( Qiagen , UK ) or DNA lysis solution ( Gentra Systems , UK ) . For quantification of xerC , avrPphB and gyrB expression , RNA was extracted using the RNAeasy kit ( Qiagen , UK ) followed by a second DNase step of 15 min at 37°C ( Promega , UK ) . cDNA was synthesised using the TaqMan Reverse Transcription Reagents kit ( Applied Biosystems ) . For quantification of the PPHGI-1 CE , DNA was extracted using the Puregene DNA isolation kit ( Gentra Systems , UK ) . Triplicate samples were taken for each variable for cDNA/DNA synthesis , and each sample was analyzed twice in separate amplifications . cDNA and DNA were quantified using a Nanodrop spectrophotometer ( Thermo Scientific , USA ) and adjusted to 100 ng/µl . When gene expression was analysed after FACS separation , cells were sorted directly into either RNA protect reagent ( Qiagen UK ) or DNA lysis solution ( Gentra Systems , UK ) and RNA/DNA extracted as above . qPCR was performed on an ABI 7300 Real-Time PCR System ( Applied Biosystems ) , calibrated using 7300 Real-Time PCR Systems Spectral Calibration kit ( Applied Biosystems ) and probes ( Table S1 ) were labelled with 3′ FAM and 5′ TAMRA TaqMan dyes . Reaction volume ( 25 µl ) consisted of 12 . 5 µl TaqMan PCR mastermix ( Applied Biosystems ) , 2 µl each primer ( 10 µM ) , 2 µl probe ( 5 µM ) and 6 . 5 µl RNase free water . Standard qPCR cycling conditions were 50°C for 2 min , 95°C for 10 min and 40× cycles of 95°C for 15 sec followed by 60°C for 1 min . Results were analysed using ABI 7300 System SDS software ( Applied Biosystems ) and compared using JMP IN 7 . 0 statistical analysis software ( www . jmpin . com ) . Average abundance of xerC and avrPphB RNA and CE DNA were expressed relative to levels of gyrB . | Bacterial pathogens evolve rapidly through the transfer of large segments , or genomic islands ( GIs ) , of DNA . We study the mobility of an island named PPHGI-1 in Pseudomonas syringae pv . phaseolicola that causes halo-blight disease of bean . The exposure of P . syringae pv . phaseolicola to plant defenses triggers the excision of PPHGI-1 , creation of a circular episomal form and finally deletion of the GI or its transfer to other bacteria . We planned to examine deletion of PPHGI-1 within infected leaves , and we generated strains that expressed differently coloured fluorescent proteins from genes in the island or elsewhere on the chromosome . Loss of the specific fluorescence derived from the GI was expected to show deletion of PPHGI-1 . However , collecting fluorescent and non-fluorescent bacteria showed that PPHGI-1 was usually not lost , but expressed its component genes very poorly when in the circularized state . Bacteria were therefore able to carry a hidden suite of genes that become activated when re-inserted into the chromosome . The “stealthy” movement of the island is beneficial to P . syringae pv . phaseolicola because genes on PPHGI-1 encode proteins that activate plant defenses . Similar gene silencing on episomes may occur in other pathogens and contribute to the evolution of microbial pathogenicity to animals and plants . | [
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"plant",
"science",
"microbial",
"evolution",
"plant",
"pathogens",
"plant",
"pathology",
"plant",
"biology",
"biology",
"microbiology"
] | 2011 | The Stealth Episome: Suppression of Gene Expression on the Excised Genomic Island PPHGI-1 from Pseudomonas syringae pv. phaseolicola |
Human cytomegalovirus ( HCMV ) US2 , US3 , US6 and US11 act in concert to prevent immune recognition of virally infected cells by CD8+ T-lymphocytes through downregulation of MHC class I molecules ( MHC-I ) . Here we show that US2 function goes far beyond MHC-I degradation . A systematic proteomic study using Plasma Membrane Profiling revealed US2 was unique in downregulating additional cellular targets , including: five distinct integrin α-chains , CD112 , the interleukin-12 receptor , PTPRJ and thrombomodulin . US2 recruited the cellular E3 ligase TRC8 to direct the proteasomal degradation of all its targets , reminiscent of its degradation of MHC-I . Whereas integrin α-chains were selectively degraded , their integrin β1 binding partner accumulated in the ER . Consequently integrin signaling , cell adhesion and migration were strongly suppressed . US2 was necessary and sufficient for degradation of the majority of its substrates , but remarkably , the HCMV NK cell evasion function UL141 requisitioned US2 to enhance downregulation of the NK cell ligand CD112 . UL141 retained CD112 in the ER from where US2 promoted its TRC8-dependent retrotranslocation and degradation . These findings redefine US2 as a multifunctional degradation hub which , through recruitment of the cellular E3 ligase TRC8 , modulates diverse immune pathways involved in antigen presentation , NK cell activation , migration and coagulation; and highlight US2’s impact on HCMV pathogenesis .
HCMV is the prototype betaherpesvirus and an important human pathogen . Following primary infection , HCMV persists for the lifetime of the host under constant control by the host immune system . In the face of this selective pressure , HCMV has evolved multiple mechanisms to evade immune detection and has emerged as a paradigm of viral immune modulation and evasion . Experimentally , only 45 of the ~170 canonical HCMV protein coding genes are required for in vitro replication [1 , 2]; most HCMV genes appear to be directed at promoting virus persistence through targeting host defenses [3–5] . Four genes clustered in the HCMV unique short ( US ) gene region use independent mechanisms to suppress MHC-I dependent antigen presentation to CD8+ cytotoxic T lymphocytes [6] . US3 is an immediate early gene product that binds and retains newly synthesized MHC-I proteins in the endoplasmic reticulum ( ER ) and blocks tapasin-dependent peptide loading [7 , 8] , whereas US6 inhibits TAP-mediated peptide translocation into the ER [9 , 10] . US2 and US11 both bind MHC-I in the lumen of the ER and hijack the mammalian ER-associated degradation ( ERAD ) machinery to promote retrotranslocation to the cytosol for proteasome degradation [11 , 12] . US2 and US11 appropriate distinct cellular ERAD pathways for MHC I dislocation . US2 utilizes the cellular E3 ligase TRC8 ( translocation in renal cancer from chromosome 8 ) to ubiquitinate and subsequently degrade MHC-I [13] , whereas US11 uses a Derlin-1-associated ERAD complex centered around the newly characterized TMEM129 E3 ligase [14–16] . Functionally US2 and US11 are distinct as US11 has a combined ER retention and degradation function [13 , 15 , 17] , while US2 is unable to retain MHC-I in the ER prior to degradation , but relies on US3 for enhanced degradation . In addition to downregulating MHC-I , US2 and US3 also target the MHC-II antigen presentation pathway [18 , 19] . US3 retains MHC-II molecules in the ER while US2 initiates the retrotranslocation of MHC-II DR-α chain and DM-α chain from the ER back to the cytosol for subsequent degradation . Since endogenous MHC-I molecules constitute the chief ligands recognized by NK cell inhibitory receptors , their downregulation has the potential to render cells more vulnerable to NK cell attack . To compensate , HCMV encodes its own MHC-I homologue ( UL18 ) and a peptide present in the UL40 signal sequence acts to stabilize and maintain cell surface expression of the NK inhibitory ligand HLA-E [20–24] . Moreover , HCMV systematically suppresses cell surface expression of ligands for NK cell activating receptors . The HCMV glycoprotein UL141 plays a major role in such protection via interaction with TRAIL death receptors , as well as CD155 ( PVR , necl5 ) and CD112 ( PVRL2 , nectin-2 ) which are both ligands for the ubiquitous NK activating receptor DNAM1 [25–27] . In isolation , UL141 is capable of suppressing both CD155 and TRAIL-R2 cell surface expression , but an additional HCMV-encoded factor is known to be required for efficient downregulation of cell surface CD112 [25] . Moreover , while CD155 and TRAIL-R2 accumulate in the ER during the course of infection , CD112 is degraded [25] . While US2 , US3 , US6 and US11 were originally defined by their capacity to inhibit cell surface MHC-I expression , classical MHC molecules are not necessarily their only targets . To gain an unbiased view of cellular receptors whose expression is altered upon viral gene expression , we recently developed ‘Plasma Membrane Profiling’ ( PMP ) , a SILAC ( Stable Isotope Labelling of Amino acids in Culture ) -based quantitative proteomics technique which compares the relative abundance of cell surface receptors between infected and uninfected cells , and therefore identifies the range of cell surface proteins downregulated upon viral infection [28–30] . PMP demonstrated that whereas US3 , US6 and US11 specifically downregulate MHC-I , US2 targets a series of novel substrates including the NK cell ligand CD112 , the anti-coagulation factor thrombomodulin and at least six integrin family members , abolishing integrin signalling , cell adhesion and migration . While US2 alone is necessary and sufficient to target most substrates , effective downregulation of CD112 requires a synergistic interaction between US2 and UL141 . UL141 retains CD112 in the ER and associates with US2 resulting in CD112 dislocation across the ER membrane for proteasome degradation . We therefore propose a role for US2 that is much broader than previously appreciated , but nevertheless depends on the common activity of TRC8 for impact . Furthermore , US2 and UL141 form a multifunctional and highly adaptive degradation hub with a substrate range much wider than previously appreciated , affecting cellular processes as broad as antigen presentation , NK cell killing , cell migration and coagulation .
To determine whether the HCMV-encoded US2 , 3 , 6 and 11 viral gene products downregulate cell surface proteins in addition to MHC molecules , we used Plasma Membrane Profiling ( PMP ) , an unbiased , proteomic technique which compares the relative abundance of plasma membrane proteins [28] . Plasma membrane proteins were isolated from THP-1 monocytic cells stably expressing HCMV US2 , US3 , US6 or US11 by sequential cell surface glycoprotein biotinylation followed by streptavidin pull-down and their relative expression was quantified by high through-put mass spectrometry . To minimise differences in sample preparation , cells were metabolically labelled prior to biotinylation by SILAC , which allows early stage sample mixing without loss of sample identity [31] . The relative abundance of plasma membrane proteins in US2 , US3 , US6 or US11 expressing cells versus control is plotted in Fig 1A with individual proteins represented by single dots . Those proteins whose expression is unaltered by viral gene expression accumulate in the centre , whereas left and right shifts represent proteins down- or up-regulated at the plasma membrane of viral gene expressing cells . In US3 , US6 and US11 expressing cells , the majority of plasma membrane proteins identified ( 370–454 ) were unchanged , with HLA-A , B and C allotypes of MHC-I molecules being the predominant proteins lost from the cell surface ( Fig 1A; left and top right panels ) . US3 also showed a decrease in MHC-II and C1q complement receptor expression . In contrast , US2 altered the expression of a multitude of cell surface receptors , with thirteen new proteins showing more than a four-fold downregulation ( Fig 1A bottom right panel , S1 Table ) . In addition to MHC-I , six integrin family members were downregulated: α1 ( ITGA1 , 4 . 9 fold ) , α2 ( ITGA2 , 10 . 7 fold ) , α4 ( ITGA4 , 15 . 3 fold ) , α5 ( ITGA5 , 3 . 9 fold ) , α7 ( ITGA7 , 4 . 3 fold ) and β1 ( ITGB1 , 5 . 2 fold ) . Other substrates downregulated by US2 include thrombomodulin ( THBD , 31 . 9 fold ) , protein tyrosine phosphatase , receptor type , J ( PTPRJ , 4 . 9 fold ) and the interleukin-12 receptor β1 ( IL12RB1 , 8 . 9 fold ) . We focused on novel US2 substrates and confirmed their downregulation by flow cytometry . Indeed US2 expressing THP-1 cells showed a robust downregulation of integrins α1 , α2 , α4 , β1 , thrombomodulin , PTPRJ and IL12 receptor β1 ( Fig 1B , grey line ) , compared to control ( black line ) . Other cell surface molecules , including the transferrin receptor and integrin αV , remained unaffected by US2 and none of the US2 substrates were affected by US11 expression ( S1 Fig ) , thus confirming the specificity of substrate down-regulation . AXL , integrin αM ( ITGAM ) and αL ( ITGAL ) expression was dysregulated by lentiviral transduction and not followed further . US2 is a type I membrane protein that co-opts the cellular ERAD degradation machinery to degrade MHC-I [12] . The ER-resident ubiquitin E3 ligase TRC8 is a critical component of the US2-mediated MHC-I degradation pathway [13] . US2 recruitment of TRC8 is required for the ubiquitination and subsequent degradation of newly synthesized MHC-I , and in the absence of this ligase MHC-I is rescued back to the cell surface [13] . In a similar manner , shRNA knock-down of TRC8 in US2-expressing THP-1 cells rescued cell surface expression of integrins , THBD , PTPRJ and IL-12Rβ1 ( Figs 1B dashed line and S2 ) . The recruitment of TRC8 thus appears to be a common and essential step in the US2 pathway . To further examine the fate of these novel US2 substrates , we focused first on the integrin family . Integrins consist of an / chain heterodimer . All integrin alpha chains downregulated by US2 share the common beta-1 chain as their binding partner ( Fig 1C ) . Integrins require α/β dimerisation prior to transport to the cell surface and loss of either may result in ER retention of the other subunit . It was therefore important to test whether the various α subunits or the β1 chain itself constitute the primary US2 substrate . On immunoblots integrin 4 , 5 and 6 expression was strongly decreased in US2-expressing THP-1 cells ( Figs 2A lane 3 and S3 ) , suggesting not only down-regulation from the cell surface but US2-dependent degradation . Indeed integrin expression was rescued by proteasome inhibition ( Fig 2B lane 7 ) or shRNA-mediated depletion of the TRC8 E3 ligase ( Fig 2A and 2B lane 4 ) . US11-expressing cells showed no change in integrin expression ( Fig 2A and 2B lane 2 ) . Unexpectedly , the integrin 1 chain was not itself degraded in US2-expressing cells , but accumulated in its faster migrating ER-resident immature form ( Fig 2A lane 3 ) . TRC8 depletion rescued this immature species to its mature form , while expression of the control integrin 3 was unaffected by US2 ( Fig 2A lanes 3 and 4 ) . We used [35S]-methionine radiolabeling and pulse-chase analysis to further examine how US2 affects 4 and 1 integrin maturation . Integrin 4 was rapidly degraded in US2-expressing cells with a marked reduction in its half-life from more than 1 hour to less than 15 minutes ( Fig 2C ) , which was prevented by the use of proteasome inhibitors ( Fig 2D ) . In contrast , in the presence of US2 , the 1 integrin was neither degraded , nor was it able to mature to its higher molecular species but remained in the ER in its immature form throughout the course of the 3 hour chase ( Fig 2E ) . Our data suggest that α integrins are direct substrates for the US2/TRC8 pathway of proteasomal degradation , whereas ER retention of the β1 integrin is likely secondary to degradation of its α integrin interaction partners . US2 rapidly degrades its target proteins , making it difficult to ascertain whether they physically interact with US2 . A truncated US2 mutant ( US2ΔC' ) , from which the cytoplasmic tail was deleted ( aa 186–199 ) is reported to be functionally inactive , but can still bind its MHC-I substrate [32] , providing a useful tool to probe US2 interactions . We initially tested whether the US2 cytosolic domain is responsible for TRC8 recruitment , which would explain the US2ΔC' loss of function . While wild-type US2 readily binds TRC8 [13] , this association is lost in the US2ΔC' mutant ( Fig 2F lanes 5 and 6 ) , explaining the loss of function phenotype . Furthermore , the US2ΔC' mutant is now found associated with , but unable to degrade integrin α4 ( Fig 2G lanes 3 and 6 ) , suggesting that US2 binds its α integrin substrate prior to recruitment of TRC8 . Ubiquitination by TRC8 triggers the US2-induced retrotranslocation and degradation of MHC-I [13] . We therefore examined whether integrin α4 is also ubiquitinated in the presence of US2 . Immune precipitation of radiolabelled integrin α4 visualized a smear of ubiquitinated species in the presence but not absence of US2 ( Fig 2H ) . These species were recovered upon denaturation and re-precipitation of integrin α4 as well as ubiquitin . Integrin α4 is thus ubiquitinated in a US2-dependent manner triggering its proteasomal degradation . To address the functional consequence of US2-mediated integrin downregulation , we focused on the signaling properties of the most dramatically down-regulated integrin , α4β1 . The importance of the α4β1 integrin in embryogenesis and disease pathogenesis derives from its role in cell adhesion and cell migration [33 , 34] . Binding of integrin α4β1 to its ligands fibronectin or vascular cell adhesion molecule-1 ( VCAM-1 ) initiates focal adhesion complex assembly and phosphorylation of paxillin [35 , 36] . This phosphorylation event is specific to integrin β1 and α4 tails , and is not stimulated by other α-integrins [33–35] . Fibronectin stimulation of vector-only and US11-transduced control cells led to the expected increase in paxillin phosphorylation ( Fig 3A , top panel , lanes 5 and 7 ) , while US2 inhibited phosphorylation of paxillin as effectively as shRNA-induced depletion of the β1 integrin ( lanes 6 and 8 ) . Since the α4β1 integrin is required for macrophage chemotaxis , we examined how US2 affects cell adhesion and migration . Adhesion of US2-expressing THP-1 cells to a fibronectin substrate was completely abolished ( Fig 3B ) and migration of these cells was significantly reduced ( Fig 3C ) compared to the empty vector and US11 controls . A similar phenotype was observed in β1 integrin depleted cells . Thus US2-induced downregulation of integrin α4β1 has dramatic functional consequences , inhibiting downstream integrin-mediated signalling , cell adhesion and migration . US2 expression alone is both necessary and sufficient for downregulation and TRC8-dependent degradation of the novel substrates . We next sought to investigate US2 function in the context of a productive HCMV infection and , in an experiment complementary to US2 single gene expression , evaluated the effect of deleting US2 from the HCMV genome using PMP . Permissive human foreskin fibroblasts ( HFF ) were infected with the HCMV strain Merlin ( HCMV wt ) or a US2 deletion virus ( HCMV ΔUS2 ) , SILAC labeled and plasma membrane proteins were isolated and quantified . Proteins whose relative abundance is higher in cells infected with HCMV ΔUS2 compared to those infected with wt-HCMV require US2 for their downregulation ( left shift in Fig 4A ) . Of 686 plasma membrane proteins identified from HCMV-infected HFFs , four integrin family members ( α2 ( 3 . 6 fold ) , α4 ( 3 . 2 fold ) , α6 ( 7 . 7 fold ) , and β4 ( 6 . 4 fold ) ( Fig 4A and S2 and S3 Tables ) as well as PTPRJ were significantly downregulated in a US2-dependent manner . In addition , other immunoglobulin superfamily members that had not been identified in THP-1 cells also required US2 for their downregulation . These included immunoglobulin superfamily member 8 ( IGSF8 ) and epithelial cell adhesion molecule ( EPCAM ) as well as butyrophilin subfamily 2 member A1 ( BTN2A1 ) , cadherin-4 ( CDH4 ) and endothelin-converting enzyme 1 ( ECE1 ) ( Fig 4A and S3 Table ) . To validate PMP results , flow cytometry was performed on HFFs infected with either an HCMV variant encoding a UL32-GFP fusion protein or the same virus additionally deleted for the US1-11 region . Gating on GFP+ HCMV-infected cells confirmed that the US1-11 region was required to downregulate MHC-I , integrin α4 and THBD ( Fig 4B ) . In a further test with the specific HCMV ΔUS2 mutant used for PMP , infected cells were distinguished by their downregulation of MHC-I . MHC-I downregulation is unaffected with the HCMV ΔUS2 mutant due to redundancy with the US3/6/11 gene products ( S2 Table ) . By gating on MHC-Ilo HCMV-infected cells we confirmed that downregulation of integrin α4 and THBD was specifically dependent on US2 ( Fig 4C ) . Thus , we confirmed that many of the key US2 targets in THP-1 cells were also downregulated in HFFs in the context of HCMV infection . HCMV infected myeloid cells are thought to play a key role in the spreading of virus in vivo . Effective mechanisms of immune evasion will be important for enabling viral reactivation in the presence of a primed immune system . We therefore tested whether the US2-induced substrate downregulation observed in whole HCMV-infected HFFs could be replicated in differentiated THP-1 cells , and specifically how US2-induced integrin downregulation affects cell adhesion . PMA-activated THP-1 cells were infected with the endothelial-tropic HCMV strain TB40 harboring a UL32-GFP marker [37] . Since a ΔUS2 mutant of this strain was not available , we inactivated US2 function by a stable TRC8 knockdown , prior to TB40 infection . Comparing control to TRC8 knock-down , we observe a striking US2/TRC8-dependent downregulation ( Fig 5A ) and degradation ( Fig 5B lanes 3 and 4 ) of integrins α2 , α4 , α6 and thrombomodulin in HCMV-infected cells . This downregulation is therefore similar to that observed following HFF infection with the HCMV Merlin strain ( Fig 4A , 4B , and 4C ) . Integrin α6 , which is not detected in basal THP-1 cells ( Fig 5B lanes 1 and 2 ) , was markedly induced upon viral infection and concomitantly downregulated via the US2/TRC8 pathway ( Fig 5B lanes 3 and 4 ) , suggesting a potent anti-viral role counteracted by US2 . A similar upregulation upon virus infection and subsequent downregulation by US2 was observed for thrombomodulin in HFF cells ( Fig 4B ) . To assess the functional consequence of US2-induced integrin downregulation in HCMV infected cells , TB40-infected THP-1 cells were allowed to adhere to fibronectin , recombinant VCAM-1 , collagen or uncoated tissue culture wells . TB40-infected THP-1 cells with a control knock-down showed significantly decreased binding to both VCAM-1 and collagen , but not uncoated wells , compared to TRC8 knock-down cells ( Fig 5C ) . VCAM-1 and collagen are respectively substrates for the US2-targetted integrins α4 and α2 . Binding to fibronectin—a less specific substrate for integrin α4—was unaffected by TRC8 knock-down . This lack of an effect is likely due to up-regulation of integrin αV in PMA-activated THP-1 cells; this β3-associated integrin binds readily to fibronectin but is not a US2 substrate . In conclusion we show α integrins including α2 , α4 and α6 are specific US2 substrates that are degraded in a TRC8-dependent manner upon both US2 single gene expression and whole HCMV infection , thereby reducing cell adhesion of myeloid cells to a variety of substrates . HCMV UL141 is a powerful NK cell evasion gene that downregulates NK cell ligands CD112 , CD155 and the death receptor TRAIL-R2 from the cell surface [25 , 26 , 38] . Our previous data indicated that UL141 requires an additional unmapped HCMV function to efficiently downregulate CD112 [25] . The PMP study revealed that US2 was also able to alter CD112 expression , both independently and in the context of HCMV infection ( Figs 1A and 4A and S1 Table ) , so we further analyzed and compared specific UL141 and US2 deletion mutants . Proteomic analysis of HFF cells infected with a HCMV UL141 deletion mutant versus wild-type HCMV confirmed UL141-dependent down-regulation of the known UL141 targets: CD155 , TRAIL-R2 and CD112 . TRAILR4 was identified as a novel UL141 target ( Fig 6A ) . While expression of the US2 gene alone caused only a modest downregulation of CD112 ( 2 . 4 fold ) ( Fig 1A and S1 Table ) , in the context of whole virus infection , the observed robust CD112 downregulation was clearly dependent on both US2 and UL141 ( Figs 4A and 6A and S3 Table ) . A further PMP experiment using a dual ΔUS2ΔUL141 HCMV deletion mutant ( Fig 6B and S3 Table ) confirmed many of the changes we observed using single gene deletion viruses , and showed that CD112 downregulation was most efficient in the presence of both viral genes ( Fig 6A , 6B , and 6C; S3 Table ) . These results suggest a requirement for both US2 and UL141 for effective CD112 downregulation . UL141 is a predominantly ER-resident viral protein which , in contrast to US2 , does not actively promote proteolysis of CD155 or TRAIL-R2 but sequesters them in the ER , thus preventing their further trafficking through the secretory pathway [25 , 26 , 38] . We hypothesized that UL141 might retain CD112 in the ER , and promote its transfer to US2 for TRC8-dependent ubiquitination and subsequent degradation . Whereas UL141 or US2 alone caused only a partial cell surface down-regulation of CD112 within the viral context ( Fig 6D; Merlin ΔUS2–3 . 6x; Merlin ΔUL141–2 . 8x ) , their combined action showed a more than additive effect ( wt Merlin -13 . 3x ) , suggesting synergy between the two immune evasion genes . Cooperativity was also observed between TRC8 and UL141 . Cell surface expression of CD112 was partially rescued by TRC8 depletion of cells infected with wild-type HCMV or HCMVΔUL141 , but not the HCMVΔUS2 deletion mutant ( Fig 6C ) . TRC8-dependent downregulation of CD112 is thus dependent on US2 within the viral context . Two versions of CD112 are produced by differential splicing; the short α and long δ variants can be distinguished using antibodies specific for their cytosolic tails [39] ( S4 Fig ) . Following wild-type HCMV infection , both CD112 α and δ forms were degraded ( Fig 7A , lane 3 vs . lane 1 ) . However , TRC8 depletion , or the absence of US2 ( HCMV ΔUS2 ) , rescued the CD112 δ form in both its mature Endo H resistant form ( Fig 7A , CD112 δ blots , lane 11 vs . 12 and 11 vs . 13 , upper bands ) and its immature , ER resident , Endo H sensitive forms ( lower bands ) . A similar pattern was seen with the CD112 α isoform , which was degraded by wild-type HCMV and restored specifically in its immature form by a HCMV US2 deletion mutant or TRC8 depletion ( Fig 7A CD112 α blots , lane 3 vs . 4 and 3 vs . 5 ) . The mature form of both the α and δ isoform was only fully restored upon combined UL141 deletion ( HCMV ΔUL141 ) and TRC8 depletion ( Fig 7A lane 3 vs . 8 and 11 vs . 16 ) . Therefore , the ability of UL141 to retain both CD112 isoforms in the ER was only revealed in the absence of US2 or following TRC8 depletion which abrogates US2-induced CD112 degradation . In contrast , CD155 downmodulation was solely dependent on UL141 for retention in the ER , and integrin α4 required only US2 in order to be degraded during virus infection ( Fig 7A , CD155 and integrin α4 blots ) . Together the data indicate that US2 and UL141 co-operate to prevent CD112 cell surface expression . Indeed UL141 and US2 appear capable of interacting , as evidenced by US2 co-immunoprecipitation with UL141 ( Fig 7B , lane 8 ) . UL141 also directly associated with its substrate CD112 ( Fig 7B , lane 7 ) . This interaction is lost in the presence of US2 due to CD112 degradation , but was rescued following TRC8 depletion ( Fig 7B , lanes 8 and 9 ) . UL141 co-precipitated US2 in both the presence and absence of TRC8 indicating UL141 itself is not degraded by US2 ( Fig 7B , lanes 8 and 9 ) . The control viral protein US11 is not found in association with UL141 . Collectively our results provide a remarkable example of cooperativity between two unrelated viral proteins with diverse functions . While the retention of CD112 in the ER by UL141 is inefficient and can easily be overcome , teaming up with US2 promoted CD112 degradation via the TRC8-dependent pathway and provides an efficient mechanism of controlling expression of this cellular protein .
This study demonstrates the power of Plasma Membrane Profiling ( PMP ) as an unbiased approach to establish a global picture of how individual viral genes modulate the cell surface proteome . Specific antibodies have traditionally been used to determine the changes in cell surface proteins upon viral infection . This straightforward approach has proven particularly useful in tracking changes in expression of critical immune effector cell ligands ( e . g . MHC-I ) during the course of an infection . By design , this candidate approach is inevitably selective and cannot , therefore , provide a complete picture of the effect of virus-encoded immunomodulatory functions on the cell . By deploying PMP we were able to demonstrate the precision with which US3 , US6 and US11 specifically target MHC molecules , whereas US2 was revealed to be a pleotropic modulator of cell surface receptors whose function extends beyond T cell evasion to impact on NK cell function , cell adhesion , signaling and coagulation ( Fig 8 ) . We identified many novel cellular substrates that require US2 for their downregulation . For all targets examined , US2-mediated degradation was dependent on the TRC8 E3 ligase , indicating that HCMV-mediated appropriation of this cellular ubiquitin ligase provides a common pathway for the ER-associated degradation of US2 substrates . While US2 alone is sufficient for the downregulation of the majority of new targets , effective removal of the NK cell ligand CD112 requires co-operation between UL141 , to retain CD112 in the ER , and US2 to initiate CD112 degradation . The recruitment of US2 by UL141 greatly enhances the efficiency with which CD112 is downregulated and the US2-dependant degradative pathway provides a potential conduit by which host proteins retained in the ER can be targeted for degradation , and may be exploited by other HCMV proteins . US3 , like UL141 , retains MHC molecules in the ER from where they are degraded by the US2/TRC8 complex [17 , 18] . Whereas US3 is an HCMV immediate early gene with expression peaking at 8 hours post-infection [40] , UL141 reaches maximum expression at 4–5 days post-infection [26] , suggesting US2 might change substrate specificity during the HCMV life cycle . Reliance on different viral retention factors might therefore enhance the flexibility of the US2/TRC8 degradation hub which may therefore be customized towards specific requirements at different stages of the viral life cycle . Antigen presentation may be an acute problem early in viral infection , requiring US3 , while NK cell killing , and the requirement for UL141 , becomes critical as MHC-I levels on the cell surface decline . While the US2/TRC8 hub induces degradation of the UL141-substrate CD112 , this mechanism is not deployed against UL141’s other targets: CD155 and TRAIL-R2 [26] . Since these three main cellular targets of UL141 are all implicated in distinct intracellular signaling pathways , there may be additional benefit to the virus in retaining CD155 and TRAIL-R2 in an intracellular compartment while targeting CD112 to the proteasome . Alternatively binding to CD155 and/or TRAIL-R2 might be incompatible with US2 binding to UL141 . Recent structural analysis suggests the Ig-like domain of UL141 is a structural mimic of TIGIT thus allowing CD155 binding [41] , whereas the interaction between UL141 and TRAIL-R2 involves a separate , non-canonical death receptor interaction site . [42] . As it is unusual for an immunomodulator to selectively target multiple unrelated receptors , further understanding of UL141 function may highlight important aspects of the evolution of immune recognition and modulation . It will be of particular interest to gain further structural insight into the CD112-UL141-US2 complex . The largest group of novel targets downregulated by US2 were the integrins , a large family of 18 α and 8 β chains that assemble into 24 different heterodimers . HCMV-induced integrin down-regulation was first reported for integrin α1β1 [43] , and we here show that US2 down-regulates a variety of integrins including α1 , α2 , α4 , α5 , α6 , α7 , β1 and β4 . αβ-heterodimer assembly is a prerequisite for integrin maturation and transport to the cell surface [44–46] . Our data suggest that α integrins are targeted by US2 for TRC8-induced ubiquitination and proteasomal degradation , whereas the β1 subunit is retained in the ER due to the absence of its α integrin binding partner . Whether each α chain is individually recognized by US2 and degraded , or US2 binds the shared β1 integrin , which is itself protected , but leads to the degradation of any associated α chain , remains unclear . Our preliminary experiments favour the former scenario , as alpha chains were still degraded in cells depleted of β1 integrin , suggesting that alpha chains are indeed direct targets of US2 . However , we cannot exclude that , despite an effective depletion , any remaining β1 integrin , might still target alpha chains for degradation . Furthermore , no common motif in α integrins targeted by US2 has been identified . It is even less clear how US2 recognizes the broad range of substrates ( MHC-I , integrins , thrombomodulin , the IL-12 receptor β1 subunit ) that share no apparent structural features . Integrins mediate cellular attachment to a wide range of extracellular proteins , and control multiple cellular functions , including morphology , migration and differentiation [47] . US2-induced α integrin degradation is predicted to have a broad impact on the physiology of HCMV infected cells . In this perspective integrin α6 is of interest as it is induced following HCMV infection and concomitantly downregulated by US2 , a pattern reminiscent of antiviral proteins . Indeed , integrin α6 is essential for dendritic cell ( DC ) migration across the laminin and collagen IV rich basement membranes to reach the draining lymph node for antigen presentation [48 , 49] . Integrin α6 specific blocking antibodies inhibit DC migration to lymph nodes [48] , and US2-induced integrin degradation should therefore counteract infection-induced DC migration . This might be particularly relevant during viral reactivation from latency when the immune system is already primed to eradicate early stage infection . Additional US2-targetted integrins may also inhibit DC migration: integrin α1 and α2 are receptors for the basement membrane component collagen IV , while integrin α4 and α5 are likely required for DC reverse migration across the endothelial cell layer into the lymph [50] . Furthermore , the α4 integrin is essential for leukocyte transendothelial migration from blood into peripheral tissue and plays a prominent role in immune surveillance [47] . Indeed a monoclonal antibody targeting α4β1 and α4β7 is in clinical use for the treatment of autoimmune diseases , and reduces inflammation by preventing leukocyte extravasation into the tissue [51 , 52] . The downregulation of α4β1 by US2 may thus prevent circulating virus-infected myeloid cells from responding to chemoattractants and homing . Cell migration remains an understudied area of viral immune evasion . A global comparison of plasma membrane proteins altered upon HCMV infection showed that cell surface proteins involved in adhesion and migration are a major target for HCMV [30] , and additional HCMV genes likely contribute to HCMV’s modulation of cell migration . In addition to integrins , downregulation of VCAM-1 , at least eight protocadherins , five plexins and two ephrins was seen [30] . Furthermore , the actin cytoskeleton of HCMV-infected cells is heavily reorganized by UL135 which hijacks the WAVE2 complex and prevents the formation of focal adhesions [53] . Furthermore , in latent HCMV infection , UL138 likely inhibits DC migration via degradation of the multidrug transporter MRP1 which is essential for leukotriene C4 ( LTC4 ) secretion [29 , 54] . Although of potential benefit to viral immune evasion , US2-mediated degradation of cell surface receptors could also potentially contribute to the pathophysiology associated with congenital HCMV infection . Several integrins targeted by US2 ( α1 , α4 , α5 , α6 and β1 ) are required for placentation or fetal development [47] . Integrin α4 deficiency for example is embryonically lethal in mice due to cardiac defects and defective placentation [55] . Depending on the cell types infected in pregnancy , US2’s ability to downregulate integrin family members could contribute to the fetal damage associated with HCMV infection . Indeed , HCMV-induced downregulation of integrin α1β1 has been associated with impaired cytotrophoblast invasion and placentation [56] . The endothelial cell surface protein thrombomodulin ( THBD ) is another US2 substrate that might contribute to HCMV-associated fetal defects . THBD alters thrombin’s substrate specificity from pro-coagulant and pro-inflammatory to anti-coagulant and anti-inflammatory [57] and its deficiency leads to lethal consumptive coagulopathy in embryonic blood vessel endothelium . Since , the endothelium is a common site of HCMV infection in vivo , US2-induced THBD degradation may contribute to both the coagulopathy and severe fetal thrombotic vasculopathy seen in congenital HCMV infection . THBD is also a marker for a human DC subset proficient in antigen cross-presentation , with similarities to mouse CD8+ DCs [58 , 59] . While the exact role of THBD in these DCs is unknown , THBD regulates substrate affinity of the TLR4 co-receptor CD14 [60] , suggesting a potential broader role in antiviral immunity , which may explain its induction in HCMV-infected cells and concomitant downregulation by US2 . Our study emphasizes the key role of US2 in combating different HCMV host defense pathways through the downregulation of multiple cell surface receptors . As we are particularly interested in HCMV evasion of the cellular immune response , this study focused on plasma membrane proteins . Additional US2 substrates may yet be identified by analysis of ER-resident proteins . Indeed , as shown for UL141 , other viral proteins which retain host receptors in the ER may also cooperate with US2 to degrade their cargo through the common pathway of TRC8-dependent degradation .
THP-1 cells and HFF cells ( System Bioscience ) were grown in RPMI-1640 or DMEM respectively ( PAA ) , with 10% heat-inactivated fetal bovine serum ( FCS; PAA ) and penicillin/streptomycin ( pen/strep , Sigma ) . The HCMV strain Merlin is designated the reference HCMV genome sequence by the National Center for Biotechnology Information [61] and is available as a BAC clone [62] . Merlin BAC derived clone Merlin wild-type used for this study contains point mutations in RL13 and UL128 , enhancing replication in fibroblasts [62] . Generation of virus recombinants and stocks was described previously [62] . All recombinants were validated by whole genome Illumina sequencing ( S4 Table ) . Merlin ΔUS2 has a deletion of the US2 ORF; Merlin ΔUL141 has deletions of the UL141 ORF; GFP tagged Merlin ΔUL16 , 18 ΔUS1-11 has deletions of the UL16 , UL18 , US1-11 ORFs and contains a UL32-GFP fusion; GFP tagged Merlin ΔUL16 , 18 ( Merlin delta UL16/UL18 , UL32-GFP ) was described previously [25] . HFF cells were infected with HCMV at indicated MOI for 72h . The GFP-tagged endothelial-tropic HCMV TB40 strain was originally created by Christian Sinzger ( University of Ulm , Germany ) [37] and a kind gift from Mark Wills ( University of Cambridge , UK ) . It contains an intact US1-11 region and a UL32-GFP fusion . For THP-1 infections , THP-1 cells were starved for 12-16h in RPMI with 2% FCS , activated for 48h with 100ng/ml PMA ( Sigma ) and infected with TB40 at MOI 25 . This resulted in infection of >95% of cells as estimated by GFP positivity and MHC-I down-regulation . Infected cells were harvested at 96h post-infection . A C-terminal myc-tagged UL141 ( HCMV Merlin strain ) and HA-tagged human integrin α4 were cloned into the pHRSIN lentivirus vector with a hygromycin B selection cassette . Untagged US2 , US2 with a deletion of the C-terminal cytoplasmic tail ( aa1-186; US2ΔC' ) , US3 , US6 and US11 were cloned into a pHRSIN lentivirus vector with an IRES CFP and puromycin selection cassette . N-terminal HA-tagged US2 ( HA-US2 ) was cloned into pHRSIN with a puromycin cassette only . The pHRSIN lentivirus expression system was used as described previously [63] . For shRNA-mediated knockdown of TRC8 and integrin β1 expression , hairpin oligonucleotides were designed as described [13 , 64] , annealed , cloned into the pHR-SIREN lentiviral vector ( a gift from Greg Towers , UCL , London ) . Lentivirus was produced as previously described in 293ET cells and used to transduce THP-1 cells . Primary antibodies used for flow cytometry were: mouse α-conformational MHC-I ( W6/32 ) , mouse α-conformational HLA-A2 ( BB7 . 2 ) , mouse α-integrin β1 ( Biolegend ) , mouse α-integrin α1 , ( BD ) , mouse α-integrin α2 , ( BD ) , mouse α-integrin α4 ( BD ) , rat α-integrin α6 ( Biolegend ) , rabbit α-integrin αV ( Santa Cruz ) , mouse α-thrombomodulin ( BD ) , mouse α-PTPRJ ( Medical & Biological Laboratories ) , mouse α-IL12 receptor β1 ( BD ) , mouse α-CD112 ( Santa Cruz ) , mouse α-CD155 ( Abcam ) and FITC-conjugated mouse α-transferrin receptor ( CD71; BD ) . AlexaFluor 647 conjugated goat anti-mouse ( Life Biosciences ) was used as a secondary antibody . Antibodies used for immunoblotting were: mouse α-MHC-I ( HC10 ) , rabbit α-calreticulin ( Thermo ) , mouse α-β-actin ( Sigma ) , mouse α-HA ( Sigma ) , rabbit α-integrin α 4 , α 5 , β1 , β3 ( Integrin antibody sampler kit; Cell Signaling ) , rabbit α-Integrin α6 ( Cell Signaling ) , goat α-CD112 ( R&D Systems ) , rabbit α- CD112 δ form ( Abcam ) , rabbit α-CD112 α form ( LifeSpan Biosciences ) , mouse α-CD155 [26] , mouse α-HCMV IE antigen ( Argene ) , mouse α-HCMV US2 ( a kind gift from Jack R . Bennink , NIH , US ) , rabbit α-HCMV US11 ( a kind gift from Emmanuel Wiertz , University of Utrecht , Netherlands ) , mouse α-HCMV UL141 [26] , mouse α-paxillin ( BD ) and rabbit α-phospho-paxillin Tyr118 ( Cell Signaling ) . Antibodies used for immune precipitation were: mouse α–HA and α-myc agarose affinity gel ( Sigma ) , mouse α-integrin β1 ( Biolegend ) and mouse α-ubiquitin ( FK1; Millipore ) in combination with protein A or protein G-conjugated sepharose ( Sigma ) . For siRNA gene depletion , cells were transfected using Oligofectamine ( Invitrogen ) at a final concentration of 40 nM and harvested at 96 h post transfection . The following siRNA oligonucleotides were used ( Dharmacon , ON-TARGET PLUS ) : CD112 siRNA-1 , 5’-GCGCUGAGCAGGUCAUCUU-3’; CD112 siRNA-2 , 5’-GCAUGAGAGCUUCGAGGAA-3’ . For immunoblots , cells were lysed in TBS ( pH7 . 4 ) with 1% NP40 , 0 . 1% SDS , 5mM IAA , 0 . 5mM PMSF ( Sigma ) and 1X complete protease inhibitor ( Roche ) for 30 min on ice . For immunoprecipitations , cells were lysed in 1% digitonin ( Calbiochem ) in TBS with 5mM IAA , 0 . 5mM PMSF ( Sigma ) and 1X complete protease inhibitor ( Roche ) for 30 min on ice . Lysates were cleared of cellular debris by centrifugation and pre-cleared using IgG-sepharose ( GE Healthcare ) . Individual proteins were immunoprecipitated using indicated antibodies in combination with Protein A or G sepharose ( Sigma ) , washed extensively and eluted in SDS reducing sample buffer . All samples were heated for 10 min at 50°C , separated by SDS/PAGE and transferred to PVDF membranes ( Millipore ) . Membranes were probed with the indicated antibodies , and reactive bands were visualized with Supersignal West Pico or West Dura ( Thermo Fisher Scientific ) . Cells were starved for 20 min in methionine-free , cysteine-free medium ( Sigma ) , labeled with [35S]methionine/[35S]cysteine ( Amersham ) for the indicated time and then chased in medium containing an excess of cold methionine and cysteine ( Sigma ) at 37°C . Samples taken at the indicated time points were lysed in 1% Triton X-100/TBS with 5mM IAA , 0 . 5mM PMSF ( Sigma ) and 1X complete protease inhibitor ( Roche ) for 30 min on ice . Immunoprecipitations were performed as above . For visualization of ubiquitinated integrin α4 , primary α-HA immune precipitations from 35S-labelled cells were eluted at 50oC for 15min in 50μl TBS , containing 1% SDS . Eluates were taken off the beads and after addition of 20μM DTT fully denatured at 70oC for 10min to dissociate interacting proteins . SDS was quenched by the addition of 1ml 1% Triton X-100/TBS with IAA/PMSF/protease inhibitor followed by secondary α-HA or α-ubiquitin immune precipitation . For cell adhesion assays , 96-well tissue culture plates were coated with 20 μg/ml fibronectin ( Sigma ) , 20 μg/ml recombinant VCAM1/Fc ( R&D Biosystems ) or 200ug/ml collagen ( Sigma ) in PBS , for 16 h at 4°C and blocked with 0 . 5% bovine serum albumin ( BSA ) in PBS for 2 h at 37°C to block nonspecific binding . THP-1 cells expressing single HCMV genes were starved 3 h in serum free RPMI prior to assays , seeded on fibronectin-coated and uncoated plates for 1 h at 37°C and washed 3 times with PBS . Adherent cells numbers were quantified by CyQUANT-NF ( Invitrogen ) according to the manufacturer’s protocol . HCMV TB40 infected THP-1 cells were harvested at 96h post-infection using enzyme-free cell dissociation buffer ( Life Biosciences ) , counted and re-seeded on fibronectin , VCAM1 , collagen or uncoated plates . Following 1 h ( fibronectin , VCAM1 , uncoated ) or 2 h incubation at 37°C , wells were washed 5 times with PBS , containing 0 . 5% BSA . Adherent cell numbers were quantified as above . For cell migration assays , Costar Transwell 24-well plates with 8 μm pore size ( Thermo ) were coated with 20 μg/ml fibronectin ( Sigma ) in PBS , for 16 h at 4°C . Cell migration was performed in RPMI with or without the chemotaxis agent MCP-1 ( Sigma ) at 10 ng/ml and 10% FBS in the lower chamber . After 6 h incubation at 37°C , cells migrated to the lower chamber were counted using a Neubauer counting chamber . All cell adhesion and migration assays were performed in triplicate and p-values were calculated using paired Student’s t-test based on independent experiments . Plasma membrane profiling was performed as described previously for THP-1 cells , with minor modifications for adherent HFFs [28 , 29] . Briefly , for THP-1 cells , 1 . 5 × 108 of each SILAC-labeled cell type were pooled in a 1:1 ratio . Labeling was as follows: ( first experiment , Fig 1A , top left and bottom right panel ) THP-control ( light label ) ; THP-US2 ( medium label ) ; THP-US11 ( heavy label ) . ( Second experiment , Fig 1A , top right and bottom left panel ) THP-US6 ( light label ) ; THP-control ( medium label ) ; THP-US3 ( heavy label ) . Surface sialic acid residues were oxidized with sodium meta-periodate ( Thermo ) then biotinylated with aminooxy-biotin ( Biotium ) . The reaction was quenched , and the biotinylated cells incubated in a 1% Triton X-100 lysis buffer . Biotinylated glycoproteins were enriched with high affinity streptavidin agarose beads ( Pierce ) and washed extensively . Captured protein was denatured with DTT , alkylated with iodoacetamide ( IAA , Sigma ) and digested with trypsin ( Promega ) on-bead overnight . Tryptic peptides were collected and fractionated ( described below ) . Glycopeptides were eluted using PNGase ( New England Biolabs ) then desalted using StageTips [65] . For HFF cells , one 150cm2 flask of HCMV-infected HFFFs per condition was washed twice with ice-cold PBS . Labeling was as follows: ( first experiment , Fig 4A ) HFF/Merlin-wt ( light label ) , HFF-Merlin ΔUS2 ( medium label ) . ( Second experiment , Fig 6A ) HFF/Merlin-wt ( medium label ) , HFF-Merlin ΔUL141 ( heavy label ) . ( third experiment , Fig 6B ) HFF/Merlin-wt ( light label ) , HFF-Merlin ΔUL141ΔUS2 ( medium label ) . Sialic acid residues were oxidized with sodium meta-periodate ( Thermo ) then biotinylated with aminooxy-biotin ( Biotium ) . The reaction was quenched , and the biotinylated cells scraped into 1% Triton X-100 lysis buffer . Biotinylated glycoproteins were enriched and digested as described above . HpRP-HPLC was performed on tryptic peptides as described previously [28] . 90% of each tryptic peptide sample was subjected to HpRP-HPLC fractionation using a Dionex Ultimate 3000 powered by an ICS-3000 SP pump with an Agilent ZORBAX Extend-C18 column ( 4 . 6 mm x 250 mm , 5 μm particle size ) . Peptides were resolved using a linear 40 min 0 . 1%-40% acetonitrile gradient at pH 10 . 5 . Eluting peptides were collected in 15s fractions . For THP-1 experiments , a total of 30 combined fractions were generated then dried using an Eppendorf Concentrator for LC-MSMS using a NanoAcquity uPLC ( Waters , MA , USA ) coupled to an LTQ-OrbiTrap XL ( Thermo , FL , UA ) . MS data was acquired between 300 and 2000 m/z at 60 , 000 fwhm with CID spectra acquired in the LTQ with MSMS switching operating in a top 6 DDA fashion . Fractionated HFF experiments were re-combined to give either 40 or 10 fractions . The 40 fraction experiments were acquired using the OrbiTrap XL as above . The 10 fraction experiments were acquired using a Q Exactive ( Thermo ) coupled to a RSLC nano3000 ( Thermo ) with MS data acquired between 400 and 1650 m/z at 75 , 000 fwhm with HCD fragment spectra acquired in a top 10 DDA fashion . Raw MS files were processed using MaxQuant version 1 . 3 . 0 . 5 . [66 , 67] . Data were searched against concatenated Uniprot human and HCMV databases , and common contaminants [67] . Fragment ion tolerance was set to 0 . 5 Da with a maximum of 2 missed tryptic cleavage sites . Carbamidomethyl cysteine was defined as a fixed modification , oxidised methionine , N-terminal acetylation and deamidation ( NQ ) were selected as variable modifications . Reversed decoy databases were used and the false discovery rate for both peptides and proteins were set at 0 . 01 . Protein quantitation utilised razor and unique peptides and required a minimum of 2 ratio counts and normalized protein ratios reported . Peptide re-quantify was enabled in all analyses . Summed intensity represents the sum of all heavy and light labelled peptide intensities for a given protein and was calculated by MaxQuant [68] . Significance B values were calculated and Gene Ontology Cellular Compartment ( GOCC ) terms added using Perseus version 1 . 2 . 0 . 16 ( downloaded from http://maxquant . org ) . Significance B identifies the significance of outlier protein ratios from the distribution of all ratios with a greater significance given to proteins with a high intensity [68] . Ratios were excluded for proteins identified by <2 unique peptides , or with a variability >150% . We assessed the number of PM proteins identified as described previously [28] . | As the largest human herpesvirus , HCMV is a paradigm of viral immune evasion and has evolved multiple mechanisms to evade immune detection and enable survival . The HCMV genes US2 , US3 , US6 and US11 promote virus persistence by their ability to downregulate cell surface MHC . We developed ‘Plasma Membrane Profiling’ ( PMP ) , an unbiased SILAC-based proteomics technique to ask whether MHC molecules are the only focus of these genes , or whether additional cellular immunoreceptors are also targeted . PMP compares the relative abundance of cell surface receptors between control and viral gene expressing cells . We found that whereas US3 , US6 and US11 were remarkably MHC specific , US2 modulated expression of a wide variety of cell surface immunoreceptors . US2-mediated proteasomal degradation of integrin α-chains blocked integrin signaling and suppressed cell adhesion and migration . All US2 substrates were degraded via the cellular E3 ligase TRC8 , and in a remarkable example of cooperativity between HCMV immune-evasins , UL141 requisitioned US2 to target the NK cell ligand CD112 for proteasomal degradation . HCMV US2 and UL141 are therefore modulators of multiple immune-related pathways and act as a multifunctional degradation hub that inhibits the migration , immune recognition and killing of HCMV-infected cells . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Plasma Membrane Profiling Defines an Expanded Class of Cell Surface Proteins Selectively Targeted for Degradation by HCMV US2 in Cooperation with UL141 |
Susceptibility loci identified by GWAS generally account for a limited fraction of heritability . Predictive models based on identified loci also have modest success in risk assessment and therefore are of limited practical use . Many methods have been developed to overcome these limitations by incorporating prior biological knowledge . However , most of the information utilized by these methods is at the level of genes , limiting analyses to variants that are in or proximate to coding regions . We propose a new method that integrates protein protein interaction ( PPI ) as well as expression quantitative trait loci ( eQTL ) data to identify sets of functionally related loci that are collectively associated with a trait of interest . We call such sets of loci “population covering locus sets” ( PoCos ) . The contributions of the proposed approach are three-fold: 1 ) We consider all possible genotype models for each locus , thereby enabling identification of combinatorial relationships between multiple loci . 2 ) We develop a framework for the integration of PPI and eQTL into a heterogenous network model , enabling efficient identification of functionally related variants that are associated with the disease . 3 ) We develop a novel method to integrate the genotypes of multiple loci in a PoCo into a representative genotype to be used in risk assessment . We test the proposed framework in the context of risk assessment for seven complex diseases , type 1 diabetes ( T1D ) , type 2 diabetes ( T2D ) , psoriasis ( PS ) , bipolar disorder ( BD ) , coronary artery disease ( CAD ) , hypertension ( HT ) , and multiple sclerosis ( MS ) . Our results show that the proposed method significantly outperforms individual variant based risk assessment models as well as the state-of-the-art polygenic score . We also show that incorporation of eQTL data improves the performance of identified POCOs in risk assessment . We also assess the biological relevance of PoCos for three diseases that have similar biological mechanisms and identify novel candidate genes . The resulting software is publicly available at http://compbio . case . edu/pocos/ .
Genome-wide association studies ( GWAS ) have a transformative effect on the search for genetic variants that are associated with complex traits , since they enable screening of hundreds of thousands of genomic variants for their association with traits of interest [1] . Recently published GWAS lead to the discovery of susceptibility loci for many complex diseases , including type 2 diabetes [2] , psoriasis [3] , multiple sclerosis [4] , and prostate cancer [5] . For improved identification of risk variants , researchers draw information from clinical , microarray , copy number , and single nucleotide polymorphism ( SNP ) data to build disease risk models , which are then used to predict an individual’s susceptibility to the disease of interest [6 , 7] . Several companies , such as deCODE genetics ( http://www . decode . com ) and 23andme ( https://www . 23andme . com ) have started using SNPs identified by GWAS , to provide personal genomic test services in the United States and health related genomic test services in Canada and the United Kingdom . An important problem with GWAS is that the identified variants account for little heritability [8 , 9] . However , empirical evidence from model organisms [10] and human studies [11] suggests that the interplay among multiple genetic variants contribute to complex traits . Epistasis among pairs of loci , i . e . , significantly improved association with the phenotype when two loci are considered together , is also shown to provide provide further insights into disease mechanisms [12–14] . Therefore , recent studies focus on identifying the interactions among pairs of genomic loci , as well as among multiple genomic loci [15–17] . These studies suggest that consideration of more than one locus together can better capture the relationship between genotype and phenotype . For this reason , genetic markers that aggregate multiple genomic loci can be used to design effective strategies for risk assessment and guide treatment decisions [18] . The Polygenic score is a commonly used method to identify the joint association of a large mass of the loci to predict disease risk [19] . The first application of polygenic score on GWAS data shows that the genetic risk for schizophrenia is a predictor of bipolar disorder [20] . There are also several studies demonstrating that polygenic risk score is a powerful tool in risk prediction [20–22] . However , polygenic score does not make use of prior biological knowledge , which may be useful in generating more robust features by incorporating the functional relationships among individual variants . Furthermore , according to a recent comparative assessment of various classification algorithms , there are no statistically significant differences between state-of-the-art classification algorithms in terms of performance in risk assessment [23] . This observation suggests that research on construction of features for risk assessment can be useful in improving the classification performance of these algorithms . Since detection of epistasis and higher order interactions is computationally expensive , many methods first assess the disease association of individual loci and then use functional knowledge to integrate these associations [24–26] . The key idea behind these methods is that functionally related variants , e . g . , those that induce dense subnetworks in protein-protein interaction ( PPI ) networks , can provide stronger statistical signals when they are considered together [27] . Based on similar insights , some researchers integrate GWAS with pathway information to identify statistically significant pathways that are associated with the disease [28 , 29] . Recently , Azencott et al . propose a method to discover sets of genomic loci that are associated with a phenotype while being connected in an underlying biological network [30] . They use an additive model to integrate the genotypes of loci and use connectivity patterns in the network to select a functionally coherent set of disease associated SNPs . While this method works on a network of genomic loci , the network is constructed based on the interactions among genes and mapping of loci to genes . For this reason , the application of these methods is limited to the variants in coding regions or in regions that are in close proximity to genes . However , 88 percent of genotyped variants in GWAS fall outside of coding regions [31] . Several risk variants are found in non-coding regions of the genome and it is shown that the functional effects of these variants are regulatory ( e . g . , mRNA expression , microRNA expression ) as opposed to directly influencing protein structure or function [32] . In this paper , we propose a new algorithm for the identification of multiple functionally related genomic variants that are collectively associated with a phenotype . The proposed method builds on the concept of “Population Covering Locus Sets” ( PoCos ) [33 , 34] . A PoCo is a set of loci that harbor at least one susceptibility allele in samples with the phenotype of interest . Here , we extend the notion of PoCos to enable adaptive identification of “susceptibility genotype” ( as opposed to susceptibility allele ) for each locus . We also develop a method for aggregating the genotypes of multiple loci in a PoCo to compute representative genotypes for use in risk assessment . Finally , in order to capture the functional relationship between genomic loci , we integrate GWAS data with human protein-protein interaction ( PPI ) network and regulatory interactions identified via expression quantitative trait loci ( eQTL ) . We use the PoCos identified by the proposed framework to construct features that can be used in risk assessment . We evaluate the performance of PoCos in risk assessment via cross-validation on seven GWAS case-control data sets obtained from the Wellcome Trust Case-Control Consortium ( WTCCC ) . We compare the risk assessment performance of models built using PoCos to that of models built using individual loci and polygenic score . Our experimental results show that PoCos significantly outperform individual loci and polygenic score in risk assessment . Furthermore , we assess the information added by the incorporation of PPI and eQTL and observe that inclusion of these data leads to more parsimonious models for risk assessment . In the next section , we describe the proposed procedure for modeling the genotypes and identifying PoCos . Then we describe how we use PoCos to develop a model for risk assessment . Subsequently , we present comprehensive experimental results on GWAS data sets for Type 2 Diabetes ( T2D ) , Psoriasis ( PS ) , Type 1 Diabetes ( T1D ) , Hypertension ( HT ) , Bipolar Disorder ( BD ) , Multiple Sclerosis ( MS ) and Coronary Artery Disease ( CAD ) . Our results show that the proposed method significantly outperforms individual variant based risk assessment model as well as the state-of-the-art polygenic score . We also observe that integrating prior biological information leads to more parsimonious models for risk assessment .
The input to the problem is a genome-wide association ( GWA ) dataset D = ( C , S , g , f ) , where C denotes the set of genomic loci that harbor the genetic variants ( e . g . , single nucleotide polymorphisms or copy number variants ) that are assayed , S denotes the set of samples , g ( c , s ) denotes the genotype of locus c ∈ C in sample s ∈ S , and f ( s ) denotes the phenotype of sample s ∈ S . Here , we assume that the phenotype variable is dichotomous , i . e . , f ( s ) can take only two values: if sample s is associated with the phenotype of interest ( e . g . diagnosed with the disease , responds to a certain drug etc . ) , s is called a “case” sample ( f ( s ) = 1 ) , otherwise ( e . g . , was not diagnosed with the disease , does not respond to a certain drug etc . ) , s is called a “control” sample ( f ( s ) = 0 ) . We denote the set of case samples with S1 and the set of control samples with S0 , where S1 ∪ S0 = S . While we focus on qualitative traits here for brevity , the proposed methodology can also be extended to quantitative traits ( i . e . , when f ( s ) is a continuous phenotype variable ) . The minor allele for a locus is usually defined as the allele that is less frequent in the population . While it is common to focus on the minor allele as the risk allele , specific genotypes can also be associated with a phenotype [35–37] . Different types of encoding may represent different biological assumptions . In an additive model , each genotype is encoded as a single numeric feature that reflects the number of minor alleles ( homozygous major , heterozygous , and homozygous minor are respectively encoded as 0 , 1 and 2 ) . This model does not capture combinatorial relationships between locus genotypes and phenotype , since the assumption is that one of the alleles quantitatively contributes to risk . In the recessive/dominant model , each genotype is encoded as two binary features ( presence of minor allele and presence of major allele ) . This model does not capture the difference between homozygous and heterozygous genotypes , since it only accounts for the presence of an allele . Here , we argue that considering the effect of all possible genotype combinations can provide more information in distinguishing case samples from control samples . The five models proposed here capture all potential relationships , in that differences in heterozygosity vs . homozygosity , presence vs . absence of a specific risk allele are represented by different genotype models . This notion is particularly useful when the genotypes of multiple loci are being integrated . For example , heterozygosity on one locus can be associated with increased susceptibility to a disease , while homozygous minor allele on another locus may be protective at the presence of heterozygosity in the former locus [38] . In this case , the interaction between the two loci can be detected by considering the association of all possible genotype combinations with the phenotype . We adaptively binarize the genotypes of each locus by considering all possible allele combinations . Given the genotype of a locus , we consider five different binary genotype models m ( i ) , i ∈ {1 , … 5} . Based on each model , we generate a binary genotype profile for each locus . Namely , we consider the following genotype models: 1 . Homozygous Minor Allele: This corresponds to the case when the possible effect of the minor allele is “recessive” , i . e . , the locus is considered to harbor a genotype of interest if both copies contain the minor allele . m ( 1 ) ( c , s ) = 1 2 if g ( c , s ) ∈ { a a } 0 otherwise ( 1 ) 2 . Heterozygous Genotype: The locus is considered to harbor a genotype of interest if the two copies contain different alleles . m ( 2 ) ( c , s ) = 1 2 if g ( c , s ) ∈ { A a } 0 otherwise ( 2 ) 3 . Homozygous Major Allele: The locus is considered to harbor a genotype of interest if both copies contain the major allele . m ( 3 ) ( c , s ) = 1 2 if g ( c , s ) ∈ { A A } 0 otherwise ( 3 ) 4 . Presence of Minor Allele: This corresponds to the case when the possible effect of the minor allele is “dominant” , i . e . , the locus is considered to harbor a genotype of interest if at least one copy contains the minor allele . This is the complement of m ( 3 ) . m ( 4 ) ( c , s ) = 1 2 if g ( c , s ) ∈ { A a , a a } 0 otherwise ( 4 ) 5 . Presence of Major Allele: The locus is considered to harbor a genotype of interest if at least one copy contains the major allele . This is the complement of m ( 1 ) . m ( 5 ) ( c , s ) = 1 2 if g ( c , s ) ∈ { A a , A A } 0 otherwise ( 5 ) Note that , although models m4 and m5 are complements of other models , we consider them separately . This is because , as we discuss in the next section , the 1s and 0s in the binary genotype profiles are considered asymmetrically while integrating the genotypes of multiple loci . Also note that “homozygous minor allele or homozygous major allele” is not considered since it is not associated with a specific risk allele . To select a genotype model for each locus , we separately assess the association of the resulting five genotype profiles with the phenotype of interest . Subsequently , we choose the model that leads to greatest discrimination between cases and controls , and use the respective binary genotype profile as the representative genotype of that locus . This process is illustrated in Fig 2 . For each locus c , binarization according to the five different genotype models produces five |S|-dimensional binary genotype profiles m ( i ) ( c ) , i ∈ {1 , … 5} . For each binary genotype profile m ( i ) ( c ) , we compute the difference in the fraction of case and control samples that harbor the genotype of interest as follows: D ( i ) ( c ) = 〈f , m ( i ) ( c ) 〉 | S 1 | - 〈 1 - f , m ( i ) ( c ) 〉 | S 0 | . ( 6 ) where 1 denotes a vector of all 1’s and < . > denotes the inner product of two vectors . We then determine the binary genotype model for each locus as the model that maximizes the difference of relative coverage between case samples and control samples , i . e . : k ( c ) = argmax i ∈ { 1 ⋯ 5 } { | D ( i ) ( c ) | } . ( 7 ) Based on the selected model for each locus , we compute the binary genotype profile accordingly: M ( c , s ) = m ( k ( c ) ) ( c , s ) . ( 8 ) Once we compute the binary genotype profiles for all loci , we identify Population Covering Locus Sets ( PoCos ) . In previous work , we define and use PoCos in the context of prioritizing locus pairs for testing epistasis [33] . In this earlier definition , the genotypes of interest are limited to the presence of the minor or major allele; i . e . , only the last two models described in the previous section are used to determine the binary genotype profile of each locus . Here , we generalize the concept of PoCo to utilize five different models for determining the genotypes of interest , as described in the previous subsection . A PoCo is a set of genomic loci that collectively “cover” a larger fraction of case samples while minimally covering control samples . Namely for a given set P ⊆ C of loci , we define the set of case and control samples covered by P respectively as E ( P ) = ∪ c ∈ P { s ∈ S 1 : M ( c , s ) = 1 } ( 9 ) and T ( P ) = ∪ c ∈ P { s ∈ S 0 : M ( c , s ) = 1 } . ( 10 ) We define a PoCo as a set P of loci that satisfies |E ( P ) | = |S1| while minimizing |T ( P ) | . Note that , since we are interested in finding all sets of loci with potential relationship in their association with phenotype , we do not define an optimization problem that aims to find a single PoCo with minimum |T ( P ) | . We rather develop an algorithm to heuristically identify all non-overlapping PoCos with minimal |T ( P ) | . To identify all non-overlapping PoCos , we use a greedy algorithm that progressively grows a set of loci to maximize the difference of the fraction of case and control samples covered by the loci that are recruited in a PoCo . In another words , we initialize P to ∅ and at each step , add to P the locus that maximizes δ ( c ) = E ( { c } ) ∩ S ′ | | S 1 | - | T ( { c } ) ∩ S ′ | | S 0 | ( 11 ) where S′ = S\ ( E ( P ) ∪ T ( P ) ) . The algorithm stops when all case samples are covered . We then record P , remove the loci in P from the dataset , and identify another PoCo . This process continues until it is not possible to find a set of loci that covers all case samples . We develop two methods to identify two different types of PoCos . The first type of PoCos ( named “network-free PoCos” ) are identifed using the greedy algorithm described above , without the use of any prior biological information . The second type of PoCos are NetPocos , which are identified by restricting the search space to connected subgraphs of a network of potential functional relationships among genomic loci . As we describe below , this network is constructed by integrating established locus-gene associations from eQTL studies and protein-protein interaction ( PPI ) data that contains functional relationships among genes . One potential utility of the PoCos is risk assessment . By construction , PoCos ( NetPocos ) contain ( functionally associated ) loci that exhibit improved power in distinguishing cases from control . Consequently , as compared to individual variants , they may provide more robust and reproducible features to be used in predictive models . To investigate the utility of these multi-locus features in risk assessment , we use PoCos to build a model for risk assessment using L1 regularized logistic regression classifier .
For each dataset , we divide the population into 5 groups while preserving the proportion of case and control samples in each group . We reserve one group for testing and we identify NetPocos on the remaining four groups . Then , we use these four groups for feature selection and model building . Finally , we test the performance on the group reserved for testing . All of the reported performance figures are averages across five different cross-validation runs . The number of PoCos identified on each dataset and the size of these PoCos are presented in Table 2 . Please note that the variance in number of PoCos does not have a significant effect on the performance ( S1 Fig ) .
In this paper , we propose a novel criterion to assess the collective disease-association of multiple genomic loci ( PoCos ) and investigate the utility of these multiple-loci features in risk assessment . We also perform extensive experiments to evaluate the effect of using network information to drive the search for multi-locus features on risk assessment . We also investigate the effect of the variants that have regulatory effects ( i . e . eQTL data ) on performance for risk assessment . Moreover , we compare the proposed method with the polygenic score which has been shown to be successful in different studies . Our result show that our method is significantly more powerful in risk assessment . Our results show that multi-locus features improve prediction performance as compared to individual locus based features . We also observe that integrating functional information provided by protein-protein interaction data and expression quantitative trait loci ( i . e . eQTL ) data leads to more parsimonious models for risk assessment . However , inclusion of functional data does not yield significant improvement in prediction performance . This may be indicative of the limitations of genomic data in risk assessment . Furthermore , since PoCos contain loci that are related to each other in the context of a phenotype , PoCos that are discovered without the inclusion of functional information also likely contain functionally related loci . However , utilization of functional information reduces the search space to render the problem computationally feasible , and brings forward PoCos that are more functionally relevant and robust , thereby leading to more parsimonious models . Based on the success of multi-locus genomic features in risk assessment , we conclude that combining these features with non-genetic risk factors and other biological data may lead to further improvements in risk assessment . The proposed method is implemented in MATLAB and provided in the public domain ( http://compbio . case . edu/pocos/ ) as open source software . | Several studies try to predict the individual disease risk using genetic data obtained from genome wide association studies ( GWAS ) . Earlier studies only focus on individual genetic variants . However , studies on disease mechanisms suggest the aggregation of genomic variants may contribute to diseases . For this reason , researchers commonly use prior biological knowledge to identify genetic variants that are functionally related . However , these approaches are often limited to variants that are in the coding regions of genes . However , several risk variants are in the regulatory region . Here , we incorporate known regulatory and functional interactions to find sets of genetic variants which are informative features for risk assessment . Our result on seven complex diseases show that our method outperforms individual variant based risk assessment models , as well as other methods that integrate multiple genetic variants . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"genome-wide",
"association",
"studies",
"medicine",
"and",
"health",
"sciences",
"protein",
"interactions",
"protein",
"interaction",
"networks",
"variant",
"genotypes",
"bipolar",
"disorder",
"alleles",
"genetic",
"mapping",
"coronary",
"heart",
"disease",
"network",
"analysis",
"genome",
"analysis",
"mood",
"disorders",
"cardiology",
"computer",
"and",
"information",
"sciences",
"protein-protein",
"interactions",
"proteins",
"proteomics",
"mental",
"health",
"and",
"psychiatry",
"genetic",
"loci",
"biochemistry",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"vascular",
"medicine",
"computational",
"biology",
"human",
"genetics"
] | 2016 | PoCos: Population Covering Locus Sets for Risk Assessment in Complex Diseases |
One of the most critical problems we face in the study of biological systems is building accurate statistical descriptions of them . This problem has been particularly challenging because biological systems typically contain large numbers of interacting elements , which precludes the use of standard brute force approaches . Recently , though , several groups have reported that there may be an alternate strategy . The reports show that reliable statistical models can be built without knowledge of all the interactions in a system; instead , pairwise interactions can suffice . These findings , however , are based on the analysis of small subsystems . Here , we ask whether the observations will generalize to systems of realistic size , that is , whether pairwise models will provide reliable descriptions of true biological systems . Our results show that , in most cases , they will not . The reason is that there is a crossover in the predictive power of pairwise models: If the size of the subsystem is below the crossover point , then the results have no predictive power for large systems . If the size is above the crossover point , then the results may have predictive power . This work thus provides a general framework for determining the extent to which pairwise models can be used to predict the behavior of large biological systems . Applied to neural data , the size of most systems studied so far is below the crossover point .
Many fundamental questions in biology are naturally treated in a probabilistic setting . For instance , deciphering the neural code requires knowledge of the probability of observing patterns of activity in response to stimuli [1]; determining which features of a protein are important for correct folding requires knowledge of the probability that a particular sequence of amino acids folds naturally [2] , [3]; and determining the patterns of foraging of animals and their social and individual behavior requires knowledge of the distribution of food and species over both space and time [4]–[6] . Constructing these probability distributions is , however , hard . There are several reasons for this: i ) biological systems are composed of large numbers of elements , and so can exhibit a huge number of configurations—in fact , an exponentially large number , ii ) the elements typically interact with each other , making it impossible to view the system as a collection of independent entities , and iii ) because of technological considerations , the descriptions of biological systems have to be built from very little data . For example , with current technology in neuroscience , we can record simultaneously from only about 100 neurons out of approximately 100 billion in the human brain . So , not only are we faced with the problem of estimating probability distributions in high dimensional spaces , we must do this based on a small fraction of the neurons in the network . Despite these apparent difficulties , recent work has suggested that the situation may be less bleak than it seems , and that an accurate statistical description of systems can be achieved without having to examine all possible configurations [2] , [3] , [7]–[11] . One merely has to measure the probability distribution over pairs of elements and use those to build the full distribution . These “pairwise models” potentially offer a fundamental simplification , as the number of pairs is quadratic in the number of elements , not exponential . However , support for the efficacy of pairwise models has , necessarily , come from relatively small subsystems—small enough that the true probability distribution could be measured experimentally [7]–[9] , [11] . While these studies have provided a key first step , a critical question remains: will the results from the analysis of these small subsystems extrapolate to large ones ? That is , if a pairwise model predicts the probability distribution for a subset of the elements in a system , will it also predict the probability distribution for the whole system ? Here we find that , for a biologically relevant class of systems , this question can be answered quantitatively and , importantly , generically—independent of many of the details of the biological system under consideration . And the answer is , generally , “no . ” In this paper , we explain , both analytically and with simulations , why this is the case .
To gain intuition into the extrapolation problem , let us consider a specific example: neuronal spike trains . Fig . 1A shows a typical spike train for a small population of neurons . Although the raw spike times provide a complete description , they are not a useful representation , as they are too high-dimensional . Therefore , we divide time into bins and re-represent the spike train as 0 s and 1 s: 0 if there is no spike in a bin; 1 otherwise ( Fig . 1B ) [7]–[9] , [11] . For now we assume that the bins are independent ( an assumption whose validity we discuss below , and in more detail in the section “Is there anything wrong with using small time bins ? ” ) . The problem , then , is to find where is a binary variable indicating no spike ( ) or one or more spikes ( ) on neuron . Since this , too , is a high dimensional problem ( though less so than the original spike time representation ) , suppose that we instead construct a pairwise approximation to , which we denote , for a population of size . ( The pairwise model derives its name from the fact that it has the same mean and pairwise correlations as the true model; see Eq . ( 15 ) . ) Our question , then , is: if is close to for small , what can we say about how close the two distributions are for large ? To answer this question quantitatively , we need a measure of distance . The measure we use , denoted , is defined in Eq . ( 3 ) below , but all we need to know about it for now is that if then , and if is near one then is far from . In terms of , our main results are as follows . First , for small , in what we call the perturbative regime , is proportional to . In other words , as the population size increases , the pairwise model becomes a worse and worse approximation to the true distribution . Second , this behavior is entirely generic: for small , increases linearly , no matter what the true distribution is . This is illustrated schematically in Fig . 2 , which shows the generic behavior of . The solid red part of the curve is the perturbative regime , where is a linearly increasing function of ; the dashed curves show possible behavior beyond the perturbative regime . These results have an important corollary: if one does an experiment and finds that is increasing linearly with , then one has no information at all about the true distribution . The flip side of this is more encouraging: if one can measure the true distribution for sufficiently large that saturates , as for the dashed blue line in Fig . 2 , then there is a chance that extrapolation to large is meaningful . The implications for the interpretation of experiments is , therefore , that one can gain information about large behavior only if one can analyze data past the perturbative regime . Under what conditions is a subsystem in the perturbative regime ? The answer turns out to be simple: the size of the system , , times the average probability of observing a spike in a bin , must be small compared to 1 . For example , if the average probability is 1/100 , then a system will be in the perturbative regime if the number of neurons is small compared to 100 . This last observation would seem to be good news: if we divide the spike trains into sufficiently small time bins and ignore temporal correlations , then we can model the data very well with a pairwise distribution . The problem with this , though , is that temporal correlations can be ignored only when time bins are large compared to the autocorrelation time . This leads to a kind of catch-22: pairwise models are guaranteed to work well ( in the sense that they describe spike trains in which temporal correlations are ignored ) if one uses small time bins , but small time bins is the one regime where ignoring temporal correlations is not a valid approximation . In the next several sections we quantify the qualitative picture presented above: we write down an explicit expression for , explain why it increases linearly with when is small , and provide additional tests , besides assessing the linearity of , to determine whether or not one is in the perturbative regime . A natural measure of the distance between and is the Kullback-Leibler ( KL ) divergence [12] , denoted and defined as ( 1 ) The KL divergence is zero if the two distributions are equal; otherwise it is nonzero . Although the KL divergence is a very natural measure , it is not easy to interpret ( except , of course , when it is exactly zero ) . That is because a nonzero KL divergence tells us that , but it does not give us any real handle on how good , or bad , the pairwise model really is . To make sense of the KL divergence , we need something to compare it to . A reasonable reference quantity , used by a number of authors [7]–[9] , is the KL divergence between the true distribution and the independent one , the latter denoted . The independent distribution , as its name suggests , is a distribution in which the variables are taken to be independent , ( 2 ) where is the distribution of the response of the neuron , . With this choice for a comparison , we define a normalized distance measure—a measure of how well the pairwise model explains the data—as ( 3 ) Note that the denominator in this expression , , is usually referred to as the multi-information [7] , [13] , [14] . The quantity lies between 0 and 1 , and measures how well a pairwise model does relative to an independent model . If it is 0 , the pairwise model is equal to the true model ( ) ; if it is near 1 , the pairwise model offers little improvement over the independent model; and if it is exactly 1 , the pairwise model is equal to the independent model ( ) , and so offers no improvement . How do we attach intuitive meaning to the two divergences and ? For the latter , we use the fact that , as is easy to show , ( 4 ) where and are the entropies [15] , [16] of and , respectively , defined , as usual , to be . For the former , we use the definition of the KL divergence to write ( 5 ) The quantity has the flavor of an entropy , although it is a true entropy only when is maximum entropy as well as pairwise ( see Eq . ( 6 ) below ) . For other pairwise distributions , all we need to know is that lies between and . A plot illustrating the relationship between , the two entropies and , and the entropy-like quantity , is shown in Fig . 3 . Note that for pairwise maximum entropy models ( or maximum entropy models for short ) , has a particularly simple interpretation , since in this case really is an entropy . Using to denote the pairwise entropy of a maximum entropy model , for this case we have ( 6 ) as is easy to see by inserting Eqs . ( 4 ) and ( 5 ) into ( 3 ) . This expression has been used previously by a number of authors [7] , [9] . The extrapolation problem discussed above is the problem of determining in the large limit . This is hard to do in general , but there is a perturbative regime in which it is possible . The small parameter that defines this regime is the average number of spikes produced by the whole population of neurons in each time bin . It is given quantitatively by where is the bin size and the average firing rate , ( 7 ) with the firing rate of neuron . The first step in the perturbation expansion is to compute the two quantities that make up : and . As we show in the section “Perturbative Expansion” ( Methods ) , these are given by ( 8a ) ( 8b ) where ( 9a ) ( 9b ) Here and in what follows we use to denote terms that are proportional to in the limit . The in Eq . ( 9a ) has been noted previously [7] , although the authors did not compute the prefactor , . The prefactors and , which are given explicitly in Eqs . ( 42 ) and ( 44 ) , depend on the low order statistics of the spike trains: depends on the second order normalized correlation coefficients , depends on the second and third order normalized correlation coefficients ( the normalized correlation coefficients are defined in Eq . ( 16 ) below ) , and both depend on the firing rates of the individual cells . The details of that dependence , however , are not important for now; what is important is that and are independent of and ( at least on average; see next section ) . Inserting Eq . ( 8 ) into Eq . ( 3 ) ( into the definition of ) and using Eq . ( 9 ) , we arrive at our main result , ( 10a ) ( 10b ) Note that in the regime , is necessarily small . This explains why , in an analytic study of non-pairwise model in which was small , Shlens et al . found that was rarely greater than 0 . 1 [8] . We refer to quantities with a superscript zero as “zeroth order . ” Note that , via Eqs . ( 4 ) and ( 5 ) , we can also define zeroth order entropies , ( 11a ) ( 11b ) These quantities are important primarily because differences between them and the actual entropies indicate a breakdown of the perturbation expansion ( see in particular Fig . 4 below ) . Assuming , as discussed in the next section , that and are approximately independent of , , and , Eq . ( 10 ) tells us that scales linearly with in the perturbative regime—the regime in which . The key observation about this scaling is that it is independent of the details of the true distribution , . This has a very important consequence , one that has major implications for experimental data: if one does an experiment with small and finds that is proportional to , then the system is , with very high probability , in the perturbative regime , and one does not know whether will remain close to as increases . What this means in practical terms is that if one wants to know whether a particular pairwise model is a good one for large systems , it is necessary to consider values of that are significantly greater than , where ( 12 ) We interpret as the value at which there is a crossover in the behavior of the pairwise model . Specifically , if , the system is in the perturbative regime and the pairwise model is not informative about the large behavior , whereas if , the system is in a regime in which it may be possible to make inferences about the behavior of the full system . As we show in Methods ( see in particular Eqs . ( 42 ) and ( 44 ) ) , the prefactors and depend on which neurons out of the full population are used . Consequently , these quantities fluctuate around their true values ( in the sense that different subpopulations produce different values of and ) , where “true” refers to an average over all possible sub-populations . Here we assume that the neurons are chosen randomly from the full population , so any set of neurons provides unbiased estimates of and . In our simulations , the fluctuations were small , as indicated by the small error bars on the blue points in Fig . 5 . However , in general the size of the fluctuations is determined by the range of firing rates and correlation coefficients , with larger ranges producing larger fluctuations . Because does not affect the mean values of and , it is reasonable to think of these quantities—or at least their true values—as being independent of . They are also independent of , again modulo fluctuations . Finally , as we show in the section “Bin size and the correlation coefficients” ( Methods ) , and are independent of in the limit that is small compared to the width of the temporal correlations among neurons . We will assume this limit applies here . In sum , then , to first approximation , and are independent of our three important quantities: , , and . Thus , we treat them as effectively constant throughout our analysis . Although the behavior of in the perturbative regime does not tell us much about its behavior at large , it is possible that other quantities that can be calculated in the perturbative regime , , , and ( the last one exactly ) , are informative , as others have suggested [7] . Here we show that this is not the case—they also are uninformative . The easiest way to relate the perturbative regime to the large regime is to ignore the corrections in Eqs . ( 8a ) and ( 8b ) , extrapolate the expressions for the zeroth order terms , and ask what their large behavior tells us . Generic versions of these extrapolations , plotted on a log-log scale , are shown in Fig . 4A , along with a plot of the independent entropy , ( which is necessarily linear in ) . The first thing we notice about the extrapolations is that they do not , technically , have a large behavior: one terminates at the point labeled , which is where ( and thus , via Eq . ( 0a ) , ; continuing the extrapolation implies negative true zeroth order entropy ) ; the other at the point labeled , which is where ( and thus , via Eq . ( 5 ) and the fact that , ) . Despite the fact that the extrapolations end abruptly , they still might provide information about the large regime . For example , and/or might be values of at which something interesting happens . To see if this is the case , in Fig . 4B we plot the naive extrapolations of and ( that is , the zeroth order quantities given in Eq . ( 11 ) , and ) , on a linear-linear plot , along with . This plot contains no new information compared to Fig . 4A , but it does elucidate the meaning of the extrapolations . Perhaps its most striking feature is that the naive extrapolation of has a decreasing portion . As is easy to show mathematically , entropy cannot decrease with ( intuitively , that is because observing one additional neuron cannot decrease the entropy of previously observed neurons ) . Thus , , which occurs well beyond the point where the naive extrapolation of is decreasing , has essentially no meaning , something that has been pointed out previously by Bethge and Berens [10] . The other potentially important value of is . This , though , suffers from a similar problem: when , is negative . How do the naively extrapolated entropies—the solid lines in Fig . 4B—compare to the actual entropies ? To answer this , in Fig . 4B we show the true behavior of and versus ( dashed lines ) . Note that is asymptotically linear in , even though the neurons are correlated , a fact that forces to be linear in , as it is sandwiched between and . ( The asymptotically linear behavior of is typical , even in highly correlated systems . Although this is not always appreciated , it is easy to show; see the section “The behavior of the true entropy in the large N limit , ” Methods . ) Comparing the dashed and solid lines , we see that the naively extrapolated and true entropies , and thus the naively extrapolated and true values of , have extremely different behavior . This further suggests that there is very little connection between the perturbative and large regimes . In fact , these observations follow directly from the fact that and depend only on correlation coefficients up to third order ( see Eqs . ( 42 ) and ( 44 ) ) whereas the large behavior depends on correlations at all orders . Thus , since and tell us very little , if anything , about higher order correlations , it is not surprising that they tell us very little about the behavior of in the large limit . To check that our perturbation expansions , Eqs . ( 8–10 ) , are correct , and to investigate the regime in which they are valid , we performed numerical simulations . We generated , from synthetic data , a set of true distributions , computed the true distance measures , , , and numerically , and compared them to the zeroth order ones , , , and . If the perturbation expansion is valid , then the true values should be close to the zeroth order values whenever is small . The results are shown in Fig . 5 , and that is , indeed , what we observed . Before discussing that figure , though , we explain our procedure for constructing true distributions . The set of true distributions we used were generated from a third order model ( so named because it includes up to third order interactions ) . This model has the form ( 13 ) where is a normalization constant , chosen to ensure that the probability distribution sums to 1 , and the sums over , and run from 1 to . The parameters and were chosen by sampling from distributions ( see the section “Generating synthetic data , ” Methods ) , which allowed us to generate many different true distributions . In all of our simulations we calculate the relevant quantities directly from Eq . ( 13 ) . Consequently , we do not have to worry about issues of finite data , as would be the case in realistic experiments . For a particular simulation ( corresponding to a column in Fig . 5 ) , we generated a true distribution with , randomly chose 5 neurons , and marginalized over them . This gave us a 10-neuron true distribution . True distributions with were constructed by marginalizing over additional neurons within our 10-neuron population . To achieve a representative sample , we considered all possible marginalizations ( of which there are 10 choose , or ) . The results in Fig . 5 are averages over these marginalizations . For neural data , the most commonly used pairwise model is the maximum entropy model . Therefore , we use that one here . To emphasize the maximum entropy nature of this model , we replace the label “pair” that we have been using so far with “maxent . ” The maximum entropy distribution has the form ( 14 ) Fitting this distribution requires that we choose the and so that the first and second moments match those of the true distribution . Quantitatively , these conditions are ( 15a ) ( 15b ) where the angle brackets , and , represent averages with respect to and , respectively . Once we have and that satisfy Eq . ( 15 ) , we calculate the KL divergences , Eqs . ( 1 ) and ( 4 ) , and use those to compute . The results are shown in Fig . 5 . The rows correspond to our three quantities of interest: , , and ( top to bottom ) . The columns correspond to different values of , with the smallest on the left and the largest on the right . Red circles are the true values of these quantities; blue ones are the zeroth order predictions from Eqs . ( 9 ) and ( 10b ) . As suggested by our perturbation analysis , the smaller the value of , the larger the value of for which agreement between the true and zeroth order values is good . Our simulations corroborate this: the left column of Fig . 5 has , and agreement is almost perfect out to ; the middle column has , and agreement is almost perfect out to ; and the right column has , and agreement is not good for any value of . Note that the perturbation expansion breaks down for values of well below ( defined in Eq . ( 12 ) ) : in the middle column of Fig . 5 it breaks down when , and in the right column it breaks down when . This is not , however , especially surprising , as the perturbation expansion is guaranteed to be valid only if . These results validate the perturbation expansions in Eqs . ( 8 ) and ( 10 ) , and show that those expansions provide sensible predictions—at least for some parameters . They also suggest a natural way to assess the significance of one's data: plot , , and versus , and look for agreement with the predictions of the perturbation expansion . If agreement is good , as in the left column of Fig . 5 , then one is in the perturbative regime , and one knows very little about the true distribution . If , on the other hand , agreement is bad , as in the right column , then one is out of the perturbative regime , and it may be possible to extract meaningful information about the relationship between the true and pairwise models . That said , the qualifier “at least for some parameters” is an important one . This is because the perturbation expansion is essentially an expansion that depends on the normalized correlation coefficients , and there is an underlying assumption that they don't exhibit pathological behavior . The order normalized correlation coefficient for the distribution , denoted , is written ( 16 ) A potentially problematic feature of the correlation coefficients is that the denominator is a product over mean activities . If the mean activities are small , the denominator can become very small , leading to very large correlation coefficients . Although our perturbation expansion is always valid for sufficiently small time bins ( because the correlation coefficients eventually becomes independent of bin size; see the section “Bin size and the correlation coeffcients , ” Methods ) , “sufficiently small” can depend in detail on the parameters . For instance , at the maximum population size tested ( ) and for the true distributions that had , the absolute error of the prediction had a median of approximately 16% . However , about 11% of the runs had errors larger than 60% . Thus , the exact size of the small parameter at which the perturbative expansion breaks down can depend on the details of the true distribution . Estimation of the KL divergences and from real data can be hard , in the sense that it takes a large amount of data for them to converge to their true values . In addition , as discussed above , in the section “The prefactors gind and gpair” , there are fluctuations in associated with finite subsampling of the full population of neurons . Those fluctuations tend to keep from being purely linear , as can seen , for example , in the blue points in Fig . 5F and 5I . We therefore provide a second set of relationships that can be used to determine whether or not a particular data set is in the perturbative regime . These relationships are between the parameters of the maximum entropy model , the and , and the mean activity and normalized second order correlation coefficient ( the latter defined in Eq . ( 19 ) below ) . Since the quantity plays a central role in our analysis , we replace it with a single parameter , which we denote , ( 17 ) In terms of this parameter , we find ( using the same perturbative approach that led us to Eqs . ( 8–10 ) ; see the section “External fields , pairwise couplings and moments , ” Methods ) , that ( 18a ) ( 18b ) where , the normalized second order correlation coefficient , is defined in Eq . ( 16 ) with ; it is given explicitly by ( 19 ) ( We don't need a superscript on or a subscript on the angle brackets because the first and second moments are the same under the true and pairwise distributions . ) Equation ( 18a ) can be reconstructed from the low firing rate limit of analysis carried out by Sessak and Monasson [17] , as can the first three terms in the expansion of the log in Eq . ( 18b ) . Equation ( 18 ) tells us that the of the and , the external fields and pairwise couplings , is very weak . In Fig . 6 we confirm this through numerical simulations . Equation ( 18b ) also provides additional information—it gives us a functional relationship between the pairwise couplings and the normalized pairwise correlations function , . In Fig . 7A–C we plot the pairwise couplings , , versus the normalized pairwise correlation coefficient , ( blue dots ) , along with the prediction from Eq . ( 18b ) ( black line ) . Consistent with our predictions , the data in Fig . 7A–C essentially follows a line—the line given by Eq . ( 18b ) . A relationship between the pairwise couplings and the correlations coefficients has been sought previously , but for the more standard Pearson correlation coefficient [7] , [9] , [11] . Our analysis explains why it was not found . The Pearson correlation coefficient , denoted , is given by ( 20 ) In the small limit—the limit of interest—the right hand side of Eq . ( 20 ) is approximately equal to . Because depends on the external fields , and ( see Eq . ( 18a ) ) and there is a one-to-one relationship between and ( Eq . ( 18b ) ) , there can't be a one-to-one relationship between and . We verify the lack of a relationship in Fig . 7D and 7E , where we again plot , but this time versus the standard correlation coefficient , . As predicted , the data in Fig . 7D and 7E is scattered over two dimensions . This suggests that , not , is the natural measure of the correlation between two neurons when they have a binary representation , something that has also been suggested by Amari based on information-geometric arguments [18] . Note that the lack of a simple relationship between the pairwise couplings and the standard correlation coefficient has been a major motivation in building maximum entropy models [7] , [11] . This is for good reason: if there is a simple relationship , knowing the adds essentially nothing . Thus , plotting versus ( but not ) is an important test of one's data , and if the two quantities fall on the curve predicted by Eq . ( 18b ) , the maximum entropy model is adding very little information , if any . As an aside , we should point out that the is a function of the variables used to represent the firing patterns . Here we use 0 for no spike and 1 for one or more spikes , but another , possibly more common , representation , derived from the Ising model and used in a number of studies [7] , [9] , [11] , is to use −1 and +1 rather than 0 and 1 . This amounts to making the change of variables . In terms of , the maximum entropy model has the form where and are given by ( 21a ) ( 21b ) The second term on the right side of Eq . ( 21a ) is proportional to , which means the external fields in the Ising representation acquire a linear that was not present in our 0/1 representation . The two studies that reported the of the external fields [7] , [9] used this representation , and , as predicted by our analysis , the external fields in those studies had a component that was linear in . An outcome of our perturbative approach is that our normalized distance measure , , is linear in bin size ( see Eq . ( 10b ) ) . This suggests that one could make the pairwise model look better and better simply by making the bin size smaller and smaller . Is there anything wrong with this ? The answer is yes , for reasons discussed above ( see the the section “The extrapolation problem” ) ; here we emphasize and expand on this issue , as it is an important one for making sense of experimental results . The problem arises because what we have been calling the “true” distribution is not really the true distribution of spike trains . It is the distribution assuming independent time bins , an assumption that becomes worse and worse as we make the bins smaller and smaller . ( We use this potentially confusing nomenclature primarily because all studies of neuronal data carried out so far have assumed temporal independence , and compared the pairwise distribution to the temporally independent—but still correlated across neurons—distribution [7]–[9] , [11] . In addition , the correct name “true under the assumption of temporal independence , ” is unwieldy . ) Here we quantify how much worse . In particular , we show that if one uses time bins that are small compared to the characteristic correlation time in the spike trains , the pairwise model will not provide a good description of the data . Essentially , we show that , when the time bins are too small , the error one makes in ignoring temporal correlations is larger than the error one makes in ignoring correlations across neurons . As usual , we divide time into bins of size . However , because we are dropping the independence assumption , we use , rather than , to denote the response in bin . The full probability distribution over all time bins is denoted . Here is the number of bins; it is equal to for spike trains of length . If time bins are approximately independent and the distribution of is the same for all ( an assumption we make for convenience only , but do not need; see the section “Extending the normalized distance measure to the time domain , ” Methods ) , we can write ( 22 ) Furthermore , if the pairwise model is a good one , we have ( 23 ) Combining Eqs . ( 22 ) and Eq . ( 23 ) then gives us an especially simple expression for the full probability distribution: . The problem with small time bins lies in Eq . ( 22 ) : the right hand side is a good approximation to the true distribution when the time bins are large compared to the spike train correlation time , but it is a bad approximation when the time bins are small ( because adjacent time bins become highly correlated ) . To quantify how bad , we compare the error one makes assuming independence across time to the error one makes assuming independence across neurons . The ratio of those two errors , denoted , is given by ( 24 ) It is relatively easy to compute in the limit of small time bins ( see the section “Extending the normalized distance measure to the time domain , ” Methods ) , and we find that ( 25 ) As expected , this reduces to our old result , , when there is only one time bin ( ) . When is larger than 1 , however , the second term is always at least one , and for small bin size , the third term is much larger than one . Consequently , if we use bins that are small compared to the temporal correlation time of the spike trains , the pairwise model will do a very bad job describing the full , temporally correlated spike trains .
That the pairwise model is always good if is sufficiently small has strong implications: if we want to build a good model for a particular , we can simply choose a bin size that is small compared to . However , one of the assumptions in all pairwise models used on neural data is that bins at different times are independent . This produces a tension between small time bins and temporal independence: small time bins essentially ensure that a pairwise model will provide a close approximation to a model with independent bins , but they make adjacent bins highly correlated . Large time bins come with no such assurance , but they make adjacent bins independent . Unfortunately , this tension is often unresolvable in large populations , in the sense that pairwise models are assured to work only up to populations of size where τcorr is the typical correlation time . Given that is at least several Hz , for experimental paradigms in which the correlation time is more than a few hundred ms , is about one , and this assurance does not apply to even moderately sized populations of neurons . These observations are especially relevant for studies that use small time bins to model spike trains driven by natural stimuli . This is because the long correlation times inherent in natural stimuli are passed on to the spike trains , so the assumption of independence across time ( which is required for the independence assumption to be valid ) breaks badly . Knowing that these models are successful in describing spike trains under the independence assumption , then , does not tell us whether they will be successful in describing full , temporally correlated , spike trains . Note that for studies that use stimuli with short correlation times ( e . g . , non-natural stimuli such as white noise ) , the temporal correlations in the spike trains are likely to be short , and using small time bins may be perfectly valid . The only study that has investigated the issue of temporal correlations in maximum entropy models does indeed support the above picture [9]: for the parameters used in that study ( to 0 . 4 ) , the pairwise maximum entropy model provided a good fit to the data ( was typically smaller than 0 . 1 ) , but it did not do a good job modeling the temporal structure of the spike trains . As mentioned in the Introduction , in addition to the studies on neuronal data , studies on protein folding have also emphasized the role of pairwise interactions [2] , [3] . Briefly , proteins consist of strings of amino acids , and a major question in structural biology is: what is the probability distribution of amino acid strings in naturally folding proteins ? One way to answer this is to approximate the full probability distribution of naturally folding proteins from knowledge of single-site and pairwise distributions . One can show that there is a perturbative regime for proteins as well . This can be readily seen using the celebrated HP protein model [20] , where a protein is composed of only two types of amino acids . If , at each site , one amino acid type is preferred and occurs with high probability , say with , then a protein of length shorter than will be in the perturbative regime , and , therefore , a good match between the true distribution and the pairwise distribution for such a protein is virtually guaranteed . Fortunately , the properties of real proteins generally prevent this from happening: at the majority of sites in a protein , the distribution of amino acids is not sharply peaked around one amino acid . Even for those sites that are sharply peaked ( the evolutionarily-conserved sites ) , the probability of the most likely amino acid , , rarely exceeds 90% [21] , [22] . This puts proteins consisting of only a few amino acids out of the perturbative regime , and puts longer proteins—the ones usually studied using pairwise models—well out of it . This difference is fundamental: because many of the studies that have been carried out on neural data were in the perturbative regime , the conclusions of those studies—specifically , the conclusion that pairwise models provide accurate descriptions of large populations of neurons—is not yet supported . This is not the case for the protein studies , because they are not in the perturbative regime . Thus , the evidence that pairwise models provide accurate descriptions of protein folding remain strong and exceedingly promising . In our analysis , we sidestepped two issues of practical importance: finite sampling and alternative measures for assessing the quality of the pairwise model . These issues are beyond the scope of this paper , but in our view , they are natural next steps in the analysis of pairwise models . Below we briefly expand on them . Finite sampling refers to the fact that in any real experiment , one has access to only a finite amount of data , and so does not know the true probability distribution of the spike trains . In our analysis , however , we assumed that one did have full knowledge of the true probability distribution . Since a good estimate of the probability distribution is crucial for assessing whether the pairwise model can be extrapolated to large populations , it is important to study how such estimates are affected by finite data . Future work is needed to address this issue , and to find ways to overcome data limitation—for example , by finding efficient methods for removing the finite data bias that affects information theoretic quantities such as the Kullback-Leibler divergence . There are always many possible ways to assess the quality of a model . Our choice of was motivated by two considerations: it is based on the Kullback-Leibler divergence , which is a standard measure of “distance” between probability distributions , and it is a widely used measure in the field [7]–[10] . It suffers , however , from a number of shortcomings . In particular , can be small even when the pairwise model assigns very different probabilities to many of the configurations of the system . It would , therefore , be important to study the quality of pairwise models using other measures .
To understand how the true entropy behaves in the large limit , it is useful to express the difference of the entropies as a mutual information . Using to denote the true entropy of neurons and to denote the mutual information between one neuron and the other neurons in a population of size , we have ( 26 ) If knowing the activity of neurons does not fully constrain the firing of neuron , then the single neuron entropy , , will exceed the mutual information , , and the entropy will be an increasing function of . For the entropy to be linear in , all we need is that the mutual information saturates with . Because of synaptic noise , this is a reasonable assumption for networks of neurons: even if we observed all the spikes from all the neurons , there would still be residual noise associated with synaptic failures , jitter in release time , variability in the amount of neurotransmitter released , stochastic channel dynamics , etc . Consequently , in the large limit , we may replace by its average , denoted . Also replacing by its average , denoted , we see that for large , the difference between and approaches a constant . Specifically , ( 27 ) where this expression is valid in the large limit and the corrections are sublinear in . Our main quantitative result , given in Eqs . ( 8–10 ) , is that the KL divergence between the true distribution and both the independent and pairwise distributions can be computed perturbatively as an expansion in powers of in the limit . Here we carry out this expansion , and derive explicit expressions for the quantities and . To simplify our notation , here we use for the true distribution . The critical step in computing the KL divergences perturbatively is to use the Sarmanov-Lancaster expansion [23]–[28] for , ( 28 ) where ( 29a ) ( 29b ) ( 29c ) ( 29d ) This expansion has a number of important , but not immediately obvious , properties . First , as can be shown by a direct calculation , ( 30 ) where the subscripts and indicate an average with respect to and , respectively . This has an immediate corollary , This last relationship is important , because it tells us that if a product of contains any terms linear in one of the , the whole product averages to zero under the independent distribution . This can be used to show that ( 31 ) from which it follows that Thus , is properly normalized . Finally , a slightly more involved calculations provides us with a relationship between the parameters of the model and the moments: for , ( 32a ) ( 32b ) Virtually identical expressions hold for higher order moments . It is this last set of relationships that make the Sarmanov-Lancaster expansion so useful . Note that Eqs . ( 32a ) and ( 32b ) , along with the expression for the normalized correlation coefficients given in Eq . ( 16 ) , imply that ( 33a ) ( 33b ) These identities will be extremely useful for simplifying expressions later on . Because the moments are so closely related to the parameters of the distribution , moment matching is especially convenient: to construct a distribution whose moments match those of up to some order , one simply needs to ensure that the parameters of that distribution , , , , etc . , are identical to those of the true distributions up to the order of interest . In particular , let us write down a new distribution , , ( 34a ) ( 34b ) We can recover the independent distribution by letting , and we can recover the pairwise distribution by letting . This allows us to compute in the general case , and then either set to zero or set to . Expressions analogous to those in Eqs . ( 31–33 ) exist for averages with respect to ; the only difference is that is replaced by . Using Eqs . ( 28 ) and ( 34a ) and a small amount of algebra , the KL divergence between and may be written ( 35 ) where ( 36 ) To derive Eq . ( 35 ) , we used the fact that ( see Eq . ( 31 ) ) . The extra term was included to ensure that and its first derivatives vanish at , something that greatly simplifies our analysis later on . Our approach is to Taylor expand the right hand side of Eq . ( 35 ) around , compute each term , and then sum the whole series ( we do not assume that either or is small ) . Using to denote the coefficients of the Taylor series , we have ( 37 ) Note that because and its first derivatives vanish at , all terms in this sum have . Because both and are themselves sums , an exact calculation of the terms in Eq . ( 37 ) would be difficult . However , as we show below , in the section “Averages of powers of ξp and ξq” ( see in particular Eqs . ( 52 ) and ( 54 ) ) , they can be computed as perturbation expansions in powers of , and the result is ( 38 ) where and are given by ( 39 ) . The last equality in Eq . ( 39 ) follows from a small amount of algebra and the definition of the correlation coefficients given in Eq . ( 16 ) . Equation ( 38 ) is valid only when , which is the case of interest to us ( since the Taylor expansion of has only terms with ) . The important point about Eq . ( 38 ) is that the and dependence follows that of the original Taylor expansion . Thus , when we insert this equation back into Eq . ( 37 ) , we recover our original function—all we have to do is interchange the sums . For example , consider inserting the first term in Eq . ( 38 ) into Eq . ( 37 ) , Performing the same set of manipulations on all of Eq . ( 38 ) leads to ( 40 ) This expression is true in general ( except for some technical considerations; see the section “Averages of powers of ξp and ξq” ) ; to restrict it to the KL divergences of interest we set to and to either or . In the first case ( set to ) , , which implies that , and thus . Because has a quadratic minimum at , when , the second two terms on the right hand side of Eq . ( 40 ) are . We thus have , to lowest nonvanishing order in , ( 41 ) with the correction coming from the last sum in Eq . ( 40 ) . Defining ( 42 ) where , recall , and inserting Eq . ( 42 ) into Eq . ( 41 ) , we recover Eq . ( 8a ) . In the second case ( set to ) , the first and second moments of and are equal . This implies , using Eq . ( 32 ) , that , and thus . Because ( see Eq . ( 36 ) ) , the first three terms on the right hand side of Eq . ( 40 ) —those involving and but not —vanish . The next order term does not vanish , and yields ( 43 ) Defining ( 44 ) and inserting this expression into Eq . ( 43 ) , we recover Eq . ( 8b ) . In this section we derive , to leading order in , expressions relating the external fields and pairwise couplings of the maximum entropy model , and , to the first and second moments; these are the expressions reported in Eq . ( 18 ) . The calculation proceeds along the same lines as in the previous section . There is , though , one extra step associated with the fact that the quadratic term in the maximum entropy distribution given in Eq . ( 14 ) is proportional to , not . However , to lowest order in , this doesn't matter . That's becausewhere is defined as in Eq . ( 29d ) except with replaced by , and we used the fact that . The second term introduces a correction to the external fields , . However , the correction is , so we drop it . We should keep in mind , though , that our final expression for will have corrections of this order . Using Eq . ( 14 ) , but with replaced by where it appears with , we may write ( after a small amount of algebra ) ( 45 ) where is the same as the function defined in Eq . ( 29a ) except that is replaced by , the subscript “ind” indicates , as usual , an average with respect to , and the two functions and are defined by ( 46 ) and ( 47 ) Given this setup , we can use Eqs . ( 55 ) and ( 56 ) below to compute the moments under the maximum entropy model . That's because both and its first derivative vanish at , which are the two conditions required for these equations to be valid . Using also the fact that , Eqs . ( 55 ) and ( 56 ) imply that ( 48a ) ( 48b ) ( 48c ) where the first term in Eq . ( 48b ) came from Eq . ( 29d ) with replaced by , the term “” in Eq . ( 48c ) came from , and for the second two expressions we used the fact that , to lowest order in , the denominator in Eq . ( 45 ) is equal to 1 . Finally , using Eq . ( 19 ) for the normalized correlation coefficient , dropping the subscript “maxent” ( since the first and second moments are the same under the maxent and true distributions ) , and inverting Eqs . ( 48b ) and ( 48c ) to express the external fields and coupling coefficients in terms of the first and second moments , we arrive at Eq ( 18 ) . Here we compute , which , as can be seen in Eq . ( 37 ) , is the key quantity in our perturbation expansion . Our starting point is to ( formally ) expand the sums that make up and ( see Eqs . ( 29b ) and ( 34b ) ) , which yields ( 49 ) The sum over is a sum over all possible configurations of the . The coefficient are complicated functions of the , etc . Computing these functions is straightforward , although somewhat tedious , especially when is large; below we compute them only for and 3 . The reason starts at 2 is that ; see Eq . ( 37 ) . We first show that all terms with superscript are . To do this , we note that , because the right hand side of Eq . ( 49 ) is an average with respect to the independent distribution , the average of the product is the product of the averages , ( 50 ) Then , using the fact that with probability and with probability ( see Eq . ( 29c ) ) , we have ( 51 ) The significance of this expression is that , for , , independent of . Consequently , if all the in Eq . ( 50 ) are greater than 1 , then the right hand side is . This shows that , as promised above , the superscript labels the order of the terms . As we saw in the section “The KL divergence in the Sarmanov-Lancaster representation” , we need to go to third order in , which means we need to compute the terms on the right hand side of Eq . ( 49 ) with and 3 . Let us start with , which picks out only those terms with two unique indices . Examining the expressions for and given in Eqs . ( 29b ) and ( 34b ) , we see that we must keep only terms involving , since has three indices , and higher order terms have more . Thus , the contribution to the average in Eq . ( 49 ) , which we denote , is given by Pulling and out of the averages , using Eq . ( 33a ) to eliminate and in favor of and , and applying Eq . ( 51 ) ( while throwing away some of the terms in that equation that are higher than second order in ) , the above expression may be written ( 52 ) Note that we were not quite consistent in our ordering with respect to , in the sense that we kept some higher order terms and not others . We did this so that we could use rather than , as the former is directly observable . For the calculation is more involved , but not substantially so . Including terms with exactly three unique indices in the sum on the right hand side of Eq . ( 49 ) gives us ( 53 ) This expression is not quite correct , since some of the terms contain only two unique indices—these are the terms proportional to —whereas it should contain only terms with exactly three unique indices . Fortunately , this turns out not to matter , for reasons we discuss at the end of the section . To perform the averages in Eq . ( 53 ) , we would need to use multinomial expansions , and then average over the resulting powers of . For the latter , we can work to lowest order in , which means we only take the first term in Eq . ( 51 ) . This amounts to replacing every with ( and similarly for and ) , and in addition multiplying the whole expression by an overall factor of . For example , if and , one of the terms in the multinomial expansion is . This average would yield , using Eq . ( 51 ) and considering only the lowest order term , . This procedure also is not quite correct , since terms with only one factor of , which average to zero , are replaced with . This also turns out not to matter; again , we discuss why at the end of the section . We can , then , go ahead and use the above “replace blindly” algorithm . Note that the factors of , and turn and into normalized correlation coefficients ( see Eq . ( 33 ) ) , which considerably simplifies our equations . Using also Eq . ( 39 ) for , Eq . ( 53 ) becomes ( 54 ) We can now combine Eqs . ( 52 ) and ( 54 ) , and insert them into Eq . ( 49 ) . This gives us the first two terms in the perturbative expansion of ; the result is written down in Eq . ( 38 ) above . Why can we ignore the overcounting associated with terms in which an index appears exactly zero or one times ? We clearly can't do this in general , because for such terms , replacing with fails—either because the terms didn't exist in the first place ( when one of the indices never appeared ) or because they averaged to zero ( when an index appeared exactly once ) . In our case , however , such terms do not appear in the Taylor expansion . To see why , note first of all that , because of the form of , its Taylor expansion can be written where is finite at ( see Eq . ( 36 ) ) . Consequently , the original Taylor expansion of , Eq . ( 37 ) , should contain a factor of ; i . e . , where the are the coefficients of the Taylor expansion of . The factor , when expanded , has the form As we saw in the previous section , we are interested in the third order term only to compute , for which . Therefore , the above multiplicative factor reduces to . It is that last factor of that is important , since it guarantees that for every term in the Taylor expansion , all indices appear at least twice . Therefore , although Eq . ( 53 ) is not true in general , it is valid for our analysis . We end this section by pointing out that there is a very simple procedure for computing averages to second order in . Consider a function that has a minimum at and also obeys . Then , based on the above analysis , we have ( 55 ) Two easy corollaries of this are: for and positive integers , ( 56a ) ( 56b ) where the sum in Eq . ( 56a ) runs over only , and we used the fact that both and are symmetric with respect to the interchange of and . As can be seen in Eq . ( 13 ) , the synthetic data depends on three sets of parameters: , and . Here we describe how they were generated . To generate the , we draw a set of firing rates , , from an exponential distribution with mean 0 . 02 ( recall that , which we set to 15 , is the number of neurons in our base distribution ) . From this we chose the external field according to Eq . ( 18a ) , In the perturbative regime , a distribution generated with these values of the external fields has firing rates approximately equal to the ; see Eq . ( 18a ) and Fig . 6 . To generate the and , we drew them from Gaussian distributions with means equal to 0 . 05 and 0 . 02 and standard deviations of 0 . 8 and 0 . 5 , respectively . Using non-zero values for means that the true distribution is not pairwise . One of our main claims is that is linear in bin size , . This is true , however , only if and are independent of , as can be seen from Eq . ( 10b ) . In this section we show that independence is satisfied if is smaller than the typical correlation time of the responses . For larger than such correlation times , and do depend on , and is no longer linear in . Note , though , that the correlation time is always finite , so there will always be a bin size below which the linear relationship , , is guaranteed . Examining Eqs . ( 42 ) and ( 44 ) , we see that and depend on the normalized correlation coefficients , and ( we drop superscripts , since our discussion will be generic ) . Thus , to understand how and depend on bin size , we need to understand how the normalized correlation coefficients depend on bin size . To do that , we express them in terms of standard cross-correlograms , as the cross-correlograms contain , in a very natural way , information about the temporal timescales in the spike train . We start with the second order correlation coefficient , since it is simplest . The corresponding cross-correlogram , which we denote , is given by ( 57 ) where is the time of the kth spike on neuron ( and similarly for ) , and is the Dirac . The normalization in Eq . ( 57 ) is slightly non-standard—more typical is to divide by something with units of firing rate ( , or ) , to give units of spikes/s . The normalization we use is convenient , however , because in the limit of large , approaches one . It is slightly tedious , but otherwise straightforward , to show that when is sufficiently small that only one spike can occur in a time bin , is related to via ( 58 ) The ( unimportant ) factor comes from the fact that the spikes occur at random locations within a bin . Equation ( 58 ) has a simple interpretation: is the average height of the central peak of the cross-correlogram relative to baseline . How strongly depends on is thus determined by the shape of the cross-correlogram . If it is smooth , then approaches a constant as becomes small . If , on the other hand , there is a sharp peak at , then for small , so long as is larger than the width of the peak . ( The factor of included in the scaling is approximate; it is a placeholder for an effective firing rate that depends on the indices and . It is , however , sufficiently accurate for our purposes . ) A similar relationship exists between the third order correlogram and the correlation coefficient . Thus , is also independent of in the small limit , whereas if the central peak is sharp it scales as . The upshot of this analysis is that the shape of the cross-correlogram has a profound effect on the correlation coefficients and , therefore , on . What is the shape in real networks ? The answer typically depends on the physical distance between cells . If two neurons are close , they are likely to receive common input and thus exhibit a narrow central peak in their cross-correlogram . Just how narrow depends on the area . Early in the sensory pathways , such as retina [29]–[31] and LGN [32] , peaks can be very narrow—on the order of milliseconds . Deeper into cortex , however , peaks tend to broaden , to at least tens of milliseconds [33] , [34] . If , on the other hand , the neurons are far apart , they are less likely to receive common input . In this case , the correlations come from external stimuli , so the central peak tends to have a characteristic width given by the temporal correlation time of the stimulus , typically 100 s of milliseconds . Although clearly both kinds of cross-correlograms exist in any single population of neurons , it is convenient to analyze them separately . We have already considered networks in which the cross-correlograms were broad and perfectly flat , so that the correlation coefficients were strictly independent of bin size . We can also consider the opposite extreme: networks in which the the cross-correlograms ( both second and higher order ) among nearby neurons exhibit sharp peaks while those among distant neurons are uniformly equal to 1 . In this regime , the correlation coefficients depend on : as discussed above , the second order ones scale as and the third as . This means that the arguments of and are large ( see Eqs . ( 42 ) and ( 44 ) ) . From the definition of in Eq . ( 36 ) , in this regime both are approximately linear in their arguments ( ignoring log corrections ) . Consequently , and . This implies that and scale as and , respectively , and so , independent of . Thus , if the bin size is large compared to the correlation time , will be approximately independent of bin size . In this section we derive the expression for given in Eq . ( 25 ) . Our starting point is its definition , Eq . ( 24 ) . It is convenient to define to be a concatenation of the responses in time bins , ( 59 ) where , as in the section “Is there anything wrong with using small time bins ? ” , the superscript labels time , so is the full , temporally correlated , distribution . With this definition , we may write the numerator in Eq . ( 24 ) as ( 60 ) where is the entropy of , the last sum follows from a marginalization over all but one element of , and is the true distribution at time ( unlike in the section “Is there anything wrong with using small time bins ? ” , here we do not assume that the true distribution is the same in all time bins ) . Note that is independent of time , since it is computed from a time average of the true distribution . That time average , which we call , is given in terms of as Inserting this definition into Eq . ( 60 ) eliminates the sum over , and replaces it with . For simplicity we consider the maximum entropy pairwise model . In this case , because is in the exponential family , and the first and second moments are the same under the true and maximum entropy distributions , we can replace with . Consequently , Eq . ( 60 ) becomes This gives us the numerator in the expression for ( Eq . ( 24 ) ) ; using Eq . ( 4 ) to write , the full expression for becomes ( 61 ) where we added and subtracted to the numerator . The first term on the right hand side of Eq . ( 61 ) we recognize , from Eq . ( 6 ) , as . To cast the second into a reasonable form , we define to be the entropy of the distribution that retains the temporal correlations within each neuron but is independent across neurons . Then , adding and subtracting this quantity to the numerator in Eq . ( 61 ) , and also adding and subtracting , we have ( 62 ) The key observation is that if , then Comparing this with Eqs . ( 8a ) and ( 9a ) , we see that is a factor of times larger than . We thus have ( 63 ) Again assuming , and defining , the last term in this expression may be written ( 64 ) Inserting this into Eq . ( 63 ) and using Eqs . ( 4 ) , ( 8a ) and ( 9a ) yields Eq . ( 25 ) . We have assumed here that ; what happens when , or larger ? To answer this , we rewrite Eq . ( 61 ) as ( 65 ) We argue that in general , as increases , becomes increasingly different from , since the former was derived under the assumption that the responses at different time bins were independent . Thus , Eq . ( 25 ) should be considered a lower bound on . | Biological systems are exceedingly complicated: They consist of a large number of elements , those elements interact in nonlinear and highly unpredictable ways , and collective interactions typically play a critical role . It would seem surprising , then , that one could build a quantitative description of biological systems based only on knowledge of how pairs of elements interact . Yet , that is what a number of studies have found . Those studies , however , focused on relatively small systems . Here , we ask the question: Do their conclusions extend to large systems ? We show that the answer depends on the size of the system relative to a crossover point: Below the crossover point the results on the small system have no predictive power for large systems; above the crossover point the results on the small system may have predictive power . Moreover , the crossover point can be computed analytically . This work thus provides a general framework for determining the extent to which pairwise models can be used to predict the behavior of large biological systems . It also provides a useful heuristic for designing experiments: If one is interested in understanding truly large systems via pairwise interactions , then make sure that the system one studies is above the crossover point . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"neuroscience/theoretical",
"neuroscience",
"neuroscience/sensory",
"systems"
] | 2009 | Pairwise Maximum Entropy Models for Studying Large Biological
Systems: When They Can Work and When They Can't |
During development , cell polarization is often coordinated to harmonize tissue patterning and morphogenesis . However , how extrinsic signals synchronize cell polarization is not understood . In Caenorhabditis elegans , most mitotic cells are polarized along the anterior-posterior axis and divide asymmetrically . Although this process is regulated by a Wnt-signaling pathway , Wnts functioning in cell polarity have been demonstrated in only a few cells . We analyzed how Wnts control cell polarity , using compound Wnt mutants , including animals with mutations in all five Wnt genes . We found that somatic gonadal precursor cells ( SGPs ) are properly polarized and oriented in quintuple Wnt mutants , suggesting Wnts are dispensable for the SGPs' polarity , which instead requires signals from the germ cells . Thus , signals from the germ cells organize the C . elegans somatic gonad . In contrast , in compound but not single Wnt mutants , most of the six seam cells , V1–V6 ( which are epithelial stem cells ) , retain their polarization , but their polar orientation becomes random , indicating that it is redundantly regulated by multiple Wnt genes . In contrast , in animals in which the functions of three Wnt receptors ( LIN-17 , MOM-5 , and CAM-1 ) are disrupted—the stem cells are not polarized and divide symmetrically—suggesting that the Wnt receptors are essential for generating polarity and that they function even in the absence of Wnts . All the seam cells except V5 were polarized properly by a single Wnt gene expressed at the cell's anterior or posterior . The ectopic expression of posteriorly expressed Wnts in an anterior region and vice versa rescued polarity defects in compound Wnt mutants , raising two possibilities: one , Wnts permissively control the orientation of polarity; or two , Wnt functions are instructive , but which orientation they specify is determined by the cells that express them . Our results provide a paradigm for understanding how cell polarity is coordinated by extrinsic signals .
For tissues and organs to be properly organized , it is often essential that cell polarity be coordinated among cell groups . In the Drosophila wing , for example , cells are polarized in the same proximal-to-distal orientation to produce hairs pointing distally [1] . Similarly , in the mammalian cochlea , stereociliary bundles form at the outer edge of all hair-producing cells [2] . Such coordinated polarizations are often controlled by the Wnt/PCP ( planar cell polarity ) pathway , which involves the polarized localization of signaling molecules such as Frizzled , Dvl/Dishevelled , and Van Gogh proteins [3]–[5] . One plausible model for cell polarity coordination is that individual cells recognize extrinsic cues that orient their polarity . Although Wnt proteins have been considered candidates for orienting molecules , their functions in regulating cell polarity are not well understood . In Drosophila , where PCP phenotypes are lacking in some Wnt mutants , including wingless , Wnt proteins are not believed to be required for regulating PCP . Instead , PCP is coordinated by communication between neighboring cells , although the presence of extrinsic cues is still anticipated . In the mammalian cochlea , Wnt7a has been suggested as a cue to instruct cell polarity orientation , based on overexpression and inhibitor studies in organ cultures [6] . However , there is no PCP phenotype in the cochlea of Wnt7a null mice [6] , suggesting that other Wnt proteins function redundantly with Wnt7a . In Xenopus and zebrafish , Wnt11 and Wnt5 are required for convergent extension movements during gastrulation , which is also regulated by the PCP pathway [7] , [8] . However , these Wnts are thought to function permissively in cell polarization , rather than providing a directional cue . The presence of global extrinsic cues that orchestrate polarity orientations has not been shown in any organism . In C . elegans , the Wnt/ß-catenin asymmetry pathway controls asymmetry in most somatic cell divisions occurring along the anterior-posterior axis [9] . In this regulation , Wnt pathway components localize asymmetrically . For example , after asymmetric divisions , the ß-catenin homologs WRM-1 and SYS-1 accumulate in the posterior daughter nuclei , while POP-1/TCF localizes more to the anterior than posterior nuclei [10] . Such localization has been observed in most cell divisions , during which cells are accordingly polarized in the anterior-posterior orientation . But how the polarity orientation is determined is not known , except in a few cases . We have shown that Wnts instructively orient the polarity of the EMS blastomere in embryos and of the T cell in larvae [11] . It has also been suggested that MOM-2/Wnt and LIN-44/Wnt expressed in the anchor cell orient the polarity of the P7 . p cell , while EGL-20/Wnt expressed near the anus antagonizes these Wnts to orient the P7 . p polarity in the opposite orientation [12] . However , it is not known whether or how Wnts globally regulate the polarity of many other cells . To elucidate the mechanisms of polarity coordination , we focused on a population of epithelial stem cells called seam cells ( V cells ) . At the L1 stage , the six seam cells V1–V6 are positioned on each lateral side of the animals , and repeatedly undergo self-renewing asymmetric cell divisions in each larval stage to produce anterior daughters that fuse with the hypodermal syncytium ( hyp7 ) and posterior daughters that remain as seam cells ( Figure 1A ) [13] . As with many other cells , the polarity of seam cells is controlled by the Wnt/ß-catenin asymmetry pathway [14] , [15] , which determines the polarized localization of WRM-1/ß-catenin to the posterior daughter nuclei . Among seam cells , the polarity of the V5 cell reverses fairly frequently in egl-20/Wnt mutants [16] . However , Wnt gene regulation of the polarity of other seam cells has not been reported . By analyzing various compound Wnt mutants , including quintuple Wnt mutants , we found that the Wnt genes lin-44 , cwn-1 , egl-20 , and cwn-2 are redundantly required to coordinate the orientation of seam cell polarity at the L1 stage , but three of their receptors are essential for generating the cells' polarity in the first place . The Wnt genes are expressed either anterior or posterior to the seam cells , and each one alone can determine the polarity orientation . Our results provide an important basis for elucidating undiscovered mechanisms in the coordination of cell polarity by Wnt genes .
To analyze the polarity of the seam cell divisions , we used elt-3::GFP , which is expressed in hyp7 but not in seam cells [17] , [18] ( Figure 1A , 1D ) . About 1 hour after the division of seam cells ( V1–V6 ) at the L1 stage in wild-type animals , the anterior daughters fuse with hyp7; their nuclei immediately begin fluorescing like those of hyp7 cells , because they incorporate GFP from the hyp7 cell . Therefore , we can unambiguously determine the daughter cell fates , from which we can deduce the division polarity type ( normal , reverse , or loss of polarity ) ( Figure 1B ) . ( In Figure 1C and the figures presented below , the proportions of the polarity types of individual seam cells were mathematically converted to RGB colors as described in the figure legend . ) The C . elegans genome contains five Wnt genes , lin-44 , cwn-1 , cwn-2 , egl-20 , and mom-2 . To understand how seam cell polarity is regulated , we first analyzed the phenotypes of animals with mutations in one of the five Wnt genes . Except for egl-20 , in which the V5 polarity was reversed [16] , the Wnt mutants showed weak phenotypes , if any ( Figure 2 ) , raising the possibility that multiple Wnt genes redundantly regulate seam cell polarity . To test this hypothesis , we constructed a strain with mutations in all five Wnt genes ( quintuple Wnt mutants ) . Because a combination of three Wnt null mutations ( cwn-1 , cwn-2 and mom-2 ) causes complete embryonic lethality [19] , we used the temperature-sensitive ( ts ) mutation mom-2 ( ne874 ) , in which endoderm production is strongly affected during embryogenesis at restrictive temperatures [20] . Because quintuple Wnt mutants only occasionally reproduce , even at permissive temperatures , we could analyze only 7 animals born from homozygous quintuple Wnt mutants . In addition , we analyzed quintuple mutants from mothers heterozygous for cwn-2 , egl-20 and mom-2 , as shown in Figure 2: lin-44; cwn-1; egl-20 cwn-2 ( +M ) ; mom-2 ( ts ) ( +M ) . We found that the polarity of all the seam cell divisions was abnormal in the quintuple Wnt mutants ( Figure 1E and Figure 2 , p<0 . 01 in V1–V6 by Fisher's exact test ) , indicating that multiple Wnts are redundantly required for appropriately oriented seam cell polarity . Although the phenotypes varied among the cells , the polarity tended to be either normal or reversed , and symmetric division was less frequent ( represented by the absence of yellowish colors in Figure 2 ) . Although we cannot exclude residual mom-2 activity in quintuple mutants with the mom-2 ( ts ) allele , the results suggest that seam cells are mostly polarized even in the absence of Wnt functions . To determine which combinations of Wnt genes are required for the properly oriented polarity of individual seam cells , we analyzed them in double , triple , or quadruple Wnt mutants . The phenotype of quadruple Wnt mutants ( lin-44; cwn-1; egl-20 cwn-2 ) was quite similar to that of quintuple mutants ( Figure 2; p>0 . 1 in V1–V6 for the abnormalities ) , suggesting that mom-2 has only minor functions , if any , in seam cell polarity . Next , we constructed triple Wnt mutants from these four Wnt mutations . Through these analyses , we found three distinct regulations that depended on cell type , grouped into V1–V4 , V5 , and V6 . To confirm that Wnt genes regulate the Wnt/ß-catenin asymmetry pathway , we analyzed POP-1/TCF localization in triple Wnt mutants ( cwn-1; egl-20 cwn-2 ) , in which the polarity of V1–V5 is disrupted . We found that POP-1 asymmetry was abnormal in V1–V5 cells in the triple Wnt mutants ( Figure 3A , 3D , 3E; p<0 . 01 in V1–V5 ) . As judged by elt-3::GFP expression ( Figure 2 ) , polarity reversal is more frequent than loss of polarity ( represented by purplish colors in Figure 3A ) . Therefore , these Wnt genes control seam cell polarity via the Wnt/ß-catenin asymmetry pathway . Since seam cells are polarized in a planar ( anterior-posterior ) orientation in contact with each other before division , interactions between neighboring cells might coordinate their polarity , as with PCP regulation in the Drosophila wing . However , in triple Wnt mutants ( cwn-1; egl-20 cwn-2 ) , we did not observe any significant correlation of polarity reversal between neighboring seam cell pairs ( data not shown ) . In addition , the polarity of the V5 cell division is not affected by laser ablation of the V6 cell [16] . Furthermore , we found that the polarity of the seam cell divisions was normal in mutants of the putative PCP components vang-1/Van Gogh ( tm1422 ) ( n = 20 ) and prkl-1/Prickle ( ok3182 ) ( n = 20 ) ( the phenotype of vang-1 was analyzed using scm::GFP , as described in Materials and Methods ) . Therefore , it is likely that the polarity of each seam cell is independently controlled by Wnt genes . To understand how Wnts control polarity , it is important to identify their receptors . The C . elegans genome contains six Wnt receptors , four Frizzled ( MIG-1 , LIN-17 , CFZ-2 , and MOM-5 ) , one Ror ( CAM-1 ) [22] , and one Derailed ( LIN-18 ) family members . Among these , it has been reported that cam-1/Ror mutations reverse the polarity of the V1 and V2 cell divisions at a low frequency [23] and that lin-17/Frizzled mutants cause mostly symmetric divisions of a tail seam cell called a T cell [24] . First , we analyzed single mutants of each receptor gene . Similar to cam-1 , the mom-5 mutation weakly affected the polarity of the V1 and V2 divisions ( p<0 . 01 in V1 and V2 ) . V1–V2 defects were enhanced in mom-5 cam-1 ( RNAi ) animals ( p<0 . 01 in V1 and V2 by the comparison with mom-5 mutants or cam-1 ( RNAi ) animals ) , indicating that MOM-5 and CAM-1 redundantly control V1–V2 polarity ( Figure 4 ) . Single mutants for the other receptors showed only minor defects , if any , in the polarity of seam cell divisions , suggesting that their functions are redundant for V3–V6 . Since lin-17 and mom-5 show a strong genetic interaction in gonad development [19] , we next analyzed lin-17 mom-5 double Frizzled mutants and found that the polarity of all the seam cell divisions was abnormal ( p<0 . 01 in V1–V6 ) ( Figure 4 ) . The mig-1 , cfz-2 , or lin-18/Derailed mutations slightly modified the phenotype of the lin-17 mom-5 mutants . However , since the mig-1; cfz-2; lin-18 triple mutants showed nearly normal polarity ( Figure 4 ) , these receptors are not essential and are likely to function redundantly with other receptors . Next , we constructed lin-17 mom-5; cam-1 triple mutants , and found that this combination was embryonically lethal . Therefore , we inhibited cam-1 by RNAi in lin-17 mom-5 , and found that all seam cell divisions were symmetric at high penetrance ( Figure 1F ) ( p<0 . 01 in V1–V6 and p<0 . 01 in V1–V4 , p<0 . 05 in V5 , p>0 . 1 in V6 for symmetric division by the comparison with wild type and lin-17 mom-5 , respectively; represented by yellowish colors in Figure 4 ) . These results indicate that LIN-17 , MOM-5 , and CAM-1 are the main receptors that redundantly regulate seam cell polarity , whereas the receptors MIG-1 , CFZ-2 , and LIN-18 weakly affect polarity in the absence of the main receptors . Most importantly , the phenotype of lin-17 mom-5; cam-1 is clearly distinct from that of quintuple Wnt mutants in which polarity orientation is randomized ( p<0 . 01 in V1–V6 for symmetric division ) . These results suggest that Wnt receptors can function even in the absence of Wnts to generate polarity , while Wnts are required to orient polarity . It was previously suggested that CAM-1 functions as a receptor for CWN-2 [25] , [26] . If this is the case for seam cell polarity , the cwn-2 mutation should have the same or stronger effects than the cam-1 mutation . However , as described above , the cwn-2 mutation alone did not affect the V1 cell , which was affected in cam-1 mutants . Furthermore , the lin-17 mom-5; cwn-2 mutants had a weaker phenotype than lin-17 mom-5 cam-1 ( RNAi ) ( p<0 . 05 in V2 and V3 , p = 0 . 066 in V4 ) ( Figure 4 ) . Therefore , it is unlikely that CAM-1 is a specific receptor for CWN-2 for seam cell polarity . It was reported that cam-1p::GFP is expressed in V cells [23] . We found that lin-17p::LIN-17::GFP is also expressed in all V cells ( Figure S2 ) . To determine whether the receptors functions in seam cells , we expressed LIN-17 specifically in seam cells using the scm promoter ( scm::LIN-17::GFP ) [11] and found that the polarity defects in the lin-17 mom-5 animals were significantly rescued in V1–V3 and V6 ( p<0 . 01 in V1 , V3 , and V6 , p<0 . 05 in V2 , p>0 . 1 in V4 and V5 ) ( Figure 4 ) , suggesting that at least LIN-17 among the Wnt receptors functions in seam cells . Wnt genes are expressed in specific regions of the animal , either in the anterior ( CWN-2 ) [25] , [26] or posterior ( LIN-44 , EGL-20 , and CWN-1 ) [16] , [27]–[29] . In addition , EGL-20 forms a posterior–to-anterior gradient [30] . We examined CWN-1 and CWN-2 protein localization by full-length translational fusion constructs ( cwn-1p::CWN-1::Venus or cwn-2p::CWN-2::Venus ) . V4 and V5 polarity defects in cwn-1; egl-20 mutants were rescued by cwn-1p::CWN-1::Venus ( Figure 5G ) , and V1–V4 defects in cwn-1; egl-20 cwn-2 mutants were partly rescued by cwn-2p::CWN-2::Venus ( Figure 5H ) , indicating that these fusion proteins are functional . As reported previously for CWN-1 promoter activity [19] , cwn-1p::CWN-1::Venus was localized to the cytoplasm and around the cell membrane in posterior muscle cells , both dorsal and ventral ( Figure 5A ) . Although there were variations between animals , the signals clearly tended to be stronger in posterior cells than in the middle of the animal , suggesting that CWN-1 expression may form a gradient . Consistent with the previous observation that the cwn-2 promoter is strongly active in the pharynx [25] , [26] , we detected puncta of cwn-2p::CWN-2::Venus mostly around the pharynx . We detected these puncta on the hypodermis , including the seam cells ( Figure 5B , white arrowheads ) , suggesting its diffusion from the pharyngeal region . This is in contrast to cwn-1p::CWN-1::Venus , whose diffusion away from its expressing cells was only occasionally observed ( white arrowheads in Figure 5A ) . Although it was also reported that the cwn-2 promoter is active in the intestine , albeit weaker than in the pharynx [25] , [26] , we detected cwn-2p::CWN-2::Venus puncta only in the anterior region , along the boundary between the intestine and muscle or hypodermis ( Figure 5C , white arrowheads ) . These observations indicate that CWN-2 is mostly distributed to the anterior side of the animal . To confirm that cwn-2 functions in the pharynx , we used a ceh-22 promoter to express cwn-2 in the pharynx of Wnt triple mutants ( cwn-1; egl-20 cwn-2 ) , and found that the phenotype was rescued in V1 , V3 and V4 cells ( p<0 . 05 in V1 , p<0 . 1 in V3 and V4 , p = 0 . 14 in V2 ) ( Figure 5H ) . The weak effects of ceh-22p::CWN-2::Venus compared to cwn-2p: CWN-2::Venus appear to reflect weaker transgene expression . These results suggest that cwn-2 is expressed and functions in the pharynx . Our results indicate that each seam cell except V5 can be polarized by a single Wnt gene expressed either anterior or posterior to the cells . For example , V1 is properly polarized merely by cwn-2 expressed nearby and at its anterior , or by egl-20 expressed posterior to and far from V1 . To determine whether the position of Wnt expression is important in regulating polarity , we expressed Wnt genes ectopically . If Wnts function permissively , abnormal polarity in Wnt compound mutants should probably be rescued irrespective of the location of Wnt expression . If the Wnts were instructive , we expected that ectopic Wnt expression opposite to its normal location would enhance polarity reversals . As reported previously , EGL-20 expressed in the pharynx by the myo-2 promoter can rescue V5 polarity defects in egl-20 mutants [16] . However , since the myo-2 promoter is also weakly active in the posterior region [11] , the appropriate interpretation of these results was uncertain . We first used the hlh-8 promoter to express egl-20 in the M cell , a mesodermal blast cell positioned between the V4 and V5 cells on the right side , in egl-20 mutants [31] . We found that this had no significant effect on V5 cell polarity ( Figure 5F ) , suggesting that egl-20 does not function ( i . e . , it is not produced , secreted , or modified ) in polarization when it is expressed in the M cell . We then expressed cwn-1 or cwn-2 ectopically in the anterior ( using the ceh-22 promoter ) [32] , [33] posterior ( using the egl-20 promoter ) [29] regions in Wnt triple mutants ( cwn-1; egl-20 cwn-2 ) . Surprisingly , the posterior expression of CWN-2 , which is normally expressed in the pharynx , efficiently rescued the triple mutant phenotype ( Figure 5H , p<0 . 01 in V1–V5 ) . Similarly , the anterior expression of CWN-1 , which is normally expressed in the posterior region , appeared to rescue the polarity defects of the V1–V3 divisions ( Figure 5H , V1 p = 0 . 1076 , V2 p<0 . 01 , V3 p<0 . 05 ) . The effect of ceh-22p::CWN-1::Venus was comparable to that of ceh-22p::CWN-2::Venus . These results seem to suggest that the position of Wnt expression is not important and that Wnt functions are not instructive , even though Wnts are required for correct polarity orientation . However , the results can also be explained by assuming that functions of Wnts are determined by the cells that express them ( see Discussion ) . It is noteworthy that , even though ceh-22p::CWN-1::Venus and ceh-22p::CWN-2::Venus express these Wnts from the same ceh-22 promoter , we detected puncta of CWN-2::Venus but not CWN-1::Venus outside of the pharynx ( white arrowheads in Figure 5D , 5E ) . Together with the efficient diffusion of cwn-2p::CWN-2::Venus but not cwn-1p::CWN-1::Venus described above ( Figure 5A and 5B ) , the results suggest that these Wnts have distinct diffusion properties . Because ceh-22p::CWN-1::Venus rescued V1–V3 polarity , CWN-1 is likely to be diffused , but at such low levels that it was undetectable . The difference may reflect CWN-1's lower diffusion or weaker ability to form puncta as compared to CWN-2 . Similar to seam cells , the Wnts regulating the polarity of Z1 and Z4 cells , which are somatic gonad precursors ( SGPs ) , have not been identified . The SGPs have a mirror-symmetric polarity , which is important for producing the mirror symmetry of the C . elegans gonad [34] . POP-1 asymmetry in the Z1 daughters is reversed compared to other cells , including Z4 . POP-1 is higher in the posterior and anterior daughters of Z1 and Z4 , respectively ( Figure 6A , 6G ) [35] . SGP polarity is also regulated by the Wnt/ß-catenin asymmetry pathway [35] , although the involvement of Wnt genes has not been demonstrated . We found that the SGP polarity was not affected in quintuple Wnt mutants from mothers heterozygous for cwn-2 , egl-20 and mom-2 , as judged by the normal POP-1 localization ( Figure 6B , 6G ) and the presence of distal tip cells ( DTCs; data not shown ) . Although we could not analyze the POP-1 asymmetry in the quintuple Wnt mutants from homozygous mothers , all such animals we examined ( n = 85 ) had two gonad arms as in wild type , indicating that normal numbers of DTCs were produced from SGPs . These results suggest that the polarity of SGPs is regulated by Wnt-independent mechanisms . To explore the polarity-regulating mechanisms in SGPs , we used mes-1 mutants , which frequently lack germ cells [36] , to analyze the roles of the germ cells Z2 and Z3 , which are positioned between Z1 and Z4 . In mes-1 mutants lacking germ cells , the polarity of both Z1 and Z4 was abnormal , although the defect in Z4 was weaker than that in Z1 ( Z1 p<0 . 01 , Z4 p<0 . 05 ) ( Figure 6E , 6F , 6G ) . Such defects were not observed in mes-1 mutants that had germ cells ( Figure 6C , 6D , 6G ) . These results suggest that non-Wnt signals from germ cells control SGP polarity and hence regulate the proper organization of the somatic gonad . We also examined the possibility that these germ cell signals function redundantly with Wnts . In quadruple Wnt mutants ( lin-44; cwn-1; egl-20 cwn-2 ) lacking germ cells due to the mes-1 mutation , polarity defects appear to be enhanced in Z4 but not Z1 as compared to mes-1 mutants , although the difference did not reach significance ( Figure 6G ) ( Z1: p = 1 . 0 , Z4: p = 0 . 11 ) , raising the possibility that Z4 polarity is redundantly controlled by Wnts and signals from germ cells . In contrast , the polarity of the Z1 cell appeared not to be affected by the Wnt mutations , and Z1 , in wild type , exhibits a reversed orientation compared with Z4 and the seam cells ( i . e . , POP-1 is higher in the posterior daughter ) . Z1 may therefore be regulated by signals from germ cells but may be insensitive to Wnt signals .
We have shown that seam cell polarity is redundantly regulated by multiple Wnt genes . The V1–V4 and V6 cells are affected only by combinations of three and four Wnt mutations , respectively . Such redundancy has been reported in other organisms [37] . For example , double knockout of Wnt1 and Wnt3a in mice causes much stronger CNS developmental abnormalities than the single knockouts [38] . Because all metazoan species have multiple Wnt genes ( e . g . , 19 in humans ) , our results suggest that Wnt genes in any organism may have undiscovered functions that can not be identified by the inhibition of one or a few of them . The defects observed in Wnt mutations in any combination were mostly randomized ( normal or reverse ) polarity , and less frequently , loss of polarity . Similar observations were reported in a mutant lacking mig-14/Wntless function , which is required for Wnt secretion [39] . Our observations are consistent with a recent report that seam cell numbers are not significantly altered in lin-44; cwn-1; egl-20 cwn-2 animals [15] , since the cell numbers were not affected by random orientations of their asymmetry . Even though quintuple Wnt mutants may contain residual mom-2 activity from the ts allele , our results strongly suggest that functions of at least four Wnts ( lin-44 , cwn-1 , cwn-2 , and egl-20 ) determine the polarity orientation of seam cells . In contrast , cell polarization itself appeared to be Wnt-independent , although we cannot eliminate the possibility that cells were not polarized in the complete absence of Wnt functions . In contrast to the randomized polarity found in compound Wnt mutants , triple receptor mutants ( lin-17 mom-5; cam-1 ) showed a severe loss of polarity . These three receptors are likely to function in the polarity generation that occurs even in the absence of Wnts . Even though the other three receptors ( MIG-1 , CFZ-2 , and LIN-18 ) appear to be involved in regulating polarity , based on genetic interactions with lin-17 mom-5 mutations , their triple mutants showed nearly normal polarity . Therefore , it is likely that LIN-17 , MOM-5 , and CAM-1 function in the regulation of polarity orientation as Wnt receptors in addition to having a role in the polarity generation that occurs even in the absence of Wnts , although their activities may be modified by the other three receptors ( MIG-1 , CFZ-2 , and LIN-18 ) . Consistent with this interpretation , strains with mutations in these three receptors showed polarity reversal: V1 and V2 in cam-1 or mom-5 single mutants , and V6 in lin-17 cam-1 double mutants . Our results strongly suggest the presence of distinct mechanisms for polarity orientation , which is Wnt-dependent , and polarity generation , which can occur independently of Wnts . Our ectopic expression experiments appear to indicate that although Wnt functions are required to correctly orient polarity , those functions are permissive . Assuming Wnts are permissive , how do they control polarity orientation in seam cells ? One model is that Wnts act indirectly through other cells that produce real polarity cues in response to Wnts ( Figure 7A ) . In this case , the same Wnt receptors should function in other cells to produce the cues , and in seam cells to generate polarity . For this model , it is strange that , even though Wnts are apparently present near the seam cells , the Wnt receptor activity to polarize seam cells appears not to be affected by Wnts . Together with our finding that LIN-17 functions in seam cells , this model appears unlikely . A second model is that Wnt receptors function only in seam cells . They have two distinct functions: one to generate polarity via the Wnt/ß-catenin asymmetry pathway , and the other to interpret intrinsic polarity cues ( which might be determined by extrinsic cues ) through an unknown pathway ( orientation pathway ) to generate polarity orientation–but only when they are activated by Wnts ( Figure 7B ) . In the absence of Wnts , the receptors still function to polarize cells , but the intrinsic cues cannot be used , resulting in randomly oriented polarity . Although BAR-1/ß-catenin , which functions downstream of LIN-17 in the migration of the Q neuroblast [40] , appears to be a good candidate for mediating the orientation pathway , bar-1 single mutants have normal seam cell divisions ( H . S . unpublished observation ) . Whatever the mechanism of the orientation pathway is , the key question regarding this model is how Wnts elicit the function of the receptors to activate the orientation pathway without affecting the receptors' function in the Wnt/ß-catenin asymmetry pathway , which generates polarity even in the absence of Wnts . Because Wnts instruct the polarity of some cells ( EMS , T , and P7 . p ) [11] , [12] , it is reasonable to imagine that Wnts also instruct seam cells . Assuming that Wnts are instructive , how are the results of ectopic expression explained ? One model would be that Wnts' functions depend on the cells that express them . For instance , CWN-2 , which is expressed in the pharynx , might receive some specific modification , say , “anterior modification , ” whereas CWN-1 , which is expressed in the posterior region , might receive a different modification , say , “posterior modification” ( Figure 7C ) . When cells receive CWN-2 with the anterior modification from their anterior side , they recognize the direction of the Wnt source as “anterior” and localize their signaling components accordingly ( e . g . POP-1 in the anterior daughter nuclei ) . When CWN-1 is ectopically expressed in the pharynx , it may receive anterior modification , like CWN-2 , and function like CWN-2 to instruct normal seam cell polarity , rather than functioning like CWN-1 with posterior modification . This model can explain EGL-20's lack of function when expressed in the M cell–assuming that the M cell cannot modify EGL-20 . In addition , we have reported that LIN-44 expressed by the egl-5 promoter ( egl-5::LIN-44 ) anterior to a T cell can efficiently reverse T cell polarity in the absence of endogenous LIN-44 expressed at the posterior of the T cell [11] . However , in the presence of endogenous LIN-44 ( LIN-44 is expressed in both sides of the T cell ) , the effect of egl-5::LIN-44 is quite weak despite egl-5's promoter activity being stronger , as judged by egl-5::GFP , than that of the lin-44 promoter , as judged by lin-44::GFP . This observation is also consistent with the model that Wnt functions depend on the cells that express it . Another possibility for cell-specific Wnt functions is that Wnt-expressing cells or their neighbors express specific cofactors of Wnts that bind tightly to Wnts and determine their functions . Even though there is no direct evidence for the above models , and other explanations may be possible , our results suggest the presence of novel mechanisms that control the orientation of cell polarity . Such mechanisms , as well as the redundancy of Wnt proteins , may also explain Wnt functions that control cell polarity in other organisms .
N2 Bristol was used as the wild-type strain [41] . The animals were cultured at 22 . 5°C , except for strains containing mom-2 ( ne874ts ) [20] . The following alleles were used: lin-44 ( n1792 ) ( nonsense ) [42]; cwn-1 ( ok546 ) ( deletion ) [43]; cwn-2 ( ok895 ) ( deletion ) [43]; egl-20 ( n585 ) ( missense , but behaves like null ) [44]; mom-2 ( or309 ) ( deletion ) [45]; mom-2 ( ne874ts ) ( missense ) ; lin-17 ( n3091 ) ( nonsense ) [24]; mig-1 ( e1787 ) ( nonsense ) [28]; mom-5 ( ne12 ) ( nonsense ) [46]; cam-1 ( gm122 ) ( nonsense ) [23]; cfz-2 ( ok1201 ) ( deletion ) [43]; lin-18 ( e620 ) ( nonsense ) [47]; mes-1 ( bn7 ) [48]; vang-1 ( tm1422 ) ( deletion ) [49]; prkl-1 ( ok3182 ) ( deletion ) ; and lin-22 ( n372 ) ( missense ) [50] . Molecular information of cwn-1 ( ok546 ) , cwn-2 ( ok895 ) , mom-2 ( or309 ) cfz-2 ( ok1201 ) , vang-1 ( tm1422 ) and prkl-1 ( ok3182 ) is described in http://www . cbs . umn . edu/CGC/index . html . The genotypes of compound strains were confirmed either by PCR ( cwn-1 , cwn-2 , cfz-2 , vang-1 , and prkl-1 ) , sequencing ( lin-44 , egl-20 , mom-2 ( ne874ts ) , mig-1 , cam-1 , and lin-18 ) , or by their phenotype ( Psa for lin-44 , maternal effect lethal for mom-2 ( or309 ) and mom-5 , bivulva for lin-17 , and maternal effect sterile for mes-1 ) . The strains containing mom-5 ( ne12 ) mom-2 ( or309 ) were maintained as heterozygotes over hT2[qIs48] and nT1[qIs51] , respectively , which are marked by GFP expression . Non-fluorescent homozygotes were analyzed for their phenotype . The quintuple Wnt mutants were maintained at 15°C as lin-44; cwn-1; cwn-2 egl-20/nT1[qIs51]; mom-2/nT1 . The phenotype was analyzed in non-fluorescent homozygotes or their progeny , which were shifted to 25°C during late embryogenesis . RNAi for cam-1 was performed by feeding RNAi ( Ahringer Lab RNAi protocol , http://www . gurdon . cam . ac . uk/~ahringerlab/pages/rnai . html ) using the RNAi clone I-6L11 . In most cases , the polarity of seam cell divisions was analyzed using elt-3::GFP ( vpIs1 ) [18] expressed in hyp7 , except for cfz-2 single mutants , which were analyzed by scm::GFP ( wIs51 ) [51]; lin-18 single mutants , analyzed by ajm-1::GFP ( ncIs13 ) [52]; vang-1 , analyzed by wIs51; and compound strains with lin-18 mutations , also analyzed by wIs51 . GFP markers , including GFP::POP-1 ( qIs74 ) [53] , cwn-1p::CWN-1::Venus , and cwn-2p::CWN-2::Venus , were analyzed by confocal microscope ( Zeiss LSM510 ) . Statistical analysis was performed with the Fisher exact test . cwn-1p::CWN-1::Venus and cwn2p::CWN-2::Venus were constructed from PCR fragments containing their promoter regions ( 1 . 8 kb and 6 . 1 kb , respectively ) , and the entire coding regions were amplified by PCR from the fosmids WRM0620cE04 and WRM0622bE06 , respectively , inserted into a pPD95 . 75::wVenus derived from pPD95 . 75 ( a gift from A . Fire ) and containing the Venus gene optimized for C . elegans codon usage in place of the GFP gene . The plasmids ceh-22p::CWN-1::Venus and ceh-22p::CWN-2::Venus contain a ceh-22 promoter fragment from pCW2 . 1 [32] and full-length cDNAs ( yk236a10 and yk343h8 , respectively ) inserted into pPD95 . 75::wVenus . For egl-20p::CWN-2::Venus , a 6 . 8 kb egl-20 promoter fragment and yk343h8 were inserted into pPD95 . 75::wVenus . The plasmids were injected as described previously [54] , with pBlueScript SK+ DNA and the co-injection markers unc-76-rescuing plasmids for cwn-1p::CWN-1::Venus and cwn2p::CWN-2::Venus injected into unc-76 ( e911 ) for expression analyses; ceh-22p::GFP for the cwn-1p::CWN-1::Venus rescue experiment; and mec-4::GFP [55] for ceh-22p::CWN-2::Venus , ceh-22p::CWN-1::Venus , and the cwn-2p::CWN-2::Venus rescue experiment . Either mec-4::GFP or egl-5::GFP [11] was used for egl-20p::CWN-2::Venus . The hlh-8p::EGL-20 plasmid contains a 1 . 3 kb PCR fragment just upstream of the start codon of the hlh-8 gene from a cosmid C02B8 and an egl-20 cDNA ( yk1183a10 ) subcloned into pPD49 . 26 ( a gift from A . Fire ) . lin-17p::LIN-17::GFP was constructed by inserting a HindIII-KpnI fragment of pSH6 [24] into pPD95 . 77 ( a gift from A . Fire ) . | Proper functions and development of organs often require the synchronized polarization of entire cell groups . How cells coordinate their polarity is poorly understood . One plausible model is that individual cells recognize extrinsic signal gradients that orient their polarity , although this has not been shown in any organism . In particular , although Wnt signaling is important for cell polarization , and Wnt signal gradients are important for the coordinated specification of cell fates , the Wnts' involvement in orienting cell polarity is unclear . In the nematode Caenorhabditis elegans , most asymmetrically dividing mitotic cells are polarized in the same anterior-posterior orientation . Here we show that multiple Wnt proteins redundantly control the proper orientation of cell polarity , but not for polarization per se , in a group of epithelial stem cells . In contrast , Wnt receptors are indispensable for cells to adopt a polarized phenotype . Most stem cells are properly oriented by Wnt genes that are expressed either at their anterior or posterior side . Surprisingly , Wnt signals can properly orient stem cell polarity , even when their source is changed from anterior to posterior or vice versa . Our results suggest the presence of novel mechanisms by which Wnt genes orient cell polarity . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"developmental",
"biology",
"genetics",
"biology",
"genetics",
"and",
"genomics"
] | 2011 | Multiple Wnts Redundantly Control Polarity Orientation in Caenorhabditis elegans Epithelial Stem Cells |
Recent analyses of human genome sequences have given rise to impressive advances in identifying non-synonymous single nucleotide polymorphisms ( nsSNPs ) . By contrast , the annotation of nsSNPs and their links to diseases are progressing at a much slower pace . Many of the current approaches to analysing disease-associated nsSNPs use primarily sequence and evolutionary information , while structural information is relatively less exploited . In order to explore the potential of such information , we developed a structure-based approach , Bongo ( Bonds ON Graph ) , to predict structural effects of nsSNPs . Bongo considers protein structures as residue–residue interaction networks and applies graph theoretical measures to identify the residues that are critical for maintaining structural stability by assessing the consequences on the interaction network of single point mutations . Our results show that Bongo is able to identify mutations that cause both local and global structural effects , with a remarkably low false positive rate . Application of the Bongo method to the prediction of 506 disease-associated nsSNPs resulted in a performance ( positive predictive value , PPV , 78 . 5% ) similar to that of PolyPhen ( PPV , 77 . 2% ) and PANTHER ( PPV , 72 . 2% ) . As the Bongo method is solely structure-based , our results indicate that the structural changes resulting from nsSNPs are closely associated to their pathological consequences .
The introduction of large-scale genome sequencing technologies has dramatically increased the number of single nucleotide polymorphisms ( SNPs ) in public databases . For example , the NCBI ( National Center for Biotechnology Information ) dbSNP database [1] , which is a major repository of human SNPs , contained data about ten thousand unique human SNPs as of Build 106 in 2002 . By October 2007 , there were about six and half million validated unique human SNPs , as of Build 128 . Although the progress of collecting SNP data has been impressive , the pace at which disease-related SNPs are annotated is much slower . So far , only a few thousand SNPs have been associated with a human genetic disorder in the OMIM ( Online Mendelian Inheritance in Man ) database [2] . Further efforts are thus required to identify disease-associated SNPs in order to understand their effects on human health . Genetic variations , such as SNPs , are likely to contribute to susceptibility to complex diseases such as cancer [3] . Single nucleotide variations in the coding regions that lead to amino acid substitutions , the so-called non-synonymous SNPs ( nsSNPs ) , may be associated with a modulation of protein function . For example , extensive studies on point mutations in P-glycoprotein have shown that amino acid variations regulate its substrate specificity and lead to a variation of drug disposition among individuals [4] . As a consequence , attention has been focused on the study of the relation between nsSNPs and disease as well as predicting their phenotypic effects . Some early approaches exploited position-specific evolutionary information contained in multiple sequence alignments [5] , [6] . Others have used predictive features of sequence and structure [7] , [8] , or machine learning algorithms [9]–[11] to classify SNPs . In addition , there are approaches that annotate nsSNPs at a genomic scale , such as LS-SNP [12] . Previous analyses have shown that methods that apply only sequence information may suffer significant reductions in accuracy when fewer than ten homologous sequences are available for the target protein [8] . Sunyaev et al . [13] have shown that disease-causing mutations often affect intrinsic structural features of proteins , while in an important study Wang and Moult [14] have demonstrated that most disease-associated mutations appear to affect protein stability rather than interfere directly with protein interactions . Following these results , others have focused on comparing the structures of wild-type and mutant-type proteins [14] , [15] or have estimated the change of protein stability by using environment-specific amino acid substitution matrices that are derived from the three-dimensional structures of homologous proteins [16] . For analyzing structural effects of nsSNPs , we have developed an approach , Bongo ( Bonds ON Graph , http://www-cryst . bioc . cam . ac . uk/˜tammy/Bongo ) , which uses graph theoretic measures to annotate nsSNPs . Graph theory has found many applications in the study of protein structures during the past two decades . For example , Ahmed and Gohlke used graphs to identify rigid clusters for modelling macromolecular conformational changes [17]; Canutescu and colleagues have predicted side-chain conformations by partitioning graphs in which vertices represent residues [18]; Vendruscolo and colleagues applied small-world networks to identify key residues that are important for protein folding [19]; Jacobs , Thorpe and their colleagues used graphs to describe bond-bending networks between atoms , so identifying the rigid and flexible regions in the proteins [20] , [21]; Kannan [22]; and Brinda and Vishveshwara [23] used the graph spectral method to identify side chain clusters that are important for protein folding and oligomerisation ; Sol and colleagues used graphs to identify key residues for allosteric communication and modular connection by the edge betweenness algorithm [24] , [25] . Bongo uses graphs to represent residue-residue interaction networks within proteins and to assign key residues that are important for maintaining the networks . The novelty lies in the application of a graph theory concept , vertex cover , by which key residues are identified for analyzing structural effects of single point mutations . Here we begin by describing the use of interaction graphs to represent protein structures . We then introduce the ‘key residues’ that Bongo uses to evaluate structural impacts of point mutations , and explain their roles in terms of stabilising protein structures . We further describe the algorithm of Bongo , where a graph concept vertex cover was adapted to identify key residues , and we calibrate Bongo over eight single point mutations that result in a range of different structural changes in the p53 core domain . We evaluate the false positive rate of Bongo for 113 mutations where wild-type and mutant-type crystal structures have been demonstrated to have negligible differences in backbone conformation . Eventually , we evaluate the performance of Bongo by testing its ability to distinguish disease- and non-disease-associated nsSNPs in protein structures in the PDB ( Protein Data Bank ) [26] . Based on the benchmark results , we also analyse the percentage of disease-associated nsSNPs that are likely to cause structural effects in proteins .
A point mutation in a protein may often give rise only to a rearrangement of amino acid side chains near the mutation site , although sometimes a more substantial movement of polypeptide backbone locally or globally results . The former changes can be analysed by looking at the inter-residue interactions that a mutation creates or abolishes between its neighbouring residues . However the same approach may not be applicable to the latter , since simply paying attention to interactions immediately around a mutation site is not sufficient to predict structural effects on a larger scale . In order to understand structural changes at a longer distance , we represent a protein as a residue-residue interaction graph , in which vertices represent residues and edges represent interactions between residues ( Figure 1 ) ( see more details in Methods ) . Of course , molecular dynamics calculations provide a powerful tool for identifying the impact of point mutations on the stability of the native states of proteins . However , these simulations are often time-consuming and require large computer power . Thus we have developed Bongo to provide an alternative approach by operating on interaction graphs , which are computationally more convenient . In our model , residue-residue interactions occur either through direct connection or through indirect links that involve intermediate residues . Such connectivity is based on ‘key residues’ that are important in maintaining the overall topology of the network , and thus the stability of the folded structure . These key residues eventually serve as reference points to evaluate whether a mutation can induce structural changes in a protein away from the mutation site . Bongo measures the impact of a mutation according to its effects on key residues; it formulates the structural changes in a protein as changes of the key residues in a corresponding interaction graph . Here we adapt a variant of the vertex cover , defined in graph theory as a minimum set of vertices ( residues ) that are crucial to forming all the edges ( interactions ) , to represent the key residues . In Figure 2 , we illustrate the notion of key residues and introduce the use of the difference between the vertex cover of wild and mutant type interaction graphs as a measure of the effects of a mutation . The example here is residue Y35 of protein 1BPI , a key residue forming several relatively strong interactions including pi-cation interactions with residues R20 and N44 and a hydrophobic interaction with residue A40 ( Figure 2A and 2B ) . The mutation Y35G removes this amino acid from the set of key residues in the graph ( Figure 2C ) as its original interactions with other secondary structure elements no longer exist . Hence , residue 35 is no longer a key residue in the mutant interaction network . Therefore , this mutation is considered structurally damaging by Bongo; we discuss the exact criteria under which a mutation is deemed damaging below . Bongo derives the interaction graph of a protein by considering each residue as a vertex and each residue-residue interaction , including hydrogen bonds , π–π , π–cation , and hydrophobic interactions , as an edge . The weight on each edge differs according to the total number of cross-secondary structure interactions as well as number of interactions with individual residues . The weighting scheme was calibrated against eight disease-associated mutations in the p53 core domain analysed by Fersht and co-workers [27] , [28] , as shown in Table 1 . The optimised weighting of inter-secondary structure interactions is 0 . 8 , 0 . 8 , 0 . 8 , 2 . 0 and 2 . 0 for H-bonds , π–π , π–cation , hydrophobic interaction , and hydrophobic core respectively . For internal interactions , H-bonds , π–π , π–cation interactions were given a weight of 0 . 6 and hydrophobic interaction a weight of 0 . 8 . This distinction between inter and intra secondary structure interactions is used to reflect concerted movement of structural motifs within proteins . Thus , a single interaction loss among two densely interacting structures is less significant than one among two sparsely interacting ones . Based on the above weighting scheme , Bongo defines the key residues as the minimum weighted vertex cover ( see the definition of vertex cover in Methods ) , which represents the minimum necessary residues to establish the interaction network . However , finding the minimum vertex cover is known to be NP-complete and hence efficient algorithms only exist for approximate solutions [29] . Therefore , we use a selection scheme which adopts an approximation algorithm based on the greedy principle to identify the key residues . The approximation algorithm is known to give vertex covers that cost no more than H ( |V| ) , where |V| denotes the size of a vertex set , times than the optimum solution where H ( n ) is the nth harmonic number . Compared to other graph theoretic constructs such as dominating sets [29] , the vertex cover gives an intuitive notion of vertex importance . In fact , we have used more advanced techniques such as spectral decomposition [29] to identify structural information that is related to protein stability change , ΔΔG . However , the results were not better than those obtained by applying the vertex cover approach ( data not shown ) . Indeed , we have observed in some cases ( Figure 3 ) that the change of vertex cover after mutation correlates well with structural data . Therefore , we believe that the vertex cover can serve as a useful approach to estimating protein structural changes . The key residues maintain the interaction networks in a protein , and each is assigned a priority value that measures its importance in determining the overall topology of the network ( see Methods ) . When a point mutation is introduced into a protein , Bongo quantifies its structural effects according to the priorities of key residues affected . Thus we expect key residues , especially those with high priorities , to have important roles in stabilising folded protein structures . In order to check if the priority of key residues reflects their roles in forming structures , we calculated the correlation between the priority and the stability change ( Each key residue was mutated to 19 other amino acids and the stability changes were calculated by I-mutant2 . 0 [30] ( http://gpcr2 . biocomp . unibo . it/˜emidio/I-Mutant2 . 0/I-Mutant2 . 0_Details . html ) , which has accuracy around 80% for predicting stability changes resulting from mutations when the three-dimensional protein structure is known . We consider only mutations that cause |ΔΔG|<3kcal/mol since they affect the stability without totally abolishing the overall structure of the protein . The median number of |ΔΔG|<3kcal/mol is used to calculate the correlation with the priority of key residues in order to avoid data skewness . ) , ΔΔG , of key residues identified from the p53 core domain ( PDB: 1TSR ) . When we considered the top half of the key residues ranked by their priorities , ΔΔG relates to the priority of key residues with a Pearson correlation r = 0 . 61 and a significantly small p-value less than 0 . 001 ( Figure 4A ) . This indicates that the correlation is statistically significant and also shows a good contrast to the low relation ( r = −0 . 04 ) between assumptive priority ( Since the non-key residues do not have priority values , they are assigned values according to those of the key residues that are nearest in the same secondary structures . If a non-key residue is flanked by two key residues , its assumed priority is the average of the priority values of its two neighbours . ) and ΔΔG of non-key residues ( Figure 4B ) . We noticed that the correlation is weaker ( r = 0 . 36 ) when the lower half of key residues , ranked by their priorities , is included . This is likely due to uncertainties in the definitions of key residues that are ranked with lower priorities: Since Bongo stops selecting key residues only when no edges are left in a graph , the key residues that have lower priorities may not have structural meaning but are simply chosen in order to complete the selection process ( covering all the edges/interactions in the graph ) . In an attempt to exclude the uncertain key residues , we analysed how far the correlation is valid by gradually including key residues that have priorities in the lower half , in order of decreasing priorities . There is an acceptable correlation r = 0 . 52 when we consider up to three fourths of overall key residues , which suggests that the bottom one quarter key residues are not reliable indicators of structural effects . Thus Bongo does not consider the bottom quarter key residues so that their uncertainty does not affect the prediction results . The distribution of key residues according to their location in secondary structures ( Figure 4A ) shows that the key residues in β-strands tend to have larger ΔΔGs and priority values compared to those in loops , whereas such differences are less clear for the case of non-key residues ( Figure 4B ) . This suggests that , in general , protein stability should be more vulnerable to mutations in β-strands than those in loops , consistent with the observation that the β-strands in the p53 core domain are the major contributors to the core region of the protein . It also indicates that priority values and ΔΔG of key residues have consistent meanings in terms of protein structure . Since the structures of the mutant proteins are not often available for nsSNPs , Bongo first uses Andante [31] to model the mutant-type protein structure by rearranging the side chain around the mutation site . The structural effects of a mutation are then analysed by comparing the wild-type and mutant-type key residues , denoted as Kwt and Kmt , respectively . If a key residue in Kwt is not found in Kmt , then it is considered to be affected by the mutation . Consequently the overall impact ( I ) of a mutation is calculated according to the key residues affected by the mutation , i . e . ( 1 ) where I is the total impact value , Kj is the priority of each key residue that is in Kwt but not in Kmt . N is the total number of key residues in Kwt , which normalise the size of proteins . Thus each mutation is systematically quantified by its impact value I ( an overview scheme of Bongo is shown in Figure 5 ) . On deriving the impact value , Bongo considers mutations with I>1 to cause structural effects , which is the criterion calibrated over mutations in the p53 core domain . In Figure 3 , we give an example , the mutation Y35G in protein 1BPI , of how a mutation can have significant impact value . In addition to residue Y35 , Bongo also predicts residues R42 to be affected by the mutation ( Figure 3A ) . These two are at the ends of β strands and also in long loops linked to them . These regions undergo conformational changes when the mutation Y35G is introduced into the protein , where the biggest movement ( 4 . 2 Å ) occurs between the wild-type and mutant-type Cα atom of residue G36 ( The movement is measured when the wild-type ( 1BPI ) and the mutant-type ( 8PTI ) are superimposed by their Cα atoms . ) . Since the impact score calculated on the basis of these residues is greater than one , Bongo considers the mutation Y35G to cause structural effects in 1BPI , which corresponds to the experimental result . In order to assess the errors due to the difference of a crystal structure of the mutant and a simulated one , we also compared the key residues of the two structures . It turns out that the differences of key residues between the modelled and the crystal structures are mostly located in the loop region , where structural changes occur when the mutation is introduced into the protein ( Figure 3B ) . The overall distribution of the key residues that are specific for the modelled structure is similar to that of the key residues specific for the crystal structure . This suggests that the structural change at a longer distance can be captured in the interaction graphs by simply modelling a point mutation as rearrangement of side chains neighbouring to the mutation site . In order to calibrate Bongo , we have used experimental data on the tumour suppressor p53 core domain , which is responsible for about 50% of mutations that lead to human cancers [32] . Owing to its importance , the wild-type and many mutant protein crystal structures have been determined . Several studies have been carried out for these point mutations within the domain , and thus make it a good calibration system for predicting structural effects of mutations . Furthermore , the structure of the p53 core domain is inherently unstable with a melting temperature of ∼42–44°C [33] . As a consequence , point mutations that cause either subtle structural changes or more dramatic effects are available for comparison . For our study we identified eight nsSNPs ( Figure 6 ) analysed experimentally by Fersht and co-workers [27] , [28] . These mutations involve several different levels of structural change in the p53 core domain: ( i ) R273H has only a minor effect on the overall structure , with root mean square deviation ( RMSD ) ≤0 . 21Å in Cα positions between wild type and mutant type crystal structures; ( ii ) G245S , R249S , and R248A destabilise the p53 core domain by 1–2 kcal/mol and lead to local structural changes; ( iii ) C242S , H168R , V143A , and I195T destabilise the structure >2 kcal/mol and lead to global unfolding of the protein at body temperature . When the structure 1TSR in PDB was used as a calibration model , Bongo identified all mutations except R273H as causing structural effects in the p53 core domain ( Table 1 ) , which corresponds well with experimental data described in the literature . For comparison we also used PolyPhen [5] to predict the effects of the same mutations . We consider PolyPhen as it uses multi-source data including three-dimensional structures , sequence alignments and SWISS-PROT annotations . Compared to other methods which either focus on protein structure or sequence information , it provides more comprehensive results . Of course , there are other methods that include even more information—for example , LS-SNP [12] also considers functional pathways , domain–domain interfaces , ligand–protein binding—but our purpose is to understand the usefulness of structural information by comparing it with a standard approach that mainly uses sequence and structural information . The results in Table 1 show that PolyPhen predicts all mutations except V143A to be probably damaging . PolyPhen's success in predicting R273H to be damaging is probably a consequence of the fact that R273 is functionally important for binding DNA and thus conserved in sequence for reasons that are not evident from consideration of the structure alone , whereas PolyPhen predicts V143A to be benign , probably as a result of comparatively weaker emphasis on structural information . We further tested the application of Bongo to single point mutations that do not affect protein structure . Our benchmark set included 113 pairs of wild-type and mutant-type crystal structures in which each of them has RMSD in their backbone Cα atoms <0 . 4Å and the lower resolution of the two structures is ≤2 . 2Å ( Dataset S1A ) . We chose these criteria in order to allow for experimental errors in the crystallographic solution of the structures of identical proteins , as suggested in the work of Hubbard and Blundell [35] . The benchmark result shows that Bongo predicts three of the single point mutations to cause structural effects , therefore yields a 2 . 7% false positive rate . Although this result may not be generalised to all the cases , it indeed encourages us to expect a low false positive prediction rate . In the previous sections , we have shown that Bongo is able to predict structural effects of single point mutations with a low false positive rate . Here we further analyse the performance of Bongo in identifying disease-associated nsSNPs . Our test-set contains 506 disease-associated nsSNPs from the OMIM ( Online Mendelian Inheritance in Man ) database [2] and 220 non-disease-associated nsSNPs available in dbSNP database [1] which have no annotations in OMIM . All the nsSNPs in the test-set can be mapped to structures in the PDB ( Dataset S1B and S1C ) since Bongo uses structure as input . For evaluation of Bongo , we calculated its sensitivity and specificity ( with definitions explained in Table 2 ) . By definition , if a method always classifies any mutation as ‘disease-associated’ , it would achieve a sensitivity score of 100% . Similarly , a method could obtain a 100% specificity score by always predicting mutations as “non-disease-associated” . In order to avoid a biased analysis , we also calculated the PPV ( positive predictive value ) and NPV ( negative predictive value; with definitions explained in Table 2 ) ; a better PPV or NPV implies a better performance in predicting positive or negative cases , respectively . The overall test results ( Table 2 ) show that Bongo has PPV and NPV of 78 . 5% and 34 . 5% , respectively , compared to that of PolyPhen of 77 . 2% and 37 . 6% , respectively . This indicates that Bongo and PolyPhen have similar accuracy in predicting disease-associated nsSNPs . Given the fact that PolyPhen also exploits sequence information that may take account of protein interactions with various substrates , macromolecules and other ligands , we believe this shows the potential of using interaction networks which consider structure alone . The similar predictive values suggest that , although the mechanisms by which nsSNPs induce diseases are complicated , structural change is an important factor in most cases . This is consistent with a previous study that shows most deleterious nsSNPs affect protein stability but not functionality [14] , which indicates that structural impact is a more important factor in causing disease . In order to assess the performance of Bongo , we also compared the use of PANTHER [36] , which is verified to have higher accuracy than PolyPhen by using Hidden Markov Model ( HMM ) for sequence scoring . The result shows that PANTHER has the PPV and NPV values ( There are 48 disease-associated and 22 non-disease-associated nsSNPs for which PANTHER did not find an HMM model to do prediction; those nsSNPs are excluded from the calculation of PPV and NPV values . ) comparable to those of PolyPhen and Bongo ( Table 2 ) , which further verifies the evaluation . In addition to the predictive value , Bongo has a low sensitivity ( 28 . 1% ) compared to that of PolyPhen ( 50 . 7% ) and PANTHER ( 76 . 6% ) , and its specificity ( 82 . 4% ) is high compared to that of PolyPhen ( 65 . 8% ) and PANTHER ( 31 . 8% ) . This suggests that , although Bongo has a similar predictive value to that of PolyPhen and PANTHER , Bongo's high specificity and low sensitivity yields many less false positive predictions . We can thus be more confident about the cases that are predicted as disease-associated by Bongo than those predicted by PolyPhen . Regarding the low sensitivity of Bongo , we suppose this is due to the fact that Bongo is not able to predict mutations that only affect the function of proteins , e . g . , the mutations in active or other interaction sites . We may improve Bongo's ability in predicting functional site mutations in the future work . Among the 506 disease-associated nsSNPs in our test-set , Bongo predicted 142 of them to cause structural effects , which suggests that about 28% of nsSNPs that are involved in Mendelian diseases resulting from single protein mutations may cause extensive structural effects in proteins . However , the figure for nsSNPs involved in multigenic diseases like diabetes may not be so high as they exist individually in the population as a whole at high levels , but contribute only rarely to multigenic diseases when occurring with several other nsSNPs . We have developed a method , Bongo , which uses graph theoretic measures to evaluate the structural impacts of single point mutations . Our approach has shown that identifying structurally important key residues in proteins is effective in predicting point mutations that cause extensive structural effects with a substantially lower false positive rate . Furthermore , our approach gives clues about the effects of nsSNPs on the structures of proteins , thus providing information complementary to methods based on sequence . By comparing our approach with PolyPhen and PANTHER in analyzing nsSNPs , we have also shown that structural information can provide results of quality comparable to those that use sequence and evolutionary information in predicting disease-associated nsSNPs .
In the residue-residue interaction graphs , Bongo considers structural information including hydrogen bonds , π–π , π–cation , and hydrophobic interactions , as well as secondary structure information . ( 1 ) Hydrogen bond: we use HBPLUS [37] to calculate hydrogen bonds , using its default settings for positioning hydrogen and minimum angles formed by the donor and acceptor at the hydrogen . ( 2 ) π–π interaction: aromatic side chains are considered to have π–π interaction if they have less than 6 Å between any atoms . We note that more accurate criteria could be applied at the expense of the calculation speed with similar results . ( 3 ) π–cation interactions are identified on the condition that there is a cation within 7 Å of any side chain atoms of an aromatic ring such that the angle between the cation and the normal vector of the aromatic ring is within 60° . The criterion is only an approximate one in order to speed up the overall calculation without sacrificing the accuracy of calibration . ( 4 ) Hydrophobic interactions are weighted according to Voronoi surfaces between non-polar residues calculated by an in-house program , Provat [38] , while hydrophobic cores are identified when a non-polar residue shares non-zero Voronoi surfaces with only non-polar residues . ( 5 ) Secondary structure elements are assigned by DSSP [39] . The weighting of the interactions were optimised by using the Least-Squares Optimisation Tool in MATLAB ( http://www . mathworks . com/products/matlab/ ) , where the best solution was chosen on the basis of the best calibration result over the eight mutations listed in Table 1 . Although calibration was carried out against only eight mutations , the performance of Bongo on the 506 disease-associated nsSNPs , which are distributed in proteins from many different families , is comparable to that of PolyPhen ( Table 2 ) . Since the mutant-type structures are not usually available , we generate them computationally using Andante [31] . Andante predicts the structure by using evolutionary information to define rotamers in clusters of side chains that are structurally compatible , so rearranging the local structure around the mutation site . It should be noted that Bongo does not benefit from sequence information by using Andante , since the rearrangement of local side chains modelled by Andante simply introduces a local rearrangement to the residue-residue interaction network of a protein , which does not affect the overall structure of the interaction graph and is independent of the process of selecting vertex cover . All the structural information is transferred into graphs by using Graphviz ( http://www . graphviz . org/ ) , which is an open source graph visualization project from AT&T Research . Bongo derives the interaction graph of a protein by considering each residue as a vertex and each residue-residue interaction as an edge . More formally , an interaction graph G = ( V , E ) is a graph such that V is the set of residues and E is a set of edges . An edge ( u , v ) is defined between residue u and v if they exhibit one of the following interactions: backbone bonding , hydrogen bonds ( H-bonds ) , π–π , π–cation , and hydrophobic interactions . Each edge is initially given a weight of 1 . We then normalise interactions between two secondary structures by dividing the weight with the total number of cross-secondary structure interactions . Intra-secondary structure interactions are normalised in the same way . For interactions involving a group of residues , namely hydrophobic interactions , we normalise them by the Vonoroi surface area of each residues . Since the key residues capture the vertices that are essential to maintain the interactions , we model them through the vertex cover set of the graph [29] . A vertex cover set S of a graph G = ( V , E ) is the set of vertices such that for every edge ( u , v ) , either u or v is included in S . In the interaction graph terms , this amounts to picking a set of residues that covers every interaction in the graph . In Bongo , since the interactions are weighted , we consider the vertex cover problem G = ( V , E , c ) where c: V → R+ is the function that assigns weight to each vertex . A vertex cover set is said to be minimum if it contains the set of vertices that covers all interactions with smallest possible weight . The algorithm used to select key residues captures the concept of pulling out one piece each time in a tower of wooden pieces , with the difference that in our case the pieces pulled out are key pieces but not redundant ones ( Figure 7 ) : The algorithm reflects the importance of key residues in order of selection: key residues selected in an earlier time are more important , in terms of having higher priorities in maintaining the interaction network , than others that are identified later . Since there is a specific order of choosing vertices , the approximate vertex cover chosen by Bongo for a specific graph will be the same when Bongo repeats the selection process again . Taking advantage of the priorities assigned to each key residue , Bongo eventually quantifies the effect of a point mutation by considering the priority of key residues affected . | Non-synonymous single nucleotide polymorphisms ( nsSNPs ) are single base differences between individual genomes that lead to amino acid changes in protein sequences . They may influence an individual's susceptibility to disease or response to drugs through their impacts on a protein's structure and hence cause functional changes . In this paper , we present a new methodology to estimate the impact of nsSNPs on disease susceptibility . This is made possible by characterising the protein structure and the change of structural stability due to nsSNPs . We show that our computer program Bongo , which describes protein structures as interlinked amino acids , can identify conformational changes resulting from nsSNPs that are closely associated with pathological consequences . Bongo requires only structural information to analyze nsSNPs and thus is complementary to methods that use evolutionary information . Bongo helps us investigate the suggestion that most disease-causing mutations disturb structural features of proteins , thus affecting their stability . We anticipate that making Bongo available to the community will facilitate a better understanding of disease-associated nsSNPs and thus benefit personal medicine in the future . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] | [
"molecular",
"biology/bioinformatics",
"genetics",
"and",
"genomics/bioinformatics",
"computational",
"biology"
] | 2008 | Prediction by Graph Theoretic Measures of Structural Effects in Proteins Arising from Non-Synonymous Single Nucleotide Polymorphisms |
Murid γ-herpesvirus-4 ( MuHV-4 ) promotes polyclonal B cell activation and establishes latency in memory B cells via unclear mechanisms . We aimed at exploring whether B cell receptor specificity plays a role in B cell susceptibility to viral latency and how this is related to B cell activation . We first observed that MuHV-4-specific B cells represent a minority of the latent population , and to better understand the influence of the virus on non-MuHV-4 specific B cells we used the SWHEL mouse model , which produce hen egg lysozyme ( HEL ) -specific B cells . By tracking HEL+ and HEL− B cells , we showed that in vivo latency was restricted to HEL− B cells while the two populations were equally sensitive to the virus in vitro . Moreover , MuHV-4 induced two waves of B cell activation . While the first wave was characterized by a general B cell activation , as shown by HEL+ and HEL− B cells expansion and upregulation of CD69 expression , the second wave was restricted to the HEL− population , which acquired germinal center ( GC ) and plasma cell phenotypes . Antigenic stimulation of HEL+ B cells led to the development of HEL+ GC B cells where latent infection remained undetectable , indicating that MuHV-4 does not benefit from acute B cell responses to establish latency in non-virus specific B cells but relies on other mechanisms of the humoral response . These data support a model in which the establishment of latency in B cells by γ-herpesviruses is not stochastic in terms of BCR specificity and is tightly linked to the formation of GCs .
The murid γ-herpesvirus-4 ( MuHV-4 , also known as MHV-68 or γHV-68 ) has led to valuable insights in understanding human γ-herpesvirus related diseases caused by Epstein-Barr virus ( EBV ) and Kaposi's sarcoma associated herpesvirus ( KSHV ) [1] . Whereas primo infection by γ-herpesviruses can be responsible for lymphoproliferative disorders in immune competent hosts , they are usually well controlled [2] . As with EBV , MuHV-4 is mainly lymphotropic and establishes latency in class-switched and germinal center ( GC ) B cells [3] , [4] . The course of the infection in mice is now well described ( see [5] and [1] ) . Upon intranasal inoculation , infection starts with an acute lung infection controlled by the CD8+ T cell response . The virus then disseminates to secondary lymph organs via serial events of lymphoid/myeloid cellular exchanges [6] where it promotes a CD4-dependent polyclonal B cell response and finally establishes latency in long-lived memory B cells [1] , [5] , [7] , [8] . This polyclonal B cell activation can lead to the emergence of auto-antibodies but MuHV-4 infection is usually not associated with the development of auto-immune diseases or lymphomas in immune competent mice [9] . CD4+ T cells , and in particular follicular helper T cells [10] , have been shown to be essential for the establishment of MuHV-4 latency . Antibody-mediated depletion experiments [11] , [12] as well as work performed on MHC class II deficient mice [13] ( which are CD4+ T cells deficient ) have led to similar observations , that the absence of CD4+ T cells leads to lower latency levels . On the virus side , few proteins have been shown to be involved in the establishment of latency [1] . Among them , M2 has received particular interest for its ability to interfere with B cell activation . Studies performed with M2-deficient MuHV-4 have shown its essential role in the establishment of latency , although it is not required for acute lung infection [14] , [15] . Biochemical analysis have established that M2 is able to interact with the Fyn/Vav , Plcγ2 and PI3K pathways , involved in BCR signaling [16]–[18] . In vivo , B cells infected by M2-deficient MuHV-4 have been shown to acquire a GC phenotype comparable with the WT virus , but were unable to class-switch and differentiate into plasma cells [19] . MuHV-4 LANA has recently been shown to stabilize cellular Myc and promotes its activity , leading to B cell proliferation , a process required for GC formation and viral latency [20] . The lower level of viral latency observed in mice deprived of CD4+ T cells as well as with M2-deficient MuHV-4 are good examples showing that the establishment of MuHV-4 latency relies on mechanisms that mix the physiologic B cell response and the intervention of viral modulators . Several questions remain to be explored to better understand this complex interaction: How does MuHV-4 trigger a polyclonal B cell activation ? Are latently infected B cells also polyclonal , or is latency restricted to MuHV-4 specific B cells ? Finally , what are the respective roles for the virus and the B cells in the establishment of latency ? Until recently , those were difficult questions to address experimentally because of two major hurdles on the virus and the B cell sides . At peak of latency ( ∼14 days post-infection in C57BL/6 mice ) , latently infected B cells represent a low percentage of total B cells [3] , [4] and tracking these cells was impossible until the development of a YFP expressing MuHV-4 [4] . On the B cell side , questions concerning BCR specificity are delicate to address due to the enormous diversity of the B cell repertoire and to the difficulty to trace one clonal population , but major improvement was made with the development of the switch hen egg lysozyme mice ( SWHEL ) [21]–[24] . Based on the MD4 model [25] , SWHEL mice have been engineered to contain up to ∼10% of HEL-specific B cells . But contrary to MD4 mice , SWHEL HEL+ B cells can perform GC reactions in a competitive environment , class-switch and differentiate in long-lived memory B cells . Moreover , HEL-specific ( HEL+ ) and non-specific ( HEL− ) populations can easily be distinguished by direct staining of the BCR with fluorescently labeled HEL . In the present study , we aimed at clarifying the role of BCR specificity in the establishment of MuHV-4 latency in B cells . Taking advantage of the SWHEL mice and the YFP-MuHV-4 we designed experiments to study in parallel how MuHV-4 influences a normal B cell repertoire ( HEL− ) and a clonal population of non-virus specific B cells ( HEL+ ) in order to determine in which population latency is established and how this relates with B cell activation .
To evaluate the frequency of MuHV-4 specific B cells in infected and non-infected populations , we challenged C57BL/6 mice with YFP-MuHV-4 . Infected and non-infected CD19+ B cells were sorted at 14 dpi based on their YFP expression and used in an ELISPOT assay to evaluate the number of total IgGs and anti-MuHV-4 IgGs secreting cells ( Figure 1A ) . Both YFP− and YFP+ B cells showed a low frequency of virus-specific antibody-secreting cells ( ASC ) cells when compared to the total ASC . That is , ∼10% of total ASC for the YFP− , and ∼1% for the YFP+ populations , showing that in both populations the majority of ASC are not MuHV-4 specific . As the frequency of GC B cells ( GL-7+ , CD95+ ) is significantly different between YFP+ and YFP− B cells ( Figure 1B ) , we tried to refine our analysis by sorting infected and non-infected GC cells based on GL-7 and CD95 expression . Yet , purified cells died quickly and could not be used for ELISPOT assays , probably due to anti-CD95 induced apoptosis [26] . Our ELISPOT assay did not include the monitoring of IgGs specific for non-structural proteins . However , it would be unlikely if they accounted for the 90 to 99% of the ASC not detected in our anti-MuHV-4 IgGs assay . These data indicate that latent infection is not restricted to MuHV-4 specific B cells and that the virus is able to promote the activation of non-virus-specific B cells independently of their infection status . In order to explore these two points , we next used the SWHEL mice [21] , which allowed us to monitor MuHV-4 influence on non-virus-specific HEL+ B cells . SWHEL mice were infected with YFP-MuHV-4 and we monitored YFP expression in HEL+ and HEL− B cells 14 dpi ( Figure 2A ) . While we did not detect YFP expression in the HEL+ population , HEL− B cells harbored a frequency of YFP+ cells comparable with what has been previously reported in WT B cells [4] . We confirmed that HEL− B cells are solely latently infected by sorting HEL+ and HEL− B cells on which we monitored reactivation of latent virus by ex vivo explant co-culture assay ( Figure 2B ) and the presence of viral DNA by limiting dilution PCR ( Figure 2C ) . It is important to note that HEL− B cells emerge from a spontaneous replacement of the Vh10 exon encoding for the HEL-specific heavy chain leading to the reconstitution of a polyclonal repertoire [21] , minimizing the impact of genetic differences between these two populations . However , HEL+ B cells only belong to the B-2 lineage , while HEL− differentiate into both B-1 and B-2 B cells [21] . It is unknown in which population of B cells MuHV-4 latency is established . To evaluate if the B-2 bias of the HEL+ B cells would account for their resistance to latent infection , we phenotyped latently infected cells in C57BL/6 mice ( Figure S1 ) . B-2 B cells represented the vast majority of YFP+ B cells , but a small fraction of latently infected cells was also detected in B-1a and B-1b B cells . The proportion of B-2 , B-1a and B-1b in YFP+ and YFP− B cells corresponded to what has been described for naïve animals , with the B-2 lineage being dominant in splenic B cells [27] . Overall , these data make it unlikely that the B-2 commitment of HEL+ B cells would explain their resistance to MuHV-4 latency . Finally , to monitor that MuHV-4 latency in SWHEL HEL− B cells reproduces what has been described in a normal repertoire , we determined the phenotype of YFP+ HEL− B cells ( Figure 3 ) . As for WT B cells ( Figure 1B ) , ∼75% of YFP+ HEL− B cells harbored a GL-7+ CD95+ phenotype , with a minor fraction harboring a plasma cell phenotype and ongoing class-switch ( CD138+ IgM− ) . Together these data show that HEL+ B cells are not latently infected by MuHV-4 while latency takes place normally in HEL− B cells . MuHV-4 poorly infects B cells in vitro [28] , but work by Frederico et al overcame this hurdle by developing an in vitro co-culture assay , and showed that MuHV-4 transits by myeloid cells in order to get access to B cells [29] . Taking advantage of this experimental setting we investigated whether the absence of latently infected HEL+ B cells in vivo was due to an intrinsic resistance of these cells to the virus . As the YFP-MuHV-4 used for in vivo experiments allows the detection of latently infected cells we used in this experiment an EF1α-eGFP+ MuHV-4 , in which GFP expression can be detected 48 h post infection . We used in parallel gp150+ and a gp150− viruses , the later leading to a better B cell infection in co-culture assay [29] . We co-cultured freshly isolated SWHEL splenocytes with infected RAW-264 or BHK-21 cells , or exposed the splenocytes to free viruses ( Figure 4 ) . 48 h post co-culture , cells were harvested and GFP expression was monitored in both HEL+ and HEL− B cells . Contrasting with our in vivo observations , we observed that both populations were equally sensitive to MuHV-4 infection when co-cultured with infected RAW-264 , while they remained not infected with exposed to free virions or co-cultured with infected BHK-21 ( Figure 4 ) . Fitting with previous observations , percentages of infection were greater with a gp150-deficient virus . Overall , these in vitro data show that the absence of latently infected HEL+ B cells in vivo is not due to an intrinsic resistance of these cells to the virus . MuHV-4 is known to induce proliferation of both T cells and B cells [30] . To evaluate the influence of MuHV-4 on early B cell activation , we performed a kinetic analysis monitoring the number and CD69 expression of HEL+ and HEL− B cells isolated from spleen and cervical lymph nodes ( CLN ) ( Figure 5 ) . We observed an increased number of both HEL+ and HEL− B cells in the spleen and CLN , which peaked at 14 dpi ( Figure 5A ) , suggesting that MuHV-4-driven B cell proliferation does not rely on the infection status . We monitored CD69 expression by two complementary methods: measuring the intensity of CD69 expression ( Figure 5B ) and by evaluating the percentage of CD69high cells ( Figure 5C ) . In the spleen , beside a small population of CD69high cells observed on HEL− B cells at 7 and 14 dpi ( Figure 5C ) , we did not detect a significant increase of CD69 expression on HEL+ and HEL− B cells . On the opposite , in the CLN we observed a peak of CD69 expression on both HEL+ and HEL− B cells at 7 dpi , which disappeared at 14 dpi . In both organs , YFP+ B cells were restricted to the HEL− B cells indicating that this transient activation of HEL+ B cells is not sufficient to allow viral latency ( Figure S2 ) . As MuHV-4 is known to establish latency in GC B cells , we next monitored the late phase of the B cell response by following the frequency of GC and plasma cells in HEL+ and HEL− B cells ( Figure 6A & 6B ) . Although our ELISPOT ( Figure 1 ) , proliferation ( Figure 5A ) and early activation ( Figure 5B & 5C ) data suggested a polyclonal B cell activation upon MuHV-4 infection , HEL+ B cells did not acquire a GC or plasma cells phenotype at 14 dpi ( Figure 6A ) . In contrast , HEL− B cells entered GC reactions and differentiated into plasma cells ( Figure 6B ) . The frequency of YFP+ cells in HEL− GC was ∼8% , in accordance to what we observed in C57BL/6 mice ( Figure 6C ) . Spatial organization of the GC is an essential component of the B cell response as it dictates the interaction between B cells and the other cellular players such as follicular helper T cells and dendritic cells [31] . To have an insight into the organization of the HEL+ and HEL− B cells in infected mice , we performed immunofluorescent staining on spleen sections from naïve and 14 dpi SWHEL mice ( Figure 7 ) . As natural YFP signal was lost during fixation , infected cells were revealed with an Alexa-488 anti-GFP antibody . In naïve mice ( Figure 7A ) , HEL+ B cells were homogeneously spread in the B cell area of the follicle , and no GFP+ or GC cells were observed . At 14 dpi , clusters of GL-7+ cells were present in the B cell area ( Figure 7B ) , in which latently infected cells were found but HEL+ B cells were excluded . The number of GFP+ cells varied greatly between GCs; a heterogeneity also seen in C57BL/6 mice [32] . However , no matter the number of GFP+ cells present , we systematically observed an exclusion of the HEL+ B cells from the GC ( Figure 7B ) . These histological data confirm our phenotypical analysis ( Figure 6 ) and overall these data show that while HEL+ B cells are sensitive to MuHV-4 infection in vitro ( Figure 4 ) and get activated in vivo ( Figure 5 ) , they do not support latent infection and do not participate to the GC reaction induced by MuHV-4 . To support our phenotypic and histologic observations , PBS or MuHV-4 challenged SWHEL mice were bled to measure plasmatic levels of anti-MuHV4 and anti-HEL IgG1 , IgG2a and IgG2b ( Figure 8 ) . We were not able to quantify the amount of circulating antibodies as no standards were available , but the magnitude of these responses was assessed by systematically analyzing the time points from identical mice together , limiting the impact of technical variations . MuHV-4 has been previously shown to trigger an anti-viral response dominated by the production IgG2a and IgG2b [8] and our kinetic analysis followed the same pattern ( Figure 8 , left graphic ) . This response appeared between 7 and 14 dpi and gradually increased . For the anti-HEL response , although we did not detect a HEL+ GC response , we observed a peak of anti-HEL IgG2a and IgG2b antibodies 14 dpi , which declined quickly thereafter ( Figure 8 , right graphic ) . Naïve SWHEL mice have a basal level anti-HEL IgGs [21] and the expansion and transient activation of HEL+ B cells observed in the CLN after MuHV-4 infection ( Figure 5 ) could account for this burst of anti-HEL IgGs . However , the fact that this anti-HEL response is transient indicates the absence of a long-term anti-HEL response , fitting with our previous observations . Our data show that HEL+ B cells are not latently infected and do not participate in the GC response induced by MuHV-4 while they are equally sensitive to infection in vitro . Moreover , our ELISPOT data show that viral latency is not restricted to virus-specific B cells , indicating that latency is established in B cells of other specificities . This set of observations leads us to propose that the establishment of latency is not a stochastic event and takes place in a restricted population of polyclonal B cells . This model implies that MuHV-4 does not overcome the stimulatory signals provided by the BCR stimulation and the cognate CD4 help , but manages to benefit from it in order to settle in long-lived memory B cells . To test if MuHV-4 could benefit from acute CD4-dependent B cell responses , we stimulated a physiological number of adoptively transferred HEL+ B cells in C57BL/6 mice with sheep red blood cells ( SRBC ) conjugated to recombinant HEL [23] . We used an adoptive transfer assay in order to avoid competition between HEL+ B cells in SWHEL mice , in particular from being in too great an excess over the available SRBC-specific CD4 help . Indeed , SWHEL immunized with SRBC-HEL showed a poor GC response ( ∼1% of GC HEL+ B cells , Figure S3A ) when compared to C57BL/6 adoptively transferred with HEL+ B cells ( ∼70% of GC HEL+ B cells , Figure S3B ) . We controlled that SRBC-HEL could not induce an endogenous HEL-specific B cell response in C57BL/6 mice by co-transferring SRBC-HEL with or without HEL+ B cells and showed that transferred HEL+ B cells were required for the emergence of HEL+ GC B cells ( Figure S3B ) . As schematized in Figure 9A , we transferred HEL+ B cells 24 h before infection and immunized the infected mice with SRBC+/−HEL at 0 , 4 , 7 or 10 dpi . We decided to test different time of immunization , as it is currently unknown when the virus/B cell encounter happens and whether it infects naïve or activated B cells in vivo . At 14 dpi , we monitored the GC differentiation and percentage of infection in HEL+ and HEL− B cells ( Figure 9B ) . Immunization with SRBC-HEL at either of the time point tested triggered the GC differentiation of HEL+ B cells when compared to mice immunized with SRBC alone ( Figure 9B , top left ) while it did not affect the GC phenotype of HEL− B cells ( Figure 9B , bottom left ) . The magnitude of the GC response observed in HEL+ B cells was different between the time of immunization , certainly due to a mixed influence of the GC dynamic and survival of the transferred cells . While we could detect for the first time a HEL+ GC response in the context of MuHV-4 infection , these cells remained YFP− ( Figure 9B , top right ) , YFP+ cells being restricted to the HEL− population ( Figure 9B , bottom right ) , in which frequency of infection was not affected by SRBC-HEL immunization . We verified that adoptively transferred B cells could get latently infected by transferring WT CD45 . 1+ splenocytes into WT CD45 . 2+ recipient mice and showed that frequency of infection and GC differentiation was equivalent between donor and recipient cells ( Figure S4 ) . These data support the fact that HEL+ GC B cells are resistant to MuHV-4 latency and that MuHV-4 does not benefit from acute B cell responses to establish latency in non-virus-specific B cells , likely relying on other mechanisms yet to be identified .
In this study , we attempted at better understanding how γ-herpesviruses establish latency in B cells with a particular focus on the role of the BCR specificity . By following HEL+ B cells in MuHV-4 infected SWHEL mice , we were able to monitor the behavior of non-virus specific B cells during the establishment of MuHV-4 latency and showed that those cells were excluded from the latently infected population . Previous studies have established that MuHV-4 latency depends on B cell activation and proliferation [33] , but it is still not clear whether MuHV-4 can drive such activation independently of BCR specificity . When we compared the proliferation and CD69 expression of HEL+ and HEL− B cells , we observed that both populations behave in a similar manner , with proliferation in both spleen and CLN and a transient CD69 upregulation in the CLN . This confirms previous work that showed CD69 upregulation on B cells exposed to MuHV-4 in vitro and a temporary B cell proliferation in MHC-II-deficient I-Ab−/− mice [7] . However , HEL+ B cells did not participate to the long-term humoral response , as they did not differentiate into GC or plasma cells . We think we observed here two distinct waves of activatory signals . The first wave triggering a non-specific activation of the global B cell population , followed by a second wave that promotes the differentiation into GC of a restricted pool of B cells . The respective role of these two waves in the establishment of latency is not completely clear , but the first wave of activation is not sufficient to allow the establishment of latency in B cells , as supported by the fact that HEL+ B cells do not get latently infected . While we do not identify the mechanism driving the first wave of activation , it has been shown that both T cells and B cells are responsible for the MuHV-4 driven splenomegaly [30] , suggesting that the first wave of activation is not due to factors specific of the B cell response , and is probably cytokine mediated . Concerning the second wave of activation , correlation analysis performed on our dataset showed that frequency of YFP+ cells in HEL− B cells correlates positively with the magnitude of the GC response ( Figure 10 ) . This is in accordance with recent observations by Collins et al who observed a positive correlation between the frequency of YFP+ cells and the frequency of follicular helper T cells , an essential player of the GC response [10] . That said , the fact that HEL+ B cells are sensitive to the virus in vitro but do not get latently infected and are excluded from the GC reaction go against a model where MuHV-4 could drive a stochastic manipulation of the B cells and would instead rely on BCR specificity . One previous study looked at the influence of MuHV-4 on non-virus specific B cells by reconstituting μMT B cell −/− mice with B cells from MD4 mice ( designated HELMET mice [34] ) . In MuHV-4 infected HELMET , HEL+ B cells expressed CD69 and proliferated but contrary to our results , MuHV-4 latency was detected in HEL+ B cells by PCR . The absence of competition in HELMET mice , where B cells are all HEL-specific could account for these discrepancies . Indeed , in SWHEL mice , HEL+ B cells coexist with a majority of polyclonal B cells . Our in vitro infection by co-culture assay supports the fact that HEL+ B cells are sensitive to MuHV-4 , suggesting that a selection mechanism might occur in vivo , leading to the disappearance of these infected cells . In HELMET mice , the absence of competition could allow for the survival of latently infected HEL+ B cells . This role for competition is supported by the work of Kim et al [35] who studied how latency evolved in mice containing CD40+ and CD40− B cells . Although the two populations got latently infected , latency was ultimately lost in CD40− B cells and GC differentiation was restricted to CD40+ B cells . To explore whether the presence of CD4 T cell dependent antigens could be involved in the selection of latently infected cells , we triggered an endogenous anti-HEL response concomitantly with MuHV-4 infection . Although we were able to induce the emergence of HEL+ GC B cells , these cells remained refractory to latent infection . These data support that , at least in our experimental setting , MuHV-4 cannot hitchhike an acute humoral response to gain access to GC B cells and might instead benefit from other antigen/BCR interactions . In humans , it is estimated that up to 20% of a normal B cell repertoire is made of self-reactive B cells that need to be constantly kept under control [36] , [37] . One of the tolerance mechanism is the induction of a functional unresponsive state known as anergy , which requires endogenous BCR signaling [38] , [39] . The high prevalence of self-reactive B cells offers a good opportunity for γ-herpesviruses to manipulate these processes in order to retrieve B cells from their anergic state and promote polyclonal B cells activation in an antigen-dependent manner . Supporting this point , recent studies performed in humans [40] and mice [41] , [42] have shown that γ-herpesviruses are found in self-reactive B cells . Two studies have explored MuHV-4 impact on anergic B cells [42] , [43] , but it is still not clear how MuHV-4 can modulate this processes . SWHEL×ML5 mice [21] , in which HEL+ B cells are anergic due to the presence of soluble HEL , could be an alternative model to study the influence of the virus on competent anergic B cells . Further work will be required to elucidate how MuHV-4 promotes GC B cells differentiation independently of their infection status , and how this event is linked to the establishment of latently infected B cells in memory B cells . γ-herpesviruses are known to be mildly pathogenic in immune-competent hosts and it has been shown that MuHV-4 latent infection does not induce autoimmune disorders and actually confers protection in lupus-prone animals [41] . This highlights the fact that the interrelationship between MuHV-4 and B cells is complex and that co-evolution between γ-herpesviruses and their host as allowed for the emergence of subtle mechanisms that promote B cell activation but limits associated immune disorders in order to establish life-long latency .
This study was carried out in strict accordance with the recommendations of the Portuguese official Veterinary Directorate , which complies with the Portuguese Law ( Portaria 1005/92 ) . The Portuguese Experiments on Animal Act strictly comply with the European Guideline 86/609/EEC and follow the FELASA ( Federation of European Laboratory Animal Science Associations ) guidelines and recommendations concerning laboratory animal welfare . All animal experiments were approved by the Portuguese official veterinary department for welfare licensing under the protocol number AEC_2010_017_PS_Rdt_General and the IMM Animal Ethics Committee . SWHEL mice [21] were obtained from Dr Antonio Freitas , Institut Pasteur , Paris , in accordance with Dr Robert Brink , Garvan Institute , Melbourne . To screen for expression of the VH10tar heavy chain and the Vκ10-κ light chain genotyping was performed on DNA isolated from mouse-tails using DirectPCR solution ( Viagen ) . Mice heterozygous for both genes were used for experiments . HEL+ B cells were identified by FACS and confocal microscopy ( see details below ) by direct labeling with recombinant HEL ( Sigma-Aldrich ) conjugated with Alexa 647 ( noted HEL-A647 ) . This conjugation was made with the Alexa Fluor Antibody Labeling Kit ( Invitrogen ) following manufacturer instructions . A Bio-Gel P-6 ( Biorad ) loaded column was used to separate HEL-A647 conjugates from free dye . C57BL/6 , CD45 . 1 and CD45 . 2 mice and were purchased from Charles Rivers Laboratories . Mice were between 7 and 15 weeks old at time of infection and were sacrificed by CO2 inhalation or cervical dislocation . The YFP expressing MuHV-4 [4] was obtained from Dr Samuel Speck , Emory Vaccine Center , Atlanta . EF1α-eGFP+ MuHV-4 and EF1α-eGFP+-gp150− MuHV-4 [29] were obtained from Philip Stevenson , University of Cambridge , Cambridge . Viral stocks were prepared by infecting BHK-21 cells and titrated by plaque assay using previously published procedures [44] , [45] . For infections , mice were anaesthetized with isoflurane and inoculated intranasally with 104 pfu of YFP-MuHV-4 under 20 µl of PBS . ELISPOT assay to enumerate MuHV-4 specific B cells was adapted from [8] . Briefly , purified MuHV-4 were disrupted for 10 min in PBS+0 . 05% Triton X-100 and plated at 5×106 PFU/well in 96-well MultiScreen HA mixed cellulose filter plates ( Millipore , Billerica , MA ) . Plates were incubated overnight at 4°C , washed with PBS and blocked for 1 h at 37°C with complete medium ( RPMI-1640+10% heat inactivated FBS , 2 mM glutamine , 100 U/ml penicillin and streptomycin and 1 mM sodium pyruvate ) . Latently infected ( CD19+ YFP+ ) and non-infected ( CD19+ YFP− ) B cells were sorted from spleens on a FACS Aria ( BD Biosciences ) . Serial four-fold dilutions of sorted cells were prepared in complete medium and added under 100 ul/well in four replicate wells per cell amount . Cells were incubated overnight at 37°C in a humid 5% CO2 incubator . Plates were washed with PBS and incubated for 2 h at room-temperature with Alkaline phosphatase ( AP ) -conjugated rabbit anti-mouse IgG ( H+L ) antibodies ( Southern Biotech ) diluted 1/500 in PBS+0 . 5% FBS . After thorough washes spots were revealed at room temperature with 1 mg/ml of 5-bromo-4-chloro-3-indolyl phosphate ( Sigma ) in diethanolamine buffer . Upon optimal spot development plates were washed and dried . Blue spots representing single antibody-secreting cells ( ASC ) were counted under an Olympus SZ51 microscope . Total number of ASC were determined as described above except that plates were coated with 0 . 5 µg/well of a goat anti-mouse κ antibodies ( Southern Biotech ) diluted in PBS . In vitro infection by co-culture assay was adapted from [29] . 24 h prior co-culture , 3 . 105 RAW-264 and BHK-21 cells were seeded in 24-well plate . After cell adhesion ( 4–6 h ) , media was removed and cell infected overnight with indicated MuHV-4 at 9 . 105 pfu per condition . The next day , spleens were harvested and single cell suspensions were prepared by spleen disruption , filtration on 100 µm cell strainer and red blood cells removal by centrifugation on ficoll gradient ( Biowest ) . Splenocytes were washed and added to cells at 106/well . As a negative control , splenocytes were exposed to free viruses . In order to have enough cells to work with , spleens from two SWHEL mice were pooled . For each experiments , two suspensions were analysed in parallel . After 48 h of co-culture , cells were harvested and stained with CD11b ( to exclude RAW-264 cells ) , CD19 and HEL . Infection of RAW-264 was systematically evaluated 24 h post infection by monitoring GFP expression in cells not co-cultured with splenocytes ( data not shown ) . Single cell suspensions were prepared from spleens . Red blood cells were lysed in hypotonic NH4Cl and stainings were performed at 4°C in PBS+4% FCS and 1 mM EDTA . Briefly , cells were blocked by 10 min incubation with FcBlock ( anti-CD16/32 , 2 . 4G2 , BD bioscience ) , washed , and stained for 20 min . For biotinilated antibodies , extra 20 min incubation with streptavidin was performed . MuHV-4 infected cells were monitored based on their endogenous YFP expression . The following antibodies were used: anti-CD69 PE ( H1 . 2F3 ) , anti-CD95 PE ( Jo2 ) , anti-CD19 APC-Cy7 or APC-H7 ( 1D3 ) , anti-IgM PE ( R6-60 . 2 ) , anti-IgD Biotin ( 11-26c . 2a ) , CD11b PE or v450 ( M1/70 ) ( BD Biosciences ) ; CD45 . 2 Brilliant Violet 510 ( 104 ) ( BioLegend ) ; anti-CD45 . 1 PeCy7 ( A20 ) , and anti-GL-7 Biotin ( Ebioscience ) . Streptavidin-Cy5 ( BD Biosciences ) was used to reveal biotinilated antibodies . HEL specific B cells were identified with the HEL-A647 conjugate described above . Samples were acquired on a FACS Canto or on a LSR Fortessa ( BD Biosciences ) , using DIVA software ( BD Biosciences ) for acquisition and Flowjo v . 6 . 4 . 7 ( Tree Star ) for analysis . Cells were gated on live cells based on FSC/SSC parameters and cell doublets were excluded based on FSC-W signal . Spleens were fixed overnight at 4°C in periodate-lysin-paraformaldehyde ( PLP ) [46] , [47] and dehydrated by successive 2 h incubation in 10% , 20% and 30% sucrose solutions at 4°C . Spleens were then embedded in OCT ( Tissue Tek ) , frozen and sectioned ( 40 µm ) . For immunofluorescence staining , sections were encircled with a Fatpen and rehydrated 10 min in phosphate buffer . All incubations were made in humid chamber , protected from light . Sections were permeabilized for 1 h at room temperature in 1% triton and blocked 1 h in 1% BSA+FcBlock ( anti-CD16/32 , 2 . 4G2 , BD bioscience ) . Stainings were performed using the following reagents: anti-GFP Alexa 488 ( Invitrogen ) , anti-GL-7 Biotin ( GL-7 , Ebioscience ) , HEL-A647 ( described above ) and Hoechst 33343 ( Invitrogen ) . Streptavidin Alexa568 ( Invitrogen ) was used for biotinilated antibodies , incubated 1 h at room temperature . Slides were washed , mounted in Fluoromount-G ( SouthernBiotech ) and kept at 4°C . Images were acquired on a LSM 510 META point scanning confocal microscopes ( Zeiss ) and analyzed using LSM Image Browser ( Zeiss ) and Photoshop CS2 ( Adobe ) . HEL+ B cells ( CD19+ , HEL+ ) and HEL− B cells ( CD19+ , HEL− ) were sorted using a FACS Aria ( BD Biosciences ) and used for in vitro reactivation assay to quantify latent infection . Serial dilutions of freshly isolated cells were co-cultured with BHK-21 in complete media supplemented with 50 µg/ml of Gentamycin ( Invitrogen ) . Lysing half of the sorted cells by a quick freeze/thaw cycle before coculture allowed us to assess the presence of preformed viral particles , indicative of lytic infection . After 5 days , BHK-21 were fixed with 4% paraformaldehyde and stained with toluidine blue for plaque counting . The number of plaques in each sample was expressed as plaques forming unit ( pfu ) /107 cells . The frequency of virus-genome-positive cells was determined from pools of 2 to 3 spleens by limiting dilution combined with real-time PCR as previously described [48] . Sorted HEL+ and HEL− B cells were serially two-fold diluted and eight replicates of each dilution were analysed by real time PCR ( Rotor Gene 6000 , Corbett Life Science ) . The primer/probe sets were specific for the MuHV-4 M9 gene ( 5′ primer: GCCACGGTGGCCCTCTA; 3′ primer: CAGGCCTCCCTCCCTTTG; probe: 6-FAM-CTTCTGTTGATCTTCC–MGB ) . Samples were subjected to a melting step of 95°C for 10 min followed by 40 cycles of 15 s at 95°C and 1 min at 60°C . Positive vs . negative reactions were scored using the Rotor Gene 6000 software . Our data were compatible with the single-hit Poisson model ( SHPM ) as tested by modeling the limiting dilution data according to a generalized linear log-log model fitting the SHPM and checking this model by an appropriate slope test as described [3] , [49] . A regression plot of input cell number against log fraction-negative samples was used to estimate the frequency of cells with viral genomes . Estimation of the cell subset frequency of MuHV-4 infection consisted of computation by maximal-likelihood estimation as follows: let f be the estimate of the cell frequency; the maximum likelihood of f is the value of f that maximizeswhere log ( L ) is the natural logarithm of the likelihood function L and Pi is given by Pi = exp ( -f xi ) according to the SHPM . The variance of f was calculated as the negative reciprocal of the second derivative of log ( L ) , var ( f ) = 1/[d2 log ( L ) /df2] . The 95% confidence interval ( CI ) for f was calculated as 95% CI ( f ) = f±1 . 96SE ( f ) . Abbreviations are as follows: k = the number of groups of replicate PCRs , numbered i = 1 , 2 , … k; ni = the number of replicate reactions; ri = the number of observed negative PCRs; and mi = the observed fraction of negatives ( mi = ri/ni ) . Freshly isolated bulk splenocytes from SWHEL mice containing 104 HEL+ B cells were transferred into C57BL/6 mice by intravenous injection , as previously described [50] . Sheep red blood cells ( SRBC ) were obtained from Miguel Fevereiro , Laboratório Nacional de Investigação Veterinária , Lisbon . Recombinant HEL was covalently conjugated to SRBC with 1-ethyl-3- ( 3-dimethylaminopropyl ) -carbodiimide hydrochloride ( Sigma-Aldrich ) as described [25] . Conjugation was confirmed by FACS by staining mock or HEL-conjugated SRBC with the HyHEL10 ( an anti-HEL IgG1 [21] ) followed by an anti-IgG1 APC ( BD Biosciences ) . HEL-A647 was used to identify transferred HEL+ B cells . To measure anti-HEL and anti-MuHV-4 antibody production , sera were regularly collected by facial-vein bleeding . To measure HEL-specific antibodies , maxisorp plates ( Nunc ) were coated with 60 µl of recombinant HEL ( 10 µg/ml ) diluted in NPP buffer ( adapted from [21] ) . For MuHV-4 antibodies , plates were coated with viral particles disrupted with 0 , 1% triton and diluted in NPP buffer ( adapted from [7] ) . Coated plates were incubated overnight at 4°C and blocked for 1 h with 100 µl PBS+1% BSA . Sera were diluted to 1/200 in PBS+0 , 1% BSA and 50 µl were incubated 2 h at room temperature . IgG1 , IgG2a and IgG2b subclasses were measured using 50 µl of anti-mouse IgG1 , IgG2a and IgG2b conjugated to alkalyne-phosphatase ( SouthernBiotech ) diluted to 1/500 and incubated 1 h at room temperature . Bound antibodies were revealed using 100 µl of 1 mg/ml P-Nitrophenyl Phosphate ( MP Biomedical ) prepared in NPP buffer and incubated 40 min at 37°C . Absorbance was measured at 405 nm . p values were calculated using non-parametric Mann-Whitney U test; ns indicates p>0 . 05 , * indicates p≤0 , 05 , ** indicates p≤0 , 005 , and *** indicates p≤0 , 001 . | Murid γ-herpesvirus-4 ( MuHV-4 ) is a good model to study infectious mononucleosis in mice , in which the virus ultimately establishes life-long latency in B cells . Whereas several viral proteins have been shown to modulate B cell behavior , in the present study we aimed at clarifying the parameters that dictate the establishment of viral latency from the B cell perspective . Indeed , the B cell repertoire is highly diverse and it remains unknown whether latency takes place randomly in B cells . To study this question , we isolated latently infected B cells in which we observed a low frequency of virus-specific B cells , suggesting that viral latency is not restricted to this population . To better understand MuHV-4 influence on non-virus specific B cells , we then followed the fate of B cells specific for a foreign antigen , hen egg lysozyme ( HEL ) . While in vitro experiments showed that HEL-specific B cells could be acutely infected by MuHV-4 , these cells were resistant to MuHV-4 latent infection in vivo . These results suggest that while establishment of γ-herpesvirus latency is not restricted to virus-specific B cells , it does not take place randomly in B cells and relies on mechanisms that remain to be identified . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"blood",
"cells",
"humoral",
"immunity",
"antibody-producing",
"cells",
"white",
"blood",
"cells",
"immune",
"cells",
"cell",
"biology",
"animal",
"cells",
"b",
"cells",
"immunity",
"immunity",
"to",
"infections",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"immunology",
"immune",
"response"
] | 2014 | Establishment of Murine Gammaherpesvirus Latency in B Cells Is Not a Stochastic Event |
Otitis media with effusion ( OME ) is the commonest cause of hearing loss in children , yet the underlying genetic pathways and mechanisms involved are incompletely understood . Ventilation of the middle ear with tympanostomy tubes is the commonest surgical procedure in children and the best treatment for chronic OME , but the mechanism by which they work remains uncertain . As hypoxia is a common feature of inflamed microenvironments , moderation of hypoxia may be a significant contributory mechanism . We have investigated the occurrence of hypoxia and hypoxia-inducible factor ( HIF ) mediated responses in Junbo and Jeff mouse mutant models , which develop spontaneous chronic otitis media . We found that Jeff and Junbo mice labeled in vivo with pimonidazole showed cellular hypoxia in inflammatory cells in the bulla lumen , and in Junbo the middle ear mucosa was also hypoxic . The bulla fluid inflammatory cell numbers were greater and the upregulation of inflammatory gene networks were more pronounced in Junbo than Jeff . Hif-1α gene expression was elevated in bulla fluid inflammatory cells , and there was upregulation of its target genes including Vegfa in Junbo and Jeff . We therefore investigated the effects in Junbo of small-molecule inhibitors of VEGFR signaling ( PTK787 , SU-11248 , and BAY 43-9006 ) and destabilizing HIF by inhibiting its chaperone HSP90 with 17-DMAG . We found that both classes of inhibitor significantly reduced hearing loss and the occurrence of bulla fluid and that VEGFR inhibitors moderated angiogenesis and lymphangiogenesis in the inflamed middle ear mucosa . The effectiveness of HSP90 and VEGFR signaling inhibitors in suppressing OM in the Junbo model implicates HIF–mediated VEGF as playing a pivotal role in OM pathogenesis . Our analysis of the Junbo and Jeff mutants highlights the role of hypoxia and HIF–mediated pathways , and we conclude that targeting molecules in HIF–VEGF signaling pathways has therapeutic potential in the treatment of chronic OM .
Chronic middle ear effusion without the symptoms of acute infection is termed otitis media ( OM ) with effusion and can be sequel to acute bacterial otitis media . Otitis media with effusion ( OME ) is the most common cause of hearing impairment in children potentially causing language delays , learning and behavioral problems [1] , [2] . About 2 . 2 million episodes of OME occur annually in the US with an annual cost estimate of $4 . 0 billion [3] . The prolonged ventilation of the middle ear with tympanostomy tubes , also known as grommets , remains the best treatment for OME [4] . Placement of tympanostomy tubes is the most common operation in the UK ( 30 , 000 procedures per annum ) however the mechanism by which they work remains uncertain . As hypoxia is a common feature of inflamed microenvironments [5] , [6] the therapeutic benefits of ventilating the middle ear may conceivably include the moderation of hypoxia as well as relieving negative pressure and fluid drainage . Responses to hypoxia are mediated by Hypoxia Inducible Factor ( HIF ) protein a transcription factor that induces genes whose products restore blood supply , nutrients and energy production to maintain tissue homeostasis . Constitutively expressed HIF-1α is modified by prolyl hydroxylase domain ( PHD ) enzymes under normoxic conditions and targeted for proteasomal degradation . Under hypoxic conditions PHD activity is limited and HIF-1α is stabilized and forms a heterodimer with HIF-1β before translocation to the nucleus where it binds to hypoxic response elements [7] . HIF signaling is also regulated by inflammation at the transcriptional level via HIF-1α interactions with the master regulator of inflammation NF-κB [8]–[10] and at the translational level by cytokines such as IL-1β and TNF-α [5] , [6] . HIF responses are adaptive and help overcome localized ischemia as well as regulating innate immune responses to microbial infections [11] but chronic hypoxic inflammation may result in dysregulated HIF signaling and lead to pathological outcomes . Examples include fibrosis via immune cell activation [6] and the progression of rheumatoid arthritis [12] via angiogenesis caused by HIF-induced vascular endothelial growth factor ( VEGF ) . Indeed , treatment using VEGF receptor ( VEGFR ) signaling inhibitors moderates experimentally-induced arthritis [13] . Although hypoxia might be expected in the inflammatory conditions of chronic OM the evidence is inconsistent . Some studies have found that OME fluids in the middle ear cavity ( bulla ) have oxygen tensions similar to venous blood , of ∼40 Torr [14] , [15] . Another study reported pO2 in mucoid and serous OME fluids were lower ∼29–32 Torr . However , these values were not significantly different than pO2 values in barotrauma bulla fluids [16] . Nevertheless , there are a few studies to suggest that the downstream HIF signaling protein VEGF plays a role in experimental and clinical OME . Injection of recombinant VEGF into the rat bulla causes fluid effusion , mucosal inflammation and an increase in vascular permeability [17] . Vegf , Vegfr1 ( also known as Flt1 ) and Vegfr2 ( also known as Kdr ) gene and protein expression are upregulated in the endotoxin-induced rat model of OME [18] , [19] and Vegf protein is elevated in mouse middle and inner ear tissue after challenge with Haemophilus influenzae [20] . Moreover VEGF mRNA and protein are detectable in bulla fluids of patients with OME [18] , [21] . However , these studies have not investigated the role of hypoxia and HIF signaling in the inflamed middle ear . There is a significant genetic component predisposing to recurrent or chronic OM in human populations [22]–[26] . However , while a number of association studies have been carried out , sample sizes are relatively small and confirmation will be required through larger scale analyses and replication . A number of underlying OM susceptibility genes have been discovered in the mouse which represents a powerful model for dissecting the underlying pathways . These genes apppear to fall into three categories; those which are involved in craniofacial development and thereby Eustachian tube morphology and function , TLR4/MyD88 pathway genes that regulate innate immune function , and TGF-β pathway genes that modulate pro-inflammatory responses [27] . The two OM mouse mutants Junbo and Jeff , generated by N′-ethyl-N′-nitrosourea mutagenesis , represent powerful models for human OM as unlike many other mouse mutants they are non-syndromic and do not show the wide-ranging pleiotropic effects often associated with middle ear inflammatory disease in other models [27] . Jeff encodes a mutation in the Fbxo11 protein [28] and Junbo encodes a mutation in the transcription factor Evi1 [29] . Heterozygote Junbo ( Jbo/+ ) and Jeff ( Jf/+ ) mice develop OM spontaneously in the absence of other organ pathology or overt immune deficiency [28] , [29] . There is an association between polymorphisms in FBXO11 , the human homologue of the Jeff mutant protein , in OME and recurrent OM [24] and severe OM [25] , but there is no such association with EVI1 polymorphisms . In this work we have analyzed the two OM mouse models Junbo and Jeff by hypoxia labeling , transcriptional profiling and , in Junbo , using small-molecule inhibitors . We have discovered that the response to chronic inflammatory hypoxia via Hif-1α signaling and VEGF pathways is critical for chronic OM . Our analysis of the two mutants provides insight into the molecular and genetic mechanisms of OM and identifies potential new therapeutic targets for OM .
We surmised that the inflamed microenvironment in chronic OM is hypoxic and proceeded to test this hypothesis by the analysis of the Junbo and Jeff mutants . To test whether the inflammatory cells that accumulate within the middle ear were hypoxic we injected mice in vivo with pimonidazole ( PIMO ) , a marker that labels cells and tissues with a pO2<10 Torr ( ∼1 . 5% O2 ) . FACS analysis revealed hypoxia in viable and apoptotic polymorphonuclear cell ( PMN ) populations in the purulent bulla fluids of Jbo/+ ( 7 . 1±1 . 7×106 cells per µl , n = 10 ) and serous effusions of Jf/+ mice ( 55±25×103 cells per µl , n = 5 ) ( Figure 1H and Table 1 ) . In addition , immunohistochemistry showed hypoxia in F4/80-positive foamy macrophages ( mΦ ) within the bulla , the epithelium and in the connective tissues of the thickened , inflamed middle ear mucosa of Jbo/+ mice ( Figure 1A , 1C , 1D , 1G ) but not in the normal thin mucosa of wild type ( +/+ ) mice ( Figure 1B ) . Hypoxia was evident at 4 wk , increased at 7–8 wk and remained chronically elevated for >30 wk ( Figure 1E ) . The only part of the tubotympanum that appeared hypoxic under normal physiological conditions was the Eustachian tube ( Figure 1F ) . In Jf/+ mice PIMO labeling was restricted to inflammatory cells in the bulla fluids and there was no detectable mucosal labeling ( Figure 1I , 1J ) . Evi1 and Fbxo11 were expressed in the inflammatory cells that accumulate within the bulla fluids of Jbo/+ and Jf/+ mice , but only Evi1 ( 23–37 fold ) was expressed at higher levels relative to a normoxic baseline control of Jbo/+ or Jf/+ venous blood white blood cells ( WBC ) . The Evi1 target genes Jun ( 28–50 fold ) and Fos ( 5–10 fold ) were also elevated in Jbo/+ and Jf/+ bulla fluid inflammatory cells relative to blood WBC ( Figure S1 ) . We found elevated expression of Hif-1α ( 6–12 fold ) and HIF responsive genes Vegfa ( 41–122 fold ) and Slc2a1 ( also known as Glut1 ) ( 8 fold ) in Jbo/+ and Jf/+ bulla fluid WBC relative to blood WBC ( Figure S1 ) and Vegf signaling arrays showed elevated expression in a wide spectrum of Vegf pathway genes ( Table S1 ) . In Jbo/+ and Jf/+ mice we obtained data for 84 and 77 genes respectively and there was a strong similarity in pattern of upregulation of genes belonging to functional groups such as Vegf/growth factors and their receptors , Akt and Pi-3-Kinases , phospholipases A2 , heat shock proteins , Hif-1α and Arnt ( Hif-1β ) . 44% of the genes were significantly elevated ( >2-fold , P<0 . 05 ) in both mutants; 18% genes were elevated in both mutants with levels in either Jf/+ or Jbo/+ achieving statistical significance; 4% of genes were elevated in both mutants but P>0 . 05 , and 8% of genes were up-regulated in Jbo/+ mice but beneath detection limits for Jf/+ . 11% of genes were unaltered in one or other mutant; 14% unaltered in both mutants and only 1 gene was significantly lower in both mutants ( Table S1 , Figure S2 ) . In 8 wk old mice , Vegfa protein was elevated ∼74-fold in Jf/+ bulla fluids compared with Jf/+ sera ( median values of 5 , 793 pg/ml versus 78 pg/ml; P<0 . 001 ) and ∼335-fold in Jbo/+ bulla fluids compared with Jbo/+ sera ( median values of 28 , 123 pg/ml versus 84 pg/ml; P<0 . 001 ) ( Figure 2 ) . The difference between Vegfa titers in Jf/+ and Jbo/+ bulla fluids did not achieve statistical significance , nor did titer differences between Jeff and Junbo mutant and wild type ( +/+ ) sera ( Kruskall Wallis ANOVA and Dunn's multiple comparison post hoc tests ) . Using inflammation arrays we obtained data for 84 genes in Jbo/+ and 79 genes in Jf/+ mice . Again there was a strong similarity in the pattern of upregulation of gene expression for chemokines , cytokines , their receptors and acute phase response mediators . 35% of genes were significantly elevated ( >2-fold , P<0 . 05 ) in both mutants; 31% genes were elevated in both mutants with either Jf/+ or Jbo/+ achieving statistical significance; 11% of genes were elevated in both mutants but did not achieve statistical significance ( P>0 . 05 ) ; 6% of genes were up-regulated in Jbo/+ mice but beneath detection limits for Jf/+ . 10% of genes were unaltered in one or other mutant , 5% unaltered in both and only 2 genes were significantly lower ( >2-fold , P<0 . 05 ) in one or both mutants ( Table S2 , Figure S3 ) . Il-1β and Tnf-α are known modulators of Hif-1α translation and array data indicated that they were significantly elevated ( P<0 . 05 ) in Jbo/+ ( Il-1β 26-fold; Tnf-α 78-fold ) but elevations in Jf/+ expression ( Il-1β 3-fold; Tnf-α 50-fold ) were not statistically significant ( Table S2 ) . We therefore went on to determine their protein titers . Il-1β and Tnf-α were elevated in Jbo/+ bulla fluid but not consistently so in Jf/+ mice ( Figure 3 ) . Two of 22 Jf/+ mice had Tnf-α bulla fluid titers of 571 and 7 , 352 pg/ml respectively whereas 20/29 Jbo/+ mice had a median bulla fluid titer of 4 , 598 pg/ml ( range 2 , 156 to 15 , 293 pg/ml ) . The Tnf-α serum titers for mutant and +/+ mice were comparable and ranged from 24 to 107 pg/ml . One of 22 Jf/+ mice had an Il-1β bulla fluid titer of 1920 pg/ml whereas 28/29 Jbo/+ mice had a median bulla fluid titer of 2 , 862 pg/ml ( range 1 , 319 to 5 , 819 pg/ml ) . The Il-1β serum titers for mutant and +/+ mice were comparable and ranged from 15 to 27 pg/ml Figure 3 ) . To investigate whether Vegf has a pro-inflammatory role in OM we employed a variety of small-molecule inhibitors of VEGFR and assessed their effects on OM when delivered systemically to the Junbo mouse mutant . The rationale for using the Junbo model and not Jeff was that the OM phenotype was more penetrant . The percentage of Jbo/+ mice with bilateral OM was higher at 78% versus 46% in Jf/+ ( Figure S4 ) making auditory brainstem response ( ABR ) measurements more robust ( see below ) . Moreover , hearing loss over the standard test period from day 28 to day 56 was greater in Jbo/+ ( averaging 7–14 dB in independent experiments ) than in Jf/+ ( ∼4 dB ) ( Figure 4 and Figure S5 ) . When Jbo/+ mice were treated with VEGFR signaling inhibitors BAY 43-9006 ( 30 mg/kg ) , SU-11248 ( 20 mg/kg ) and PTK787/ZK 222584 ( 50 mg/kg or 75 mg/kg ) ( hereafter referred to as PTK787 ) there was a significant moderation of hearing loss ( Figure 4 ) . The trial with BAY 43-9006 was terminated after 2 wk when mice suddenly became piloerect . Although BAY 43-9006 was not as well tolerated as PTK787 and SU-11248 , the positive therapeutic response to three separate VEGFR signaling inhibitors confirms our data , indicating that HIF mediated VEGF is a critical pathway in OM pathogenesis . We also proceeded to target HIF signaling directly using a HSP90 inhibitor , 17-DMAG . HSP90 is a chaperone of HIF-1α . We found that its use also moderated hearing loss ( Figure 4 ) . We went on to examine the middle ear mucosal changes in mice treated with VEGFR inhibitors . Morphometric analysis of the mucosal histology was performed on 50 mg/kg and 75 mg/kg PTK787 treatment groups ( Figure 5 ) . ANOVA analyses revealed significant reductions in blood vessel number at the higher 75 mg/kg dose; lymphatic vessel number was reduced at both dosages; but neither the mucosal thickness nor lymphatic vessel diameter was reduced by PTK787 treatment ( Figure 5 ) . In the BAY 43-9006 trial , treated Jbo/+ mice had reduced lymphatic vessel number ( 10 . 8±1 . 1 n = 11 mice versus 16 . 4±1 . 0 n = 11 , P = 0 . 0012 ) and lymphatic vessel dilation ( 9 . 9±1 . 0 µm n = 11 versus 14 . 2±0 . 9 µm n = 11 , P = 0 . 0072 ) compared with sham treated controls but mucosal thickness and blood vessel number were not altered . To qualitatively assess the effect of drug treatment on bulla fluid accumulation , the middle ears were sampled in the 75 mg/kg PTK787 , SU-11248 and 17-DMAG treated and sham treated Jbo/+ mice . In each trial , a significantly lower ( P<0 . 05 ) proportion of treated Jbo/+ mice yielded bulla fluid samples than sham treated Jbo/+ controls ( Figure S6 ) .
Single gene mutations in mouse Eya4 , Tlr4 , p73 , MyD88 , Fas , E2f4 , Plg , Fbxo11 and Evi1 give rise to chronic spontaneous OM phenotypes , in several cases as part of a spectrum of pleiotropic effects , and are candidate susceptibility genes for human OM . In human populations there are significant associations between OM and polymorphisms in FBXO11 , TLR4 and PAI1 . However , the mechanisms and pathways by which these mutations result in chronic middle ear inflammatory disease are poorly understood . It has been proposed that they may act by a variety of different mechanisms including altered Eustachian tube function and reduced clearance of middle ear pathogens , dysregulation of innate immunity via TLR4/MyD88 pathways and dysregulation of anti-inflammatory mechanisms via TGF-β pathways [27] . We have analyzed the Junbo and Jeff mutants using a number of approaches , including transcriptional profiling and , in Junbo mice , small-molecule inhibitors to dissect the genetic pathways and pathophysiological processes leading to chronic OM . The characteristic lesion of OM is the accumulation of fluid and inflammatory cells in the bulla and mucosal inflammation . At other sites of inflammation , hypoxia is likely to occur as a result of the uptake of oxygen by inflammatory cells coupled with their physical separation from an underlying vascular bed [30] . Using PIMO labeling we have identified cellular hypoxia in inflammatory cells in the purulent Jbo/+ and serous Jf/+ fluids that accumulate within the 5–6 µl bulla [31] . However mucosal hypoxia was only detectable in Jbo/+ mice . The driver of mucosal hypoxia may be the unmet oxygen demand of inflammatory cells in bulla fluids which in turn is presumably a function of their numbers and viability . The cellularity of Jbo/+ bulla fluids is certainly >100-fold higher than in Jf/+ mice but there are substantial apoptotic PMN cell populations ( ranging from 20–40% ) and a necrotic cell population ( 7–8% ) which may affect overall oxygen requirements ( Table 1 ) . Apoptosis/cell death pathways and oxidative stress pathways would be expected to be upregulated as part of the inflammatory process . Human chronic OME effusions ( with or without bacterial infection ) range from purulent to serous and mucoid and contain viable and degenerative inflammatory cells [32] , [33] and VEGF protein [21] which is a critical downstream mediator of hypoxia signaling . Our results provide direct evidence of cellular hypoxia in bulla fluid inflammatory cells whereas the data for pO2 in human OME bulla fluids is inconsistent [14]–[16] . Mucosal gas exchange is the main method of ventilation of the normal tubotympanum and the resting oxygen tension of the middle ear is comparable to that of venous blood [34] . Surgical ventilation causes relative hyperoxia of the middle ear [35] and a change in oxygen tension might also be an important mechanism in the down-regulation of HIF signaling . One therapeutic benefit may be reduced mucin secretion as conserved promoter regions of respiratory mucin genes expressed in human middle ear bind to HIF-1α [36] , [37] . While the influx of inflammatory cells into the bulla lumen may be a key event in the development of hypoxia and activation of HIF signaling via stabilization of HIF-1α protein the activation of inflammatory cells and upregulation of Il-1β , Tnf-α and Nfκb in particular may further modulate HIF signaling [5] , [6] . Transcriptional profiling showed upregulation of inflammatory gene networks in the bulla fluids of Jbo/+ and Jf/+ mice relative to blood WBC . Il-1β and Tnf-α serum titers are comparable in mutant and +/+ mice suggesting that OM is not a cause of systemic inflammation , nor is it part of an ongoing systemic inflammatory condition in Junbo and Jeff models . A number of inflammatory genes associated with OM have been published; for a review , see [38] and [39]–[63] and our array data adds another 20 genes to this list ( Table S2 ) . However , middle ear inflammation appeared less pronounced in Jf/+ mice . In line with Jf/+ serous bulla fluids containing fewer inflammatory cells , protein titers for the key cytokines Il-1β and Tnf-α were only elevated in a minority of mice . This degree of biological variation between individual Jf/+ mice may explain the variability between pooled samples in which elevated expression levels for genes such as Il-1β and Tnf-α failed to achieve statistical significance . The relative contributions of Il-1β and Tnf-α to hypoxic modulation of Hif-1α and Vegf signaling in the middle ear may be greater in Jbo/+ mice . Nevertheless in both mutants upregulation of HIF signaling was evident from the elevated expression of multiple Vegf signaling pathway genes ( Figure S1 and Table S1 ) including Vegfa and in Jbo/+ its principle receptor Kdr ( Vegfr2 ) . Elevated Vegfa gene expression was accompanied by elevated Vegfa protein in Jbo/+ and Jf/+ mice ( Figure 2 ) . VEGF acts to induce angiogenesis , increases vascular permeability and recruitment of neutrophils and macrophages [64] , [65] and may therefore contribute to OM by the accumulation of fluid and inflammatory cells within the bulla causing conductive hearing loss and secondary cochlear dysfunction via diffusion of cytokines through the round window [66] , [67] . We tested the hypothesis that VEGF signaling contributes to OM pathogenesis by treating Junbo mice , which have highly penetrant OM , with the VEGFR signaling inhibitors PTK787 , SU-11248 , and BAY 43-9006 and the HSP90 inhibitor 17-DMAG . Their use reduced hearing loss ( Figure 4 ) . Histological analysis of the middle ear mucosa in PTK787 treated Jbo/+ mice revealed reduced blood vessel formation ( at the higher 75 mg/kg dose ) and lymphatic vessel formation ( at 50 mg/kg and 75 mg/kg dosages ) consistent with the anti-angiogenic effects of VEGFR signaling inhibitors ( Figure 5 ) . Only lymphatic vessel number and diameter were significantly moderated by BAY 43-9006 but this may be a reflection of the initial acute inflammatory change taking place in the first 2 wk which was the end point of this trial . SU-11248 treated mice were not examined by histology . Another effect of treatment with PTK787 , SU-11248 and 17-DMAG was to reduce the proportion of Jbo/+ mice that yielded bulla fluid samples ( Figure S6 ) . This may reflect moderation of VEGF induced vascular permeability in treated mice . The implication is that bulla fluids recoverable from treated Jbo/+ mice come from those which are less responsive to treatment , and this would confound comparisons of inflammatory cell numbers and gene expression between treated and control mice . The range of molecular targets for VEGFR and HSP90 inhibitors will require further clarification . VEGF receptors are members of the Receptor Tyrosine Kinase ( RTK ) superfamily and small-molecule VEGFR inhibitors have multi-kinase inhibitor profiles against different VEGF receptors as well as other RTK families . PTK787 is an inhibitor of VEGFR1 , VEGFR2 , VEGFR3 , PDGFR-β and c-Kit; SU-11248 acts as a VEGFR2 , PDGFR-β , FLT3 and c-Kit inhibitor; and BAY 43-9006 acts as a VEGFR2 , FLT3 , PDGFR-β , c-Kit and Raf1 inhibitor [68] . VEGFR inhibitors therefore have the potential to disrupt additional pathways [69] that might contribute to OM pathogenesis . We therefore also targeted HIF-VEGF signaling pathways using 17-DMAG treatment to inhibit HSP90 . HSP90 chaperones a number of proteins involved in HIF-VEGF signaling including HIF-1α itself , the mitogenic signaling protein AKT , and RAF-1 in the RAS/RAF/MEK/ERK MAPK pathway [70]–[72] . In addition , phosphorylation of HSP90 by its client protein VEGFR2 is required for receptor signaling to endothelial NO synthase [73] . However , 17-DMAG can also attenuate inflammatory pathways [74] and may also contribute to the amelioration of OM observed in Junbo mice . The expression of mutant Evi1 and Fbxo11 proteins in inflammatory cells in bulla fluids has the potential to perturb a variety of signaling pathways that may affect the response to hypoxia and contribute to OM pathogenesis . The Fbxo11 gene is a member of the large F-box family which are specificity factors for the SCF E3 ubiquitin ligase complex , and in homozygote Jeff mutants there are developmental defects in palate , eyelid and lung airway as a result of perturbed Tgf-β signaling [75] . EVI1 is a co-transcriptional repressor of SMAD3 and the mutation in Evi1 in Junbo mice may also exert effects via TGF-β signaling . EVI1 has two zinc-finger domains and a central transcription repression domain . Repressor activities via the proximal N-terminal zinc-finger domain include c-Jun N-terminal kinases ( JNK ) and TGF-β signaling via direct binding of SMAD3 . SMAD3 activity is also reduced by recruitment of the co-repressor CtBP by the central repressor domain [76] . There is considerable cross-talk between TGF-β and HIF-1α pathways . For example , SMAD3 and HIF-1α are co-activators of VEGF expression [77] , [78] and mutations affecting TGF-β signaling might be expected to perturb hypoxia responses . The distal zinc-finger domain of EVI1 has three zinc-finger motifs [79] and the Junbo mutation is a non-conservative Asn763Ile change located within three amino acids of a contact residue in the second zinc-finger motif . Interactions with the distal zinc finger domain raise AP-1 activity by increased expression of Jun and Fos [80] . AP-1 and Jun also interact with HIF pathways [81] , [82] and play a role in the pathogenesis of inflammatory bone and skin disease [83] . We found Evi1 , and its target genes Jun and Fos were relatively upregulated in the bulla of both Jbo/+ and Jf/+ mice . However we cannot usefully speculate on the possibility of differential expression of Jun and Fos by mutant Evi1Jbo/+ and wild type Evi1+/+ protein . Interpretation is problematic because bulla gene expression levels were normalized to their respective blood baselines , and we have no Evi1 protein data . Our studies on the mutants Junbo and Jeff , highlights chronic inflammatory hypoxia as a key mechanism of OM pathogenesis and underlines the role of Hif-1α signaling in the underlying genetic and pathophysiological mechanisms that predispose to chronic OM . Jeff has a less pronounced inflammatory OM phenotype , nevertheless the underlying hypoxic signaling mechanism acting via VEGF appears similar to the Junbo model . As a consequence we have identified potential new therapeutic targets for OM . The practical clinical implications for using small-molecule VEGFR signaling inhibitors or other anti-VEGF agents and HSP90 inhibitors are limited in pediatric applications as they are used principally for the treatment of cancer [68] , [69] , [84] . Ototopical delivery appears to be the most likely way forwards to achieve therapeutic levels of small-molecule inhibitors in the bulla fluids whilst reducing any adverse effects caused by systemic administration . In summary , our findings on the genetic bases for OM in the Junbo and Jeff mutants have underlined the importance of hypoxia mechanisms in the development of chronic OM and as a consequence have revealed potential new therapeutic strategies that merit further exploration .
The humane care and use of mice in this study was under the appropriate UK Home Office license . Junbo mice were congenic on a C3H/HeH background [29] and Jeff mice were on a mixed C3H/HeH and C57BL/6J genetic background [28] . The mice were specific pathogen free and had normal commensal nasopharyngeal flora [29] . Blood was collected from the retro-orbital sinus of mice under terminal anesthesia induced by an i . p . overdose of sodium pentobarbital . After removal of any adherent material on the external surface of the tympanic membrane , a hole in the membrane was made by removing the malleus with a clean pair of forceps and collecting the bulla fluid with a pipette . Bulla fluid volume was measured by collecting 0 . 5 µl aliquots and the total pooled samples from both ears generally ranged between 0 . 5–2 . 0 µl . Bulla fluid was collected into 100 µl aliquots of the appropriate buffer for each analysis ( see below ) or into 20 µl of RNase free water for RNA isolation . Whole blood for RNA isolation was collected in RNAlater ( Qiagen ) . Samples of bulla fluids from 8 wk old Jbo/+ and Jf/+ mice were analyzed for total WBC counts on an Advia 120 hematology analyzer ( Bayer ) . Cytology preparations of bulla fluids were made on electrostatically charged slides ( Superfrost Plus , Menzel Glaser ) , methanol fixed then stained with rat anti-mouse F4/80 Mab ( MCA497 ) ( AbD Serotech ) and counterstained with haematoxylin . Jbo/+ , Jf/+ and their respective wild type ( +/+ ) controls were labeled 3 h in vivo by i . p . injection with 60 mg/kg pimonidazole ( PIMO ) ( Hypoxyprobe , HPI Inc ) dissolved in 100 µl of sterile PBS . For FACS , bulla fluid samples were collected into 100 µl aliquots of ice cold FACS buffer then stained with anti-PIMO FITC , anti-mouse Ly6G and Ly6C PerCP-Cy5 . 5 ( BD Pharminogen ) and anti-Annexin V Biotin ( BD Pharminogen ) /Streptavidin Pacific Blue ( Invitrogen ) . Propidium iodide ( BD Pharminogen ) was used to assess necrotic cells . 50 µl EDTA blood samples were diluted in 100 µl FACS buffer then treated with RBC lysis buffer ( BD Pharminogen ) . Unlabeled bulla fluid PMN and non-staining peripheral ( normoxic ) PMN from PIMO-labeled mice served as negative controls . For histology , the head with the tympanic membranes left intact was fixed for 48 h in 10% neutral buffered formalin then decalcified with Formical ( Decal Corp ) for 72 h . Wax embedded 3 µm dorsal plane sections of the middle ear were immunostained for PIMO or stained with haematoxylin and eosin . Total RNA from 4 independent pooled samples of Jbo/+ and Jf/+ bulla fluids was isolated using Nucleospin RNA/protein isolation kits ( Macherey-Nagel ) . Individual blood samples from Jbo/+ and Jf/+ mice were extracted using Mouse RiboPure kits ( Ambion ) then the RNA was made into 3 separate sample pools . Each sample pool comprised 10–15 Jbo/+ mice or 5–9 Jf/+ mice . RNA quantity was measured on a Nanodrop 8000 ( Thermo Fisher Scientific ) and the integrity assessed by gel electrophoresis . 1 µg of RNA from each pool was used to synthesize double stranded cDNA with a High Capacity cDNA archive kit ( AB ) . RT-qPCR was performed using TaqMan gene expression assays using Fast Universal PCR Master Mix on a 7500 Fast Real-Time PCR System ( AB ) . Three technical replicates were performed for each TaqMan assay . Data was normalized using Ppia as the endogenous control and fold changes of expression ( ddCts ) of bulla fluid WBC over blood WBC were calculated using AB 7500 software v2 . 0 . 1 . This software allowed us to average the technical replicates for each pool and then average the biological replicates for the n = 4 bulla fluid sample pools and n = 3 blood sample pools . The fold change data is shown by mean relative quantification ( RQ ) ± min/max error bars representing 95% Confidence Limits ( CL ) . Using the biological replicate pools of bulla fluids and bloods described above , Vegf signaling ( PAMM-091c ) and Inflammation Response and Autoimmunity ( PAMM-077c ) arrays ( RT2-qPCT™ , SA Biosciences ) were performed . For each plate , 0 . 5 µg of RNA was converted to double stranded cDNA using the RT2 first strand synthesis kit . After mixing with the SABiosciences RT2 qPCR mastermix , the cDNA was pipetted into the 96 well profile plate and run on a 7500 Fast Real-Time PCR System ( AB ) . Data was normalized using β-actin as an endogenous control and fold changes of expression of bulla fluid WBC over blood WBC were calculated using SA Biosciences online software ( http://pcrdataanalysis . sabiosciences . com/pcr/arrayanalysis . php ) . The significance of the fold change is shown as a P value based on a Student's t-test of the replicate 2∧ ( −dCt ) values for each gene in the n = 3 control blood and n = 4 bulla fluid sample pools . Blood was collected into serum-gel clotting activator tubes ( Sarstedt ) . Measured volumes of bulla fluid were added to 100 µl of ice cold PBS , then vortexed and centrifuged at 500×g for 5 min at 4°C . Bulla fluid supernatants and serum samples were stored at −80°C until assay using Quantikine mouse Vegfa , Il-1β and Tnf-α ELISA kits ( R&D Systems ) . Some serum and bulla fluid samples had cytokine titers beneath the lowest assay standard ( 23 pg/ml for Tnf-α and 12 pg/ml for Il-1β ) and according to the manufacturer's instructions these results are not reportable . 27–29 d old +/+ and Jbo/+ mice were dosed by oral gavage once a day with 30 mg/kg BAY 43-9006 , 20 mg/kg SU-11248 , 50 or 75 mg/kg PTK787 , or 10 mg/kg 17-dimethylaminoethylamino-17-demethoxy-geldanamycin ( 17-DMAG ) . +/+ mice were treated with drug as a control for unforeseen ototoxicity . DMSO stock solutions of BAY 43-9006 and SU-11248 ( LC Laboratories ) or aqueous solutions of PTK787 and 17-DMAG were frozen at −20°C then diluted 10-fold in 2% methyl cellulose for administration . Drug and sham Jbo/+ groups were matched for age , gender and pre-trial ABR threshold ( range 30–60 dB ) and the sham group received vehicle alone . The anesthetized mouse was placed in right lateral recumbency with the speaker positioned 1 . 5 cm from the right ear , and a click-evoked ABR performed [85] . ABR measurements were made one day before the first treatment and one day after the last treatment . For ABR with recovery , anesthesia was induced by i . p . injection with a mixture of 10 mg/kg xylazine and 100 mg/kg ketamine and was reversed by 5 mg/kg atipamezole hydrochloride . To assess whether treatment with SU-11248 , 75 mg/kg PTK787 , or 17-DMAG treatment altered bulla fluid accumulation a note was made whether fluids were recoverable . Middle ear histology was assessed in the BAY 43-9006 , and the 50 or 75 mg/kg PTK787 trial mice . Morphometric evaluation was by blinded assessment of a standard 1000 µm length of middle ear mucosa ( avoiding the cochlea and the region close to the Eustachian tube ) , the mucosal thickness was averaged from 5 measurements , and the numbers of capillaries and lymphatic vessels ( and their diameter ) were recorded . D'Agostino & Pearson omnibus normality tests were performed on PTK787 histology data and Vegfa titer data . Blood vessel number , lymphatic vessel diameter and mucosal thickness were normally distributed and this data was subsequently analyzed using one-way ANOVAs and Bonferroni's multiple comparison tests for post hoc testing . Lymphatic vessel number and Vegfa titers were not normally distributed and a Kruskall-Wallis test was performed followed by Dunn's multiple comparison tests for post hoc testing . Arcsine transformed proportion data from FACS was analyzed using Student t-tests . Chi-squared tests were used to analyze tympanic membrane appearance ( cloudy or clear ) and the presence or absence of bulla fluids . All other data including ABR measurements ( where interval data was in 5 dB increments ) was analyzed using Mann Whitney U tests . In the drug trials , 1-tailed tests were used to test positive response to therapy , otherwise 2-tailed tests were used and values P<0 . 05 were considered significant . Data are presented as mean ± SEM ( n ) or in the case of Vegfa , Tnf-α and Il-1β protein titers with box and whisker plots . | Otitis media with effusion ( OME ) is the commonest cause of hearing loss in children , and treatment using grommets remains the commonest surgical procedure in children . Chronic forms of OM are known from human population studies to have a significant genetic component , but little is known of the underlying genes or pathways involved . We have analyzed two chronic OM mouse models , the Junbo and Jeff mutants , and have found that both demonstrate hypoxia and hypoxia-inducible factor ( HIF ) mediated responses . There is upregulation of inflammatory pathways in the mutant middle ears and in Junbo elevation of cytokines that modulate Hif-1α . Hif-1α levels are raised in the middle ear as well as downstream targets of HIF such as Vegfa . We explored the effects of small-molecule inhibitors of HSP90 and VEGF receptor signaling in the Junbo mutant and found significant reductions in hearing loss , the occurrence of bulla fluid , and moderation of vascular changes in the inflamed middle ear mucosa with the VEGF receptor inhibitors . The study of the Junbo and Jeff mutants demonstrates the role of hypoxia and HIF mediated pathways in OM pathogenesis , and it indicates that targeting the HIF–VEGF pathway may represent a novel approach to therapeutic intervention in chronic OM . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"of",
"disease",
"medicine",
"model",
"organisms",
"otorhinolaryngology",
"drugs",
"and",
"devices",
"gene",
"expression",
"genetics",
"biology",
"otolaryngology",
"gene",
"networks",
"genetics",
"and",
"genomics"
] | 2011 | HIF–VEGF Pathways Are Critical for Chronic Otitis Media in Junbo and Jeff Mouse Mutants |
European and African descendants settled the continental US during the 17th-19th centuries , coming into contact with established Native American populations . The resulting admixture among these groups yielded a significant reservoir of Native American ancestry in the modern US population . We analyzed the patterns of Native American admixture seen for the three largest genetic ancestry groups in the US population: African descendants , Western European descendants , and Spanish descendants . The three groups show distinct Native American ancestry profiles , which are indicative of their historical patterns of migration and settlement across the country . Native American ancestry in the modern African descendant population does not coincide with local geography , instead forming a single group with origins in the southeastern US , consistent with the Great Migration of the early 20th century . Western European descendants show Native American ancestry that tracks their geographic origins across the US , indicative of ongoing contact during westward expansion , and Native American ancestry can resolve Spanish descendant individuals into distinct local groups formed by more recent migration from Mexico and Puerto Rico . We found an anomalous pattern of Native American ancestry from the US southwest , which most likely corresponds to the Nuevomexicano descendants of early Spanish settlers to the region . We addressed a number of controversies surrounding this population , including the extent of Sephardic Jewish ancestry . Nuevomexicanos are less admixed than nearby Mexican-American individuals , with more European and less Native American and African ancestry , and while they do show demonstrable Sephardic Jewish ancestry , the fraction is no greater than seen for other New World Spanish descendant populations .
Native Americans inhabited the area that now makes up the continental US for thousands of years prior to the arrival of the first European settlers . The ancestors of modern Native Americans are thought have arrived in the Americas from Asia , by way of the Bering Strait , in several successive waves of migration [1] . The current model , based on archaeology and comparative genomic studies , holds that the earliest ancestors of Native Americans arrived in the Americas ~23 , 000 years ago [2] . The earliest evidence for Native Americans in the continental US dates to ~14 , 000 years ago [3] . The much later arrival of Europeans in the Americas , followed shortly thereafter by Africans who were brought by force via the trans-Atlantic slave trade , had a drastic effect on the demographic makeup of the region . Native American population numbers declined rapidly in the face of continuous immigration , settlement , and conflict , and as a result the modern US population is made up mainly of descendants of European and African immigrants . Europeans arrived in the Americas more than 20 , 000 years after the first Native Americans . The first European settlers to reach the continental US were Spaniards led by the conquistador Ponce de León , who claimed Florida for the Spanish crown in 1513 [4] . British settlers arrived more than 70 years later , initially establishing the ill-fated colony of Roanoke in 1585 and later the permanent settlement of Jamestown in 1607 [5] . An estimated 400 , 000 British had migrated to the US by the end of the 17th century . The first Africans were brought to Jamestown in 1619 by Dutch pirates who traded them to the British settlers as indentured servants [6] . The social status of Africans in the US changed quickly , with slavery first legally sanctioned by 1640 . The trans-Atlantic slave trade would eventually bring ~400 , 000 enslaved Africans to the continental US [7] . The arrival of Europeans and Africans in the Americas , and the conflict that followed , would prove to be catastrophic for the indigenous population . It has been estimated that 10–100 million Native Americans may have died in the first 150 years after Columbus’ arrival in the New World , amounting to a 95% reduction in the population [8] . This massive Native American population decline is mainly attributed to the introduction of European and African endemic infectious diseases–e . g . malaria , measles , and smallpox–for which the indigenous population had little or no immune defense [8 , 9] . The story of conflict between Native Americans and European settlers and enslaved Africans , along with the devastating consequences for the indigenous population , is by now well-known . However , there is another , perhaps less appreciated , aspect of the encounter between these population groups that has also had profound consequences for the genetic demography of the Americas . Here , we are referring to the process of genetic admixture , whereby individuals from previously isolated population groups reproduce , resulting in the novel combination of ancestry-specific haplotypes within individual genomes . Admixture has been a fundamental feature of human evolution and migration [10] . Whenever previously isolated human populations meet , no matter what the circumstances , they mix and give rise to individuals with a mosaic of different genetic ancestries . As European and African descendants settled the continental US , they inevitably came into contact with established Native American populations resulting in admixture and the introduction of Native American genomic sequence into the expanding US population . Accordingly , the genomes of European and African descendants in the US are expected to contain some fraction of Native American ancestry . In other words , a significant reservoir of Native American ancestry currently exists outside of recognized indigenous communities . In this study , we ask how the historical processes of migration and settlement affected the distribution of Native American admixture across the continental US ( S1 Fig ) . We address this question for the three largest genetic ancestry groups in the modern US population: African descendants ( AD ) , Western European descendants ( WD ) , and Spanish descendants ( SD ) .
The first aim of our study was to characterize the major genetic ancestry groups for the continental US based on observable patterns of ancestry and admixture seen for the 15 , 620 individuals from the Health and Retirement Study ( HRS ) analyzed here . The Health and Retirement Study data is sponsored by the National Institute on Aging ( grant number U01AG009740 ) and is conducted by the University of Michigan . Having defined the US genetic ancestry groups , we then considered the distribution of Native American admixture within and between ancestry groups and among geographic regions . We provide a detailed description ( S1 Text ) , along with supporting results ( S3–S8 Figs , S2 and S3 Tables ) , of how we defined the three main US ancestry groups–African descendants , Western European descendants , and Spanish descendants–in the Supplementary Material . The ancestry distribution of HRS individuals among the three largest US genetic ancestry groups is shown in Fig 1 . Visual inspection of the continental ancestry fractions seen for members of the three groups supports our approach to genetic ancestry-based classification ( Fig 1A ) . For example , the majority of Spanish descendant individuals show substantially higher Native American ancestry compared to Western European descendants ( Fig 1A ) ; the median Native American ancestry for the Spanish descendant group is 38% compared to 0 . 1% for the Western European descendant group ( Fig 1B ) . In addition , individuals from the Spanish descendant group cluster tightly with the Mexican reference population from the 1KGP , along the second axis between the European and Native American populations in the principal components analysis ( PCA ) plot of the pairwise genome distances ( Fig 1C ) . It is important to note that we did not use Native American ancestry for the purposes of classification . Rather , European ancestry alone was sufficient to recapitulate known levels of Native American ancestry for Spanish descendants . Individuals from the African descendant group show medians of 85% African ancestry , 14% European ancestry , and 1% Native American ancestry ( Fig 1B ) . Most of these individuals group along the first PCA axis separating the African and European reference populations . In contrast to the admixed Spanish and African descendant groups , Western European descendants show extremely low levels of admixture with non-European populations , with a median value of 99 . 8% European ancestry . Given their relatively low numbers ( S2 Fig ) , as well as their relatively late historical arrival in the continental US , we did not consider Asian descendants further in this study . Individuals assigned to the three main genetic ancestry groups show distinct geographic distributions across the continental US , which are largely consistent with demographic data for the country . The proportion of African descendants is highest in the three southern census regions , Western European descendants in the two north central regions , and Spanish descendants in the Mountain census region , which includes Arizona and New Mexico ( Fig 1D ) . We compared the patterns and extent of sex-biased admixture among the three US genetic ancestry groups by comparing the continental ancestry fractions–African , European , and Native American–seen for the X chromosomes versus the autosomes . For any given ancestry component , a relative excess of X chromosome ancestry is indicative of female-biased admixture , whereas an excess of autosomal ancestry reflects male-biased admixture [11] . This was only done for admixed individuals that had two or more continental ancestry fractions at >1 . 5% of the overall ancestry . Almost all individuals from the African and Spanish descendant groups met this criterion , but only a small minority of Western European descendant individuals with Native American admixture did . African and Spanish descendant groups showed marked patterns of sex-biased admixture , whereas the Western European descendants did not show any appreciable evidence of sex-biased admixture ( Fig 2 ) . The strongest pattern of sex-biased admixture was seen for Spanish descendants , with female-biased Native American admixture and male-biased European admixture . African descendants show female-biased African ancestry and male-biased European ancestry . For each US genetic ancestry group , we considered three distinct characteristics of Native American ancestry across the continental US: ( 1 ) the relative levels of Native American ancestry genome-wide , ( 2 ) the patterns of Native American allele frequencies , and ( 3 ) the phylogenetic relationships among US populations based on their Native American ancestry . As we showed previously , overall Native American ancestry is highest for the Spanish descendant group ( median 38% , SD = 20 . 1 ) , followed by the African descendant ( 1% , SD = 4 . 4 ) and Western European descendant groups ( 0 . 1% , SD = 2 . 7 ) ( Fig 1B ) . Among all three ancestry groups , the highest levels of Native American ancestry are seen for the West-South-Central ( WSC; including Texas ) , Pacific ( PAC; including California ) , and Mountain ( MNT; including Arizona and New Mexico ) census regions ( Fig 3 ) . Native American ancestry levels show the highest variability among regions for the Spanish descendant group ( coefficient of variation [c . v . ] = 1 . 08 ) , followed by the Western European descendant ( c . v . = 0 . 65 ) and African descendant ( c . v . = 0 . 60 ) groups . We characterized the ancestry-specific and genome-wide haplotype heterozygosity ( HH ) for each of the admixed populations to interrogate how admixture has affected the diversity of the populations ( S9 Fig , S4 Table ) . Where present , the African-specific HH was the highest for each population and the Native American HH was the lowest , consistent with previous observations of present day populations [12] . The genome-wide HH was significantly higher than any ancestry-specific HH for the African descendant populations , consistent with the introduction of novel haplotypes into the already diverse African background . Spanish descendant genome-wide HH was significantly higher than both the European and Native American-specific HH , but lower than African , which contributes only a small fraction of the total ancestry in the present-day Spanish descendant populations . The Western European descendant populations show a relatively very small amount of Native American ancestry; accordingly , the genome-wide HH shows no significant difference from the European HH , but is nevertheless higher than the Native American HH . We measured the patterns of Native American allele frequencies across the continental US using ADMIXTURE analysis of Native American haplotypes for individuals from the three ancestry groups . Visualization of the ancestry vectors produced by ADMIXTURE shows that the African and Western European descendant groups have patterns that are similar to each other ( Fig 4A , top panel; S10 Fig ) and distinct from the patterns seen for the Spanish descendant group ( Fig 4B , top panel; S11 Fig ) . Comparing the ADMIXTURE vectors of these two population groups to those of the Spanish descendant populations shows that African descendant and Western European descendant populations are significantly closer to each other than either is to the Spanish descendant populations ( S1 Text , S12 and S13 Figs ) . Furthermore , the African descendant and Western European descendant groups show ancestry patterns that are intermediate to the Canadian and Northern Mexican Native American reference populations , whereas the Spanish descendant group shows Native American ancestry patterns that are more similar to the Mexican reference population , Mexican Native American populations , or the admixed Puerto Rican population . This is consistent with the fact that we use Native American reference populations from outside the US to identify Native American haplotypes in US population groups . There is substantial regional variation in Native American ancestry seen in the Spanish descendant group , with characteristically Mexican patterns seen in the Pacific ( PAC ) and West South-Central ( WSC ) regions and a strongly Puerto Rican pattern in the Mid-Atlantic ( MA ) region . At K = 9 , ADMIXTURE is able to resolve the two Northern Mexican Native American reference populations , as well as reveal a unique ancestry in the Spanish descendant population from the Mountain ( MNT ) region . This population shows a distinct pattern of Native American ancestry under any K ( S11 Fig ) , which we explore in more detail in the following section . The phylogenetic relationships of the Native American ancestry in modern US populations were inferred by calculating the fixation index ( FST ) between pairs of populations based on their masked Native American haplotypes ( Fig 5 ) . The Canadian and Amazonian Native American reference populations occupy the most distant clades on the phylogeny with the admixed Mexican and Mexican Native American reference populations adjacent to the Amazonian group . African descendant populations from all of the census regions form a single clade , along with Western European descendants from the Southeast region ( SE , WD ) . Western European descendant populations from the West North-Central ( WNC , WD ) and East North-Central ( ENC , WD ) regions group most closely with the Canadian Native American reference populations . Western European descendant populations from the Western US ( West South-Central ( WSC , WD ) , Pacific ( PAC , WD ) , and Mountain ( MNT , WD ) regions ) are intermediate between the African descendant clade and the Spanish descendant of populations . Spanish descendant populations from most of the US census regions group closely with Mexican populations , with the exception of the Mid-Atlantic region ( MA ) which groups most closely with the Puerto Rican and Amazonian reference populations . To quantify the affinities of the Native American ancestry in admixed US populations we computed outgroup f3-statistics of the form f3 ( African; admixed , reference ) and D-statistics of the form D ( African , admixed; Native American , reference ) using the masked Native American haplotypes and AdmixTools [13] . The f3 and D-statistics agree well with the inferred phylogeny ( S14 & S15 Figs ) . Western European descendants from WNC and ENC regions showed the highest affinity for Canadian Native American populations . Consistent with the single clade observed in the phylogeny , African descendant populations showed generally lower affinity to the reference populations . Spanish descendant populations showed the highest affinities for Mexican reference populations , apart from the MA population which showed higher affinity for Amazonian groups . The ADMIXTURE results for the Spanish descendant group in the Mountain region ( MNT ) point to the presence of two distinct sub-populations , one of which is clearly of Mexican descent , whereas the second group has a pattern distinct from any other group analyzed here ( Figs 4B and 6A ) . If these two apparent Spanish descendant Mountain sub-populations are considered separately , they form distinct phylogenetic groups ( Fig 6B ) . One group clearly falls into the clade with the other Mexican origin populations ( MNT , Mexican ) , whereas the distinct group is basal to the Mexican clade and intermediate between the Western US and Mexican clades ( MNT , Nuevomexicano ) . The results of the ADMIXTURE and phylogenetic analyses are consistent with historical records indicating the presence of a unique group of Spanish descendants in the American Southwest , known as the ‘Hispanos of New Mexico’ or Nuevomexicanos . This population is descended from very early Spanish settlers to the Four Corners region of the US , primarily New Mexico and southern Colorado , and distinct from Mexican-American immigrants who arrived later [14] . Members of the Nuevomexicano population have maintained a distinct cultural identity for centuries , and the ability to isolate individuals from this group based on analysis of their genotypes allowed us to address open questions related to their ancestry . In addition to characterizing their distinct pattern of Native American ancestry , we also compared the levels of Native American admixture between Nuevomexicanos and the other nearby Spanish descendant groups , which show a Mexican pattern of Native American ancestry . Consistent with previous results [15] , we show that Nuevomexicanos have significantly more European ancestry and less Native American ancestry than other Spanish descendant groups from the Western Census regions ( Fig 6C ) . Nuevomexicanos also show significantly lower levels of African ancestry compared to the other Spanish descendant groups . Nuevomexicano cultural and historical traditions suggest that many of the early Spanish settlers in the region were Conversos , Jewish individuals who ostensibly converted to Catholicism in an effort to avoid religious persecution and pogroms , while secretly maintaining Jewish identity and traditions [16] . We interrogated this idea by comparing the extent of Sephardic Jewish admixture found among individuals with the Nuevomexicano ancestry pattern compared to other Spanish descendant populations . Sephardic Jewish admixture was measured by comparing European haplotypes from Spanish descendant individuals to a reference panel including both European and Sephardic Jewish populations . Nuevomexicanos show elevated levels of matching to Jewish haplotypes compared to Spanish and other European populations , consistent with substantial Converso ancestry among New World Spanish descendant populations [17] ( Fig 6d ) . However , Nuevomexicanos do not show a higher level of Converso ancestry compared to the other New World Spanish descendant populations .
We were able to delineate three predominant genetic US ancestry groups–African descendant , Western European descendant , and Spanish descendant–using comparative analysis of whole genome genotypes from >15 , 000 individuals from across the continental US . Each of these different groups of people experienced distinct historical trajectories in the US , which we found to be manifested as group-specific patterns of Native American ancestry . Individuals from the African descendant group show low ( Fig 1B ) and relatively invariant ( Fig 3A ) levels of Native American ancestry across the continental US . The patterns of Native American ancestry seen for the African descendant group are also more constant among US census regions compared to individuals from the other two ancestry groups ( Fig 4A ) . With respect to the Native American component of their ancestry , African descendant populations from all US census groups form a single clade , along with the Southeast Western European descendant population ( SE , WD ) ( Fig 5 ) . Considered together , these results point to a most likely scenario whereby African descendants admixed with local Native American groups in the antebellum South . Early admixture with Native Americans in the South was followed by subsequent dispersal across the US during the Great Migration in the early to mid-twentieth century [18] . The genetic legacy of the Great Migration has previously been explored based on overall patterns of African American genetic diversity [19] . Here , we were able to uncover traces of this same history based solely on the relatively low Native American ancestry component found in the genomes of African descendants . Of the three US ancestry groups characterized here , the Western European descendant group shows the lowest levels of Native American ancestry ( Fig 1B ) , consistent with a large and fairly constant influx of European immigrants to the US along with social and legal prohibitions against miscegenation [20] . Compared to African descendants , individuals from the Western European descendant group show more variant levels of Native American ancestry among US census regions ( Fig 3B ) along with substantially more region-specific patterns of Native American ancestry ( Fig 4A ) . Their region-specific patterns of Native American ancestry are also reflected in the Native American ancestry-based phylogeny , whereby the Western European descendant populations are related according to their geographic origin across the country ( Fig 5 ) . These results point to a historical pattern of continuous , albeit infrequent , admixture between local Native American groups and European settlers as they moved westward across the continental US . As can be expected , the Spanish descendant group shows by far the highest ( Fig 1B ) and most variable ( Fig 3C ) levels of Native American ancestry across the US . Individuals from this group show highly regional-specific patterns of Native American ancestry ( Fig 4B ) , consistent with known demographic trends . For example , analysis of the Native American component of Spanish descendant ancestry is sufficient to distinguish Puerto Rican immigrants from the Mid-Atlantic census region from Mexican Americans who predominate in the western census regions . Perhaps most striking , the patterns of Native American ancestry seen for the Mountain census regions were alone sufficient to distinguish descendants of very early Spanish settlers to the region , the group known as Hispanos or Nuevomexicanos , from subsequent waves of Spanish descendants who arrived later from Mexico . The three main US ancestry groups are also distinguished by their patterns of sex-biased ancestry in a way that reflects the unique history of each group ( Fig 2 ) . Western European descendants show very little evidence for sex-biased ancestry , along with very low levels of overall admixture , compared to the African and Spanish descendant groups . Sex-bias for Spanish descendants is characterized by a strong female-bias for Native American ancestry coupled with European male-biased ancestry . The pattern that we observe here is similar to what has been reported in a number of previous studies and is consistent with the history of male-biased migration to the region dating back to the era of the conquistadors [21 , 22] . The African descendant group shows female-biased African ancestry and male-biased European ancestry , a pattern which has also been documented previously and tied to the legacy of slavery and racial oppression in the US [23 , 24] . It has not been previously possible to directly compare the extent of sex-biased admixture among the three largest ancestry groups in the US as we have done here . As such , it is interesting to note that the history of the Spanish colonization in Latin America had a stronger impact on sex-biased ancestry than the legacy of slavery in the US . Our ability to distinguish Nuevomexicanos from the HRS dataset , using their distinct Native American ancestry , allowed us to address a number of open questions and controversies regarding the history and culture of this interesting population . Nuevomexicanos from the American southwest are historically defined as the descendants of early Spanish settlers , those who arrived in the period from 1598 to 1848 , as opposed to immigrants from Mexico who arrived the region considerably later . The two distinct patterns of Native American ancestry seen for Spanish descendant individuals from the Mountain census region are very much consistent with this historical definition . The Nuevomexicanos show a pattern of Native American ancestry that is intermediate to the Canadian and Mesoamerican reference populations analyzed here , whereas the Mexican American individuals from the same region are more closely related to Mesoamerican reference populations . This is consistent with early admixture with local Native American groups in the US southwest , for the Nuevomexicanos , versus admixture with Mesoamerican groups in Mexico for the later Mexican immigrants . A more precise characterization of Nuevomexicanos’ Native American ancestry would require access to genomic data from US Native American reference populations , which are not readily available owing to cultural resistance to genetic testing for ancestry among these groups [25] . Historically , Nuevomexicanos have identified strongly with their European ( Spanish ) ancestry , while downplaying ancestral ties to Native Americans [26] . This tradition of exclusive European identity is rooted in the colonial era when Spanish descendants in the region were preoccupied with the notion of maintaining so-called pure blood , and the local aristocracy identified as Castilian . The Spanish preoccupation with admixture in the Americas was codified into the so-called Sistema de Castas , whereby mixed-race individuals were categorized into a complex hierarchical system , with tangible legal and social implications , based on their parents’ ancestry [27] . Mexicans , on the other hand , have long identified as Mestizo with an explicit recognition of their Native American heritage [28] . Our comparative analysis of genetic ancestry for Nuevomexicanos and Mexican ancestry groups yielded results that are partly consistent with this historical narrative . On the one hand , Nuevomexicanos do have a substantial amount of Native American ancestry , with a median of just under 40% ( Fig 6C ) , which is far more than seen for the African descendant and Wester European descendant groups analyzed here . The fraction of Native American ancestry seen in the Nuevomexicanos is also higher than in several populations in South America ( Medellín [29] and Chocó [30] , Colombia ) and the Caribbean ( Cuba , the Dominican Republic , and Puerto Rico [29] ) . Nevertheless , the Nuevomexicanos have significantly less Native American ancestry , and more European ancestry , than nearby Mexican descendant populations ( Fig 6C ) . Our results are consistent with a recent study that used microsatellite-based ancestry analysis on a much smaller sample of self-identified Nuevomexicanos , who were also found to have higher European ancestry and lower Native American ancestry compared to Mexican Americans [15] . Interestingly , we found that the Nuevomexicanos also have significantly less African ancestry than Mexican descendant populations , which likely reflects higher levels of early African admixture in Mexico [31] . We investigated this apparent differing population history by inferring the timings and proportions of admixture with the TRACTS utility [32] . The best models from the TRACTS analysis indicated a European and Native American admixture 10–11 generations ago , followed shortly by a small African admixture ( S16 Fig ) . All models for the Mexican populations converged on an admixture time of 10–11 generations . The best Nuevomexicano model suggests a slightly older admixture , though with the same ordering , 11–12 generations ago , while the best Nuevomexicano model for 10–11 generations produced a significantly worse model ( log-likelihood of -413 vs . -390 ) . Regardless , this suggests that the timing of admixture in the Mexican populations and the Nuevomexicano population was similar , consistent with historical records , while the Native American source populations were different , consistent with their geographical origins . While the admixture timing estimates for these groups are within the range of previous estimates , they are younger than what has been previously reported for Mexican populations [33] . Nevertheless , as can be expected for a Caribbean population , the Puerto Rican descendant MA population showed a much older admixture , ~15 generations ago , very similar to the 1KGP Puerto Rican population ( S17 Fig ) . Perhaps the most controversial aspect of Nuevomexicano history relates to the influence of Conversos on the culture and traditions of the local community . Conversos are Jewish people who converted to Catholicism under intense pressure from religious persecution in Spain , and elsewhere in Europe , and many Spanish Conversos immigrated to the New World [34] . Despite their forced conversion to Catholicism , some New World Conversos apparently maintained Jewish religious traditions over the centuries since their immigration from Spain . For example , the persistence of rituals and symbols related to Jewish traditions in New Mexico has been taken as evidence for an influential presence of Conversos among the Nuevomexicanos , a position championed by the historian Stanley Hordes[16] . On the other hand , the folklorist Judith Neulander and others have been fiercely critical of this narrative based on what they perceive to be misunderstandings of the origins of many of the cultural traditions tied to Jewish rituals and even deliberate misrepresentations of facts [35] . Neulander’s interpretation relates the notion of Converso identity among Nuevomexicanos back to the colonial assertions of pure Spanish ancestry given that the Sephardim are Spanish and would presumably be loath to marry outside of their religious group [36] . We evaluated the extent of Sephardic Jewish ancestry among Nuevomexicanos , via comparative analysis of their European haplotypes to both European and Sephardic Jewish reference populations , in attempt to assess the genetic evidence in support of the Converso narrative . While we did find more Sephardic Jewish ancestry among Nuevomexicanos compared to Spaniards or other Europeans , they did not show any more Sephardic Jewish ancestry than Mexican descendants from nearby regions ( Fig 6D ) . Our results are consistent with a recent study that used haplotype-based ancestry methods to uncover widespread Converso ancestry in Latin American populations [17] . Taken together , we interpret these results to indicate that , while Nuevomexicanos do in fact have a demonstrable amount of Jewish ancestry , they show no more , or less , Jewish ancestry than other New World Latin American populations . Of course , we cannot weigh in on the strength of evidence for or against the persistence of Jewish cultural traditions among Nuevomexicanos based on our genetic evidence alone . Nevertheless , there does not seem to be anything particularly unusual , at least from the genetic perspective , with respect to the extent of Sephardic Jewish heritage among Nuevomexicanos . Much of the genetic legacy of the original inhabitants of the area that is now the continental US can be found in the genomes of the descendants of European and African immigrants to the region . In this study , we analyzed signals of Native American genetic ancestry in a comparative analysis of genomes from the three largest US ancestry groups: African descendants , Western European descendants , and Spanish descendants . Our study was enabled by the use of haplotype-based methods for genetic ancestry inference and leveraged a large dataset of whole genome genotypes . This approach allowed for detailed analysis of Native American ancestry patterns even when the per-genome levels of Native American ancestry were quite low . Each of the three genetic ancestry groups analyzed here shows distinct profiles of Native American ancestry , which reflect population-specific historical patterns of migration and settlement across the US . Analysis of the Native American ancestry component for members of these groups allowed for the delineation of region-specific subpopulations , such as the Nuevomexicanos from the American southwest , and facilitated the interrogation of specific historical scenarios .
This study was approved by the Georgia Institute of Technology Central Institutional Review Board , #H17029 . Data were provided by third party sources and no additional ethical approval was required . Whole genome genotype data of US individuals from the Health and Retirement Study ( HRS ) dataset ( n = 15 , 620 ) were merged with whole genome sequence variant data from the 1000 Genomes Project ( 1KGP ) [37 , 38] ( n = 1 , 718 ) and whole genome genotype data from the Human Genome Diversity Project ( HGDP ) [12 , 39 , 40] ( n = 230 ) ( S1 Table ) . Individual HRS genotypes are provided along with geographical origin data for sample donors from the nine census regions in the continental US . A collection of Native American genotypes from 21 populations across the Americas was taken from a comprehensive study on Native American population history [2] ( n = 314 ) . These Native American genotype data were accessed according to the terms of a data use agreement from the Universidad de Antioquia . Whole genome genotype data from 5 populations of Sephardic Jewish individuals ( n = 40 ) were also included as reference populations [41] . The genotypes from HRS individuals were merged with the comparative genomic data sources using PLINK version 1 . 9 [42] , keeping only those sites common to all datasets and correcting SNP strand orientations for consistency as needed . The final merged dataset includes 228 , 190 SNPs across 17 , 882 individuals . Pairwise distances between individuals was calculated using the–dist option of PLINK [42] , and principal component analysis carried out using the prcomp function of R [43] . The merged genotype dataset was phased using ShapeIT version 2 . r837 [44] . SNPs that interfered with the ShapeIT phasing process were excluded from subsequent analyses . ShapeIT was run without reference haplotypes , and all individuals were phased at the same time . Individual chromosomes were phased separately , and the X chromosome was phased with the additional ‘-X’ flag . The RFMix algorithm [45] is able to accurately characterize the local ancestry of admixed individuals but is prohibitively slow when run on a dataset of the size used here . To reduce the runtime , we modified RFMix version 1 . 5 . 4 so that the expectation-maximization ( EM ) procedure samples from , and creates a forest for , the entire set of individuals rather than each individual . This modified RFMix was run in the PopPhased mode with a minimum node size of five , using 12 generations and the “—use-reference-panels-in-EM” for two rounds of EM , generating local ancestry inference for both the reference and admixed populations . Continental African , European , and Native American populations were used as reference populations . Contiguous regions of ancestral assignment , “ancestry tracts , ” were created where RFMix ancestral certainty was at least 95% . Genome-wide ancestry estimates from the modified RFMix algorithm closely correlate with those from ADMIXTURE ( S18 Fig ) . The present Native American reference populations may not be close to the actual ancestral Native American populations for all of the HRS regions . To evaluate how a distant reference population would affect the LAI , we carried out the RFMix procedure a second time , but using only East Asian populations as the reference for Native American ancestry . The local ancestry inferred in this was very similar to that inferred when using actual Native American populations as references ( S19 Fig ) , indicating that the choice of reference population does not greatly affect the LAI . For each admixed population , rephased genotypes from the final output of RFMix were used to compute the haplotype heterozygosity ( HH ) for both the masked ancestry-specific genomes and for the unmasked whole-genome . Haplotypes were found by considering sets of 5–15 consecutive variants with a maximum recombination rate between any two variants of 0 . 5 cM/mB as in [46] , resulting in 11 , 816 haplotypes . Significance in HH between ancestry-specific genomes was assessed using a Wilcoxon rank-sum test . The extent of Sephardic Jewish ( Converso ) ancestry in individuals from the Spanish descendant group in HRS ( as defined in the genome-wide ancestry section below ) , and Latin American populations from 1KGP , was inferred via ancestry-specific haplotype comparisons with Sephardic Jewish reference populations using the program ChromoPainter2 [10] ( kindly provided by Garrett Hellenthal ) . First , African and Native American haplotypes were masked from the RFMix output . Then , the remaining European haplotypes were compared against genomes from the European reference populations together with the Sephardic Jewish populations . The extent of Jewish ancestry for any individual genome is defined as the ‘copying fraction’ from the Sephardic Jewish populations , where the copying fraction is taken as the fraction of sites with best matches to the Sephardic Jewish reference genomes . It should be noted that this procedure results in a relative fraction of Sephardic Jewish ancestry for all individuals under consideration , which is directly comparable among individuals but likely to be an overestimate of the total ancestry derived from a single source population . ADMIXTURE [47] version 1 . 3 . 0 was used with K = 4 to infer continental ancestry fractions for individuals in the dataset via comparison with reference populations from Africa , Europe , the Americas , and East Asia . Sub-continental ancestry was inferred independently for each of the three major continental ancestry components–African , European , and Native American–using an ancestry-specific masking procedure that we developed as previously described [30] . This procedure relies on the local continental ancestry assignments , along with the re-phased genotypes , generated by RFMix as described above . Sub-continental ancestry was characterized by first masking out two of the three continental ancestries ( African , European , and/or Native American ) at a time and then analyzing the genomic regions ( haplotypes ) corresponding to the remaining continental ancestry . For sub-continental ancestry analysis of any given continental ancestry component , only those individuals with at least 1 . 5% genome-wide ancestry for that same continental group were used . This 1 . 5% threshold was chosen empirically based on observed ancestry assignments in the reference populations . As this work was focused on Native American ancestry , we chose a threshold higher than the Native American ancestry inferred in any of the European or African reference populations ( max = 1 . 4% in a Spanish individual ) . While lowering this threshold would likely include a number of additional individuals with genuine Native American ancestry , we chose this stricter cutoff to avoid any possible ambiguity . We developed a novel machine learning based approach to distinguish Spanish from other ( primarily Western ) European descendants in the HRS dataset via analysis of European-specific haplotypes . First , ADMIXTURE was run with K = 5 on the RFMix characterized European haplotypes for the HRS individuals to stratify sub-continental European ancestries based on comparison with Northern ( Finnish and Russian ) , Western ( French and British ) , Spanish , and Southern ( Italian and Sardinian ) European reference populations from the 1KGP and HGDP datasets . The ADMIXTURE results at K = 5 were used as one of the ADMIXTURE components was substantially different between the Spanish and Italian reference populations ( Fig 4 , S3 and S4 Figs ) . A Support Vector Machine ( SVM ) classifier [48] was then trained using the resulting ADMIXTURE ancestry vectors for the European reference populations from the four sub-continental groups: Northern , Western , Spanish , and Southern . The European-specific ADMIXTURE ancestry vectors for the HRS individuals were then classified into one of the four European sub-continental groups defined by the SVM classifier . A confidence threshold of 0 . 8 was used for sub-continental group assignments in order to minimize the number of misclassified individuals; while a lower threshold would allow for additional individuals to be included , a threshold below 0 . 7 lead to a higher missassignment rate while validating the classifier . For the purpose of analysis here , we consider two major groups of European descendants in the HRS data set: Spanish descendants ( SD ) and all others . Non-Spanish HRS individuals with <5% African ancestry are defined as Western European descendant ( WD ) , whereas non-Spanish HRS individuals with at least 20% African ancestry were defined as African descendant ( AD ) . It should be noted that this approach to defining genetic ancestry groups , as opposed to relying on self-identified race/ethnicity groups , is likely to yield ancestry classifications that correspond very well to self-identified race/ethnicity labels for the vast majority of individuals analyzed here . But our African descendant group will not include a small fraction of self-identified African Americans with little or no African ancestry . For example , there are 11 individuals in HRS who self-identify as African American but have no discernable African ancestry . We chose to rely on genetic ancestry , as opposed to self-identified race/ethnicity , in such cases in an effort to be as consistent as possible when delineating the three broad ancestry groups . We discuss this issue at more length in the Supplementary material ( S1 Text ) . Sex-biased ancestry contributions were inferred by comparing the RFMix characterized fractions of each continental ancestry component on the X chromosomes versus the autosomes as previously described [22 , 46] . For each individual genome , and each ancestry component , the normalized difference between the X chromosome ancestry fraction and the autosomal ancestry fraction ( ΔAdmix ) is defined as: ΔAdmix=Fanc , total× ( Fanc , X-Fanc , auto ) / ( Fanc , X+Fanc , auto ) where Fanc , total , Fanc , X , and Fanc , auto are the genome-wide , X chromosome , and autosome ancestry fractions , respectively . We used the RFMix defined Native American haplotypes for individuals from the HRS and reference populations to infer the phylogenetic relationships between populations . Using the masked Native American haplotypes , the FST was found between each population using smartpca from the EIGENSOFT package [49] . The resulting FST distance matrix was used to create a neighbor-joining tree [50] with the program MEGA6 [51] . Clade bootstrap values were calculated by resampling sites from the data , recalculating FST , and counting the occurrences of each clade using prop . part and part . clades of the Ape package [52] . The TRACTS method was used to infer the timing of admixture events with ancestry tracts defined by RFMix [32] . For the admixed Nuevomexicano , Mexican ( 1KGP ) , MA Spanish descendant , and Puerto Rican ( 1KGP ) populations , three possible orderings of admixture were evaluated with TRACTS: ( 1 ) European , Native American , and African; ( 2 ) European , African , and Native American; and ( 3 ) African , Native American , and European . For each ordering , TRACTS was used to evaluate possible admixture timing from 14 to six generations ago , in 1000 bootstrap attempts . From the bootstrap attempts , the most likely series of admixture events was chosen to represent the population . | The post-Colombian settling of North America brought African , European , and Native American populations into close proximity for the first time . The inevitable admixture among these groups resulted a reservoir of Native American ancestry in modern US populations , outside of traditional Native American groups . Here we characterize that Native American ancestry in a geographically diverse set of African descendant , Western European descendant , and Spanish descendant populations . We show that Native American ancestry in the US population is not monomorphic , strongly related to geography , and suggestive of frequent historical admixture between European settlers and local Native American groups . We also show the presence of a unique , admixed Spanish population in the Southwestern US , the modern Nuevomexicanos , that is distinct from other Spanish descendant groups . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"united",
"states",
"evolutionary",
"biology",
"population",
"genetics",
"geographical",
"locations",
"census",
"genetic",
"mapping",
"ethnicities",
"north",
"america",
"research",
"design",
"latin",
"american",
"people",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"genomics",
"comparative",
"genomics",
"native",
"american",
"people",
"people",
"and",
"places",
"haplotypes",
"survey",
"research",
"heredity",
"genetics",
"population",
"groupings",
"mexican",
"people",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"europe"
] | 2019 | Native American admixture recapitulates population-specific migration and settlement of the continental United States |
Central players of the adaptive immune system are the groups of proteins encoded in the major histocompatibility complex ( MHC ) , which shape the immune response against pathogens and tolerance to self-peptides . The corresponding genomic region is of particular interest , as it harbors more disease associations than any other region in the human genome , including associations with infectious diseases , autoimmune disorders , cancers , and neuropsychiatric diseases . Certain MHC molecules can bind to a much wider range of epitopes than others , but the functional implication of such an elevated epitope-binding repertoire has remained largely unclear . It has been suggested that by recognizing more peptide segments , such promiscuous MHC molecules promote immune response against a broader range of pathogens . If so , the geographical distribution of MHC promiscuity level should be shaped by pathogen diversity . Three lines of evidence support the hypothesis . First , we found that in pathogen-rich geographical regions , humans are more likely to carry highly promiscuous MHC class II DRB1 alleles . Second , the switch between specialist and generalist antigen presentation has occurred repeatedly and in a rapid manner during human evolution . Third , molecular positions that define promiscuity level of MHC class II molecules are especially diverse and are under positive selection in human populations . Taken together , our work indicates that pathogen load maintains generalist adaptive immune recognition , with implications for medical genetics and epidemiology .
The major histocompatibility complex ( MHC ) genes in vertebrates encode cell surface proteins and are essential components of adaptive immune recognition [1] . MHC proteins are endowed with highly variable peptide-binding domains that bind short protein fragments . The MHC region is one of the most polymorphic gene clusters in vertebrate genomes [2] . Co-evolutionary arms race with pathogens is considered largely responsible for the observed exceptionally high levels of genetic diversity [3–6] , yet it cannot fully account for the observed geographic differences in human MHC genetic diversity [7 , 8] . This indicates that , beyond MHC allelic diversity , other MHC-related factors contribute to the capacity of human populations to withstand pathogens . In this paper , we argue that peptide-binding repertoire size ( or , shortly , promiscuity ) of MHC alleles is one important factor . Recent empirical studies demonstrated that there is a substantial variation in the size of the bound and presented antigen repertoire across MHC class I alleles . Certain MHC class I alleles appear to be promiscuous and are capable of binding an exceptionally large set of epitope peptide segments [9 , 10] . For example , Paul and colleagues carried out bioinformatics analysis to predict the binding capacity of common HLA-A and HLA-B alleles to a set of 30 , 000 dengue virus–derived peptides [11] . The analysis revealed over 16-fold variation in the number of peptides bound by the different alleles , indicating significant variation in epitope repertoire size across HLA molecules . The authors selected three alleles for further study in an in vivo transgenic mouse model . Immunization of the corresponding HLA transgenic mouse strains with a set of dengue virus–derived peptides revealed a positive relationship between epitope repertoire size and immunogenicity . Similarly , Kosmrlj and colleagues computed the fraction of self-peptides that bind to various HLA-B molecules , and found that this fraction varies extensively across four HLA-B alleles [12] . The authors then demonstrated that the self-peptide–binding repertoire of HLA-B shapes the native repertoire of T-cell clones developed in the thymus , with implications for recognizing human immunodeficiency virus ( HIV ) epitopes . Their results could explain why individuals carrying HLA-B*57 alleles can maintain low HIV RNA without therapy . Remarkably , analogous MHC class I alleles with the HLA-B*27 superfamily is widespread in Chinese rhesus macaques , animals which show especially slow progression of simian immunodeficiency virus ( SIV ) /HIV [13] . Finally , by focusing on seven chicken MHC class I haplotypes and four human HLA-B alleles , Chappel and colleagues demonstrated that MHC class I molecules that can bind a wide range of viral epitopes show lowered expression on the cellular surfaces of immune cells , such as monocytes and lymphocytes [9] . The authors suggested that the breadth of epitope-binding repertoire shapes genetic susceptibility to Marek’s disease virus in chickens and HIV disease progression in humans . More generally , by recognizing more peptide segments , promiscuous MHC molecules may promote immune response against a broader range of pathogens and are hence generalists [9 , 10] . Prior case studies in chicken indicate that this may be so [14–17] . However , it remains to be established whether this relationship generally holds across MHC class I and II alleles and a wide range of infectious diseases . Specifically , we propose that in regions of high pathogen diversity , human populations should carry promiscuous MHC alleles . Moreover , as migrating human populations have been exposed to changing sets of pathogens [18] , shifts in MHC promiscuity level should have occurred repeatedly and in a rapid manner during the course of human evolution . To test these predictions , we first focused on the human HLA class II DRB1 gene , for several reasons . First , DRB1 is the most variable HLA class II locus , with over 2 , 000 registered alleles [19] . Together with HLA-DRA , HLA-DRB1 encodes the heterodimeric HLA-DR protein complex , but HLA-DRA is basically invariant . Second , DRB1 shows the strongest general signature of selection among HLA class II loci [20] , while at the same time showing the weakest evidence for divergent allele advantage , an alternative mechanism at the genotype level for presenting a broader set epitopes [21] . Third , DRB1 has diversified very rapidly in the human lineage [22] . Many of the DRB1 alleles appear to be human specific and most likely evolved after the migration of ancestral human populations out of Africa [22] . These periods have been associated with human populations encountering numerous new pathogens [18 , 23] . For other HLA class II loci , the level of genetic diversity is lower [19 , 24] , probably driven by selection for functions partly unrelated to pathogens . Notably , HLA-DQ has a fundamental role in the development of immune tolerance [25 , 26] , while HLA-DP contributes to the presentation of epitopes of intracellular origins [27–29] . Fourth , epitope-binding prediction algorithms show higher accuracy for DRB1 than for other HLA class II loci [30 , 31] . Finally , the abundance of DRB1 on the cell surface is especially high compared with other HLA class II molecules [32–34] . Subsequently , we also evaluated promiscuity patterns of HLA class I molecules . Estimates on epitope-binding promiscuity were derived from two sources: experimental assays that measured individual peptide–MHC interactions in vitro and systematic computational predictions . In a series of analyses , we show that predictions of our hypothesis are upheld , regardless of how HLA-DRB1 promiscuity level is estimated .
Given that large-scale experimental assays to measure individual peptide–MHC interactions are extremely tedious , we first employed established bioinformatics tools to predict the binding affinities of experimentally verified epitope peptides for a panel of 162 nonsynonymous HLA-DRB1 alleles , all of which are present at detectable frequencies in at least one human population [35–37] . The set of investigated epitopes was derived from the Immune Epitope Database ( IEDB ) and contains 2 , 691 peptide epitopes of 71 pathogens known to be bound by certain HLA class II alleles [38] ( S1 Data ) . Epitopes showing high levels of amino acid similarity to each other were excluded from the analysis ( See Methods ) . Most included epitopes are 15 to 20 amino acids long and are found in only one of the 71 pathogens ( S1 Fig , S1 Data ) . The NetMHCIIpan algorithm was used to predict individual epitope–MHC interactions [30] , not least because it outperforms other prediction algorithms [31] . The breadth of epitope-binding repertoire or , shortly , the level of promiscuity of individual HLA-DRB1 alleles was estimated as the fraction of epitopes with a binding affinity stronger than 50 nM to the given MHC molecule . This threshold corresponds to high-affinity binding , which is frequently observed in MHC molecules displaying immunodominance [39] . We found large variation in promiscuity levels across HLA-DRB1 alleles ( S2 Fig , S2 Data ) . Using a smaller dataset with information from both approaches , we show that the computationally predicted and the empirically estimated promiscuity values are strongly correlated with each other ( Spearman’s rho: 0 . 78 , P = 0 . 004 , S3 Fig ) . Moreover , our results are robust to changes in the affinity threshold ( S4A to S4C Fig ) , usage of other prediction algorithms ( S4D Fig ) , and variation in the epitope data employed ( S4E Fig ) . As expected , promiscuous HLA-DRB1 alleles can present epitopes from a broader range of pathogen species ( S5 Fig ) . Reassuringly , there was no correlation between allele promiscuity values and the amount of data per allele used for the training of the algorithm ( Spearman’s rho: −0 . 38 , P = 0 . 21 ) . Taking advantage of the confirmed reliability of computational predictions , we next investigated the geographic distribution of HLA-DRB1 alleles . We first collected high-quality HLA-DRB1 allele prevalence data of 96 human populations residing in 43 countries from two databases and an article [35–37] . The weighted average of promiscuity level in each population was calculated based on the promiscuity values and allele frequencies of individual alleles in the population ( See Methods ) . The analysis revealed a large variation in mean promiscuity across geographical regions and the corresponding human populations ( S1 Table ) . Importantly , several distantly related but highly promiscuous alleles contribute to this pattern ( S1 Table ) . Notably , an especially high allelic promiscuity level was found in Southeast Asia , an important hot spot of emerging infectious diseases [40] . To minimize any potential confounding effect of high genetic relatedness between neighboring populations , we merged populations with similar HLA allele compositions for all further analyses ( See Methods ) . Using the Global Infectious Diseases and Epidemiology Network ( GIDEON ) , we compiled a dataset on pathogen richness in the corresponding 43 geographic regions [41] . It consists of 95 diseases caused by 168 extracellular pathogens , including diverse bacterial species , fungi , protozoa , and helminthes . Using the same protocol , we additionally compiled a dataset on the prevalence of 149 diseases in the same regions caused by 214 viral and other obligate intracellular pathogens . The dataset and methodology employed for the analysis are standardized and have been used previously in similar contexts [7 , 8 , 42] . We report a strong positive correlation between extracellular pathogen diversity and mean promiscuity: HLA-DRB1 alleles that can bind epitopes from a broader range of pathogens are more likely to be found in regions of elevated pathogen diversity ( Fig 1A ) . This pattern is unlikely to be explained by confounding factors , such as country size or HLA-DRB1 genetic diversity across countries ( S2 Table ) . By contrast , we found no significant association between HLA-DRB1 promiscuity level and diversity of intracellular pathogens ( Fig 1B ) . We conclude that the geographical distribution of promiscuous HLA-DRB1 alleles has been mainly shaped by the diversity of extracellular pathogens . The above results hold—and are even stronger—when estimates on promiscuity were derived from empirical in vitro MHC binding data ( shortly , in vitro promiscuity ) , downloaded from the IEDB database [38] ( Fig 1C and 1D , S2 Table and S3 Data ) . However , these results do not exclude the possibility that the geographical link between pathogen diversity and promiscuity is indirect . More direct support on the causal relationship between the two variables comes from analysis of prior human genetic studies . To investigate this issue , we focused on two geographically widespread allelic groups with exceptionally high ( HLA-DRB1*12 ) and exceptionally low ( HLA-DRB1*03 ) promiscuity values , respectively , and conducted literature mining on their reported associations with infectious diseases ( S3 Table ) . As expected , HLA-DRB1*12 was associated with protection against at least five infectious diseases , while HLA-DRB1*03 was associated with susceptibilities to eight infectious diseases , which is highly unlikely by chance ( Fisher test , P = 0 . 003 ) ( S3 Table , S5 Data ) . The data also indicate local adaptation towards elevated promiscuity under diverse pathogen pressure . The HLA-DRB1*12:02 allele is prevalent in specific regions of Southeast Asia . Compared with other alleles detected in this region , HLA-DRB1*12:02 has a relatively high promiscuity value ( top 20% , Fig 2A , S2 Data ) . The high frequency of HLA-DRB1*12:02 has been previously suggested to reflect pathogen-driven selection during the migration of a Mongolian population to South China [43] . Indeed , this allele is associated with protection from recurrent pulmonary tuberculosis , recurrent typhoid fever , and hepatosplenic schistomiasis ( S5 Data ) , all of which are endemic diseases in Southeast Asia [44–46] . Remarkably , the frequency of this allele increases with extracellular pathogen diversity in this region ( Fig 2B ) . Together , these observations support the hypothesis that promiscuous epitope binding of HLA-DRB1 alleles is favored by selection when extracellular pathogen diversity is high . An important unresolved issue is how promiscuity has changed during the course of human evolution . Under the assumption that local pathogen diversity drives the evolution of epitope recognition of HLA class II alleles , promiscuity as a molecular trait should have evolved rapidly as human populations expanded into new territories . To investigate this issue , we combined an established phylogeny of HLA-DRB1 alleles [47] with predicted epitope-binding promiscuity values . We found that alleles with a high promiscuity level have a patchy distribution across the tree ( S6 Fig ) , indicating that high promiscuity has multiple independent origins . To investigate this observation further , we selected a set of 96 HLA-DRB1 alleles with a detectable frequency in at least one human population and appropriate sequence data ( see Methods ) . A comparison of all pairs of these alleles revealed that even very closely related alleles show major differences in promiscuity levels ( Fig 3A ) . For example , alleles belonging to the HLA-DRB1*13 group show over 98% amino acid sequence identity to each other , but display as much as 57-fold variation in the predicted promiscuity levels . We conclude that the switch between high and low promiscuity levels has occurred repeatedly and in a rapid manner during the allelic diversification of the HLA-DRB1 locus . We next asked how selection on promiscuity has shaped the genetic diversity along the epitope-binding region of HLA molecules . To quantify protein sequence variability at each amino acid position , we calculated the Shannon entropy index based on the alignment of the 96 selected HLA-DRB1 alleles from above . For each position , we also calculated promiscuity fragility , that is , the median impact of single amino acid substitutions on promiscuity ( see Methods ) . A strong positive correlation was found between Shannon entropy and promiscuity fragility ( Fig 3B , Spearman’s rho = 0 . 76 , P = 0 . 0001 ) . Importantly , this conclusion does not depend on how sequence polymorphism was estimated ( S7 Fig ) . Accordingly , amino acid positions with a large impact on epitope-binding promiscuity are highly variable in human populations . Furthermore , those sites in the epitope-binding region that are under positive selection [48] tend to have high promiscuity fragility values ( Wilcoxon rank sum test , P = 0 . 0012 , Fig 3B; see also S8 Fig ) . The above data suggest a link between allele promiscuity and HLA-DRB1 diversification , probably as an outcome of selection for locally optimized promiscuity levels . Finally , we note that several variable molecular sites in the binding region of HLA-DRB1 affect epitope-binding characteristics without any major impact on promiscuity per se . For example , our computational analysis indicates that mutations at amino acid site numbers 9 and 47 do not seriously affect promiscuity level ( Fig 3B ) . However , several mutations at these sites are associated with binding self-peptides and thereby shape vulnerability to specific autoimmune diseases [49 , 50] . The relationship between pathogen diversity and epitope-binding promiscuity may be more general , as similar results hold for the HLA-A locus . HLA-A is one of the three types of classical human MHC class I molecules and is mainly involved in the presentation of epitopes from intracellular pathogens [51] . In agreement with expectation , we report a positive correlation between local intracellular pathogen diversity and the HLA-A promiscuity level of the corresponding human populations ( S9A and S9B Fig , S3 Data ) . No marked positive correlation was found for two other MHC class I genes ( HLA-B and HLA-C , see S9C to S9F Fig , and S3 Data ) . Therefore , other unrelated evolutionary forces may shape the geographical distribution of promiscuous HLA-B and HLA-C alleles ( S1 Text ) .
Central players of the adaptive immune system are the groups of proteins encoded in the MHC . By binding short peptide segments ( epitopes ) , MHC molecules guide both immune response against pathogens and tolerance to self-peptides . The genomic region encoding these MHC molecules is of special interest , for two reasons . It harbors more disease associations than any other regions in the human genome , including associations with infectious diseases , autoimmune disorders , tumors , and neuropsychiatric diseases [52 , 53] . A growing body of literature is now revealing that certain MHC class I alleles can bind a wider range of epitopes than others , but the functional implications of this variation remain largely unknown [10] . By recognizing a larger variety of epitopes , such promiscuous MHC alleles promote immune response against a broader range of pathogens at the individual level . Therefore , promiscuous epitope binding of MHC molecules should be favored by selection in geographic regions where extracellular pathogen diversity is high . Importantly , this mechanism is conceptually distinct from the well-established concept of heterozygote advantage at the MHC [54] , as it concerns individual alleles and not allele combinations or genotypes . To test this hypothesis , we combined data on the geographic distribution of human MHC class II alleles and prevalence of extracellular pathogens , empirical/computational estimates of epitope-binding promiscuity , and phylogenetic analyses . Our main findings , strongly supporting our hypothesis , are as follows . First , in geographical regions of high extracellular pathogen diversity , human HLA-DRB1 alleles have exceptionally high epitope-binding repertoires . This suggests that the geographical distribution of promiscuous HLA-DRB1 alleles has been shaped by the diversity of extracellular pathogens . The HLA-DRB1*12:02 allele highlights this point . HLA-DRB1*12:02 is a promiscuous allele that has been associated with protection from certain infectious diseases ( S5 Data ) . As expected , this allele is especially prevalent in regions of Southeast Asia with elevated pathogen load ( Fig 2B ) . It is well established that antigens presented by HLA class II molecules derive mainly from extracellular proteins [1] . However , HLA class II molecules have well-established roles in controlling immune response against viruses [55 , 56] . Additionally , viral peptides are reported to be processed and presented also by the HLA class II pathway [57] . Therefore , it remains to be established why intracellular pathogen diversity has no major impact on the global distribution of HLA-DRB1 alleles . Notably , the relationship between pathogen load and epitope-binding promiscuity may be more general , as similar results hold for the HLA-A locus: we found a positive correlation between local intracellular pathogen diversity and the HLA-A promiscuity level of the corresponding human populations ( S9A and S9B Fig , S3 Data ) . Second , a phylogenetic analysis revealed major differences in promiscuity levels of very closely related HLA-DRB1 alleles . This suggests that high promiscuity level in HLA-DRB1 has evolved rapidly and repeatedly during human evolution . Finally , amino acid positions with a prominent role in shaping HLA-DRB1 promiscuity level are especially variable in human populations and tend to be under positive selection . In sum , we conclude that HLA promiscuity level is a human trait with paramount importance during adaptation to local pathogens . Our work has important ramifications for future studies . MHC is the most variable region of the human genome , and the variation is associated with numerous infectious and immune-mediated diseases [52 , 53 , 58–62] . The impact of MHC promiscuity level on population allelic diversity is an interesting area for future research . In a similar vein , MHC allelic diversity is associated with olfaction-based mating preferences in human and other animals [63] . The roles of MHC promiscuity in mating success and mating preferences are a terra incognita . We note that the most promiscuous HLA-DRB1 alleles are rare in certain human populations ( S1 Table; S2 Data ) . This suggest that these alleles are not particularly favored by natural selection in these areas . Why should it be so ? First , high promiscuity may not be able to cope with the rise of novel and highly virulent pathogens . In such cases , displaying a particular epitope might be the most efficient way to achieve resistance , and high promiscuity might be suboptimal due to a reduced specificity [9 , 10] . Second , high promiscuity level may elevate the risk of immune reactions against host tissues and non-harmful proteins [9 , 64] . Clearly , future work should elucidate the evolutionary trade-offs between protection from pathogens and genetic susceptibility to autoimmune diseases . This will require high-throughput experimental methods to determine epitope-binding repertoire [65] , and HLA transgenic mice studies on the role of promiscuity in immune response [66] . Finally , genetic variation within particular MHC genes influences vaccine efficacy [67] , rejection rates of transplanted organs [68] , susceptibility to autoimmune diseases [49] , and antitumor immunity [28 , 69 , 70] . Our work raises the possibility that , by altering the maturation and functionality of the immune system , the size of the epitope-binding repertoire of MHC alleles itself could have an impact on these processes . The exact role of MHC promiscuity in these crucial public health issues is an exciting future research area .
The IEDB has collected the results of individual and systematic studies on epitope binding by MHC alleles [38] . The experimental studies include HLA-binding assays , T-cell activation assays , and immunopeptidomic studies as well . Epitopes of all available viral , bacterial , and eukaryotic pathogens known to be bound by at least one HLA-I or HLA-II allele were extracted from IEDB . Reference proteomes of pathogenic species that carry at least one of the collected epitope sequences were retrieved from the Uniprot database ( 102 for HLA-I and 71 for HLA-II epitopes ) [71] . Only epitopes of these species were analyzed further . All proteomes were scanned for each epitope sequence , and epitope sequences found in only one proteome ( i . e . , species-specific epitopes ) were kept for further analysis . Highly similar epitope sequences were identified using Clustal Omega [72] and excluded as follows . A protein distance matrix was created and epitopes were discarded iteratively . In each iteration , the epitope pairs with the lowest k-tuple distance were identified . Then , the epitope with the highest average similarity to all other sequences was excluded . Iterations were repeated until distance values less than 0 . 5 ( corresponding to greater than approximately 50% sequence identity ) were eliminated from the matrix [73] . Note that this filtering procedure was carried out separately for epitope sequences bound by HLA-I and HLA-II . Binding affinities of the remaining 3 , 265 HLA-I epitope sequences to 346 HLA-A , 532 HLA-B and 225 HLA-C alleles were predicted with the NetMHCpan-4 . 0 algorithm [74] . The binding of 2 , 691 HLA-II epitope sequences to 162 HLA-DRB1 alleles was predicted using the NetMHCIIpan-3 . 1 algorithm [30] . All 162 alleles are present in at least one of the human populations studied here ( see below ) . The “pep” sequence input format was used for both HLA-I and HLA-II epitope-binding prediction . A binding affinity threshold of 50 nM was applied , yielding peptides that are likely to be immunodominant [39] . For alternative binding threshold definitions , see S4 Fig . For each binding threshold , epitope-binding promiscuity was defined as the fraction of the epitope set bound by a given allele . To determine the epitope-binding promiscuity of HLA-DRB1 alleles based on previously published experimental data , we used the IEDB database [38] . Specifically , we downloaded all MHC ligand and T-cell assay data , which was available for 48 HLA-DRB1 alleles . Binding data of 20 alleles screened for at least 100 ligands each were further analyzed . The epitope set of each allele was filtered for highly similar sequences , as described above . As the majority of in vitro assay data were available in a binary format ( i . e . , presence or absence of binding ) , promiscuity was calculated as the fraction of positive binding assays for a given allele . To calculate population-level promiscuity values , we obtained HLA allele frequency data from the Allele Frequency Net Database ( AFND ) and the International Histocompatibility Working Group ( IHWG ) populations [35 , 36] . Haplotype-level data of the 13th International HLA and Immunogenetics Workshop ( IHIW ) populations were downloaded from dbMHC ( National Center for Biotechnology Information [NCBI]; ftp://ftp . ncbi . nlm . nih . gov/pub/mhc/mhc/Final%20Archive ) . Additionally , allele frequency data of the 14th and 16th IHIW populations , as published by Riccio and colleagues [37] , and populations in the AFND were used in the analyses . To avoid potential confounding effects of recent genetic admixture and migration , we focused on native populations , similarly to previous studies ( S1 Table ) [7 , 8] . We excluded IHWG populations reported to deviate from Hardy-Weinberg equilibrium [37] . Among the AFND populations and IHWG populations without haplotype-resolution data ( 14th and 16th IHIW ) , those comprising less than 100 genotyped individuals or those in which the sum of allele frequencies deviated from 1 by more than 1% were excluded . Populations reported in multiple databases were included only once in the analysis . For each HLA loci , we calculated mean population promiscuity by averaging promiscuity values of alleles weighted by their relative frequencies in the populations . In all of these calculations , we used standardized ( i . e . , z-score ) promiscuity values to make the in silico and in vitro values more easily comparable . Finally , when calculating population-level promiscuity based on in vitro promiscuity data , we excluded populations for which in vitro promiscuity values could be assigned to less than 50% cumulative allele frequency . To tackle the issue of nonindependence of data points , we focused on populations instead of countries and grouped those populations that have highly similar HLA allele compositions , based on standard measures of genetic distance ( see below ) . We merged populations with highly similar HLA allele compositions , allowing us to avoid pseudoreplication of data points while retaining informative allele frequency differences between populations residing in the same broad geographical areas . To this end , we first generated a genetic distance matrix between populations with the adegenet R library using allele frequency data of the examined locus . We used the Rogers’ genetic distance measure [75] because it does not assume that allele frequency changes are driven by genetic drift only , an unlikely scenario for HLA genes . Next , populations were merged using a network-based approach . Populations were treated as nodes and two nodes were connected if their genetic distance was under a cutoff value . Populations were grouped in an iterative manner . In each iteration , all maximal cliques ( i . e . , subsets of nodes that are fully connected to each other ) in the network were identified . Maximal cliques represent groups of populations in which all populations have similar allele compositions to each other . Then , mean genetic distance within each clique was calculated . The clique with the lowest average distance was selected and its populations were grouped together . Then , this clique was deleted from the network . Iterations were repeated until no maximal cliques remained in the network . Grouping of populations was carried out using different distance value cutoffs ( 1st , 5th , 10th , and 15th rank percentile of all distance values ) . The resulting population groups and the individual populations that remained in the network were treated as independent data points in subsequent statistical analyses . Mean promiscuity level in population groups was calculated by averaging population promiscuity values . Unless otherwise indicated , all figures are based on population groups using the 15th percentile genetic distance cutoff value . Importantly , using different cutoffs has no impact on our results ( S3 Data ) . Finally , we note that genetic differences among human populations mostly come from gradations in allele frequencies rather than from the presence of distinctive alleles [76] . Therefore , traditional clustering of populations based on HLA composition would have been ill-suited for our purposes . Data on 309 infectious diseases were collected from GIDEON [41] . For each disease , the number of causative species or genera ( when species were not listed for the genus ) was determined using disease information in the GIDEON database , as described previously [42] . Causative agents were classified into obligate intracellular and extracellular pathogen groups based on a previous study [7] and literature information . Putative facultative intracellular pathogens were excluded from the analysis . Diseases caused by agents that could not be clearly classified were also excluded from the analysis . Extracellular and intracellular pathogen diversity ( richness ) of each country was approximated by the number of identified endemic extracellular and intracellular species , respectively . Finally , we assigned country-level measures of pathogen and HLA diversity to population groups as follows . For each population group , extracellular and intracellular pathogen counts were calculated by averaging the corresponding country-level values across the populations within the group . For example , if a population group contained two populations residing in neighboring countries , then we assigned the average pathogen diversity of the two countries to it . To examine associations between selected HLA allele groups and infectious diseases , we carried out a systematic literature search on PubMed database using the following terms: “assoc* drb1 12 02” , “assoc* drb1 1202” , “assoc* drb1 12 01” , “assoc* drb1 1201” , “assoc* dr12” , “assoc* drb1*12” , “assoc* drb1 03 01” , “assoc* drb1 0301” , “assoc* dr3” , “assoc* drb1*03” , “assoc* dr17” , “infect* drb1 12 02” , “infect* drb1 1202” , “infect* drb1 12 01” , “infect* drb1 1201” , “infect* dr12” , “infect* drb1*12” , “infect* drb1 03 01” , “infect* drb1 0301” , “infect* dr3” , “infect* drb1*03” , and “infect* dr17” . “assoc*” and “infect*” refer to any word beginning with these letters . Each resulting paper containing HLA association data was examined , and statistically significant associations between allele groups ( DRB1*03 or DRB1*12 ) or common alleles in allele groups ( DRB1*12:01 , DRB1*12:02 , DRB1*03:01 ) and infectious diseases were collected . Associations with diseases caused by intracellular pathogens were excluded from the analysis . HLA-disease associations were classified as beneficial or detrimental , if all related studies supported the beneficial or detrimental role of HLA allele/allele group in the development or course of the given disease . Otherwise , the association was classified as controversial . The results were summarized ( S3 Table ) , and statistical association between beneficial/detrimental effects and high/low promiscuity across allele groups was determined by a Fisher’s exact test . We used amino acid distance as a proxy for phylogenetic distance between pairs of DRB1 alleles . To this end , nucleotide sequences of DRB1 alleles that contained full exon 2 and 3 regions were downloaded from the IPD-IMGT/HLA database [19] . To limit our analyses to alleles that have an impact on the inferred promiscuity level of a population , we considered only those sequences that had a nonzero frequency in at least one human population ( see above ) . From allele groups that code for the same protein sequence ( synonymous differences , differentiated by the third set of digits in the HLA nomenclature ) , one of the alleles was randomly chosen . This selection procedure resulted in 96 alleles . Multiple alignment of nucleotide sequences was performed using the MUSCLE algorithm as implemented in the MEGA software [77] and converted to protein sequence alignments . Amino acid distance was calculated using the Jones-Taylor-Thornton substitution model in MEGA [77] ( Fig 3A ) . Epitope-binding region sites—as defined previously [30]—were excluded when calculating amino acid distance . The rationale behind this exclusion is that these sites are known to be under positive selection [78 , 79] and are therefore less informative on evolutionary distance . Additionally , by removing these sites , the amino acid distance remains independent of promiscuity predictions . Finally , as intragenic recombination may distort the inference of evolutionary distance , we identified such events across all alleles following the protocol of Satta and colleagues [80] using GENECONV [81] and RDP algorithms [82] as implemented in the RDP software [83] . Recombinant alleles were removed when calculating amino acid distance . We first defined the epitope-binding region of HLA-DRB1 alleles , as previously [30] . To estimate sequence diversity along the epitope-binding region , we employed two measures: standard Shannon entropy [84] and nucleotide diversity ( π ) , a widely employed measure of genetic variation [85] . Using the protein sequence alignment of the 96 alleles defined above , we calculated amino acid sequence variability as the Shannon entropy of the given amino acid site as follows: ∑i=1MPilog2Pi where Pi is the fraction of residues of amino acid type i at a given site , and M is the number of amino acid types observed at that site . Nonsynonymous nucleotide diversity ( πA ) measures the average number of nonsynonymous nucleotide differences per nonsynonymous site between two randomly chosen protein coding DNA sequences from the same population [85 , 86] . πA was calculated for each amino acid site in the epitope-binding region for each population using DnaSP software [87] and custom-written R scripts . Nucleotide sequences of DRB1 alleles were downloaded from the IPD-IMGT/HLA database [19] . The calculation is based on the work of Nei and colleagues [85] using the equation πA=∑i , jxi*xj*πAij where xi and xj are the frequencies of the ith and jth alleles in the population , respectively , and πAij is the number of nonsynonymous nucleotide differences per nonsynonymous nucleotide site between the two codon sequences of the given amino acid site in the ith and jth alleles . To calculate πA for each population , allele frequency data of human populations were obtained , as described earlier ( see above ) . An overall nucleotide diversity index was calculated by averaging πA across populations . To calculate each amino acid site’s impact on epitope-binding promiscuity ( promiscuity fragility ) , promiscuity was predicted for each 19 possible amino acid change along the epitope-binding region of each of 96 alleles . The fold difference in promiscuity resulting from each amino acid substitution was calculated . The median promiscuity fold difference of each possible allele and amino acid change combination ( 96 × 19 ) was used to estimate promiscuity fragility at each amino acid position . As some of the 19 possible amino acid changes are not accessible via a single nucleotide mutation , and the accessible amino acid changes can have different likelihoods based on the codon sequence of the site and the genetic code , we also calculated promiscuity fragility based on each nonsynonymous nucleotide substitution of the codon instead of each amino acid substitution of the site . All statistical analyses were carried out in R version 3 . 2 . 0 [88] . Smooth curves were fitted using the cubic smoothing spline method [89] . | Variation in the human genome influences our susceptibility to infectious diseases , but the causal link between disease and underlying mutation often remains enigmatic . Major histocompatibility complex II ( MHC class II ) molecules shape both our immune response against pathogens and our tolerance of self-peptides . The genomic region that encodes MHC molecules is of particular interest , as it is home to more genetic disease associations than any other region in the human genome , including associations with infectious diseases , autoimmune disorders , cancers , and neuropsychiatric diseases . Here , we propose that MHC class II molecules can be categorized into two major types; specialists initiate effective immune response against only relatively few pathogens , while generalists provide protection against a broad range of pathogens . As support , we demonstrate that generalist MHC class II variants are more prevalent in human populations residing in pathogen-rich areas . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"sequencing",
"techniques",
"biogeography",
"medicine",
"and",
"health",
"sciences",
"ecology",
"and",
"environmental",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"intracellular",
"pathogens",
"pathogens",
"immunology",
"population",
"genetics",
"clinical",
"medicine",
"molecular",
"biology",
"techniques",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"ecological",
"metrics",
"infectious",
"diseases",
"geography",
"major",
"histocompatibility",
"complex",
"bioinformatics",
"biological",
"databases",
"phylogeography",
"short",
"reports",
"species",
"diversity",
"molecular",
"biology",
"nucleotide",
"sequencing",
"sequence",
"databases",
"clinical",
"immunology",
"ecology",
"earth",
"sciences",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology"
] | 2019 | Pathogen diversity drives the evolution of generalist MHC-II alleles in human populations |
Eukaryotic cells respond to genomic and environmental stresses , such as DNA damage and heat shock ( HS ) , with the synthesis of poly-[ADP-ribose] ( PAR ) at specific chromatin regions , such as DNA breaks or HS genes , by PAR polymerases ( PARP ) . Little is known about the role of this modification during cellular stress responses . We show here that the nucleosome remodeler dMi-2 is recruited to active HS genes in a PARP–dependent manner . dMi-2 binds PAR suggesting that this physical interaction is important for recruitment . Indeed , a dMi-2 mutant unable to bind PAR does not localise to active HS loci in vivo . We have identified several dMi-2 regions which bind PAR independently in vitro , including the chromodomains and regions near the N-terminus containing motifs rich in K and R residues . Moreover , upon HS gene activation , dMi-2 associates with nascent HS gene transcripts , and its catalytic activity is required for efficient transcription and co-transcriptional RNA processing . RNA and PAR compete for dMi-2 binding in vitro , suggesting a two step process for dMi-2 association with active HS genes: initial recruitment to the locus via PAR interaction , followed by binding to nascent RNA transcripts . We suggest that stress-induced chromatin PARylation serves to rapidly attract factors that are required for an efficient and timely transcriptional response .
The activity of eukaryotic genomes is regulated by dynamic changes in chromatin structure . A multitude of nucleosome remodeling enzymes , histone modifying activities and chromatin binding proteins cooperate to establish , maintain and reprogram chromatin structures that determine genome activity . Drosophila heat shock ( HS ) genes provide a textbook example of how dramatic changes in the organismal and cellular environment affect chromatin structure in a manner that promotes transcriptional activation of genes coding for molecular chaperones required during the HS response . Upon temperature shift , the HS loci of polytene chromosomes form transcriptionally active “puffs” . This rapid chromatin decondensation correlates with a strong decrease in nucleosome density [1] . Puff formation can be uncoupled from transcription and much of the nucleosome loss at the hsp70 gene occurs prior to the first round of transcription [1] , [2] . Recently , heat shock factor ( HSF ) , GAGA factor and poly-[ADP-ribose] polymerase ( PARP ) have been shown to be required for the rapid removal of nucleosomes upon activation of the hsp70 gene [1] . In addition , HS puffs accumulate PARylated proteins and puff formation depends on PARP activity [3] . The mechanisms underlying PARP action during HS gene activation are not clear . It has been suggested that PARylation may be removing proteins , including histones - which are themselves a good PARP substrate - thereby promoting chromatin opening [1] . The accumulation of PARylated proteins at HS loci has recently been proposed to build up a “transcription compartment” which hinders the diffusion of proteins into and out of the compartment , thus favouring factor recycling [4] . In addition to histone displacement and transcription compartment formation at HS genes , recent evidence suggests that PARylation could also act as a signaling scaffold for the recruitment of PAR-sensing factors during DNA damage . In mammals PARylation at DNA damage sites can mediate the recruitment of several ATP-dependent nucleosome remodeling enzymes [5]–[10] . Here we sought to address whether and how nucleosome remodelers may be recruited to PARP activation sites upon environmental stresses other than DNA damage . We have investigated a paradigm of environmental stress , the activation of HS loci in Drosophila and have analyzed the mechanism through which the nucleosome remodeler dMi-2 is recruited to HS genes . Mi-2 ( CHD3/CHD4 ) is a conserved ATP-dependent nucleosome remodeler . In both vertebrates and invertebrates , it is a subunit of Nucleosome Remodeling and Deacetylation ( NuRD ) complexes . NuRD complexes repress cell type specific genes during differentiation [11]–[13] . dMi-2 is also a subunit of the Drosophila-specific Mep-1 complex ( dMec ) which represses neuron-specific genes during differentiation of the peripheral nervous system [12] , [14] . Mi-2 containing complexes lack subunits with sequence-specific DNA binding activity . Two main mechanisms for their recruitment to chromatin have been suggested . First , NuRD complexes contain subunits with methylated DNA binding domains ( MBD ) which direct NuRD to methylated DNA [15] , [16] . This is unlikely to be a major recruitment mechanism for Drosophila Mi-2 complexes , however , given the low and transient levels of DNA methylation in this organism [17] . A second mode of Mi-2 recruitment involves interactions with DNA bound transcription factors [11] , [12] , [14] , [18]–[22] . In addition , SUMOylation of transcription factors can increase their affinity for Mi-2 complexes [21] , [22] . Despite its well established role in repression , dMi-2 localises to actively transcribed chromosome regions suggesting an unexpected potential function of dMi-2 in transcription [23] . Here we sought to establish how dMi-2 is recruited to actively transcribed chromatin and to clarify its role in transcriptional activation using genetic , biochemical and pharmacological assays . We show that dMi-2 rapidly associates with activated HS loci , covering the entire transcribed region of the hsp70 gene . dMi-2 recruitment is not affected when transcriptional elongation is blocked but is abrogated when PARP is inhibited . Indeed , we find that dMi-2 specifically binds PARP's oligomeric product PAR in vitro . Significantly , a dMi-2 mutant unable to bind PAR is not recruited to active HS loci in vivo . We have identified several regions of dMi-2 that bind PAR in vitro . These include the chromodomains and a series of K/R-rich motifs near the N-terminus . Further , dMi-2 depletion or expression of an inactive enzyme greatly decreases transcript levels , suggesting that dMi-2 actively supports efficient HS gene expression . Indeed , dMi-2 associates with nascent hsp70 transcripts in vivo and ablation of dMi-2 function results in inefficient RNA processing . RNA and PAR compete for dMi-2 binding suggesting a two step process of dMi-2 association with HS genes: intial recruitment of dMi-2 is effected by its binding to PAR which is produced prior to the onset of transcription , dMi-2 then switches to interacting with the emerging nascent transcripts . Taken together , our results uncover PAR binding as a novel mechanism for the recruitment of the nucleosome remodeler dMi-2 to targeted sites of PARP activitation upon environmental stress and demonstrate that dMi-2 acts as a co-activator for the full transcriptional activation of HS genes . This study provides the first evidence for an in vivo function of PARylation in promoting the recruitment of a nucleosome remodeler to support the transcription of stress induced genes .
As shown previously , dMi-2 colocalised with active RNA polymerase II ( Pol II ) on polytene chromosomes [23] ( Figure 1A ) . In addition , dMi-2 significantly colocalised with different forms of elongating Pol II ( Ser2P and Ser5P ) and elongation factors ( Spt5 ) . This suggests that dMi-2 may play an unanticipated role in active transcription . Upon HS , dMi-2 associated with the loci 87A and 87C which contain multiple copies of the hsp70 gene ( Figure 1B ) , further strengthening a potential link between dMi-2 and active transcription . Chromatin immunoprecipitation ( ChIP ) analysis of dMi-2 binding to the activated hsp70 gene in Kc cells revealed an enrichment of dMi-2 in the transcribed region ( Figure 2A and 2B ) . dMi-2 association was detected as early as 2 min after HS and progressively increased for 20 min ( Figure S1 ) . We considered three recruitment mechanisms: First , dMi-2 might bind histone modifications enriched in actively transcribed genes , such as H3K4me3 or H3K36me3 . However , we did not find a methylation sensitive interaction of recombinant dMi-2 with histone peptides in pulldown assays ( data not shown ) . Second , dMi-2 might bind and travel with RNA Pol II or elongation factors . This hypothesis predicts that HS-dependent dMi-2 recruitment to the transcribed part of hsp70 is transcription-dependent . To test this hypothesis , we inhibited transcriptional elongation with DRB ( Figure 2C ) . Although this treatment efficiently ablated production of hsp70 transcripts , it did not significantly reduce HS-dependent recruitment of dMi-2 . In addition , we failed to detect robust biochemical interactions of dMi-2 with RNA Pol II or elongation factors in co-immunoprecipitation assays ( data not shown ) . We conclude that the HS-dependent recruitment of dMi-2 to the hsp70 gene can be uncoupled from the transcriptional activity of hsp70 . Third , dMi-2 might be recruited by interaction with PAR , a modification that rapidly accumulates over the hsp70 locus upon HS [3] . We therefore treated Kc cells with the small molecule PARP inhibitor PJ34 ( Figure 2D ) . This led to a significant decrease of global PARylation levels , but did not abrogate hsp70 transcription or nucleosome depletion ( Figure 2D and Figure S2 ) . Nevertheless , dMi-2 recruitment to hsp70 was severely decreased during HS , suggesting that efficient PARylation of the locus is a requirement for stress-dependent enrichment of dMi-2 . To determine whether dMi-2 binds PAR directly , we auto-PARylated PARP1 in vitro and incubated the reaction with immobilised dMi-2 . mH2A1 . 1 which contains a macrodomain known to interact with PAR was used as a positive control in this assay . Western blot analysis revealed that dMi-2 , like mH2A1 . 1 , bound PARylated PARP1 efficiently ( Figure 3A ) . We confirmed that dMi-2 also interacted with radioactively labeled PARylated PARP1 ( Figure S3 ) . To ensure that dMi-2 interacted directly with the PAR polymer , we assayed binding to purified PAR using a dot blot assay ( Figure S4 ) . This verified the apparent direct interaction between dMi-2 and PAR . Next , we sought to define the dMi-2 region required for PAR binding . We tested an array of dMi-2 truncation mutants for their ability to interact with PARylated PARP1 in vitro ( Figure 3B ) . This revealed that the N-terminal region had a high affinity for PAR . Within this part of dMi-2 , both the PHD finger containing region N-terminal of the chromodomains ( aa 1-485 ) and ( to a lesser extent ) the chromodomains ( aa 484-690 ) were capable of binding PAR . To verify these results we also tested binding of dMi-2 mutants to PAR in dot blot assays ( Figure S4 ) . We conclude that dMi-2 possesses at least two PAR-sensing regions that can function independently of each other . To assess the functional importance of dMi-2′s PAR binding activity , we compared recruitment of GFP-dMi-2 fusion proteins to the activated hsp70 loci in transgenic flies ( Figure 3C ) . GFP fused to full length dMi-2 and a GFP-dMi-2 fusion lacking the N-terminal PAR-binding regions were expressed to similar levels in 3rd instar larvae and correctly localised to salivary gland nuclei ( Figure S5 ) . Full length GFP-dMi-2 was enriched at active HS loci , the PAR binding mutant , however , failed to accumulate . This supports the notion that dMi-2 binding to PAR makes an important contribution to the recruitment of this nucleosome remodeler to the stress-activated hsp70 gene . The N-terminal PAR binding region of dMi-2 contains two highly conserved domains , a pair of PHD fingers ( residues 377 to 484 ) and a tandem chromodomain ( residues 488 to 673 ) . We generated GST fusions containing these domains and tested their ability to bind PAR in dot blot assays ( Figure S6 ) . This confirmed that the chromomodomains can bind PAR independently . However , the PHD fingers did not display PAR binding activity . We next sought to better define the PAR binding region near the N-terminus of dMi-2 . The N-terminal 375 residues of dMi-2 are characterised by a high content in charged residues ( 24% D/E , 21% R/K ) . This general feature is conserved between dMi-2 and mammalian CHD4 proteins ( Figure 4A ) . In addition , these proteins share a region with high sequence similarity , the CHDNT domain ( Pfam family PF08073 ) . The function of this domain is not known . A number of diverse PAR binding motifs have recently been identified [24]–[26] . A common feature of these motifs is that they all contain several R/K residues that are interspersed by hydrophobic residues which often play critical roles in mediating PAR binding [24]–[26] . We subjected different dMi-2 fragments to the PAR binding assay , including four K/R-rich fragments ( K/R I to IV in Figure 4A ) . This analysis revealed strong PAR binding activity for three of the four K/R-rich fragments ( K/R I , K/R II and K/R IV; Figure 4B ) . By contrast , K/R-rich fragment II and a fragment encompassing the CHDNT domain failed to interact with PAR . Taken together , our results suggest that dMi-2 contains multiple PAR binding regions in its N-terminus: three are characterised by a high content of basic amino acid residues ( K/R I , K/R III and K/R IV ) and one region containing the tandem chromodomain . PARylation of the hsp70 locus has been proposed to assist in the opening of chromatin structure and to increase access of factors to DNA and nascent hsp70 transcripts [1] . Given that dMi-2 localises to the entire transcribed region and given that PAR exhibits chemical and structural similarity to RNA , we speculated that dMi-2 , once recruited , might interact with nascent hsp70 RNA . We immunoprecipitated dMi-2 from nuclear extracts of heat shocked Kc cells and probed for the co-precipitation of nascent ( unprocessed ) hsp70 and hsp83 RNA ( Figure 5A ) . Indeed , two independent dMi-2 antibodies precipitated these transcripts arguing for a physical , potentially direct interaction . In agreement with this , dMi-2 bound to single-stranded hsp70 RNA in an electrophoretic mobility shift assay in vitro ( Figure 5B ) . Next , we performed competition assays to gain insight into the relative affinities of dMi-2 for DNA , RNA and PAR and to determine if dMi-2 can bind to several types of nucleic acid simultaneously or if binding is competitive . First , we tested dMi-2 binding to RNA and DNA , respectively , in the presence of increasing amounts of PAR in electrophoretic mobility shift assays ( mass ratios 1∶1 , 1∶2 and 1∶4; Figure 5C ) . In this assay , PAR was able to compete with RNA and DNA for dMi-2 binding . However , whereas dMi-2 no longer bound to DNA at a DNA:PAR mass ratio of 1∶2 , residual dMi-2/RNA complexes were still detectable at an RNA:PAR mass ratio of 1∶4 . This suggests that dMi-2 has a higher binding affinity for RNA than for DNA . We confirmed this hypothesis by incubating dMi-2 with different mass ratios of RNA and DNA ( Figure 5C ) : At a DNA:RNA mass ratio of 1∶1 , dMi-2/RNA complexes formed readily but dMi-2/DNA complexes were not detected . dMi-2/RNA complexes formed even at DNA:RNA mass ratios of 4∶1 . To test if RNA or DNA can compete with dMi-2 for binding to the branched PAR polymer we performed dot blot assays ( Figure S7 ) . RNA competed with immobilised PAR for binding to dMi-2 whereas DNA failed to do so . Taken together , our results suggest that dMi-2 has a higher affinity for binding to RNA and PAR than for binding to DNA . In addition , dMi-2 appears to bind RNA and PAR in a mutually exclusive manner . These results are consistent with the hypothesis that dMi-2 is first recruited to HS loci by interaction with PAR ( which is produced prior to and independent of transcription ) and , once RNA synthesis has been strongly activated , switches to binding the nascent RNA . We hypothesised that dMi-2 binding to nascent RNA might influence hsp70 transcription or processing . We used transgenic fly lines to deplete dMi-2 by RNAi ( Figure 6A ) . We subjected transgenic larvae to HS and determined the HS gene transcription by RT-QPCR . Although hsp70 , hsp26 and hsp83 genes were all activated by HS , transcript levels were severely reduced in dMi-2 depleted larvae compared to controls . Importantly , transcription of a housekeeping gene was not significantly affected . We conclude that dMi-2 makes a positive contribution to transcription and is essential for full HS gene activation in larvae . We next determined whether dMi-2 enzymatic activity was required to activate HS genes . We generated transgenic fly lines overexpressing wild type dMi-2 or a dMi-2 mutant carrying a point mutation in the ATP binding site ( K761R ) predicted to prevent ATP binding ( Figure 6B ) . Indeed , dMi-2K761R could not hydrolyse ATP in vitro ( Figure S8 ) . We subjected 3rd instar larvae to HS and determined effects on HS gene transcription as before . Whereas overexpression of wild type dMi-2 had little effect , levels of HS gene transcripts were greatly reduced in larvae overexpressing the enzymatically inactive dMi-2 ( Figure 6B ) . We conclude that the ATPase activity of dMi-2 is essential for full HS gene activation . Next , we sought to assess whether dMi-2 influences RNA processing . Because dMi-2 depletion and expression of enzymatically inactive dMi-2 resulted in an overall reduction of hsp70 transcript levels we determined the ratio of 3′ unprocessed to total hsp70 RNA as a measure of RNA processing efficiency . We reasoned that a mere reduction in hsp70 activation ( e . g . a reduction in the number of initiation events per time ) would not change the ratio of unprocessed to total hsp70 RNA . By contrast , processing defects might give rise to a higher relative proportion of unprocessed RNA and , therefore , to a higher unprocessed:total RNA ratio . Depletion of dMi-2 increased the relative proportion of unprocessed hsp70 RNA ( Figure 7A ) . An even more striking effect was observed in larvae overexpressing inactive dMi-2 , whereas overexpression of wild type dMi-2 was of little consequence . Similar effects on 3′ RNA processing were observed with the hsp83 gene ( data not shown ) . Hsp83 is one of the few HS genes possessing an intron . Therefore , we determined the ratio of unspliced to total hsp83 transcripts in transgenic larvae ( Figure 7B ) . Again , we observed a significant increase in the relative proportion of unspliced RNA in dMi-2-depleted larvae and in larvae overexpressing inactive enzyme . This suggests that dMi-2 activity is required for the efficient processing of HS gene transcripts and that dMi-2 affects both RNA 3′ end cleavage and splicing .
Mi-2 is strongly linked to transcriptional repression in both vertebrate and invertebrate organisms . Within NuRD and dMec complexes it contributes to the repression of cell type-specific genes [11] , [14] , [19]–[21] . Therefore , the widespread colocalisation of dMi-2 with active Pol II and elongation factors at many chromosomal sites is surprising and suggests that dMi-2 might play an unappreciated role during active transcription , at least ( or specifically ) during environmental stresses such as HS . Indeed , dMi-2 is recruited to HS genes within minutes of HS . This property is not shared by other chromatin remodelers: Brahma ( BRM ) is not enriched at HS puffs and HS gene activation is independent of BRM function ( [27] and data not shown ) . Moreover , although imitation switch ( ISWI ) containing complexes are important for HS gene transcription , ISWI does not accumulate to high levels at active HS loci ( [28] , [29] and data not shown ) . Recruitment to HS puffs has previously been reported for Drosophila CHD1 [30] . Thus , accumulation at active HS genes is shared by at least two members of the CHD family of nucleosome remodelers but not by SWI/SNF and ISWI proteins . Depletion of dMi-2 or a reduction of dMi-2 recruitment does not significantly perturb hsp70 transcription in Kc cells and , therefore , dMi-2 is dispensable for HS gene activation in this system ( Figure 2D and data not shown ) . By contrast , depletion of dMi-2 in larvae strongly decreases hsp70 , hsp26 and hsp83 activation ( Figure 6A ) . It is possible , that the RNAi-mediated depletion of dMi-2 is more efficient in transgenic flies compared to cell lines . In addition , it is believed that several factors contributing to HS gene activation are highly abundant or redundant in Kc cells but more limiting in other contexts . Accordingly , FACT and Spt6 are required for a HS gene activation in flies but are not essential in Kc cells [31] , [32] . The strong decrease of HS gene activation in dMi-2 RNAi larvae indicates a positive contribution of dMi-2 to transcription in vivo . Overexpression of inactive dMi-2 also results in reduced HS gene transcription implying that its enzymatic activity is critical ( Figure 6B ) . It is presently unclear whether this reflects a requirement for dMi-2 catalysed nucleosome remodeling or whether its activity is directed towards different substrates . While dMi-2 could indirectly influence transcription by remodeling nucleosomes within the transcribed part of hsp70 , its physical association with nascent HS gene transcripts argues for a more direct effect . Indeed , dMi-2 is not only required for high HS gene mRNA levels , but also affects the efficiency of co-transcriptional 3′ end formation and splicing . A role of chromatin remodelers in splicing has been suggested before: Both CHD1 and BRG1 bind components of the splicing apparatus [33] , [34] . CHD1 associates with Pol II and binds nucleosomes containing H3K4me3 , which are enriched near the 5′ end of active genes [34] , [35] . BRG1 is present at the coding region of genes and influences splice site choice [33] , [36] . It has been proposed that CHD1 and BRG1 physically recruit splicing factors but it is unclear if their ATPase activities play a role . Indeed , inactive BRG1 retains the ability to affect exon choice [33] , [34] . Inefficient processing of the hsp70 and hsp83 transcripts is not only observed in larvae expressing reduced levels of dMi-2 . Importantly , even stronger processing defects are generated by overexpression of inactive dMi-2 ( Figure 7 ) . This strongly suggests , for the first time , that the catalytic activity of a chromatin remodeler is required for correct co-transcriptional RNA processing . It remains to be determined whether dMi-2 nucleosome remodeling activity influences RNA processing indirectly , e . g . by altering Pol II elongation rates , or whether it has a more direct role . A series of complementary results support our hypothesis that dMi-2 interacts with PAR polymers that are rapidly synthesized at activated HS loci . First , the broad distribution of dMi-2 over the entire transcribed region correlates with the distribution of PAR polymer [3] . Second , pharmacological inhibition of PARP greatly decreases dMi-2 binding to activated hsp70 . Third , dMi-2 directly binds PAR polymers in vitro . Fourth , an dMi-2 mutant unable to bind PAR also fails to localise to active HS loci . As discussed above , dMi-2 physically associates with nascent HS gene transcripts and binds RNA in vitro . While this interaction is potentially important for the efficiency of transcription and processing , it likely plays a minor role in dMi-2 targeting . Accordingly , inhibition of transcriptional elongation has no significant effect on dMi-2 recruitment ( Figure 2C ) . It is important to note , that while our results argue for an important role of PAR binding in the recruitment of dMi-2 to HS loci , we cannot exclude that protein-protein interactions with histone or non-histone proteins also play a role . Our analysis indicates that dMi-2 harbours several PAR binding motifs in its N-terminal region . Polo and colleagues have recently demonstrated that human CHD4 is recruited to double stranded DNA breaks in a PARP-dependent manner [10] . They have mapped PAR binding activity to the region N-terminal of the ATPase domain of CHD4 . This agrees well with our data and suggests that the PAR binding function of CHD4/dMi-2 has been conserved in evolution . Two structural protein modules directly interact with PAR , the macrodomain and the PBZ domain; however , these domains are not present in dMi-2 [5] , [7] , [37] . In addition , several shorter PAR binding motifs have been identified [5] , [26] . These motifs bear little sequence similarity but share the presence of several K/R residues which are interspersed by hydrophobic residues . Our results have uncovered three K/R-rich regions with PAR binding activity near the N-terminus of dMi-2 . Two of these three K/R-rich regions ( K/R III and K/R IV ) consist of interspersed basic and hydrophobic residues and are therefore reminiscent of the previously described PAR binding motifs [24] , [25] , the third ( K/R I ) lacks hydrophobic residues completely . None of the three K/R regions matches the consensus PAR binding motifs . It is possible that a consensus motif should generally be chosen less stringently and that a high content of K and R-residues in these regions is sufficient to provide PAR binding activity in vitro . Further characterisation of these regions will be required to resolve this issue . In addition to the K/R regions , the tandem chromodomains of dMi-2 bind PAR in vitro . We have previously shown that the chromodomains are required for interacting with nucleosomal DNA in vitro [38] . Our new data suggests that these domains can interact with different nucleic acids . Several potential molecular functions of PARylation at HS genes have been suggested . First , PARP activity is required for the rapid loss of nucleosomes at hsp70 within the first two minutes after HS [1] . It has been suggested that PARylation of histones aids rapid nucleosome disassembly [1] . Second , at later stages of the HS response ( 20–60 minutes after HS ) , PARP activity is required to establish a compartment which restricts the diffusion of factors such as Pol II and Spt6 and promotes efficient factor recycling [4] . Our results suggest that PARylation carries out a third task , namely , to recruit factors via their direct interaction with PAR . The earliest time point when we can detect dMi-2 binding to hsp70 is between 2 and 5 minutes after HS . This places dMi-2 recruitment between the early PARP-dependent nucleosome removal ( 0–2 minutes after HS ) and effects of the transcription compartment ( 20–60 minutes after HS ) . The ability of dMi-2 to bind both PAR and RNA and the finding that RNA can compete for PAR binding to dMi-2 is consistent with the hypothesis that dMi-2 association with active HS genes is a two step process ( Figure 8 ) . We propose that dMi-2 is initially recruited via interaction with PAR polymers . Synthesis of these starts prior to the onset of hsp70 transcription [1] . This results in a rapid local increase of the dMi-2 concentration . In the second step , when hsp70 transcripts are produced by elongating RNA polymerase II at high rates , dMi-2 can switch from binding PAR to interacting with nascent transcripts . Severe cellular stresses , such as DNA strand breaks and acute HS , must be dealt with quickly and efficiently . In both cases , a multitude of factors are rapidly recruited to orchestrate the repair of DNA and the massive transcriptional activation of HS genes , respectively . We postulate that rapid synthesis of PAR polymers at both DNA damage sites and HS genes affords an efficient mechanism to recruit chromatin remodelers and other factors . It has recently been shown that PARylation of DNA breaks is instrumental in recruiting chromatin remodelers , including mammalian dMi-2 homologs , to damaged sites [8] , [9] , [10] , [39] , [40] , [41] . Here , we show that dMi-2′s recruitment to activated HS genes requires PARP activity and that dMi-2 binds PAR directly . The high local concentration of PAR polymers at DNA breaks and HS genes might exploit the general affinity of dMi-2 for nucleic acids . Indeed , dMi-2 binds both DNA and RNA as well as PAR in vitro ( [38] and this study ) . In this manner , PAR polymers might act as a scaffold to redirect dMi-2 to chromatin regions where high levels of dMi-2 activity are required , thus acting as a stress-dependent , transient affinity site for chromatin remodeling and possibly RNA processing activities ( Figure 8 ) . Our results highlight a signaling and scaffolding function for PARP activity during transient environmental stresses other than DNA damage , suggesting that PARylation carries out important modulatory functions in the stress-dependent reprogramming of nuclear activities .
Kc cell HS treatment and ChIP was performed as decribed using dMi-2C antibody [14] , [42] . For primer sequences see Dataset S2 . Triplicate mean values of percentage input DNA and standard deviations are plotted . dMi-2 knockdown by RNAi was described previously [23] . For RNAi primer sequences see Dataset S4 . Kc cells were treated with 125 µM DRB ( Sigma ) to inhibit transcription and with 5 µM PJ34 ( Alexis ) to inhibit PARP activity for 20 min before subjecting cells to HS . Chromosomes were prepared as before [23] . The following antibodies were used: Primary antibodies: anti-dMi-2N ( rabbit ) 1∶200 , anti-pol II ( mouse H5 , Covance ) 1∶50 , anti-GFP ( rabbit , Abcam ) 1∶50 , anti-Spt5 ( guinea pig ) 1∶200 . Secondary antibodies: Alexa Fluor 488 goat anti-rabbit 1∶200 , Alexa Fluor 546 goat anti-mouse or anti-guinea pig 1∶200 ( Invitrogen ) . Analysis was performed with a Zeiss fluorescence microscope ( Axioplan ) . For baculovirus production , dMi-2 mutants ( aa 1-691 ) and ( aa 1-485 ) were generated by PCR using appropriate sets of primers and cloned with NotI and XbaI into the pVL1392 transfer vector . Vectors for dMi-2 WT and other mutants were described previously [38] . dMi-2 GST-fusion fragments were generated by PCR using appropriate sets of primers and cloned with NotI and SalI into the pGEX4T1 vector . All constructs were verified by DNA sequencing . For primer sequences see Dataset S1 . Protein extracts from 3rd instar larvae were prepared as described in [14] . Purification of recombinant dMi-2 and ATPase assays are described [23] . Recombinant mH2A1 . 1 was purified as in [43] . GST-fusion proteins were expressed in E . coli BL21 ( DE3 ) and purified with Glutathione Sepharose 4 Fast flow ( GE Healthcare ) according to the manufacturer's instructions . A typical DNA or RNA binding reaction ( 25 µl ) was performed in the presence of 0 . 2 µg of dMi-2F and 80 ng of nucleic acid ( DNA or ssRNA ) in 40 mM KCl , 20 mM Tris pH 7 . 6 , 1 . 5 mM MgCl2 , 0 . 5 mM EGTA , 10% glycerol , BSA ( 200 ng/µl ) , 1 mM DTT ( supplemented with 0 . 4 units of RNAsin ) . For competition assays , samples were preincubated for 15 min at 26°C before the different amounts of competitor ( PAR or DNA or RNA ) were added . Reactions were further incubated at 26°C for 75 min . Products were analyzed on 6% native PAA gel and visualized with ethidium bromide ( EtBr ) staining . ssRNA was synthesized by in vitro transcription using a fragment of hsp70 DNA as a template . This template ( also used for the DNA bandshift assays ) was produced by PCR amplification of cDNA derived from heat shocked Kc cells using the following primers: T7-hsp70_f - TAATACGACTCACTATAGGGCCTACGGACTGGACAAGAAC and hsp70_r -AGGGTTGGAGCGCAGATCCTTCTTGTAC . Total RNA was isolated from 3rd instar larvae using PeqGold total RNA Kit ( PeqLab ) . 10-12 larvae from each cross were pestled in 400 µl of lysis buffer before loading the material on the column . 1 µg of RNA was reverse transcribed by incubation with 0 . 3 µg of random primers ( Invitrogene ) and 100 U of M-MLV reverse transcriptase ( Invitrogen ) . cDNA synthesis was performed according to the manufacturer's protocol . cDNA was analyzed by QPCR using Absolute SybrGreen Mix ( Thermo Fisher ) and the Mx3000P real-time detection system ( Agilent ) . For primer sequences used in RT-QPCR see Dataset S3 . All amplifications were performed in triplicate using 0 . 6 µl of cDNA per reaction . Triplicate mean values were calculated according to the ΔΔCt quantification method using rp49 gene transcription as reference for normalization . Relative mRNA levels in uninduced control larvae were set to 1 and other values were expressed relative to this . The RT-QPCR results were reproduced several times using independent fly crosses and representative data sets are shown . RNA immunoprecipitation was performed as described previously [44] . Briefly , Kc cells were crosslinked as for ChIP . Cells were washed once with PBS buffer and lysed on ice for 15 min in FA buffer ( 50 mM Hepes- KOH , pH 7 , 6 , 140 mM NaCl , 1% Triton X-100 , 0 , 1% sodium deoxycholate , proteinase inhibitors , RNAsin ( 100 u/ml of buffer ) ) . Cells were sonicated , spun down and chromatin was digested with DNAse I . The chromatin containing solution was adjusted to 25 mM MgCl2 and 5 mM CaCl2 . 1 ul of DNAse I ( Qiagen ) was added and reactions were incubated for 10 min at room temperature and then stopped with 20 mM EDTA . Chromatin was spun down for 10 min ( 13000 rpm ) at 4°C . 300 µl of chromatin was used for IP . 2 µl of anti-dMi-2 ( C ) and anti-dMi-2 ( N ) antibodies , 2 µl rabbit IgG , 2 µl rabbit preimmuneserum and beads only ( control ) and were used for IP . Samples were incubated over night at 4°C . RNA-protein complexes were precipitated with 30 ul of 50% protein G Sepharose beads for 2 hr at 4°C . IPs were washed 5 times in FA buffer , twice with TE buffer and eluted twice with 100 µl of elution buffer ( 100 mm Tris HCl , pH 8 , 0 , 10 mM EDTA , 1% SDS ) – once at room temperature and once at 65°C . All buffers were supplemented with RNAse inhibitor ( RNAsin , Promega ) . All samples were digested with proteinase K for 1 hr at 42°C and decrosslinking was performed at 65°C over night . Immunoprecipitated RNA was purified using PeqGold total RNA Kit ( PeqLab ) , digested with DNAse on the column and eluted with 30 µl of RNAse free dH20 . cDNA was synthesized with 10 µl of eluted RNA and 2 µl of input with random hexamers and analysed by Q-PCR with appropriate primer pairs . Non-radioactive PAR synthesis was performed according to the standard protocol [45] . Briefly , PARP reactions were set up in a final volume of 0 . 5 ml: 2 µg recombinant Parp1 , 100 mM Tris-HCl , pH 7 . 5 , 50 mM NaCl , 10 mM MgCl2 , 2 µg/ml DNA oligonucleotides , 1 mM NAD+ , 1 mM DTT . Reactions were incubated at 37°C for 25 min . PJ34 inhibitor was added before the reaction to control samples to a final concentration of 5 µM . All reactions were stopped with PJ34 . Control beads and beads with bound proteins ( dMi-2 , dMi-2 mutants and mH2A1 . 1 or GST fusions ) were equilibrated in binding buffer ( 50 mM Tris , pH 8 . 0 , 0 , 2 mM DTT , 4 mM MgCl2 , 200 mM NaCl , 0 , 1% NP-40 ) . 10 µl of bead-bound proteins were used for each pulldown . Pulldowns were performed with the whole PARP reaction ( 0 . 5 ml ) and 500 µl of binding buffer ( for baculovirus expressed proteins ) or in 250 µl of PARP reaction and 250 µl of binding buffer ( for GST fusions ) for 1 hr at 4°C . After extensive washing ( 5 times ) , beads were boiled in SDS loading buffer , loaded on 4–12% gradient SDS-Page gels and analysed by Western blot . For Western blot anti-PAR ( 10H , 1∶500 ) antibodies were used . mH2A1 . 1 was used as a positive control . Radioactive pulldown reactions were prepared in the same way , in the presence of 2 µl of radioactive NAD+ ( PerkinElmer ) . After washing , samples were resuspended in 30 µl of SDS-loading buffer and 10 µl was resolved by SDS PAGE . The gel was dried and and subjected to autoradiography . Generation of transgenic fly strains and the PAR dot blot assay are described in Text S1 . | Cells respond to elevated temperatures with the rapid activation of heat shock genes to ensure cellular survival . Heat shock gene activation involves the synthesis of poly-[ADP-ribose] ( PAR ) at heat shock loci , the opening of chromatin structure , and the coordinated recruitment of transcription factors and chromatin regulators RNA polymerase II and components of the RNA processing machinery . The molecular roles of PAR and and ATP-dependent chromatin remodelers in heat shock gene activation are not clear . We show here that the chromatin remodeler dMi-2 is recruited to Drosophila heat shock genes in a PAR–dependent manner . We provide evidence that recruitment involves direct binding of dMi-2 to PAR polymers and identify novel PAR sensing regions in the dMi-2 protein , including the chromodomains and a series of motifs rich in K and R residues . Upon HS gene activation , dMi-2 associates with nascent transcripts . In addition , we find that dMi-2 and its catalytic activity are important for heat shock gene activation and co-transcriptional RNA processing efficiency . Our study uncovers a novel role of PAR during heat shock gene activation and establishes an unanticipated link between chromatin remodeler activity and RNA processing . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biochemistry",
"developmental",
"biology",
"model",
"organisms",
"genetics",
"biology",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] | 2011 | Stress-Induced PARP Activation Mediates Recruitment of Drosophila Mi-2 to Promote Heat Shock Gene Expression |
Genome-wide association studies ( GWAS ) have identified loci reproducibly associated with pulmonary diseases; however , the molecular mechanism underlying these associations are largely unknown . The objectives of this study were to discover genetic variants affecting gene expression in human lung tissue , to refine susceptibility loci for asthma identified in GWAS studies , and to use the genetics of gene expression and network analyses to find key molecular drivers of asthma . We performed a genome-wide search for expression quantitative trait loci ( eQTL ) in 1 , 111 human lung samples . The lung eQTL dataset was then used to inform asthma genetic studies reported in the literature . The top ranked lung eQTLs were integrated with the GWAS on asthma reported by the GABRIEL consortium to generate a Bayesian gene expression network for discovery of novel molecular pathways underpinning asthma . We detected 17 , 178 cis- and 593 trans- lung eQTLs , which can be used to explore the functional consequences of loci associated with lung diseases and traits . Some strong eQTLs are also asthma susceptibility loci . For example , rs3859192 on chr17q21 is robustly associated with the mRNA levels of GSDMA ( P = 3 . 55×10−151 ) . The genetic-gene expression network identified the SOCS3 pathway as one of the key drivers of asthma . The eQTLs and gene networks identified in this study are powerful tools for elucidating the causal mechanisms underlying pulmonary disease . This data resource offers much-needed support to pinpoint the causal genes and characterize the molecular function of gene variants associated with lung diseases .
Recent genome-wide association studies ( GWAS ) have identified loci that harbor susceptibility genes for asthma and other pulmonary conditions [1]–[16] . Many of the genes at these loci have unknown function and have not previously been considered biologically plausible candidates for disease pathogenesis . Extensive linkage disequilibrium ( LD ) within these loci makes it difficult to identify the actual susceptibility genes , let alone which genetic variants are responsible for altered expression or function of their protein products . Moreover , the associated polymorphisms can only explain a relatively small proportion of the variability of the phenotype in the population [9] and of its heritability [17] . Integrative genomics is a promising new approach to identify true causal genes and variants . By using gene expression as a phenotype and examining how DNA polymorphisms contribute to both gene expression ( expression quantitative trait loci – eQTLs ) and disease phenotype , true causal relationships can be discovered [18]–[20] . In the present study we performed genome-wide genotyping and lung-specific gene expression on a large dataset of lung tissue ( 1 , 111 human subjects ) to explore effects of genetic variation on gene expression and their joint relationship to asthma . Although the lung tissue came from a heterogeneous group of subjects , the discovery of lung tissue eSNPs could elucidate the causal molecular pathways in a variety of pulmonary disorders .
Demographic and clinical data for the 1 , 111 patients who passed clinical , genotyping and gene expression quality control assessments are summarized in Table 1 according to the three sites of recruitment , Laval University , University of British Columbia , and University of Groningen ( henceforth referred to as Laval , UBC , and Groningen , respectively ) . The majority of the subjects were smokers or former-smokers . There were some differences in the clinical characteristics across the sites including age , lung function and smoking status . To account for this heterogeneity between sites , we performed a meta-analysis rather than a pooled analysis for eQTL discovery . The overall analysis workflow is illustrated in Figure 1 . We identified cis- and trans-acting eQTLs using established methods [19] . For simplicity we assumed that each transcript could have at most one cis-eQTL . We considered all association signals from SNPs within 1 Mb up and downstream , of the transcription probeset as a single cis-eQTL . Trans-eQTLs were defined as association signals from SNPs located greater than 1 Mb from the probeset . The eSNP was identified as the SNP that was most significantly associated ( lowest P value ) with the expression trait . A summary of the eQTLs identified at 10% false discovery rate ( FDR ) in the three cohorts as well as the meta-analysis are reported in Table 2 . There were variations in the numbers of eQTLs across the cohorts . For example , we identified 10 , 630 , 5 , 655 and 7 , 953 cis-eQTLs in the Laval , Groningen and UBC cohorts , respectively . There were also variations in the number of detectable trans-eQTLs ( Table 2 ) . The P values reported for the trans-eQTLs are lower than those for cis-eQTLs , owing to the higher statistical threshold required to achieve statistical significance . In general , a single SNP ( eQTL peak ) explained between 10 and 20% of the transcript's expression variance for cis- and trans-eQTLs , respectively ( Table S1 ) . However 6 . 8% of the cis acting SNPs explained more than 30% of the variance in the transcript and 15 . 3% of the trans acting SNPs explained more than 30% of a transcript's variance ( Figure S1 ) . Further , we investigated the consistency of the eQTLs derived from the cohorts . Rather than simply comparing the existence of eQTLs , we adopted replication criteria similar to GWAS . A successful replication was defined as an eQTL in which the relationship between the same SNP and gene expression ( with the same sign ) was observed in more than one cohort at a P of ≤1×10−4 . Using this criterion , 54 . 5% of the Laval cis-eQTLs were replicated in at least one other cohort and 79 . 5% of the Laval trans-eQTLs were replicated in another cohort . With respect to the Groningen eQTLs , the replication rates of the cis- and trans-eQTLs were 72 . 3% and 86 . 0% , respectively . The replication rates of the UBC cis- and trans-eQTLs were 65 . 4% and 90 . 5% , respectively . The eQTLs with higher R2values ( variance of gene expression level explained ) were more likely to be replicated than those with lower R2 values . There was a higher replication rate with trans-eQTLs than with cis-eQTLs . The meta-analysis revealed many more eQTLs owing to the increased sample size ( Table 2 ) . We identified 17 , 049 cis-eQTLs and 534 trans-eQTL and the details of all eQTLs are listed in Table S2 . 68 . 7% of these cis-eQTLs and 31 . 7% of trans-eQTLs were identified in at least one cohort , suggesting that the meta-analysis detected many additional eQTLs . The ‘non-combinability’ among cohorts is quantified in terms of Cochran's Q and the τ statistics ( Table S2 ) . Cochran's Q , was calculated as the weighted sum of squared differences among individual study effects and the pooled effect across studies , with the weights proportional to the inverse of variance of the eQTL effect . Q follows a chi-square statistic with k-1 degrees of freedom ( k is the number of studies ) , and by these means , we calibrated the p value ( termed Q . pvalue ) . We detected modest degrees of heterogeneity among the three cohorts . 16% of the QTLs had Q . pvalue<0 . 05 and 2% of the eQTLs showed Q . pvalue<0 . 001 . We also present the τ statistic , which is the moment-based estimate of the between-study variance and is not dependent on the number of studies . Both the fixed effect and random effect meta-analysis results are presented in Table S2 , and we used the fix effect results in the downstream analysis . We previously showed that the number of eQTLs detectable at a certain FDR was related directly to the sample size ( i . e . statistical power ) [21] . The meta-analysis yielded a greater relative increase in the number of trans activating eQTLs than of cis-eQTLs , suggesting that sample size is particularly important for detecting trans-eQTLs . Among the 51 , 627 non-control probesets on our array , 33 . 0% showed a cis-eQTL . 74 . 0% , 64 . 4% , 56 . 4% of the 51 , 627 probesets were “present” in at least 20% , 50% , and 80% of the tissue samples . Using these three cutoffs , 40 . 2% , 42 . 5% , and 43 . 4% of the “present” transcripts were found to be controlled by cis-eSNPs , respectively . Accordingly we believe that we have identified the majority of the strong lung cis-eQTLs . As reported previously [22] , [23] , eQTLs were often found with the expression of more than one gene underlying a GWAS signal ( Table 3 ) . For example , the rs7216389-T allele on chromosome 17q was reported to increase asthma risk with an odds ratio ( OR ) of 1 . 45 [10] . This SNP was significantly associated with the expression levels of four genes , ORMDL3 , GSDMA , GSDMB , and CRKRS ( Figure 2a ) . Although ORMDL3 was originally suggested [10] to be the gene mediating rs7216389-T's association with asthma , this gene demonstrated the weakest eQTL signal of the four genes in our dataset . Rs7216389-T was positively associated with expression levels of ORMDL3 ( P = 1×10−7 ) , CRKRS ( P = 1 . 76×10−9 ) and GSDMB ( P = 4×10−15 ) , consistent with the results of Moffat et al . [10] However , this same allele was inversely related to the level of expression of GSDMA ( P = 8 . 78×10−25 ) ( Figure 2b ) . The strongest eSNP in the 17q asthma susceptibility locus was rs3859192 located in intron 6 of the GSDMA gene governing the expression levels of this gene ( P = 3 . 55×10−151 ) ( Figure 2b ) . To determine the cellular source and relative expression of GSDMB and GSDMA in human lung we performed real-time PCR , and Western blots on primary normal airway epithelial cells and immunohistochemistry on formalin-fixed normal lung tissue ( see Text S1 for details ) . Figure 3 shows abundant mRNA and protein expression of GSDMA but little GSDMB from primary human lung epithelial cells ( Figure 3a and 3b respectively ) . Figure 3c shows that GSDMA is expressed in both apical and basal airway epithelial cells in the conducting airways . While most GWAS publications only report the top signals , the GABRIEL study [9] released the asthma association results on all SNPs investigated , allowing an in-depth analysis . For this analysis , we did not limit the list of eSNPs to the most significant SNP for each probeset; all cis eSNPs which passed the 10% FDR for any probe-set were considered . Using this strategy 60 , 530 of the 567 , 589 GABRIEL SNPs were eSNPs in lung tissue . These 60 , 530 eSNPs were enriched for significant association with asthma in the GABRIEL study ( Figure 4 ) . This is consistent with previous studies showing that SNPs associated with complex traits are more likely to be eQTLs [24] . One of the drawbacks of GWAS is the reliance on a large number of statistical tests , which puts the threshold for significance at an extremely low level thereby increasing the chance of missing real associations . Given that asthma is a pulmonary disorder it is reasonable to assume that important molecular drivers are expressed in lung tissue . Therefore , instead of filtering primarily by P values to identify loci/genes that explain asthma , we filtered the loci in the GABRIEL [9] dataset by their status as a cis acting eSNP . Specifically , all SNPs from the GABRIEL study associated with asthma with P<0 . 01 were translated into genes via our lung eSNP list . A total of 7 , 613 SNPs were linked to the expression of 739 unique genes . Using Bayesian networks constructed on gene expression data we attempted to find the molecular underpinnings of asthma . For this discussion we define a Bayesian network as a probabilistic graphical model that represents a set of random variables ( gene expression in this case ) and their conditional dependencies ( edges ) via a directed acyclic graph ( see Text S1 ) . The workflow is illustrated in Figure S2 . We first identified each gene from the GABRIEL GWAS dataset and collected all genes within 3 edges of that gene on the Bayesian Networks and found the largest coherent sub-network ( i . e . the largest network with an observable path between every gene in the network ) that contained the highest proportion of GABRIEL genes ( Figure S3 ) . We choose to use 3 edges so that we had enough genes to lead to a reasonable biological annotation of the genes in the network , i . e . limiting to 2 edges is too restrictive and results in a small number of relevant genes; on the other hand , using 4 edges leads to broad gene sets and makes it hard to pinpoint pathways . From this sub-network , we performed Key Driver Analyses ( KDA ) to capture key regulators of asthma [25] , [26] . KDA is a method that captures the “hub nodes” conditioning on the direction that one gene influences over other genes and isolating those genes with the strongest influence over the entire network . KDA determines the regulatory components in a directed network for a particular set of genes ( i . e . the genes whose expression levels were control by GABRIEL GWAS SNPs ) . With this analysis , we identified well documented asthma candidate genes [27] that drive genes discovered in GABRIEL . These genes are denoted as canonical asthma genes . Figure 5 shows the top 6 key driver genes ( yellow nodes ) that control the asthma canonical genes ( blue nodes ) as well as other genes that are one edge away from the key driver genes ( grey nodes ) . The network when annotated with GeneGo intuitively is described as “Immune Response” ( the number of overlap genes is 25 , from a 1 , 346 genes set , with the background being 16 , 606 , P = 3 . 53×10−20 ) . The KDA data in Table S3 shows the six genes that remain significant after adjustments for multiple testing . Networks provide a means for data integration . Above we used the networks to define a central node of the network that drives asthma by integrating lung Bayesian networks , asthma GWAS and literature ( canonical genes ) to arrive at set of genes that drive the asthma genes in the lung Bayesian network . Similarly we can use the Bayesian networks to rank competing genetic hypotheses around the molecular basis of asthma . It has been shown that the neighborhood milieu of proteins in a network can predict the probable function of a protein for which no function is known [28] . Following the same logic we assessed the sub-network surrounding each candidate causal gene for asthma . The gene with the highest connection to other asthma genes is most likely a causal asthma gene . Chromosomes 2q12 and 17q21 both have strong GWAS hits for asthma . However , these loci harbor a number of genes that could potentially explain these hits10 , 11 . Following the workflow shown in Figure S4 [27] , we drew inference on causal genes underlying asthma GWAS signals on these two chromosomes ( Table 4 ) . Clearly IL1RL1's subnetwork is enriched for canonical asthma genes ( P = 1 . 8×10−07 ) making it the most likely gene driving the asthma association on 2q12 . The result for 17q21 is less clear , but still points away from ORMDL3 , the gene first suggested as causal by Moffatt et al . [10] . The sub-network around GSDMA is significantly enriched for canonical genes ( P = 0 . 005 ) suggesting that GSDMA is the most likely driver behind the asthma association on 17q21 .
One of the primary goals of genetic studies is to identify genes and pathways which contribute to susceptibility to human diseases and towards this goal many GWA studies have discovered SNPs in novel genes [1]–[16] . However , the mechanism by which these SNPs lead to disease susceptibility cannot be directly inferred from GWAS . The discovery of eQTLs has been shown to be a powerful tool in addressing this gap [18]–[20] . Although eQTL analysis of readily available tissue types such as peripheral blood leukocytes and lymphoblastoid cell lines has contributed to the understanding of how genes modulate risk of respiratory diseases , the discovery of lung-specific eQTL is probably more revealing in understanding the pathogenesis of respiratory disease . In this paper , we determined genetic variation and interrogated gene expression in a large collection ( n = 1 , 111 ) of human lung samples to systematically characterize the genetic architecture of gene expression in this tissue . Owing to the large sample size , we discovered over 17 , 000 eQTLs . These eQTLs are gene variants that directly or indirectly govern gene expression in the lung and the data set represents a unique resource which we have made available for the use of lung researchers . To illustrate the power of the resource we merged these data with SNPs known to be strongly associated with asthma and were able to identify the most likely causal gene variants . Chromosome 17q21 is the most consistent locus associated with asthma [9] , [10] , [29]–[36] . In the original GWAS study [10] , the SNPs associated with asthma in the 17q21 susceptibility locus were also associated with transcript levels of ORMDL3 in lymphoblastoid cell lines , suggesting that ORDML3 was the causal gene . However , more refined analysis of the same samples , exploiting data generated by the 1000 genomes project , suggested that GSDMB , in close proximity to ORMDL3 , could be the causal gene [9] . Another eQTL study performed in white blood cell RNA samples suggested that many SNPs in the 17q21 regions are associated with transcript levels of both ORMDL3 and GSDMB [32] . Verlaan et al . [37] showed that SNPs in the region demonstrate domain-wide cis-regulatory effects suggesting long-range chromatin interactions and they found allele-specific differences in nucleosome distribution and binding of the insulator protein CTCF . A recent study also showed that asthma risk alleles on 17q21 were associated with increased ORMDL3 and GSDMA gene expression and elevated IL-17 secretion in cord blood mononuclear cells [38] . To refine the causal gene/variant , we have interrogated our eQTL results within this locus . Interestingly , the SNPs most strongly associated with the expression of ORMDL3 are located in the promoter ( rs4794820 ) and intron 1 ( rs12603332 ) of the ORMDL3 gene , but the P values are not convincing ( most significant SNP rs4794820 P = 1 . 0×10−7 ) . In addition the SNP rs4794820 is more strongly associated with mRNA expression levels of GSDMB ( P = 9 . 3×10−12 ) and GSDMA ( P = 2 . 2×10−46 ) . Rs4794820 is located between ORMDL3 and GSDMA on 17q21 ( see Figure 2a for the location of genes ) . Considering the transcription orientation of ORMDL3 , GSDMA , and GSDMB , rs4794820 lies in the promoter of the three genes and is possibly a pleiotropic regulatory variant or , more likely , in LD with such a regulatory element . We observed a concordance between SNPs associated with transcript levels of ORMDL3 , GSDMB and GSDMA ( Figure 2a ) . These genes are likely co-regulated , as proposed previously [32] , [37] . However , the most compelling eQTL on 17q21 was observed in the association between the asthma susceptibility SNP ( rs3859192 – p = 1 . 11−12 for association with asthma ) and the level of expression of GSDMA ( P = 3 . 55×10−151 , Figure 2b ) . In fact , many SNPs on 17q21 were strongly associated with the expression of GSDMA ( Figure 2a ) . Thus our data strongly suggest that the SNPs across this whole locus are associated with asthma because they modulate GSDMA expression in the lung , clearly showing the value of our lung eQTL dataset to refine previous GWAS hits for asthma . It should be noted that correlations between GWAS and eQTL results must be interpreted with caution . Here we related eQTLs in the lung to GWAS studies for asthma which is an airway disease but which is known to involve other tissue types including immune cells . The genetic control of gene expression is tissue specific . Regulatory variants for GSDMA identified in the lung may regulate a different gene ( s ) in another relevant tissue or cell-type for asthma . More functional work in multiple tissues and cell-types measuring gene expression with or without various stimuli will be required to confirm the causal asthma gene on 17q21 . The case for GSDMA as the susceptibility gene underlying the 17q GWAS signal is supported by biologic plausibility . GSDMA is a member of the gasdermin family of proteins first identified in the mouse [39] . They have been reported as being expressed in the upper gastrointestinal tract and skin where they are involved in the regulation of apoptosis and act as tumor suppressors [40] . Further support for GSDMA as an asthma susceptibility gene is our observation that GSDMA is robustly expressed in human airway epithelium ( Figure 3 ) . In addition its network neighborhood is enriched in genes involved in immune responses . We examined the genes around GSDMA using Bayesian and co-expression networks , and performed functional annotation of these genes ( Table S4a–S4b ) . The GSDMA network neighborhood is enriched for a number of immune response pathways , which are highly relevant to asthma . By integrating asthma GWAS results with our eSNP and Bayesian network , we were able to identify a network of 34 genes that highlights the molecular underpinnings of asthma ( Figure 5 and Table S3 ) . This network of genes is annotated as inflammatory response by the online tool DAVID [41] with an enrichment score of 3 . 27 ( P = 7×10−6 ) . Of particular interest is one Key Driver node , SOCS3 ( suppressor of cytokine signaling 3 ) . SOCS3 belongs to the SOCS family of genes that are cytokine-inducible negative regulators of cytokine signaling and play an important role in TH2-mediated allergic responses through control of the balance between TH1 and TH2 cells . It is implicated in both asthma and atopic dermatitis , as well as in regulating serum IgE levels [42] . Moriwaki and colleagues found that down-regulation of Socs3 in ovalbumin sensitized mice caused attenuation of eosinophilia and airway hyperresponsiveness generated by ovalbumin challenge [43] . Important discoveries in the field of asthma were made using eQTL mapping in lymphoblastoid cell lines [10] . In this study , we used lung tissue to map lung eQTLs , which is likely the most relevant tissue to study the genetics of lung diseases like asthma . Ideally one would want to relate SNPs to gene expression and to disease phenotype using tissue or cells from individuals affected by the disease of interest [44] . In this study , only a small percentage of the subjects had self-reported asthma . It should be noted that eQTL mapping in liver tissues was successful to identify new susceptibility genes for type 1 diabetes , coronary artery disease and blood lipid levels [19] , [20] , [45] . These discoveries were made regardless of the medical condition of patients from whom the liver tissues were explanted . Similarly , lung tissue from any source ( normal or diseased ) is a valid approach to identify the genetic variants that influence gene expression irrespective of disease status . The heterogeneous nature of the tissue we profiled with varying environmental exposures ( e . g . smoking ) and disease status ( e . g . lung cancer ) would not be expected to generate associations between SNPs and gene expression but would rather tend to introduce noise into these relationships . In summary , the present study reports on a comprehensive set of lung eQTLs that complement previous large scale studies of eQTLs in other tissue types [18] , [19] and which can be used to shed light on GWAS findings in lung diseases . Using the results of the largest asthma GWA study as an example we show how the lung tissue eQTL dataset can be used to identify the most likely causal genes and pathways . This dataset constitutes an invaluable tool to provide new insights into the pathogenesis of other lung diseases such as chronic obstructive pulmonary disease , lung cancer and cystic fibrosis .
All lung tissue samples were obtained in accordance with Institutional Review Board guidelines at the three sites: Laval University ( Quebec , Canada ) , University of British-Columbia ( Vancouver , Canada ) and Groningen University ( Groningen , The Netherlands ) . All patients provided written informed consent and the study was approved by the ethics committees of the Institut universitaire de cardiologie et de pneumologie de Québec and the UBC-Providence Health Care Research Institute Ethics Board for Laval and UBC , respectively . The study protocol was consistent with the Research Code of the University Medical Center Groningen and Dutch national ethical and professional guidelines ( “Code of conduct; Dutch federation of biomedical scientific societies”; http://www . federa . org ) . Lung tissue was collected from patients who underwent lung resectional surgery at the three participating sites; Laval University , University of British Columbia , and University of Groningen . Lung specimens from the Laval site were taken from the Respiratory Health Network Tissue Bank of the FRQS ( www . rsr . chus . qc . ca ) . Subjects' enrollment and lung tissues processing at the three sites are described in Text S1 . Gene expression profiles were obtained using a custom Affymetrix array ( see GEO platform GPL10379 ) testing 51 , 627 non-control probesets . DNA samples were genotyped on the Illumina Human1M-Duo BeadChip array . 409 , 363 , and 339 patients had both genotyping and gene expression data that passed standard quality controls ( see Text S1 ) in Laval , Groningen , and UBC , respectively . These final datasets ( total n = 1 , 111 ) were used to discover eQTLs and to identify eSNPs . The normalized expression data were adjusted for age , gender and smoking status in a robust linear model to accommodate potential outliers in expression level ( Figure 1 ) . In parallel , we performed genotype QC to exclude SNPs of low call rate ( <0 . 9 ) and deviating from Hardy-Weinberg equilibrium ( P<1×10−6 ) . Genotype imputations were based on the cleaned sets . In the end , assayed and imputed genotypes were used to identify cis and trans acting expression quantitative trait loci ( eQTLs ) following a method similar to that previously described in Schadt et al . [19] . The strategy for assembling of the asthma candidate gene list ( Table S5 ) and for constructing Bayesian networks , and co-expression modules are given in Text S1 . | Recent genome-wide association studies ( GWAS ) have identified genetic variants associated with lung diseases . The challenge now is to find the causal genes in GWAS–nominated chromosomal regions and to characterize the molecular function of disease-associated genetic variants . In this paper , we describe an international effort to systematically capture the genetic architecture of gene expression regulation in human lung . By studying lung specimens from 1 , 111 individuals of European ancestry , we found a large number of genetic variants affecting gene expression in the lung , or lung expression quantitative trait loci ( eQTL ) . These lung eQTLs will serve as an important resource to aid in the understanding of the molecular underpinnings of lung biology and its disruption in disease . To demonstrate the utility of this lung eQTL dataset , we integrated our data with previous genetic studies on asthma . Through integrative techniques , we identified causal variants and genes in GWAS–nominated loci and found key molecular drivers for asthma . We feel that sharing our lung eQTLs dataset with the scientific community will leverage the impact of previous large-scale GWAS on lung diseases and function by providing much needed functional information to understand the molecular changes introduced by the susceptibility genetic variants . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"genetics",
"genetics",
"and",
"genomics",
"biology",
"genomics",
"respiratory",
"medicine",
"pulmonology"
] | 2012 | Lung eQTLs to Help Reveal the Molecular Underpinnings of Asthma |
Increasing the intracellular Zn2+ concentration with zinc-ionophores like pyrithione ( PT ) can efficiently impair the replication of a variety of RNA viruses , including poliovirus and influenza virus . For some viruses this effect has been attributed to interference with viral polyprotein processing . In this study we demonstrate that the combination of Zn2+ and PT at low concentrations ( 2 µM Zn2+ and 2 µM PT ) inhibits the replication of SARS-coronavirus ( SARS-CoV ) and equine arteritis virus ( EAV ) in cell culture . The RNA synthesis of these two distantly related nidoviruses is catalyzed by an RNA-dependent RNA polymerase ( RdRp ) , which is the core enzyme of their multiprotein replication and transcription complex ( RTC ) . Using an activity assay for RTCs isolated from cells infected with SARS-CoV or EAV—thus eliminating the need for PT to transport Zn2+ across the plasma membrane—we show that Zn2+ efficiently inhibits the RNA-synthesizing activity of the RTCs of both viruses . Enzymatic studies using recombinant RdRps ( SARS-CoV nsp12 and EAV nsp9 ) purified from E . coli subsequently revealed that Zn2+ directly inhibited the in vitro activity of both nidovirus polymerases . More specifically , Zn2+ was found to block the initiation step of EAV RNA synthesis , whereas in the case of the SARS-CoV RdRp elongation was inhibited and template binding reduced . By chelating Zn2+ with MgEDTA , the inhibitory effect of the divalent cation could be reversed , which provides a novel experimental tool for in vitro studies of the molecular details of nidovirus replication and transcription .
Zinc ions are involved in many different cellular processes and have proven crucial for the proper folding and activity of various cellular enzymes and transcription factors . Zn2+ is probably an important cofactor for numerous viral proteins as well . Nevertheless , the intracellular concentration of free Zn2+ is maintained at a relatively low level by metallothioneins , likely due to the fact that Zn2+ can serve as intracellular second messenger and may trigger apoptosis or a decrease in protein synthesis at elevated concentrations [1] , [2] , [3] . Interestingly , in cell culture studies , high Zn2+ concentrations and the addition of compounds that stimulate cellular import of Zn2+ , such as hinokitol ( HK ) , pyrrolidine dithiocarbamate ( PDTC ) and pyrithione ( PT ) , were found to inhibit the replication of various RNA viruses , including influenza virus [4] , respiratory syncytial virus [5] and several picornaviruses [6] , [7] , [8] , [9] , [10] , [11] . Although these previous studies provided limited mechanistic information , this suggests that intracellular Zn2+ levels affect a common step in the replicative cycle of these viruses . In cell culture , PT stimulates Zn2+ uptake within minutes and inhibits RNA virus replication through a mechanism that has only been studied in reasonable detail for picornaviruses [11] , [12] . In vitro studies with purified rhinovirus and poliovirus 3C proteases revealed that protease activity was inhibited by Zn2+ [13] , [14] , which is in line with the inhibition of polyprotein processing by zinc ions that was observed in cells infected with human rhinovirus and coxsackievirus B3 [11] . The replication of segmented negative-strand RNA viruses such as influenza virus , however , does not depend on polyprotein processing and the effect of PDTC-mediated Zn2+ import was therefore hypothesized to result from inhibition of the viral RNA-dependent RNA polymerase ( RdRp ) and cellular cofactors [4] . Moreover , an inhibitory effect of Zn2+ on the activity of purified RdRps from rhinoviruses and hepatitis C virus was noted , but not investigated in any detail [15] , [16] . Details on the effect of zinc ions are currently largely unknown for nidoviruses . This large group of positive-strand RNA ( +RNA ) viruses includes major pathogens of humans and livestock , such as severe acute respiratory syndrome coronavirus ( SARS-CoV ) , other human coronaviruses , the arteriviruses equine arteritis virus ( EAV ) , and porcine reproductive and respiratory syndrome virus ( PRRSV ) [17] , [18] . The common ancestry of nidoviruses is reflected in their similar genome organization and expression strategy , and in the conservation of a number of key enzymatic functions in their large replicase polyproteins [19] . A hallmark of the corona- and arterivirus replicative cycle is the transcription of a 5′- and 3′-coterminal nested set of subgenomic ( sg ) mRNAs from which the viral structural and accessory protein genes are expressed [20] , [21] . Analogous to picornaviruses [13] , [22] , zinc ions were demonstrated to inhibit certain proteolytic cleavages in the processing of the coronavirus replicase polyproteins in infected cells and cell-free systems [23] , [24] . In this study we report that the zinc-ionophore pyrithione ( PT ) in combination with Zn2+ is a potent inhibitor of the replication of SARS-coronavirus ( SARS-CoV ) and equine arteritis virus ( EAV ) in cell culture . To assess whether - besides a possible effect on proteolytic processing - nidovirus RTC subunits and RNA synthesis are directly affected by Zn2+ , we employed in vitro systems for SARS-CoV and EAV RNA synthesis that are based on membrane-associated RTCs isolated from infected cells ( from here on referred to as RTC assays ) [25] , [26] . In addition , we used in vitro recombinant RdRp assays to directly study the effect of zinc ions on the RdRps of SARS-CoV and EAV [27] , [28] . Using these independent in vitro approaches , we were able to demonstrate that Zn2+ directly impairs nidovirus RNA synthesis , since it had a strong inhibitory effect in both RTC and RdRp assays . Interestingly , the Zn2+-mediated inhibition could be reversed through the addition of a Zn2+ chelator ( MgEDTA ) . We therefore applied this compound to stop and restart the in vitro RNA-synthesizing activity at will . This convenient tool allowed us to study various mechanistic aspects of arteri- and coronavirus RNA synthesis in more detail . Additionally , the zinc-mediated inhibition of nidovirus RNA synthesis described here may provide an interesting basis to further explore the use of zinc-ionophores in antiviral therapy .
Zinc ions are involved in many different cellular processes , but the concentration of free Zn2+ is maintained at a relatively low level by metallothioneins [1] . Zn2+ and compounds that stimulate import of Zn2+ into cells , such as PT , were previously found to inhibit replication of several picornaviruses , including rhinoviruses , foot-and-mouth disease virus , coxsackievirus , and mengovirus in cell culture [6] , [7] , [8] , [9] , [10] , [11] . To determine whether Zn2+ has a similar effect on nidoviruses , we investigated the effect of PT and Zn2+ on the replication of EAV and SARS-CoV in Vero-E6 cells , using reporter viruses that express green fluorescent proteins ( GFP ) , i . e . , EAV-GFP [29] and SARS-CoV-GFP [30] . EAV-GFP encodes an N-terminal fusion of GFP to the viral nonstructural protein 2 ( nsp2 ) , one of the cleavage products of the replicase polyproteins , and thus provides a direct readout for translation of the replicase gene . In SARS-CoV-GFP , reporter expression occurs from sg mRNA 7 , following the replacement of two accessory protein-coding genes ( ORFs 7a and 7b ) that are dispensable for replication in cell culture . We first assessed the cytotoxicity of a range of PT concentrations ( 0–32 µM ) in the presence of 0 to 8 µM ZnOAc2 . Treatment with PT of concentrations up to 32 µM in combination with <4 µM ZnOAc2 did not reduce the viability of mock-infected cells after 18 h ( Fig . 1A ) , as measured by the colorimetric MTS ( 3- ( 4 , 5-dimethylthiazol-2-yl ) -5- ( 3-carboxymethoxyphenyl ) -2- ( 4-sulfophenyl ) -2H-tetrazolium ) viability assay . As elevated Zn2+ concentrations are known to inhibit cellular translation , we also used metabolic labeling with 35S-methionine to assess the effect of PT and Zn2+ on cellular protein synthesis . Incubation of Vero-E6 cells for 18 h with the combinations of PT and Zn2+ mentioned above , followed by a 2-h metabolic labeling , revealed no change in overall cellular protein synthesis when the concentration of ZnOAc2 was <4 µM ( data not shown ) . Using these non-cytotoxic conditions we subsequently tested the effect of PT and ZnOAc2 on EAV-GFP and SARS-CoV-GFP replication . To this end , Vero-E6 cells in 96-well plates were infected with a multiplicity of infection ( m . o . i . ) of 4 . One hour post infection ( h p . i . ) , between 0 and 32 µM of PT and 0 , 1 , or 2 µM ZnOAc2 were added to the culture medium . At 17 h p . i . , a time point at which GFP expression in untreated infected cells reaches its maximum for both viruses , cells were fixed , and GFP fluorescence was quantified . The reporter gene expression of both SARS-CoV-GFP and EAV-GFP was already significantly inhibited in a dose-dependent manner by the addition of PT alone ( Fig . 1B and C ) . This effect was significantly enhanced when 2 µM of Zn2+ was added to the medium . We found that addition of ZnOAc2 alone also reduced virus replication , but only at levels that were close to the 50% cytotoxicity concentration ( CC50 ) of ZnOAc2 in Vero-E6 cells ( ∼70 µM , data not shown ) . This is likely due to the poor solubility of Zn2+ in phosphate-containing medium and the inefficient uptake of Zn2+ by cells in the absence of zinc-ionophores . The combination of 2 µM PT and 2 µM ZnOAc2 resulted in a 98±1% and 85±3% reduction of the GFP signal for EAV-GFP and SARS-CoV-GFP , respectively . No cytotoxicity was observed for this combination of PT and ZnOAc2 concentrations . From the dose-response curves in Fig . 1 , a CC50 value of 82 µM was calculated for PT in the presence of 2 µM zinc . Half maximal inhibitory concentrations ( IC50 ) of 1 . 4 µM and 0 . 5 µM and selectivity indices of 59 and 164 were calculated for SARS-CoV and EAV , respectively . We previously developed assays to study the in vitro RNA-synthesizing activity of RTCs isolated from cells infected with SARS-CoV or EAV [25] , [26] . In these RTC assays [α-32P]CMP is incorporated into both genomic ( replication ) and sg mRNA ( transcription ) ( Fig . 2 ) . This allowed us to monitor the synthesis of the same viral RNA molecules that can be detected by hybridization of RNA from nidovirus-infected cells . A benefit of these assays is that the activity does not depend on continued protein synthesis and that it allows us to study viral RNA synthesis independent of other aspects of the viral replicative cycle [26] . To investigate whether the inhibitory effect of PT and zinc ions on nidovirus replication in cell culture is reflected in a direct effect of Zn2+ on viral RNA synthesis , we tested the effect of Zn2+ addition on RTC activity . For both EAV ( Fig . 2A ) and SARS-CoV ( Fig . 2B ) , a dose-dependent decrease in the amount of RNA synthesized was observed when ZnOAc2 was present . For both viruses , a more than 50% reduction of overall RNA-synthesis was observed at a Zn2+ concentration of 50 µM , while less than 5% activity remained at a Zn2+ concentration of 500 µM . Both genome synthesis and sg mRNA production were equally affected . To test whether the inhibition of RTC activity by Zn2+ was reversible , RTC reactions were started in the presence or absence of 500 µM Zn2+ . After 30 min , these reactions were split into two aliquots and magnesium-saturated EDTA ( MgEDTA ) was added to one of the tubes to a final concentration of 1 mM ( Fig . 3A ) . We used MgEDTA as Zn2+ chelator in these in vitro assays , because it specifically chelates Zn2+ while releasing Mg2+ , due to the higher stability constant of the ZnEDTA complex . Uncomplexed EDTA inhibited RTC activity in all reactions ( data not shown ) , most likely by chelating the Mg2+ that is crucial for RdRp activity [27] , [28] , whereas MgEDTA had no effects on control reactions without Zn2+ ( Fig . 3B , compare lane 1 and 2 ) . As shown in Fig . 2 , the EAV RTC activity that was inhibited by Zn2+ ( Fig . 3B&C , lane 3 ) could be restored by the addition of MgEDTA ( Fig . 3B , lane 4 ) to a level observed for control reactions without Zn2+ ( Fig . 3B , lane 1 ) . Compared to the untreated control , the EAV RTC assay produced approximately 30% less RNA , which was consistent with the 30% shorter reaction time after the addition of the MgEDTA ( 100 versus 70 min for lanes 1 and 4 , respectively ) . Surprisingly , SARS-CoV RTC assays that were consecutively supplemented with Zn2+ and MgEDTA incorporated slightly more [α-32P]CMP compared to untreated control reactions ( Fig . 3C; compare lane 1 and 4 ) . This effect was not due to chelation of the Zn2+ already present in the post-nuclear supernatant ( PNS ) of SARS-CoV-infected cells , as this increase was not observed when MgEDTA was added to a control reaction without additional Zn2+ ( Fig . 3C , lane 2 ) . To establish whether inhibition of RTC activity might be due to a direct effect of Zn2+ on nidovirus RdRp activity , we tested the effect of Zn2+ on the activity of the purified recombinant RdRps of SARS-CoV ( nsp12 ) and EAV ( nsp9 ) using previously developed RdRp assays [27] , [28] . Using an 18-mer polyU template , the EAV RdRp incorporated [α-32P]AMP into RNA products of up to 18 nt in length ( Fig . 4A ) . Initiation was de novo , which is in line with previous observations and the presence of a conserved priming loop in the nsp9 sequence [28] . Unlike the EAV RdRp nsp9 , the in vitro activity of the SARS-CoV RdRp nsp12 - which lacks a priming loop - was shown to be strictly primer-dependent [27] . Thus , to study the RdRp activity of SARS-CoV nsp12 , a primed polyU template was used ( Fig . 4B ) , thereby allowing us to sample [α-32P]AMP incorporation as described previously [27] . As specificity controls , we used the previously described SARS-CoV nsp12 mutant D618A [27] , which contains an aspartate to alanine substitution in motif A of the RdRp active site , and EAV nsp9-D445A , in which we engineered an aspartate to alanine substitution at the corresponding site of EAV nsp9 [28] , [31] . Both mutant RdRps showed greatly reduced [α-32P]AMP incorporation in our assays ( Fig . 4 ) , confirming once again that the radiolabeled RNA products derived from nidovirus RdRp activity . Addition of ZnOAc2 to RdRp assays resulted in a strong , dose-dependent inhibition of enzymatic activity for both the EAV and SARS-CoV enzyme ( Fig . 5A and B , respectively ) , similar to what was observed in RTC assays . In fact , compared to other divalent metal ions such Co2+ and Ca2+ , which typically bind to amino acid side chains containing oxygen atoms rather than sulfur groups , Zn2+ was the most efficient inhibitor of SARS-CoV nsp12 RdRp activity ( Supplemental Fig . S1 ) . To test whether , as in the RTC assay , the RdRp inhibition by zinc ions was reversible , RdRp assays were pre-incubated with 6 mM Zn2+ , a concentration that consistently gave >95% inhibition . After 30 min , 8 mM MgEDTA was added to both a control reaction and the reaction inhibited with ZnOAc2 , and samples were incubated for another 30 min ( Fig . 5C ) . As shown in Fig . 5D , the inhibition of EAV RdRp activity by Zn2+ could be reversed by chelation of Zn2+ ( Fig . 5D; compare lanes 3 and 4 ) . The amount of product synthesized was consistently 60±5% of that synthesized in a 60-min control reaction ( Fig . 5D; compare lanes 1 and 4 ) , which was within the expected range given the shorter reaction time . The inhibition of the SARS-CoV RdRp was reversible as well . During the 30-min incubation after the addition of MgEDTA , SARS-CoV nsp12 incorporated 40±5% of the label incorporated during a standard 60-min reaction ( Fig . 5E ) . This was slightly lower than the expected yield and may be caused by the elevated Mg2+ concentration , which was shown to be suboptimal for nsp12 activity [27] and results from the release of Mg2+ from MgEDTA upon chelation of Zn2+ . For EAV , close inspection of the RdRp assays revealed a less pronounced effect of Zn2+ on the generation of full-length 18-nt products than on the synthesis of smaller reaction intermediates ( Fig . 5A ) . This suggested that Zn2+ specifically inhibited the initiation step of EAV RNA synthesis . To test this hypothesis , an RTC assay was incubated for 30 min with unlabeled CTP ( initiation ) , after which the reaction was split in two . Then , [α-32P]CTP was added to both tubes ( pulse ) , 500 µM Zn2+ was added to one of the tubes , and samples were taken at different time points during the reaction ( Fig . 6A ) . Fig . 6B shows that in the presence of Zn2+ [α-32P]CMP was predominantly incorporated into nascent RNA molecules that were already past the initiation phase at the moment that Zn2+ was added to the reaction . No new initiation occurred , as was indicated by the smear of short radiolabeled products that progressively shifted up towards the position of full-length genomic RNA . This suggested that Zn2+ does not affect the elongation phase of EAV RNA synthesis and that it specifically inhibits initiation . This also explains the relatively weak signal intensity of the smaller sg mRNA bands ( e . g . , compare the relative change in signal of RNA2 to RNA7 ) produced in the presence of Zn2+ , since multiple initiation events are required on these short molecules to obtain signal intensities similar to those resulting from a single initiation event on the long genomic RNA , e . g . , 16 times more in the case of RNA7 . In contrast to EAV , the effect of Zn2+ on RNA synthesis by SARS-CoV RTCs was not limited to initiation , but appeared to impair the elongation phase as well , given that the addition of Zn2+ completely blocked further incorporation of [α-32P]CMP when added 40 min after the start of the reaction ( Fig . 6C ) . In the RdRp assays , the short templates used made it technically impossible to do experiments similar to those performed with complete RTCs . However , we previously noticed that at low concentrations of [α-32P]ATP ( ∼0 . 17 µM ) SARS-CoV nsp12 RdRp activity was restricted to the addition of only a single nucleotide to the primer [27] . EAV nsp9 mainly produced very short ( 2–3 nt long ) abortive RNA products and only a fraction of full length products , as is common for de novo initiating RdRps [28] . This allowed us to separately study the effect of Zn2+ on initiation and elongation by performing an experiment in which a pulse with a low concentration of [α-32P]ATP was followed by a chase in the presence of 50 µM of unlabeled ATP , which increased processivity and allowed us to study elongation ( Fig . 7A and C ) as described previously [27] . The results of these experiments were in agreement with those obtained with isolated RTCs . For EAV , with initiation and dinucleotide synthesis completely inhibited by the presence of 6 mM Zn2+ ( Supplemental Fig . S2A ) , the amount of reaction intermediates shorter than 18 nt diminished with time , while products from templates on which the RdRp had already initiated were elongated to full-length 18-nt molecules ( Fig . 7B , right panel ) . This was consistent with the observation that the EAV RdRp remained capable of extending the synthetic dinucleotide ApA to trinucleotides in the presence of Zn2+ ( Supplemental Fig . S2B ) . Likely due to the absence of reinitiation in the reactions shown in Fig . 7B , the low processivity of the EAV RdRp , and the substrate competition between the remaining [α-32P]ATP and the >200 fold excess of unlabeled ATP , the differences between the 5- and 30-min time points were small . In the absence of Zn2+ , the RdRp continued to initiate as indicated by the ladder of smaller-sized RNA molecules below the full-length product ( Fig . 7B , left panel ) and the time course shown in Supplemental Fig . S2A . In contrast , the addition of Zn2+ to a SARS-CoV RdRp reaction also blocked elongation , since extension of the radiolabeled primer as observed in the absence of Zn2+ ( Fig . 7D , left panel ) no longer occurred ( Fig . 7D , right panel ) . To assess whether Zn2+ affects the interaction of recombinant SARS-CoV nsp12 with the template used in our assays , we performed electromobility shift assays ( EMSA ) in the presence and absence of Zn2+ ( Fig . 8A ) . To measure the binding affinity of the RdRp for the template , we determined the fraction of bound template at various protein concentrations and observed a 3–4 fold reduction in RNA binding when Zn2+ was present in the assay ( Fig . 8B ) . We also assessed whether pre-incubation of the RdRp or RNA with Zn2+ was a requirement for this drop in binding affinity , but found no significant difference with experiments not involving such a preincubation ( data not shown ) . No binding was observed when a similar RNA binding assay was performed with purified EAV RdRp . Likewise , nsp9 did not bind RNA in pull-down experiments with Talon-beads , His6-tagged nsp9 , and radiolabeled EAV genomic RNA or various short RNA templates including polyU , whereas we were able to detect binding of a control protein ( SARS-CoV nsp8 , which has demonstrated RNA and DNA binding activity [32] ) using this assay . It presently remains unclear why we are not able to detect the binding of recombinant EAV nsp9 to an RNA template .
Although a variety of compounds have been studied , registered antivirals are currently still lacking for the effective treatment of SARS and other nidovirus-related diseases [33] . RdRps are suitable targets for antiviral drug development as their activity is strictly virus-specific and may be blocked without severely affecting key cellular functions . Several inhibitors developed against the polymerases of e . g . human immunodeficiency virus ( HIV ) and hepatitis C virus are currently being used in antiviral therapy or clinical trials [34] , [35] , [36] . Therefore , advancing our molecular knowledge of nidovirus RdRps and the larger enzyme complexes that they are part of , and utilizing the potential of recently developed in vitro RdRp assays [25] , [26] , [27] , [28] could ultimately aid in the development of effective antiviral strategies . Zinc ions and zinc-ionophores , such as PT and PDTC , have previously been described as potent inhibitors of various RNA viruses . We therefore investigated whether PT-stimulated import of zinc ions into cells also inhibited the replication of nidoviruses in cell culture . Using GFP-expressing EAV and SARS-CoV [29] , [30] , we found that the combination of 2 µM PT and 2 µM Zn2+ efficiently inhibited their replication , while not causing detectable cytoxicity ( Fig . 1 ) . Inhibition of replication by PT and Zn2+ at similar concentrations ( 2–10 µM ) was previously observed for several picornaviruses such as rhinoviruses , foot-and-mouth disease virus , coxsackievirus , and mengovirus [6] , [7] , [8] , [9] , [10] , [11] . The inhibitory effect of Zn2+ on the replication of picornaviruses appeared to be due to interference with viral polyprotein processing . In infections with the coronavirus mouse hepatitis virus ( MHV ) , Zn2+ also interfered with some of the replicase polyproteins cleavages [24] , albeit at a much higher concentration ( 100 µM Zn2+ ) than used in our studies . Since impaired replicase processing will indirectly affect viral RNA synthesis in the infected cell , we used two recently developed in vitro assays to investigate whether Zn2+ also affects nidovirus RNA synthesis directly . Our in vitro studies revealed a strong inhibitory effect of zinc ions on the RNA-synthesizing activity of isolated EAV and SARS-CoV RTCs . Assays with recombinant enzymes subsequently demonstrated that this was likely due to direct inhibition of RdRp function . The inhibitory effect could be reversed by chelating the zinc ions , which provides an interesting experimental ( on/off ) approach to study nidovirus RNA synthesis . Addition of Zn2+ following initiation of EAV RNA synthesis had little or no effect on NTP incorporation in molecules whose synthesis had already been initiated in the absence of Zn2+ ( Fig . 6 and 7 ) , indicating that Zn2+ does not affect elongation and does not increase the termination frequency , as was previously found for Mn2+ [25] . Therefore , Zn2+ appears to be a specific inhibitor of the initiation phase of EAV RNA synthesis . In contrast , Zn2+ inhibited SARS-CoV RdRp activity also during the elongation phase of RNA synthesis , probably by directly affecting template binding ( Fig . 8 ) . In coronaviruses , zinc ions thus appear to inhibit both the proper proteolytic processing of replicase polyproteins [23] , [24] and RdRp activity ( this study ) . Contrary to the RTC assays , millimolar instead of micromolar concentrations of ZnOAc2 were required for a nearly complete inhibition of nucleotide incorporation in RdRp assays . It has been well established that DNA and RNA polymerases use a triad of conserved aspartate residues in motifs A and C to bind divalent metal ions like Mg2+ , which subsequently coordinate incoming nucleotides during the polymerization reaction [37] , [38] . Mg2+ is also the divalent metal ion that is required for the in vitro activity of isolated SARS-CoV and EAV RTCs and recombinant RdRps [25] , [26] , [27] , [28] , although de novo initiation of EAV nsp9 is primarily Mn2+-dependent . Zn2+ could not substitute for Mg2+ or Mn2+ as cofactor as it was incapable of supporting the polymerase activity of nidovirus RTCs and RdRps in the absence of Mg2+ ( data not shown ) , as was also reported for the poliovirus RdRp [39] . Moreover , inhibition of nidovirus RdRp activity by Zn2+ was even observed at low concentrations and in the presence of a more than 25-fold excess of Mg2+ , suggesting that either the affinity of the active site for Zn2+ is much higher or that Zn2+ does not compete for Mg2+-binding and binds to another zinc ( -specific ) binding site in the protein . Specific protein domains or pockets that contain zinc ions may be involved in protein-protein interactions , protein-RNA/DNA interactions , or conformational changes in enzyme structures . Zinc-binding domains commonly consist of at least three conserved cysteine and/or histidine residues within a stretch of ∼10–30 amino acids , such as in zinc-finger motifs and metalloproteases [2] , [40] , [41] . However , in RdRps there are only few precedents for the presence of zinc-binding pockets , such as those identified in the crystal structure of the Dengue RdRp [42] . Sequence analysis of the EAV nsp9 amino acid sequence revealed that it lacks patches rich in conserved cysteines and/or histidines . In contrast , inspection of the SARS-CoV nsp12 amino acid sequence revealed two such patches , namely H295-C301-C306-H309-C310 and C799-H810-C813-H816 . A crystal structure for nsp12 is presently unavailable , but a predicted structure that represents the C-terminal two-thirds of the enzyme has been published [31] . Interestingly , in this model , C799 , H810 , C813 and H816 are in a spatial arrangement resembling that of the Zn2+ coordinating residues in the Zn2 zinc-binding pocket found in motif E of the Dengue virus RdRp ( see Supplemental Fig . S3 ) . Clearly , an in-depth analysis of nidovirus RdRps , e . g . through structural analysis and subsequent mutational studies targeting aforementioned cysteines and histidines , is required to provide further insight into and a structural basis for the Zn2+-induced inhibitory effects on RdRp activity documented in this study . Such studies may , however , be complicated when Zn2+ binding proves to be very transient in nature and not detectable with currently available methods . In summary , the combination of zinc ions and the zinc-ionophore PT efficiently inhibits nidovirus replication in cell culture . This provides an interesting basis for further studies into the use of zinc-ionophores as antiviral compounds , although systemic effects have to be considered [43] , [44] and a water-soluble zinc-ionophore may be better suited , given the apparent lack of systemic toxicity of such a compound at concentrations that were effective against tumors in a mouse xenograft model [45] . In vitro , the reversible inhibition of the RdRp by Zn2+ has also provided us with a convenient research tool to gain more insight into the molecular details of ( nido ) viral RNA synthesis , and revealed novel mechanistic differences between the RdRps of SARS-CoV and EAV .
Vero-E6 cells were cultured and infected with SARS-CoV ( strain Frankfurt-1; accession nr . AY291315 ) or SARS-CoV-GFP as described previously [46] . All procedures involving live SARS-CoV were performed in the biosafety level 3 facility at Leiden University Medical Center . BHK-21 or Vero-E6 cells were cultured and infected with EAV ( Bucyrus strain; accession nr . NC_002532 ) or EAV-GFP [29] as described elsewhere [25] . One day prior to infection , Vero-E6 cells were seeded in transparent or black ( low fluorescence ) 96-well clusters at 10 , 000 cells per well . The next day , cells were infected with SARS-CoV-GFP or EAV-GFP with an m . o . i . of 4 , and 1 h p . i . the inoculum was removed and 100 µl of medium containing 2% fetal calf serum ( FCS ) was added to each well . In some experiments 0–32 µM of pyrithione ( Sigma ) was added in addition to 0–2 µM ZnOAc2 . Infected cells were fixed at 17 h p . i . by aspirating the medium and adding 3% paraformaldehyde in PBS . After washing with PBS , GFP expression was quantified by measuring fluorescence with a LB940 Mithras plate reader ( Berthold ) at 485 nm . To determine toxicity of ZnOAc2 and PT , cells were exposed to 0–32 µM PT and 0–8 µM ZnOAc2 . After 18 h incubation , cell viability was determined with the Cell Titer 96 AQ MTS assay ( Promega ) . EC50 and CC50 values were calculated with Graphpad Prism 5 using the nonlinear regression model . RNA oligonucleotides SAV557R ( 5′-GCUAUGUGAGAUUAAGUUAU-3′ ) , SAV481R ( 5′-UUUUUUUUUUAUAACUUAAUCUCACAUAGC-3′ ) and poly ( U ) 18 ( 5′-UUUUUUUUUUUUUUUUUU-3′ ) were purchased from Eurogentec , purified from 7 M Urea/15% PAGE gels and desalted through NAP-10 columns ( GE healthcare ) . To anneal the RNA duplex SAV557R/SAV481R , oligonucleotides were mixed at equimolar ratios in annealing buffer ( 20 mM Tris-HCl pH 8 . 0 , 50 mM NaCl and 5 mM EDTA ) , denatured by heating to 90°C and allowed to slowly cool to room temperature after which they were purified from 15% non-denaturing PAGE gels . SARS-CoV and EAV RTCs were isolated from infected cells and assayed for activity in vitro as described previously [25] , [26] . To assess the effect of Zn2+ , 1 µl of a ZnOAc2 stock solution was added to standard 28-µl reactions , resulting in final Zn2+ concentrations of 10–500 µM . When Zn2+ had to be chelated in the course of the reaction , magnesium-saturated EDTA ( MgEDTA ) was added to a final concentration of 1 mM . After RNA isolation , the 32P-labeled reaction products were separated on denaturing 1% ( SARS-CoV ) or 1 . 5% ( EAV ) agarose formaldehyde gels . The incorporation of [α-32P]CMP into viral RNA was quantified by phosphorimaging of the dried gels using a Typhoon scanner ( GE Healthcare ) and the ImageQuant TL 7 software ( GE Healthcare ) . SARS-CoV nsp12 and EAV nsp9 were purified essentially as described elsewhere [27] , [28] , but with modifications for nsp9 . In short , E . coli BL21 ( DE3 ) with plasmid pDEST14-nsp9-CH was grown in auto-induction medium ZYM-5052 [47] for 6 hours at 37°C and a further 16 hours at 20°C . After lysis in buffer A ( 20 mM HEPES pH 7 . 4 , 200 mM NaCl , 20 mM imidazole , and 0 . 05% Tween-20 ) the supernatant was applied to a HisTrap column ( GE Healthcare ) . Elution was performed with a gradient of 20–250 mM imidazole in buffer A . The nsp9-containing fraction was further purified by gel filtration in 20 mM HEPES , 300 mM NaCl and 0 . 1% Tween-20 on a Superdex 200 column ( GE Healthcare ) . The fractions containing nsp9-CH were pooled , dialyzed against 1000 volumes of buffer B ( 20 mM HEPES , 100 mM NaCl , 1 mM DTT and 50% glycerol ) and stored at −20°C . RdRps with a D618A ( SARS-CoV ) or D445A ( EAV ) mutation were obtained by site-directed mutagenesis of the wild-type ( wt ) plasmid pDEST14-nsp9-CH [28] with oligonucleotides 5′-TACTGCCTTGAAACAGCCCTGGAGAGTTGTGAT-3′ and 5′-ATCACAACTCTCCAGGGCTGTTTCAAGGCAGTA-3′ , and plasmid pASK3-Ub-nsp12-CHis6 with oligonucleotides 5′-CCTTATGGGTTGGGCTTATCCAAAATGTG-3′ and 5′-CACATTTTGGATAAGCCCAACCCATAAGGA-3′ , as described elsewhere [27] . Mutant proteins were purified parallel to the wt enzymes . Standard reaction conditions for the RdRp assay with 0 . 1 µM of purified SARS-CoV nsp12 are described elsewhere [27] . To study the effect of Zn2+ in this assay , 0 . 5 µl of a dilution series of 0–80 mM ZnOAc2 was added to the 5 µl reaction mixture , yielding final Zn2+ concentrations of 0–8 mM . The EAV RdRp assay contained 1 µM nsp9 , 1 µM RNA template poly ( U ) 18 , 0 . 17 µM [α-32P]ATP ( 0 . 5 µCi/µl; Perkin-Elmer ) , 50 µM ATP , 20 mM Tris-HCl ( pH 8 . 0 ) , 10 mM NaCl , 10 mM KCl , 1 mM MnCl2 , 4 mM MgOAc2 , 5% glycerol , 0 . 1% Triton-X100 , 1 mM DTT and 0 . 5 units RNaseOUT . ZnOAc2 was added to the reaction to give a final concentration of 0–6 mM . To chelate Zn2+ during reactions , MgEDTA was added to a final concentration of 8 mM . Reactions were terminated after 1 hour and analyzed as described [27] . SARS-CoV RdRp was incubated with 0 . 2 nM 5′ 32P-labeled SAV557R/SAV481R RNA duplex , for 10 minutes at 30°C either in presence or absence of 6 mM ZnOAc2 . Reactions were analyzed as described previously [27] . | Positive-stranded RNA ( +RNA ) viruses include many important pathogens . They have evolved a variety of replication strategies , but are unified in the fact that an RNA-dependent RNA polymerase ( RdRp ) functions as the core enzyme of their RNA-synthesizing machinery . The RdRp is commonly embedded in a membrane-associated replication complex that is assembled from viral RNA , and viral and host proteins . Given their crucial function in the viral replicative cycle , RdRps are key targets for antiviral research . Increased intracellular Zn2+ concentrations are known to efficiently impair replication of a number of RNA viruses , e . g . by interfering with correct proteolytic processing of viral polyproteins . Here , we not only show that corona- and arterivirus replication can be inhibited by increased Zn2+ levels , but also use both isolated replication complexes and purified recombinant RdRps to demonstrate that this effect may be based on direct inhibition of nidovirus RdRps . The combination of protocols described here will be valuable for future studies into the function of nidoviral enzyme complexes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"virology/viral",
"replication",
"and",
"gene",
"regulation",
"molecular",
"biology/rna-protein",
"interactions",
"virology/antivirals,",
"including",
"modes",
"of",
"action",
"and",
"resistance",
"virology",
"biochemistry/replication",
"and",
"repair"
] | 2010 | Zn2+ Inhibits Coronavirus and Arterivirus RNA Polymerase Activity In Vitro and Zinc Ionophores Block the Replication of These Viruses in Cell Culture |
Protein thermostability is a crucial factor for biotechnological enzyme applications . Protein engineering studies aimed at improving thermostability have successfully applied both directed evolution and rational design . However , for rational approaches , the major challenge remains the prediction of mutation sites and optimal amino acid substitutions . Recently , we showed that such mutation sites can be identified as structural weak spots by rigidity theory-based thermal unfolding simulations of proteins . Here , we describe and validate a unique , ensemble-based , yet highly efficient strategy to predict optimal amino acid substitutions at structural weak spots for improving a protein’s thermostability . For this , we exploit the fact that in the majority of cases an increased structural rigidity of the folded state has been found as the cause for thermostability . When applied prospectively to lipase A from Bacillus subtilis , we achieved both a high success rate ( 25% over all experimentally tested mutations , which raises to 60% if small-to-large residue mutations and mutations in the active site are excluded ) in predicting significantly thermostabilized lipase variants and a remarkably large increase in those variants’ thermostability ( up to 6 . 6°C ) based on single amino acid mutations . When considering negative controls in addition and evaluating the performance of our approach as a binary classifier , the accuracy is 63% and increases to 83% if small-to-large residue mutations and mutations in the active site are excluded . The gain in precision ( predictive value for increased thermostability ) over random classification is 1 . 6-fold ( 2 . 4-fold ) . Furthermore , an increase in thermostability predicted by our approach significantly points to increased experimental thermostability ( p < 0 . 05 ) . These results suggest that our strategy is a valuable complement to existing methods for rational protein design aimed at improving thermostability .
Thermostability is a crucial factor for a wealth of biotechnological enzyme applications [1 , 2] . Protein engineering aimed at improving thermostability is thus an important field of research in biotechnology [3 , 4] . There , methods of directed evolution are usually applied , which mimic natural evolution [5–8] . However , directed evolution is limited in that out of the extraordinarily large number of possible variant proteins , only a small subset can be experimentally tested [9] . Alternatively , rational approaches have been successfully pursued [10–13] but the major challenge here remains the prediction of mutation sites and the optimal amino acid substitution at such sites [14 , 15] . As to the prediction of mutation sites , we developed the rigidity theory-based Constraint Network Analysis ( CNA ) approach [16–21] ( available as a web service at http://cpclab . uni-duesseldorf . de/cna/ [16–21] ) , which identifies residues in a protein that are structural “weak spots” . For this , a protein is modeled as a network of sites ( atoms ) and constraints ( covalent and noncovalent interactions ) [22] . Rigid atom clusters and flexible regions in between are then rigorously determined by rigidity analysis [23–25] . By successively removing non-covalent constraints from the network , the thermal unfolding of the protein is simulated ( Fig 1a and 1b ) [16 , 18 , 19 , 26] . From the unfolding trajectory , a phase transition temperature Tp is identified , which relates to the ( thermodynamic ) thermostability , as are the weak spots ( Fig 1c ) . Mutating such weak spots should likely improve a protein’s thermostability [16 , 18 , 19] . Here , we describe and validate a novel and unique strategy based on the CNA approach to predict optimal amino acid substitutions at these weak spots . At variance with other rational approaches that rely upon calculating free energies for predicting effects of mutations on a protein’s thermostability [27–33] , we exploit the fact that in the majority of cases an increased structural rigidity of the folded state has been identified as the underlying cause for thermostability [34] . To this end , we add a highly efficient , ensemble-based second step by generating structural models of single-point site-saturation mutations at identified weak spots , filtering the models with respect to their structural quality , and screening for variants with increased structural rigidity ( Fig 1d–1f , see below for detailed descriptions ) . Using the recently developed ENTFNC approach [35] that performs rigidity analyses on an ensemble of network topologies generated from a single input structure using fuzzy network constraints , rather than a structural ensemble , this second step only takes about 1 h on a single core per variant and can be performed in parallel for multiple variants . We applied this strategy prospectively on lipase LipA from Bacillus subtilis ( BsLipA ) ; BsLipA has considerable biotechnological importance [36 , 37] and has been extensively studied with respect to thermostability [6 , 15 , 38–43] , which makes BsLipA a prominent model system . Out of 589 BsLipA variants screened in silico , twelve were suggested for experimental testing . Of these , three showed a significant increase of up to 6 . 6°C in thermostability with respect to the wild-type enzyme ( WT ) . We thus achieved both a high success rate in predicting thermostabilized lipase variants and a remarkably large increase in the thermostability of such variants . This demonstrates the value of the novel strategy , which extends the existing portfolio of methods for rational protein design aimed at improving thermostability .
BsLipA has a minimal α/β hydrolase fold in which a central parallel β-sheet of six β-strands is surrounded by six α-helices [44] . For identifying weak spots on BsLipA , a thermal unfolding simulation was carried out by CNA on an ensemble of 2000 WT BsLipA structures extracted from a molecular dynamics ( MD ) trajectory of 100 ns length ( Fig 1a ) . The ensemble-based CNA was pursued to increase the robustness of the rigidity analyses [19 , 35 , 45] . The unfolding trajectory ( Figs 1b and 2 ) reveals the early segregation of loops from the largest rigid cluster , followed by the segregation of α-helices and , finally , the segregation and disintegration of the β-sheet region . This order of segregation is in agreement with experimental findings on the unfolding of other α/β hydrolase proteins [46 , 47] . The realistic description of WT BsLipA thermal unfolding encouraged us to identify weak spots at major phase transitions along the unfolding trajectory ( Fig 1c ) . By visual inspection of the unfolding trajectory , we identified five major transitions ( T1–T5 ) at which helices αA , αF , αD and αE , αB , αC as well as the central beta sheet segregate from the largest rigid cluster at temperatures 316 , 318 , 334 , 336 , and 338 K , respectively ( Table 1 and Fig 2 ) . Weak spot residues were then identified as those residues that are in the neighborhood of the largest rigid cluster from which they segregate at the respective major transition . These residues are particularly promising for increasing BsLipA’s thermostability considering that their mutation can improve the interaction strength with the largest rigid cluster and , hence , delay the disintegration of that cluster with increasing temperature . In total , 36 weak spots were found , which are located on α-helices and loops joining α-helices and β-strands ( Fig 2 ) . The weak spot residues are very diverse in size ( ranging from Gly to Trp ) and physicochemical properties ( charged , uncharged polar , and hydrophobic ) ( Table 1 ) . Of these , weak spot residues at highly conserved sequence positions were discarded ( Figs 1d and S1; Table 1 ) because conserved residues are usually important for function and/or stability of a protein and , hence , should not be mutated [48 , 49] . For each of the remaining 31 weak spots ( ~17% of all BsLipA residues ) , computational site saturation mutagenesis was performed by generating structures of all possible single-point amino acid substitutions using the SCWRL program ( Fig 1e ) [50] . SCWRL constructs variant models by predicting backbone-dependent side-chain conformations with the help of a rotamer library . This resulted in 589 single point variants . 67 variant structures were discarded based on the evaluation of residue-wise non-local interaction energies by the ANOLEA server ( S1 Fig ) [51 , 52] . In such structures , the mutation apparently does not fit into the environment of the other residues . The remaining 522 variants were subjected to thermal unfolding simulations on ensembles of network topologies using the ENTFNC approach [35] implemented in CNA . Differences in the phase transition temperatures ΔTp = Tp ( variant ) − Tp ( WT ) were averaged over 1000 simulations started from different network topologies generated for each variant ( see “Materials and Methods section”; Fig 1f ) . A map of ΔTp values of all variants is shown in S1 Fig . In total , this procedure yielded a predicted thermostabilization with respect to WT BsLipA for 75 out of the 522 mutations ( ~14% ) investigated . In order to further reduce the number of mutations for experimental validation only the mutation with the highest ΔTp was chosen from all mutations with ΔTp > 1 K at a weak spot . The sole exception is G104 located in the active site , for which two mutations were chosen . This resulted in twelve lipase variants of which the most are associated with weak spot residues on helix αB identified during the late transition T4 ( Table 2; S1 Fig ) . As a negative control , we also predicted 10 variants with negative ΔTp , i . e . , where a mutation according to the thermal unfolding simulations leads to a decrease in thermostability with respect to WT ( S1 Table ) . Six of these mutations were chosen from the above analyses of 522 variants such that they have the most negative ΔTp; four were chosen with the most negative ΔTp from analyses of variants with a mutation not at a weak spot . Initially , specific activities of WT BsLipA and the twelve variants ( Table 2 ) for hydrolysis of p-nitrophenyl-palmitate ( pNPP ) were measured at temperatures between 40 and 60°C after keeping them at the respective temperatures for 5 min . WT BsLipA showed the highest specific activity ( 246 U/mg ) among all BsLipA variants at the temperature of maximum activity Tmax ( 40°C ) ( S2 Fig ) . At temperatures above 55°C , the activity begins to drop , which is probably due to an unfolding already within 5 min of preincubation . However , two variants , F58I and V96S , showed higher activities than the WT at temperatures above 58°C ( S2 Fig ) , which may originate from them being more stable at high temperatures . Next , thermostability was assessed by measuring the activity of each BsLipA variant at temperatures between 40 and 60°C after incubating the respective variant at these temperatures for 30 min . Three variants , V54H , F58I , and V96S , were more thermostable than WT; they consistently showed higher activities than the WT at temperatures above 48°C ( Figs 3a and S3 ) . The largest differences between thermostabilities of WT and variants of BsLipA was observed at 53 . 5°C where the activities of V54H and V96S were twice as a high as that of the WT , and the activity of F58I was four times higher ( Fig 3 ) . The kinetic constants of these variants were derived from initial rate measurements for hydrolysis of p-nitrophenyl-decanoate ( pNPD ) at 40°C ( see S1 Text ) . No significant impact on the Michaelis constant ( KM ) was observed , and the turnover numbers ( kcat ) were reduced by at most 25% ( Table 3 ) . Thus , the thermostability of the variants has been increased without significantly influencing kcat / KM at 40°C . Still , two of the three thermostable variants showed lower activities than WT at temperatures below ~45°C ( Fig 3 ) . This may have been caused by a rigidification of the lipase structure in the thermostable variants ( see section “Analysis of thermostability changes at the structural level” below ) , which may also influence the flexibility of the active site . Similarly , in a series of five orthologs of 2-deoxy-d-ribose-5-phosphate aldolase ( DERA ) from psychrophilic , mesophilic , and hyperthermophilic organisms investigated by us recently in terms of biochemical , structural , and rigidity properties , an anticorrelation between specific activity at temperatures ≤ 40°C and experimental or computed melting temperature was observed [53] . In that study , both the analysis of local rigidity by CNA and B-factor analysis of X-ray structures provided independent clues that psychrophilic DERAs have a more flexible environment of the substrate binding pocket . Thus , it may depend on the actual operating temperature of an enzymatic process whether it is worth to apply thermostable variants with increased activities at high ( er ) temperatures only . Finally , the thermostability of BsLipA variants was quantified by T′50 values; these values report on the temperature at which the fraction of the activity to the initial activity ( at 40°C ) is 50% after incubation for 30 min . This is different from the T50 values normally used for characterizing the thermostability of proteins [15 , 54 , 55] in that the activity here is measured at the temperature of incubation , not at room temperature after cooling . T′50 thus reports on the thermo-tolerance of an enzyme during operational bioprocesses carried out at elevated temperatures for a longer duration of time , e . g . , as done in the lipid processing industry [56] . The three variants V54H , F58I , and V96S showed T′50 values higher by 5 . 7 , 6 . 6 , and 3 . 6°C , respectively , than WT BsLipA ( Fig 3c; Table 2 ) . The predicted ΔTp values for these variants were similar to each other , in agreement with the similar T′50 values found , but at the lower end of all predicted ΔTp ( Table 2 ) . For the variants used as a negative control ( S1 Table ) [57] , the thermostability was quantified by T′50 values; these values report on the temperature of incubation for 20 min after which the fraction of the activity at room temperature to the initial activity is 50% . With respect to the T′50 values used above , a significant and very good correlation was obtained for T′50 ( see S1 Text ) For nine out of ten variants , significantly lower thermostabilities were measured , with the largest decrease being 7 . 3°C for the N48R variant ( S1 Table ) . The three thermostable variants involve mutations at weak spots identified at later phase transitions T4 and T5 during the thermal unfolding simulation . This finding supports our previous reasoning that it is the late phase transition ( s ) involving the final decay of the rigid core during thermal unfolding that mostly determine ( s ) the thermodynamic thermostability of a protein [16 , 18 , 19] . Accordingly , mutations that strengthen contacts of weak spot residues identified at late phase transitions should particularly improve thermostability . A sound discussion of this implication requires X-ray structural data of the variants , which is not yet available . Still , using the modeled variant structures , we observed that the three variants V54H , F58I , and V96S do have in general stronger “rigid contacts” between neighboring residues than the WT ( a “rigid contact” denotes that two residues belong to one rigid cluster ) : On average , the mutations V54H , F58I , and V96S increased the strength of rigid contacts of neighboring residues by 2 . 0 , 1 . 2 , and 0 . 4 K , respectively , compared to WT ( S4 Fig; see section “Constraint Network Analysis: Local rigidity indices” for an explanation how these values were calculated ) . Considering the most thermostable variant F58I in more detail , the strengthening holds true for local contacts as well as contacts that arise from a long-range stabilization . As to local contacts , Ile at position 58 along with residues of the neighboring loop β4-αB ( A38 , V39 , D40 ) are part of a rigid cluster , which persists to a temperature ~3 K higher than the rigid cluster formed by F58 of WT and the same loop residues ( Figs 4a , 4b , S4b and S6a ) . The persistence at higher temperature results from a better side-chain packing ( Fig 4c ) . In particular , in variant F58I , V39 forms four hydrophobic contacts with three different residues ( V7 , S16 , F41 ) , whereas in WT it only forms two such hydrophobic contacts ( Fig 4c ) . However , not all F58I mutation-induced changes lead to stabilization ( Fig 4d ) . As to contacts that arise from a long-range stabilization , residues of several pairs of secondary structure ( αA/β strands 3 , 4 , 5; αB/αC; loop αB-β5/loop αC-β6; loop αC-β6/loop αD-β7 ) remain part of one rigid cluster for temperatures 2–5 K higher in the variant F58I than in WT ( Figs 4d , S4b and S5b–S5e ) . This demonstrates the inherent long-range aspect to rigidity percolation [23 , 45 , 58–60] , i . e . , a local change on one end of a network can affect the stability all across the network . Recently , we described the unfolding pathway of BsLipA in detail as deduced from thermal unfolding simulations [61] ( see also page 9 , Fig 3 in that publication ) . We observed that α-helices αD and αE first segregate to form individual small rigid clusters , followed by αA and αF . The giant rigid cluster at this temperature is formed by the central β-sheet region and the two helices αB and αC . Next , the β-sheet region becomes sequentially flexible , beginning with β4 and β8 , followed by the remaining β-strands in the order β3 , β7 , and β5−β6 , finally leading to a completely flexible β-sheet region . As described above , several of these secondary structural elements are involved in the thermal stabilization of the variant F58I ( S4 Fig ) . Furthermore , the amino acids forming the catalytic triad in BsLipA are S77 located between strand β5 and helix αC , D133 between strand β7 and helix αE , and H156 between strand β8 and helix αF [62] . Stabilization of these secondary structural elements due to introducing mutation F58I ( S4 Fig; in particular , loops αB-β5 and αC-β6 ( S5d Fig ) , loops αC-β6 and αD-β7 ( S5e Fig ) , and helices αB and αC ( S5c Fig ) ) may thus delay the unfolding of the active site . Five mutations at weak spots identified at transitions T4 and T5 resulted in lower T′50 values than that of WT BsLipA ( Table 2 ) . This result appears to contradict our reasoning that mutations which strengthen connections of weak spot residues identified at late phase transitions should particularly improve thermostability . In each case , however , a small amino acid was substituted by a large amino acid , which likely could not be accommodated by the fold . This calls for improved modeling considering backbone relaxation [63] for variant construction in future studies with the aim to improve discrimination between amino acid substitutions in already densely packed regions , which could not accommodate small-to-large residue mutations , and substitutions in the vicinity of a protein cavity , where small-to-large residue mutations are an established strategy to increase protein stability [39 , 64] . Along the same lines , the two variants G104I and G104L out of the three variants that showed a nearly complete loss of activity at room temperature , and no residual activity after 30 min incubation at temperatures between 40–60°C , involved a residue located in the active site . While at the opposite side of the catalytic triad , introducing larger residues may occlude the substrate binding region . Such weak spots can be filtered out in future studies based on their location in the protein [65] .
We developed a novel rational approach based on increasing structural rigidity for improving a protein’s thermostability and applied it prospectively to BsLipA . The approach combines ensemble- and rigidity theory-based weak spot prediction by CNA , filtering of weak spots according to sequence conservation , computational site saturation mutagenesis , assessment of variant structures with respect to their structural quality , and screening of the variants for increased structural rigidity by ensemble-based CNA . Two reasons account for its high computational efficiency: In the first step , the number of potential mutation sites is dramatically reduced due to concentrating only on structural weak spots . In the second step , the use of ensembles of network topologies , rather than structural ensembles , alleviates the need for costly conformation sampling . As a result , about one mutation per hour can be processed on one core once weak spots have been detected ( Table 4 ) ; this task is trivially parallelizable for multiple mutations . From a methodological point of view , this majorly distinguishes our approach from other state-of-the art methods for predicting effects of mutations on protein stability [27–33] in that these methods would need to consider all potential mutation sites due to the lack of an equivalent “step one” . Furthermore , these methods either do not consider ensemble representations of the protein [28–33] or use structural ensembles [27] . Finally , our approach does not require weighting or fitting parameters , in contrast to other methods [27 , 30 , 31 , 66] . As to the application to BsLipA , our approach resulted in three out of twelve experimentally tested single-point mutations with significantly increased thermostability with respect to WT , yielding 6 . 6°C as the largest increase . This increase compares favorably to the median increase in the apparent melting temperature of 8°C found for 93 cases of engineered proteins , most of which contain more than one mutation [67] . Considering all tested single-point mutations , our approach yielded a success rate as to significantly increased thermostability of 25% , which raises to 60% if the five small-to-large residue mutations and the two mutations in the active site are excluded . These success rates are markedly higher than the 5% of mutations showing an increase in protein stability found within 1285 variants of ten different proteins [68 , 69] . It is also instructive to compare our results to those obtained by testing a complete site saturation mutagenesis library of BsLipA for improved detergent tolerance , where the success rate amounts to 2% [57] . Furthermore , for state-of-the-art methods for predicting the sign of stability change due to a mutation , impressive accuracies of over 80% have been reported [28] . These values result from the methods being very good at predicting destabilizing mutations and the prevalence of such mutations in the investigated data sets [28] . In line with this , for our predicted negative controls , we found a success rate as to significantly decreased thermostability of 90% . In contrast , the methods’ performances are much worse in predicting stabilizing mutations , yielding an average success rate for such mutations of 36% over 12 methods [28] . Evaluating the performance of our approach as a binary classifier [70] ( S2 Table ) , our approach discriminates between mutations leading to increased thermostability versus those leading to decreased thermostability with a sensitivity of 83% , a specificity of 56% , and an accuracy of 63% considering all variants in Tables 2 and S1 , and a sensitivity of 100% , a specificity of 77% , and an accuracy of 83% if the small-to-large residue mutations and the two mutations in the active site are excluded . In our view , this signifies that our approach provides for a robust binary classifier . Our approach has a precision ( predictive value for increased thermostability ) of 42% ( 60% if the small-to-large residue mutations and the two mutations in the active site are excluded ) ( S2 Table ) , which leads to a gain in precision with respect to a random classifier of a factor of 1 . 6 ( 2 . 4 ) . Furthermore , a Mann–Whitney U test [71] demonstrates that predicted positive ΔTp significantly points to increased experimental thermostability ( p < 0 . 05 ) ( see S2 Text ) . An approach related to CNA is the distance constraint model ( DCM ) [72] , which reaches average percent errors of 1 . 1% ( Pearson correlation coefficient R = 0 . 72 ) [73] and 4 . 3% ( R = 0 . 64 ) [74] for melting point predictions of protein variants with single and multiple mutations , corresponding to an error of ~4 K [73] and ~14 K [74] . This model requires a system-specific fitting to experimental heat capacity curves from differential scanning calorimetry , however [73 , 74] . Over all variants predicted ( including the negative controls but excluding the three variants for which no activity could be measured ( Table 2 ) ) , our approach , which does not require fitting parameters , yields a significant ( R = 0 . 48 , p = 0 . 02 ) correlation between predicted and experimental thermostabilities; if small-to-large residue mutations and the two mutations in the active site are excluded , the correlation improves further ( R = 0 . 62 , p = 0 . 02; S6 Fig ) . These results show that our approach can reproduce experimental trends with sufficient accuracy . The effectiveness of our approach is also demonstrated when comparing it to the study by Reetz and coworkers [15] applying iterative saturation mutagenesis to BsLipA . The largest increase in T50 they have found for a variant containing a single point mutation in the first step was 4 . 3°C; our largest increase of 6 . 6°C compares favorably to this value . Four more steps of optimization and screening of about 8000 colonies then yielded two variants carrying five and seven mutations that showed an increase of T50 by 45°C . The study of Reetz et al . also differs from ours in a fundamental aspect: in the former study , those residues that showed the highest crystallographic B-factors , i . e . , were the most mobile , were chosen as weak spots . In our study , weak spots constitute residues that segregate from large rigid , i . e . internally immobile , clusters during thermal unfolding . In summary , these results suggest that our approach is a valuable , orthogonal complement to existing methods for rational protein design aimed at improving thermostability . The more thermostable variants can serve as starting points for further engineering of substrate scope and/or enantioselectivity by directed evolution , exploiting that enhanced thermostability promotes the ease of evolvability [75] .
Constraint Network Analysis ( CNA ) predicts rigid and flexible regions within a biomolecule , which allows linking these static characteristics to the molecule’s stability and function [17 , 21] . CNA has been described in detail in refs . [17 , 21 , 35 , 76] . The approach has been used previously to predict the ( thermodynamic ) thermostability of proteins and to identify weak spot residues that , when mutated , are likely to improve thermostability [16 , 18 , 19] . In CNA , a protein is modeled as a body-and-bar network of bodies ( atoms ) and bars ( covalent and noncovalent interactions ) . Each atom has six degrees of freedom , and each bar removes one degree of freedom [22] . An interaction between two atoms can be modeled as any number of bars between one and six depending on the strength of the interaction . Here , single covalent bonds ( double and peptide bonds ) were modeled as five ( six ) bars , hydrogen bonds and salt bridges ( together referred to as “hydrogen bonds” ) as five bars , and hydrophobic interactions as two bars . For hydrogen bonds a hydrogen bond energy EHB is computed by a modified version of the potential by Mayo and coworkers [77] as described in ref . [26] . By successively removing noncovalent constraints from a network , a thermal unfolding of the protein is simulated [16 , 18 , 19 , 26] . Hydrogen bonds are removed from the network in increasing order of their strength [77] , i . e . , hydrogen bonds with an energy EHB > Ecut ( σ ) are discarded from the network of state σ . In the present study , Ecut values ranging from −0 . 1 kcal mol−1 to −6 . 0 kcal mol−1 with a step size of 0 . 1 kcal mol−1 were used . Ecut can be converted to a temperature using a linear relation introduced by Radestock and Gohlke [16 , 18] , according to which the range of Ecut used in this study is equivalent to increasing the temperature of the system from 302 K to 420 K with a step size of 2 K . The rigidity of each network state σ during the thermal unfolding simulation is analyzed by the pebble game algorithm [23 , 24] as implemented in the FIRST program [25] . From these analyses , the change in the global rigidity characteristics is monitored by the cluster configuration entropy Htype2 [76] . Finally , a phase transition temperature Tp is identified as the temperature when a largely rigid network becomes largely flexible . We showed that Tp can be used for predicting the thermodynamic thermostability of and identifying structural weak spots in a protein [16 , 18 , 19] . Usually , multiple phase transitions occur during the thermal unfolding of a protein because of its modular architecture , i . e . , secondary structure elements can segregate from the largest rigid cluster as a whole [18] . In contrast to global indices , local indices monitor rigidity at a residue level . One such index , the rigidity index ri , is defined for each covalent bond i between two atoms as the Ecut value during the thermal unfolding simulation at which the bond changes from rigid to flexible [76] . For a Cα atom-based representation , the average of the two ri values of the two backbone bonds is taken . As a two-dimensional itemization of ri , a stability map rcij indicates for all residue pairs the Ecut value at which a rigid contact between the two residues i , j is lost , i . e . , when the two residues stop belonging to the same rigid cluster [76] . From rcij , a rigid cluster decomposition , i . e . , a set of rigid clusters and flexible links in between , can be computed for each network state σ during the thermal unfolding simulation . When the stability map rcij is filtered such that only rigid contacts between residues that are at most 5 Å apart from each other ( measured as the distance between the closest atom pair of the two residues ) are considered , a neighbor stability map results . This map helps focusing on short-range rigid contacts that can be directly modulated by mutagenesis with the aim to stabilize them for improving the overall stability of a protein . In this study we use neighbor stability maps to analyze the ( local ) effect of mutations on the stability of rigid contacts of neighboring residues ( S4 Fig ) . The increase in the strength of rigid contacts is calculated as the average over differences in rcij of the variant versus WT for all neighboring residue pairs ( lower triangles in S4 Fig ) . The increase in the strength is measured in K . Rigidity analyses using CNA are sensitive with respect to the input structure [45 , 78] . One way to improve the robustness is to carry out CNA on a structural ensemble derived from molecular dynamics ( MD ) simulations; then results ( Tp values and stability maps ) are averaged [19] . In the present study , MD simulations of WT BsLipA were performed using the GPU accelerated version of PMEMD [79] of the AMBER 11 suite of programs [80 , 81] together with the ff99SB force field [82] . The X-ray crystal structure of BsLipA with the highest resolution ( PDB ID: 1ISP; resolution 1 . 3 Å ) was used as input structure [83] . Hydrogen atoms were added using the REDUCE program [84] during which side-chains of Asn , Gln , and His were flipped if necessary to optimize the hydrogen bond network . Then , the system , neutralized by addition of sodium counter-ions , was solvated by a truncated octahedral box of TIP3P [85] water such that a layer of water molecules of at least 11 Å width covers the protein surface . The particle mesh Ewald method [86] was used with a direct-space non-bonded cutoff of 8 Å . Bond lengths involving hydrogen atoms were constrained using the SHAKE algorithm [87] , and the time step for the simulation was 2 fs . After equilibration , a production run of unrestrained MD in the canonical ensemble ( NVT ) was performed to generate a trajectory of 100 ns length , with conformations extracted every 40 ps from the last 80 ns resulting in a structural ensemble of 2000 conformations . The ensemble was used to predict weak spot residues on BsLipA . According to our strategy ( Fig 1 ) , the structural ensemble of 2000 conformations of WT BsLipA ( Fig 1a ) was initially submitted to CNA for weak spot identification and prioritization . An average stability map was generated from individual stability maps for each conformation in the ensemble . A thermal unfolding trajectory showing average rigid cluster decompositions during the thermal unfolding simulation was reconstructed from the average stability map ( Figs 1b and S7 ) . For this we exploited that rigid cluster decompositions can be reconstructed from stability maps as the latter store Ecut ( or temperature , according to the above mentioned linear relation ) values for all residue pairs at which these residues cease to be in one rigid cluster during the thermal unfolding . The thermal unfolding trajectory was visually inspected for identifying transitions at which the rigidity of WT BsLipA is substantially reduced using VisualCNA [88] . The inspection was done with a view that rigidifying contacts between the largest rigid cluster and residues that segregate at these substantial phase transitions should improve the thermostability of the protein by delaying the disintegration of the largest rigid cluster . Accordingly , at every transition , residues that are in the neighborhood of , and whose side-chains point towards the largest rigid cluster from which they segregated , were identified as potential weak spots ( Fig 1c ) . Weak spot residues that showed a high sequence conservation ( ≥ 80% identity ) in a multiple sequence alignment of 296 lipase class 2 sequences obtained from the Pfam database [89] were not considered further ( Fig 1d ) . Next , for modeling single-point site-saturation mutations , structures of all possible mutations at each weak spot residue were generated by the SCWRL program [50] using WT BsLipA ( PDB ID: 1ISP ) as a template ( Fig 1e ) . Conformations of side-chains of all residues within 8 Å of a mutated residue were re-predicted in order to allow for a local structural relaxation . The goodness of fit of the mutated side-chain in its environment was assessed using the ANOLEA server [51 , 52] . A variant was discarded if its average ANOLEA energy of the neighboring residues ( ≤ 5 Å of the mutation ) is higher than the average energy of the same residues in WT by ≥ 2 kcal mol-1 . For all variant structures , hydrogen atoms were added using REDUCE [84] in an identical way as done for WT BsLipA ( see section “Generation of a structural ensemble of WT BsLipA” for details ) . The structures were minimized by 2000 steps of conjugate gradient minimization ( including an initial steepest descent minimization for 100 steps ) or until the root mean-square gradient of the energy was < 1 . 0*10-4 kcal mol-1 Å-1 . The energy minimization was carried out with Amber11 [80] using the ff99SB force field [82] and the GBOBC generalized Born model [90] . Finally , the generated variant structures were subjected to thermostability prediction and prioritization . In order to circumvent compute-intensive MD simulations for generating structural ensembles of each of the BsLipA variants , the more efficient ENTFNC approach [35] was used in connection with thermal unfolding simulations by CNA [21] . Ensembles of 1000 network topologies of all single point variants of BsLipA were analyzed; for consistency , the WT BsLipA structure was treated in the same way ( including an energy minimization as described above ) . For each variant and WT , Tp was identified as the inflection point of the sigmoid with the larger change in Htype2 using a double sigmoid function [19] fitted to Htype2 vs . T curves . That way , in most cases , a late transition involving the final decay of the largest rigid cluster is identified as Tp [18] except when a very large loss of rigidity occurs during an early transition . Based on ensemble-averaged Tp ( Fig 1f ) , variants were selected for experimental characterization of their thermostability and Michaelis-Menten kinetics . See S1 Text for details . Table 4 summarizes the required computing times . | Protein thermostability is a crucial factor for biotechnological enzyme applications . However , performance studies of computational approaches for predicting effects of mutations on protein ( thermo ) stability have suggested that there is still room for improvement . We describe and validate a novel and unique strategy to predict optimal amino acid substitutions at structural weak spots . At variance with other rational approaches , we exploit the fact that in the majority of cases an increased structural rigidity of the folded state is the underlying cause for thermostability . When applied prospectively on lipase LipA from Bacillus subtilis , a high success rate in predicting thermostabilized lipase variants and a remarkably large increase in their thermostability is achieved . This demonstrates the value of the novel strategy , which extends the existing portfolio of methods for rational protein design . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"chemical",
"bonding",
"chemical",
"compounds",
"enzymes",
"enzymology",
"organic",
"compounds",
"mutation",
"substitution",
"mutation",
"amino",
"acid",
"substitution",
"network",
"analysis",
"amino",
"acids",
"hydrogen",
"bonding",
"physical",
"chemistry",
"computer",
"and",
"information",
"sciences",
"proteins",
"chemistry",
"lipases",
"mutagenesis",
"biochemistry",
"biochemical",
"simulations",
"point",
"mutation",
"hydrolases",
"organic",
"chemistry",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology"
] | 2016 | Application of Rigidity Theory to the Thermostabilization of Lipase A from Bacillus subtilis |
Proteins secreted by pathogens during host colonization largely determine the outcome of pathogen-host interactions and are commonly called ‘effectors’ . In fungal plant pathogens , coordinated transcriptional up-regulation of effector genes is a key feature of pathogenesis and effectors are often encoded in genomic regions with distinct repeat content , histone code and rate of evolution . In the tomato pathogen Fusarium oxysporum f . sp . lycopersici ( Fol ) , effector genes reside on one of four accessory chromosomes , known as the ‘pathogenicity’ chromosome , which can be exchanged between strains through horizontal transfer . The three other accessory chromosomes in the Fol reference strain may also be important for virulence towards tomato . Expression of effector genes in Fol is highly up-regulated upon infection and requires Sge1 , a transcription factor encoded on the core genome . Interestingly , the pathogenicity chromosome itself contains 13 predicted transcription factor genes and for all except one , there is a homolog on the core genome . We determined DNA binding specificity for nine transcription factors using oligonucleotide arrays . The binding sites for homologous transcription factors were highly similar , suggesting that extensive neofunctionalization of DNA binding specificity has not occurred . Several DNA binding sites are enriched on accessory chromosomes , and expression of FTF1 , its core homolog FTF2 and SGE1 from a constitutive promoter can induce expression of effector genes . The DNA binding sites of only these three transcription factors are enriched among genes up-regulated during infection . We further show that Ftf1 , Ftf2 and Sge1 can activate transcription from their binding sites in yeast . RNAseq analysis revealed that in strains with constitutive expression of FTF1 , FTF2 or SGE1 , expression of a similar set of plant-responsive genes on the pathogenicity chromosome is induced , including most effector genes . We conclude that the Fol pathogenicity chromosome may be partially transcriptionally autonomous , but there are also extensive transcriptional connections between core and accessory chromosomes .
Plant pathogenic fungi are genetically adapted to infect their host plant , but are also in a constant arms race with that host to stay virulent . For this , pathogens need to allow accelerated evolution of pathogenicity-related genes , without affecting the function of housekeeping genes . One possibility is to spatially separate these different functional groups of genes into subgenomic compartments with different rates of evolution . One of the fastest evolving determinants of pathogenicity is the effector repertoire . Definitions vary , but a typical effector is a small , secreted protein that affects the interaction between the pathogen and its host . In many plant pathogenic fungi effector genes indeed reside in specific genomic regions , generally distinguished by one or more of the following characteristics: lineage-specific or accessory , rich in transposable elements , a different GC content and/or codon bias from the rest of the genome , depleted for housekeeping genes and associated with particular chromatin modifications [1–3] . Accumulating evidence suggests that these genomic environments evolve more rapidly than the rest of the genome and facilitate adaptation [2 , 4 , 5] . In addition , in several fungi these types of regions—or parts thereof—have been shown or suggested to transfer horizontally between different strains or even species [6–11] . Another hallmark of effector genes is a plant specific expression pattern [12–14] . How coordinated expression of effector genes is regulated and how this is related to the specific genomic environment of these genes is poorly understood . For some effector genes it has been shown that their genomic environment is key for regulated expression , through histone modifications [1 , 15] . On the other hand , a small number of transcription factors required for effector gene expression has been identified . In Ustilago maydis the heterodimer bE/bW , the forkhead transcription factor Fox1 , and the zinc finger transcription factors Rbf1 and Mzr1 are involved in transcriptional regulation of pathogenicity-related genes and/or effector genes [16–19] . In Leptosphaeria maculans and Stagnospora nodorum , homologs of StuA , a bHLH ( basic helix-loop-helix ) type of transcription factor , regulate expression of several effector genes [20 , 21] . Most is known about the role of Wor1 orthologs in effector gene expression . Wor1 is a conserved fungal transcription factor from Candida albicans , with a WOPR type of DNA binding domain [22] . In plant pathogenic fungi from the genus Fusarium ( putative ) effector genes and/or secondary metabolite gene clusters are regulated by an ortholog of this transcription factor: F . oxysporum f . sp . lycopersici ( Sge1 ) , F . graminearum ( Fgp1 ) , F . verticillioides ( FvSge1 ) and F . fujikuroi ( FfSge1 ) [23–26] . Also in the plant pathogenic fungi Botrytis cinerea ( Reg1 ) , Verticillium dahliae ( VdSge1 ) , Cladosporium fulvum ( CfWor1 ) , Zymoseptoria tritici ( ZtWor1 ) , Ustilago maydis ( UmRos1 ) and Magnaporthe oryzae ( MoGti1 ) deletion of the gene for this transcription factor ( partially ) perturbs expression of effector genes [27–32] . Mutant strains deleted for this gene are mostly non-pathogenic ( except Δffsge1 ) , although for CfWOR1 this may be a secondary effect of a developmental phenotype . In C . albicans , Wor1 was originally discovered as a ‘master regulator’ of the morphological switch from white to opaque cells . Also in Saccharomyces cerevisiae and Histoplasma capsulatum the Wor1 orthologs ( Mit1 and Ryp1 , respectively ) regulate a morphological transition , which , both in C . albicans and H . capsulatum , is associated with differences in virulence towards humans [33] . This led to the idea that Wor1 orthologs in plant pathogenic fungi are also master regulators of a lifestyle switch , from saprotrophic to pathogenic . In Fusarium oxysporum f . sp . lycopersici ( Fol ) , the causal agent of Fusarium wilt in tomato , effector genes ( called SIX genes for ‘Secreted In Xylem’ ) reside on an accessory chromosome that can be transferred horizontally between strains . Upon receipt of this accessory chromosome of Fol , a non-pathogenic strain can acquire pathogenicity towards tomato [7] . This means that effector gene expression must be ensured in different genomic environments ( i . e . in the original strain and in the recipient strain ) . Two different but not mutually exclusive strategies to ensure effector gene expression are: i ) to rely on conserved transcription factors encoded on the core genome or ii ) to encode the transcription factors necessary for effector gene expression on the accessory chromosome itself . As mentioned above , Fol effector gene expression requires the presence of the core-encoded conserved transcription factor Sge1 [26] . However , also on the accessory chromosome transcription factors are encoded , 13 in total [34] . One of these transcription factor genes , FTF1 , is associated with highly virulent strains of F . oxysporum f . sp . phaseoli and is up-regulated during infection [35 , 36] . In addition , FTF1 is present in three variants on the Fol pathogenicity chromosome and all three genes are located close to single or small groups of effector genes [34] . Although the pathogenicity chromosome of Fol is transcriptionally connected to the core genome via Sge1 , the presence of numerous transcription factor genes on the chromosome itself suggests that this accessory chromosome might be transcriptionally semi-autonomous . To see whether this may be the case , we investigated the role of the transcription factors encoded on the pathogenicity chromosome of Fol in effector gene expression .
To see if the pathogenicity chromosome of Fol ( chromosome 14 in reference strain Fol4287 ) may be transcriptionally autonomous , we inventoried the transcription factors it encodes . We found 13 predicted transcription factor genes that cluster into nine families . The transcription factor gene families were numbered TF1 to TF9 and include one homolog of EBR1 ( TF8 ) and three homologs of FTF1 ( TF1 ) ( Fig 1 , S1 Data: tab ‘TF table’ ) [35 , 37] . Most gene families encode proteins containing zinc finger DNA binding domains; four are Cys2His2 zinc finger DNA binding domains ( Tf3 , Tf4 , Tf6 and Tf7 ) and two are Zn ( 2 ) Cys ( 6 ) zinc finger DNA binding domains ( Tf1 and Tf8 ) . Additionally , there are two gene families encoding transcription factors with a basic leucine zipper ( bZIP ) DNA binding domain ( Tf5 and Tf9 ) and one gene family encoding forkhead transcription factors ( Tf2 ) . All transcription factor genes on the pathogenicity chromosome have a homolog on the core genome , except TF3 ( Fig 1 ) . Four of the transcription factor gene families have also expanded on other accessory chromosomes of Fol4287 ( TF1 , TF7 , TF8 and TF9 ) . The F . oxysporum species complex encompasses many different formae speciales ( ff . spp . ) , each with a specific plant host and specific accessory genomic material . All transcription factor gene families ( except TF3 ) have one core- and one or more accessory-encoded homologs in other ff . spp . of F . oxysporum investigated . ( Fig 2 , S1 Fig , S1 Data:tab ‘TF table’ ) . From here on , we refer to accessory homologs as aTF and to core homologs as cTF . The other Fusarium species analysed ( F . solani , F . graminearum and F . verticillioides ) each have only one homolog of most of the transcription factors; the expansion we see on the accessory regions in F . oxysporum has not occurred . Regardless of the forma specialis , core-encoded transcription factors form one clade ( Fig 2 , S1 Fig , indicated with a grey bar ) and show little sequence divergence . In general , the F . verticillioides homologs are closest to this clade , consistent with the species phylogeny . Accessory chromosome-encoded homologs show more divergence , both within and between strains , and are also more diverse in sequence than the core encoded homologs in F . oxysporum and F . verticillioides . This suggests that either i ) the expansion and divergence of the transcription factor genes on the accessory chromosomes is older than the separation between F . oxysporum and F . verticillioides , similar to what was suggested for accessory genes in general [7] , or ii ) the rate of evolution is accelerated in the accessory regions compared to the core genome , or iii ) a combination of the two . As transcription factor gene families have expanded and diverged on the accessory chromosomes , we wanted to investigate whether some of these genes may have neofunctionalized . To see if homologous core- and accessory-encoded transcription factors regulate different target genes , we set out to determine the DNA binding sites of each transcription factor encoded on the pathogenicity chromosome and its core-encoded homolog . Transcription factor coding sequences were cloned from cDNA , in vitro translated as a GST-fusion and hybridised with two different oligo arrays ( called HK & ME ) [38] . Binding enrichments were inferred for each possible 8-mer . Of 13 transcription factors the cDNA could be amplified and cloned . For aTf1 , aTf8 and aTf9 , cloning of any of the homologs on the pathogenicity chromosome was unsuccessful . Possibly this was due to both low transcript levels and the presence of homologous transcripts; only hybrid PCR products were amplified . However , a partial cDNA encoding the predicted DNA binding domain from another aTf1 homolog was obtained ( FOXG_17084 on accessory chromosome 6 ) . Tf1 homologs separate in two groups based on the length of the coding sequence . The first group ( ~3200 bp , we refer to this as the longer coding sequence ) include the core homolog , two homologs on the pathogenicity chromosome and two identical genes on other accessory chromosomes . The remaining aTF1 genes are shorter ( ~2800 bp ) , because they have a more downstream startcodon and a more upstream stopcodon ( S2 Fig ) [35] . All Tf1 homologs have highly similar DNA binding domains ( S3 Fig ) . The cloned aTF1 cDNA ( FOXG_17084 ) has a short coding sequence . For nine of the cloned transcription factors a reliable DNA binding site could be inferred from one or both arrays ( Fig 3A , S1 Data: tab ‘DNA binding assay’ ) . In all cases both arrays yielded similar top 8-mers . For the remaining four transcription factors no significant enrichment was found . The DNA binding sites of homologous transcription factors were the same or very similar in all four families for which both homologs yielded a DNA binding site , indicating that no diversification in recognition specificity has occurred . The DNA binding site of aTf2 and its homolog cTf2 overlap with the consensus DNA binding site of forkhead transcription factors ( RYMAAYA ) [39] . The aTF5/cTF5 DNA binding site is almost palindromic , which is in accordance with the dimeric structure of leucine zippers . Tf1 has a Gal4-like Zn ( 2 ) Cys ( 6 ) DNA binding domain , which in Gal4 binds as a dimer to the CGGN11CCG consensus sequence [40] . The DNA binding site of aTf1 and cTf1 ( TRCCG ) overlaps with half of this consensus . Interestingly , the aTf1 DNA binding site overlaps with a motif found earlier to be enriched in the promoters of effector genes: aacTGCCGa [34] . Note that the top 8-mer for both aTf1 and cTf1 is perfectly contained in this motif ( Fig 3A , S1 Data: tab ‘DNA binding assay’ ) . We attempted to determine the DNA binding site of Sge1 in a similar way , but no significant 8-mer enrichments were detected . Sge1 has an unusual fungi-specific WOPR type DNA binding domain . The DNA binding site and protein crystal structure of several homologs of Sge1 have been resolved [33 , 41–43] . The amino acids interacting with DNA are highly conserved among Wor1/Sge1 homologs from different fungi [23 , 27 , 29 , 30 , 41 , 43 , 44] . Consistent with this , the DNA binding sites of Sge1 orthologs in Saccharomyces cerevisiae ( Mit1 ) and the filamentous fungus Histoplasma capsulatum ( Ryp1 ) are the same as for Wor1 [33] . To test whether Sge1 can also bind to the Wor1-DNA binding site , we have adopted an in vivo transcriptional activation assay developed previously [42] . In short , Wor1 or Sge1 is produced constitutively in yeast together with a reporter construct consisting of the Wor1-DNA binding site and an UAS-less CYC1 promoter fused to the lacZ gene . The two SGE1 homologs in yeast ( YEL007 or Mit1 and YHR177 ) are deleted from the yeast strain to avoid potential cross-activation . We found that Sge1 can induce the reporter gene to the same level as Wor1 , and this ability is lost when the Wor1 DNA-binding site is mutated ( Fig 3B ) . This confirms that Sge1 binds the same DNA sequence as its orthologs and is a transcriptional activator . Determination of the DNA binding site of several transcription factors showed that DNA binding specificity has not or hardly diverged between accessory-encoded and core-encoded transcription factors . However , if accessory-encoded transcription factor homologs fulfill an important role in transcriptional regulation of the accessory chromosomes , target genes of these transcription factors could be enriched there . To test this , the number of genes with minimally one , two or three binding sites in 1 kb upstream of the annotated transcriptional start site was counted ( Fusarium Comparative Sequencing Project , Broad Institute of Harvard and MIT ( http://www . broadinstitute . org/ ) , annotation 3 ) ( S1 Data: tab ‘promoter DBS’ ) . Because some of the sequence motifs contain relatively little information we found many occurrences throughout the genome . Nevertheless , for several DNA binding sites ( of aTf1 , cTf4 and aTf5 ) significant enrichment ( hypergeometric test , P value < 0 . 01 after Bonferroni correction ) was found on accessory chromosomes , and for the aTf2 DNA binding site enrichment was found on the core genome ( Fig 3C , S1 Data: tab ‘promoter DBS’ ) . Specific enrichment of DNA binding sites on the pathogenicity chromosome , however , was not detected . In an effort to reduce noise levels for the smaller sequence motifs we also looked at multiple occurrences , and indeed the observed enrichments were also present under these criteria . Interestingly , also the Sge1 DNA binding site is enriched on the accessory chromosomes . To test whether the observed enrichments are specific for promoter regions , we performed the same test for the 1000 bp downstream the ATG of each gene . We found no DNA binding site enrichments in the coding regions , except for the aTf1 DNA binding site , which was found enriched both in the promoter and in the coding regions of accessory genes ( S4A and S4B Fig , S1 Data: tab ‘ORF DBS’ ) . Since the pathogenicity chromosome and the effectors encoded on it are important for the infection of tomato plants , enrichment of DNA binding sites among genes that are up-regulated during infection and those that are down-regulated was also tested [45] . Only the aTf1 and the Sge1 DNA binding sites were significantly enriched among up-regulated genes , and none of the DNA binding sites was enriched among genes down-regulated during infection ( Fig 3C , S1 Data ) . Taken together , this statistical analysis suggests that several transcription factor gene families on the pathogenicity chromosome may be involved in regulating targets on accessory chromosomes . In addition , aTf1 and/or cTf1 and Sge1 in particular may target genes involved in pathogenicity . The effectors encoded on the pathogenicity chromosome are important determinants for pathogenicity and are under strict transcriptional regulation [26 , 46–50] . To determine whether transcription factors encoded on the pathogenicity chromosome can induce effector gene expression , strains ectopically expressing these transcription factors from a constitutive promoter—from hereon called ‘overexpressors’–were generated and tested for their ability to induce effector gene expression . For seven out of the nine transcription factor families encoded on the pathogenicity chromosome we were able to make constructs with the constitutive FEM1 promoter [51] to drive expression ( cloning of aTF8 and aTF9 was unsuccessful ) . Three aTF1 homologs are present on the pathogenicity chromosome and one of these was selected for overexpression ( FOXG_17458 , long coding sequence ) . Also one ‘short’ aTF1 gene from another accessory region was chosen for overexpression ( FOXG_17084 ) , the same gene that was used for the DNA binding site determination . To facilitate screening for the induction of effector gene expression , transcription factor overexpression constructs were transformed into a strain carrying a reporter gene . In this Fol strain , the reporter ( GFP ) replaces the ORF of the effector gene SIX1 –a representative plant-induced effector gene–so that it is expressed in locus from the SIX1 promoter [46] . For each transcription factor , 11 to 23 independent transformants were inspected for increase of GFP expression using fluorescence microscopy ( S5A Fig ) . GFP signals varied between transformants with the same construct . The three to ten transformants that appeared to respond most strongly to each transcription factor were used to quantify GFP levels spectrophotometrically ( Fig 4 ) . Most accessory transcription factors tested did not affect SIX1 expression when expressed from the FEM1 promoter . In contrast , the long version of aTF1 very strongly induced SIX1 expression , and the shorter version conferred some induction . Also SGE1 , tested in the same way , can induce SIX1 expression , albeit to a lesser extent than aTF1 . To confirm overexpression of the transcription factor genes in the transformants that showed no GFP induction , transcript levels were determined with quantitative RT-PCR . In all cases an increase of transcription factor gene expression compared to the background strain was observed in at least one out of three transformants tested ( S5B Fig ) . Some of the strains expressing SGE1 or the long version of aTF1 from the FEM1 promoter showed slight growth retardation , and strains overexpressing aTF7 ( FOXG_14275 ) consistently grow slower on both rich and minimal medium ( S5C Fig ) . The same transformants were tested for their ability to infect tomato plants . Only aTF7 overexpression altered this ability; aTF7-overexpressors were less virulent , which may be explained by their retarded growth ( S5D Fig ) . To see whether the phenotypes caused by aTF1 or aTF7 overexpression could also be induced by overexpression of their core homologs , strains expressing cTF1 ( FOXG_09390 , long gene model ) or cTF7 ( FOXG_17774 ) from the FEM1 promoter were generated and tested as described above ( S5 Fig , Fig 4 ) . cTF1 can activate the SIX1 promoter almost to the levels of the long version of aTF1 , and cTF7 overexpressors have a similar growth retardation and reduced virulence phenotype as those overexpressing aTF7 . This suggests a similar function for core and accessory homologs at least in these two cases . Since expression of aTF1 and cTF1 from the FEM1 promoter was shown to potently induce the SIX1 promoter , and the SIX1 promoter contains several aTf1 DNA binding sites , we proceeded to test the results of the DNA binding array in an independent system . For this we used the aTF1 homolog ( FOXG_17458 ) used for overexpression and cTF1 in the in vivo transcriptional activation assay described above for Sge1 . In this case , a short fragment was cloned from the promoter of the SIX1 effector gene . This sequence includes two Tf1-DNA binding motifs that overlap with the previously identified motif aacTGCCGa and are separated by 17 basepairs . We also constructed a version with mutated Tf1 DNA-binding sites ( two basepair substitutions; Fig 5 , lower panel ) . Both aTf1 and cTf1 are able to activate transcription from the construct with the wild type SIX1 promoter fragment , but not the mutated fragment ( Fig 5 ) . This assay was performed in the yeast strain lacking SGE1 homologs ( YEL007 or Mit1 and YHR177 ) , showing that DNA binding and transcriptional activation by aTf1 and cTf1 does not require ( a homolog of ) Sge1 . We found that only SGE1 , aTF1 and its core homolog cTF1 bind to motifs enriched in the promoters of plant-induced genes and genes on accessory chromosomes . Also , expression of any of these three transcription factor genes from the FEM1 promoter induces effector ( SIX1 ) gene expression . Of both SGE1 and cTF1 a role in pathogenicity has been established previously: the Fol Δsge1 mutant is no longer pathogenic and a Δctf1 ( Δftf2 ) mutant is reduced in pathogenicity [26 , 45] . Expression of all three transcription factor genes is increased upon infection of tomato plants [26 , 35 , 36] . To investigate which genes are regulated by these transcription factors , we compared the transcriptomes of SGE1 , aTF1 and cTF1 overexpressors with the transcriptome of the background strain . Overexpressors were preferred over gene deletion strains in this setup because effector genes are only expressed during colonization of the plant , and the plant signal that induces this up-regulation is unknown . Since the deletion mutants are not pathogenic or reduced in pathogenicity , expression cannot reliably be studied in planta with gene deletion strains either . Given that expression levels of SGE1 , aTF1 and cTF1 increase during infection , we assume that strains overexpressing either of these transcription factor genes partially mimic the in planta state . RNA was isolated from two-day old cultures growing in minimal medium . For each transcription factor two independent overexpressors were used , and for each strain three independent biological replicates were sampled . Per sample between 2 . 8*10^7 and 4 . 8*10^7 reads were paired-end sequenced ( Illumina ) . Reads were mapped to the reference genome ( Fol4287 ) and differentially expressed genes were called for each pairwise comparison with the background strain , using DEseq software [52] . The different pairs yielded between 396 and 691 differentially expressed genes ( S6A Fig ) . To compare the fold induction of expression of the overexpressed transcription factor genes to their induction during infection , we isolated RNA from Fol-infected tomato plants and from axenic cultures . Both transcriptomes of three biological replicates were sequenced and sequencing reads were mapped to the same annotation of the reference genome as above , using the same parameters . As can be seen in Fig 6 , expression of each of the three transcription factor genes is increased in the respective overexpressors as well as during infection . The increase in expression is significant ( p<0 . 05 ) for all three transcription factors in their respective overexpressors , but during infection only for aTF1 . Q-RT-PCR confirmed a significant increase of aTF1 expression in aTF1 overexpressors and during infection , but could not confirm significant differences for cTF1 and SGE1 transcript abundance ( S7 Fig ) . Our inability to measure a ( significant ) increase of cTF1 and SGE1 transcripts above their basal expression level by Q-RT PCR was unexpected . SIX1 ( GFP ) induction in the independent cTF1 and SGE1 overexpressors may be caused by an only modest increase in transcript levels , as suggested by our RNAseq analysis . For the aTF1 overexpressors , expression of both SIX1 ( GFP ) and another effector gene ( SIX3 ) does correlate well with aTF1 levels ( S8 Fig ) . Also , only in transformants expressing aTF1 , cTF1 or SGE1 from the FEM1 promoter did we ever observe high GFP levels in the Psix1:GFP strain ( Fig 4 , S5A Fig ) . Still , we wished to exclude the possibility that the transcriptome changes in the aTF1 , cTF1 and SGE1 overexpressors could be due to spontaneous changes unrelated to the transcription factor . We therefore clustered gene expression data of significantly differentially expressed genes for all six overexpressors ( S8 Fig ) . Both aTF1 and both cTF1 overexpressors cluster together , showing that genome wide changes in gene expression in these strains are related to the overexpressed transcription factor . The SGE1 overexpressors , however , do not form a single clade , so we could not link the transcriptional changes in these strains to SGE1 with this approach . To further test the connection between the selected transcription factors and the changes in the transcriptome , we determined whether the DNA binding site of each transcription factor is enriched in the upstream regions of the genes that are differentially expressed in the aTF1 , cTF1 or SGE1 overexpressors . For this we only considered those genes that were significantly up- or down-regulated in both overexpressors of each transcription factor to be consistent transcription factor-specific effects and this selection was used in all further analysis . Under these criteria , each transcription factor up-regulates 65 to 116 genes , and down-regulates 273 to 347 genes , when overexpressed ( S6B Fig ) . We have analysed single occurrence of the Sge1 DNA binding site , triple occurrence of the ( relatively short ) Tf1 DNA binding site , and single occurrence of the longer motif found in effector gene promoters ( that overlaps with the Tf1 DNA binding site ) for enrichment in 1kb promoter regions or coding regions ( Fig 7 , S4C &S4D Fig , S1 Data: tab ‘promoter DBS’ and ‘ORF DBS’ ) . Both the short and the long Tf1 DNA binding sites are significantly and specifically enriched in the promoters of genes up-regulated in both the aTF1 and cTF1 overexpressors , but not in the SGE1 overexpressor . The Sge1 DNA binding site , on the other hand , is specifically enriched in the promoters of SGE1- but also aTF1- and cTF1-up-regulated genes . This links the presence of DNA binding sites to changes in gene expression for all three transcription factors . For none of the down-regulated genes a significant enrichment of DNA binding sites was found , neither for down-regulated genes on the pathogenicity chromosome nor for down-regulated genes on the core chromosomes . This is consistent with the observation that all three transcription factors act as transcriptional activators . Finally , the enrichment of the Tf1 and Sge1 DNA binding site among three other groups of genes was tested ( Fig 7 ) . The three groups are: i ) genes coding for small secreted proteins; ii ) genes with a ( partial ) miniature impala ( MIMP ) , a non-autonomous transposable element , in the upstream region ( up to 2 kb from the ATG—all Fol effector genes are associated with a MIMP , but also other , mainly accessory genes , of which around half is also up-regulated during infection ( S6B Fig , [34] ) ) ; iii ) genes of which the protein has been detected in xylem sap of infected plants [34] . The Tf1 DNA binding site is not enriched among these categories , except the long effector motif among small secreted proteins . Remarkably , the Sge1 DNA binding site is enriched in all these categories ( Fig 7 , lower panel ) , suggesting that Sge1 may target pathogenicity-associated genes more specifically than aTf1 or cTf1 . We have shown that the genome-wide transcriptional changes in the aTF1 and cTF1 overexpressors are correlated with expression of the respective transcription factor from the FEM1 promoter , while the Sge1 binding site is enriched in upstream regions of genes upregulated in the SGE1 overexpressors and of pathogenicity-associated genes . We now compared the gene sets differentially expressed in the SGE1 , aTF1 and cTF1 overexpressors and during infection ( Fig 8A and S6B Fig ) . Strikingly , the majority of the plant-responsive genes on the pathogenicity chromosome is also up-regulated in all three overexpressors , including almost all SIX effector genes ( indicated in purple text in Fig 8B ) . The SIX genes that are not induced are SIX13 in all overexpressors and SIX2 in the SGE1 overexpressors . On the other accessory regions ( chromosome 3 , 6 , 15 plus small regions on chromosome 1 and 2 ) the overlap between plant responsive genes and Sge1- , aTf1- or cTf1-responsive genes is much smaller . Here a large group of genes is down-regulated in all transformants but not during infection . Of this group many genes are generally weakly expressed . We noticed that lower expression levels of these genes–as observed in the overexpressors–was also found in wild type strains ( S9 Fig ) . This suggests that the apparent down-regulation of expression from this region may not be due to increased abundance of one of the three transcription factors , but rather to general variability in gene expression in this region between transformants . Another transcription factor-unrelated change in expression was observed for a part of the pathogenicity chromosome . In Fig 8B , the top rows of the enlargement of the pathogenicity chromosome show a group of genes up-regulated in one of the cTF1 overexpressors and one of the SGE1 overexpressors . Closer examination showed that these genes are part of a region on supercontig 22 that is generally higher expressed in these two transformants ( S10 Fig ) . Given the regional nature and the lack of correlation to a particular transcription factor , we suspect these differences may be either caused by changes in chromatin state , or by ‘spontaneous’ duplication of these regions . Spontaneous duplications in accessory regions have been reported previously in F . oxysporum [53] . The majority of all plant-induced genes is located on the core chromosomes , but only a small portion of those is differentially regulated in the aTF1 , cTF1 or SGE1 overexpressors . Remarkably , whereas these three transcription factors seem to target the same set of genes on the pathogenicity chromosome , on the core genome the overlap between the genes of which expression is altered by the different transcription factors is much smaller , and each mostly induces a specific set of genes , especially Sge1 ( S6A Fig ) . Still , the genes up-regulated by the transcription factor genes on core chromosomes are enriched for plant-responsive genes ( hypergeometric test , P value < 0 . 05 after Bonferroni correction ) . To see which functional categories of genes apart from effector genes are targeted by aTf1 , cTf1 and Sge1 , a FunCat analysis was performed [54] ( S1 Data: tab ‘FungiFun’ ) . Of the genes up-regulated in SGE1 overexpressors , genes in the categories heme binding ( including genes coding for cytochrome P450 proteins ) , catalase reactions and electron transport ( including ATPases and oxidoreductases ) are overrepresented . Among the predicted functions of the genes induced in the cTF1 overexpressors there is enrichment in secondary metabolism and C-compound and carbohydrate metabolism ( including polysaccharide metabolism and protein glycosylation ) . The set of genes induced by aTf1 is also enriched for genes in secondary metabolism and C-compound and carbohydrate metabolism ( including polysaccharide and chitin metabolism as well as pectate lyases ) . In the SGE1 overexpressors a peroxisome biogenesis factor ( PEX11 ) is up-regulated . Peroxisome function is required for pathogenicity in Fol [55] . Both aTf1 and cTf1 up-regulate expression of the shorter aTF1 gene on the pathogenicity chromosome ( FOXG_16414 ) , although not to the same level as during infection . Apart from this aTF1 homolog , expression of only one other transcription factor gene is significantly altered in any of the overexpressors: aTf1 up-regulates FOXG_04965 . Strikingly , this transcription factor is required for pathogenicity [45] . The set of down-regulated genes on the pathogenicity chromosome and the core chromosomes was not significantly enriched for certain categories for aTf1 and cTf1 regulated genes . Sge1-repressed genes were enriched for genes in polysaccharide metabolism and amino saccharide metabolism . Taken together , all three transcription factors influence effector gene expression and secondary metabolism and target genes predicted to affect both the fungal and the plant cell wall ( S1 Data: tab ‘FungiFun’ ) . Although very different transcription factors , aTf1/cTf1 and Sge1 induce expression of a large , overlapping set of genes on the pathogenicity chromosome . Together with the observation that some regions may be prone to activation that is not linked to a specific transcription factor gene , this made us wonder whether the pathogenicity chromosome is transcriptionally activated as a whole , rather than gene by gene . The accessory chromosomes are very rich in transposable elements , and we decided to investigate expression of transposable elements as a proxy for chromosome-wide transcriptional activation . Besides , we were interested to see whether general up-regulation of genes on the pathogenicity chromosome ( for instance during infection ) could jeopardize genome integrity by induction of transposon activity . One problem , however , is that in our RNAseq analysis the number of transcripts derived from transposable elements was probably highly underestimated , because: i ) many transposons are not annotated , ii ) reads of identical transposons are distributed over all copies , obscuring activation of individual copies , and iii ) reads mapping to more than ten different genomic locations were excluded . To circumvent this , a fasta file was generated where the sequence of each repetitive element , plus all previously annotated transposable elements [34] is present only once . To this file , all sequence reads from the overexpressors as well as reads from infected plant material were mapped ( S11 Fig and S6D Fig ) . We have found no evidence for a general increase of transposable element transcription in the aTF1 , cTF1 or SGE1 overexpressors or during plant infection . We have shown that expression of Sge1 , aTf1 or cTf1 from the FEM1 promoter is correlated with induced expression of a large overlapping set of genes , including effector genes . Whereas aTf1 and cTf1 are homologs and have the same DNA binding specificity , Sge1 is a very different transcription factor , with a strongly conserved , fungal specific DNA binding domain . This raises the question what causes this overlap in transcriptional activation . Expression of for example SIX1 and SIX3 is induced in aTF1 , cTF1 and SGE1 overexpressors . However , whereas SIX3 has both Tf1 and Sge1 DNA binding sites present in the promoter , for SIX1 only Tf1 DNA binding sites are present . To test whether the presence of Sge1 is required for aTF1 overexpression-mediated induction of SIX1 and SIX3 , we deleted SGE1 in an aTF1 overexpressing strain . Without the presence of SGE1 , the expression of the effector genes SIX1 and SIX3 was reduced to low ( wild type ) levels , while overexpression of aTF1 itself remained unchanged ( Fig 9 ) . This shows that Sge1 is required for aTF1 overexpression-mediated activation of effector gene expression , but not necessarily via an Sge1 DNA binding site . Also , deletion of SGE1 resulted in loss of pathogenicity in both WT and in aTF1 overexpressing strains ( S12 Fig ) . In summary , we have demonstrated that some pathogenicity chromosome-encoded transcription factors have regulating potential on the accessory chromosomes themselves , and that aTF1 and the core-encoded cTF1 and SGE1 can induce effector gene expression and expression of other plant-responsive genes upon constitutive expression , possibly via direct binding of the transcription factors to these promoters and subsequent transcriptional activation .
The large overlap in the sets of target genes of aTf1/cTf1 and Sge1 on the pathogenicity chromosome suggests these transcription factors somehow work together . This is substantiated by the observation that aTF1 overexpression-mediated induction of effector gene expression requires Sge1 . Overexpression of SGE1 orthologs in F . graminearum , F . verticillioides and F . fujikuroi induces the expression of secondary metabolite genes [23–26] . Apart from SGE1 , expression of secondary metabolite gene clusters depends on specialized transcription factors often physically associated with the cluster [56] . Reminiscent of this , SIX effector genes sometimes occur in mini clusters of two or three SIX genes , accompanied by a aTF1 homolog . Genes that are located between or physically close to clustered SIX genes ( ORX1 , SHH1 ) are co-induced , both during infection and in the overexpressors . Transcriptional regulation of secondary metabolite clusters partially takes place on the chromatin level [57 , 58] . Some observations hint at chromatin-mediated regulation of expression of genes on the pathogenicity chromosome of Fol as well . First , there is the transcription factor unrelated up-regulation of parts of SC22 on the pathogenicity chromosome ( in one cTF1 and one SGE1 overexpressor , this study ) . Second , the BLE resistance gene behind the constitutive GPD promoter located near the SIX1 locus is co-up-regulated in some of the overexpressors ( S1 Data , tabs: ‘total mapped reads’ and ‘RPKM’ . Finally , physical interactors of the Saccharomyces cerevisiae Sge1 ortholog Mit1 include histones and histone acetyl transferase complexes and are enriched for the GO terms chromosome organization , chromatin assembly or disassembly and chromatin organization ( yeastgenome . org ) . This raises the question whether the large overlap of genes on the pathogenicity chromosome affected by expression of aTF1/ cTF1 and SGE1 from the FEM1 promoter may be caused by a shared influence on chromatin structure . It will be interesting to see if facultative heterochromatin functions as a layer of effector gene expression regulation also in F . oxysporum . Another explanation for the overlap between the aTF1 , cTF1 and SGE1 targets on the pathogenicity chromosome is that Sge1 induces effector gene expression via aTF1/cTF1 or vice versa . Despite the fact that SGE1 has potential aTf1 binding sites in its promoter and one aTF1 ( FOXG_17458 ) has potential Sge1 DNA binding sites in its promoter , neither of the genes is constitutively up-regulated by overexpression of the other . Moreover , Sge1 is still required for effector gene expression when aTF1 is overexpressed . Also , there are clear differences between the different overexpressors , for example SIX11 and SIX2 are induced in aTF1- and cTF1-overexpressors , but not in SGE1-overexpressors . SIX11 and SIX2 indeed have no Sge1 DNA binding site in their promoter . SGE1 is , however , required for the induced expression of SIX2 upon exposure of Fol to tomato cell cultures [26] . Alternatively , Sge1 and aTf1/cTf1 may be simultaneously required at promoters for transcription activation . Sge1 and cTf1 both have basal ( non-induced ) expression levels , and overexpression of either might cause recruitment of the other ( perhaps by direct interaction or through chromatin modification ) leading to transcription activation . However , we have shown that Sge1 as well as Tf1 can bind DNA and activate transcription independent of each other . A similar situation has been described for the Sge1 homolog Ryp1 and two velvet-like transcription factors in H . capsulatum [59] . A subset of the genes induced by overexpression of Ryp1 is also induced by overexpression of either of the two velvet-like transcription factors . These shared target genes have DNA binding sites for each transcription factor in their promoter , and the three transcription factors physically interact . To unravel the question of Sge1/Tf1 co-operation , it will be necessary to determine physical interactions , both between them and with promoters , under basal conditions and upon overexpression , in combination with analysis of chromatin structure at the same promoters . Next to ( nearly ) identical DNA binding sites , several observations support an overlap in function between homologous transcription factors encoded on accessory and core genomes . aTF7- and cTF7-overexpressing strains show a similar growth retardation and reduction of virulence , and aTf1 and cTf1 both are potent inducers of effector gene expression . In addition , a deletion mutant of cTF1 is less virulent , but can still colonize the plant [45 , 60] . The partial loss of virulence could be a result of redundancy between cTF1 and aTF1 . Deletion of either of two of the shorter aTF1 homologs ( FOXG_16414 and FOXG_17123 ) does not affect pathogenicity [45] , but recently silencing affecting both aTF1 and cTF1 was shown to reduce virulence in both f . sp . lycopersici and f . sp . phaseoli [60] . Also host-induced gene silencing of aTF1 , potentially silencing the entire gene family , has been reported in F . oxysporum f . sp . cubense on banana , which led to complete absence of disease symptoms [61] . If there indeed is such a large functional overlap between core and accessory homologs , then why does Fol contain transcription factor genes on the pathogenicity chromosome at all ? And why—in some cases–even multiple homologs ? One possibility is that the presence of the transcription factor ( and other ) genes on accessory chromosomes did not result from selective pressure on a particular functional advantage but is simply provoked by certain features of accessory chromosomes , such as high transposable element density or chromatin structure . A different , but not mutually exclusive , view is a selective advantage of gene duplications because of a dosage effect . In F . oxysporum f . sp . phaseoli , a correlation between virulence and the number of aTF1 homologs has been reported [35] . It should be noted , however , that although very similar , functions of core and accessory-encoded homologous transcription factors may not be completely redundant . For example , overexpression of aTF1 ( FOXG_17458 ) induces expression of a transcription factor required for pathogenicity , whereas overexpression of cTF1 does not . Also , the short aTF1 ( FOXG_17084 ) homolog can bind DNA with the same specificity , but is far less efficient in inducing SIX1 expression . At present , it cannot be excluded that the short homolog might even function as a negative regulator . It is currently unclear over how large a phylogenetic distance the pathogenicity chromosome can be transferred . Up to now , transfer as only been experimentally demonstrated within the F . oxysporum species complex [7] . In such a case the cTF1 and SGE1 homolog in the recipient strain will be nearly identical to the chromosome donor strain , only accessory-encoded aTf1 homologs on regions other than the pathogenicity chromosome will be absent or different in the recipient . Interesting in this respect is a previous observation: when the pathogenicity chromosome is transferred to the non-pathogenic strain Fo-47 , this strain gains pathogenicity , but is not very aggressive . Virulence is higher in those strains that gained not only the pathogenicity chromosome , but also a second small chromosome , corresponding to the duplicated region of chromosome 3/6 in the reference strain Fol4287 [7] . This region contains few if any effector genes , but many transcription factor genes , including several homologs of aTF1 , aTF4 , aTF6 , aTF7 , aTF8 ( EBR2 ) and aTF9 . Genome analyses also suggest at least one ancient transfer event from a Fusarium species at a phylogenetic distance from Fol somewhere between F . verticillioides and F . graminearum [7] . If such a transfer were to occur again , the cTF1 homolog in the recipient would still be very similar to the cTF1 homolog in the donor strain , at least more similar than cTF1 is to aTF1 . No additional accessory aTF1 homologs would be present in the recipient . We speculate that the transcription factors on accessory chromosomes do contribute to virulence , but many of their roles could be also partially fulfilled by their core-encoded homologs within the genus Fusarium . For SGE1 homologs , significant functional differences have been reported between closely related species , like different Fusarium spp , but also between C . albicans and S . cerevisiae [23–25 , 33] . As described above , the DNA binding domain at the N-terminus is very conserved , and so is–as far as it has been tested–the DNA binding site [33] , whereas the C-terminal domain is very variable [24 , 25] . For the yeasts C . albicans and S . cerevisiae , differences in target genes are caused by differences in the presence or absence of cis-acting promoter elements [33] . For F . graminearum and Fol , however , differences in target genes are caused by differences in the C-terminal domain , and transcomplementation can partially restore pathogenicity only in some cases [24] . Also in F . fujikuroi and F . verticillioides overexpression of their respective SGE1 homologs regulates a different set of genes [23 , 25] . Since the differences between the Sge1 homologs between different Fusarium species are so substantial that Sge1-mediated transcription regulation is extensively rewired , this may be a limiting factor to expression of effector genes from the Fol pathogenicity chromosome in another Fusarium species . Interestingly , transcomplementation of the Δsge1 mutant of Fol with CfWOR1 from C . fulvum resulted in restoration of effector gene expression but not in restoration of pathogenicity [28] . Previously , Jonkers and co-workers have looked at differentially expressed genes in a Fol SGE1 deletion mutant compared to WT using a microarray [24] . This set ( 1213 genes ) is very different from the differentially expressed genes in the SGE1 overexpressors that we found ( 168 genes ) ; only 15 genes are present in both sets . This shows that the different approaches are complementary for identification of target genes . The genes differentially expressed between WT and deletion strain are predominantly present on the core genome . Also , the Sge1 DNA binding site is not significantly enriched among up- or down-regulated genes in the deletion mutant ( 104 [out of 394] down-regulated genes with a Sge1 DNA binding site and 171 [out of 820] up-regulated genes , on a total of 4870 [out of 20935] genes with a Sge1 DNA binding site ) , in contrast to the SGE1 overexpressor . This may be because several transcription factor genes are differentially expressed in the deletion mutant , potentially regulating secondary targets . Also , the comparison of WT and SGE1 deletion mutant was of necessity conducted under axenic growth conditions which excludes finding targets that are not or very weakly expressed under those conditions , including most genes on the accessory genome . We have shown that aTf1 can activate effector gene expression . However , many more transcription factors are encoded on the pathogenicity chromosome . It is tempting to speculate that these transcription factors may control expression of some of the other plant-responsive genes . The core-encoded homologs of some of the other accessory transcription factors have been implicated in pathogenicity . cTF8 ( EBR1 ) is required for full pathogenicity of Fol and F . graminearum , and orthologs of cTF4 and cTF9 are required for pathogenicity of F . graminearum ( FGSG_10057 , FGSG_10517 and FGSG_06651 respectively ) [37 , 62] . Of these three transcription factors , we have only obtained a DNA binding site for cTf4 , which is , like the DNA binding sites of aTf2 , aTf5 and aTf7 , not enriched among genes up- or down-regulated during infection . It is of course possible that there is not always a significant correlation between the presence of the DNA binding site and changes in expression during infection . This may occur when loss of pathogenicity is an indirect effect ( via a second , downstream transcription factor ) , when only a particular combination of transcription factors regulates plant responsive genes , or when too many apparent binding sites are not functional , precluding detection of a significant association between binding site and gene regulation . Of two transcription factors ( cTf4 and aTf5 ) , DNA binding sites are enriched on the accessory chromosomes , suggesting they may also act on genes of the accessory chromosomes . In F . graminearum , global gene regulatory network modelling revealed that species-specific genes are most often controlled by species-specific regulators , whereas genes conserved between species are controlled by conserved regulators [63] . Possibly , such a compartmentalized network structure of gene regulation may also apply to F . oxysporum . Other putative functions of the transcription factors encoded on the pathogenicity chromosome could be downregulation of effector gene expression and returning to or maintaining a repressed , saprotrophic state . They may also be promoting horizontal transfer . It would be interesting to see if the pathogenicity chromosome can be lost , and if so , in which ways this may affect the phenotype of Fol and expression of core genes .
In order to determine the DNA binding sites of the different transcription factors , the ORFs were cloned from cDNA . For this , RNA was isolated from axenic cultures and from susceptible infected tomato plants , between 1–2 weeks after inoculation . cDNA was generated as described below . For some transcription factors alternative start codons were tried , and the longest obtained PCR product was selected for cloning . For many transcription factors , 35 PCR cycles was insufficient to amplify clonable amounts of DNA , therefore , a re-amplification was done . Primers were designed in such a way that the subsequent product could be cloned with AscI , BamHI or SbfI , in frame with a N-terminal GST-tag in an Escherichia coli T7 expression vector pTH6838 . The cloned ORF was checked by sequencing . To express the transcription factor genes in Fol , the binary vector pRW2h was modified , yielding a plasmid with a right border ( facilitating Agrobacterium tumefaciens mediated transformation ) , the FEM1 promoter , a multiple cloning site including XbaI , AscI , StuI , SbfI , BglII and ApaI , followed by the SIX1 terminator , the HPH resistance cassette and the left border . The transcription factor ORF was then cut out of pTH6838 with the same enzymes used to clone it in , and cloned into pRW2h_Pfem_MCS_Tsix1 , again with the same enzymes . For those transcription factors of which the gDNA ORF was cloned , a PCR was done on Fol007 gDNA , isolated as described in [64] using the same primers initially designed for cloning of the cDNA . The PCR product was digested and cloned directly into pRW2h_Pfem_MCS_Tsix1 . The cloned transcription factor ORFs were checked by sequencing . Fol was transformed via Agrobacterium mediated transformation , as described previously [65] . Transformants were monospored by pipetting 10–20 μl of sterile water on the emerging colony , and spreading this on a fresh PDA plate supplemented with cefotaxime and Hygromycine . After two days of growth at 25 degree Celsius , single colonies were picked and transferred to fresh plates . From these plates glycerol stocks were made and these are the transformants we worked with . Transformants were only selected on antibiotic resistance , and should contain one ( or in rare cases more than one ) random ectopic insertion of the T-DNA construct [26] . Cultures for RNA isolation were grown as described below , except for experiments described in S8A Fig and Fig 9 . These RNA isolations were done as described under ‘Deletion of SGE1 in aTF1 overexpressor background’ later in this section . For all other cases a small cube with mycelium from a PDA plate was used to inoculate a preculture of 50 ml liquid minimal medium ( 1%KNO3 , 3% sucrose 0 . 17% YNB w/o amino acids or NH4 ) and grown for 3 to 5 days at 25°C , shaking 150–175 rpm . From this preculture microconidia were isolated by filtering the culture over miracloth and pelleting the microconidia in the filtrate at 2000 rpm for 10–15 min . To harvest mycelium for RNAseq analysis of transcription factor overexpressors or for quantification of transcription factor transcripts by Q-RT-PCR , microconidia were suspended in a small volume of minimal medium , counted , and used to inoculate 100 ml liquid minimal medium with 2 . 5 * 10^8 microconidia . This culture was grown for 2 days at 25°C , shaking 150–175 rpm before mycelium was harvested by filtering the culture over a double layer of miracloth . The mycelium in the filter was washed once with 50–100 ml sterile water , scraped from the miracloth and snap frozen in liquid nitrogen . Of each condition , three independent biological replicates were sampled . To harvest material for RNAseq analysis of infected plants , Fol4287-infected tomato roots were harvested nine days post inoculation . Infections were performed as described below ( for the bioassays ) . Fol4287 mycelium from axenic cultures was harvested from five day old cultures inoculated from plate in 100 ml liquid minimal medium ( 1%KNO3 , 3% sucrose 0 . 17% YNB w/o amino acids or NH4 ) , 25°C , 150–175 rpm . RNA was isolated as described earlier , using a Trizol extraction on mycelium ground in liquid nitrogen , followed by DNase treatment and purification over RNeasy RNA purification columns , according to the instructions of the manufacturer ( Qiagen ) . Synthesis of cDNA was performed using 1 μg of RNA , poly dT primers , Promega RNasin ( ribonuclease inhibitor ) and Gibco Superscript II RNase H− Reverse transcriptase , according to instructions of Gibco . Of 2 μg of total RNA of each biological replicate , polyadenylated RNA was amplified and ligated to adapters to make a library suitable for multiplex illumina paired-end sequencing . Each sample was barcoded and sequenced in 8 different lanes . After de-multiplexing , total reads of the different lanes were combined . This rendered one file of reads per biological replicate . Quantitative PCR was performed with a model 7500 Real Time PCR system ( Applied Biosystems ) and Solis BioDyne 5x HOT FIREPol Eva Green qPCR Mix Plus ( ROX ) . Primers used for Q-RT-PCR where designed to amplify fragments of approximately 100 bp and tested for primer efficiency and melting curve ( S1 Data ) . 1 μL of cDNA was used per sample , two technical replicates were performed for each sample . Transcription elongation factor 1α ( EF1α ) gene expression was used as a reference , and RNA that was not transcribed into cDNA as a gDNA contamination control . The following formula was used to calculate the amount of DNA: [DNA] = ( 1/E^ Ct_sample ) - ( 1/E^Ct_control ) , with E = primer efficiency , Ct_sample = Ct value of the test sample , using WT or TF overexpressor cDNA as a template and Ct_control = Ct value of no cDNA control sample ( to check for gDNA contamination ) , with the same primer pair . The comparison with EF1α was made as follows: DNA_TF/DNA_EF1α . Standard deviations of the two technical replicates per sample were calculated with the following formula: Standard deviation = √ ( ( stdev DNA_TF /average DNA_TF ) ^2 + ( stdev DNA_EF1α /average DNA_EF1α ) ^2 ) ) * ( DNA_TF /DNA_EF1 α ) . The Illumina reads ( 125 bp paired end , insert size around 200–500 bp ) were mapped to the annotated genome of Fol4287 ( Fusarium Comparative Sequencing Project , Broad Institute of Harvard and MIT ( http://www . broadinstitute . org/ ) , annotation 3 ) using CLC Genomics Workbench version 6 . 5 . 1 module ( CLC bio , Aarhus , Denmark ) . Reads were imported as: illumina ( pipeline 1 . 8 and later ) , paired-end reads , insert size 100–600 bp , remove failed reads . Imported reads were trimmed to remove any remaining adapter sequences or low quality reads . Quality scores and ambiguous nucleotides were trimmed with standard settings ( limit 0 . 05 , ambiguities 2 ) . Adapters were trimmed by checking for the presence of the Truseq Universal adapter ( minus strand ) and the presence of the Truseq index adapter ( plus strand ) with the following parameters: mismatch = 2 , gapcost = 3 , cutoff ns , cutoff at end 6 , action: remove adapter . The following gene models were manually added to the annotation: Supercontig_2 . 36 135864–136180 minus strand FOXG_SIX14; Supercontig_2 . 51 62412–62758 minus strand FOXG_SIX12; Supercontig_2 . 51 65216–65878 plus strand FOXG_SIX7; Supercontig_2 . 22 806692–807024 minus strand FOXG_SIX11; Supercontig_2 . 22 44647–44970 minus strand FOXG_PEG4 ( Putative Effector Gene 4 ) ; Supercontig_2 . 36 468654–469317 minus strand FOXG_SIX8-36; Supercontig_2 . 51 6999–7662 minus strand FOXG_SIX8-51 . Reads were mapped to the annotated genome with parameters: Organism type = eukaryote . Exon discovery according to standard settings: relative expression level = min 0 . 2 , min reads = 10 , min length = 50 bp . Additional downstream bases = 0 . Additional upstream bases = 0 . Minimum length fraction = 0 . 9 . Minimum similarity fraction = 0 . 95 . Minimum number of reads = 10 . Map only intact pairs , count paired reads as one . Unspecific match limit = 10 . Expression value = Total number of mapped reads and reads per kb per million mapped reads ( RPKM ) . Differentially expressed genes were called for pairwise comparisons of the total number of mapped reads ( three replicates ) , using the Bioconductor DEseq software [52] . After normalization for library size ( estimateSizeFactors ) and variance estimation ( estimateDispersion ) with parameters “method = blind , sharingMode = fit-only” , Nbinominal testing and the Benjamin-Hochberg multiple testing adjustment procedure were used [52] . For the transcription factor gene overexpressing transformants ( all compared pairwise to the Fol007 Psix1GFP samples ) , all genes that had an adjusted p value <0 . 1 for both overexpressors were considered significant . For the samples from infected plants ( compared to the Fol 4287 control samples ) all genes with an adjusted p value < 0 . 05 were considered significant . The output file gives the mean normalized mapped reads per sample ( mean of the three replicates ) , pvalue and adjusted pvalue . For all pairwise comparisons the normalized total mapped reads were collected , and every count of zero reads was replaced by 0 . 1 ( 0 . 1 is roughly half of the lowest number of normalized total mapped reads in all comparisons , this allows calculation of -an approximate- fold change for each gene ) . Fold change and log2 fold change were calculated . To visualize expression differences in a heatmap , all genes considered differentially expressed in one of the comparisons were listed and the log2 fold change for each condition was listed in seven subsequent columns . In this list , each value not reaching the significance threshold was replaced by ‘0’ ( indicating no fold change ) . This list was separated into three lists based on subgenome and each of these lists were clustered on gene and condition in Gene Cluster 3 . 0 , uncentered correlation , average linkage . Results were visualized in Java Treeview . Next to this , the differentially expressed genes were subdivided in lists of up-regulated and down-regulated genes . These lists were used to count the contributions of different subgroups to each category ( for example: number of aTF1 up-regulated genes that is located on the pathogenicity chromosome ) . Hypergeometrical distribution tests were used to determine significant enrichments or depletions among different categories . The adjusted p value was reached by multiplying the p value with the total number of tests performed . To determine which genes have a MIMP in their promoter , two kb upstream the ATG of each gene was searched for the presence of complete ( both inverted repeats present ) or partial MIMPs . Any missing SIX gene , of which a MIMP had been demonstrated in the promoter earlier [34] , was manually added to the list . The Illumina reads ( 125 bp paired end , insert size around 200–500 bp ) were mapped to the fasta file with all repetitive/transposable elements as described above , except the following parameters: Organism type = prokaryote , unspecific match limit = 30 . Reads were normalized as the number of reads per 20*10^6 uniquely mapped reads to the total genome for that particular sample . For the more detailed analysis the same pairwise comparisons were made as described above . Every count of zero reads was replaced by 0 . 1 , the data was log10 transformed and a T-test was performed on the log10 transformed normalized total mapped reads . All sequences that compared differently ( p<0 . 05 ) were counted . Bioassays were performed as described earlier [66] . Briefly , tomato seedlings of 10 to 11 days were trimmed at the main root and dipped in a Fol microconidia suspension of 0 . 5*10^7 microconidia/ml for at least 1 minute . The seedlings were potted in soil in individual pots and grown in the greenhouse at 25°C for three weeks . At the time of harvest , the above ground parts were cut off at the cotelydons and scored for fresh weight and disease index . The disease index ranges from 0 ( no symptoms ) , 1 ( thickening of hypocotyl , formation of lateral roots ) , 2 ( one brown vessel ) , 3 ( up to ¾ of the vessels show browning , asymmetric development ) to 4 ( all vessels brown , severe growth retardation , death ) . Growth assays were performed by positioning a droplet of spores on the middle of a PDA or CDA plate , growing the fungus at 25°C for 5 days , and measuring the colony diameter . GFP fluorescence was measured on a Fluostar optima platereader ( BMG Labtech ) . For this dilutions of 10^8 , 10^7 , 10^6 and 10^5 microconidia per ml were made and 200 μl of each suspension was pipetted in a sterile , flat bottom , black 96 well plate ( Greiner ) . The plate was shortly mixed and measured from the bottom , with 470–10 nm excitation and a 510–10 nm emission filter . The plates were kept o/n at 25°C , and measured again the next day , same settings . No differences were observed between the days ( apart from a slight increase in fluorescence due to growth ) . Details of the design and use of PBMs have been described elsewhere [38 , 67–69] . Here , we used two different universal PBM array designs , designated 'ME' and 'HK' , after the initials of their designers , as described in [70] . Briefly , we used 150 ng of plasmid DNA in a 15 μl in vitro transcription and/or translation reaction using a PURExpress In Vitro Protein Synthesis Kit ( New England BioLabs ) supplemented with RNase inhibitor and 50 μM zinc acetate . After a 2-h incubation at 37°C , 15 μl of the mix was added to 155 μl of protein-binding solution for a final mix of PBS/2% skim milk/0 . 2 mg per ml BSA/50 μM zinc acetate/0 . 1% Tween-20 . This mixture was added to an array previously blocked with PBS/2% skim milk and washed once with PBS/0 . 1% Tween-20 and once with PBS/0 . 01% Triton-X 100 . After a 1-h incubation at room temperature , the array was washed once with PBS/0 . 5% Tween-20/50 mM zinc acetate and once with PBS/0 . 01% Triton-X 100/50 mM zinc acetate . Cy5-labeled anti-GST antibody was added , diluted in PBS/2% skim milk/50 mM zinc acetate . After a 1-h incubation at room temperature , the array was washed three times with PBS/0 . 05% Tween-20/50 mM zinc acetate and once with PBS/50 mM zinc acetate . The array was then imaged using an Agilent microarray scanner at 2 μm resolution . Images were scanned at two power settings: 100% photomultiplier tube ( PMT ) voltage ( high ) , and 10% PMT ( low ) . The two resulting grid images were then manually examined , and the scan with the fewest number of saturated spots was used . Image spot intensities were quantified using ImaGene software ( BioDiscovery ) . Calculation of spot intensities was done as described in [38] . In summary , bad spots ( spots that had scratches , dust flecks or other imperfections ) were flagged manually and removed from subsequent analysis . The PBM signal intensity at each spot was normalized by the corresponding amount of dsDNA . To correct for any possible nonuniformities in hybridization , these normalized PBM intensities were then adjusted according to their positions on the microarray . Each spot was considered to be at the center of a block of spots [70] . The difference between the median normalized intensity of the spots within the block and the median normalized intensity of all spots on the microarray was subtracted from the normalized intensity at that particular spot . Calculation of 8-mer Z- and E-scores was performed as previously described [38 , 71] . Z-scores are derived by taking the average spot intensity for each probe containing the 8-mer , then subtracting the median value for each 8-mer , and dividing by the standard deviation , thus yielding a distribution with a median of zero and a standard deviation of one . E-scores are a modified version of the AUROC statistic , which consider the relative ranking of probes containing a given 8-mer , and range from −0 . 5 to +0 . 5 , with E > 0 . 45 taken as highly statistically significant [67] . DNA binding sites were determined as described in [72] . The oligo-binding array returns for each transcription factor a set of eightmers and corresponding E-scores that indicate the likelihood that the protein binds this eightmer . This set of eightmers contains both the forward and reverse binding eightmers and may represent different binding motifs for a single transcription factor . Hence to infer binding motifs for a transcription factor from a set of eightmers , we need to first cluster eightmers into similar groups , where we expect at least two clusters ( ‘forward’ and ‘reverse’ ) for each transcription factor , unless the binding motif is a palindrome . First we remove unreliable eightmers ( those that have a score < 0 . 45 ) . We then perform pairwise Smith-Waterman alignments using the water program from the EMBOSS package ( with options: -nobrief -gapopen 5 . 0 -gapextend 2 . 0 ) to obtain a sequence similarity measure ( the Smith-Waterman score ) for each eightmer-pair . We take 40 . 0 –the Smith-Waterman score as a pairwise distance and use hierarchical clustering as implemented in scipy ( average linkage: UPGMA ) to obtain a hierarchical clustering of the eightmers . We split the resulting clustering trees into two clusters and manually checked whether these correspond to ‘forward’ and ‘reverse’ strands . We find that this is the case for all transcription factors except FOXG_15625 that probably has a palindromic binding site and FOXG_04904 for which we did not find ‘reverse’ strand eightmers . In the cases where we could identify ‘forward’ and ‘reverse’ strands we added the reverse complement of ‘forward’ strand sequences to ‘reverse’ strand sequences , in other cases we simply merged both clusters as they were . We aligned the sequences and obtained a sequence logo ( as shown in Fig 3A ) for these alignments with WebLogo [73] . The following sequence motifs were used to search the regions 1000 bp upstream or 1000bp downstream of the annotated transcriptional start site ( Fusarium Comparative Sequencing Project , Broad Institute of Harvard and MIT ( http://www . broadinstitute . org/ ) , annotation 3 ) ; aTF2: CAAAC , cTF4: AGCC[A , G , C , T]TA , aTF5: CACGT , aTF7: G[A , G , C]GGCT , aTF1:T[A , G]CCG , SGE1: TTA[A , G][A , G][G , C]TT , effector motif: AACTGCCGA . For the upstream regions we used fasta files with promoter regions that we downloaded from the Broad Institute . We used custom Python scripts to append promoter regions for SIX genes that were not part of the annotation , based on the reported locations in [34] . We used custom Python scripts to make a fasta file with sequences that correspond to the first 1000 bp downstream from the first ATG based on the transcript gtf-file downloaded from the Broad Institute or–for SIX genes that were not part of the annotation–based on locations reported in [34] . We counted the number of genes with one or more motifs in the upstream regions , forward and reverse orientation separately . We did the same for two or more motifs and three or more motifs . Significant enrichment of genes with binding sites in the upstream regions in certain subgroups ( accessory genes , plant-induced genes , etc . ) was tested with a Hypergeometric test , with a P value < 0 . 01 after Bonferroni correction . To define TF families we used blastp to search for homologs in 12 Fusarium oxysporum species ( Fusarium Comparative Sequencing Project , Broad Institute of Harvard and MIT ( http://www . broadinstitute . org/ ) [74] . We only included hits that have an E-value < 1e-5 and for which the alignment returned by BLAST spans more than 60% of both the query and the subject sequence . We used a custom Python script to cluster all hits into families using single linkage clustering and used Clustal Omega to construct multiple sequence alignments per family [75] . We trimmed the alignments using trimAl ( -strictplus ) [76] , inspected and manually curated them . We used PhyML ( with options: -q -b -2 -v e -a e ) to infer phylogenies [77] . For large families we pruned the tree such that we keep the last common ancestor of the TFs that lie on chromosome 14 and their nearest neighbour that lies on the core , as root of the pruned tree . For each of the seven transcription factor families tested for DNA binding , we searched for the occurrence of conserved domains from the Pfam database ( version 27 . 0 ) using hmmscan from the hmmer package . We manually checked for presence of residues that are conserved in the Pfam seed alignment of the DNA binding domain in the multiple sequence alignments ( S3 Fig ) . We compiled a list of sequences corresponding to putative transposable elements by extracting DNA sequences for elements identified in a thorough analysis of chromosome 14 [34] . We combined these elements by elements identified by running RepeatMasker ( with RepBase19 . 11 ) and extracting all sequences that were not denoted as a low-complexity region or simple repeat . We filtered out multiple occurrences of identical sequences . For the transcription activation assay in yeast , plasmids ( Ptef1-expression vector [B3909] with and without WOR1 , Wor1 DNA binding site–WT , mutated or empty–in front of UAS-less CYC1 promoter followed by the LacZ reporter gene [B3946] ) and strains ( S . cerevisiae Sigma 2000 ΔYEL007 /ΔYHR177 ) were a kind gift from Alexander D . Johnson and are described in [42] . To express SGE1 in yeast , the SGE1 ORF was amplified from gDNA and cloned behind the TEF1 promoter in plasmid B3909 using SpeI and XhoI restriction enzymes . To express aTF1 or cTF1 in yeast , for each gene both exons were amplified from gDNA and fused together in an overlap PCR , to create an intronless sequence . This sequence was then cloned behind the TEF1 promoter in plasmid B3909 using SpeI and SalI restriction enzymes . To clone the Tf1 DNA binding site ( WT or mutated ) , oligo pairs were ordered corresponding to part of the SIX1 promoter ( -335 to –290 relative to the ATG ) containing two Tf1 motifs flanked by 5 bp on each end plus sticky ends corresponding to the XhoI restriction site . Oligos were phosphorylated using T4 PNK ( Fermentas ) , ligated into XhoI-digested and phosphatase treated B3946 plasmid . All plasmids were sequenced to verify the insert sequence and orientation . Oligos are listed in S1 Data , tab: ‘primers’ . Plasmids containing WOR1 , SGE1 , aTF1 or cTF1 and LacZ reporter plasmids were co-transformed to yeast strain Sigma 2000 ΔYEL007 /ΔYHR177 according to [78] . LacZ activity was assayed as follows . Transformed yeast cells were grown in SD medium lacking Uracil and Histidine . Before cells were harvested , OD600 was measured and cells were pelleted . The cell pellet was resuspended in 150μl Z-buffer with β-mercaptoethanol ( 60 mM Na2HPO4 , 40 mM NaH2PO4 , 10 mM KCl , 1mM MgSO4 , 1 mM β-mercaptoethanol , pH 7 ) , 50 μl chloroform and 20 μl 0 . 1% SDS were added , tubes were vortexed for 15 seconds and 700 μl pre-warmed ( 30°C ) Z-buffer ( 60 mM Na2HPO4 , 40 mM NaH2PO4 , 10 mM KCl , 1mM MgSO4 , pH 7 ) with ONPG ( 1 mg/ml ) was added at t = 0 . Tubes were incubated at 30°C until the reaction started to turn yellow . The reaction was stopped with 500 μl 1M Na2CO3 and the time recorded , Tubes were centrifuged 2 min . at 14000 rpm and the OD of the supernatant was measured at 420 nm . Miller units were calculated as follows: ( A420*1000 ) / ( A600*minutes*ml culture ) . To make aTF1 overexpressors in the Fol4287 wild type background , the hygromycin resistance cassette ( HPH ) in the plasmid described above ( Pfem1aTF1-HPH cassette ) was exchanged for a phleomycine resistance cassette ( BLE ) , using XbaI plus PacI ( insert ) and SpeI plus PacI ( vector ) restriction enzymes . The resulting plasmid ( Pfem1aTF1-BLE cassette ) was transformed to Fol4287 and transformants were selected on zeocine . An empty plasmid ( pRW1p: containing only the BLE cassette [79] ) was transformed to Fol4287 in parallel , as a negative control . Ten independent zeocine resistant colonies of each transformation were monospored and checked for SIX1 and SIX3 expression . For this , liquid cultures ( 100 ml 1% KNO3 , 3% sucrose , 0 . 17% YNB w/o NH4 and aa . ) were inoculated from plate and mycelium was harvested after 5 days at 25°C , 150–175 rpm . RNA isolation , cDNA synthesis and Q-RT-PCR were performed as described above . SIX1 and SIX3 levels were normalized to EF1-α . Two out of ten aTF1 transformants showed induction of SIX1 and SIX3 expression . One of those was selected for deletion of SGE1 . Deletion of SGE1 in the selected Fol4287 aTF1 overexpressor was done as described in [26] , using the same deletion construct and the same PCR control for deletion of SGE1 . To check transformants ( ectopic and in locus ) for SIX1 and SIX3 expression , liquid cultures ( 100 ml 1% KNO3 , 3% sucrose , 0 . 17% YNB w/o NH4 and aa . ) were inoculated from plate and mycelium was harvested after 5 days at 25°C , 150–175 rpm . RNA isolation , cDNA synthesis and Q-RT-PCR were performed as described above . SIX1 , SIX3 , SGE1 and aTF1 levels were normalized to EF1-α . Primers used are listed in S1 Data , tab: ‘primers’ . | Eukaryotic genomes are organised . Genomic regions may differ in spatial organisation , chromatin condensation , rate of evolution , GC content , gene expression and transposon density . Many plant pathogenic fungi maintain genomic subcompartments containing specialized genes to facilitate host colonisation . These genes , for example effector genes , show concerted transcriptional up-regulation during infection . An extreme case is the tomato pathogen Fusarium oxysporum f . sp . lycopersici , which carries accessory chromosomes , of which one encodes all effector genes and can be transferred horizontally between strains . We investigated the transcriptional connections between this accessory chromosome and the core genome , particularly with respect to effector gene expression . Several of the transcription factors encoded on this accessory chromosome bind to motifs enriched on accessory chromosomes , suggesting the accessory chromosomes may be partially transcriptionally independent . Only one of these , Ftf1 , can induce the expression of effector genes and binds to a motif that is enriched in their promoters . Also Sge1 –a conserved regulator of fungal lifestyle switches and required for infection–can activate the expression of effector genes . Both transcription factors induce a largely overlapping set of genes , including many of the host-induced genes on the accessory chromosome including the effector genes . This demonstrates the existence of extensive transcriptional connections between accessory and core chromosomes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"and",
"health",
"sciences",
"chemical",
"characterization",
"pathology",
"and",
"laboratory",
"medicine",
"gene",
"regulation",
"regulatory",
"proteins",
"dna-binding",
"proteins",
"dna",
"transcription",
"transcription",
"factors",
"genetic",
"elements",
"dna",
"research",
"and",
"analysis",
"methods",
"dna",
"binding",
"assay",
"chromosome",
"biology",
"proteins",
"gene",
"expression",
"pathogenesis",
"binding",
"analysis",
"biochemistry",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"transposable",
"elements",
"biology",
"and",
"life",
"sciences",
"genomics",
"mobile",
"genetic",
"elements",
"chromosomes"
] | 2016 | Transcription Factors Encoded on Core and Accessory Chromosomes of Fusarium oxysporum Induce Expression of Effector Genes |
Mammalian and fungal prions arise de novo; however , the mechanism is poorly understood in molecular terms . One strong possibility is that oxidative damage to the non-prion form of a protein may be an important trigger influencing the formation of its heritable prion conformation . We have examined the oxidative stress-induced formation of the yeast [PSI+] prion , which is the altered conformation of the Sup35 translation termination factor . We used tandem affinity purification ( TAP ) and mass spectrometry to identify the proteins which associate with Sup35 in a tsa1 tsa2 antioxidant mutant to address the mechanism by which Sup35 forms the [PSI+] prion during oxidative stress conditions . This analysis identified several components of the cortical actin cytoskeleton including the Abp1 actin nucleation promoting factor , and we show that deletion of the ABP1 gene abrogates oxidant-induced [PSI+] prion formation . The frequency of spontaneous [PSI+] prion formation can be increased by overexpression of Sup35 since the excess Sup35 increases the probability of forming prion seeds . In contrast to oxidant-induced [PSI+] prion formation , overexpression-induced [PSI+] prion formation was only modestly affected in an abp1 mutant . Furthermore , treating yeast cells with latrunculin A to disrupt the formation of actin cables and patches abrogated oxidant-induced , but not overexpression-induced [PSI+] prion formation , suggesting a mechanistic difference in prion formation . [PIN+] , the prion form of Rnq1 , localizes to the IPOD ( insoluble protein deposit ) and is thought to influence the aggregation of other proteins . We show Sup35 becomes oxidized and aggregates during oxidative stress conditions , but does not co-localize with Rnq1 in an abp1 mutant which may account for the reduced frequency of [PSI+] prion formation .
Prions are infectious agents composed of misfolded proteins . They are associated with a group of neurodegenerative diseases in animals and humans that have common pathological hallmarks , typified by human Creutzfeldt-Jakob Disease ( CJD ) . The presence of the misfolded prion protein ( PrPSc ) underlies the development of prion diseases in a mechanism which involves conversion of the normal prion protein ( PrP ) into its infectious PrPSc conformation [1 , 2] . Aggregated , protease-resistant PrPSc seeds are believed to act as templates that promote the conversion of normal PrPC to the pathological PrPSc form , which is rich in β-sheets and resistant to chemical and enzymatic degradation . PrPc can adopt an alternative conformational state by spontaneous misfolding event ( s ) that might be triggered by mutation , mistranslation , environmental stresses and/or by disruption of the chaperone network [3] . This ‘protein-only’ mechanism of infectivity also explains the unusual genetic behaviour of several prions found in the yeast Saccharomyces cerevisiae [4–8] . At present , several yeast proteins are known to form prions with many other proteins classified as potential prion candidates [9] . Additionally , the [Het-s] prion that controls vegetative incompatibility has been described in Podospora anserina , an unrelated fungal species [10] . [PIN+] and [PSI+] are the best studied yeast prions , which are formed from the Rnq1 and Sup35 proteins , respectively [4 , 11 , 12] . Sup35 is the yeast eERF3 which functions in translation termination and hence [PSI+] formation influences the recognition of translation stop codons . [PSI+] formation requires the presence of an another prion , termed [PIN+] , which is often present as the prion form the Rnq1 protein whose native protein function is unknown [13–15] . However , a number of prions can be designated as [PIN+] that are required for the de novo formation of [PSI+] [16–18] . Several studies have demonstrated the infectious behavior of the fungal prion associated with a particular phenotype adding further weight to the ‘protein-only’ mechanism of prion propagation [19–21] , How prions form spontaneously without underlying infection or genetic change is poorly understood at the molecular level , yet if we are to develop effective preventative measures for human and animal amyloidoses , this mechanism must be established . Of particular importance is identifying what can trigger this event . Several different environmental stress conditions , including heat , oxidative and salt stresses , increase the frequency of yeast [PSI+] prion formation [22] . A number of mutants have been identified which increase the frequency of [PSI+] formation [22] . This includes a number of mutations in the protein homeostasis network including mutations in chaperones and the autophagy system [7 , 23] . Additionally , formation of the yeast [GAR+] prion can be induced by bacterial exposure in a chemical induction mechanism and the [GAR+] prion can be lost upon desiccation [24–26] . The spontaneous formation of prions may therefore occur as a result of random protein misfolding events which are normally dealt with by the cellular protein quality control systems . One strong possibility underlying the de novo formation of prions is that oxidative damage to the non-prion form of a protein may be an important trigger influencing the formation of its heritable prion conformation [27] . For example , methionine oxidation of mammalian PrP has been proposed to underlie the misfolding events which promote the conversion to PrPSc [28–30] while methionine oxidation destabilizes native PrP facilitating misfolding and transition to the PrPSc conformation [31] . Methionine oxidation is also a common factor in many protein misfolding diseases [32–34] and an age-dependent increase in methionine oxidation has been detected in various model systems [35] . Oxidative stress has been shown to increase the frequency of yeast [PSI+] prion formation [22] . The de novo formation of the [PSI+] prion is also significantly increased in yeast mutants lacking key antioxidants suggesting that endogenous reactive oxygen species ( ROS ) can trigger the de novo formation of the [PSI+] prion [36–38] . Preventing methionine oxidation by overexpressing methionine sulphoxide reductase abrogates the shift to the prion form indicating that the direct oxidation of Sup35 may trigger structural transitions favouring its conversion to the transmissible amyloid-like form [37 , 38] . Hence , protein oxidation may be a common mechanism underlying the aggregation of some mammalian and some yeast amyloid-forming proteins . The frequency of de novo appearance of the [PSI+] prion is increased by overexpression of Sup35 in [PIN+][psi-] strains which increases the probability of forming prion seeds [4] . This frequency can be influenced by components of the actin cytoskeleton which physically associate with Sup35 including various proteins of the cortical actin cytoskeleton ( Sla1 , Sla2 , End3 , Arp2 , Arp3 ) that are involved in endocytosis [39] . Loss of some of these proteins decreases the aggregation of overexpressed Sup35 and de novo [PSI+] formation . This is particularly interesting given the increasing evidence suggesting that cytoskeletal structures provide a scaffold for the generation of protein aggregates . Insoluble aggregates of amyloid-forming proteins including prions are targeted to the IPOD as part of the cells’ protein quality control system [40 , 41] . The IPOD is located at a perivacuolar site adjacent to the preautophagosomal structure ( PAS ) where cells initiate autophagy [42] . Prion conversion has been proposed to occur at the cell periphery in association with the actin cytoskeleton , prior to deposition at the IPOD [43] . The actin cytoskeleton has also been implicated in the asymmetric inheritance of oxidatively-damaged proteins [44] . Actin organization therefore appears to play an important role in the aggregation of damaged proteins , which can result in prion formation . Oxidative stress provides a powerful tool to examine the de novo formation of prions since it does not necessitate overexpression or mutation of the normally soluble version of the prion protein . In this current study , we have used a mutant lacking the Tsa1 and Tsa2 antioxidants to isolate the proteins which aggregate with Sup35 . We used a tsa1 tsa2 antioxidant mutant to enrich for factors which associate with oxidized Sup35 and therefore might be important for the conversion of Sup35 to the [PSI+] prion . Our data suggest a key role for the cortical actin cytoskeleton since we identified a number of components of the Arp2/3 actin-nucleation complex which specifically associate with Sup35 in the antioxidant mutant . We show that loss of several of these factors abrogates the increased frequency of [PSI+] prion formation which is normally observed in response to oxidative stress conditions . However , these mutants do not affect the increased frequency of [PSI+] prion formation induced in response to Sup35 overexpression . We show that Sup35 oxidative damage and aggregation occurs in actin-nucleation complex mutants in response to oxidative stress conditions , but the aggregates do not appear to form normally at the IPOD . Our data suggest that the cortical actin cytoskeleton is important for the formation of a propagating [PSI+] conformer following oxidant-induced misfolding and aggregation of Sup35 .
To address the mechanism by which Sup35 forms the [PSI+] prion during oxidative stress conditions , we used tandem affinity purification ( TAP ) and mass spectrometry to identify proteins which associate with Sup35 in a tsa1 tsa2 mutant . For this analysis , we used [PIN+][psi-] versions of wild-type and tsa1 tsa2 mutant strains containing genomically-tagged Sup35 . We have previously confirmed that TAP-tagging Sup35 does not affect reversible [PSI+] prion formation [37] . Freshly inoculated strains were grown for 20 hours ( approximately ten generations ) and Sup35 affinity-purified from both strains using TAP chromatography . The associated proteins were identified from three repeat experiments and were considered significant if they were identified in at least two independent experiments . This resulted in the identification of 63 and 47 proteins which co-purify with Sup35 in the wild-type and tsa1 tsa2 mutant strains , respectively ( S1 Table ) . We searched for functional categories that were enriched in the Sup35 co-purifying proteins using MIPS category classifications ( Fig 1A and S2 Table ) . The overlap between the wild type and tsa1 tsa2 datasets is 18 proteins and , as might be expected , this included functions related to protein fate ( >3-fold enrichment; Fischer's exact test , P<10−7 ) such as chaperones and stress-related proteins ( Sti1 , Sse1 , Cdc48 , Ssa2 , Hsc82 , Ssb2 , Ssa1 , Hsp60 ) many of which have well characterized roles in prion formation and propagation in yeast [5] . Similar functional proteins were identified as part the unfolded protein response ( 21-fold; P = 1 . 5×10−8 ) and stress response categories ( 4-fold; P = 2×10−5 ) . Over represented functions in the wild-type strain included protein synthesis ( 2-fold; P = 2×10−5 ) and protein folding and stabilization ( 9-fold; P = 2×10−10 ) . The protein synthesis category included a number of ribosomal proteins and translation factors which might be expected to associate with the Sup35 translation termination factor . Interestingly , proteins associated with the actin cytoskeleton were overrepresented in the tsa1 tsa2 mutant ( 8-fold; P = 4×10−6 ) , but not in the wild-type strain and were therefore subject to further analysis . Our data suggest an important role for the cortical actin cytoskeleton based on the cytoskeleton-related proteins which co-purify with Sup35 in the tsa1 tsa2 mutant strain ( Act1 , Sac6 , Crn1 , Abp1 , Arc40 , Arp2 , Arp3 and Arc35 ) . These data strongly implicate the Arp2/3 complex , which is a seven-protein complex containing two actin-related proteins ( Arp2 and Arp3 ) and five non-actin related proteins including Arc35 and Arc40 [45] . Abp1 is an actin-binding protein of the cortical actin cytoskeleton which is important for activation of the Arp2/3 complex [46] . Crn1 and Act1 were identified with purified Sup35 in both the wild-type and tsa1 tsa2 mutant strains . Crn1 is an actin-binding protein which regulates the actin filament nucleation and branching activity of the Arp2/3 complex through its interaction with the Arc35 subunit [47] . Act1 is encoded by ACT1 , a single essential gene in yeast . We validated our Sup35-interacting proteins for a number of proteins . Sup35-TAP was immunoprecipitated from the wild-type and tsa1 tsa2 mutant and possible interactions examined using Western blot analysis ( Fig 1B ) . This analysis confirmed that Abp1 , Arp3 and Sap190 co-purify with Sup35 in the tsa1 tsa2 mutant strain and Act1 co-purifies with Sup35 in both the wild-type and tsa1 tsa2 mutant strains . In order to test whether the Sup35-associated proteins affect oxidative stress-induced prion formation , mutant strains were constructed in a [PIN+][psi-] yeast strain ( 74D-694 ) which is commonly used to study yeast prion biology . We were unable to make arp2 , arp3 , arc35 or arc40 deletion mutants in this strain background in agreement with previous observations suggesting that components of the Arp2/Arp3 complex are essential for normal growth and viability [39 , 48 , 49] . We therefore focused on Abp1 and Crn1 , which were identified in our Sup35 immunopurification experiments . The induction of [PSI+] prion formation was quantified by analysing the formation of Ade+ colonies which arise due to nonsense suppression of the ade1-14 mutant allele . [PSI+]-mediated suppression can be differentiated from nuclear-encoded nonsense suppressor mutations by their elimination in guanidine hydrochloride ( GdnHCl ) . The control [PIN+][psi-] strain was grown in the presence of 100 μM hydrogen peroxide for 20 hours prior to scoring [PSI+] prion formation . This oxidative stress treatment increased the frequency of [PSI+] prion formation by approximately ten-fold ( Fig 2A ) , similar to our previous observations [38] . The basal frequency of spontaneous [PSI+] prion formation was reduced by approximately 150-fold in the abp1 mutant and 30-fold in the crn1 mutant . Furthermore , loss of ABP1 or CRN1 abrogated the peroxide-induced increase in [PSI+] prion formation ( Fig 2A ) . The nucleation of actin patches by the Arp2/3 complex is enhanced by the activity of nucleation promoting factors such as Abp1 [45 , 50–52] . We therefore tested whether loss of another nucleation promoting factor , Pan1 , similarly reduced the frequency of [PSI+] prion formation . The frequency of spontaneous [PSI+] prion formation was significantly reduced in the pan1 mutant and no induction was observed in response to hydrogen peroxide stress ( Fig 2A ) . We ruled out any effects on [PSI+] propagation by following [PSI+]-maintenance in [PSI+] versions of wild-type and abp1 mutant strains . After four days of culture , the formation of [psi-] cells was comparable in the wild-type and abp1 mutant strains ( Fig 2B ) . Since our Sup35-interacting proteins were identified in a tsa1 tsa2 mutant , we examined whether deletion of ABP1 affects the high frequency of spontaneous [PSI+] formation normally observed in a tsa1 tsa2 mutant [36] . The frequency of [PSI+] formation was increased approximately 30-fold in a tsa1 tsa2 mutant compared with the wild-type strain ( Fig 2C ) . [PSI+] formation was significantly decreased in a tsa1 tsa2 abp1 mutant suggesting that Abp1 is required for the elevated frequency of [PSI+] formation in a tsa1 tsa2 mutant . Since mutations which disrupt the cortical actin cytoskeleton reduce oxidative stress induced [PSI+] formation , we next examined whether the formation of another prion unrelated in sequence to the Sup35/[PSI+] prion , is similarly affected . Rnq1 can switch to the [PIN+] prion , which is formed at relatively high frequencies compared with the de novo formation of [PSI+] [53 , 54] . The de novo formation of [PIN+] prions , which is also dependent on oxidative status of the cells [37] , was detected in approximately 6% of control [pin-] cells , compared with 1 . 7% of abp1 mutant cells ( Fig 2D ) . Hydrogen peroxide treatment increased the frequency of [PIN+] prion formation in both strains , but the frequency of [PIN+] formation was significantly lower in the abp1 mutant compared with the wild-type strain . Loss of ABP1 therefore appears to decrease the spontaneous formation of both the [PSI+] and [PIN+] prions , but has a greater effect on the oxidative stress induced formation of [PSI+] compared with [PIN+] . [PSI+] prion formation can be induced by the overexpression of Sup35 in [PIN+] [psi-] strains due to the increased possibility for prion seed formation [4] . We therefore tested whether loss of ABP1 , CRN1 or PAN1 similarly affected the frequency of overexpression-induced [PSI+] prion formation . SUP35NM-GFP was induced for 24 hours and visible fluorescent aggregates were observed in 6 . 1% of wild-type cells examined ( Fig 3A ) . This included large fluorescent foci which arise due to decorating existing aggregates ( 3 . 7% ) , as well as rod- and ribbon-like aggregates ( 2 . 4% ) characteristic of the de novo formation of [PSI+] [39 , 43 , 55] . Loss of CRN1 or PAN1 resulted in modest decreases in Sup35 aggregation including the formation of fewer visible puncta and rod and ribbon-like aggregates ( Fig 3A ) . A stronger effect was seen with the abp1 mutant , with no rod or ribbon-like aggregates detected , although 1% of abp1 mutant cells still contained visible SUP35NM-GFP puncta ( Fig 3A ) . As a control , Western blot analysis was used to confirm that similar levels of Sup35NM-GFP were induced in the wild-type , abp1 , crn1 and pan1 mutant strains ( Fig 3B ) . Rhodamine-phalloidin staining was used to visualize the cortical actin cytoskeleton in the wild-type and mutant strains . Multiple bright rhodamine-phalloidin–stained puncta were detected in the wild-type strain ( S1 Fig ) , typical of the cortical actin patches normally observed in wild-type yeast cells [56 , 57] . In comparison , fewer , fainter cortical actin patches were detected in the abp1 and crn1 mutant strains , and fewer brighter patches were detected in pan1 mutant cells . In some cells , the formation of Sup35NM-GFP aggregates was coincident with actin patches but this was difficult to differentiate since rhodamine-phalloidin–stained puncta covered a large proportion of the cellular cortex ( S1 Fig ) . This is similar to previous studies which have shown that few Sup35-GFP puncta [39] or no Sup35-GFP puncta [55] co-localize with actin patches . Similarly some examples of co-localized Sup35NM-GFP/rhodamine-phalloidin–stained puncta were observed in the abp1 , crn1 and pan1 mutants . Although it is difficult to determine whether Sup35 aggregates are associated with the cortical actin cytoskeleton , our mutants which disrupt cortical actin patch formation do not appear to significantly affect overexpression-induced Sup35 aggregate formation . The presence of fluorescent SUP35-GFP aggregates is not necessarily indicative of [PSI+] prion formation since some cells with fluorescent dots will die , and some contain non-productive or non-amyloid aggregates [58] . We therefore quantified [PSI+] prion formation using the ade1-14 mutant allele as described above . [PSI+] formation was strongly induced by approximately 65-fold in the wild-type [PIN+][psi-] strain in response to Sup35 overexpression ( Fig 3C ) . A similar induction of [PSI+] prion formation was observed in the crn1 and pan1 mutants . This induction was reduced in the abp1 mutant compared with the wild-type strain , although a 21-fold increase in the frequency of [PSI+] prion formation was still observed in response to Sup35 overexpression ( Fig 3C ) . Taken together , these data indicate that in contrast to oxidative stress-induced prion formation , mutations which disrupt cortical actin patch formation only modestly affect the induction of [PSI+] prion formation in response to Sup35 overexpression . Treatment of yeast cells with latrunculin A ( LTA ) disrupts the formation of actin cables and patches [59] . This has been used to show that the actin cytoskeleton plays a role in [PSI+] propagation since disrupting the actin cytoskeleton by treatment with LTA causes the loss of [PSI+] from yeast cells [60] . This effect was observed at relatively high concentrations of LTA ( 40–200 μM ) and so we wanted to test whether a lower concentration of LTA , which does not significantly affect [PSI+] propagation , might disrupt oxidative-stress induced prion formation . We first tested the effect of growing a [PSI+] strain in the presence of 10 μM LTA for 20 hours . This concentration of LTA resulted in modest curing ( 9 . 3 ± 0 . 3% ) of [PSI+] and so we reasoned that we could use this concentration of LTA to test whether it affects the induction of [PSI+] . Rhodamine-phalloidin staining was used to visualize the cortical actin cytoskeleton and to confirm that the 10 μM LTA treatment disrupted the formation of actin patches ( Fig 4A ) . The wild-type [PIN+][psi-]-strain was grown in the presence of 10 μM LTA and 100 μM hydrogen peroxide for 20 hours to induce prion formation . We first examined Sup35 puncta formation by expressing SUP35NM-GFP for the final two hours of the oxidant treatment . Approximately 8% of wild-type cells contained visible Sup35 aggregates following exposure to hydrogen peroxide . This frequency was somewhat reduced in cells treated with 10 μM LTA , where 3 . 9% of cells examined contained visible Sup35 aggregates ( Fig 4B ) . [PSI+] prion formation was quantified under the same conditions , using the ade1-14-based assay . LTA treatment decreased the basal frequency of [PSI+] prion formation by 3-fold and also abrogated the oxidant-induced increase in the frequency of [PSI+] prion formation ( Fig 4C ) . For comparison , we examined whether a 10 μM LTA treatment affected overexpression-induced [PSI+] prion formation . [PSI+] formation was strongly induced in response to Sup35 overexpression and only a modest decrease in induction frequency was observed in the presence of LTA ( Fig 4D ) . Western blot analysis was used to confirm that LTA does not affect Sup35 overexpression ( Fig 4E ) . Thus , LTA strongly disrupts oxidant-induced [PSI+] prion formation , but only modestly affects overexpression-induced [PSI+] prion formation . Disrupting the actin cytoskeleton may potentially decrease oxidative stress-induced [PSI+] prion formation in a number of different ways including preventing Sup35 oxidative damage , altering the formation of Sup35 protein aggregates or by disrupting the formation of heritable [PSI+] propagons . We have previously shown that Sup35 oxidative protein damage is an important trigger for the formation of the heritable [PSI+] prion in yeast [37 , 38] . We therefore examined the extent of Sup35 oxidative damage in response to oxidative stress conditions , to determine whether disrupting the cortical actin cytoskeleton influences protein oxidative damage . Protein carbonylation is a commonly used measure of protein oxidative damage [61] . Carbonyl groups on proteins can be detected by Western blot analysis using an antibody against the carbonyl-specific probe DNPH . Using this assay we found that oxidative stress increased Sup35-carbonylation in response to oxidative stress as might be expected ( Fig 5A ) . A similar increase in carbonylation was detected in the wild-type , abp1 , crn1 and pan1 mutant strains suggesting that Sup35 protein oxidative damage is not altered in the mutant strains . To address whether disrupting cortical actin patch formation influences Sup35 aggregate formation , the subcellular distribution of Sup35 was examined biochemically during oxidative stress conditions . We used a protocol which separates soluble fractions from SDS-insoluble high-molecular weight forms [62] . Sup35 was predominantly detected in the soluble fraction in wild-type cells as expected . In response to oxidative stress conditions , a small fraction of Sup35 was present in an SDS-insoluble high-molecular weight form ( Fig 5B ) . Surprisingly , a significant proportion of Sup35 was already present in this SDS-insoluble high-molecular weight form in the abp1 mutant in the absence of stress , and there was no further increase in response to hydrogen peroxide treatment ( Fig 5B ) . This suggests that Sup35 aggregates in an abp1 mutant but is not converted to the heritable [PSI+] prion form . For comparison to Sup35 aggregation , we next examined the aggregation of a non-amyloidogenic protein in the abp1 mutant . We used a thermolabile allele of UBC9 fused to GFP ( GFP–Ubc9ts ) which was expressed under the control of the GAL1 galactose-regulated promoter [63] . At permissive temperatures , GFP–Ubc9ts is native and diffuse , whereas , shifting cells to 37°C causes GFP–Ubc9ts to misfold and to form puncta visible by fluorescence microscopy . These protein quality control structures are referred to as Q-bodies and do not contain amyloid aggregates [63] . Cells were grown in raffinose medium prior to inducing GFP–Ubc9ts expression for three hours following galactose addition . We found that approximately 90% of wild-type cells formed Q-bodies following a temperature shift to 37°C for 30 minutes ( Fig 5C ) . A similar number of cells containing Q-bodies were also detected in the abp1 mutant suggesting that loss of ABP1 does not affect non-amyloidogenic protein aggregation . Given that Sup35 is oxidized and aggregates in an abp1 mutant strain we assessed the intracellular localization of Sup35 in an abp1 mutant . We first quantified Sup35 puncta formation using the SUP35NM-GFP fusion to visualize aggregate formation . The wild-type and abp1 mutant strains were treated with 100μM hydrogen peroxide for 20 hours and SUP35NM-GFP expression induced for the final two hours by copper addition . Following hydrogen peroxide treatment , approximately 6% of wild-type cells contained visible Sup35 aggregates ( Fig 6A ) . This was reduced in abp1 mutant cells where 0 . 5% of cells examined contained visible Sup35 aggregates . The amyloidogenic [PIN+] prion form of Rnq1 localizes to the IPOD and is thought to influence the aggregation of other proteins [41 , 64] . This is further supported by the observation that during overexpression induced [PSI+] prion formation , approximately 60% of Sup35-RFP puncta co-localize with Rnq1-GFP puncta [58] . Newly formed Sup35-RFP puncta were found to perfectly co-localize with Rnq1-GFP puncta , whereas , mature Rnq1-GFP and Sup35-RFP puncta were found that did not co-localize . We therefore visualized the relationship of Sup35 and Rnq1 during oxidant-induced [PSI+] prion formation . [PIN+][psi-]-versions of the wild-type and abp1 mutant strains were grown in the presence of 100 μM hydrogen peroxide for 20 hours to induce prion formation . Sup35NM-RFP and Rnq1-GFP were expressed under the control of the GAL1 promoter and were induced for three hours to visualize Sup35 and Rnq1 aggregate formation . Similar to overexpression-induced [PSI+] prion formation , 76% of Sup35 puncta co-localized with Rnq1 puncta following oxidant treatment of the wild-type strain ( Fig 6B ) . In contrast , 17% of Sup35 puncta co-localized with Rnq1 puncta following oxidative stress conditions in the abp1 mutant strain . We also searched for rare Sup35 puncta formation in the absence of stress conditions and found a similar result where 71% of Sup35 puncta colocalized with Rnq1 puncta in a wild-type strain , compared with just 13% co-localization the abp1 mutant strain .
Our data suggest an important role for the Arp2/3 complex in prion formation since deletion of the ABP1 , CRN1 or PAN1 genes abrogate oxidant-induced [PSI+] prion formation ( Fig 2A and 2B ) . The Arp2/3 complex contains two actin-elated proteins ( Arp2 and Arp3 ) and five non-actin related proteins [45] . It is required for the motility and integrity of actin cortical patches , and for actin-dependent processes such as endocytosis and organelle inheritance . For example , conditional Arp2/3 mutants are deficient in actin patch formation suggesting that Arp2/3 is required for the assembly and organization of cortical actin filaments [65] . Nucleation-promoting factors such as Abp1 and Pan1 associate with the Arp2/3 complex and stimulate actin nucleation [50 , 52] . Crn1 regulates actin filament nucleation and the branching activity of the Arp2/3 complex through its interaction with the Arc35 subunit [47] . We found that disrupting the actin cytoskeleton by deletion of ABP1 , or treating cells with LTA , decreased the frequency of [PSI+] prion formation during oxidative stress conditions ( Figs 2A , 2B and 4C ) . This suggests that the cortical actin cytoskeleton is required for the conversion of oxidatively damaged Sup35 into its heritable prion form . Previous studies have implicated the Arp2/3 complex in prion formation and shown that Sup35 physically interacts with various proteins of the cortical actin cytoskeleton [39] . This includes Arp2 and Arp3 which were shown to interact with the N-terminal prion-forming domain of Sup35 during normal non-stress conditions using a two-hybrid assay . In contrast , we found that Arp2 and Arp3 are enriched in the Sup35-interacting proteins identified under oxidative stress conditions ( Fig 1 ) . This difference might arise since we have used native Sup35 expressed under the control of its own promoter , rather than a fragment of Sup35 fused to the DNA domain of Gal4 in a two-hybrid assay . Additionally , Gal4-activation domain fusion proteins are unlikely to assemble into normal actin complex structures in the nuclear two-hybrid assay [39] . An interaction between actin and overexpressed Sup35 was also demonstrated using immunoprecipitation experiments [39] similar to our finding that Sup35 interacts with actin in wild-type and tsa1 tsa2 mutant cells ( Fig 1 ) . We also found that Crn1 interacts with Sup35 in both the wild-type and tsa1 tsa2 mutant strains . Arp2/3 complex-related proteins therefore appear to be common Sup35-interacting proteins , although certain components show an increased interaction under oxidative stress conditions . Disrupting the cortical actin cytoskeleton , either genetically by deletion of ABP1 , CRN1 or PAN1 , or chemically , by treating cells with LTA , prevented oxidative stress-induced [PSI+] prion formation . In contrast , similar disruption of the cortical actin cytoskeleton did not significantly alter Sup35-overexpression-induced [PSI+] prion formation ( Figs 3C and 4D ) . This suggests that the mechanism underlying the conversion of the soluble protein to its amyloid form is different for overexpression-induced versus oxidative stress-induced prion formation . However , there are mechanistic similarities for these two induction pathways since cortical actin cytoskeleton mutants do influence overexpression-induced [PSI+] prion formation . For example , an abp1 mutant abrogated Sup35 ring formation , although the frequency of [PSI+] prion formation was still strongly induced in this mutant . Other studies have linked cortical actin patch formation with overexpression-induced [PSI+] prion formation . For example , loss of LAS17 or SAC6 abrogates overexpression-induced [PSI+] prion formation [66 , 67] . Las17 is a nucleation-promoting factor similar to Pan1 and Abp1 , although Las17 is a stronger nucleation-promoting factor compared with Pan1 and Abp1 [68] . Sac6 , an actin-bundling protein , is the major F-actin crosslinking protein in budding yeast [69] . Additionally , loss of other genes affecting actin patch formation including SLA1 , SLA2 and END3 decreases the frequency of Sup35 overexpression-induced [PSI+] formation [39 , 43] . Taken together , these data strongly implicate actin cytoskeletal networks in de novo [PSI+] prion formation , although there appear to be mechanistic differences between overexpression and oxidative stress-induced prion formation . Much previous research has made use of the Sup35NM-GFP fusion that we used to visualize [PSI+] prion formation . As with wild-type Sup35 , the Sup35NM-GFP fusion protein retains the unstructured PrD that has a high propensity to misfold [39 , 55 , 58 , 70] . Overexpression of Sup35NM-GFP is frequently used to visualize [PSI+] formation since the spontaneous formation of prions is very low making it difficult to observe during normal growth conditions . Oxidative stress conditions increase the frequency of [PSI+] formation making it possible to visualize [PSI+] formation without Sup35 overexpression . We observed cells containing puncta , which included examples of cells containing few ( sometimes one ) large dots and cells containing multiple smaller dots ( Fig 6 ) . The frequency of puncta formation was reduced in the abp1 mutant compared with a wild-type strain , although it was still significantly higher than the frequency of [PSI+] prion formation observed in the abp1 mutant in response to the same oxidative stress conditions . This is not surprising because Sup35-GFP puncta are not necessarily indicative of amyloidogenic-aggregation since GFP-puncta may arise due to amorphous aggregation or the formation of other granules such as stress granules . In fact , cell fractionation experiments revealed that Sup35 was more prevalent in an SDS-insoluble high-molecular weight form in an abp1 mutant during both non-stress and oxidative stress conditions compared with a wild-type strain ( Fig 5B ) . Together with our observation that Sup35 protein oxidation is similar in wild-type and abp1 mutant strains ( Fig 5A ) it does not seem likely that alterations in protein oxidation and aggregation account for the reduced frequency of [PSI+] prion formation in cortical actin cytoskeleton mutants . The Arp2/3 complex may therefore be required to provide the driving force for aggregate movement via growing actin filament formation . In the absence of the Arp2/3 complex , Sup35 aggregates are formed but are not transported to protein quality control compartments where prion formation occurs . Accumulating evidence suggests that eukaryotic cells defend themselves against protein aggregation by sequestering misfolded proteins into defined quality control compartments . Studies in yeast cells have revealed intricate protein quality control systems where insoluble proteins are partitioned into defined sites in the cell . Amyloid and amorphous aggregates are believed to be processed via distinct cytosolic protein inclusion bodies [40 , 41] . Upon proteasome inhibition , JUNQ serves as a sequestration site for ubiquitinated proteins , whereas , the IPOD sequesters terminally misfolded and amyloidogenic proteins . When proteasomes are active , misfolded proteins aggregate into Q-bodies [63 , 71 , 72] . Overexpressed Sup35NM-GFP is initially soluble , but the PrD has a high propensity to misfold and the misfolded protein is thought to be targeted to the IPOD via Myo2-based actin cable transport [64 , 73] . Other prion proteins including Ure2 and Rnq1 localize to the IPOD which facilitates nucleation and [PSI+] induction [41 , 64] . Accordingly , following Sup35NM-GFP overexpression , most newly induced Sup35 dots overlap with Rnq1 dots [58] . Similarly , we found that most oxidant-induced Sup35 dots co-localized with Rnq1 dots in a wild-type strain ( Fig 6 ) . In contrast , relatively few oxidant-induced Sup35 dots were found to co-localize with Rnq1 dots in an abp1 mutant . This may explain the reduced frequency of [PSI+] prion formation since Sup35 forms aggregates in response to oxidative stress conditions , but these aggregates do no efficiently localize to the IPOD in an abp1 mutant and hence do not form heritable propagons . The IPOD is formed on a perivacuolar site adjacent to the pre-autophagosome ( PAS ) where cells initiate autophagy [40 , 64] . One possibility is that the PAS may serve to recruit aggregated prion proteins prior to autophagic turnover . Previous studies may be complicated by overexpressing Sup35NM-GFP to follow [PSI+] formation which might overwhelm or impair autophagic flux , and it has been suggested that the IPOD provides a storage site for excess aggregates [73] . It is known that autophagy protects against de novo formation of [PSI+] and [PIN+] , and conversely , increasing autophagic flux by treating cells with the polyamine spermidine suppresses prion formation in mutants which normally show a high frequency of de novo prion formation [23 , 74] . Growth under anaerobic conditions in the absence of molecular oxygen prevents Sup35 protein damage and suppressed the high frequency of [PSI+] formation in an autophagy mutant further reinforcing the idea that oxidatively damaged Sup35 is cleared by autophagy to protect against the structural transitions favouring its conversion to the propagatable [PSI+] form [23] . Disrupting the cortical actin cytoskeleton not only prevented oxidant-induced [PSI+] formation , but also reduced the spontaneous frequency of de novo [PSI+] formation ( Fig 2A ) . Similarly , the co-localization of Sup35 and Rnq1 was also disrupted in an abp1 mutant during non-stress conditions ( Fig 6B ) . Mammalian and fungal prions arise de novo and for example sporadic forms of CJD account for approximately 80% of all recognised prion disease . However , the mechanism is poorly understood in molecular terms . It is therefore tempting to speculate that oxidant-induced prion formation provides a model for these spontaneous events . Elucidating the underlying mechanism of de novo prion formation following protein oxidation and how it depends on the balance between clearance of misfolded proteins mediated by autophagy and the formation of transmissible propagons , will enable a more mechanistic understanding of de novo prion formation .
The wild-type yeast strain 74D-694 ( MATa ade1-14 ura3-52 leu2-3 , 112 trp1-289 his3-200 ) was used for all experiments . Strains deleted for TSA1 ( tsa1::LEU2 ) and TSA2 ( tsa2::kanMX ) and containing Sup35 tagged at its C-terminus with a tandem affinity purification ( TAP ) tag have been described previously [37] . Strains deleted for ABP1 , CRN1 and PAN1 were constructed in 74D-694 using standard yeast methodology . The yeast plasmid CUP1-SUP35NM-GFP [URA3] expressing the Sup35NM domain conjugated to GFP under the control of the CUP1 promoter has been described previously [75] as has the yeast plasmid p2018 containing GAL1-SUP35NM-RFP [LEU2] [58] . Rnq1 was visualized using a yeast plasmid containing GAL1-RNQ1-EGFP [URA3] which expresses Rnq1-GFP under the control of the GAL1 promoter [76] . The yeast plasmid expressing RFP-tagged Hsp104 ( pRP1186 , Hsp104-RFP ) has been described previously [77] . A thermo-labile allele of UBC9 fused to GFP ( GFP–Ubc9ts ) was expressed under the control of the GAL1 galactose-regulated promoter [63] Strains were grown at 30°C with shaking at 180 rpm in rich YEPD medium ( 2% w/v glucose , 2% w/v bactopeptone , 1% w/v yeast extract ) or minimal SD ( 0 . 67% w/v yeast nitrogen base without amino acids , 2% w/v glucose ) supplemented with appropriate amino acids and bases . SRaf media contained 2% w/v raffinose and SGal media contained 2% w/v galactose . Media were solidified by the addition of 2% ( w/v ) agar . Strains were cured by five rounds of growth on YEPD agar plates containing 4 mM guanidine hydrochloride ( GdnHCl ) . Where indicated , strains were grown in the presence of 100 μM hydrogen peroxide for 20 hours prior to analysing [PSI+] prion formation . Cells were treated with 10 μM latrunculin to disrupt the actin cytoskeleton . The frequency of spontaneous [PSI+] prion formation was scored by growth in the absence of adenine . Diluted cell cultures were plated onto SD plates lacking adenine ( SD-Ade ) and incubated for 7–10 days . Colonies which grew on SD-Ade plates were counted and then picked onto new SD-Ade plates before replica-printing onto SD-Ade and SD-Ade containing 4mM GdnHCl . Colonies that grew on SD-Ade , but not on SD-Ade with GdnHCl were scored as [PSI+] . [PSI+] colonies were also scored by visual differentiation of red/white colony formation on YEPD plates and by the conversion of pink/white [PSI+] colonies to red [psi-] colonies on YEPD plates containing GdnHCl . For oxidant induced prion assays , cultures were grown in the presence of 100 μm hydrogen peroxide for 20 hours prior to scoring [PSI+] formation . For Sup35 overexpression-induced prion assays , cultures were grown in the presence of 50 μM copper sulphate for 20 hours to induce CUP1-SUP35NM-GFP expression prior to scoring [PSI+] formation [23] . De novo [PIN+] formation was performed as previously described [37] . Briefly , Sup35NM-GFP was overexpressed in [pin–] [psi–] strains in order to detect cells that generate [PSI+] de novo . Since [PSI+] formation is dependent on cells being [PIN+] [53] , the rate of [PIN+] formation was estimated based on the number of [PSI+] cells which arise . [PSI+] and [PIN+] formation was calculated based on the means of at least three independent biological repeat experiments . Rhodamine phalloidin staining of actin was performed as described previously [78] . Sup35 aggregate-formation was visualized using CUP1-SUP35NM-GFP following 50 μM copper sulphate addition to induce the CUP1 promoter [37] . Sup35 and Rnq1 co-localization experiments were conducted using plasmids containing GAL1-SUP35NM-RFP and GAL1-RNQ1-EGFP . Strains were grown in the presence of absence of 100μM hydrogen peroxide for 20 hours in SRaf media before switching to SGal media for three hours to induce the expression of GAL-SUP35NM-RFP and GAL1-RNQ1-EGFP . Visualization of the aggregation of a non-amyloidogenic protein , Ubc9 , was performed using GAL1-GFP–Ubc9ts . Strains were grown in SRaf media before switching to SGal media for 3 hours to induce the expression of GAL1-GFP–Ubc9ts . The temperature was shifted to 37°C for the final 30 minutes to trigger Ubc9 misfolding . Cells were washed and immobilised on 10% poly-L-lysine-coated slides . All images were acquired on a Delta Vision ( Applied Precision ) restoration microscope using a 100x/NA 1 . 42 Plan Apo objective and fluorescein isothiocyanate ( FITC ) and Texas Red band pass filters from the Sedat filter set ( Chroma ) . The images were collected using a Coolsnap HQ ( Photometrics ) camera with a Z optical spacing of 0 . 2μm . Raw images were then deconvolved using the Softworx software and maximum intensity projections of these deconvolved images are shown in the results . Sup35-TAP affinity purification was performed as described previously [37] . Sup35-interacting proteins were identified in the wild-type and tsa1 tsa2 mutant by mass spectrometry in triplicate for each strain . For protein identification , protein samples were run a short distance into SDS-PAGE gels and stained using colloidal Coomassie blue ( Sigma ) . Total proteins were excised , trypsin digested , and identified using liquid chromatography-mass spectrometry ( LC-MS ) performed by the Biomolecular Analysis Core Facility , Faculty of Biology , Medicine and Health , University of Manchester . Proteins were identified using the Mascot mass fingerprinting programme ( www . matrixscience . com ) to search the NCBInr and Swissprot databases . Final datasets for each condition were determined by selecting proteins that were identified in at least two of the three replicates . Protein extracts were electrophoresed under reducing conditions on SDS-PAGE minigels and electroblotted onto PVDF membrane ( Amersham Pharmacia Biotech ) . Bound antibody was visualised by chemiluminescence ( ECL , Amersham Pharmacia Biotech ) . Primary antibodies used were Sup35 [62] , Tsa1 [79] , Tef1 [80] , Abp1 ( abcam ) , Arp3 ( Santa Cruz Biotechnology ) , Sap190 ( affinity-purified polyclonal antibody raised against Sap190 peptides ) , Act1 ( ThermoFisher Scientific ) and Pgk1 ( ThermoFisher Scientific ) . The analysis of Sup35 aggregates by subcellular fractionation was performed essentially as described previously [62] . Briefly , exponential phase cells ( A600 ∼0 . 5 ) were collected by centrifugation , washed once with distilled water , and resuspended in buffer ST ( 10mM sodium phosphate buffer , pH7 . 5 , 250mM NaCl , 2% w/v SDS , 1% w/v Triton X-100 , 2mM PMSF ) . Cells were broken with glass beads using a Minibead beater ( Biospec Scientific , Bartlesville ) for 30 s at 4°C and centrifuged at 3000g for 3 minutes at 4°C . The supernatant ( total ) was centrifuged at 13000g for 45 minutes at 4°C to separate the soluble ( supernatant ) and insoluble ( pellet ) fractions . The soluble and insoluble fractions were resuspended in an equal volume of ST buffer prior to western blot analysis . Protein carbonylation was measured by reacting carbonyl groups with 2 , 4-dinitrophenyl-hydrazine ( DNPH ) based on previously described methods [81 , 82] . Briefly , exponential phase cells ( A600 ∼0 . 5 ) were broken with glass beads in 10% trichloroacetic acid ( TCA ) using a Minibead beater . The supernatant was centrifuged at 13000g for 15 mins at 4°C and the protein pellet washed with acetone to remove residual TCA . The pellet was dried and resuspended in 70 μl of 6% ( w/v ) SDS . 70 μl of 10mM 2 , 4-dinitrophenyl hydrazine ( DNPH ) in 10% trifluoroacetic acid was added and incubated at room temperature for 20 minutes . 45 μl of 2 M Tris/30% glycerol was added to the suspension and mixed to neutralize the DNPH reaction . SDS-PAGE sample buffer was added prior to western blot analysis using rabbit anti::DNPH ( Dako ) antibodies to detect carbonylation . Data are presented as mean values ± standard deviation ( SD ) . Statistical analysis for multiple groups was performed using one-way ANOVA with pair-wise comparisons of sample means via the Turkey HSD test . An unpaired two-tailed t-test was used for statistical analyses of two groups of samples . Results were considered statistically significant with a p-value less than 0 . 05 . | Prions are infectious agents which are composed of misfolded proteins and have been implicated in progressive neurodegenerative diseases such as Creutzfeldt Jakob Disease ( CJD ) . Most prion diseases occur sporadically and are then propagated in a protein-only mechanism via induced protein misfolding . Little is currently known regarding how normally soluble proteins spontaneously form their prion forms . Previous studies have implicated oxidative damage of the non-prion form of some proteins as an important trigger for the formation of their heritable prion conformation . Using a yeast prion model we found that the cortical actin cytoskeleton is required for the transition of an oxidized protein to its heritable infectious conformation . In mutants which disrupt the cortical actin cytoskeleton , the oxidized protein aggregates , but does not localize to its normal amyloid deposition site , termed the IPOD . The IPOD serves as a site where prion proteins undergo fragmentation and seeding and we show that preventing actin-mediated localization to this site prevents both spontaneous and oxidant-induced prion formation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"chemical",
"compounds",
"oxides",
"oxidative",
"stress",
"cell",
"disruption",
"mutation",
"hydrogen",
"peroxide",
"cellular",
"structures",
"and",
"organelles",
"cytoskeleton",
"research",
"and",
"analysis",
"methods",
"contractile",
"proteins",
"infectious",
"diseases",
"actins",
"mutant",
"strains",
"zoonoses",
"specimen",
"preparation",
"and",
"treatment",
"proteins",
"chemistry",
"biochemistry",
"cytoskeletal",
"proteins",
"mechanical",
"treatment",
"of",
"specimens",
"specimen",
"disruption",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"oxidative",
"damage",
"peroxides",
"prion",
"diseases"
] | 2017 | Disrupting the cortical actin cytoskeleton points to two distinct mechanisms of yeast [PSI+] prion formation |
Lassa and Junín viruses are the most prominent members of the Arenaviridae family of viruses that cause viral hemorrhagic fever syndromes Lassa fever and Argentine hemorrhagic fever , respectively . At present , ribavirin is the only antiviral drug indicated for use in treatment of these diseases , but because of its limited efficacy in advanced cases of disease and its toxicity , safer and more effective antivirals are needed . Here , we used a model of acute arenaviral infection in outbred guinea pigs based on challenge with an adapted strain of Pichindé virus ( PICV ) to further preclinical development of T-705 ( Favipiravir ) , a promising broad-spectrum inhibitor of RNA virus infections . The guinea pig-adapted passage 19 PICV was uniformly lethal with an LD50 of ∼5 plaque-forming units and disease was associated with fever , weight loss , thrombocytopenia , coagulation defects , increases in serum aspartate aminotransferase ( AST ) concentrations , and pantropic viral infection . Favipiravir ( 300 mg/kg/day , twice daily orally for 14 days ) was highly effective , as all animals recovered fully from PICV-induced disease even when therapy was initiated one week after virus challenge when animals were already significantly ill with marked fevers and thrombocytopenia . Antiviral activity and reduced disease severity was evidenced by dramatic reductions in peak serum virus titers and AST concentrations in favipiravir-treated animals . Moreover , a sharp decrease in body temperature was observed shortly after the start of treatment . Oral ribavirin was also evaluated , and although effective , the slower rate of recovery may be a sign of the drug's known toxicity . Our findings support further development of favipiravir for the treatment of severe arenaviral infections . The optimization of the experimental favipiravir treatment regimen in the PICV guinea pig model will inform critical future studies in the same species based on challenge with highly pathogenic arenaviruses such as Lassa and Junín .
A limited number of phylogenetically distinct viruses that belong to the Arenaviridae , Bunyaviridae , Filoviridae , and Flaviviridae families can cause a severe hemorrhagic fever syndrome that often results in death . Among the arenaviruses , two Old World ( Lassa and Lujo ) and several New World ( Junín , Machupo , Guanarito , Sabiá , and Chapare ) viruses are the etiologic agents of viral hemorrhagic fever in endemic areas of Africa and South America , respectively [1] , [2] , [3] . Estimates of the number of yearly Lassa virus infections and associated deaths in West Africa range up to 2 , 000 , 000 and 10 , 000 , respectively [4] . The highest disease burden in the New World is associated with Junín virus ( JUNV ) infection in the Pampas agricultural regions of Argentina . Although an effective vaccine has curtailed the number of cases of Argentine hemorrhagic fever ( AHF ) , cases continue to be reported annually [5] . Immune plasma has been used with some success but has been associated with a late neurological syndrome [6] . Ribavirin ( 1-β-d-ribofuranosyl-1H-1 , 2 , 4-triazole-3carboxamide ) is the only licensed antiviral with reported activity against Lassa virus ( LASV ) , JUNV , and Machupo virus ( MACV ) [7] , [8] , [9] , and could be used off-label in the event of an emergency [10] . Although the adverse effects in humans treated with ribavirin are generally considered to be mild and reversible with termination of treatment [11] , [12] , [13] , [14] , teratogenicity and embryotoxicity are of concern [15] , [16] . In addition , intravenous ribavirin is not widely available and is often very expensive [17] . Favipiravir ( T-705; 6-flouro-3-hydroxy-2-pyrazinecarboxamide ) is a pyrazine derivative presently being developed clinically for the treatment of influenza virus infections . Orally administered favipiravir has shown efficacy in experimental mouse and hamster models of arenavirus , phlebovirus , flavivirus , and influenza virus infections [18] , [19] , [20] , [21] , [22] , [23] . We were able to demonstrate a limited protective antiviral effect when treating advanced Pichindé virus ( PICV ) infection in hamsters [24] . In Vero cell culture experiments , we have demonstrated micromolar range activity of favipiravir against the JUNV vaccine strain , Candid 1 [19] . We have now confirmed this activity with a pathogenic strain of JUNV , as well as isolates of other South American hemorrhagic fever viruses [25] . Although several groups have recently reported on the development of mouse models of LASV and JUNV infection , these systems are based on challenge of immunocompromised animals [26] , [27] . Guinea pig infection models have been described for LASV , JUNV and Guanarito virus ( GTOV ) [28] , [29] , [30] , [31] , and are the best-suited small animal models to further investigate the activity of favipiravir . Due to the maximum containment requirement and high costs associated with conducting studies with highly pathogenic arenaviruses , the aim of the present work was to evaluate favipiravir in the guinea pig PICV infection model [32] , [33] , as a means to optimize treatment conditions to assist in the planning of future studies in biosafety level 4 ( BSL-4 ) containment . Our PICV stock derived from a single additional passage of a previously described guinea pig-adapted virus [33] was uniformly lethal in outbred guinea pigs , and we characterized the natural history of disease to establish the model in our laboratory for use in the evaluation of favipiravir .
All animal procedures complied with USDA guidelines and were conducted at the AAALAC-accredited Laboratory Animal Research Center at Utah State University under protocol 1393 , approved by the Utah State University Institutional Animal Care and Use Committee . Outbred male Hartley strain guinea pigs weighing ∼300–350 g were obtained from Charles River ( Wilmington , MA ) . Animals were sorted prior to the start of all experiments so that the average group weight was similar across all groups . For all experiments , IPTT-300 electronic transponders were subcutaneously implanted for identification and temperature measurement in conjunction with the DAS 6002 scanner ( BMDS , Seaford , DE ) . Guinea pig-adapted PICV , passage 18 ( p18 ) , was provided by Dr . Robert Tesh ( World Reference Center for Emerging Viruses and Arboviruses , University of Texas Medical Branch , Galveston , TX ) . The p18 strain was derived from 2 additional passages of a p16 guinea pig spleen suspension of the CoAn 4763 Munchique strain obtained from the U . S . Army Medical Research Institute of Infectious Diseases ( USAMRIID ) . A p19 spleen homogenate was prepared from a single ill p18-infected guinea pig euthanized on day 12 post-infection . The p19 stock ( ∼4 . 8×106 plaque-forming units ( PFU ) /ml ) was used for all challenge studies . Sequencing of viral RNA isolated from the p19 stock was performed by SeqWright DNA Technologies Services ( Houston , TX ) using standard fluorescent dye-terminator DNA sequencing chemistry following RT-PCR amplification . GenBank accession numbers for the p19 S and L segments are JN378747 and JN7378748 , respectively . Favipiravir ( T-705 ) was provided by the Toyama Chemical Company , Ltd . ( Tokyo , Japan ) . Ribavirin was supplied by ICN Pharmaceuticals , Inc . ( Costa Mesa , CA ) . Both were suspended in GERBER NatureSelect 1st FOODS carrot food ( ingredients: carrots and water ) for oral administration . Favipiravir toxicity was assessed in guinea pigs following twice-daily treatments for 10 days . Groups of five guinea pigs each were dosed orally with 500 , 250 , 100 and 0 ( placebo ) mg/kg/day of favipiravir . Treatments were administered using 1 ml tuberculin syringes by placement of the doses in carrot food vehicle in the back of the oral cavity . During the 10-day dosing period and for seven days following , guinea pigs were monitored closely for signs of toxicosis and weights and temperatures were recorded daily . Seven days after the final dose was administered , animals were euthanized by CO2 asphyxiation , whole blood and serum were collected for hematology and blood chemistry analyses , and necropsies and pathological examination were performed at the Ross A . Smart Veterinary Diagnostic Laboratory ( Logan , UT ) . Virus titrations were performed to determine 50% and 90% lethal doses ( LD50 and LD90 ) of the p19 PICV stock . Groups of three to four guinea pigs each were challenged by bilateral intraperitoneal ( i . p . ) injections with log10 PICV quantities ranging from 0 . 05 to 50 , 000 PFU prepared in minimal essential medium ( MEM ) . Body weight and temperature , and morbidity and mortality were monitored for 28 days following infection . Clinical signs of illness were weight loss , pyrexia , ruffling of fur , and lethargy . In this and all other experiments , animals were considered moribund and euthanized when they lost 20% of their starting body weight or their body temperature dropped to 36°C or less . For survival analysis , animals were counted as dead the day after euthanasia . LD50 and LD90 values were determined by regression analysis . Based on the titration data , guinea pigs were infected with 500 PFU of PICV for the natural history study and the challenge efficacy experiments . For the natural history study , PICV-challenged guinea pigs were sacrificed daily ( n = 3/day ) , with the exception of day 10 and 11 of infection , on which 4 and 2 guinea pigs were euthanized , respectively , due to one of the day 11 animals having reached the 20% weight loss euthanasia criteria on day 10 . Whole blood ( in both citrate- and EDTA-coated tubes; Sarstedt Inc . , Newton , NC ) , serum , livers , lungs , kidneys , spleens , and brains were harvested . Whole spleens were weighed prior to sectioning . Sections from each tissue were preserved in 10% formalin and sent to the Ross A . Smart Veterinary Diagnostic Laboratory ( Logan , UT ) for histologic analysis . The other sections were stored at −80°C and virus titers determined as described below . Whole blood was analyzed for coagulation and hematologic parameters , and serum was analyzed for viremia and comprehensive blood chemistry as described below . Two independent studies were performed to investigate the efficacy of favipiravir in guinea pigs challenged with PICV . In the first study , groups of 8 guinea pigs each were treated twice daily with 100 and 30 mg/kg/day of favipiravir on days 4–7 of infection . Due to continued deterioration of the animals despite favipiravir treatment , the doses were increased to 300 and 90 mg/kg/day , respectively , for the remainder of the treatment schedule ( days 8–17 ) . For comparison , 8 guinea pigs each were treated twice daily with 50 mg/kg/day of ribavirin or carrot food placebo on days 4–17 after challenge . Body weight and temperature , and morbidity and mortality were monitored for 29 days post-challenge . Serum was collected by saphenous vein puncture from all animals on day 11 of infection , with the exception of two animals in the placebo group that had to be put down on day 10 . Serum was analyzed for viral burden , aspartate aminotransferase ( AST ) and alanine aminotransferase ( ALT ) as described below . Serum was also collected from surviving animals at the conclusion of the experiment ( day 29 ) for virus titer analysis . In the second study , groups of 8 guinea pigs each were treated orally with 300 or 150 mg/kg/day favipiravir , or vehicle placebo , divided into two daily doses for 14 days beginning on day 7 after infection . In an attempt reduce the high viral loads encountered at the start of treatment , the 150-mg/kg/day favipiravir group received a loading dose of 300 mg/kg on the first day , with a shift to the lower maintenance dose thereafter . This strategy is commonly used in the clinic and has been employed for the treatment of cases of Lassa fever and AHF [17] , [34] . A group of 7 guinea pigs received 50 mg/kg/day of ribavirin , with treatment also starting on day 7 and given twice daily for 14 days . Guinea pigs were monitored for signs of illness and weights and temperatures were recorded as previously described . Serum was collected on day 10 from all animals by saphenous vein puncture and analyzed for viremia and AST concentration . Sera , spleens , livers , lungs , kidneys , and brains were collected from surviving animals at the end of the study ( day 36 ) for virus titer analysis . Virus titers were determined using an infectious cell culture assay as previously described [19] . Briefly , tissues were homogenized 1∶10 w/v in MEM . Serum and homogenized tissue samples were serially log10 diluted and plated in triplicate wells on Vero cell monolayers ( American Type Culture Collection , Manassas , VA ) in 96-well microtiter plates . Plates were incubated for 7 days and viral cytopathic effect ( CPE ) was determined for calculation of 50% endpoints by the Reed-Muench method [35] . Assay detection range was 1 . 8 to 8 . 5 log10 50% cell culture infectious dose ( CCID50 ) /ml of serum or 0 . 1 g of tissue . Whole blood in citrate-coated tubes was analyzed for prothrombin time ( PT ) and activated partial thromboplastin time ( aPTT ) using the VetScan VSpro and PT/aPTT cartridges ( Abaxis Inc . , Union City , CA ) . Due to a technical problem , PT clot times were not detected in two of three animals on day 1 . For the characterization and efficacy studies , whole blood in EDTA-coated tubes was analyzed for hematology using the VetScan HMT ( Abaxis Inc . ) . Hematology factors included white blood cells ( WBC ) , lymphocytes ( Lym ) , monocytes ( Mon ) , granulocytes ( Gra ) , red blood cells ( RBC ) , mean corpuscular volume ( MCV ) , hematocrit ( Hct ) , mean corpuscular hemoglobin ( MCH ) , mean corpuscular hemoglobin concentration ( MCHC ) , red cell distribution width ( RDW ) , hemoglobin ( Hb ) , platelets ( PLT ) , mean platelet volume ( MPV ) , plateletcrit ( PCT ) , and platelet distribution width ( PDW ) . For the favipiravir toxicity study , hematology was performed using a HEMAVET HV 950 ( Drew Scientific , Dallas , TX ) and parameters evaluated were equivalent to those above , except WBC were further broken down to include neutrophils ( Neu ) , eosinophils ( Eos ) and basophils ( Bas ) . Serum was analyzed for comprehensive blood chemistry or individual analytes using the DRI-CHEM 4000 chemistry analyzer ( Heska , Loveland , CA ) . Blood analytes included AST , amylase ( AMY ) , alkaline phosphatase ( ALP ) , ALT , blood urea nitrogen ( BUN ) , calcium ( Ca ) , creatinine ( CRE ) , gamma-glutamyl transferase ( GGT ) , glucose ( GLU ) , total protein ( TP ) , total bilirubin ( TBIL ) , albumin ( ALB ) , cholesterol ( CHO ) , and inorganic phosphate ( IP ) . For the favipiravir toxicity study , blood chemistry analysis was performed using the VetScan VS2 and comprehensive diagnostic profile rotors ( Abaxis Inc . ) . Similar analytes were profiled with the exception of CHO and the addition of phosphate ( PHOS ) , sodium ( Na+ ) , potassium ( K+ ) , and globulin ( GLOB ) . Three guinea pigs were treated by placement of a 50 mg/kg dose of favipiravir , suspended in carrot food vehicle , towards the back of the oral cavity with a 1 ml syringe . Plasma was collected from each animal at 15 min , 30 min , 1 h , 2 h , and 4 h by saphenous vein puncture . Each sample was mixed with equal volume of 1∶1 methanol∶acetonitrile for deproteinization . Samples were centrifuged ( 10 , 000× g ) for 10 min and supernatants transferred to new tubes for evaporation . The contents were then resuspended in HPLC buffer for analysis as previously described [24] . Favipiravir plasma concentrations were extrapolated using a standard curve from samples containing known amounts of favipiravir . Area under the curve ( AUC ) analysis and half-life ( t1/2 ) estimation were performed using Prism ( GraphPad Software , La Jolla , CA ) . The Mantel-Cox log-rank test was performed to analyze the survival data . Hematology , blood chemistry , virus titer , coagulation , and spleen weight data were analyzed using one-way analysis of variance ( ANOVA ) followed by Bonferroni multiple comparison test . Correlation of PLT count with other platelet parameters ( PCT , MPV and PDW ) were performed according to the Pearson rank correlation method . All statistical evaluations were done using Prism ( GraphPad Software ) .
Because of the reported variability in lethality caused by PICV infection of outbred guinea pigs [32] , [36] , [37] , [38] , [39] , we sought to establish a uniformly lethal model that would facilitate the evaluation of favipiravir . We prepared a virus stock from a single passage of the p18 guinea pig-adapted strain in Hartley guinea pigs . The p19 PICV was derived from the spleen of clinically ill guinea pig with an advanced infection on day 12 post-challenge . The p19 stock was found to be highly virulent causing severe disease in guinea pigs with 100% mortality at challenge doses ≥500 PFU ( Figure 1 ) . The LD50 of the p19 stock was ∼5 PFU with an LD90 of ∼200 PFU . Complete sequence analysis of the L and S segments from the p19 virus stock revealed only a single substitution in the consensus sequence compared to the previously reported p18 sequences [39] . The substitution was present in the L segment and was heterogeneous matching either the p18 or p2 sequences . We next investigated the natural history of disease in guinea pigs challenged with 500 PFU of the p19 virus . PICV-induced disease in guinea pigs was marked by elevated temperatures beginning on day 3 after infection , with weight loss and anorexia becoming evident by day 6 ( Figure 2A ) . Clot times for extrinsic ( PT ) and intrinsic ( aPTT ) coagulation pathways were increased as the infection progressed past day 7 ( Figure 2B ) . Hematological analysis revealed additional alterations affecting the coagulation system . Marked thrombocytopenia was observed starting on day 5 post-infection with dramatic decreases in platelet counts ( PLT; Figure 2C , Table S1 ) and plateletcrit ( PCT; Table S1 ) ; the latter measure being directly related to the total number of platelets . Platelet distribution width ( PDW ) , a marker for platelet activation [40] , concomitantly decreased and became increasingly variable as the infection progressed and , due to low PLT , could not be measured past day 7 ( Table S1 ) . On the other hand , mean platelet volume ( MPV ) , considered an indicator of platelet function [41] , did not significantly change over the course of the course of the study and did not correlate with PLT ( Table S1 ) . PLT concentration had a strong correlation with both PCT ( r = 1 . 0; P = 0 . 0001 ) and PDW ( r = 0 . 83; P = 0 . 0015 ) . The only other notable hematologic findings observed were spikes in total WBC and granulocytes ( Gra ) on day 7 of infection; however , both parameters returned to the normal range the following day ( Table S1 ) . In addition to the depleted PLT and PCT levels , and the prolonged coagulation times observed during the course of acute PICV infection , we found blood in the stools of several ill animals with advanced clinical disease signs ( ruffling of fur and anorexia ) . Evidence of hemorrhaging within individual tissues was not observed upon pathological examination , suggesting that internal bleeding into tissues was not contributing substantially to the demise of the animals . Severe hemorrhagic manifestations are not often seen in human arenaviral hemorrhagic fever cases [42] , [43] . Comprehensive blood chemistry analysis revealed a dramatic increase in serum AST concentration starting on day 8 of PICV infection ( Figure 2D ) . In contrast , serum ALB decreased gradually through the course of the infection ( Table S2 ) , which may reflect alterations in vascular permeability or nutritional status . No other significant changes in the blood chemistry parameters evaluated were observed ( Table S2 ) . Viremia and tissue virus titers were also assessed on a daily basis . As shown in Figure 2E , PICV replication was observed in all tissues examined . Onset of viremia occurred on day 5 post-challenge and persisted through day 11 . The spleen supported vigorous replication as PICV could be detected as early as day 1 , and having the greatest viral loads throughout the acute infection period , with titers of >108 CCID50/g . The liver , lungs , and kidneys also had substantial viral burdens that crested on day 8 and persisted through the end of the study . Infectious virus was detectable in the brain of several animals starting on day 7 , but the low titers may be attributable to blood-borne virus . This trend was similarly observed in LASV-infected and PICV ( p8 strain ) -infected strain 13 guinea pigs [28] , [32] . Consistent with the high viral loads measured in spleen tissue homogenates , the spleens of PICV-infected guinea pigs were grossly enlarged and weighed significantly more on days 5–11 of infection compared to day 1 ( Figure 2F ) . As described in previous studies using lower passage strains of adapted PICV and inbred strain 13 guinea pigs [32] , [44] , we also found the liver and spleen to be most affected histologically by p19 PICV infection ( not shown ) , although the degree of damage in these tissues was minor and less than would have been expected based on viral titers ( Figure 2E ) . Few to moderate numbers of necrotic cells were observed in the interstitium and periarteriolar sheaths of the spleen and acute multifocal hepatic necrosis was observed as early as day 6 and necrotic areas were more prominent in later days of infection ( not shown ) . Several guinea pigs exhibited hepatic lipidosis on days 10 and 11; however , this change is most likely due to the mobilization of fat for energy as the sick animals greatly reduce food consumption . All other tissues appeared normal upon histologic examination . Our observations are consistent with those noted in human Lassa fever , in which histopathologic findings are generally not severe enough to account for death [45] . Having established a uniformly lethal guinea pig PICV infection model and characterized the timing of disease progression , the second objective was to evaluate the efficacy of oral favipiravir using this model . Because of a small palatal ostium , oral gavage of guinea pigs is very difficult and generally contraindicated . Thus , we devised a method to treat the animals by suspending test drug in a carrot food vehicle for administration as described in detail in the methods section . Using this method of drug delivery , we were able to confirm gastric absorption of favipiravir ( AUC = 44 µg/ml h; t1/2 = 1 . 42 h ) with peak plasma levels in the range of 40 µg/ml ( 256 µM ) within 15–30 min of treatment with a 50-mg/kg dose ( Figure S1 ) . To reach favipiravir concentrations well above the reported 50% effective concentration ( EC50 ) of ∼17 µM for arenavirus inhibition in cell culture [19] , [46] , we treated PICV-infected guinea pigs with 100 or 30 mg/kg/day . This dose range has also been previously shown to be effective in treating PICV infection in hamsters by oral gavage [24] . Favipiravir , placebo , and ribavirin ( positive control ) treatments were initiated 4 days after challenge and dosed twice daily for 14 days ( Figure 3A ) . Because we did not observe any signs of toxicity following 10 days of treatment with a favipiravir dose of 500 mg/kg/day ( Table S3 ) , we extended the duration of treatment to facilitate complete clearance of the virus . Unexpectedly , despite 4 days of therapy , guinea pigs treated with favipiravir began to lose weight , became lethargic , and developed high fevers similar to animals treated with placebo ( Figure 3B , C ) . In contrast , guinea pigs treated with ribavirin ( 50 mg/kg/day ) did not develop fever or show signs of illness . Consequently , we decided to triple the dose of favipiravir for the remaining 10 days of therapy starting on the morning of day 8 post-infection , as no clinical signs of adverse effects , histopathology ( not shown ) , or changes in laboratory values were observed in toxicity studies when guinea pigs were treated with up to 500 mg/kg/day ( Table S3 ) . The increase in the favipiravir dose resulted in a rapid reduction in fever in the 300-mg/kg/day group ( Figure 3C ) , with all animals recovering completely ( Figure 3A ) , as reflected by robust weight gain at a rate greater than ribavirin ( Figure 3B ) . The 90-mg/kg/day dose of favipiravir provided a reduced , yet significant degree of protection . Notably , the reduction in fever seen starting on day 9 for both the placebo and low-dose favipiravir groups is principally due to decreasing temperatures in animals as they become moribund ( Figure 3C ) . Guinea pigs receiving ribavirin therapy all survived the challenge , but recovered more slowly as demonstrated by the shallower weight gain trend relative to the high-dose favipiravir group . Serum AST and virus titers measured on day 11 were significantly lower in all drug-treated groups compared to the placebo , with a clear dose response evident with the favipiravir-treated animals ( Figure 3D , E ) . All surviving guinea pigs had undetectable virus in the serum at the conclusion of the experiment ( not shown ) . A second efficacy study was conducted wherein treatment was initiated one week after virus challenge , to assess the ability of favipiravir to treat more advanced PICV infection and disease in guinea pigs exhibiting clear clinical signs of illness , including anorexia and sustained fever . Similar to the first experiment , we were able to successfully treat lethal PICV challenge with a 300 mg/kg/day favipiravir regimen ( Figure 4A ) , even when delaying treatment until a time when animals presented with considerable viral loads , splenomegaly , and were thrombocytopenic , febrile , and losing weight ( Figure 2A , C , E , F ) . The intermediate dose of 150 mg/kg/day of favipiravir , which included a 300-mg/kg loading dose on the first day of treatment , also provided a significant level of protection ( Figure 4A ) . All animals treated with ribavirin at 50 mg/kg/day survived the challenge . Most guinea pigs began to lose weight beginning on day 6 after PICV challenge ( Figure 4B ) . The high-dose favipiravir group decreased in body weight until day 9 , at which point the animals began to steadily recover throughout the remainder of the experiment . Guinea pigs in the intermediate-dose favipiravir treatment group that succumbed to illness steadily decreased in body weight , while surviving counterparts began to recover as early as day 11 . Guinea pigs in the ribavirin treatment group maintained fairly steady weights through day 21 , when they began to gradually gain weight through the rest of the observation period . In both studies , slower rate of weight gain compared to the high-dose favipiravir group was observed . Animals in the placebo group sharply decreased in body weight prior to succumbing to the infection . Most guinea pigs presented with elevated temperatures of >40°C by day 6 of PICV infection ( Figure 4C ) . Similar to the initial efficacy study , fever was almost immediately reduced following the onset of therapy with 300 mg/kg/day of favipiravir . As before , it is important to note that prior to succumbing to the infection , animals in the intermediate-dose favipiravir and placebo groups dropped in temperature as they approached the terminal stage of the disease . Serum AST , reflective of the extent of tissue damage , was significantly lower in all drug-treated groups compared to the placebo when measured on day 10 post PICV challenge ( Figure 4D ) . Serum virus titers were also significantly lower in all treatment groups , with average titers of 6 . 4 log10 CCID50/ml in the placebo , and 4 . 3 , 5 . 1 , and 4 . 8 log10 CCID50/ml in the high- and intermediate-dose favipiravir , and ribavirin groups , respectively ( Figure 4E ) . The treated guinea pigs that survived the 36 day observation period were all found to be devoid of systemic and tissue virus titers ( not shown ) .
Because there are presently no other small animal model options based on challenge of immune competent animals with highly pathogenic viral hemorrhagic fever-causing arenaviruses , future studies evaluating favipiravir as an antiviral therapy for the treatment of severe arenaviral infections will likely first be done in guinea pig JUNV , GTOV , or LASV infection models that require BSL-4 maximum biocontainment facilities [47] . To this end , we sought to establish a guinea pig PICV infection model similar to those previously described [32] , [33] . PICV infection in guinea pigs has proven to be useful for the study of acute arenaviral disease [37] , [38] , [39] , [44] and for preclinical efficacy evaluations [48] , [49] , as the virus can be handled safely in BSL-2 containment . There have been mixed reports on the lethality of guinea pig-adapted PICV in the readily available Hartley outbred guinea pig strain [32] , [36] , [37] , [38] , [39] . In addition to other factors , the variation in the stringency of the criteria used to define the terminal endpoints likely contributed to the reported variability . We found that our PICV stock prepared from the spleen of a clinically ill Hartley guinea pig sacrificed on day 12 of infection was uniformly lethal when inoculated at i . p . at doses of 500 PFU or more , but the genetic make-up of the virus did not vary substantially from previously reported sequences . Because most of the previous studies with PICV in guinea pigs that investigated pathogenesis , pathophysiology , virology , and clinical chemistry have used inbred animals that varied in age , gender , and/or infectious dose of PICV , we first characterized the p19 PICV infection in male 350 g outbred guinea pigs . The results provided a detailed picture into the evolution of the clinical disease and pathophysiology specific to the present model , facilitating the evaluation of favipiravir with the goal of demonstrating anti-arenavirus activity in guinea pigs and optimizing the dosing method , level , frequency , and duration of treatment to inform future BSL-4 studies . With an understanding of the natural history of disease in our guinea pig p19 PICV infection model , we assessed the anti-arenavirus activity of orally administered favipiravir . Although previously we were able to show limited efficacy with favipiravir in treating advanced PICV disease in a hamster model [24] , here we were able to demonstrate complete protection from lethal disease in guinea pigs when treatment was initiated well after the onset of fever and the beginning of weight loss . Notably , PICV-infected hamsters do not develop fever , and weight loss is not apparent until the day prior to death [50] . To this end , the insidious progression of the human disease is better reflected in guinea pigs , including the development of fever , a hallmark of the clinical arenaviral hemorrhagic fever diseases . Importantly , favipiravir significantly reduced viremia and systemic AST concentrations , which are prognostic indicators for severe disease and lethality in Lassa fever patients [51] . The efficacy studies conducted provide the foundation for more advanced evaluations with favipiravir employing guinea pig infection models based on challenge with authentic arenaviral hemorrhagic fever viruses , such as the JUNV-guinea pig model actively being used to study AHF [30] . Although we were able to effectively treat PICV-infected guinea pigs with favipiravir , the dosage required was higher than expected based on previous studies in hamsters [19] , [24] . A dose of 300 mg/kg/day was needed to achieve 100% survival in guinea pigs challenged with PICV , whereas a 100-mg/kg/day therapeutic regimen afforded the same level of protection when initiated as late as day 5 following infection in hamsters [24] . To put these findings into perspective , the dosage presently being used for clinical evaluation of favipiravir for influenza treatment in humans is 40 mg/kg/day on the first day , with a reduction to 27 mg/kg/day for an additional 4 days . Nevertheless , the equivalent 300-mg/kg/day guinea pig dosage based on body surface area translation [52] would be 65 mg/kg/day . For hamsters , the 100-mg/kg/day dosage would be 14 mg/kg/day . The higher dose requirement of favipiravir in the guinea pig model is not likely due to differences in the virus stocks used , as our analysis of the An 4763 strain used to infect hamsters and the p19 guinea-pig adapted strain used in the present study were equally sensitive to the inhibitory effects of favipiravir in cell culture ( M . Mendenhall , unpublished data ) . The evidence to date suggests that favipiravir acts as a purine nucleoside analog targeting the viral RNA-dependent RNA polymerase [25] , [53] . We hypothesize that the difference in effective dosage of favipiravir is most likely due to a less efficient conversion of the parent compound , T-705 , to its active triphosphate form ( T-705RTP ) in guinea pigs , and/or a more rapid systemic elimination . However , we cannot rule out the possibility of better absorption and more favorable biodistribution of favipiravir in 0 . 4% carboxymethyl cellulose vehicle when given by oral gavage , as previously described for hamsters [19] , [24] . Notwithstanding , our present findings , coupled with having recently demonstrated favipiravir activity in cell culture against JUNV , GTOV , and MACV [25] , have us positioned to investigate activity against arenaviruses that are the etiologic agents of Argentine and Venezuelan hemorrhagic fevers in humans , in existing guinea pig infection models [30] , [31] . | Several viruses in the Arenaviridae family cause severe life-threatening hemorrhagic fever syndromes , which are considered neglected tropical diseases in endemic areas of Africa and South America . Ribavirin , the only licensed antiviral indicated for use has limited efficacy when treating advanced cases of disease and is associated with toxicity . In the present study , we use a model of acute arenaviral disease in guinea pigs based on infection with an adapted strain of the Pichindé arenavirus ( PICV ) to further preclinical development of a promising broad-spectrum antiviral drug candidate , favipiravir . Oral favipiravir was highly effective in the treatment of sick animals with marked fevers , as all recovered fully from lethal PICV infection even when therapy was initiated one week after virus challenge . Antiviral activity and reduced disease severity was evidenced by dramatic reductions in serum virus loads and serum aspartate aminotransferase , an enzyme released into the bloodstream following tissue damage and a marker for severe arenaviral infections . Moreover , a sharp decrease in fever was observed shortly after the onset of treatment . Our findings support further development of favipiravir for the treatment of severe arenaviral infections , for which there are presently no safe and effective therapies for treating advanced cases of disease . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"viral",
"hemorrhagic",
"fevers",
"infectious",
"diseases",
"venezuelan",
"hemorrhagic",
"fever",
"neglected",
"tropical",
"diseases",
"argentine",
"hemorrhagic",
"fever",
"viral",
"diseases",
"lassa",
"fever"
] | 2011 | Effective Oral Favipiravir (T-705) Therapy Initiated after the Onset of Clinical Disease in a Model of Arenavirus Hemorrhagic Fever |
The neovascular ( wet ) form of age-related macular degeneration ( AMD ) leads to vision loss due to choroidal neovascularization ( CNV ) . Since macrophages are important in CNV development , and cytomegalovirus ( CMV ) -specific IgG serum titers in patients with wet AMD are elevated , we hypothesized that chronic CMV infection contributes to wet AMD , possibly by pro-angiogenic macrophage activation . This hypothesis was tested using an established mouse model of experimental CNV . At 6 days , 6 weeks , or 12 weeks after infection with murine CMV ( MCMV ) , laser-induced CNV was performed , and CNV severity was determined 4 weeks later by analysis of choroidal flatmounts . Although all MCMV-infected mice exhibited more severe CNV when compared with control mice , the most severe CNV developed in mice with chronic infection , a time when MCMV-specific gene sequences could not be detected within choroidal tissues . Splenic macrophages collected from mice with chronic MCMV infection , however , expressed significantly greater levels of TNF-α , COX-2 , MMP-9 , and , most significantly , VEGF transcripts by quantitative RT-PCR assay when compared to splenic macrophages from control mice . Direct MCMV infection of monolayers of IC-21 mouse macrophages confirmed significant stimulation of VEGF mRNA and VEGF protein as determined by quantitative RT-PCR assay , ELISA , and immunostaining . Stimulation of VEGF production in vivo and in vitro was sensitive to the antiviral ganciclovir . These studies suggest that chronic CMV infection may serve as a heretofore unrecognized risk factor in the pathogenesis of wet AMD . One mechanism by which chronic CMV infection might promote increased CNV severity is via stimulation of macrophages to make pro-angiogenic factors ( VEGF ) , an outcome that requires active virus replication .
Angiogenesis , the formation of blood vessels , plays a critical role in embryonic development , wound healing , and normal physiologic processes associated with natural growth and development . On the other hand , new blood vessel growth ( neovascularization ) contributes to a number of pathologic conditions that include atherosclerosis and tumor formation [1] , [2] . The eye is also particularly sensitive to neovascularization during which abnormal blood vessel growth within retinal or choroidal tissues leads to vision loss or blindness . Sight-threatening diseases of the eye associated with abnormal neovascularization include diabetic retinopathy [3] , retinopathy of prematurity [4] , and age-related macular degeneration ( AMD ) [5] . Of these , AMD is the leading cause of severe irreversible central vision loss and legal blindness in individuals 65 years of age or older in the United States and other developed countries [6]–[9] . Since the number of elderly persons will double by 2020 , AMD is expected to become a major public health problem . Two forms of AMD are recognized [5]–[10] . The non-neovascular form ( also known as “dry” or “nonexudative” ) represents an early form of AMD usually associated with little visual acuity loss . It is characterized by atrophic abnormalities of the retinal pigment epithelium ( RPE ) and drusen , small lesions at the level of the RPE that contain granular and vesicular lipid-rich material . Over time , however , this form of AMD often progresses to the neovascular ( also known as “wet” or “exudative” ) form of AMD that results in significant vision loss due to the appearance of choroidal neovascularization ( CNV ) . Although the precise events that contribute to the development of AMD remain uncertain , recent studies have implicated various immunological and inflammatory mechanisms . For example , complement deposition has been demonstrated within drusen and the choriocapillaris , and several publications have demonstrated that polymorphisms in complement factor-H are associated with an increased risk of AMD [11]–[14] . Several investigators have also identified macrophages in association with drusen as well as choroidal neovascular membranes [15]–[18] suggesting a role for macrophages in the pathophysiology of both forms of AMD . In support of this hypothesis , we [19] , and others [20] , have shown in a mouse model of experimental CNV that depletion of macrophages significantly decreases the size and severity of lesions . Macrophages are immune cells of monocyte origin that are classically associated with innate immune responses , particularly inflammation [21] , but they may also exhibit pro-angiogenic as well as anti-angiogenic activities [22] . Thus , macrophages may exist in different activation states [23] , and individuals may therefore vary in activation states as defined by expression of cytokine transcripts as well as inducible cytokine production [24] . In fact , phenotypically polarized macrophages have been broadly classified into two main groups: classically activated ( M1 ) macrophages and alternatively activated ( M2 ) macrophages that are further subdivided into three subtypes [25] . Moreover , M1 macrophages exhibit an anti-angiogenic phenotype , whereas M2 macrophages exhibit a pro-angiogenic phenotype [22]–[27] . It is therefore possible that individuals with macrophages of one activation state will have a relative protective effect in AMD while individuals with macrophages of another activation state will be at risk for progressive complications . This idea is supported by our observation that the presence of highly activated macrophages is associated with a 5-fold increase in risk of having wet AMD [24] . The mechanism of macrophage activation is clearly multifactorial involving genetics , systemic health cofactors , and environmental cofactors including infection [15] , [28]–[30] . Infectious pathogens have been implicated in several vascular diseases , especially atherosclerosis [29]–[32] . Chlamydia pneumoniae , human cytomegalovirus ( HCMV ) , and Helicobacter pylori all have been implicated in promoting severity of atherosclerosis and inducing complications such as myocardial infarction [31] , [33] , [34] . These findings prompted us to perform a seroepidemiologic study to investigate a possible association for these infectious pathogens with neovascular AMD , a study that subsequently demonstrated a significant association with high HCMV IgG serum titers [35] . This finding differed from that of Kalayoglu and coworkers [36] whose study suggested an association between chlamydia and neovascular AMD . Given our clinical findings [35] and our long-standing interests in the immunology and pathogenesis of cytomegalovirus retinal disease [37] , we sought to test the hypothesis that chronic infection with HCMV , a common β-herpesvirus that targets myeloid lineage cells that give rise to activated macrophage cell populations in tissues [38] , is a heretofore unrecognized risk factor for onset and progression of neovascular AMD . This hypothesis was tested herein using an established mouse model of laser-induced CNV [39] to evaluate the effect of systemic infection by murine cytomegalovirus , a mouse β-herpesvirus whose genomic structure and cellular/tissue tropisms parallels those of HCMV [38] , on the severity of CNV lesions during acute and chronic virus infection . We observed that mice with chronic MCMV infection developed more severe CNV , and that macrophages collected from chronically infected animals were activated as determined by expression of high levels of transcripts for a number of pro-inflammatory and pro-angiogenic factors , especially the pro-angiogenic cytokine vascular endothelial growth factor ( VEGF ) . In vitro studies confirmed that MCMV infection of a mouse macrophage cell line resulted in significant upregulation of VEGF mRNA and VEGF protein production . Subsequent in vivo and in vitro studies using the antiviral ganciclovir demonstrated that increased production of VEGF by splenic macrophages collected from chronically infected mice and by MCMV-infected mouse macrophages grown in culture was ganciclovir-sensitive , findings that suggest that active virus replication is indeed required for stimulation of VEGF production by macrophages .
We first explored the effect of systemic MCMV infection on experimental CNV . Three times relative to systemic virus inoculation were chosen for this study , one at time of acute systemic infection ( 6 days postinfection ) and two at times of chronic systemic infection ( 6 weeks postinfection and 12 weeks postinfection ) [38] . Groups of C57BL/6 mice were inoculated intraperitoneally with a sublethal dose of MCMV , and their eyes were subjected to laser-induced CNV [39] at 6 days , 6 weeks , or 12 weeks after systemic MCMV infection . In this study , control mice received UV-inactivated MCMV . Four weeks after laser treatment , propidium iodide-stained flatmounts of the posterior pole were prepared of all laser-treated eyes , and groups were compared for severity of CNV . Results are shown in Figure 1A–D and Figure 2A . As expected , mice inoculated with UV-inactivated MCMV exhibited small CNV lesions ( 1 . 8±0 . 1 disc areas ) . In comparison , mice inoculated with infectious MCMV exhibited CNV lesions of increased size . Lesion size also increased with time after MCMV infection . Whereas mice with MCMV infection of 6-days duration exhibited CNV lesions of moderate enlargement ( 2 . 7±0 . 2 disc areas ) four weeks after laser treatment , progressively larger lesions were observed in mice with MCMV infection of 6-weeks ( 3 . 1±0 . 2 disc areas ) and 12-weeks ( 4 . 4±0 . 6 disc areas ) duration prior to laser treatment ( Figure 2A ) . A statistical comparison of lesion sizes observed in animals with MCMV infection of 12-weeks duration versus control animals revealed significance ( p = <0 . 0001 ) . The frequency of large lesions also increased with progression of MCMV . Whereas only 10% of the total number of lesions in mice inoculated with UV-inactivated virus exceeded 2 . 2 disc areas ( Figure 2B ) ( representing the 95% confidence interval for lesion size in control mice ) , 57 . 5 , 92 , and 100% of animals with systemic MCMV infection of 6-days , 6-weeks , and 12-weeks duration prior to laser treatment , respectively , developed large CNV lesions . Similar findings were observed when flatmounts were evaluated for degree of vascular size ( Figure 2C ) and vascularity ( data not shown ) , although cellular density remained constant ( Figure 2D ) . Taken together , these results suggest that systemic MCMV infection results in more severe CNV in mice , even during acute infection where a trend in increased severity is also observed . The most severe and statistically significant of CNV lesions , however , is found in mice with chronic MCMV infection of 12-weeks duration . Histopathologic analysis of CNV lesions ( Figure 3A ) paralleled those of flatmount findings . When compared with mice that received UV-inactivated virus ( Figure 3B ) , mice with laser-induced CNV at 12 weeks after infection demonstrated a near doubling of CNV surface area ( 64 , 977±7 , 267 pixels2 versus 119 , 149±8 , 578 pixels2; p = <0 . 0004 ) . Importantly , neither active nor chronic systemic MCMV infection changed the typical morphological appearance of experimental CNV . There was an absolute absence of MCMV-induced cytopathology as well as retinal necrosis . Since systemic MCMV infection was found to induce more severe CNV , we explored the possibility that direct virus infection of choroidal tissues might be responsible for this outcome . Choroidal tissues as well as several key tissues and cell populations known to be associated with MCMV pathogenesis [38] , [40] were sampled for detection of MCMV DNA using primers for virus-specific immediate-early 1 ( IE1 ) and glycoprotein H ( gH ) gene sequences in PCR assays . As expected , samples of spleen tissue , lung tissue ( Figure 4 ) , and salivary gland tissue as well as splenic macrophages collected from animals at time of acute MCMV infection ( 6 days postinfection ) or chronic MCMV infection ( 12 weeks postinfection ) provided positive signals for MCMV-specific DNA ( Table 1 ) indicating extensive systemic MCMV infection . Purified CD34+ cells of bone marrow origin collected from mice with acute and chronic MCMV infection were also positive for MCMV DNA . In comparison , choroidal tissues from eyes of acutely infected mice were indeterminant for MCMV-specific DNA , and MCMV-specific DNA sequences could not be detected in choroidal tissues from eyes of chronically infected animals ( Table 1 ) . The apparent lack of MCMV infection of choroidal tissues taken from chronically infected animals was confirmed by our inability to recover infectious virus from whole eyes of parallel groups of chronically infected animals individually homogenized and individually inoculated onto MEF monolayers . Thus , no evidence was found for direct MCMV infection of choroidal tissues or subsequent active virus replication within the eye at time of chronic infection when CNV was found to be most severe . Alternatively , systemic MCMV infection could contribute to increased severity of CNV indirectly via activation of macrophages to produce pro-angiogenic factors . It is well known for both HCMV and MCMV that peripheral blood monocytes are vehicles for systemic dissemination of virus during acute infection [38] , [40] , and these cells can harbor virus during chronic infection and with the potential to become activated macrophages within various tissues [23] . Since sufficient numbers of macrophages could not be collected from individual eyes of acutely and chronically infected mice for analysis , we subjected enriched populations of splenic F4/80+ macrophages collected from acutely and chronically infected mice to real time RT-PCR assay for detection and quantification of transcripts to several pro-inflammatory and pro-angiogenic cytokines and mediators associated with neovascular AMD . In this study , results were compared with baseline transcript levels established for splenic macrophages collected from mice inoculated with UV-inactivated virus . As shown in Table 2 , significant differences were observed in the patterns of synthesis for a number of macrophage-associated transcripts examined during acute and chronic MCMV infection . Of importance was the finding of significant upregulation of VEGF ( p = 0 . 04 ) and VEGFR1 ( p = 0 . 05 ) transcripts that progressed from acute to chronic infection . This was associated with a concomitant significant upregulation of matrix metalloproteinase-9 ( MMP-9 ) ( p = 0 . 02 ) , cyclooxygenase-2 ( COX-2 ) ( p = 0 . 05 ) , and tumor necrosis factor alpha ( TNF-α ) ( p = 0 . 03 ) transcripts , but only during chronic infection . Interestingly , macrophage-associated VEGFR2 transcript was significantly downregulated ( p = 0 . 01 ) during acute and chronic infection . These results suggest that an increase in CNV size and severity during chronic MCMV infection may be due to virus-induced activation of macrophages that favor neovascularization . Although splenic macrophages collected from mice with chronic systemic MCMV infection exhibited an approximate 20-fold increase in VEGF mRNA levels when compared with splenic macrophages collected from control mice , it is possible that increased VEGF mRNA production was not due to active MCMV replication . To explore directly the ability of mouse macrophages to produce increased amounts of VEGF during active virus replication , monolayers of IC-21 mouse macrophages , a macrophage cell line of C57BL/6 origin [41] , were either mock-infected ( control ) , treated with LPS ( positive control ) , inoculated with UV-inactivated MCMV ( negative control ) , or inoculated with infectious MCMV at a dose resulting in a low level of infection ( 2 . 5 PFU/cell ) . All monolayers were quantified at 24 hr and 48 hr later for levels of TNF-α mRNA and VEGF mRNA by quantitative RT-PCR assay . Results are shown in Figure 5A . When compared with mock-infected monolayers , monolayers of IC-21 mouse macrophages were activated by LPS treatment as demonstrated by large increases in VEGF mRNA and TNF-α mRNA levels , but parallel monolayers inoculated with UV-inactivated virus produced only low levels of VEGF mRNA and TNF-α mRNA suggesting little-to-no activation . In comparison , MCMV-infected monolayers of IC-21 mouse macrophages at 24 hr postinfection showed a 13-fold increase in VEGF mRNA levels , but interestingly failed to duplicate an increase in TNF-α mRNA production as seen in LPS-treated MCMV-infected monolayers . The same pattern of cytokine mRNA synthesis was observed in MCMV-infected IC-21 mouse macrophages at 48 hr postinfection . At this time after virus infection , VEGF mRNA levels were >50-fold greater than levels found in mock-infected monolayers ( p = <0 . 04 ) , but TNF-α mRNA levels were only ∼3-fold greater . This pattern of activation is consistent with a M2 phenotype of macrophage activation [25] since further analysis of MCMV-infected IC-21 macrophages when compared with mock-infected cells revealed increased levels of IL-10 and IL-1RA mRNA levels , equivalent levels of IL-23 mRNA production , and no detectable IL-21 mRNA production ( data not shown ) . Confirmation that MCMV infection of IC-21 mouse macrophages resulted not only in a significant increase in VEGF mRNA levels , but also in a significant increase in VEGF protein , was provided by ELISA analysis of supernatants collected at 48 hr postinfection ( p = 0 . 01 ) ( Figure 5B ) . Taken together , these results provide proof-of-principal that the increase in VEGF mRNA levels observed in mice with chronic systemic infection could arise from direct MCMV infection , active virus replication , and subsequent macrophage activation associated with the M2 phenotype , a pro-angiogenic phenotype [25] . To further explore VEGF production by MCMV-infected mouse macrophages in culture , monolayers of IC-21 mouse macrophages were either MCMV-infected ( 2 . 5 PFU/cell ) or mock-infected and subjected to immunostaining analysis for detection of VEGF production and for quantification of VEGF-positive cells at 24 hr and 48 hr postinfection . Results are shown in Figure 6 . When compared with MCMV-infected and mock-infected cells reacted with control antibody , MCMV-infected cells reacted with anti-VEGF antibody at 24 hr and 48 hr postinfection exhibited positive cytoplasmic staining for VEGF . Whereas staining was generally stronger in MCMV-infected cells at 48 hr postinfection when compared with MCMV-infected cells at 24 hr postinfection , the strongest staining was observed in foci of MCMV-infected cells at 48 hr postinfection showing early stages of cytopathology during plaque formation . It is noteworthy that individual macrophages at 48 hr postinfection not involved in plaque formation were also VEGF positive . Quantification studies revealed that ∼55% and ∼93% of MCMV-infected IC-21 mouse macrophages exhibited positive staining for VEGF at 24 hr and 48 hr postinfection , respectively , whereas mock-infected controls showed background levels of VEGF production of ∼10% . We found in studies described above that splenic macrophages collected from acutely infected mice and chronically infected mice produced significantly more VEGF mRNA when compared with splenic macrophages collected from control mice ( Table 2 ) . In addition , monolayers of MCMV-infected IC-21 mouse macrophages produced significantly more VEGF mRNA and VEGF protein when compared with monolayers of mock-infected cells ( Figures 5–6 ) . If stimulation of VEGF production in vivo and in vitro is induced directly by active virus replication , we hypothesized that stimulation of VEGF production should be sensitive to treatment with ganciclovir , an antiviral that inhibits HCMV and MCMV replication at the level of virus DNA synthesis [42] , [43] . To test this hypothesis in vivo , a study was performed in which groups of C57BL/6 mice were either inoculated intraperitoneally with a sublethal dose of MCMV or mock-infected with maintenance medium ( control ) . Unlike the study summarized in Table 2 , it is noteworthy that this study did not use inoculation with UV-inactivated virus as a control . At 12 weeks postinfection , groups of MCMV-infected mice or mock-infected mice were treated intraperitoneally with ganciclovir ( 40 mg/kg/day ) for 7 days [43] . Parallel groups of untreated control MCMV-infected mice or mock-infected mice were not treated with ganciclovir , but instead received daily intraperitoneal injections of phosphate-buffered saline for 7 days . Following the 7-day regimen of ganciclovir or phosphate-buffered saline treatment , splenic macrophages were collected from ganciclovir-treated and untreated chronically infected mice and compared by quantitative real time RT-PCR assay for levels of VEGF mRNA and TNF-α mRNA . In agreement with our previous study summarized in Table 2 , splenic macrophages collected from untreated chronically infected mice showed dramatic stimulation of VEGF mRNA production as well as TNF-α mRNA production ( Figure 7 ) . In fact , the degree of stimulation for both VEGF mRNA and TNF-α mRNA production was greater than that observed in our previous study ( Table 2 ) , especially with respect to TNF-α mRNA production . This difference might be due to the different controls used in the two separate studies , UV-inactivated virus ( Table 2 ) versus maintenance medium ( Figure 7 ) . When compared with untreated virus-infected animals , however , ganciclovir treatment resulted in a significant inhibition of VEGF mRNA production ( p = ≤0 . 009 ) , specifically an approximate 44-fold decrease in VEGF mRNA production . A similar degree of inhibition of TNF-α mRNA production was also observed in the presence of ganciclovir treatment ( p = ≤0 . 009 ) . Importantly , this significant inhibition of VEGF mRNA and TNF-α mRNA production in ganciclovir-treated animals could not be attributed to drug-related toxicity since splenic macrophages collected from these animals were found to be >95% viable at time of enrichment and just prior to RT-PCR assay when analyzed by the trypan blue exclusion and MTS assays ( data not shown ) . An in vitro study was performed to confirm our in vivo ganciclovir treatment findings . Monolayers of IC-21 mouse macrophages were inoculated with either a low dose of MCMV ( 2 . 5 PFU per cell ) or mock-infected , and all monolayers were treated at 1 hour postinfection with either 0 , 15 , 30 , or 60 uM of ganciclovir . At 24 hr postinfection , all monolayers were harvested and subjected to quantitative RT-PCR assay for comparison of VEGF mRNA levels . In agreement with in vivo ganciclovir treatment findings , increasing amounts of the antiviral reduced in a relatively dose-dependent manner the amounts of VEGF mRNA produced when compared with untreated MCMV-infected mouse macrophages ( Figure 8A ) . As expected , untreated MCMV-infected mouse macrophages produced VEGF mRNA at increased levels , and at levels equivalent to that observed for MCMV-infected macrophages at 24 hr postinfection as shown in Figure 5A . With increasing doses of ganciclovir , however , amounts of VEGF mRNA were dampened , ultimately being reduced by ∼5-fold at the highest doses of ganciclovir , 30 and 60 uM . This reduction in VEGF mRNA production could not be attributed to drug-induced toxicity since mock-infected ganciclovir-treated IC-21 mouse macrophages remained >95% viable at all doses tested when subjected to the trypan blue exclusion and MTS assays ( data not shown ) . Since ganciclovir treatment appeared to reduce , but not eliminate , VEGF mRNA production by MCMV-infected IC-21 mouse macrophages in a relatively dose-dependent manner , we sought to determine if VEGF could be detected within ganciclovir-treated MCMV-infected IC-21 mouse macrophages , albeit at reduced levels , with increasing doses of drug . We therefore performed an immunostaining study to visualize VEGF production within monolayers of MCMV-infected IC-21 mouse macrophages at 24 hr postinfection following treatment with 0 , 15 , 30 , or 60 uM of ganciclovir at 1 hr after virus inoculation . Results are shown in Figure 8B . In agreement with previous findings ( Figure 6 ) , MCMV-infected IC-21 mouse macrophages not treated with drug exhibited prominent cytoplasmic staining for VEGF . In comparison , increasing doses of ganciclovir treatment appeared to dampen VEGF protein production within the MCMV-infected cells . Nonetheless , positive staining for VEGF could still be detected within MCMV-infected IC-21 mouse macrophages treated with the highest dose of ganciclovir , 60 uM . Only faint background staining or no detectable staining was observed in parallel control monolayers of mouse macrophages that were either not infected with virus or virus-infected and reacted with control antibody ( data not shown ) . Taken together , these in vitro findings suggest that VEGF mRNA and VEGF protein production during MCMV infection of IC-21 mouse macrophages are indeed ganciclovir-sensitive , although VEGF production is not completely eliminated in the presence of the antiviral . Moreover , the significant reduction of VEGF mRNA and VEGF protein during ganciclovir treatment of MCMV-infected IC-21 mouse macrophages in culture is in agreement with in vivo findings , thereby supporting the hypothesis that upregulation of VEGF mRNA within splenic macrophages collected from MCMV-infected mice with chronic infection is due to active virus replication .
The number of investigations of angiogenesis in the eye has increased significantly in recent years due to findings that neovascularization of the retina and choroid plays a central role in the development of a number of major blinding diseases . These include AMD as well as diabetic retinopathy , polypoidal choroidal vasculopathy , myopic choroidal neovascularization , neovascular glaucoma , retinopathy of prematurity , and ocular tumorigenesis ( all reviewed in [44] ) . Since a seroepidemiologic clinical study by us demonstrated an apparent association between HCMV infection and neovascular AMD [35] , we used an experimental C57BL/6 mouse model of CNV to test the hypothesis that systemic MCMV infection will contribute to the severity of CNV . It has not escaped our attention that mouse strain-dependent factors might play a factor in CNV development during systemic MCMV infection since macrophages from C57BL/6 mice ( a prototypical Th1 mouse strain ) and macrophages from BALB/c mice ( a prototypical Th2 mouse strain ) exhibit distinct M1- or M2-dominant responses [45] . Nonetheless , our results collectively showed that systemic MCMV infection of C57BL/6 mice did indeed result in more severe CNV , and , more importantly , chronically infected mice showed the greatest severity of CNV . Although MCMV DNA sequences could not be detected within choroidal tissues of chronically infected animals , splenic macrophages collected from chronically infected animals produced increased amounts of transcripts to several pro-inflammatory and pro-angiogenic cytokines including VEGF . That MCMV infection of mouse macrophages will modulate a pro-angiogenic M2 phenotype that included significant stimulation of VEGF production was shown directly by in vitro studies using a mouse macrophage cell line of C57BL/6 origin . Further evidence that virus infection induced stimulation of VEGF production both in vivo and in vitro was provided by ganciclovir treatment studies that demonstrated sensitivity of VEGF production to the antiviral both in vivo and in vitro . Thus , our findings are novel with respect to chronic eye disease since they provide for the first time new data that suggests that chronic cytomegalovirus infection can contribute to the pathogenesis of wet AMD , possibly via activation of macrophages towards a pro-angiogenic phenotype and stimulation of VEGF production . While we have not yet demonstrated in our model direct visualization of MCMV-infected , VEGF-producing macrophages associated with areas of CNV , several observations would argue that this is a likely occurrence . Firstly , we [19] and others [20] have shown previously that macrophages are essential for development of CNV . Secondly , we have shown previously in the context of MCMV retinitis that IC-21 macrophages infected with a β-galactsidase-expressing LacZ recombinant MCMV will travel to ocular tissues of C57BL/6 mice following tail vein injection [46] . Finally , we show herein that MCMV infection of IC-21 macrophages stimulates VEGF production , a stimulation that is also observed in splenic macrophages collected from chronically infected mice with severe CNV . Future in vivo immunostaining studies will directly address this important issue . The concept that infectious agents might contribute to the pathogenesis of vascular diseases has become an intense and controversial area of investigation . Two major hypotheses have emerged . One hypothesis proposes that vascular disease is caused by direct infection of the target tissue [36] , while the second hypothesis proposes a bystander effect caused by infection at a distant tissue [47]–[49] . In atherosclerosis , direct infection of the atheromatous plaque by Chlamydia pneumoniae has been suggested as a stimulus for recruitment of inflammatory cells . Arguing against this hypothesis , however , are antibiotic treatment trials designed to suppress Chlamydia pneumoniae infection and failing to demonstrate a measurable clinical effect on preventing myocardial infarction or other sequelae [34] . On the other hand , patients with chronic periodontal infection and inflammation have provided evidence suggesting that chronic infection at a distant site may play a role in vascular disease . In this patient population , infection by a variety of different organisms appeared to lead to more severe vascular disease [34] , [50] , [51] . Since in our study , MCMV-specific DNA sequences could not be detected in choroidal tissues of eyes with the most severe choroidal neovascularization , we propose a similar bystander hypothesis for the role of HCMV infection in chroroidal neovascularization of the eye . HCMV is a common β-herpesvirus that persists for the life of its host following primary infection . While chronic HCMV infection of healthy , immunologically normal persons was initially thought to have no significant disease consequence , chronic HCMV infection has now been associated with a growing number of long-term diseases that include the vascular disease atherosclerosis , restenosis following angioplasty , transplant vascular sclerosis associated with chronic allograft rejection of solid organ grafts ( reviewed in [52] ) , and possibly tumor formation ( reviewed in [53] ) . Evidence for a link between HCMV and vascular disease was first provided by Melnick , DeBakey , and coworkers [54] when virus antigen was detected within arterial tissues from carotid artery plaques obtained from patients with atherosclerosis . Since this fundamental observation of ∼20 years ago , however , it has been difficult to determine the precise mechanisms by which HCMV might participate in the pathophysiology of vascular disease because the etiologies of chronic diseases are complex and multifactorial . Nonetheless , seropositive HCMV persons are two to three-times more likely to develop coronary artery disease when compared with HCMV seronegative patients [55] . In support of this association are recent findings that 76% of patients with ischemic heart disease have detectable HCMV DNA within their vascular tissues [56] , and up to 53% of carotid artery atherosclerotic lesions are positive for HCMV DNA [57] . A number of animal studies have also provided compelling evidence that cytomegalovirus plays an important role in the pathophysiology of atherosclerosis , including several studies that have demonstrated more severe atherosclerosis in apoE −/− mice following systemic MCMV infection [58]–[61] . While an association has been recognized between cytomegalovirus infection and atherosclerosis , the strongest association of cytomegalovirus in vascular disease is with the development of restenosis and transplant vascular sclerosis . Several clinical studies have shown that HCMV infection is involved in accelerating both acute and chronic graft failure in all types of solid organ transplants by promoting vascular disease associated with rejection [52] , probably by virus originating from the vasculature of transplanted organs from HCMV seropositive donors [62] . For example , HCMV infection was shown to double the 5-year rate of graft failure in cardiac allograft recipients due to accelerated transplant vascular sclerosis [63] . Similarly , kidney transplant allograft survival was decreased in asymptomatic HCMV-infected recipients during the first 100 days after transplantation when compared with recipient patients who had no evidence for HCMV infection , an outcome suggesting that HCMV infection , even when asymptomatic , has a negative impact on graft survival [64] . These clinical findings have been supported by a number of rat models of heart , kidney , lung , and small bowel transplantation in which infection with rat cytomegalovirus ( RCMV ) significantly decreased the mean time to graft failure while concomitantly increasing the degree of vasculopathy within the allograft tissue [65] , [66] . Neovascularization is a complex , multi-step process of angiogenesis that rapidly takes place in response to inflammation and tissue injury , and involves many cell types , cytokines , chemokines , and proteases that work in concert to form new blood vessels from existing blood vessels . In brief ( reviewed in [52] ) , angiogenesis is initiated by release of pro-angiogenic factors from activated endothelial cells and tissue-resident macrophages , followed by removal of pericytes that surround the existing blood vessels . This results in the breakdown of the basement membrane of the existing blood vessel wall through activation of several proteases including matrix metalloproteinases ( MMPs ) . The release of extracellular remodeling proteins during continued degradation of the blood vessel wall leads to the release of growth factors that promote endothelial cell migration toward the angiogenic stimulus and ultimately mediates endothelial cell proliferation that drives the formation of neotubules . These neotubules in turn release additional growth factors such as platelet-derived growth factor ( PDGF ) that recruit vascular smooth muscle cells and pericytes that stabilize the newly formed blood vessel . Importantly , pro-angiogenic M2 macrophages have been shown recently to act as bridging cells that promote the fusion of neotubules into one continuous blood vessel [67] . Cytomegalovirus infection could therefore enhance neovascularization at various stages of angiogenesis through a number of direct and indirect mechanisms . Monocytes are the primary target in vivo for HCMV ( and MCMV and RCMV ) infection [68] , [69] . They serve as a site for virus latency and persistence [70] , and help to disseminate virus throughout the host including the vasculature . When virus-infected monocytes enter the vasculature , they mature , and during the maturation process to become macrophages , they initiate an activation program that also serves to stimulate virus replication [71] . In this manner , infected macrophages may disseminate virus to other cells of the vasculature that are involved in angiogenesis and vascular disease . These include endothelial cells , smooth muscle cells , pericytes , and fibroblasts [52] . Given this complexity , the precise temporal relationship between virus infection of individual cell types and disease pathogenesis remains obscure and difficult to determine . Nonetheless , it is known that HCMV infection of endothelial cells induces the expression of adhesion molecules ICAM-1 and VCAM-1 [72] that serve to magnify transendothelial cell migration of inflammatory cells including monocytes . These monocytes become resident macrophages that promote angiogenesis by secretion of VEGF and other pro-angiogenic factors such as IL-6 [52] . During virus replication , the HCMV-encoded chemokine receptor US28 also plays a prominent yet multifaceted role in angiogenesis . Firstly , US28 has been shown to stimulate VEGF production directly by induction of COX-2 via activation of the NF-κB pathway [73] . Secondly , this HCMV-encoded chemokine receptor promotes the migration of macrophages in response to the CX3CL1 chemokine Fractalkine [52] , a function that may help to attract additional HCMV-infected macrophages to areas of inflammation and thereby amplify angiogenesis . Thirdly , US28 also promotes the migration of vascular smooth muscle cells [74] , but does so by binding to CC-chemokines and not Fractalkine [75] . Thus , US28 appears to stimulate the migration of both macrophages and vascular smooth muscle cells , but in a ligand-dependent manner . Whereas US28-induced migration of macrophages takes place after ligation with Fractalkine , but not CC-chemokines , US28-induced migration of vascular smooth muscle cells is mediated by binding to CC-chemokines , but not Fractalkine . Since HCMV-encoded US28 apparently plays multiple roles in promoting angiogenesis , we postulate the same is true for M33 , the MCMV homologue of US28 [76] . Ongoing studies are therefore oriented toward testing the hypothesis that MCMV-encoded M33 plays significant roles in the pathophysiology and increased severity of CNV during chronic MCMV infection . Additional direct and indirect mechanisms by which cytomegalovirus might contribute to angiogenesis and vascular disease are suggested by other studies . Examples include studies that have shown that HCMV infection induces a reduction of endothelial nitric oxide synthase activity commonly observed during cardiovascular disease [77]; RCMV induces the stimulation of a number of proteases including MMPs that are involved in degradation of the basement membrane required during the angiogenesis process [78]; HCMV induces an upregulation of a number of cellular chemokines including macrophage inflammatory protein 1 alpha ( MIP1-α ) , MIP1-β , RANTES , and IL-2 that play critical roles in angiogenesis and development of vascular disease [74] , [79]; and HCMV infection of coronary artery smooth muscle cells stimulates VEGF expression [80] . Since angiogenesis in health and disease is a process of great complexity that offers a number of mechanisms by which cytomegalovirus infection of multiple cell types might serve as a stimulatory cofactor in the development of more severe choroidal neovascularization , we elected to focus our study on a possible role for macrophages during chronic systemic MCMV infection . Macrophages can be either pro-angiogenic or anti-angiogenic depending on their polarization phenotype [25] that is regulated by the cytokine patterns encountered by macrophages within the resident tissue milieu [23] , [26] . Classically activated macrophages , or M1 macrophages , exhibit an anti-angiogenic phenotype and produce high amounts of IL-12 , IL-23 , IL-6 , and TNF-α , but low amounts of IL-10 [81] . In comparison , alternatively activated macrophages , or M2 macrophages , exhibit a pro-angiogenic phenotype and produce high amounts of IL-10 , but low amounts of pro-inflammatory cytokines such as IL-6 and TNF-α [81] . Moreover , M1 macrophages inhibit angiogenesis by inducing a cell-death program in endothelial cells , whereas M2 macrophages promote angiogenesis by stimulating production and release of pro-angiogenic factors such as VEGF that encourage endothelial tip cell formation [52] . In this regard , Fantin and coworkers [67] have recently made the extraordinary observation that M2 macrophages may also play a critical role during formation of new blood vessels by serving as bridge cells to properly position and fuse neotubules into one continuous blood vessel , possibly via activation of the DII4-a ligand and expression of Notch receptors [82] . Thus , cytomegalovirus infection of monocytes and macrophages may influence angiogenesis-related activities by several possible mechanisms . For example , HCMV infection of monocytes appears to influence the polarization phenotype of the activated macrophage by modulating in a selective manner many M1/M2-associated factors [52] , [83] , thereby inducing angiogenesis through stimulation of VEGF production and other angiogenic factors . Importantly , MCMV-infected IC-21 mouse macrophages exhibited a pro-angiogenic M2 phenotype in our studies . Alternatively , HCMV infection could conceivably have a detrimental on the normal angiogenic process by promoting inflammation . HCMV infection of endothelial cells may also enhance the stability of newly formed blood vessels through stimulation and release of several cytokines and growth factors including the Notch 2 receptor [83] . We therefore postulate that chronic MCMV infection results in more severe choroidal neovascularization in our study by driving monocytes toward a M2 macrophage phenotype that favors angiogenesis through stimulation and release of pro-angiogenic factors that includes VEGF . It has not escaped our attention , however , that chronic MCMV infection might also cause more severe choroidal neovascularization by direct or indirect mechanisms associated with endothelial cell infection , a focus of future studies . Splenic macrophages collected from chronically infected mice with the most severe choroidal neovascularization in our study showed significant increases in the amounts of transcripts to MMP-9 and COX-2 , two proteins known to be involved in angiogenesis [52] . The most dramatic increase in transcript level , however , was observed for that of VEGF , a critical pro-angiogenic factor . This observation was confirmed in a second independent animal study by us that demonstrated an even greater increase in VEGF transcript production in splenic macrophages collected from chronically infected animals . One interpretation of these reproducible findings is that when chronically infected monocytes are recruited to choroidal sites of laser-induced damage , their activation programs are initiated and oriented toward the pro-angiogenic M2 phenotype . Since they are also chronically infected with MCMV , this activation program stimulates virus replication , an event that leads to enhanced production and secretion of several pro-angiogenic factors including VEGF . Inoculation of cultures of human foreskin fibroblasts or cultures of coronary artery smooth muscle cells with HCMV has been shown to result in stimulation of functionally active VEGF production [78] . It is therefore not surprising in the present study that MCMV infection of cultures of IC-21 mouse macrophages significantly stimulated production of VEGF mRNA and VEGF protein . Additional observations made during immunostaining studies also demonstrated that VEGF is indeed produced in high amounts by MCMV-infected IC-21 mouse macrophages , especially those in the early stages of cytopathology during plaque formation . Of particular interest , however , was the additional observation that monolayer cells too early to be infected with MCMV ( given the low multiplicity of infection used ) were also VEGF-positive , an observation suggesting the attractive hypothesis that uninfected bystander macrophages might also be stimulated by adjacent MCMV-infected macrophages to produce enhanced amounts of VEGF during virus infection . Thus , MCMV infection of resident macrophages of tissues of the lung and spleen , and even bone marrow cells , could conceivably contribute to macrophage activation during chronic infection . MCMV-infected bone marrow cells , especially stromal cells , could favor a pro-angiogenic microenvironment that induces bystander activation during development within the marrow since stromal cells serve as a substrate upon which monocytes are induced to differentiate [84] . In addition , due to their high vascularity , both lung and spleen experience high monocyte traffic , and chronic MCMV infection of these tissues could induce bystander activation . We therefore postulate that chronic MCMV infection of monocytes and macrophages distant from the eye serves as an important mechanism for macrophage activation of the M2 phenotype that would contribute to the pro-angiogenic microenvironment of the choroidal tissues of the eye . Treatment with ganciclovir , a potent inhibitor of active cytomegalovirus replication and HCMV disease in the clinical setting [42] , [43] , has been shown to delay the time to development of allograft rejection in heart transplant recipients [85] , [86] , a finding that underscores the importance for active HCMV replication in acceleration of vascular disease . Additional studies using experimental rat transplant models have provided similar data showing that ganciclovir therapy also reduced or prevented RCMV-associated acceleration of tissue rejection when compared with RCMV-infected animals not treated with the antiviral [87] , [88] . Since we hypothesize that MCMV infection of macrophages plays a central role in amplifying the severity of experimental choroidal neovascularization in mice by stimulation of pro-angiogenic factors including VEGF , we used a similar antiviral approach to demonstrate that VEGF-specific transcript production by splenic macrophages collected from chronically infected mice was indeed ganciclovir-sensitive . This outcome strongly supports the need for active MCMV virus replication in stimulation of production of pro-angiogenic factors such as VEGF that is required for increased severity of choroidal neovascularization . These in vivo findings were duplicated and extended in culture using ganciclovir-treated , MCMV-infected monolayers of IC-21 mouse macrophages , and in a relatively dose-dependent manner . In summary , the findings reported herein using an experimental mouse model of CNV serve to clarify our previous seroepidemiologic clinical study in which a significant association was identified between high titers of anti-HCMV IgG and development of neovascular AMD [35] . The presence of high anti-HCMV titers may indicate a subset of patients who harbor a greater total body burden of chronic HCMV infection , or who have experienced a recent , significant reactivation event . In either case , we hypothesize that the blood load of circulating HCMV-infected monocytes would be exceptionally high in this subset of patients . Upon recruitment to sites of drusen formation in patients who manifest the dry form of AMD , HCMV-infected monocytes would mature into tissue-resident macrophages with active virus replication , become polarized toward the pro-angiogenic M2 phenotype , and become a major source for production of a number of pro-angiogenic factors including VEGF that would amplify choroidal neovascularization associated with the wet form of AMD . We therefore believe that HCMV infection should be considered as a heretofore unrecognized risk factor for development of neovascular AMD . If true , subsets of patients who harbor a low virus load of HCMV would be predicted to experience decreased onset and progression of choroidal neovascularization , an occurrence that would impact their clinical outcome in terms of time of onset of visual loss and degree of visual loss . It is therefore possible that antiviral treatment might be effective in suppressing choroidal neovascularization associated with wet AMD in a fashion similar to that for suppression of allograft rejection in heart transplant recipients . Future studies will be oriented toward this investigation .
All animal procedures were performed in strict accordance with the Association for Research in Vision and Ophthalmology Statement for the Use of Animals in Ophthalmic and Vision Research , and with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . Animal research protocols were approved by the Institutional Animal Care and Use Committees of the University of Miami Miller School of Medicine ( A3324-01 ) and Georgia State University ( A3914-01 ) . All laser treatments were performed under anesthesia ( intramuscular administration of ketamine hydrochloride , xylazine , and adepromazine ) , and all efforts were made to minimize suffering . Adult female C57BL/6 mice were purchased from the National Institute of Aging ( Bethesda , Maryland ) , and used throughout this investigation . Mice were allowed unrestricted access to food and water and maintained in alternating 12-hour light-dark cycles . Stocks of MCMV were prepared in mouse salivary glands as described previously [89] . Briefly , BALB/c mice ( Taconic Farms , Germantown , New York ) were infected intraperitoneally with 1×102 to 1×103 plaque-forming units ( PFU ) of the Smith strain of MCMV ( American Type Culture Collection , Manassas , VA ) contained within a 0 . 2-ml volume . Approximately 14 days later , the salivary glands were removed aseptically , homogenized ( 10% wt/vol ) in Dulbecco's modified Eagle's tissue culture medium containing 10% fetal bovine serum ( DMEM ) , clarified by centrifugation , and 0 . 25 ml aliquots of the supernatant stored in liquid N2 . Virus stocks were titered on monolayers of mouse embryo fibroblasts ( MEF ) grown in DMEM . Fresh aliquots of MCMV stock were thawed and used for single experiments . UV-inactivated virus was prepared by exposure of aliquots of MCMV stock to ultraviolet radiation for 30 min to inactivate virus infectivity as determined by no detectable plaque formation on MEF monolayers 7 days after undiluted inoculation . Plan 1: To evaluate CNV severity after acute or chronic infection with MCMV , four groups of mice ( n = 10 mice per group ) were injected intraperitoneally with 40 ul of a non-lethal dose of infectious MCMV ( 1 . 5×106 plaque-forming units ) or with an equivalent dose of UV-inactivated MCMV ( controls ) . At 6 days ( acute infection ) and at 6 weeks and 12 weeks ( chronic infection ) after inoculation , the eyes of age-matched mice were subjected to bilateral laser treatment to induce CNV as described below . Mice were matched in age ( 10 months ) at time of laser treatment . Four weeks later , the right eyes were collected for flat-mount analysis , and the left eyes were collected for histopathologic analysis . Plan 2: To evaluate macrophages for their patterns of production of various pro-angiogenic factors during CNV at time of acute versus chronic MCMV infection , the eyes of groups of mice ( n = 10 mice per group ) were subjected to bilateral laser treatment at 6 days or at 12 weeks after intraperitoneal injection with infectious MCMV . The control group for this study consisted of groups of mice injected intraperitoneally with UV-inactivated virus . Mice were matched in age ( 10 months ) at time of laser treatment for these animal groups . Four weeks after CNV induction , splenic macrophages were collected from all animals for quantitative RT-PCR assay analysis of several gene transcripts relevant to inflammation and/or neovascularization . Whole eyes , choroidal tissues , tissues from various organs ( salivary glands , lung , spleen ) , and bone-marrow cells ( CD34+ cells ) were also collected from mice of the same animal groups and analyzed by standard plaque assay for detection of infectious virus or analyzed by PCR assay for detection of MCMV-specific DNA sequences . Plan 3: To confirm mouse macrophages as a source for VEGF production following MCMV infection , monolayers of the IC-21 mouse macrophage cell line ( American Type Culture Collection , Manassas , VA , USA ) [41] were inoculated either with MCMV ( moi = 2 . 5 ) , UV-inactivated MCMV , maintenance medium only , or maintenance medium containing lipopolysaccharide ( LPS ) ( 100 ng/ml ) . All cells were harvested at 24 or 48 hrs postinfection and subjected to quantitative real time RT-PCR assay for quantification of VEGF mRNA and TNF-α mRNA , standard ELISA for quantification of VEGF protein production , or immunostaining for detection and pattern of VEGF production . Plan 4: To determine the effect of antiviral treatment on production of VEGF mRNA and TNF-α mRNA by splenic macrophages at time of chronic MCMV infection , groups of mice ( n = 10 mice per group ) were injected intraperitoneally with 40 ul of a non-lethal dose of infectious MCMV ( 1 . 5×106 plaque-forming units ) or maintenance medium ( mock infected ) . At 12 weeks after inoculation , MCMV-infected or mock-infected mice were treated intraperitoneally with ganciclovir for 7 days at a dose of 40 mg/kg/day , a dose that reflects the relative decreased sensitivity of MCMV to ganciclovir when compared with the sensitivity of HCMV to ganciclovir [43] . Untreated control MCMV-infected or mock-infected mice received daily intraperitoneal injections of phosphate-buffered saline for 7 days . Following the 7-day regimen of ganciclovir or phosphate-buffered saline treatment , splenic macrophages were collected from ganciclovir-treated and untreated chronically infected mice and compared by quantitiative real time RT-PCR assay for levels of VEGF mRNA production . To determine the effect of antiviral treatment on production of VEGF mRNA and TNF-α mRNA by mouse macrophages during acute MCMV infection , monolayers of IC-21 mouse macrophages were inoculated either with MCMV ( moi = 2 . 5 ) or mock-infected with maintenance medium . At 1-hr postinfection , MCMV-infected and mock-infected monolayers were treated either with 15 , 30 , or 60 uM of ganciclovir or treated with phosphate-buffered saline ( control ) . At 24 hr postinfection , all monolayers were harvested and subjected to quantitative RT-PCR assay for quantification and comparison of levels of VEGF mRNA production . At 6 days , 6 weeks , or 12 weeks after injection with infectious or UV-inactivated MCMV , diode red laser was used to create choroidal thermal burns bilaterally and induce experimental CNV as described previously [39] . Four weeks after laser application , mice were euthanized , and subjected to histopathologic analysis as well as flat-mount analysis of surface area , vascularity , and cell density of CNV . All images were digitally acquired ( Axiovision , Zeiss ) and recompiled ( Photoshop version 6 . 0; Adobe , San Jose , California ) . Surface area of CNV lesions was determined by using either fluorescein-isothiocyanate ( FITC ) -dextran ( Sigma , St . Louis , Missouri ) fluorescence or propidium iodide ( PI , Sigma ) fluorescence , and outlining the margins of the lesion with a computer analysis software ( Photoshop 6 . 0 ) . The area in pixels was normalized by dividing the average of the optic disc measured in 10 independent eyes . Five eyes were examined 4 weeks after laser treatment to determine the average spot size ( 0 . 48 disc areas ) . A CNV was determined to be present if the surface area of an individual lesion was greater than 0 . 50 disc areas . Four weeks after bilateral laser treatment of groups of mice infected systemically with MCMV for 6 days , 6 weeks , or 12 weeks , left eyes were carefully removed from all animals following euthanasia , fixed in 10% buffered formalin , paraffin embedded , sectioned with hematoxylin and eosin , and examined by light microscopy for detection and quantification of areas of CNV . Following removal of spleens under sterile conditions from euthanized mice , a Spectra/Mesh macroporus 210 µm filter ( Spectrum Laboratories , Inc . , Los Angeles , California ) was used to obtain splenic macrophages after maceration of individual spleens in a Hanks balance salt solution ( HBSS ) medium containing 1 M HEPES , 1 M NaAZ , and fetal bovine serum . ACK buffer was added to the spleen suspension to lyse red blood cells . The remaining cells were centrifuged and resuspended in HBSS medium containing rat anti-mouse F4/80 conjugated with PE ( Caltag , Burlingame , California ) . Splenic macrophages were then purified by magnetic column separation using MACS Anti-PE Microbeads ( Miltenyi , Auburn , California ) as specified by manufacturer's instructions . At the time of euthanasia and under sterile conditions , tibias and femurs were dissected and bone marrow was extracted by slowly flushing the dyaphyseal channel with HBSS medium using a 27-gauge needle . Bone marrow was homogenized , filtered , centrifuged , and resuspended in HBBS medium . Red blood cells were lysed with ACK buffer , and the remaining cells were incubated with rat anti-mouse CD34 conjugated with PE ( BD Biosciences , Pharmingen , San Diego , California ) . CD34+ vascular precursor cells were then purified by magnetic column separation using MACS Anti-PE Microbeads ( Miltenyi ) as specified by manufacturer's instructions . Whole eyes collected from mice at 30 days after laser-induced CNV were frozen individually at −80°C . At time of quantitative plaque assay , eyes were thawed , homogenized individually in 1 . 0 ml of cold Dulbecco's modified Eagle's medium ( DMEM ) containing 10% fetal bovine serum , and clarified by centrifugation . Ten-fold dilutions of the resulting supernatants were titered in duplicate onto monolayers of MEF contained within 6-well plates , allowed to absorb for 1 hour at 37°C , overlaid with methylcellulose containing DMEM , and incubated for 5 or 6 days at 37°C in a humidified CO2 atmosphere . Monolayers were screened daily for 7 days using an inverted light microscope for detection of plaques of MCMV-induced cytopathology . DNA was extracted from whole eyes , salivary glands , lungs , bone marrow , spleens , and isolated macrophages collected from euthanized mice using the QIAamp Tissue kit ( QIAGEN GmbH , Valencia , California ) according to manufacturer's instructions and subjected to PCR assay to detect MCMV-specific DNA using primers for immediate early 1 ( IE1 ) and glycoprotein H ( gH ) genes . The primers used were kindly provided by Dr . Daniel D . Sedmak , Ohio State University College of Medicine , Columbus , Ohio . The primer pair for MCMV IE1 gene was 5′-TAGCCAATG ATATCTTCGAGCG-3′ and 3′-ATCTGGTGCTCCTCAGATCAGCTAA-5′ , and the primer pair for MCMV gH gene was 5′-TTCAGTTCAACTCGAA-3′ and 3′-GGGAAGAAGTACTCGACCGG-5′ . PCR amplification of β-actin was performed as an internal control . Actin primers consisted of 5′-ATTGTGATGGACTCCGGTGA-3′ and 3′-AGCTCATAGCTCTTCTCCAG-5′ . DNA extracted from tissue homogenates was eluted in 100 µl of distilled water , and stored at −20 C until analysis . DNA was amplified in a total volume of 25 µl with 200 nM of each primer and 1 . 0 U of Taq DNA polymerase ( Gibco BRL ) added in 2 . 5 µl of a PCR buffer ( 50 mM KCL , 20 mM Tris-HCl [pH 8 . 4] , and 1 . 5 mM MgCl2 ) . PCR assays were performed on a Perkin Elmer 9600 thermocycler ( PE Applied Biosystems ) . PCR assay conditions consisted of an initial denaturation step of 4 min at 94 C , followed by 35 cycles , with 1 cycle consisting of 30 sec at 94 C , 30 sec at 53 C , and 30 sec 72 C . Amplification products were separated by electrophoresis through 1% agarose gels , and stained with ethidium bromide for visualization . Total RNA was extracted from whole bone marrow cells ( CD34+ cells ) , splenic macrophages , or MCMV-infected IC-21 mouse macrophage monolayers using Tri-Reagent and prepared for quantitative RT-PCR reactions as described previously [90] . Real time RT-PCR assay was used to quantify several cellular transcripts of interest that included mouse tumor necrosis factor-alpha ( TNF-α ) , matrix metalloproteinase-9 ( MMP-9 ) , vascular endothelial growth factor ( VEGF ) , VEGF receptor 1 ( VEGFR1 ) , VEGF receptor 2 ( VEGFR2 ) , platelet-derived growth factor-beta ( PDGF-β ) , cyclooxygenase ( COX-2 ) , and inducible nitric oxide synthase ( iNOS ) . Real time RT-PCR assays were performed for TNF-α , VEGFR1 , VEGFR2 , and COX-2 mouse transcripts using commercially available kits ( Perkin Elmer Applied Biosciences ) . The primer pair for real time RT-PCR assay of mouse PDGF-β mRNA was 5′-AAGCACACGCATGACAAG-3′ and 3′-GGGGCAATACAGCAAATAC-5′; for VEGF mRNA was 5′- CGAAACCATGAACTTTCTGC-3′ and 3′-CCTCAGTGGGCACACACTCC-5′; for MMP-9 mRNA was 5′-CAGGATAAACTGTATGGCTTCTGC-3′ and 3′- GCCGAGTTGCCCCCA-5′; and for iNOS mRNA was 5′-TGACGCCAAACATGACTTCAG-3′ and 3′-GCCATCGGGCATCTGGTA . Transcripts of these molecules were normalized to 18S ribosomal RNA transcripts via standard curves generated using serially diluted samples of mRNA ( 0 . 001–100 ng ) . Real time RT-PCR assays were performed in duplicate with quantitative values determined for each molecule as the ratio of the mean values for a specific mRNA versus 18S mRNA . Median values for each molecule were calculated and normalized to samples obtained from sham-inoculated control animals ( 100% ) . MCMV-infected and mock-infected monolayers of IC-21 mouse macrophages grown on 6-well chamber slides were harvested at 24 and 48 hr postinfection , fixed in cold ethanol , dried , and reacted with 5% normal goat serum containing 0 . 2% Triton X-100 . Following three washings in phosphate-buffered saline , slides were incubated for 1 hr with either rabbit anti-mouse VEGF IgG ( 1∶100 dilution ) ( Santa Cruz Biotechnology , Santa Cruz , CA ) or normal rabbit IgG ( 1∶100 dilution ( Santa Cruz Biotechnology , Santa Cruz , CA ) , washed three times with phosphate-buffered saline , and reacted with biotinylated anti-rabbit IgG secondary antibody using the Rabbit ABC Staining system ( Santa Cruz Biotechnology , Santa Cruz , CA ) . Chamber slides were mounted on standard microscope slides and cell nuclei were counterstained using Vectashield Mounting Medium containing DAPI ( Vector Laboratories , Burlingame , CA ) . All slides were examined and photographed using a Nikon Eclipse 50i microscope equipped with an X-Cite Series 120 Epi-fl illuminator . Morphometric data for individual lesions in each eye were averaged to provide one value per eye . Mean and standard deviation values for each group was calculated and p values were determined using student t-test and one-way analysis of variance+Dunnett's multiple comparison post-hoc test ( GraphPad Prism 4 . 0 , San Diego , CA ) . Values of p≤0 . 05 were considered statistically significant for all forms of statistical analysis used . | Neovascular age-related macular degeneration ( AMD ) is the leading cause of vision loss in the elderly . Onset of AMD is due to local production of vascular endothelial growth factor ( VEGF ) that promotes formation of new blood vessels in the retina , thereby leading to retinal tissue destruction and blindness . Since a clinical study by us showed that AMD patients have high amounts of antibodies to human cytomegalovirus ( HCMV ) , we postulated that infection with HCMV might be a risk factor for AMD . To investigate this possibility , mice were infected with murine cytomegalovirus ( MCMV ) , and at various times after infection , subjected to laser treatment of the eye to induce choroidal neovascularization , an experimental model of AMD . Most severe CNV developed in mice with chronic MCMV infection , a time when MCMV gene sequences could not be detected within eye tissues . However , splenic macrophages collected from mice with chronic MCMV infection produced high levels of gene transcripts to several pro-angiogenic factors including VEGF . MCMV infection of mouse macrophages in culture also produced high amounts of VEGF . Stimulation of VEGF production in vivo and in vitro was sensitive to antiviral treatment . Chronic HCMV infection may therefore promote AMD by stimulation of VEGF production by activated macrophages . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"medicine",
"biology"
] | 2012 | Macrophage Activation Associated with Chronic Murine Cytomegalovirus Infection Results in More Severe Experimental Choroidal Neovascularization |
Tsetse flies are the notorious transmitters of African trypanosomiasis , a disease caused by the Trypanosoma parasite that affects humans and livestock on the African continent . Metacyclic infection rates in natural tsetse populations with Trypanosoma brucei , including the two human-pathogenic subspecies , are very low , even in epidemic situations . Therefore , the infected fly/host contact frequency is a key determinant of the transmission dynamics . As an obligate blood feeder , tsetse flies rely on their complex salivary potion to inhibit host haemostatic reactions ensuring an efficient feeding . The results of this experimental study suggest that the parasite might promote its transmission through manipulation of the tsetse feeding behavior by modifying the saliva composition . Indeed , salivary gland Trypanosoma brucei-infected flies display a significantly prolonged feeding time , thereby enhancing the likelihood of infecting multiple hosts during the process of a single blood meal cycle . Comparison of the two major anti-haemostatic activities i . e . anti-platelet aggregation and anti-coagulation activity in these flies versus non-infected tsetse flies demonstrates a significant suppression of these activities as a result of the trypanosome-infection status . This effect was mainly related to the parasite-induced reduction in salivary gland gene transcription , resulting in a strong decrease in protein content and related biological activities . Additionally , the anti-thrombin activity and inhibition of thrombin-induced coagulation was even more severely hampered as a result of the trypanosome infection . Indeed , while naive tsetse saliva strongly inhibited human thrombin activity and thrombin-induced blood coagulation , saliva from T . brucei-infected flies showed a significantly enhanced thrombinase activity resulting in a far less potent anti-coagulation activity . These data clearly provide evidence for a trypanosome-mediated modification of the tsetse salivary composition that results in a drastically reduced anti-haemostatic potential and a hampered feeding performance which could lead to an increase of the vector/host contact and parasite transmission in field conditions .
Tsetse flies ( Diptera: Glossinidae ) are obligate blood feeding insects that are important disease vectors given their involvement in the transmission of different pathogenic trypanosome species that cause human sleeping sickness and livestock trypanosomiasis in Africa . Trypanosomes of the Trypanosoma brucei group – including the two human-pathogenic subspecies T . b . gambiense and T . b . rhodesiense – have to go through a complex developmental cycle in the alimentary tract and salivary glands of the tsetse fly [1] . The salivary gland is the tissue in which T . brucei parasites undergo the final developmental phase , i . e . a continuous cycle of multiplication and cellular differentiation into the metacyclic form that is infective for the mammalian host [2] . Once this trypanosome population has been established in the salivary glands , it is continuously maintained at high density throughout the remaining life span of the tsetse fly . In the naive salivary gland micro-environment , saliva components are present that enhance the infection onset upon trypanosome inoculation in the host skin [3] . Other constituents are essential for the hematophagous behavior of the tsetse fly by counteracting host responses such as vasoconstriction , platelet aggregation and coagulation reactions involving serine proteases such as thrombin [4] . Several compounds have been implicated in facilitating blood feeding: a thrombin inhibitor [tsetse thrombin inhibitor ( TTI ) ] [5] , [6] and salivary apyrases [5′nucleotidase related protein , Glossina morsitans morsitans salivary gland protein 3 ( Sgp3 ) ] including at least one with fibrinogen receptor ( GPIIb/IIIa ) antagonistic properties ( 5′Nuc ) [7] . Other abundant salivary components include putative endonucleases [tsetse salivary gland proteins 1 and 2 ( Tsal1 and Tsal2 ) ] [8] , putative adenosine deaminases [tsetse salivary gland growth factors 1 and 2 ( TSGF-1 and TSGF-2 ) ] [9] and an antigen5-related allergen [tsetse Antigen5 ( TAg5 ) ] [10] . However , there is no information on the importance of these major tsetse saliva proteins in their interplay with the trypanosome life cycle . To date , a growing number of studies demonstrate the ability of vector-borne parasites to alter phenotypic traits of their insect vectors in a way that increases vector-host contact frequency and hence increases the probability of parasite transmission [11] , [12] . This type of parasite-induced modulation of the vector physiology and feeding behavior has already been documented for the Leishmania-sandfly model [13] , the Plasmodium-mosquito model [14]–[16] and other pathogen-vector models [reviewed in [17] , [18]] . A recurrent feature of infected vectors is a modified feeding behavior that results from the physical obstruction of the alimentary tract and interference with mechanoreceptors that are required to regulate the blood meal uptake . Indeed , Leishmania promastigotes produce a secretory gel , mainly composed of a filamentous proteophosphoglycan that blocks the foregut and impairs the phagoreceptors , thereby reducing the arthropod feeding efficiency [19] . Similarly , a proportion of plague-transmitting fleas display obstructed proventiculi as a result of Yersinia biofilm surrounded by an extracellular matrix [20] . In the tsetse fly-trypanosome interaction , mouthpart obstruction and interference with labral mechanoreceptors has been documented upon infection with Dutonella and Nannomonas subgenera of Trypanosoma ( T . congolense and T . vivax ) that form rosettes and colonize the tsetse fly labrum [21]–[26] . However , limited and contradictory data have been reported on the feeding behavior of tsetse flies infected with T . brucei parasites ( including the human pathogens ) which belong to the Trypanozoon subgenus and display a different developmental cycle in the vector than T . congolense and T . vivax [1] , [2] . Jenni et al . observed a more frequent probing behavior of T . brucei infected tsetse flies and hypothesized that this resulted from physical interference of trypanosomes with the function of the labral mechanoreceptors [27] . However , other experimental results suggested that T . brucei parasites in the salivary glands did not significantly alter the tsetse feeding [22] , [28] . In this study , we investigated whether T . brucei parasites alter the tsetse fly blood feeding behavior in a way that would favor parasite transmission within the mammalian host population . Next , we determined the impact of a T . brucei salivary gland infection on the saliva composition and the biological activities related to anti-haemostasis . The obtained data provide evidence that the trypanosome parasites drastically modulate the tsetse salivary composition and anti-haemostatic activity resulting in an alteration of the feeding behavior that favors parasite transmission .
The feeding efficiency of salivary gland infected ( SG+ ) tsetse flies ( n = 50 ) was compared to that of controls that did not develop a salivary gland infection ( SG- , n = 48 ) upon feeding on a Trypanosoma brucei brucei AnTAR1 parasitemic mouse . As a read-out , two variables were measured: ( i ) the time necessary to obtain a full blood meal including the probing behavior that precedes the actual blood ingestion and ( ii ) the size ( mass ) of the blood meal . Despite a considerable variability in both experimental groups , the blood meal acquisition was significantly slower ( p<0 . 05 , Table 1 ) for SG+ flies ( 267±23 s . ) than for SG- flies ( 210±16 s . ) , especially resulting from a prolonged probing behavior ( visual observation ) . No differences in ingested blood masses were observed ( p = 0 . 83 ) . The presence of a T . b . brucei infection in the salivary glands significantly compromised ( p<0 . 05 ) the expression of genes that encode the major G . m . morsitans saliva proteins ( Figure 1A ) . Expression levels were decreased by 63% ( tsal 2 ) up to 95% for the 5′nuc apyrase gene ( Figure 1 ) . In two independent experiments , threshold cycle values for actin and tubulin housekeeping genes did not significantly change as a result of the SG+ infection status . Concomitant to the reduced transcription of the major saliva genes , the saliva of SG+ flies contained 70% less protein ( p<0 . 01 ) as compared to the SG- flies ( 0 . 9±0 . 2 versus 3 . 0±0 . 5 µg per salivary gland , Figure 1B ) . A more detailed analysis of the SG- and SG+ saliva composition was performed using Tricine-SDS-PAGE ( Fig . 2A ) combined with either Coomassie ( Fig . 2A , lanes section 1 ) or Silver based staining methods ( Fig . 2A , lanes section 2 ) . Densitometry analysis of the Coomassie stained protein profiles revealed a generalized reduction of 70–97% in protein band intensities for SG+ saliva samples ( Fig . 2B ) . In addition , several protein and peptide bands that are visible in the SG- saliva profiles upon silver staining , are no longer detectable in SG+ saliva . Western blot analysis using anti-T . b . brucei infectome immune serum could not detect the appearance of trypanosome-derived components in SG+ saliva . Different biological activities ( apyrase , adenosine deaminase and anti-thrombinase ) that were previously described or suggested to be present in tsetse saliva , were quantified in SG- and SG+ samples . Based on the quantification of Pi-release from the individual substrates ATP and ADP as read-out for apyrase ( ATP diphosphohydrolase ) activity , an approximate 5-fold reduction ( p<0 . 01 ) in salivary apyrase was observed in trypanosome infected salivary glands ( Figure 3A ) . For the adenosine deaminase activity that was present in the SG- saliva at 6 . 0±1 . 0 mU/salivary gland , a similar reduction ( 82% , p<0 . 01 ) was observed in the SG+ flies exhibiting an activity of only 1 . 1±0 . 5 mU/salivary gland ( Figure 3B ) . The thrombinase-inhibitory properties of tsetse fly saliva were assayed with respectively 1/80 and 1/400 dilutions . The 1/80 SG- saliva dilutions almost completely inhibited the human thrombinase activity ( assayed by the release of pNA from thrombin-specific substrate ) at the concentration of 500 mU/ml ( Figure 3C ) . In contrast , a significant increase ( 83% , p<0 . 01 ) in thrombinase activity was observed for the same SG+ saliva dilution , suggesting a potentiation of the thrombin enzymatic activity in the used assay conditions . For the 1/400 SG+ dilution , an increase could still be detected although less pronounced ( 27% , p<0 . 05 ) . The enhancement of thrombinase activity by SG+ saliva did not depend on a trypanosome-derived enzyme with the same substrate-specificity , as saliva from SG+ flies by itself did not convert the chromogenic substrate ( data not shown ) . The salivary anti-thrombotic and anti-coagulant activities were monitored in human plasma using respective in vitro read-out assays . The aggregation of platelets in human platelet rich plasma ( PRP ) supplemented with ½ serial SG- and SG+ saliva dilutions ( 1/100–1/400 ) was analyzed in response to 10 µM ADP , revealing an approximate 3-fold reduction in anti-platelet aggregating capacity of SG+ saliva ( Figure 4 ) . Coagulation in human platelet poor plasma ( PPP ) , induced by 25 mU/ml thrombin in the presence or absence of ½ serial SG- ( 1/400–1/6400 ) and SG+ saliva dilutions ( 1/50–1/6400 ) , revealed a striking decrease of anti-coagulant activity in tsetse fly saliva upon trypanosome infection . Indeed , while all tested SG- saliva dilutions ( 1/400–1/6400 ) markedly increased the coagulation lag times ( Figure 5A ) , all SG+ saliva dilutions from 1/400 downwards ( 1/800–1/6400 ) exerted negligible anti-coagulant activity ( Figure 5B ) . Comparison of the coagulation lag times revealed a 16- to 32-fold reduction of anti-coagulant activity in SG+ as compared to SG- saliva ( Figure 5C ) . Moreover , thrombin did not induce maximal coagulation responses in PPP in the presence of the 1/400–1/6400 SG- saliva samples within a 3 hour reaction time ( Figure 5D ) , while endpoint O . D . values at 405 nm were even slightly higher when thrombin was incubated with 1/800–1/6400 SG+ saliva samples ( Figure 5D ) . These slightly increased endpoint O . D . values did not result from clotting of salivary components , as no thrombin-induced coagulation was observed in SG- and SG+ saliva ( data not shown ) .
African trypanosomes including the human-infectious Trypanosoma brucei subspecies , exploit the obligate blood feeding behavior of tsetse flies ( Glossina sp . ) for their transmission . These tsetse fly vectors rely on a pool feeding strategy which involves the laceration of the skin with their proboscis and blood ingestion from a superficial lesion . Once the skin is pierced , the proboscis is often partially withdrawn before being thrust again at a slightly different angle to probe for suitable blood vessels and to enhance the blood pool formation [29] . During these events , about 4 µg of salivary proteins are inoculated at the bite site in order to neutralize the complex anti-haemostatic host reactions that would lead to blood clotting and vasoconstriction [30] . In the case of feeding on a parasitemic host , tsetse flies can acquire a trypanosome infection which depends on a complex sequence of differentiation and migration that ends in the insect salivary glands [1] , [2] . Once the salivary glands are colonized by metacyclic T . brucei parasites ( SG+ ) , the tsetse fly can transmit parasites throughout its entire lifespan at each vector/host contact . Despite the epidemiological importance , information on the impact of the salivary gland infection on the tsetse feeding behavior and trypanosome transmission is scanty and contradictory . While Moloo et al . did not observe significant feeding behavioral differences as a result of the SG+ status [22] , Jenni et al . [27] reported that T . brucei-infected flies probed more frequently ( 2 to 3 fold increase ) before feeding and subsequently fed more voraciously as compared to uninfected ( SG- ) flies . The authors suggested that these effects resulted from the association of some trypanosomes with labral mechanoreceptors that play a role in the feeding and gorging response , analogous to what was reported for T . congolense infected flies . Indeed , the increased probing activity of T . congolense infected G . morsitans flies [31] may be caused by physical interference of the parasite with phagoreceptors in combination with a reduced diameter of the tsetse labrum due to the presence of parasite rosettes [22]–[23] , [32] . However , in contrast to T . congolense , T . brucei parasites never permanently colonize the tsetse fly mouthparts where the mechanoreceptors are localized [1] , [2] , which is not supportive for Jenni's hypothesis . In our study , we could confirm Jenni's observation that a T . brucei infection in tsetse fly salivary glands does significantly disturb the fly feeding behavior . Indeed , SG+ tsetse flies needed significantly longer times ( >25% longer ) to complete blood feeding due to a prolonged pre-feeding probing phase . Our experimental data clearly suggest that this altered feeding phenotype is the consequence of a changed protein content of the tsetse saliva due to the presence of a trypanosome infection , resulting in a much less potent anti-haemostatic activity . This reduced saliva production was confirmed using Tricine-SDS-PAGE , revealing a generalized suppression ( 70–97% ) of all protein bands in tsetse saliva which was found to be associated with a severely reduced ( 63%–95% ) transcription of the major tsetse fly salivary genes . Especially the 5′nuc gene that encodes an important tsetse fly salivary apyrase with GPIIb/IIIa ( fibrinogen receptor ) antagonistic properties [7] and another putative apyrase gene ( sgp3 ) were strongly suppressed ( >90% ) resulting in an overall 80% down regulation in the saliva apyrase [AT ( D ) Pase] activity . A similar phenomenon has been described for Plasmodium infected mosquitoes , where the salivary apyrase activity was reduced by three fold and which was also associated with prolonged probing times [14] , [33] . Salivary apyrase activity underlies one of the major anti-haemostatic strategies in a blood feeding insect [34] given that these enzymes inhibit purinergic thrombocyte triggering by hydrolyzing ATP and ADP , haemostatic triggers that are released from injured cells and activated platelets [35] . As such , the reduced apyrase activity in the SG+ tsetse saliva seriously affected the normally powerful capacity to inhibit the blood platelet aggregation demonstrated in an in vitro aggregation studies using human platelets . The significant suppressed adenosine deaminase activity in the trypanosome-infected saliva could also be a contributing factor in the decreased platelet aggregation inhibition . Indeed , adenosine deaminases convert adenosine into inosine , a nucleoside that was recently suggested to modulate platelet responses against various agonists including ADP and collagen [36] . The inhibition of the thrombin activity is another key anti-haemostatic activity of normal tsetse saliva . Indeed , a femtomolar affinity thrombin inhibitor ( TTI ) has been previously characterized in tsetse fly salivary gland extracts and shown to potently inhibit thrombinase activity and thrombin-induced haemostatic reactions [6] . In our study , we demonstrate that the presence of trypanosomes in the salivary glands severely impairs this ability of saliva to inhibit human thrombin and even modifies saliva to enhance the activity of this thrombinase in an in vitro pNA-release assay . This observed increase in thrombin activity was not related to the presence of a trypanosome-derived enzyme since SG+ saliva by itself did not hydrolyze the chromogenic thrombin substrate ( data not shown ) . Corroborating the observed effects of salivary gland infection on the measured enzymatic activities in the biochemical assays , the anti-coagulant potency of SG+ saliva was severely compromised in human plasma coagulation assays using human thrombin as a trigger . Indeed , while all tested SG- saliva dilutions significantly inhibited thrombin-induced coagulation , several SG+ saliva dilutions ( 1/800–1/6400 ) failed to inhibit this haemostatic reaction and even slightly increased the maximal coagulation response induced by thrombin . As such , both the biochemical and plasma coagulation assays suggested the presence of a parasite-derived or infection-induced procoagulant factor in the saliva of SG+ flies . Known thrombin activity enhancing cofactors include glycoprotein Ibα , fibrin and Na+ [37] . Given that experiments were performed under physiological salt conditions ( 150 mM ) with very low saliva concentrations , the influence of Na+ ions can be ruled out . Strikingly , tsetse fly transcriptome analyses revealed an abundant representation of a fibrinogen-domain-containing protein family that is enriched in the salivary gland tissue ( 197 ESTs ) as compared to other organs ( 16 ESTs in midgut , none in the fat body ) [38] . Possibly , these or other ( tissue or parasite-derived ) proteins might modulate thrombin activity through exosite binding and allosteric activation or even contribute as substrate in the coagulation reaction . The possibility that SG+ and SG- saliva by itself undergoes coagulation in response to thrombin was excluded experimentally . An experimental approach based on SG-/SG+ differential salivary proteome analyses and/or affinity purification using thrombin as bait could possibly unveil the identity of this thrombin enhancing factor . Collectively , we have demonstrated that upon colonization of the tsetse salivary glands with Trypanosoma brucei , the protein content and anti-haemostatic activity of the saliva change resulting in an altered insect vector feeding behavior . We assume that the reduced anti-haemostatic activity precludes the SG+ tsetse fly from efficiently generating and maintaining a primary blood pool as prerequisite in the feeding process . The observed prolonged probing/feeding time might result in an increased host contact as a result of interrupted feeding and partial blood acquisition and contribute to a higher probability of parasite transmission . To experimentally demonstrate the latter in a natural setting , i . e . to evidence the link between the behavioral modifications of tsetse flies and a more successful parasite transmission , is not obvious . However , field studies have indicated that tsetse flies are highly responsive to host defensive behavior and are prone to interrupted feeding [39] . Given that T . brucei salivary gland infected tsetse flies need longer times to feed successfully compared to non-infected ones , this high sensitivity to the host defensive behavior might result in a higher probability of interrupted blood feeding and of alternative host seeking . In other words , it might result in an increased biting rate of the infected tsetse within the available host population . As such , an infected tsetse fly is more likely to probe on multiple hosts during a single feeding cycle . Given that probing alone was proven to be sufficient to infect a mammalian host and that successive probing of the same fly on different hosts results in multiple infections [27] , the parasite-induced change in tsetse biting behavior might result in an enhanced trypanosome transmission . Here , it is clear that multiple transmission of the parasite in a single tsetse feeding cycle increases its survival and circulation within the natural mammalian host population . In the case of the human pathogenic T . brucei sp . , where the numbers of salivary gland infected tsetse flies in the natural population are extremely low [<0 . 1% , [40]–[42]] , the increased biting rate of the infected tsetse could be a major epidemiological factor . Currently , we do not know the molecular mechanism that underlies the trypanosome-induced modification of saliva composition and biological activities . Possibly , the high density of actively metabolizing parasites causes physiological stress to the salivary gland cells resulting in a suppression of salivary gene transcription and translation . In addition , the significant enhancement of the thrombin activity in the chromogenic thrombinase assay suggests that an activating factor is directly released or induced by the parasites in the saliva .
Animal ethics approval for the tsetse fly feeding on live animals and infection with T . brucei parasites was obtained from the Animal Ethical Committee of the Institute of Tropical Medicine , Antwerp ( Belgium ) ( Ethical clearance nrs . PAR013-MC-M-Tryp and PAR014-MC-K-Tryp ) . All tsetse fly infection studies were performed in compliance with the regulations for biosafety and under approval from the Environmental administration of the Flemish government ( licencenr . SBB 219 . 2007/1410 ) . Male Glossina morsitans morsitans ( Westwood ) from the colony at the Institute of Tropical Medicine ( Antwerp , Belgium ) were used in all experiments . This colony originated from pupae collected in Kariba ( Zimbabwe ) and Handeni ( Tanzania ) [43] . Flies were fed 4 days per week on rabbits and are maintained at 26°C and 65% relative humidity . Animal ethics approval for the tsetse fly feeding on live animals was obtained from the Animal Ethical Committee of the Institute of Tropical Medicine , Antwerp ( Belgium ) . The pleiomorphic Trypanosoma brucei brucei AnTAR1 strain , derived from the EATRO 1125 stock [44] , was used for the infection experiments . This strain was previously demonstrated to develop efficiently in the tsetse fly , resulting in >20% salivary gland infections [45] . Freshly emerged flies were offered their first blood meal on an anaesthetized mouse showing a pleiomorphic T . b . brucei parasitaemia of approximately 108 trypanosomes/ml blood with >80% intermediate/stumpy forms . Only fully engorged flies were further maintained at 26°C and 65% relative humidity and were fed 3 days per week on a naive rabbit . Thirty days after the infective blood meal , individual flies were evaluated for the presence of metacyclic trypanosomes in their salivary glands by salivation on pre-warmed ( 37°C ) glass slides [modification of the method of Burtt et al . [46]] . This allowed us to obtain two experimental fly groups of equal age and feeding history but with a different trypanosome infection status in the salivary glands ( SG+ and SG- ) . These flies were subsequently used for feeding efficiency analysis and for the dissection of salivary glands to assess salivary protein expression and associated biological activities . All tsetse fly infection studies were performed in compliance with the regulations for biosafety and under approval from the Environmental administration of the Flemish government . The feeding efficiencies of individualized SG+ and SG- flies of the same age and exactly the same feeding history were compared three days after the last blood meal on anaesthetized mice . Feeding efficiencies were monitored by direct observation by two observers ( JVDA and GC ) . Each observer contributed half of the observations for each experimental group , thereby excluding inter-group differences as a result of the different observers . For each fly , the total probing and feeding time was measured with a chronometer ( accuracy of 1 sec ) by direct observation . In order to determine the blood meal size , individual fly masses were measured to an accuracy of 0 . 1 mg before and immediately after blood feeding using an analytical balance ( Sartorius ) as described previously [3] . Three days after the last blood meal and following a 10 minute cold shock at 4°C , salivary glands of SG+ and SG- flies were dissected , pooled by 3 pairs in 30 µl sterile physiological H2O and incubated on ice for two hours before centrifugation ( 500 ×g , 2 min at 4°C ) . The supernatants were centrifuged an additional time to obtain saliva devoid of trypanosomes ( 2500 ×g , 2 min at 4°C ) . Saliva samples were stored at −80°C and only thawed once for analysing protein content and enzymatic activities . SG- and SG+ samples were always handled and tested in parallel in all subsequent analyses . Pellets ( salivary gland tissue ) were further processed to extract RNA for RT-qPCR purposes . The harvested salivary gland tissue was homogenised with a Teflon pestle in 1 ml Tripure reagent ( Roche ) followed by total RNA extraction according to the manufacturer's protocol . Six-hundred nanogram of each RNA sample was used for primary cDNA synthesis using 100 pmol oligo ( dT ) 15 primer ( Promega ) and 10 units Transcriptor Reverse Transcriptase ( Roche ) . For transcript-analysis , we made a selection of genes based on ( i ) the available literature data on identified genes that encode soluble saliva proteins , ( ii ) their relative contribution to tsetse fly proteome in terms of abundance and ( iii ) their predicted involvement in the blood feeding physiology . According these criteria we selected the identified thrombin inhibitor ( TTI ) , a highly abundant allergen ( TAg5 ) , two putative adenosine deaminases ( TSGF1&2 ) that might modulate adenosine-mediated platelet responses , two highly abundant putative endonucleases ( Tsal1&2 ) that might contribute to the blood feeding process by producing a defibrotide-like mixture of DNA haptamers and one predicted and one confirmed apyrase ( Sgp3 and 5′Nuc related protein ) . Relative transcript quantification was performed on an iCycler iQ detection system ( Bio-Rad ) and using the Bio-Rad software version 3 . 1 . RT-qPCR was performed on triplicate samples in a 25 µl reaction volume , containing 1 . 5 to 15 ng primary cDNA ( depending on the gene ) , 12 . 5 µl of iQ SYBR Green Supermix ( Bio-Rad ) and an optimized primer pair concentration for one of the respective saliva genes: tti [500 nM TTI_FW ( 5′- TTTATCTGATAGTTGCCGCAC -3′ ) and TTI_REV ( 5′- AAAGCCTTATGCCAGGAATC -3′ ) ] , tag5 [300 nM TAg5FW ( 5′-GTGGGTTGTGCCGCTTCTG-3′ ) and TAg5REV ( 5′-TTGACCTCGTATTTCTCGTTGG-3′ ) ] , tsal1 [700 nM Tsal1FW ( 5′-CTGATACCTCGATGATCACTC-3′ ) and Tsal1REV ( 5′-AGGCTCTTACATAATCCTTAAC-3′ ) ] , tsal2 [500 nM Tsal2FW ( 5′-CCAAGAACTGGCTGACCAA-3′ ) and Tsal2REV ( 5′-CTGCCAGCAGATTGTGTAAC-3′ ) ] , tsgf1 [300 nM TSGF1_FW ( 5′-CGGTTGTAAATCCGAATCTGT-3′ and TSGF1_REV ( 5′-GCGGCTGGCAAATAATGTAGA-3′ ) ] , tsgf2 [500 nM TSGF2_FW ( 5′-CAAACGCTCCGGTGTTGACGT-3′ ) and TSGF2_REV ( 5′-GCGGCTGGCAAATAATGTAGA-3′ ) ] , 5′nuc [300 nM 5NucFW ( 5′-CGGGTAATAAAGTTCTGGTCGTA-3′ ) and 5NucREV ( 5′-TTGGCAAGTCCACATTTGTTCTC-3′ ) ] and sgp3 [500 nM Sgp3_FW ( 5′- GCTATGGAACCATGGAAGGA -3′ ) and Sgp3_REV ( 5′- TTCTGATTCGCCTTCGTCTT -3′ ) ] . For normalization , G . m . morsitans actin and tubulin genes were amplified using respectively 700 nM and 300 nM of the following primer pairs: actinFW ( 5′-CGCTTCTGGTCGTACTACT-3′ ) and actinREV ( 5′-CCGGACATCACAATGTTGG-3′ ) , tubulinFW ( 5′-GATGGTCAAGTGCGATCCT-3′ ) and tubulinREV ( 5′-TGAGAACTCGCCTTCTTCC-3′ ) . The PCR conditions comprised an initial 10 min polymerase activation at 95°C followed by 35 cycles , each consisting of a denaturation step at 95°C for 15 s , 60 s annealing at 60°C and 60 s elongation at 72°C . In the data analysis , both actin and tubulin housekeeping genes were included to calculate an integrated normalization factor using the geNorm software v . 3 . 5 . Protein concentrations in the saliva extracts were determined using the BCA protein assay reagent kit ( Pierce Biotechnology ) . Saliva samples of SG- and SG+ flies were analyzed by Tricine-SDS-PAGE , using Novex tricine gels 10–20% ( 1 mm/10 well , Invitrogen ) and 100 mM Tris pH 8 . 3 100 mM Tricine 0 . 1% SDS as running buffer . Gels were run at 125 V in an XCell Surelock Mini-Cell ( Invitrogen ) . In parallel , the prestained PageRuler protein ladder and Spectra Multicolor Low Range Protein Ladder ( Fermentas ) were applied to the gels . Gels were either stained with 0 . 025% Coomassie dye R-250 in 10% acetic acid according to an established protocol [47] or Silverstained using the PageSilver kit ( Fermentas ) after a 30 minute fixation in 5% glutaraldehyde . The different Coomassie-stained protein profiles were digitalised as 300 dpi greyscale TIFF-files and analysed with the ImageMaster 1D Elite 3 . 01 programme ( Amersham Pharmacia Biotech ) . In this analysis , the size and intensity of each protein band was quantified by densitometry and expressed as integrated peak density values representing the amount of protein in the respective band . Different biological activities in tsetse saliva that were previously demonstrated or that could be predicted by EST-database analysis were assayed: thrombin-inhibitory ( TTI ) , apyrase ( 5′nucleotidase , gmmspg3 ) and adenosine deaminase activity ( TSGF1/2 ) . Salivary apyrase activity was quantified by assessing the dephosphorylation rate of 20 µM ATP and ADP at 27°C in a 25 mM Tris/HCl pH 7 . 8 buffer supplemented with 2 . 5 mM CaCl2 . ATP/ADP-conversion was monitored after 1 hour by quantifying the release of inorganic phosphate ( Pi ) using the Malachite green phosphate assay kit ( Gentaur ) and O . D . measurement ( λ = 650 nm ) in an Multiskan Ascent microplate reader ( ThermoScience ) . The ATPase and ADPase activities in tsetse saliva samples were expressed as pmole Pi release/min × salivary gland . Adenosine deaminase activity in tsetse saliva samples was measured spectrophotometrically by a direct kinetic assay , monitoring the change in O . D . ( λ = 265 nm ) upon conversion of adenosine into inosine . This ADA activity assay was performed in 10 mM HEPES 150 mM NaCl buffer ( pH 7 . 5 ) containing 100 µM adenosine and O . D . values were recorded at 15 s interval over a period of 5 min in a microplate reader ( ThermoScience ) . The ADA activity in the saliva samples was expressed as milliUnits ADA/salivary gland , where 1 Unit ADA will deaminate 1 µmole of adenosine to inosine per minute at pH 7 . 5 ( millimolar extinction coefficient of adenosine at 265 nm = 8 . 1 ) . The thrombin inhibitory potential of saliva ( 1/80–1/400 dilution ) was quantified in 96-well plates by a kinetic assay at 37°C that monitors the release of p-nitroanilide ( pNA ) from 750 nM of thrombin chromogenic substrate ( β-Ala-Gly-Arg-p-nitroanilide diacetate , Sigma ) by the proteolytic activity of human thrombin ( Roche , 500 mU/ml ) in PBS . pNA-release was measured for at least 1 hour at λ = 405 nm in a microplate reader ( ThermoScience ) . The thrombin inhibitory potential of the saliva samples was expressed relative to the pNA release obtained with thrombin ( 100% activity ) . The platelet aggregation was monitored in a 96-well flat-bottom microplate assay as described elsewhere [48] . Platelet-rich plasma ( PRP ) was prepared from venous human blood that was anticoagulated in Monovette coagulation tubes ( Sarstedt ) . Aggregation of platelets was induced at 37°C with 10 µM ADP ( in 150 mM NaCl ) in the presence or absence of serial saliva dilutions ( 1/100–1/400 , in 150 mM NaCl ) from SG+ and SG- flies . Reduction in optical density ( increase in transmission ) at 650 nm wavelength was monitored as a measure for platelet aggregation . Human platelet-poor plasma ( PPP ) , prepared by pelleting the platelets in PRP ( see above ) at 1500 ×g for 15 min , was used for thrombin-induced coagulation assays . Briefly , coagulation was triggered in a total volume of 180 µl by the 1/3 addition of PPP to 10 mM HEPES ( pH 7 . 4 ) 12 . 5 mM CaCl2 supplemented with thrombin at a 25 mU/ml final concentration in the presence or absence of 1/50–1/6400 dilutions of saliva from SG+ and SG- flies . Coagulation was measured as a steep increase in absorbance ( λ = 405 nm ) . The lag phase preceeding coagulation onset was determined as a measure for anti-coagulation activity in the respective saliva samples . tsal1 ( AF259958 ) , tsal2 ( EF409243 ) , tsgf1 ( AF140521 ) , tsgf2 ( AF140522 ) , 5nuc ( AF384674 ) , sgp3 ( EF398273 ) , tag5 ( AF259957 ) , tti ( AF054616 ) . | Human African Trypanosomiasis , or sleeping sickness , is a devastating parasitic disease that is fatal if left untreated . Infections are acquired via the bite of an obligate blood feeding fly , the tsetse fly , that is exclusively present on the African continent . In this insect vector , the trypanosome parasite has a complex development ending in the salivary glands . In this experimental study we demonstrate that the Trypanosoma brucei parasites change the composition of the tsetse fly saliva making it less efficient to keep the blood fluid at the biting site in the mammalian host . This results in a more difficult blood feeding process and favors the fly biting activity on multiple hosts , thereby promoting the survival and circulation of the parasite within the natural host population . These findings give us a better understanding of how trypanosome infections in the human population can be maintained given the fact that only very few tsetse flies are actually carrying the parasite . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"infectious",
"diseases/protozoal",
"infections",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"microbiology/parasitology"
] | 2010 | Trypanosoma brucei Modifies the Tsetse Salivary Composition, Altering the Fly Feeding Behavior That Favors Parasite Transmission |
Cytotoxic T-lymphocyte associated protein 4 ( CTLA4 ) is a negative regulator of T-cell proliferation . Polymorphisms in CTLA4 have been inconsistently associated with susceptibility to rheumatoid arthritis ( RA ) in populations of European ancestry but have not been examined in African Americans . The prevalence of RA in most populations of European and Asian ancestry is ∼1 . 0%; RA is purportedly less common in black Africans , with little known about its prevalence in African Americans . We sought to determine if CTLA4 polymorphisms are associated with RA in African Americans . We performed a 2-stage analysis of 12 haplotype tagging single nucleotide polymorphisms ( SNPs ) across CTLA4 in a total of 505 African American RA patients and 712 African American controls using Illumina and TaqMan platforms . The minor allele ( G ) of the rs231778 SNP was 0 . 054 in RA patients , compared to 0 . 209 in controls ( 4 . 462×10−26 , Fisher's exact ) . The presence of the G allele was associated with a substantially reduced odds ratio ( OR ) of having RA ( AG+GG genotypes vs . AA genotype , OR 0 . 19 , 95% CI: 0 . 13–0 . 26 , p = 2 . 4×10−28 , Fisher's exact ) , suggesting a protective effect . This SNP is polymorphic in the African population ( minor allele frequency [MAF] 0 . 09 in the Yoruba population ) , but is very rare in other groups ( MAF = 0 . 002 in 530 Caucasians genotyped for this study ) . Markers associated with RA in populations of European ancestry ( rs3087243 [+60C/T] and rs231775 [+49A/G] ) were not replicated in African Americans . We found no confounding of association for rs231778 after stratifying for the HLA-DRB1 shared epitope , presence of anti-cyclic citrullinated peptide antibody , or degree of admixture from the European population . An African ancestry-specific genetic variant of CTLA4 appears to be associated with protection from RA in African Americans . This finding may explain , in part , the relatively low prevalence of RA in black African populations .
Cytotoxic T-lymphocyte associated protein 4 ( CTLA4 , CD152 ) is a negative regulator of T-cell activation . As the T-cell activation signal propagates due to costimulatory B7 molecule ( CD80 , CD86 ) binding of CD28 , cell surface expression of CTLA4 increases to compete with CD28 [1] . CTLA4 also prevents further clonal expansion of effector T-cells , including regulatory T cells ( Treg ) [2] , [3] , and can inhibit osteoclast formation [4] . Genetic variation in CTLA4 ( Chromosome 2q33 ) could contribute to unchecked T cell or osteoclast activation with resultant onset of autoimmune disease such as rheumatoid arthritis ( RA ) . CTLA4 was modestly associated with RA in a recent genome wide association study ( GWAS ) of RA in Caucasians [5] . CTLA4 single nucleotide polymorphisms ( SNP ) , such as rs231775 ( +49A/G ) , have been associated with multiple autoimmune conditions including RA , Addison's disease , autoimmune pancreatitis [6] , autoimmune thyroid disease , celiac disease , chronic inflammatory arthritis [7] . multiple sclerosis [8] , type I diabetes mellitus , Sjögren's syndrome [9] , and systemic lupus erythematosus ( SLE ) [10] . An association with another SNP , rs3087243 ( +60C/T ) , and RA was found in a Chinese Han population [11]; however , these results were not replicated in Irish [7] , United States Caucasian [12] , or , when corrected for multiple testing , British Caucasian [13] populations . Analysis of a much larger group of Caucasians from North America and Sweden associated this marker with RA [particularly with the anti-cyclic citrullinated peptide ( anti-CCP ) antibody positive RA subset] [14] . Given the association of CTLA4 with multiple diseases in various populations , we sought to characterize the genetic contribution of CTLA4 to RA in African Americans – a population not yet explored . RA is purported to be less prevalent in African Americans than in Caucasians based on clinical observation and data in black continental Africans [15]–[19] . African-specific protective alleles might explain the lower disease prevalence among persons of African ancestry and should be evaluated in genetic studies with this population . In this study , we genotyped CTLA4 haplotype tagging SNPs ( htSNPs ) in two groups totaling 505 African American patients with RA and 712 African American healthy controls . We found and replicated a novel protective association at an ethnic-specific intronic SNP , rs231778 , in both independent groups . While this SNP is polymorphic only in the HapMap Yoruba population , we confirmed a lack of variation by genotyping 530 Caucasians . Importantly , we did not detect significant confounding for association of rs231778 when our patients were stratified by level of European admixture or by RA subclassification such as presence of the HLA-DRB1 shared epitope ( SE ) or anti-cyclic citrullinated peptide ( anti-CCP ) antibodies [20] . We also did not find association with two SNPs ( rs3087243 and rs231775 ) previously reported to have disease associations with RA in European ancestry populations or with other autoimmune diseases . Our data reveal a protective African ancestry-specific allele that may contribute to the purportedly lower prevalence of RA in persons of African ancestry and provide suggestions for future research into the relationship between T cell regulation and RA pathogenesis .
The Consortium for the Longitudinal Evaluation of African Americans with Early Rheumatoid Arthritis ( CLEAR ) Registry enrolled self-identified African Americans with RA who met the American College of Rheumatology ( ACR ) 1987 diagnostic criteria [21] . Participants for CLEAR were recruited from the University of Alabama at Birmingham ( UAB ) [coordinating center]; Emory University/Grady Hospital ( Atlanta , GA ) ; University of North Carolina at Chapel Hill; Medical University of South Carolina ( Charleston , SC ) ; and Washington University ( St . Louis , MO ) . Recruitment occurred in two phases: enrollment of patients with early RA ( <2 year disease duration ) followed longitudinally until 5 years disease duration , from 2000 to 2007 ( CLEAR I ) ; and enrollment of patients with RA of any duration from the same sites as part of a cross-sectional analysis from 2007 to present ( CLEAR II ) . Comprehensive demographic , clinical , and radiographic data are being collected on all CLEAR participants , and serum and DNA samples are being stored [22] . These data allow for stratification of RA patients [20] by presence of the HLA-DRB1 SE and anti-CCP antibody positivity . We have also measured estimated global admixture using a panel of ancestry informative markers ( AIMs ) , as previously reported [23] . A group of healthy African American controls , for the longitudinal arm of this study , with similar sex , age , and geographic location has been recruited , as previously described [23] . All participants were recruited with informed consent under the approval of each respective Institutional Review Board . Genomic DNA was isolated using standard methods and stored at −70°C . This study included 282 African American RA patients and 149 African American controls from the CLEAR longitudinal study ( CLEAR I ) and 223 African American RA patients from the CLEAR cross-sectional study ( CLEAR II ) . We also obtained DNA samples from an additional 563 healthy African Americans from Alabama recruited for a case-control study of SLE [24] to use as controls for the CLEAR II RA patients . Demographics for CLEAR I and CLEAR II RA patients are presented in Table 1 . Controls were younger than the RA patients ( mean age: CLEAR I = 45±14 years , CLEAR II = 35±11 years ) . Similar to the patient groups , both of the control sets were predominantly female ( percent female: CLEAR I = 82% , CLEAR II = 74% ) . In total , we analyzed 505 African American RA patients and 712 African American controls . We used RA patients and controls from CLEAR I as an initial test set and RA patients from the CLEAR II and additional Alabama controls as a replication group . All SNPs within the CTLA4 region ( ±2 kb ) that have a minor allele frequency ( MAF ) ≥0 . 05 in the Yoruba HapMap population ( Phase II/Release 21 ) were genotyped: rs231775 , rs231776 , rs231777 , rs231778 , rs231779 , and rs3087243 . Data from the resequencing of CTLA4 in both African and European populations contracted to SeattleSNP ( Dr . Debbie Nickerson , University of Washington ) were kindly provided from the Population Genetics Study coordinated at UAB ( Drs . Richard Kaslow and Robert Kimberly ) . CTLA4 SNPs detected by SeattleSNP that capture information on polymorphisms not present in HapMap for Africans with a MAF ≥0 . 05 were additionally genotyped: rs11571319 , rs231772 , rs231780 , rs34031880 , rs733618 , and *5251 . *5251 is not yet listed in dbSNP: its physical location is 54945227 in NCBI contig file NT_005403 , and its surrounding sequence is ATGGTAGCCTTGCTTATTGT [G/T] GGTGGCAACCTTAATAGCAT . Genotyping was performed by the Illumina FastTrack GoldenGate BeadXpress genotyping service ( San Diego , CA ) for CLEAR I for SNPs from the International HapMap Consortium . All other genotyping was performed using Applied Biosystems TaqMan Allelic Discrimination Assays ( Foster City , CA ) on an ABI 7900HT Genetic Analyzer . Overall , between both platforms for all SNPs , our genotyping success rate was 99 . 4% . We successfully genotyped rs231778 among 74 samples using both platforms with 100% reproducibility . To confirm the monomorphic nature of rs231778 , we genotyped this SNP in 530 Caucasian samples from the UAB Treatment of Early Aggressive Rheumatoid Arthritis ( TEAR ) study . Fisher's exact tests were performed on SAS 9 . 0 ( Cary , NC ) and exact logistic regression tests performed on LogXact 8 . 0 ( Cambridge , MA ) . We controlled for potential confounding by HLA-DRB1 status , anti-CCP antibody positivity , and genetic admixture following the approach of Redden et al . [25] . Linkage Disequilibrium and haplotype analyses were performed with HaploView v3 . 31 [26] . All SNPs were in Hardy-Weinberg Equilibrium ( tested with Chi squared tests ) , except rs231776 ( HWE p = 0 . 0085 ) , which was excluded from further analysis .
Data available from the International Haplotype Mapping Consortium ( HapMap ) , as accessed in February 2008 , appear incomplete with regard to coverage of CTLA4 . Only SNPs present from the 5′ region through intron 1 ( rs231775 , rs231776 , rs231777 , rs231778 , rs231779 ) are represented with detailed genotyping data . HapMap does not provide data for SNPs among the remaining exons and introns of CTLA4 but does present information for polymorphisms in the 3′ end of the gene , such as rs3807243 . To select htSNPs that cover the remaining interior portions of this gene , we accessed resequencing data available from SeattleSNP that provided detailed genotypes on YRI and CEU populations . SeattleSNP routinely resequences only 500 basepairs into each end of a given intron . A portion of intron 1 ( the longest intron ) is the only region of CTLA4 not completely resequenced by SeattleSNP; however , intron 1 was completely covered by HapMap , allowing the combination of these two resources to provide the most detailed haplotype tagging strategy for this gene . See Figure 1 . Due to the limited public information on CTLA4 in African Americans , we used HaploView to calculate linkage disequilibrium ( LD ) across all genotyped SNPs . A plot representing the LD ( r2 values ) of SNPs is included as Figure 2 . We detected a protective effect for RA in African Americans with the G allele of rs231778 in both CLEAR study groups ( longitudinal and cross-sectional ) independently and together ( CLEAR I and CLEAR II combined Fisher's exact p = 4 . 46×10−26 ) . See Table 2 . Because homozygotes for the G allele were rare , we compared the frequency of persons with genotypes GG and AG to those with genotype AA . From the odds ratios of the two groups combined , it can be seen that the presence of the G allele confers a protective effect ( OR = 0 . 19 , 95% CI: 0 . 13–0 . 26 , p = 2 . 4×10−28 , Fisher's exact ) . See Table 3 . rs231778 is not in LD with any other SNP , which suggests any genetic effect it confers is likely independent . See Figure 2 . The G allele of rs231778 is relatively specific for African populations as only the A allele is detected among Asians and Caucasians genotyped in the International HapMap Project and in Caucasians genotyped by SeattleSNP . Since variation at rs231778 was not found in the HapMap ( n = 24 ) or Perlegen ( n = 60 ) based European samples , we genotyped an additional 530 self-identified Caucasians to assess ethnic specific variation at this site . Among these 530 subjects , only 3 were heterozygous at rs231778 , and none were homozygous for the G allele , which yields a MAF of 0 . 0028 . In the 697 healthy African American individuals we genotyped , the MAF is 0 . 209 illustrating the ethnic specificity of this marker . Since the presence of the SE has been associated with susceptibility to RA in our population [23] and known to confound association with RA at other immunologically relevant loci such as PTPN22 [27] , we evaluated our findings in CTLA4 for possible confounding by the HLA-DRB1 SE , the strongest known genetic risk factor for RA . We found that the MAF of rs231778 was not different within cases or controls when stratified for number of SE alleles present . Because only 4 control samples have two SE alleles , we cannot rule out any possible influence of the SE on the genetic contribution of CTLA4 in RA susceptibility , but it appears to be unlikely . See Table 4 . Since our study focuses on African Americans , a group with known recent population admixture [28] , we assessed percentage of European admixture as a confounding factor in the association of RA with rs231778 . Data from a genome-wide admixture panel performed at the Broad Institute from our previously reported work [23] allowed calculations of global admixture estimates ( percent European ancestry ) for 282 cases and 94 controls ( total N = 366 ) . Of these 366 with admixture data , there was successful genotyping for the CTLA4-containing region of Chromosome 2 in 266 cases and 81 control samples ( total N = 347 ) . We show the mean percentage of European ancestry segregated by genotype for cases and controls in Table 5 . The degree of admixture was not associated with rs231778 genotype ( Fisher's exact p = 0 . 2367 ) . We confirmed that admixture difference between cases and controls was not significant using the robust Welch test , which produced a value of 2 . 308 ( degrees of freedom = 215 . 578 , p = 0 . 130 ) . We did not find a significant association with RA of the G allele among the 347 samples with complete admixture data and CTLA4 genotypes ( asymptotic p = 0 . 0674 ) ; we suspect that this is due to the reduced statistical power of analysis of a smaller number of subjects and controls . When we based calculations upon the 366 samples used in our previous admixture-based manuscript [23] , this small increase in sample size regained statistical significance of association with RA ( asymptotic p = 0 . 0183 ) . To illustrate further the lack of significance among the 347 samples is due to lack of power , frequency counts of genotypes among the 366 samples are incorporated in Table 5 to demonstrate a similar pattern of genotype distributions with and without these additional samples . Nonsynonymous SNPs previously associated in other populations and autoimmune phenotypes ( rs3087243 and rs231775 ) were not associated with RA in our study . See Table 6 . We also found no association when we analyzed data based upon deduced haplotypes or at any individual SNP when stratified by RA subclassification ( SE status , anti-CCP antibody status , or percent European ancestry ) as has been observed with RA associations at other sites in the genome [23] and with CTLA4 SNPs in Caucasian populations [14] . See Table 6 . We found no significant association with the G allele of rs3087243 , even when stratified for presence of anti-CCP antibody , as previously reported in Europeans with RA [14]; the distribution of genotypes and allele frequencies of this SNP were similar in anti-CCP antibody-positive and anti-CCP antibody-negative RA patients . Similarly , we found no significant differences in allele frequency between anti-CCP positive RA patients and anti-CCP negative RA patients at the SNP associated with RA in our study ( rs231778 ) . In the initial analysis of the CLEAR longitudinal arm , we found a protective effect ( lower allele frequency in patients than controls ) of the minor allele ( G ) of rs231780 allele ( Fisher's Exact p = 0 . 0123 ) . However , upon replication in the CLEAR cross-sectional arm , this difference in MAF between cases and controls was not significant in the cross-sectional arm or in both arms combined [Fisher's exact p = 0 . 0667; MAF 0 . 097 in patients , 0 . 124 in controls] . See Table 6 . Of note , the rs231780 SNP appears to be African-ancestry specific as well , with a MAF of 0 . 17 in Africans and ∼0 . 00 in Europeans among HapMap subjects . It is possible that our lack of association at this marker is due to a true negative state or due to lack of power for detecting a positive association , as our the p value is bordering on significance ( p = 0 . 07 ) . Although rs231780 is also an ethnic-specific SNP , there is not significant LD between it and the strongly associated SNP rs231778 ( r2 = 0 . 107 , D′ = 0 . 445 ) . The lack of association of the African-specific SNP , rs231780 , with RA might be sufficient to rule out genetic admixture as the cause of the association at rs231778 . We also found an association with rs231776 ( Fisher's exact p = 0 . 0418 ) when both study groups were combined; however , this SNP was not in Hardy-Weinberg equilibrium ( HWE p = 0 . 0085 ) , complicating interpretation of these results .
We detected a significant novel genetic association with RA in African Americans at the CTLA4 SNP rs231778 . In this case-control study , African Americans with at least one minor ( G ) allele were 0 . 19 times as likely to have RA as those without a minor allele ( 95% CI 0 . 13 0 . 26 , Fisher's Exact p = 2 . 437×10−28 ) . This P value does not appear to be subject to the inaccuracy introduced by cancellation error by complementation [29] . Our study is limited in sample size due to its exclusive focus on a minority population , which may introduce influence by bias in sample collection , genotyping errors , and lack of power . However , due to our efforts in matching patients and controls and validating our genotyping results ( 100% reproducibility in 74 samples on different genotyping platforms ) , we believe such biases have been minimized . We believe that our study is sufficiently powered to detect associations as we found a statistically significant result in two separate arms of the study . The associated SNP , rs231778 , is located in intron 1 and is not in LD with any genotyped SNP in CTLA4 such as the disease-associated rs3087243 and rs231775 markers . See Figures 1 and 2 . It is possible , however , that LD could span farther than assessed in this study allowing the possibility that rs231778 is a surrogate marker for another associated polymorphism well outside of the gene boundaries of CTLA4 . LD has been shown to span several megabases in African Americans , which supports this possibility [30] . Additional genotyping of 5–10 AIMs in this chromosomal region in a large number of African Americans may allow a better understanding of the long-range haplotype structure . Our study did include five African-specific SNPs ( rs231772 , rs231776 , rs231780 , rs34031880 , 5251* ) and one AIM , defined as a difference in MAF >0 . 20 between populations , ( rs3087243 ) that did not associate with RA . The association of the African-specific allele of rs231778 and RA and the lack of association at these ethnic-specific markers supports the idea that the association of rs231778 is independent from bias by genetic admixture . Interestingly , rs231778 is monomorphic in both Asian and Caucasian populations , according to genotype data from HapMap and SeattleSNP , and virtually absent in our genotyping of 530 Caucasians . Given the ethnic-specific status of this SNP , it is possible that our finding helps to explain the purported , but as yet unproven , observation of a lower prevalence of RA in African Americans compared to Caucasians . We would anticipate the association of such African-specific protective alleles with resistance to RA . Racial or ethnic differences have now been suggested in the association of RA with several genes , including PTPN22 [31] , PADI4 [32] , SLC22A4 and RUNX1 [33] , and in CTLA4 , particularly between Asian and Caucasian populations [10] , [34] . These data highlight the need for additional research into the genetic background of RA in various populations such as African Americans to uncover additional ethnic-specific associations . Our study included 697 healthy African American controls that possessed a MAF of 0 . 209 at rs231778 . This finding is surprising since public resources such as the International HapMap Consortium has a MAF of 0 . 09 in their panel of 60 Yorubans and 0 . 00 in 60 Caucasians . African Americans are considered to be an admixed population with an African background and contribution of approximately 20% European genetic ancestry . In fact , we calculated that European ancestry contributes 15±5% of the genetic composition of African Americans in the CLEAR study . Therefore , we would expect a MAF for rs231778 to be between 0 . 00 and 0 . 09 . Given that our participants were collected at multiple centers across the Southeastern United States ( with each center having similar MAFs ) , that we genotyped 74 samples with 100% reproducibility on dual platforms ( TaqMan and Illumina ) , and that our study included a larger number of samples ( n = 697 ) than public resources ( n = 60 ) , we believe our results are accurate . Such a difference from the expected MAF may be due to reduced power in HapMap compared to this work or due to population stratification ( i . e . the MAF of 0 . 09 for Yorubans in Nigeria could be markedly lower than elsewhere on the continent from where ancestors of our participants may have lived ) . More work into the genetic population structure across Africa and in admixed populations such as African Americans is needed to appreciate such differences . Population-based differences in susceptibility to RA are observed through previous reports that show an association between RA and rs3087243 ( +60C/T ) , a polymorphism known to affect the expression levels of soluble CTLA4 protein [35] , in Swedish and North-American populations [14] or a lack of association at this locus in studies based in Massachusetts or Northern Ireland [7] , [13] . We failed to find an association of rs3087243 in RA among African Americans . Even when stratifying for a clinical subclassification more strongly associated with CTLA4 [14] ( anti-CCP positivity ) , we could not reproduce these results in African Americans . This non-replication finding may be due to genuine population-specific differences in allele frequency or different patterns of LD among African and European ancestry individuals , but our relatively small sample size precludes definitive conclusions . For example , to detect a small genetic effect [OR = 1 . 08 ( 95% CI: 1 . 01–1 . 17 ) ] in a meta-analysis of genotypes , Plenge et al . analyzed ∼4 , 000 Caucasian RA samples [14] , a much higher number of subjects than is available for our analysis . We also failed to find an association with the nonsynonymous SNP , rs231775 ( +49A/G ) , which has been implicated in multiple autoimmune diseases , again possibly due to small sample size . CTLA4 is an important molecule in preventing an inappropriate immune response and in dampening osteoclast formation [4] , both of which may have implications for the pathogenesis of RA . CTLA4 stimulation functions in regulatory T cell development including proliferation and frequency [2] , [3] , [36] , providing another possible mechanism for this protein to influence RA pathogenesis . While we do not address possible functional consequences of this polymorphism , future work may reveal a relationship between rs231778 and T cell/osteoclast development or linkage disequilibrium with a SNP outside of the CTLA4 gene boundaries that influences expression or function . In conclusion , our results suggest a need for greater understanding of CTLA4 function and of the ethnic-specific genetic contributions to RA including relationship to disease pathogenesis . | Rheumatoid arthritis ( RA ) is a systemic autoimmune condition affecting the synovial membranes of diarthrodial joints . The etiology of RA is unclear but is thought to result from an environmental trigger in the context of genetic predisposition . We report that a single nucleotide polymorphism ( SNP ) ( rs231778 ) in CTLA4 , which encodes a negative regulator of T cell activation , is associated ( p = 2 . 4×10−28 ) with protection from developing RA among African Americans . rs231778 is only polymorphic in populations of African ancestry . Protective alleles such as this one may contribute to the purported lower prevalence of RA in African Americans . Our finding appears to be independent from confounding by linkage with the HLA-DRB1 shared epitope or by genetic admixture . Furthermore , we did not replicate associations of CTLA4 SNPs with RA or other autoimmune diseases previously reported in Asians and Caucasians , such as rs3087243 ( +60C/T ) and rs231775 ( +49A/G ) . The associations of different SNPs with RA susceptibility specific to different populations highlight the importance of CTLA4 in the pathogenesis of RA and demonstrate the ethnic-specific genetic background that contributes to its susceptibility . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"rheumatology/rheumatoid",
"arthritis",
"genetics",
"and",
"genomics/genetics",
"of",
"the",
"immune",
"system"
] | 2009 | An African Ancestry-Specific Allele of CTLA4 Confers Protection against Rheumatoid Arthritis in African Americans |
MicroRNA cluster mirn23a has previously been shown to promote myeloid development at the expense of lymphoid development in overexpression and knockout mouse models . This polarization is observed early in hematopoietic development , with an increase in common lymphoid progenitors ( CLPs ) and a decrease in all myeloid progenitor subsets in adult bone marrow . The pool size of multipotential progenitors ( MPPs ) is unchanged; however , in this report we observe by flow cytometry that polarized subsets of MPPs are changed in the absence of mirn23a . Additionally , in vitro culture of MPPs and sorted MPP transplants showed that these cells have decreased myeloid and increased lymphoid potential in vitro and in vivo . We investigated the mechanism by which mirn23a regulates hematopoietic differentiation and observed that mirn23a promotes myeloid development of hematopoietic progenitors through regulation of hematopoietic transcription factors and signaling pathways . Early transcription factors that direct the commitment of MPPs to CLPs ( Ikzf1 , Runx1 , Satb1 , Bach1 and Bach2 ) are increased in the absence of mirn23a miRNAs as well as factors that commit the CLP to the B cell lineage ( FoxO1 , Ebf1 , and Pax5 ) . Mirn23a appears to buffer transcription factor levels so that they do not stochastically reach a threshold level to direct differentiation . Intriguingly , mirn23a also inversely regulates the PI3 kinase ( PI3K ) /Akt and BMP/Smad signaling pathways . Pharmacological inhibitor studies , coupled with dominant active/dominant negative biochemical experiments , show that both signaling pathways are critical to mirn23a’s regulation of hematopoietic differentiation . Lastly , consistent with mirn23a being a physiological inhibitor of B cell development , we observed that the essential B cell transcription factor EBF1 represses expression of mirn23a . In summary , our data demonstrates that mirn23a regulates a complex array of transcription and signaling pathways to modulate adult hematopoiesis .
Non-coding RNAs , including microRNAs ( miRNAs ) , play a critical role in regulating hematopoietic gene expression networks . MiRNAs are ~22 nucleotide long RNA molecules that negatively regulate gene expression post-transcriptionally by binding to the 3’ untranslated region ( UTR ) of target mRNAs[1 , 2] . We previously screened for miRNAs regulated by the hematopoietic transcription factor PU . 1 ( Sfpi1 ) that could be involved in immune cell fate acquisition . We identified the mirn23a miRNA cluster as a PU . 1 activated gene[3] . The mirn23a gene is located on murine chromosome 8 and codes for 3 pre-miRNAs: miR-23a , miR-24-2 , and miR-27a . Overexpression of mirn23a in hematopoietic progenitors biases cell fate decisions towards the myeloid lineage at the expense of the lymphoid lineage both in vitro and in vivo[3] . Recently we reported that when mirn23a is deleted from mice , there is an increase in B cell development and a concomitant decrease in myelopoiesis in the bone marrow that persists in the periphery . This differentiation bias occurs early during hematopoietic development , as an increase in common lymphoid progenitors ( CLPs ) is observed in mirn23a-/- bone marrow , as well as a decrease in common myeloid progenitors ( CMPs ) , granulocyte/ monocyte progenitors ( GMPs ) and megakaryocyte/ erythroid progenitors ( MEPs ) . No differences in hematopoietic stem cells ( HSC ) and multipotential potential progenitors ( MPP ) were observed . The overexpression and knockout results suggest that mirn23a miRNAs promote myelopoiesis through repressing lymphopoiesis . However , the hematopoietic targets that mirn23a regulates to drive myeloid cell development are not clear . Commitment of bone marrow hematopoietic stem and progenitor cells ( HSPCs ) to specific lineages is tightly controlled by a complex network of cell intrinsic and extrinsic cues that regulate downstream signaling cascades and transcriptional pathways . [4–7] The initial differentiation event towards a committed hematopoietic lineage is from the MPP into the CLP or CMP . It has recently become appreciated that the MPP pool can be sub-divided into 4 distinct MPP subpopulations that have varying polarizations towards different hematopoietic lineages[8] . Commitment to the CLP from the MPP is heavily dependent on the expression of several hematopoietic transcription factors , including PU . 1 ( Sfpi1 ) , Ikzf1 ( Ikaros ) Mef2c , Satb1 and Runx1[9–13] . Commitment to the B cell lineage from the CLP is dependent on B cell fate determinants E2A , EBF1 , and FoxO1[14–16] . Pax5 along with Bach1 and Bach2 , repress myeloid genes in CLPs and early B cell precursors to lock them into the lymphoid cell fate[17 , 18] . In this study , we follow up previous overexpression and knockout experiments in an attempt to elucidate hematopoietic pathways regulated by mirn23a . Although we previously reported that mirn23a loss does not alter the pool of MPPs , here we observe that it does result in changes in the polarized MPP subsets . In vitro culture of MPPs and in vivo transplants reveals that loss of mirn23a inhibits myeloid differentiation and promotes lymphoid development from the MPP pool . Gene and protein analysis from multipotential EML cell lines[19] generated from mirn23a+/+ and mirn23a-/- mice show that misregulation of Ikzf1 , Bach1 , Satb1 , and Runx1 in the absence of mirn23a contribute to an increased output of lymphoid cells . Additionally , we observed that mirn23a regulated downstream effectors of both the PI3K/Akt and BMP/Smad pathways , suggesting that , mirn23a agonizes PI3K/Akt while simultaneously antagonizing BMP/Smad signaling in order to promote myelopoiesis . Our data suggest a model where mirn23a regulates hematopoietic differentiation through buffering the expression of transcription factors and coupling the activation and repression of 2 critical signal transduction pathways .
We previously observed that overexpression of mirn23a in hematopoietic progenitor’s increased myeloid development at the expense of B cell development[3] . Similarly , analysis of immune cell populations in mirn23a-/- mice showed increased B cell development at the expense of myeloid development[20] . Analysis of hematopoietic stem and progenitor cell ( HSPC ) populations revealed increased CLPs and decreased CMPs in mirn23a-/- mice , while MPP populations were unchanged . Since MPP populations were unchanged in mirn23a-/- mice , this suggested that mirn23a null MPPs are more biased towards the CLP lineage than wildtype MPPs . It has recently become appreciated that the MPP pool consists of at least 4 distinct polarized populations , with the MPP3 being biased toward myeloid differentiation and the MPP4 being biased toward lymphoid differentiation[8] . To examine whether these populations were altered in the bone marrow mirn23a-/- mice , we stained nucleated hematopoietic cells with the previously described markers to identify MPP1-4 populations ( Fig 1A . ) This analysis revealed no significant difference in the total MPP pool , consistent with our previous results ( Fig 1B ) . However , when the MPP populations were compartmentalized into MPP1-4 populations , this surprisingly revealed a significant increase in the MPP1 and MPP3 ( myeloid-biased ) populations , and an accompanying decrease in the MPP2 and MPP4 ( lymphoid-biased ) bone marrow populations in the mirn23a-/- mice compared to wildtype ( Fig 1C ) . To examine the differentiation potential of mirn23a-/- MPPs , we performed in vitro differentiations . Previously we observed that culturing a pool of mirn23a-/- stem and progenitor cells ( Lineage negative ) on OP9s resulted in increased B lymphopoiesis with decreased myelopoiesis compared to wildtype cells . To determine if MPPs were responsible for this , we sorted 1 , 10 , or 100 MPP ( LSK CD34+ ) cells onto 96 well plates coated with OP9 stromal cells in the presence of IL-7 and Flt3L for 8 or 20d ( Fig 2A ) . After 20d less than 5% of the wells in which 1 MPP was sorted gave rise to enough cells to analyze , which was not enough to generate any significant conclusions . After 8d , the 10 and 100 cell well cultures contained sufficient cells per well for analysis . Consistent with previous in vitro stromal culture of MPPs , the cultures were predominantly myeloid ( CD11b+ ) at 8d[16] . At 8d , the 10 cell/well mirn23a+/+ cultures consist of ~65–70% CD11b+ myeloid cells , while mirn23a-/- cultures only contain ~40–45% CD11b+ cells , suggesting that loss of mirn23a hinders or slows myeloid differentiation ( Fig 2B ) . There are no detectable B220+ cells in either culture , but the CD11b-B220- double negative population is significantly increased in mirn23a-/- cells , further suggesting the inhibition of myeloid development from MPP ( Fig 2B ) . Similar results were obtained with 100 cell/well cultures ( S1A and S1B Fig ) . At 20d , there were enough cells in the 10 cell/well culture for flow cytometric analysis ( 100 cells/well were too dense with extensive cell death at this timepoint precluding meaningful analysis ) . After 20d , lymphoid cells were present in the cultures ( B220+ ) . Individual mirn23a+/+ MPP cultures generated ~25% B220+ cells , while mirn23a-/- cultures generated ~45% B220+ B cells ( Fig 2C ) . These cultures have no detectable CD11b+ populations . This suggests that loss of mirn23a promotes lymphoid differentiation from the MPP pool . To investigate whether mirn23a null M5PPs have increased lymphoid potential in vivo , we sorted LSK CD34+ MPPs ( same gating scheme in Fig 2A ) from CD45 . 2 mirn23a-/- mice and transplanted 2 , 000 MPPs with 2x105 CD45 . 1 support marrow cells into lethally irradiated CD45 . 1 recipient mice . Contribution to the B cell and myeloid cell lineage was examined 4 weeks after transplant and revealed a significant increase in B220+ B cells from mirn23a-/- MPPs ( Fig 2D and 2E ) . Since the contribution to the lymphoid lineage was low compared to the myeloid lineage ( ~3% vs ~70% respectively in WT mice ) , decreases in the myeloid lineage were harder to observe , and although we observed a slight decrease in CD11b+ myeloid cells , these changes were not statistically significant ( Fig 2E ) . To investigate whether defects in proliferation could contribute to these phenotypes , we conducted in vivo BrdU experiments that revealed no difference in proliferation between wildtype and mirn23a-/- LSK cells ( S2A and S2B Fig ) . Further analysis by FACS for cell surface markers Annexin V and 7AAD revealed no differences in apoptosis ( S2C and S2D Fig ) . Together , this work shows that mirn23a-/- MPPs have increased lymphoid potential in vitro and in vivo compared to wildtype MPPs , despite the population having an increase in myeloid-biased cells based on cell surface phenotype . To elucidate the molecular mechanisms driving increased CLP production from the MPP pool in mirn23a null mice , we created multipotent erythroid-myeloid-lymphoid ( EML ) cell lines from wild type and mirn23a-/- mouse bone marrow[19] . These cells , along with being multipotent , express stem cell markers c-Kit and CD34 , while not expressing any committed lineage markers , including CD11b and B220 , suggesting that they model MPPs ( S3A Fig ) . Consistent with our primary bone marrow cell data , differentiation of wildtype and mirn23a-/- EML cells to the myeloid lineage revealed that mirn23a-/- EMLs have impaired myeloid differentiation ( Fig 3A ) . Conversely , differentiation of EML cells to the lymphoid lineage was enhanced in mirn23a-/- EMLs ( Fig 3B ) . Mechanistically , development of the MPP to the CLP is dependent on the presence of essential hematopoietic transcription factors . Among these critical transcription factors are Runx1 , Satb1 , Bach1 , and Ikzf1 , all of which are required for normal lymphoid development[9 , 11 , 17 , 21] . Runx1 and Ikzf1 have been shown to be targets of mirn23a miRNAs in hematopoietic cells , and Satb1 was shown to be a direct target in osteosarcoma cells [22–24] . Direct targeting by mirn23a miRNAs for each gene was validated by luciferase-UTR assays . The Bach1 3’UTRs contains a conserved targeting site for miR-23a/b as predicted by the targetscan algorithm[25] . We observe that expression of the entire mirn23a cluster or miR-23a alone represses expression of a luciferase transcript containing the Bach1 3’UTR ( Fig 3C ) . To determine if mirn23a loss affected expression of these factors in multipotent EML cells , we collected RNA from wildtype and mirn23a-/- EML cells and analyzed for expression of transcription factors by qRT-PCR . This analysis revealed a significant increase in Satb1 , Runx1 , Bach1 , and Ikzf1 mRNA expression in mirn23a-/- EMLs ( Fig 3D ) . Furthermore , reintroduction of mirn23a to mirn23a-/- EMLs decreased expression of these critical lymphoid transcription factors ( Fig 3D ) . To validate the increase in these factors at the protein level , we collected whole cell lysates from wildtype and mirn23a null EML cells and analyzed for protein expression by immunoblot . This analysis confirmed that BACH1 , RUNX1 , SATB1 , and IKZF1 protein levels were all increased in mirn23-/- EML cells ( Fig 3E ) . To examine whether these critical factors were changed in vivo , we collected RNA from sorted LSK CD34+ MPP populations and analyzed gene expression by qRT-PCR . Bach1 , Satb1 , and Runx1 expression were all significantly increased , while no change was observed in Ikzf1 expression ( Fig 3F ) . Taken together , these results suggest that loss of mirn23a results in increased expression of transcription factors that skew the MPP differentiation towards the CLP at higher frequencies than observed in wildtype cells . Previously we observed that the majority of the increased CLP population consists of Ly6D+ B cell biased progenitors ( BCPs ) [20] . To further identify B cell transcription networks inhibited by mirn23a that direct commitment of the CLP to the B cell lineage , we overexpressed the individual cluster miRNAs in the 70Z/3 pre-B lymphoblast cell line using murine stem cell virus ( MSCV ) retroviruses co-expressing GFP . High expressing lines were generated through limiting dilution and 2 unique lines for each miRNA and a control line infected with empty retrovirus were analyzed for genome wide RNA expression by microarray analysis using Affymetrix Mouse Genome 430 2 . 0 Arrays ( S1–S3 Tables ) . Examining genes differentially expressed 2 fold or more we observed that exogenous miR-24-2 expression results in upregulation of myeloid cell associated genes , including Lyz , Ccl9 , and Csf1r . This is consistent with our previous study showing that miR-24 alone could enhance myeloid development in vitro[3] . To validate the upregulation of myeloid specific genes , RNA was prepared from control or mirn23a cluster ( miRs-23a , -24-2 , and -27a ) overexpressing 70Z/3 cells and gene expression was assayed by qRTPCR . MiR-24 expression is ~8 fold overexpressed in these cells ( S4A Fig ) . Myeloid genes Ccl9 , Csf1r , and Lyz1 were upregulated in cells overexpressing all 3 cluster miRNAs ( Fig 4A ) . We also examined expression of the transcription factors Sfpi1 ( PU . 1 ) and Runx1 that are known to regulate these genes , but did not observe significant changes . However we cannot rule out that activity and/ or protein expression of PU . 1 and RUNX1 is affected . Overexpression of the individual cluster miRNAs demonstrated that miR-24 alone could enhance Ccl9 and Csf1r expression ( S4B Fig ) . The increased myeloid gene expression could potentially be due to decreased expression of essential B cell transcription factors E2A ( Tcf3 ) , Ebf1 and Pax5[16 , 26 , 27] . We isolated RNA from 70Z/3 MSCV , or MSCV-mirn23a cluster expressing cells and assayed for gene expression of E2A ( Tcf3 ) , Ebf1 , Pax5 , and Ikzf1 . This analysis revealed significantly decreased Ebf1 , Pax5 , and Ikzf1 expression in mirn23a expressing cells , while no change was observed in E2A expression ( Fig 4B ) . In addition , since the transcription factors Bach1 was shown to repress myeloid genes in lymphoid cells , we analyzed their expression in 70Z/3 cells and observed that it is downregulated by mirn23a expression ( Fig 4B ) . In contrast to the myeloid genes examined above , the majority of the lymphoid genes were downregulated by the expression of a single miRNA member of the cluster ( S4C Fig ) . To investigate whether lymphoid transcription factors were increased in primary cells lacking mirn23a , we sorted CLP populations from wildtype and mirn23a-/- bone marrow and analyzed gene expression by qRT-PCR . This analysis revealed significantly increased expression of EBF1 ( Fig 4C ) . Bach1 was not observed to be affected by loss of mirn23a , however the related factor Bach2 , which also represses myeloid associated genes , was overexpressed in mutant CLPs[17] . To identify pathways affected by mirn23a that could promote lymphoid development , we performed a gene set enrichment analysis ( GSEA ) [28] with the expression data obtained from the 70Z/3 microarrays . A heat map of a ranked list of the top 50 upregulated and downregulated genes in the miRNA expressing 70Z/3 lines compared to the empty vector expressing lines is shown ( S5 Fig ) . The miRNA expressing lines were enriched for genes associated with IL2/Stat5 signaling pathway in control cells compared to miRNA overexpressing cell lines ( S6 Fig ) . Examination of gene expression in in mirn23a-/- CLPs showed that Stat5b is upregulated in the absence of mirn23a miRNAs ( Fig 4C ) . Previously Stat5b was shown to be critical downstream of the IL7 receptor for commitment of lymphoid progenitors to the B cell lineage[29] . Exogenous expression of Ebf1 in hematopoietic progenitors increases B-cell development and antagonizes myeloid promoting transcription factors C/EBPα and PU . 1 . [16] Since Ebf1 is not a validated or predicted target of the mirn23a cluster , we examined the list of mirn23a downregulated genes in 70Z/ overexpressing lines that have the potential of regulating Ebf1 expression . Tribbles 3 ( Trib3 ) was downregulated in the miR-24-2 expressing 70Z/3 cells as determined by microarray analysis , and has previously been shown to be a miR-24-2 target in vascular smooth muscle cells[30] . In addition , we previously observed that forced expression of Trib3 in OP9 cocultures results in increased B cell development at the expense of myeloid development , consistent with the phenotype of mirn23a-/- mice . Trib3 has the potential to regulate Ebf1 through negative regulation of Akt[31] , which antagonizes nuclear forkhead box protein O1 ( FoxO1 ) [31] . FoxO1 has previously been shown to drive Ebf11 expression in hematopoietic cells , which is necessary for early B-cell maturation and peripheral immune cell function[15 , 31 , 32] . Trib3 also has the potential to regulate B cell development through negative regulation of Smurf1 and the BMP/SMAD pathway , as BMP4 promotes B cell development in ex vivo cultures[33] . Through these mechanisms , we hypothesized that mirn23a regulation of Trib3 could help sustain commitment to the B cell lineage ( as opposed to T cell/ NK cell ) downstream of the MPP . We validated that miR-24-2 and mirn23a cluster expression in 70Z/3 cells decreases Trib3 expression using qRTPCR ( Fig 5A ) . To examine if Trib3 alone had the potential to regulate myeloid gene expression networks , we generated 70Z/3 pLKO . 1 control or pLKO . 1-shTRIB3 knockdown cells and isolated RNA for qRTPCR analysis . These cells have 4–5 fold knockdown of Trib3 expression ( S7A Fig ) . This revealed significantly increased expression of myeloid genes Lyz , Ccl9 , and Csf1R , similar to what was observed with mirn23a overexpression ( Fig 5B ) . Examination of B cell genes in these cells revealed decreased expression of Ebf1 and Pax5 ( Fig 5C ) . To examine whether overexpression of Trib3 could lead to decreased myeloid gene expression , we switched to the 32Dcl myeloid cell line that has high levels of endogenous myeloid gene expression , making decreases in myeloid gene expression more readily observable ( S7B Fig ) Overexpression of Trib3 in 32D myeloid cells resulted in an ~2 fold reduction in Csf1R and Ccl9 expression , while Lyz did not amplify ( Fig 5D ) . We next wanted to confirm whether Trib3 could regulate essential B cell gene expression networks in an additional cell line , so we chose the A20 B lymphoma cell line since A20 cells express high levels of endogenous Ebf1 and Pax5 . A20 pLKO . 1-shTrib3 cells were generated and showed a ~3 fold decrease in Trib3 expression ( S7C Fig ) . Consistent with mirn23a overexpression leading to decreased B cell gene expression , knockdown of Trib3 results in a >2-fold reduction in Ebf1 and Pax5 ( Fig 5E ) . To examine the effect of Trib3 knockdown on FoxO1 protein expression in these cells , we performed immunoblots and observed decreased expression of FoxO1 in shTrib3 samples ( Fig 5F ) . Together , these results suggest that miR-24 target Trib3 regulates immune cell gene expression networks that sustain lymphoid commitment while repressing myeloid development in mice . Trib3 has previously been shown to negatively regulate Smurf1 , an E3 ubiquitin ligase capable of targeting SMAD1/5 and antagonizing the BMP/SMAD pathway[34] . Activation of the BMP/SMAD pathway has previously been shown to promote B cell development in ex vivo cultures[33] . To examine whether mirn23a expression affects Smurf1 protein levels in hematopoietic cells , we performed immunoblot analysis on mirn23a+/+ and mirn23a-/- primary EML cells and observed decreased Smurf1 protein in mirn23a-/- cells ( S8A Fig ) . As expected decreased Smurf1 levels correlated with an increase in SMAD5 levels , while SMAD1 expression is not detectable in these cells ( S8B Fig ) . As discussed above , mirn23a also has the potential to regulate immune cell development through the AKT/FoxO1 pathway , which is directly targeted by Trib3 . Analysis of FoxO1 protein expression by immunoblot in primary EML cells revealed increased FoxO1 expression in mirn23a-/- EMLs ( S8C Fig ) . Mirn23a-/- EMLs also have decreased expression of active p-AKT ( inactivates FoxO1 through phosphorylation ) and p-FoxO1 ( the inactive cytoplasmic form ) ( S8D and S8E Fig ) . To functionally test whether these pathways were critical contributors to the mirn23a knockout mouse phenotype , we isolated primary Lin- cells from the BM of wildtype and mirn23a-/- mice and transduced them with pMIEV empty vector ( EV ) control or Smurf1 overexpressing retrovirus . We then cocultured these cells on OP9 stromal cells for 9d with IL-7 and Flt3L . These conditions promote the growth of pro-B-cells from hematopoietic progenitors under normal conditions[35] . By 9d , there is a mix of both myeloid and lymphoid cell . Culturing Lin- hematopoietic cells with pMIEV EV control for 9d shows that mirn23a null cells have significantly increased B cell development when compared to wildtype cells ( Fig 6A and 6B ) . However , overexpression of Smurf1 in these cells results in decreased overall B cell development and negates the phenotype of increased B cells in mirn23a-/- cultures ( Fig 6A and 6B ) . Additionally , directly antagonizing the BMP pathway with the LDN193189 BMP inhibitor during OP9 coculture ( with no retroviral inserts ) results in decreased B-cell development and similar numbers of B220+ B cells in both wildtype and mirn23a-/- cultures , showing that inhibition of the BMP pathway abrogates the increased B cell development phenotype seen in the absence of mirn23a ( Fig 6C ) . This suggests that BMP signaling is critical for both normal lymphoid development and the enhanced lymphoid development in mirn23a-/- mice . To test whether AKT/FoxO1 signaling contributes to the mirn23a-mediated phenotypes , we overexpressed a constitutively active AKT ( myrAKT ) in primary Lin- cells and observed B cell differentiation after 9d of OP9 coculture . In the absence of mirn23a , Trib3 levels rise repressing AKT activity leading to accumulation of FoxO1 in the nucleus . Expressing myrAKT was expected to reverse the stabilization of FoxO1 in mirn23a-/- cells . Like Smurf1 overexpression , this analysis revealed that overexpression of activated AKT decreases B cell development and abrogates the effect of mirn23a loss on B cell development ( Fig 6D and 6E ) . We conducted similar experiments with FoxO1 dominant negative ( DN ) [36] overexpression , but the effect of this overexpression was almost a complete block in B cell differentiation in both wildtype and mutant cultures . However , we performed OP9 cocultures with primary cells in the presence of DMSO or pharmacological FoxO1 inhibitor AS1842856 , which revealed that FOXO1 inhibition also abrogates the increased B cell development observed in mirn23a-/- cultures ( Fig 6E and 6F ) . Together , these results suggest that the BMP/SMAD pathway and the AKT/FoxO1 pathway are both critical for mirn23a’s effect on B cell development . We previously observed that mirn23a gene expression is positively regulated by the transcription factor PU . 1 . We reasoned that if mirn23a repressed B cell development , it should be a target for repression by B cell specific transcription factors . Examining ENCODE ChIP-seq[37] data for binding of B cell transcription factors binding to the mirn23a locus revealed that E2A ( Tcf3 ) , EBF1 , and PAX5 associated with the mirn23a locus , whereas other factor such as IRF4 and IKZF1 are not detected at the locus ( Fig 7A ) . Additionally , a ChIP on chip experiment with a chromosome 19 array had identified a conserved region upstream of mirn23a that bound the E2A related transcription factor Heb ( Tcf12 ) [38] . To confirm binding to the mirn23a promoter , we collected chromatin from A20 cells and performed ChIPs with pulldowns for IgG ( negative control ) , EBF1 , E2A , PAX5 , and IKZF1 ( additional negative control ) . Analysis by qRTPCR revealed increased binding of EBF1 and PAX5 over IgG control , while no differences were observed in E2A or Ikzf1 binding ( Fig 7B ) . In addition , we examined the ability of these B cell transcription factors to regulate mirn23a regulatory regions in luciferase reporter assays . We generated a luciferase reporter plasmid with an 888bp murine mirn23a promoter fragment cloned upstream of the luciferase gene transcription start site . In addition to examine E2A regulation of mirn23a we isolated a conserved upstream region of mirn23a reported to bind HEB and E2A , and cloned it upstream of the mirn23 promoter in the luciferase reporter plasmid . In transient transfections , E2A protein E47 had a modest effect on activity of the luciferase reporter containing both the promoter and conserved upstream element with a reduction of ~20% ( Fig 7C ) . E2A protein E47 can heterodimerize with the transcription factor SCL[39] . The two proteins are co-expressed together in early B cell progenitors where downregulation of mirn23a would be necessary[40] . SCL interacts with the corepressor ETO2 , which allows it to repress gene transcription[41] . Coexpressing E47 and SCL had an enhanced effect on repression of mirn23a regulation with an ~50% reduction in reporter activity . SCL alone generated similar results as expression of E47 alone . The effects of E47/SCL were rather modest so we also tested B cell factors EBF1 and PAX5 . Both PAX5 and EBF1 are implicated in repressing myeloid gene expression in developing B lymphocytes[16 , 27] . Transient transfections were performed with a reporter gene regulated by only the 888bp mirn23a reporter . Increasing amounts of Ebf1 plasmid in transient transfections resulted in a graded decrease of mirn23a promoter activity ( Fig 7D ) , whereas increasing Pax5 plasmid had a more modest effect on activity ( Fig 7E ) . Since EBF1 had the strongest effect in transient transfection , we followed up this observation by examining the effects on endogenous mirn23a expression when EBF1 levels were modified . We knocked down EBF1 in A20 B lymphoma cells using lentiviral shRNA and collected RNA for analysis by qRTPCR . A20 cells expressing shEBF1 had ~30% expression of control Ebf1 levels ( Fig 8A ) . Knockdown of EBF1 resulted in significantly increased expression of miR-23a , miR-24 , and miR-27a , consistent with EBF1 negatively regulating mirn23a ( Fig 8B ) . To evaluate whether overexpression of EBF1 leads to decreased expression of mirn23a miRNAs , we transiently transfected EBF1 into NIH/3T3 fibroblasts and collected RNA 48hr post transfection . Overexpression of Ebf1 in transfected cells was confirmed by RT-PCR ( Fig 8C ) . Cells overexpressing EBF1 showed a significant decrease in miR-23a , miR-27a , and miR-24 expression ( Fig 8D ) . Taken together , these results show that EBF1 represses mirn23a transcription , creating a regulatory feedback loop between mirn23a and EBF1 .
The miR-23a miRNA cluster promotes myeloid development at the expense of B cell development , as evidenced by overexpression and genetic knockout studies[20 , 42] . However , the pathways targeted by mirn23a that are critical to this process were previously unknown . In this study , we used a combination of bioinformatics , gene expression analysis , and functional primary cell studies to identify critical mirn23a targets and pathways that drive myeloid development at the expense of B cell development . Our results revealed that the mirn23a miRNA cluster regulates several transcription factor genes in multipotent cells , including Runx1 , Satb1 , Ikzf1 , and Bach1 . For these gene expression studies , we utilized multipotent EML cell lines derived from wildtype and mirn23a-/- mice , as well as primary sorted MPPs . While we observed 2–4 fold changes in target gene expression in the EML cell line model , this did not directly correlate with the primary MPP gene expression analysis . Increased expression of all transcription factors seen in the mutant EML cells were observed in mirn23a-/- MPPs except for Ikzf1 . However , the increases though significant , were only in the range of 1 . 4–1 . 6 fold . The less robust increase in gene expression observed in primary MPPs is likely due to the differentiation block in EML cell lines , which allows them to accumulate levels of essential transcription factors beyond what induces differentiation in vivo . The EML cells require strong differentiation promoting conditions , such as high levels of all trans retinoic acid for myeloid development or coculture with stromal cells with exogenously supplied cytokines for B cell development . However , it is important to note that these experiments , along with previous overexpression and knockout experiments with these lymphoid promoting transcription factors , do not define the precise levels of expression needed to direct differentiation . While the precise levels needed to direct differentiation have not been established , we postulate that mirn23a miRNAs buffer the levels of these transcription factors in MPPs within a narrow threshold to influence differentiation , which is likely less than 2 fold changes in expression . This is based on growing evidence that modest differences in concentration of transcription factors have distinct effects on cellular differentiation , and thus , their levels of expression must be tightly controlled[43] . This was first shown with Oct-4 in embryonic stem cells where less than a 2-fold increase in protein would direct differentiation into primitive mesoderm and endoderm , whereas a decrease would lead to development of trophectoderm[44] . This showed that a precise level of Oct4 is required to maintain pluripotency . We have also previously showed the importance of levels of PU . 1 in GMP cells in directing differentiation into monocytes and granulocytes[45] . Additionally , B cell promoting factors Bach1/2 , Ikzf1 , and Ebf1 have been shown to compete with myeloid factors to direct MPPs to differentiate into B cell progenitors , demonstrating that their concentrations relative to myeloid transcription factors is critical for B cell commitment[16 , 46–48] . Lastly , Satb1 is expressed in both HSCs and CLPs and is critical for the function of both cell types[11 , 49] . However , as mentioned previously , these studies do not delineate the precise level of lymphoid gene expression necessary for differentiation from multipotent progenitors . Currently , our model suggests that mirn23a miRNAs are part of a mechanism that maintains levels of B cell commitment factors to allow the cells to remain multipotent . In the absence of mirn23a , we postulate that normal fluctuations of protein levels are able to accumulate above a threshold level more often to commit MPPs to CLPs . Previously we observed that the majority of the CLP population in mirn23a-/- mice consists of B cell biased progenitors ( BCPs ) [20] . This bias toward B cells appears to be occurring through mirn23a inversely regulating 2 signaling pathways: PI3K/Akt and BMP/Smad . Trib3 modulation is in part responsible for this effect . Trib3 is an important downstream mediator of mirn23a as we previously reported that Trib3 overexpression could enhance in vitro B cell development similar to loss of mirn23a[20] . A recent genetic knockout study of Trib3 revealed no differences in myeloid vs lymphocyte composition , suggesting that Trib3 effects on B cell development are likely downstream of the MPP[50] . The B cell developmental populations , as well as B cell versus T cell cell fate decisions from the CLP stage , remain unknown in Trib3-/- mice . This data suggests , however , that mirn23a regulation of Trib3 is likely critical in the CLP/Pro-B stage , while the initial differentiation from the MPP is dependent on transcription factors Bach1 , Ikzf1 , Satb1 , and Runx1 . In addition , here we show that knockdown of Trib3 expression has similar effect on gene expression as overexpressing the mirn23a miRNAs ( Fig 3 ) . Mirn23a miRNA miR-24 targets Trib3 , which represses Akt kinase and Smurf1 E3 ubiquitinase activity[20 , 30 , 31 , 34 , 51] . Trib3 repression of Smurf1 leads to increased levels of BMP regulated Smads 1 and 5 in vascular smooth muscle[34 , 52] . This regulation also occurs in hematopoietic cells as loss of mirn23a results in decreased levels of Smurf1 and increased levels of SMAD1 in EML cells ( Fig 4 ) . Consistent with repression of Akt , we observed an increase in the downstream-regulated protein FoxO1 in mutant cells . Mirn23a also regulates these pathways independent of Trib3 . The PI3K/Akt pathway can be upregulated by mirn23a targeting the pathway inhibitors Pten ( miR-23 ) , and PPP2RSE ( regulatory subunit of PP2A , miR-23 ) [53 , 54] . MiR-27a synergizes with Akt by targeting transcription factors FoxO1 ( miR-27 ) that is repressed by Akt phosphorylation[55 , 56] . Mirn23a downregulates the BMP/Smad pathway through the targeting of the common Smad4 ( mir-27a , miR-24 ) , which is an obligate heterodimer partner for activated Smad1 and 5[57 , 58] . Additionally , miR-23a has been shown to target Smad5[59] . Mirn23a’s regulation of these pathways is consistent with the promotion of myelopoiesis over B lymphopoiesis . Several lines of evidence support a role for PI3K/Akt in promoting myelopoiesis . Pten deletion or expression of a dominant active Akt results in a decrease in lymphoid development and enhancement of monocyte/granulocyte development in vivo[60 , 61] . Additionally , deletion of the Akt-repressed targets FoxO1 , O3 , and O4 in HSCs results in increased myeloid progenitors[62] . Two studies report a role for BMP/Smad signaling in the commitment of MPPs to the myeloid and lymphoid lineage . Overexpression of Smad7 ( inhibitory Smad ) in human hematopoietic progenitors blocks B cell development and enhances monocyte/ granulocyte development similar to mirn23a overexpression[63] . Secondly OP9 cells support in vitro B cell growth through secretion of BMP4[33] . Our inhibitor studies suggest that both pathways are critical to mediating the differentiation phenotype observed in mirn23a-/- mice . Absence of mirn23a results in decreased Akt activity and increased FoxO1 . Treatment of wildtype and mirn23a-/- cells with FoxO1 inhibitor or myrAKT ( phosphorylates FoxO1 to reduce nuclear accumulation ) results in an ablation of the differences in B lymphopoiesis seen in OP9 cultures ( Fig 4 ) . Similarly , inhibiting BMP signaling with LDN193189 or Smurf1 E3 Ubiquitinase expression ablates the differences in lymphoid development versus myeloid development in wildtype versus mirn23a-/- cultures . This study adds to the mounting evidence that BMP signaling is critical for early B lymphopoiesis[33 , 63] . It will be important to address the molecular mechanisms regulated by SMAD transcription factors during adult B lymphopoiesis and examine whether FOXO1 and SMAD1/5 cooperate to regulate B lineage genes . Comparing the MPP populations between wildtype and mutant mice , we were expecting that the lymphoid biased MPP4 population would be increased in the mutant bone marrow , but were surprised to observe that the myeloid biased MPP3 population was enhanced and the MPP4 population decreased in the mutants ( Fig 1 ) . However , the in vitro and in vivo functional results with pooled MPPs , as well as molecular data from EML cells and primary MPPs , may explain these findings . Loss of mirn23a may be increasing MPP4 polarization through increased levels of key transcription factors while simultaneously decreasing Akt activity , leading to increased nuclear FoxO1 accumulation , which stimulates the transcription of B cell fate determinant Ebf1 . This may be speeding up differentiation of the MMP4 to B cell biased progenitors , making the MPP4 population unstable . Similarly , higher levels of transcription factor Bach1 , which inhibits myeloid genes , may be slowing down the differentiation of the MPP3 cell to the myeloid lineage , explaining their accumulation in the mutant mouse . The accumulation of MPP3 may become more pronounced as the mouse tries to compensate for the decreased mature myeloid population . Consistent with the idea that differentiation of MPP3 is slowed down , we observed an increase in the population of cells not expressing myeloid or lymphoid markers in the OP9 cultures of mirn23a-/- MPPs in the early stages of these cultures when myeloid development is predominant ( Fig 2 ) . This interpretation would explain why transplanted MPPs isolated from mirn23a-/- mice have increased contribution to the B lymphoid lineage compared to wildtype MPPs although initially consisting of more myeloid biased MPP3s . Lastly , the importance of mirn23a in commitment of progenitors to the myeloid lineages is underscored by the negative transcriptional regulation of mirn23a by key B cell transcription factors . If mirn23a regulates myeloid commitment through repression of B cell specific genes , one would expect that it would be repressed by B cell promoting factors . Analysis of ChIP-seq data from the ENCODE project demonstrated that E2A , EBF1 , and PAX5 are associated with the promoter and conserved regions of the mirn23a locus . This was confirmed by ChIP assays in this report . Furthermore , transient transfection analysis demonstrated that E2A , EBF1 , and PAX5 all could repress transcriptional activity with EBF1 having the most potent effect . Both overexpression and underexpression of EBF1 confirmed its ability to negatively regulate the endogenous mirn23a gene . Overall , this study shows that mirn23a regulates several complex pathways to drive immune cell fate decisions in the mouse . In the MPP , many critical factors are targeted by all members of the mirn23a cluster . However , sustained commitment to the lymphoid lineage appears heavily dependent on miR-24 and its direct target Trib3 . Trib3 modulates this sustained commitment by regulation of both the AKT/FoxO1 and BMP/SMAD pathways , which appear to synergistically work together to reinforce B cell commitment . The inverse regulation by mirn23a of the PI3K/Akt and BMP/ SMAD pathways is intriguing and may be also of great importance in other developmental systems . The mirn23a regulation of lymphoid and myeloid gene expression has potentially useful medical implications . MiRNAs appear to be promising therapeutic targets based on their small size , high conservation , and ability to manipulate gene expression in a physiological range[64] . Based on its role in immune cell development , mirn23a mimics and antagonists may be used to treat disorders such as cytopenias , immune deficiencies , and hematological malignancies . Understanding the role of miRNAs in hematopoiesis and their mechanism of action will be critical to manipulating miRNA expression in the treatment of human disease .
70Z/3 and A20 cell lines were obtained from ATCC ( Manassas , VA ) . 32Dcl3 cells were a gift from Allan Friedman ( Johns Hopkins ) . Unless stated otherwise , cell culture media and additives were obtained from Invitrogen ( Carlsbad , CA ) . 70Z/3 cells were grown in RPMI-1640 media ( Sigma , St . Louis , MO ) supplemented with 10% FBS , 0 . 1 mM glutamax , 10 mM HEPES , 1 mM sodium pyruvate , and 55 uM 2-Mercaptoethanol ( BME ) . A20 cells were grown in RPMI supplemented with 10% FBS and 55uM BME . 32Dcl3 cells were cultured in IMDM supplemented with 10% FBS , 10% Wehi-3B conditioned media and 55μM BME . Generation of cell lines overexpressing mirn23a miRNAs was described previously[65] . All media contained 50 U/ml penicillin and 50mg/ml streptomycin . Wildtype and mirn23a-/- C57BL/6 mice were intraperitoneally injected with 5mg 5-fluorouracil in 100 uL 1X phosphate buffered saline ( PBS ) . Bone marrow was harvested 3 days later and red blood cells were removed by ammonium chloride lysis . Nucleated blood cells were cultured in IMDM supplemented with 20% horse serum , 10% HM3 conditioned media ( as a source of GM-CSF ) , 20ng/mL human IL-6 , and 10 ng/mL murine IL-1b . Recombinant mouse cytokines obtained from R&D Systems ( Minneapolis , MN , USA ) or Invitrogen . Cells were then spin transduced with RARα403 from supernatants of GP + E86 producer lines for 2 hours at 32 degrees Celsius at 3200 RPM with 5ug/mL polybrene . Following spin transduction , cells were transferred to IMDM supplemented with 20% horse serum , 10% KSL CM ( as a source of SCF ) , 10% WEHI-3B ( as a source of IL-3 ) , and 8 U/mL human erythropoietin . Cells were then passaged every 2–3 days in IMDM supplemented with 20% horse serum , 10% COS-KSL ( as a source of SCF ) . Cells were selected for neomycin resistance by treating cells with G418 for 10 days and continuing cultures with the live cells . EML phenotype was confirmed by flow cytometry . Multipotent EML cell lines were generated from wildtype and mirn23a-/- mice as previously described[19 , 66] . These cells were differentiated to the myeloid lineage by plating 50 , 000 cells/mL in EML media ( IMDM supplemented with 20% horse serum , 10% COS-KSL ) supplemented with 10uM ATRA and 10% WEHI-3B conditioned media for 48h . Following 48h , cells were washed out of WEHI-3B conditioned media and replated with EML media supplemented with 10uM ATRA and 10% HM5 conditioned media for 5 more days . Cells were analyzed on day 7 for CD11b cell surface expression . For B cell differentiations , EML cells were plated onto 30 , 000 OP9 cells in EML media supplemented with 2ng/mL IL7 and 10ng/mL Flt3L . Cells were passaged onto fresh OP9 every 3 days and B cell differentiation was analyzed 7 days after culture . Construction of MSCV-23a , 27a , 24–2 , and mirn23a cluster retroviral plasmids were previously described[3] . MSCV-myrAKT[60] , pBABE-FOXO1dn[36] , and pRL-Smurf1[67] were obtained from Addgene . pMIEV-Smurf1 was generated by subcloning the Smurf1 cDNA from pRL-Smurf1 to the MSCV-based retroviral vector pMIEV which co-expresses GFP . Retroviruses were generated by co-transfecting 293FT cells with the retroviral plasmid along with retroviral packaging vector pCL-Eco ( Imgenex , San Diego , CA ) using Lipofectamine 2000 ( Invitrogen , Carlsbad , CA ) . Lentiviral shRNA vectors pLKO . 1 , pLKO . 1-shTrib3 , and pLKO . 1shEBF1 ( Dharmacon , Inc . , Lafayette , CO ) were co transfected into 293FT cells with pVSV-G , pMDL , and pRSV-REV plasmids ( National Gene Vector Biorepository , Indianapolis , IN ) using a Calcium Phosphate Transfection Kit according to manufacturer’s instructions ( Invitrogen , Carlsbad , CA ) . For both retroviral and lentiviral preparations , 48h and 72 h post-transfection viral supernatants were harvested and concentrated using Centricon Plus-70 filters ( Millipore , Billerica , MA ) . Cell lines were transduced by spin transduction for 2 hours at 32 degrees Celsius at 3200 RPM with 5μg/mL polybrene . Stable cell lines were generated by sorting for GFP+ cells or selecting for puromycin resistance . pBabe DN-FOXO1-HA neo was a gift from Kevin Janes ( Addgene plasmid # 45814 ) . pRK-Myc-Smurf1 was a gift from Ying Zhang ( Addgene plasmid # 13676 ) . pMSCV-flag-myr-Akt1-IRES-GFP was a gift from Kira Gritsman & Jean Zhao ( Addgene plasmid # 65063 ) . NIH/3T3 fibroblasts ( 5x105 ) were plated onto 10cm plates the night before transfection . Cells were then treated with Lipofectamine LTX ( Invitrogen , Carlsbad , CA ) transfection reagent with the addition of 10μg MigR1-GFP or Mig8-EBF1 DNA , along with 5μg of Helper II DNA . Cells were cultured in Opti-MEM media overnight and replaced with DMEM media with 10% FBS the following day . Transfection efficiency was assessed by expression of GFP under a fluorescent microscope and RNA was collected 72 hours post transfection . RNA was collected by TRIzol extraction and used in subsequent qRTPCR assays . Nucleated cells from the femur and tibia of 5–6 week old mice were lineage depleted with a MACS lineage cell separation kit according to manufacturer’s instructions ( Miltenyi Biotec , Auburn , CA ) . Lineage depleted cells were cultured onto 30 , 000 OP9 cells ( plated night before ) in IMDM supplemented with 10% defined FBS , 55mM BME , 50 U/ml penicillin , 50mg/mL streptomycin , 0 . 1mM Glutamax , 5ng/mL Flt3L , and 1 ng/mL IL-7 . Cultures were treated with DMSO control , 1 uM of BMP inhibitor LDN 193189 ( Abcam , Cambridge , MA ) , or 100nM FoxO1 inhibitor AS1842856 ( Calbiochem , Billerica , MA ) . Mouse cytokines were obtained from R&D Systems or Invitrogen . Cells were transferred onto fresh OP9 cells every 3 days . To evaluate myeloid and B-cell differentiation in cultures , cells were analyzed 6 or 9 days after culture with B220-APC and CD11b-APC/cy7 antibodies ( BioLegend , San Diego , CA ) by flow cytometry on the Beckman Coulter FC500 flow cytometer . Independently derived 70Z/3 lines infected with MSCV , or MSCV-miR-24-2 were generated previously[65] . Four MSCV lines and 2 MSCV-miR-24-2 lines were examined . Total RNA was prepared using Trizol reagent ( Invitrogen , Carlsbad , CA ) . RNA quality was evaluated using the Lab-on-a-Chip Bioanalyzer 2100 ( Agilent , Palo Alto , CA ) . Biotinylated cRNA was prepared according to the standard Affymetrix protocol using the Superscript Choice System ( Invitrogen ) and the RNA transcript labeling kit ( ENZO , Farmingdale , NY ) for cRNA preparation . Following fragmentation 10ug of cRNA was hybridized to Affymetrix Mouse 430 2 . 0 Gene Chip . Microarray analysis was performed using the Affymetrix GeneArray Scanner G2500A . The Affymetrix Expression Console Software Version 1 . 4 . 0 was used to create summarized expression values ( CHP-files ) from expression array feature intensities ( CEL-files ) . Raw data were normalized with robust multichip analysis ( RMA ) with Affymetrix transcriptome analysis console ( TAC ) software . DNA microarray and sample annotation data were deposited in GEO under the accession number GSE65874 . Total RNA was isolated from in vitro cell lines using TRIzol ( Invitrogen , Carlsbad , CA ) according to the manufacturer’s protocol . Complementary DNA ( cDNA ) was reverse transcribed from 1ug of RNA using Taqman microRNA reverse transcription kit according to manufacturer's protocol ( Applied Biosystems , Carlsbad , CA ) . Quantitative analysis was performed using gene specific Taqman ( Applied Biosystems , Foster City , CA ) or SYBR Green ( Applied Biosystems ) reagents . All experiments were performed in triplicate using BioRad CFX96 C1000 System ( BioRad , Hercules , CA ) . Relative gene expression was calculated using the ΔΔCT method . GAPDH was used to normalize expression across different RNA preparations . Relative values are presented as SEM of three independent experiments . An 888bp fragment of the murine mirn23a promoter was amplified from NIH3T3 genomic DNA with PCR using the following primers: GAGCTCTAAACGTGAGCCACCAACTG and AAGCTTGCACAGGGTCAGTTGGAAAT . Restriction sites SacI and HindIII were included in the primers . PCR fragment was cloned into pCR2 . 1 with TOPO TA Cloning Kit ( Invitrogen , Carlsbad , CA ) . The mirn23a promoter fragment was subcloned into the pGL3-basic luciferase ( Promega ) plasmid into the SacI and HindIII sites to generate pGL3-mirn23aPR . A putative upstream mirn23a enhancer was isolated by PCR using the primers: GGTACCCTGTGACTCAGCCTCATT and GGACACTTGTGGAAGCTGGA and was cloned into pCR2 . 1 . A KpnI/ SacI fragment was isolated and subcloned into pGL3-mirn23aPR to generate pGL3-mirn23aPRcns . Putative transcription factors that regulate mirn23a were identified by examining ENCODE ChIP-seq data using the Integrative Genomic Viewer ( IVG ) software version 2 . 3 . 68[68] . All transient transfections of 293T cells were carried out in 24 well plates with Lipofectamine 2000 according to manufacturer’s instructions . For E47/ SCL assays , cells were co-transfected with 100ug pGL3-CNS14mirn23apr888 , 350ug MigR1-E47 and/or 350ug pCAPP-SCL , and 5 ng pRL-tk . Total DNA content for each transfections was kept constant with 350ug of MigR1 and/or 350ug pCAPP . For EBF1 experiments , cells were co-transfected with 200ug pGL3-mirn23apr888 , 0-750ng pSport-hEBF1 , and 5ng tkRL . For each condition DNA content was kept constant with 0-750ng pcDNA3 . 1 ( CMV promoter ) . For PAX5 experiments , cells were co-transfected with 200ug pGL3-mirn23apr888 , 0-750ng MigR1-Pax5 , and 5ng tkRL . For each condition DNA content was kept constant with 0-750ng MigR1 . Each experimental condition was performed with 3 independent transfections . For all conditions , 48h post-transfection cell lysates were harvested using Promega cell lysis buffer . Firefly and renilla luciferase activity was measured using the Dual-Luciferase Assay System ( Promega ) . Firefly luciferase values were normalized to renilla luciferase values in order to adjust for potential differences in transfection efficiency . For Bach1 3’ UTR luciferase assays , a LightSwitch reporter assay ( Activ Motif , Carlsbad , CA ) containing the human 3’ UTR of Bach1 was used . 1x105 293T cells were transfected in a 24 well plate with a LightSwitch control ( no 3’ UTR ) or experimental ( with Bach1 3’ UTR ) reporter , along with an empty vector miRNA control ( MSCV ) , MSCV-miR23a27a24 , or MSCV-miR23a . Cells were transfected using Lipofectamine 2000 according to manufacturer’s protocol . 48 hours after transfection , luciferase activity was determined using the MISSION LightSwitch luciferase assay reagent ( Sigma Aldrich ) kit according to manufacturer’s protocol . Whole-cell extracts were prepared by lysing cells in RIPA buffer ( 50 mM Tris pH 7 . 5 , 150 mM NaCl , 1% NP-40 , 0 . 5% EDTA , 0 . 1% SDS ) and protease inhibitor cocktail ( Roche ) . Protein concentration was determined using the Bicinchoninic Acid ( BCA ) protein assay ( Pierce , Rockford , IL ) . 75 μg of whole cell lysates was separated by SDS-PAGE and transferred to nitrocellulose membrane . Membranes were blotted with the following antibodies: Ikzf , FoxO1 , Smurf1 , Smad1 , Satb1 , β-actin ( Cell Signaling , Danvers , MA ) , Ebf1 , and Runx1 ( Santa Cruz , Dallas , TX ) . Horseradish peroxidase conjugated secondary antibodies were purchased from GE Healthcare UK Ltd ( Buckinghamshire , England ) . Clarity Western ECL substrate enabled detection of antibodies ( Thermo Scientific , Rockford , IL ) . Analysis was performed using BIORAD Chemidoc XRS+ System using Imager Lab Software ( Hercules , CA ) or ImageJ ( NIH , Bethesda , MD ) . ChIP assays were performed using Active Motif ( Carlsbad , CA ) ChIP-IT Express kit according to manufacturer’s protocol . A20 cells were grown in 15cm plates to ~80–90% confluency before crosslinking cells . Chromatin was sheared by sonication ( 10 pulses , 25% Amp , 20 seconds/pulse , with 30 seconds rest between pulses ) . IPs were done using antibodies against IgG ( p120-101-Bethyl laboratories ) , E47 ( clone G127-32 –BD Biosciences ) , EBF1 ( Clone C-8 , Santa Cruz Biotechnology ) , Pax5 ( clone A-11 , Santa Cruz Biotechnology ) , and Ikzf1 ( clone H-100 , Santa Cruz Biotechnology ) . Following crosslink reversal , relative DNA was analyzed by qRT-PCR . Primers ( Probe: 5’-/56-FAM/CATTTGGCC/ZEN/TGCTTTGGGCTCAG/3lABkFQ/-3’ primer 1: 5’-CCTCCCTCAGCTTCCTCT-3’ primer 2: 5’-GCTTCCCACTCTGCTTCTATC-3’ ) were obtained from IDT ( Coralville , Iowa ) . CD45 . 1 recipient mice were irradiated by X-ray irradiation ( 1000 gy ) the night before transplant . MPPs ( LSK CD34+ ) were collected from CD45 . 2 wildtype and mirn23a-/- mice through FACS sorting . 2000 MPP cells were retro-orbitally injected along with 2x105 CD45 . 1 support marrow cells into the lethally irradiated CD45 . 1 recipient mice . Mice were sacrificed 4 weeks after transplant and analyzed by flow cytometry for contribution to the B cell ( B220 ) and myeloid ( CD11b ) lineages . Mice were injected with 2ug of BrdU ( BD Biosciences , Billerica , MA ) diluted in 200uL of sterile phosphate buffered saline by intraperitoneal injection 16 hours prior to analysis . Bone marrow cells were harvested from the femurs and tibias of these mice and cell fixation and permeabilization was done according to BD Pharmingen BrdU flow kit protocol ( BD Biosciences , Billerica , MA ) . BrdU incorporation in HSPCs was evaluated by flow cytometry using a panel including Sca1 ( D7 ) -FITC , Lineage cocktail- Biotin , Avidin-Texas Red , BrdU-APC , and c-Kit ( 2B8 ) -APC/Cy7 . Primary bone marrow cells were harvested from the femurs and tibias of mice 5–6 weeks of age . Staining for annexin V and 7AAD was done according to the FITC Annexin V apoptosis detection kit with 7-AAD ( Biolegend , San Diego , CA ) . Analysis by flow cytometry was done using a combination of antibodies including Annexin V- FITC , Sca1- PE , Lineage cocktail-biotin , TR-Avidin , 7AAD , and c-Kit-APC/Cy7 . Statistical data are presented as the mean +/- standard error of the mean ( SEM ) . Differences between sample groups were determined by performing an unpaired student t-test . Analysis was performed using PRISM software version 6 . 0 ( Graphpad software ) . The use of mice in these experiments was approved by the Indiana University School of Medicine and University of Notre Dame Institutional Animal Care and Use Committees ( Protocols 13–017 and 16–022 ) . Mice used for this study were euthanized using a commercial Euthanex CO2 machine , ensuring ethical euthanasia of all mice . | MicroRNAs ( miRNAs ) are small ~22 nucleotide long RNA molecules that are involved in regulating multiple cellular processes through inhibiting the expression of target proteins . We previously identified a gene ( mirn23a ) that codes for 3 miRNAs that control the development of immune cells in the bone marrow . The miRNAs promote the development of innate immune cells , macrophages and granulocytes , while repressing the development of B cells . Here we show that mirn23a miRNAs negatively affect the expression of multiple proteins that are involved in directing blood progenitor cells to become B cells . Additionally , we observed that modulation of FoxO1 and Smad proteins , downstream effectors of two signaling pathways ( PI3 kinase/ Akt and BMP/ Smad ) , is critical to direct immune cell development . This is the first observation that these pathways are potentially coregulated during the commitment of blood progenitors to mature cells of the immune system . Consistent with mirn23a being a critical gene for committing progenitors to innate immune cells at the expense of B cells , we observed that a critical B cell protein represses the expression of mirn23a . In conclusion , we demonstrate the mirn23a regulation of blood development is due to a complex regulation of both transcription factors and signaling pathways . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"gene",
"regulation",
"regulatory",
"proteins",
"immunology",
"dna-binding",
"proteins",
"cell",
"differentiation",
"bone",
"marrow",
"cells",
"developmental",
"biology",
"micrornas",
"transcription",
"factors",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"hyperexpression",
"techniques",
"gene",
"expression",
"molecular",
"biology",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"gene",
"expression",
"and",
"vector",
"techniques",
"antibody-producing",
"cells",
"biochemistry",
"rna",
"cell",
"biology",
"b",
"cells",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"non-coding",
"rna"
] | 2017 | The miR-23a~27a~24-2 microRNA cluster buffers transcription and signaling pathways during hematopoiesis |
Tracking and isolating live cells based on their proliferative history in live animals remains a technical challenge in animal studies . We have designed a genetic marking system for tracking the proliferative frequency and history of lymphocytes during their development and homeostatic maintenance . This system is based on activation of a fluorescent marker after Cre-dependent recombination between sister chromatids at a specially designed tandem loxP site , named Tlox . We have demonstrated the utility of the Tlox system in tracking proliferative windows of B and T lymphocyte development . We have further applied the Tlox system in the analysis of the proliferative behavior and homeostatic maintenance of Vγ1 . 1 positive γδ T cells . Our data show that Vγ1 . 1 T cells generated in neonatal but not adult life are able to expand in the thymus . The expanded Vγ1 . 1 T cells are preferentially maintained in the liver but not in lymphoid organs . It has been shown that numbers of Vγ1 . 1 T cells were dramatically increased in the lymphoid organs of Id3 deficient mice . By combining BrdU and Tlox assays we show that this phenotype is primarily due to enhanced neonatal expansion and subsequent retention of Vγ1 . 1 T cells . Thus , the Tlox system provides a new genetic tool to track clonal expansion within a defined cell population or tissue type in live animals .
Cell proliferation is a tightly regulated process in tissue development and maintenance of tissue functions . Knowing the frequency and history of cell division is not only important in the study of normal tissue development but also in the investigation of tissue regeneration and tumorogenesis . The most commonly used lineage tracking methods are based on Cre mediated activation of reporters in progenitor cells [1]–[3] . By restricting Cre activity to the progenitor cells , this method is highly effective in tracking clonal expansion of the labeled progenitors [3] . However , reporter activation is not linked to cell cycle and thus alone cannot be used to report the proliferative status of the progenitor population . Thus far , methods available for tracking cell proliferation in live animals are still limited and incompatible with recovery of live cells for subsequent analysis . The most commonly used methods for tracking cell proliferation are based on either incorporation of a nucleotide analog such as bromodeoxyurodine ( BrdU ) or tritiated thymidine during DNA replication [4] or natural dilution of a genetically activated protein marker such as GFP [5] . For example , BrdU pulse labeling method has been successfully used to define the proliferative windows in thymic T cell development [6] . The BrdU method is simple and generally applicable to all tissues , but the detection of chemical labels is incompatible with retrieval of live cells for further studies . In contrast , detection of GFP expression can be carried out with live cells and therefore has the potential to be combined with additional functional assays . For example , GFP intensity has been used to trace the age of naïve T cells based on the dilution of temporarily activated GFP signals associated with homeostatic proliferation of peripheral T cells [7] . However , for cells that have undergone extensive rounds of proliferation , like lymphocytes during antigen responses or cancer stem cells coming out of a dormant phase , they will lose all GFP signals and become indistinguishable from unlabeled cells in the background . Cre-mediated mitotic recombination provides another way to permit genetic marking of proliferation events in live mice [8] . In this case , activation of a reporter is strictly dependent on mitotic recombination between homologous chromosomes . This system has been successfully used in labeling and tracking progenitor cells that give rise to tumors [9] . However , the overall recombination efficiency of Cre-mediated mitotic recombination is below 1% of the proliferating population when tested in a broad range of cell types [8] including lymphocytes [10] . While the method is powerful in mosaic analysis , the low frequency of mitotic recombination makes the system less effective as a generic method to evaluate proliferative status of progenitor populations for most tissue types . To overcome these limitations and to enable tracking cell cycle in live cells , we have designed a sister chromatid recombination system to directly link cell cycle with permanent activation of a fluorescent protein marker . This system is based on the fact that Cre/lox mediated recombination can occur between sister chromatids during cell cycle [11] , [12] . Cre-mediated sister chromatid exchange occurs at a much higher frequency than Cre-mediated mitotic recombination between homologous chromosomes [12] , [13] . In our design , a non-equal exchange between the marked sister chromatids produces a fraction of progeny that acquire the fluorescent marker . We have tested the system in both cultured fibroblasts and developing lymphocytes in live mice . The mouse lymphoid system represents one of the best experimental models for understanding normal and abnormal cell proliferation in a living organism . Using mice expressing lymphoid specific Cre , we have shown that permanent activation of the fluorescent marker after Cre-mediated recombination is correlated with the well-defined windows of cell cycle . As a proof of principle , we further applied this newly established cell tracking system in the study of the expansion of γδ T cells induced by deletion of the Id3 gene .
Cre recombinase has been shown to be capable of driving recombination between sister chromatids via duplicated loxP sites [11] , [12] . This recombination system provides an opportunity for Cre mediated activation of a genetic marker during cell cycle . To ensure that Cre/lox mediated recombination occurs exclusively during cell cycle , we have designed a tandem overlapping loxP cassette named Tlox , which contains a stop codon at the beginning of the second loxP unit ( Fig . 1A ) . This Tlox cassette provides a translational stop between the GFP and tdTomato markers ( Fig . 1B ) . The stop codon is eliminated only after the Tlox has been reduced to a single loxP unit . Because recombination requires a Cre tetramer complex [14] , intra Tlox recombination cannot occur due to structural constrain . However , Cre-mediated Tlox recombination can occur between two Tlox units present on neighboring sister chromatids during the S/G2/M phase of cell cycle . The recombination will result in either equal or unequal exchange between the sister chromatids ( Fig . 1C ) . While tdTomato remains silent following equal exchange , unequal exchange between sister chromatids should lead to daughter cells inheriting either a single loxP or triple loxP sequences . Whereas cells inheriting a triple loxP remain tdTomato negative , cells inheriting a single loxP should permanently turn on the tdTomato marker . It is anticipated that triple loxP sequences will have additional chances to be converted to a single loxP in subsequent cell cycles through unequal exchange . In principle , this method will label up to one quarter of the progeny from each proliferating cycle if Cre is fully active during S/G2/M phases of each cell cycle . Because the expression of tdTomato is strictly dependent on cell cycle and is permanent , this method offers a new way to identify and isolate cells that have undergone proliferation in the developmental window defined by Cre expression . We first performed a proof-of-principle test for this genetic design by introducing the Tlox expressing cassette via a retroviral vector into 3T3 cells . As expected , only the GFP signal is detectable in Tlox infected cells in the absence of Cre ( Fig . 1D , E ) . To test the efficiency of sister chromatid recombination , we isolated cells stably expressing GFP and transduced these cells again with a Cre-expressing retroviral vector . Efficiency of transduction was monitored with a hCD2 marker co-expressed with Cre on the retroviral vector [12] . Without cell cycle synchronization , the doubling time for NIH3T3 cells has been reported to be approximately 17 hours [15] . We found 32 . 5+/−0 . 5% of Cre transfected cells expressed tdTomato 48 hours post viral transduction , which was up from 24+/−0 . 7% scored at 24 hours post transduction ( Fig . 1F ) . This number is below the predicted maximal possible frequency ( 25% for the first cycle and 44% for the second cycle ) of tdTomato activation per cell cycle ( Fig . 1C ) . Thus , the labeling frequency observed in real experiments is an underestimation of true proliferation frequency . PCR analysis of the Tlox cassette showed that Cre expression indeed induced generation of both single and triple loxP sequences ( Fig . 1G , lane 2 ) . FACS sorted tdTomato positive cells showed exclusively single loxP in the same PCR assay ( Fig . 1G , lane 4 ) , supporting the idea that all tdTomato positive cells were the result of Tlox conversion to single loxP . To introduce a single Tlox cassette into the mouse genome , we used the piggyBac transposon vector [16] to deliver the expression cassette through microinjection of fertilized eggs . FACS analysis of blood samples from founder transgenic mice identified two positive founders ( Fig . S1A ) . We mapped four independent insertion sites through inverse PCR cloning and chose a chromosome 19 insertion ( Fig . S1B and C ) for subsequent tests . The timing and frequency of cell proliferation during lymphocyte development are dynamically regulated during the generation and maintenance of polyclonal lymphocytes . Because the proliferative windows in lymphocyte development have been mapped previously with BrdU labeling methods , we chose lymphocyte development as a test model to validate our Tlox system . We used mb1Cre [17] and LckCre [18] to drive B cell and T cell specific recombination , respectively . mb1Cre initiates Cre expression at the pro-B cell stage and keeps Cre on throughout B cell development . tdTomato expression was not detected in splenic B and T lymphocytes in the absence of mb1Cre , confirming that spontaneous exchanges between sister chromatids are negligible ( Fig . 2A , middle column ) . tdTomato positive cells were exclusively found among B but not T cells in mice carrying both the Tlox transgene and mb1Cre . Activation of tdTomato was coupled with a further upregulation of GFP signals ( Fig . S2 ) , which could be a result of increased stability of mRNA after removal of the stop codon in the Tlox cassette and thus eliminating nonsense mediated RNA decay [19] . To further evaluate whether tdTomato expression is linked to cell proliferation , we examined B cell development in the bone marrow . B cell development proceeds in a sequential order from pro-B to pre-B , and then to mature B cells , which are marked as CD43hiB220low , CD43medB220low , and CD43lowB220hi fractions , respectively [20] ( Fig . 2B ) . We found that tdTomato was expressed in all three fractions with the ratio gradually increasing from 30+/−1 . 9% in pro-B , to 38+/−9 . 0% in pre-B , and to 43+/−5 . 3% in mature B cells ( Fig . 2B and Fig . S3 ) . The increased frequency of tdTomato expression is correlated with the well-defined pro-B and pre-B windows of cell proliferation [20] . tdTomato expression frequency in mature B was maintained at around 46+/−7 . 3% in the spleen . The lack of further increase of tdTomato expression in mature B cells from bone marrow to periphery is consistent with their non-cycling state in the absence of antigen stimulation . This result further confirms that Cre cannot drive Tlox recombination in non-cycling cells . T cell development in the thymus proceeds through several developmental stages with clearly defined windows of proliferation [21] . T lineage commitment and initial expansion of the committed T cell progenitors occur at the DN1 and DN2 stages , respectively . Separation between αβ and γδ T lineage fate is completed at the end of the DN3 stage . Cells choosing the αβ fate expand further at the DN4 stage before entering the non-proliferating DP stage . Clonal selection at the DP stage leads to the formation of either CD4 helper T cells or CD8 cytolytic T cells with each cell expressing a unique TCR . The LckCre transgene has been shown to initiate Cre expression during the DN3 stage of T cell development in the thymus and to support continuing Cre expression thereafter [18] , [22] ( Fig . 2C ) . BrdU based analysis has demonstrated that proliferation occurs almost exclusively in the DN2 and DN4 stage of T cell development , although the mature CD8+ single positive ( SP ) T cells have also exhibited a detectable level of proliferation [6] , [23] . Mice double positive for the LckCre and the Tlox transgenes were analyzed for tdTomato expression among different cell fractions representing progressive maturation status ( Fig . 2C ) . We found that the earliest stage to detect tdTomato expression was DN4 but not DN3 , which is consistent with the idea that LckCre cannot act on the non-proliferating DN3 cells . CD4CD8 double positive ( DP ) cells are the progeny of rapidly proliferating DN4 cells and have greatly reduced cell size comparing with other immature T cell fractions . Because the overall signal of GFP and tdTomato was dramatically reduced , it was difficult to separate tdTomato positive cells from the negative fractions even though tdTomato expression was clearly visible among DP cells . After progressing to the CD4 or CD8 single positive ( SP ) stage , cells showed a significantly higher frequency of tdTomato expression than DN4 cells . During the DP to SP transition , immature T cells remain in a non-proliferating state . Therefore , this elevated frequency of tdTomato+ SP cells reflected the proliferative transition between DN4 and DP cells . Within the SP populations , the percentage of tdTomato expression in CD8+ SP cells is approximately two times higher than that of CD4+ SP cells , suggesting additional rounds of cell proliferation have occurred in CD8 lineage cells . This result is consistent with BrdU pulse labeling studies , which also showed higher proliferation rate associated with CD8 SP cells [23] . The frequency of tdTomato expression in both CD4+ and CD8+ cells showed a moderate increase as they move from the thymus to the spleen . This finding is consistent with the knowledge that naïve T cells rarely proliferate under homeostatic conditions . Overall , the timing and capacity of immature T cell proliferation measured by tdTomato expression from the Tlox cassette are in agreement with the previously defined characteristics of various thymocyte populations during their development . To gain additional evidence that tdTomato activation is dependent on cell cycle , we compared LckCre-induced with CD4Cre-induced Tlox recombination . CD4Cre is known to support Cre expression starting between the DN3 and DN4 stage [24] , which is slightly later than that of LckCre . Correspondingly , the efficiency of tdTomato activation was found lower than that of LckCre at the DN4 stage ( Fig . 2C bottom panel ) . The difference between these two Cre drivers persists even after cells have reached to the CD4 or CD8 stage . This persistent difference cannot be explained by inefficient Cre expression in these later stages since both LckCre and CD4Cre are equally capable of deleting the flox-stop cassette embedded in the R26ZsGreen reporter [25] ( Fig . 2D ) , a process independent of cell cycle . Thus , activation of the Tlox reporter requires Cre to be expressed during the proliferative phase of T cell development . To obtain direct evidence that tdTomato activation is dependent on cell cycle , we performed in vitro cell proliferation assay with sorted tdTomato positive or negative CD4 T cells from the Tlox;CD4Cre mice ( Fig . S4 ) . TCR-induced cell proliferation was tracked by dilution of the CellTrace Violet dye . tdTomato positive cells retained tdTomato expression during cell cycle ( Fig . S4A ) . tdTomato negative cells showed increased frequency of tdTomato expression after each round of cell cycle ( Fig . S4A and B ) . In fact , tdTomato expression can be detected as early as when cells entering the blasting phase ( Fig . S4C ) . During the same time frame , tdTomato expression was not observed among tdTomato negative cells when they were kept alive without proliferation ( Fig . S4B ) . This study further demonstrates that activation of the Tlox reporter is strictly associated with cycling cells . We next used the Tlox reporter system to monitor the generation and maintenance of γδ T cells in neonatal and young adult mice . Although majority of γδ T cells found in adult tissues are descendants of fetal derived γδ T cells , γδ T cells are continuously generated in parallel to αβ T cells in postnatal thymus and exported to secondary lymphoid organs [26] . Little is known about the homeostatic maintenance of these post-natal derived γδ T cells . A fraction of γδ T cells enriched for the Vγ1 . 1 usage ( thus referred as Vγ1 . 1 cells ) has been classified as innate-like or NK-like γδ T cells due to their ability to express mixed cytokines including IL4 and γ interferon [27] . These cells are produced in late fetal and neonatal life and detected in the thymus , secondary lymphoid organs , and liver . However , it is not clear whether these cells are continuously generated in postnatal life or maintained exclusively through self-renewal of the fetal-derived population . We first used BrdU pulse labeling method to assess the cell cycle status of γδ T cells in one-week old neonates and five-week old young adult mice ( Fig . 3A&B ) . A similar frequency of BrdU labeling was observed in Vγ1 . 1 T cells of both age groups . To further evaluate the lineage relationship between the two age groups we used the Tlox assay to evaluate proliferative history of the γδ T cells . The frequency of tdTomato positive Vγ1 . 1 cells was found significantly higher in one-week old than in 5-week old thymus . While this result is consistent with the earlier studies that Vγ1 . 1 cells undergo expansion in late fetal and neonatal life [27] , the decreased frequency also suggests that Vγ1 . 1 cells present in young adult thymus are not related to neonatal derived Vγ1 . 1 T cells . Further analysis of 5-week old mice revealed that Vγ1 . 1 T cells from other lymphoid organs including lymph nodes and the spleen exhibited a similar low frequency of dtTomato as in the thymus ( Fig . 3C&D ) . In contrast , the frequency of dtTomato positive Vγ1 . 1 T cells present in the liver is as high as in neonatal thymus . Thus , circulating Vγ1 . 1 T cells and liver resident Vγ1 . 1 T cells can be separated into two distinct populations based on their proliferative history . Recently , several studies have shown that deletion of the Id3 gene led to a significant expansion of Vγ1 . 1 T cells in adult animals [28] , [29] . The Id3 gene encodes a nuclear protein , which regulates lymphocyte development through direct inhibition of E-protein transcription factors [30] . How Id3 knockout promotes development and/or expansion of Vγ1 . 1 T cells is still not clear . We thought to further investigate this issue by combining the traditional BrdU method with the newly established Tlox system . LckCre was used to induce T lineage specific deletion of Id3 and activation of the Tlox marker . We first analyzed cell cycle status with BrdU pulse labeling . A significant increase in BrdU positive cells was observed in Id3 deficient one-week old neonates in comparison with the wild type controls ( Fig . 4A , middle column ) . However , analysis of young adult mice showed a moderate decrease in BrdU incorporation in Id3 deficient mice ( Fig . 4B , middle column ) . We then used the Tlox system to track the proliferative history of Vγ1 . 1 cells . Analysis of Vγ1 . 1 T cells from one-week old neonates revealed a similar frequency of tdTomato expression between Id3 deficient and wild type controls ( Fig . 4A , right column ) , suggesting that Vγ1 . 1 T cells on both backgrounds have gone through similar numbers of cell cycles at the neonatal stage . This result is in contrast to the BrdU data , which detects a higher percentage of proliferating cells in Id3 deficient Vγ1 . 1 T cells than in wild type controls within the 4 hour window of pulse labeling . Thus , the proliferation rate revealed by BrdU pulse labeling may not reflect the proliferative history of the cell population . The differential outcomes from these two assays become even more dramatic in the analysis of 5-week old mice . In contrast to the neonates , the frequency of tdTomato positive Vγ1 . 1 T cells in 5-week old mice was significantly increased in Id3 deficient mice even though BrdU labeling frequency has decreased ( Fig . 4B , middle and right panels ) . This result supports the idea that Id3 deficiency promotes development and expansion of Vγ1 . 1 T cells during neonatal life and their subsequent maintenance in postnatal life .
Our study demonstrated that the Tlox design is an effective genetic tool to track proliferative history of Cre expressing cells in both tissue culture and live animals . This method predicts that the maximal labeling efficiency of dividing cells is 25% per cell cycle ( Fig . 1C ) . While our tracking data clearly shows that Tlox activation is correlated with increased numbers of cell cycle , the overall labeling efficiency is below the expected rate , particularly in animal models . We speculate that multiple factors may contribute to the efficiency of Cre mediated Tlox recombination between sister chromatids . These factors , such as the accessibility of Tlox during S/G2 phase , the expression level of Cre recombinase , and the duration of the S/G2 phase , may vary during development and between tissue types . Thus , the quantitative readout from Tlox assay represents an empirical value associated with the specified developmental system . Once the relative frequency is determined for relevant tissue types in the wild type mice , this reporter assay is particularly useful in assessment of proliferative behavior associated with novel mutations . When combined with other methods such as BrdU incorporation assays , this system can effectively reveal cell cycle behaviors , some of which would have been otherwise missed or misinterpreted by using the BrdU method alone . Our analysis of developing B cells and αβ T cells confirmed that the frequency of tdTomato expression from the Tlox marker is correlated with Cre expression and windows of cell cycles defined previously by the BrdU method [20] , [23] , [31] . Using this newly established Tlox assay , we further revealed a change in proliferative behavior of γδ T cells between neonatal and young adult mice . In particular , it has been shown that generation of Vγ1 . 1 T cells from donor hematopoietic stem cells requires neonatal thymic environment [27] . This observation led to the general hypothesis that Vγ1 . 1 T cells are produced in late fetal and neonatal life and maintained through self-renewal in postnatal life . Our analysis with the Tlox marker provided strong evidence indicating that most thymic resident Vγ1 . 1 T cells are not maintained through expansion of preexisting population . They are most likely continuously derived from thymic precursors and quickly turned over in circulation . In contrast to circulating Vγ1 . 1 T cells , liver resident Vγ1 . 1 T cells are maintained as a distinct population , which share similar features with Vγ1 . 1 T cells found in neonatal thymus . Our finding is consistent with the report that liver resident Vγ1 . 1 T cells lack N addition in their TCR , a feature associated with fetal derived T cells [27] . It remains to be determined whether circulating Vγ1 . 1 T cells in adult animals can be converted to tissue resident γδ T cells under certain circumstances such as in response to infection or tissue damage . Tlox mediate tracking of expanded populations may assist future investigation of function and homeostatic maintenance of Vγ1 . 1 T and other lymphoid populations . Id3 deficient mice have been characterized to exhibit excess amount of Vγ1 . 1 T cells in the thymus and peripheral lymphoid organs [28] , [29] and develop high incidence of γδ T cell lymphoma at older age [32] . Our study using both BrdU labeling and Tlox tracking methods revealed that Id3 deletion promotes the development and proliferative expansion of Vγ1 . 1 T cells in neonatal life . At adult age , this population is apparently maintained by accumulation of cells with a reduced frequency of cell cycle . This result provides a strong evidence to support the finding that Vγ1 . 1 T cells detected in Id3 deficient mice exhibit highly restricted TCR usage and often lack N nucleotide addition [29] . Our data support the idea that Id3 deficiency promotes clonal expansion of Vγ1 . 1 T cells in the neonatal thymus and , more importantly , their slow expansion and long-term maintenance in adult life . Such a proliferative behavior could contribute to the generation of γδ T cell lymphoma observed in aged Id3 knockout mice [32] . Our study of Id3 deficient mice established the Tlox system as a new tool for tracking clonal expansion and possibly for monitoring malignant transformation in live animals . A major advantage of our Tlox system is the identification and isolation of live cells that have undergone proliferation in a defined window of development . However , additional efforts are still needed for broad applications of this reporter system . Preliminary studies suggest that the Tlox system is inefficiently activated by tamoxifen inducible CreER systems , although the reasons for this and possible steps to optimize efficiency are still under investigation . In addition , the Tlox design can be further adapted to drive expression of other markers or enzymes for easy detection of proliferating cells in tissues other than lymphocytes . It would be particularly attractive to use this method to label and then isolate slowly proliferating somatic stem cells or tumor initiating clones when combined with appropriate Cre drivers . Finally , this recombination system could be combined with live imaging techniques in tracking cell proliferation in situ in mice and other model organisms .
The mb1Cre knockin was generated in Dr . Reth's group [17] . The LckCre transgenic [18] and Id3 conditional knockout [33] alleles were produced in Zhuang lab as previously described . R26ZsGreen strain [25] was purchased from the Jackson Labs . Tlox transgenic lines were produced by microinjection of circular PB donor construct mixed with a helper plasmid ( PCX-PBase ) at a ratio of 3∶1 [16] . The transgenic procedure was performed by the Duke Transgenic Mouse Facility . Animals were bred and maintained in the SPF facility managed by Duke University Division of Laboratory Animal Research . All animal procedures were approved by the Duke University Institutional Animal Care and Use Committee . The antibodies used were as follows: PE/Cy7 anti-human CD2 ( TS1/8 ) , APC/Cy7 anti-mouse TCRβ ( H57-597 ) , APC/Cy7 anti-mouse CD4 ( GK1 . 5 ) , PE/Cy7 anti-mouse CD8a ( 53–6 . 7 ) , APC anti-mouse B220 ( RA2-6B2 ) , PE/Cy7 anti-mouse CD43 Activation-associated Glycoform ( 1B11 ) , PE/Cy7 anti-mouse/human CD44 ( IM7 ) , APC anti-mouse CD25 ( 3C7 ) , PE/Cy5 anti-mouse NK-1 . 1 ( PK136 ) , PE/Cy5 anti-mouse Ly-6G/Ly-6C ( Gr-1 ) ( RB6-8C5 ) , PE/Cy5 anti-mouse CD11b ( M1/70 ) , APC anti-mouse TCRδ ( GL3 ) , PE/Cy5 anti-mouse TCRδ ( GL3 ) , APC anti-mouse TCRVγ1 . 1/Cr4 ( 2 . 11 ) , PE anti-mouse TCRVγ1 . 1/Cr4 ( 2 . 11 ) , APC anti-mouse TCRβ ( H57-597 ) were purchased from Biolegend . The APC BrdU Flow Kit was from BD Biosciences . The PGK promoter was cloned into an MSCV retrovirus backbone . The GFP coding sequence together with a 2A-Tandom loxP was amplified containing multiple cloning sites ( Cla I-target cassette-BamH I-Nsi I-Cla I ) . The fragment of GFP-2A-Tandem loxP was cloned downstream of a PGK promoter using Cla I . The coding sequence of tdTomato was amplified and cloned downstream of the Tlox site using the BamH I-Nsi I linker . For transgenic experiment , the entire GFP-2A-Tlox-tdTomato cassette was subcloned downstream of the Actin promoter present in the piggyBac vector [16] . Bosc cells were cultured in complete DMEM medium ( containing 10% fetal bovine serum ) at 60–70% confluence on 10 cm plates . 10 µg targeting plasmid and 2 µg helper plasmid ( pCL-Eco ) were transfected into Bosc cells using CaCl2 . 48 hours later , the supernatant was harvested and filtered with a 0 . 45 µm-sterile syringe filter . 3T3 cells were infected with virus containing 8 µg/ml polybrene . The medium was changed back to complete DMEM medium 24 hours after infection . GFP+ cells were sorted after 2 weeks of subculture . Stable GFP+ cells were further infected with either a control virus or Cre-ires-hCD2 virus . FACS analysis and fluorescence imaging were performed at the indicated times . Single-cell suspensions were prepared from thymus , spleen , peripheral lymph nodes and bone marrow , and suspended in cold FACS buffer ( 1×PBS supplemented with 5% bovine calf serum ) . 1×106 cells were stained with Abs in the dark at 4°C for 30 min . After washing with cold FACS buffer , cell suspensions were analyzed on a FACSCanto II flow cytometer ( BD Biosciences ) . Flowjo software ( Tree Star ) was used for data analysis . Mice used for BrdU assays were injected i . p . with 1 mg BrdU 4 hours prior to sample collection . BrdU staining was performed according to the manufacture's instruction ( BD Biosciences ) . | Identification and isolation of live cells based on their proliferative history remains a technical challenge in genetic analysis of animal models . We have designed a novel genetic tool for tracking dividing cells in live animals . The experimental system is based on a fluorescent reporter , whose expression requires both the activity of Cre recombinase and genome replication . We have successfully tested the reporter system in developing lymphocytes and revealed a unique phenomenon of population expansion involving the innate γδ T lymphocytes generated in neonatal life . The experimental system is adaptable to the analysis of any tissue types when combined with appropriate Cre drivers . It provides a new tool for tracking clonal expansion associated with tissue regeneration or neoplastic growth during the normal life span of animals . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2013 | Tracking Proliferative History in Lymphocyte Development with Cre-Mediated Sister Chromatid Recombination |
Human malaria parasites proliferate in different erythroid cell types during infection . Whilst Plasmodium vivax exhibits a strong preference for immature reticulocytes , the more pathogenic P . falciparum primarily infects mature erythrocytes . In order to assess if these two cell types offer different growth conditions and relate them to parasite preference , we compared the metabolomes of human and rodent reticulocytes with those of their mature erythrocyte counterparts . Reticulocytes were found to have a more complex , enriched metabolic profile than mature erythrocytes and a higher level of metabolic overlap between reticulocyte resident parasite stages and their host cell . This redundancy was assessed by generating a panel of mutants of the rodent malaria parasite P . berghei with defects in intermediary carbon metabolism ( ICM ) and pyrimidine biosynthesis known to be important for P . falciparum growth and survival in vitro in mature erythrocytes . P . berghei ICM mutants ( pbpepc- , phosphoenolpyruvate carboxylase and pbmdh- , malate dehydrogenase ) multiplied in reticulocytes and committed to sexual development like wild type parasites . However , P . berghei pyrimidine biosynthesis mutants ( pboprt- , orotate phosphoribosyltransferase and pbompdc- , orotidine 5′-monophosphate decarboxylase ) were restricted to growth in the youngest forms of reticulocytes and had a severe slow growth phenotype in part resulting from reduced merozoite production . The pbpepc- , pboprt- and pbompdc- mutants retained virulence in mice implying that malaria parasites can partially salvage pyrimidines but failed to complete differentiation to various stages in mosquitoes . These findings suggest that species-specific differences in Plasmodium host cell tropism result in marked differences in the necessity for parasite intrinsic metabolism . These data have implications for drug design when targeting mature erythrocyte or reticulocyte resident parasites .
The malaria-causing apicomplexan parasites Plasmodium spp . have a dynamic life cycle which is reflected in stage-specific morphologies , transcriptomes , proteomes and metabolomes [1–8] . These changes , particularly in their metabolome , reflect the nutritional needs and biological processes of the parasite during intracellular development that in turn influences , or is influenced by , the physiological state of the host cell [6] . Perhaps due to their parasitic life-style , Plasmodium spp . have a simplified and reduced metabolic capacity when compared to higher non-parasitic organisms . They are auxotrophic for purines , vitamins and many amino acids [9 , 10] , but have retained core pathways of carbon metabolism such as glycolysis [11] , the citric acid cycle [7 , 12] , lipid synthesis [13 , 14] , the pentose phosphate pathway [15] , pyrimidine biosynthesis [16] and glycosylation [17] . Plasmodium spp . are obligate intracellular parasites and their metabolism is interlinked with that of their host cell and is heavily dependent on the availability of external nutrients . As a result , intracellular Plasmodium establish systems such as the new permeation pathways with the purpose of accessing host cell and environmental nutrients [18]; in fact the parasite genome encodes >120 predicted membrane transport proteins , a subset of which are located on the plasma membrane [19] . Erythrocyte invasion is a prerequisite for establishment of infection by Plasmodium merozoites and the roles of different merozoite and host surface proteins in this invasion process have been intensively studied [20–25] . Multiple partially overlapping erythrocyte invasion pathways have been described in P . falciparum with consequent functional redundancy [26] . Many Plasmodium spp . including P . falciparum preferentially invade reticulocytes [27] which is also capable of invading and replicating within all stages of erythrocyte development including mature cells . However , P . vivax has a strict requirement for growth in reticulocytes , expresses reticulocyte binding proteins [28] and requires a host Duffy blood group glycoprotein for invasion [29] . P . vivax infection causes accelerated remodelling of very young reticulocytes , a process that normally takes 24 hours in uninfected reticulocytes [30] . The rodent model malaria parasite , P . berghei is also 150 times more likely to invade reticulocytes and establish infection in the presence of equal numbers of mature erythrocytes and reticulocytes [31] and has therefore been long thought of as a suitable model for P . vivax blood stage biology [32] . Mature erythrocytes , comprising almost 98% of the circulating red blood cells , can be considered “simplified” cells; they are metabolically active but lack intracellular organelles found in the bone marrow erythroid precursors cells [33] and enucleated reticulocytes ( maturing erythrocytes ) that are present in peripheral circulation [34] . Reticulocytes undergo many changes after their release into the peripheral circulation as they mature and this is associated with a 20% decrease in total surface area and acquisition of a biconcave shape with consequent increase in shear membrane resistance , the progressive loss of organelles ( mitochondria , ribosomes and lysosomes ) , the loss or reduced abundance of up to 30 membrane proteins , and decreased levels of membrane cholesterol [34 , 35] . This maturation process is associated with a general streamlining of cellular metabolism; mature erythrocytes are highly dependent on glycolysis [36] and the pentose phosphate pathway [37] for both energy and redox balance and lack many other pathways of carbon metabolism , such as citric acid cycle [38] . Reticulocytes are thus expected to contain a richer repertoire of carbon sources and other essential nutrients than mature erythrocytes which might be exploited or even required by reticulocyte preferent Plasmodium spp . Limited comparative metabolomics of the erythroid lineage has been attempted before but focussed on sickle cell disease and cord blood reticulocyte physiology [39 , 40] . Therefore , in order to establish whether there are metabolic differences between reticulocytes and mature erythrocytes that could influence the tropism of different Plasmodium spp . , we undertook a non-targeted , high coverage , comprehensive analysis of the metabolomes of these host cells . Comparison of the metabolomes of very young , uninfected rat and human reticulocytes and their mature erythrocyte counterparts revealed major biochemical differences that could be exploited by intracellular parasite stages . This was tested using reverse genetics to disrupt parasite metabolism and establish the broad ability of P . berghei to utilise the products of reticulocyte metabolism and ( in part ) explain differing profiles of drug susceptibility between parasites in mature erythrocyte and reticulocyte environments .
Induction of reticulocytosis was achieved through administration of phenylhydrazine-HCl ( PHZ , 100 mg/kg body weight ) to Wistar rats and cells were harvested when the percentage of reticulocytes in the peripheral blood reached a maximum at day 5 ( ~35% reticulocytes ) . This was monitored by FACS analysis using the reticulocyte surface marker transferrin receptor ( CD71 ) , which is lost as reticulocytes mature[35] . More than 90% of the 35% reticulocyte population generated by PHZ treatment were CD71-high at the time of harvest ( Fig A-A in S1 Text ) corresponding to the youngest of the four forms of reticulocytes that have been identified [39] and are from here on referred to as Reticulocyte enriched Erythrocyte Population ( REP ) Material was also collected for comparison with blood from non-enriched ( ~1% reticulocytes ) animals- wild type Erythrocyte Population ( wtEP ) ( Fig 1A ) . All samples were uniformly depleted of leucocytes . Metabolite extracts of REP and wtEP were analysed in parallel by liquid chromatography mass spectrometry ( LC-MS ) and gas chromatography mass spectrometry ( GC-MS ) , providing overlapping , as well as complementary coverage of the metabolomes of wtEP and REP . LC-MS data was processed using XCMS , MZMatch and IDEOM while GC-MS data was processed using PyMS matrix generation and Chemstation Electron Ionisation ( EI ) spectrum match analysis ( described in detail in methods ) . A total of 333 metabolites were provisionally identified from a total of 4 , 560 mass features and peaks . The volcano plot in Fig 1B shows the distribution of abundance of detected metabolites in REP compared to wtEP . Almost half of all detected metabolites ( 147 , ~45% ) were found to be more than 2-fold more abundant in REP ( with a p<0 . 05 ) ( Fig 1B and A-C in S1 Text and S1 Table ) . Only 5 ( ~1% ) metabolites were over 2-fold more abundant in wtEP than in REP ( with p<0 . 05 ) . The rest of the metabolites did not show a significant difference between REP and wtEP . Similar changes were observed when all mass features and peaks ( ~4 , 560 peaks ) were included in the analyses . Specifically , of the ~4 , 230 unassigned mass features/peaks , 1 , 051 ( ~23% ) were up-regulated and 91 peaks ( ~2% ) down regulated in REP ( Fig A-B in S1 Text ) . As the blood from reticulocytosis-induced rats still contained a major fraction of mature erythrocytes ( 1:2 final ratio of reticulocytes to mature erythrocytes ) the level of metabolite enrichment in reticulocytes was actually much greater ( column four , S1 Table ) . 20 representative metabolites up-regulated in rodent REP showed a similar ‘trend’ towards up-regulation in very young human reticulocytes grown in vitro from CD34+ stem cells [41] analysed using LC-MS ( Fig 1C ) , except carnitine derivatives . All identified metabolites were charted on metabolic pathways known to exist in Plasmodium and mammalian host cell from biochemical studies [6 , 7 , 12 , 42 , 43] and genomic data [44] , although it is expected that not all detected metabolites are endogenously synthesised , as plasma metabolites from other tissues , the microbiome , the diet and environment may also accumulate in erythrocytes . Cell fractions from rodent REP contained elevated levels of glycolytic , pentose phosphate pathway and TCA cycle intermediates ( S1 Table ) . The presence of the latter indicates that reticulocytes have a functional TCA cycle and associated intermediary carbon metabolism , consistent with the presence of a residual population of mitochondria in reticulocytes that are largely lost in mature erythrocytes [34] . Increases in the levels of intermediates of the purine and pyrimidine metabolic pathways in reticulocytes presumably originate either from biosynthesis in the preceding erythropoiesis stages or from catabolism of nucleic acid to their constituent nucleobases [45] . A number of intermediates of phospholipid metabolism were also elevated in reticulocytes compared to mature erythrocytes . Other notable changes included elevated levels of intermediates in glutathione and arginine metabolism in reticulocytes ( S1 Table ) . In addition , many carnitine derivatives were found to be up-regulated in rodent ( although interestingly not in human ) reticulocytes which may relate to fatty acid catabolism by β-oxidation in the mitochondria or peroxisomes of these cells . Although decreased levels of carnitines have previously been found in human erythrocytes derived from normal subjects compared to individuals with Sickle-Cell ( HbSS ) disease [40] , the procedures used for production of rodent reticulocytes ( in vivo ) and human reticulocytes ( in vitro ) cannot be ruled out as the reason for this difference observed between the two species as carnitine is produced in mammalian tissues ( skeletal muscle , heart , liver , kidney , and brain ) [46] a contributory factor missing in in vitro conditions . Almost 65% of the other metabolite ions detected in the HbSS study were also found to be present in erythrocytes in our analysis ( S1 Table ) and around 17% of metabolites detected in our analysis were also reported in erythrocytes in that study [40] . This difference in coverage could be due to the chromatographic and detection methods which differ between the analyses . Taken together these data demonstrate that the reticulocyte contains elevated levels of many metabolites that could potentially be scavenged by the invading malaria parasite . Furthermore , there was a marked overlap in metabolic pathways observed in the reticulocyte and those predicted in the parasite [43 , 44] . Common pathways might therefore be uniquely dispensable to Plasmodium during its growth in the reticulocyte compared with that in mature erythrocytes . To test this hypothesis , we used reverse genetics to target several metabolic pathways in intermediary metabolism and pyrimidine biosynthesis in P . berghei whose intermediates were significantly up-regulated in reticulocytes . Asexual red blood cell stages of Plasmodium spp . catabolize glucose via the intermediary carbon metabolic pathways depicted in Fig 2A and express the cytosolic enzymes , phosphoenolpyruvate carboxylase ( pepc PBANKA_101790 ) , malate dehydrogenase ( mdh PBANKA_111770 ) and aspartate amino transferase ( aat PBANKA_030230 ) . De novo synthesis of aspartate is likely to be important for nucleic acid synthesis as this amino acid is utilised in both purine salvage and as a carbon skeleton in pyrimidine biosynthesis [47] and inhibition of aat has been shown to be lethal to P . falciparum [48] . Malate produced by these pathways either enters mitochondria to participate in the TCA cycle or is excreted [7 , 42] . Metabolites involved in TCA cycle and intermediary carbon metabolism ( ICM ) , including malate and aspartate , were found to be substantially higher in REP compared to wtEP ( Fig 2A ) . The elevated levels of these intermediates may possibly explain the previous observation that disruption of the TCA cycle in P . berghei blood stages through deletion of flavoprotein ( Fp ) subunit of the succinate dehydrogenase , pbsdha ( PBANKA_051820 ) , had little effect on parasite viability in blood stage forms , although ookinete development was impaired [49] . To further explore the possibility that P . berghei has potential access to the anapleurotic substrates of reticulocyte ICM , attempts were made to delete pepc , mdh and aat in P . berghei and assess the importance of these parasite enzymes throughout the life cycle ( Fig 2A ) . P . berghei mutants lacking both pepc and mdh were generated ( Fig B in S1 Text ) , while deletion of aat proved refractory . Both the pepc- and mdh- mutant parasites caused severe cerebral malaria in CD57/B6 mouse model with similar dynamics to wt parasites ( Fig 3B ) . Interestingly , the growth of the pepc- mutant was compromised compared to wild type parasites , as the pepc- mutant , but not the mdh- mutant was overgrown by the wt parasite in an in vivo sensitive single host competitive growth assay ( Fig 3A and C-A in S1 Text ) . The number of merozoites observed in mature schizont stages in both pepc- ( 17 . 02 ± 1 . 8 ) and mdh- ( 17 . 41 ± 1 . 7 ) mutants are similar to wt ( 17 . 4±1 . 8 ) ( Fig C-C in S1 Text ) . Scrutiny of the growth phenotype detected in the pepc- mutants showed that they have a prolonged asexual cycle ( 4 h longer than wt ) ( p<0 . 05 ) ( Fig C-B in S1 Text ) . The number of gametocytes formed in blood stages was also reduced in pepc- mutants by almost 50% but unaffected in mdh- ( p>0 . 05 ) ( Fig 3C ) with no notable difference in male to female ratio in either mutant . Further phenotypic analyses showed reduction of exflagellation ( pepc- mutants 84% less than wt , p<0 . 0005; mdh- mutants 56% less than wt , p<0 . 005 ) ( Fig 3D ) . DNA replication in male gametocytes as observed by FACS analysis was reduced by 50% compared to wt at the 8 minute time point and further delayed taking up to 16 minutes to complete ( Fig C-E and C-D in S1 Text ) . Ookinete development in in vitro cultures of pepc- mutants was also severely affected while in mdh- mutants , ookinetes were formed but the number was reduced by about 50% compared to wt ( Fig 4A ) . To determine if this defect was sex specific , crosses of pepc- and mdh- were performed with P . berghei lines RMgm-348 ( Pb270 , p47- ) which produces viable male gametes but non-viable female gametes and RMgm-15 ( Pb137 , p48/45- ) which produces viable female gametes but non-viable male gametes [50] . Mutants of pepc- were found to produce severely reduced numbers of ookinetes in either cross suggesting that gametes of both genders are affected and that the activity of the protein is essential for viable gamete formation . This was not the case for mdh- mutants where although crossing experiments showed that lack of MDH protein affected both genders , they mimicked the parental phenotype producing 50% fewer mature ookinetes ( Fig 4B ) . The pepc- parasites were defective in development within the mosquitoes , forming small numbers of oocysts in mosquito midguts and no salivary gland sporozoites . However , parasites lacking mdh could complete transmission through the mosquito and infect mice generating blood stage asexual forms in 48–72 hours similar to wt despite producing reduced numbers of oocysts when compared to wt ( Fig 4C and 4D and D-A and D-B in S1 Text ) . Overall , these results suggest that two key enzymes in P . berghei ICM are at least partially redundant during stages of infection in which the parasites resides primarily in reticulocytes , but that they become essential as parasite differentiates and proliferates within other host or vector cell types . Plasmodium spp . are heavily dependent on nucleic acid synthesis during blood stage asexual growth and either salvage ( i . e . purines ) or synthesize ( i . e . pyrimidines ) the requisite bases . A schematic representation of the pyrimidine biosynthesis pathway is given in Fig 2B . Five out of six enzymes of this pathway have been shown to be essential for P . falciparum growth in standard in vitro cultures , based on pharmacological studies [51] . Interestingly , most of these inhibitors are markedly less potent in the in vivo P . berghei model , a feature that has been attributed to reduced bio-availability of inhibitors in mice or apparent differences in target enzyme structures [52 , 53] . However , increased resistance to pyrimidine biosynthetic inhibitors could also reflect higher concentrations of pyrimidine precursors ( bar glutamine ) in the reticulocyte population selectively colonized by this species ( Fig 2B ) [16 , 51] . To investigate this possibility we attempted to delete in P . berghei 6 genes encoding enzymes involved in pyrimidine biosynthesis; carbamoyl phosphate synthetase II ( cpsII ) ( PBANKA_140670 ) , aspartate carbamoyltransferase ( act ) ( PBANKA_135770 ) , dihydroorotase ( dhoase ) ( PBANKA_133610 ) , dihydroorotate dehydrogenase ( dhodh ) ( PBANKA_010210 ) , orotate phosphoribosyltransferase ( oprt ) ( PBANKA_111240 ) and orotidine 5′-monophosphate decarboxylase ( ompdc ) ( PBANKA_050740 ) . While the first four enzymes in this pathway were refractory to deletion , the last two enzymes in pyrimidine biosynthesis , orotate phosphoribosyltransferase ( oprt ) and orotidine 5′-monophosphate decarboxylase ( ompdc ) could be deleted ( Fig B in S1 Text ) . The oprt- and ompdc- mutant parasites grew slowly ( asexual cycle prolonged by approximately 4–5 hours compared to wt ( p<0 . 05 ) ) , were rapidly outgrown in a competition growth assay with wt parasites ( Fig 3A ) and based on gray value-1 of staining intensity as observed by Giemsa staining ( p<0 . 0005 ) , seem to invade very young reticulocytes ( Fig C-E in S1 Text ) . However , these infected reticulocytes could not be classified as CD71-high possibly due to the accelerated loss of the CD71 as observed with P . vivax infected reticulocytes [30] . Furthermore , both oprt- mutants ( 15 . 9 ± 2 . 0 , p<0 . 0005 ) and ompdc- mutants ( 15 . 2 ± 2 . 5 , p<0 . 0005 ) were found to generate , on average , significantly fewer merozoites than wt parasites ( 17 . 5 ± 1 . 8 ) per schizont ( counted after completion of asexual cycle ) ( Fig C-C in S1 Text ) and the asexual parasites also took longer to mature to schizonts ( Fig C-B in S1 Text ) . Both mutants showed altered lethality in the C57/B6 mouse model as the mice infected with the mutants did not manifest the symptoms of experimental cerebral malaria ( ECM ) but died between days 14–20 as a result of severe anaemia and hyperparasitemia ( Fig 3B ) . The process of transmission was also affected by the loss of ompdc and oprt . Gametocytemia was significantly reduced only in oprt- parasites ( Fig 3C ) but no change was seen in male- female ratio . Exflagellation ( the production of mature male gametes ) was found to be severely affected in oprt- and completely blocked in ompdc- parasites ( Fig 3D ) and DNA replication during male gametogenesis was severely reduced ( Fig 3E ) . Consistent with the defects in male gametogenesis , very few ookinetes were formed in in vitro cultures in oprt- parasites and no ookinetes were observed in ompdc- ( Fig 4A ) . Genetic crosses of oprt- and ompdc- mutants were performed as above with P . berghei lines RMgm-348 and RMgm-15 which showed that viable male gametes ( from RMgm-348 ) were able to rescue the ookinete conversion defect in both mutant lines suggesting that formation of male gametes is impaired in both oprt- and ompdc- mutant parasites while female gametes remain unaffected ( Fig 4B ) . Infectivity to the mosquito was significantly reduced in oprt- and completely blocked in ompdc- mutants as seen by observing oocysts in infected mosquito midguts and salivary gland sporozoites ( Fig 4C and 4D and D-C and D-D in S1 Text ) and infection to naïve mice was found to be completely blocked . However , when ookinetes from p47- x oprt- or ompdc- crosses were fed to mosquitoes , they failed to develop into mature oocysts ( Fig E in S1 Text ) hence , did not complete sporogony indicating that lack of both oprt and ompdc in the female lineage results in an allelic insufficiency in a growing oocyst . We also tested the effect of a previously published inhibitor of pyrimidine biosynthesis 5-fluoroorotate ( 5FOA ) [54] on asexual growth of both P . falciparum and P . berghei . The comparisons were carried out in vitro to prevent bioavailability of the inhibitor confounding in vivo data in mice . We tested the activity and found that the IC50 value of 5FOA in vitro was almost 90-fold higher in P . berghei ( 32 . 2 ± 0 . 9 nM ) compared to P . falciparum ( 0 . 37 ± 0 . 01 nM ) ( Fig 5 ) . A dihydroartemisinin control showed no major difference in inhibition between P . berghei ( 6 . 6 ± 0 . 1 nM ) and P . falciparum ( 2 . 8 ± 0 . 2 nM ) . These data strongly suggest that P . berghei can access pyrimidine precursors from the reticulocyte and are consistent with a role of host cell metabolism in the differential activity of 5FOA , although differences in sensitivity of P . falciparum [55] and P . berghei [56] thymidylate synthase or differences in drug uptake could also contribute to the differential lethality .
Although a small number of metabolomics studies have been undertaken on erythrocytes , including a comparison of normal and HbSS erythrocytes , the number and range of metabolites detected in these studies were relatively small ( 20–90 ) [39 , 40] . Here we used complementary LC-MS and GS-MS analytical platforms to maximise coverage , generating the most comprehensive coverage of REP and wtEP undertaken to date ( 333 metabolites ) . These studies revealed a much higher degree of metabolic complexity in reticulocytes compared to mature erythrocytes covering nearly all major pathways in central carbon metabolism . Whilst glycolysis is the main pathway for carbon metabolism in erythrocytes [36] , both human [57] and rodent [58] erythrocytes retain a residual proteomic signature of TCA cycle and ICM enzymes and our metabolomics data suggests that these pathways are much more active in reticulocytes , leading to elevated levels of TCA intermediates ( including citrate , malate ) and ICM products ( e . g . aspartate ) . The functional significance of increased metabolic complexity in reticulocytes was subsequently tested by generating P . berghei mutants with specific defects in metabolism showing that the increased availability of complementary metabolites in reticulocytes can explain the non-essential nature of the P . berghei pepc and mdh genes , which are involved in regulating intracellular levels of oxaloacetate and malate . In contrast , PEPC is essential for normal intra-erythrocytic survival of P . falciparum in vitro , although this can be bypassed by malate supplementation of P . falciparum infected mature erythrocytes [42] . It should be noted that whilst the P . berghei pepc- mutant retained its virulence , it still showed a significant growth defect compared with wild type parasites ( similar to the P . falciparum mutant [42] ) resulting at least in part from a prolongation of the asexual blood stage cycle as revealed by our sensitive single host competitive growth assay . It would be interesting to use this assay to compare asexual growth dynamics of other available metabolic mutants such as the pbsdha- with wild type which might reveal additional defects to those reported [49] . The P . berghei pepc- mutant also failed to complete transmission through mosquitoes as a result of defects in gametocyte production , male gamete formation , female gamete viability resulting in trace oocyst formation and failure to enter sporogony , which extends our understanding of the importance of this metabolic enzyme for parasite development beyond the asexual blood stages previously investigated [42] . A possible explanation for this phenotype is that the pepc- mutant is unable to by-pass the need for de novo synthesized aspartate for nucleotide biosynthesis by salvage from different host cells during its sexual and asexual life cycle ( Fig 2A ) . The demonstration of pbpepc- growth in reticulocytes suggests that the equivalent P . falciparum mutant might be a suitable candidate for an attenuated slow growing parasite vaccine that would permit generation of significant anti-parasitic immune responses . In line with this suggestion , we were unable to delete pbaat which is also required for de novo synthesis of aspartate . The essential nature of pbaat suggests that either the apparently higher levels of aspartate in reticulocytes are insufficient to meet the demands of a growing asexual stage parasite or that , as in P . falciparum intra-erythrocytic stages , P . berghei is not readily able to access host cytoplasmic pools of aspartate [59] . Production of aspartate in Plasmodium pepc- mutants can still be achieved through generation of the oxaloacetic acid precursor by mitochondrial malate: quinone oxidoreductase ( MQO ) or the reverse reaction of cytosolic MDH . However , this is apparently a suboptimal solution for the pepc- parasite resulting in slow growth in the blood stage and failure to develop in the mosquito . Plasmodium AAT can also generate methionine from aspartate , glutamate and other amino acids which can act as effective amino group donors [60] and regulate glutamine/glutamate metabolism . These functions may not be rescued by simple aspartate salvage from the host and further support the essentiality of aat as a key enzyme for the parasite and a possible drug target . The P . falciparum gene encoding MDH has proved refractory to deletion under any circumstances so far , suggesting that it is essential for these parasites . In marked contrast , the P . berghei mutants lacking pbmdh were readily generated , suggesting that this species may scavenge reticulocyte pools of malate or other intermediates in the TCA cycle . The pbmdh mutant exhibited a very modest growth phenotype and was able to develop into mosquito infective stages , although it produced 30% fewer oocysts than wt parasites . The continued viability of the pbmdh- mutants during transmission in the absence of reticulocyte-based compensatory sources of the metabolite can be explained by continued TCA derived production of malate and NADH+ H+ reducing equivalents given the increased flux through the TCA metabolism in gametocytes and probably later sexual stages [7 , 12] . Conditional silencing or disruption of pfmdh or degradation of PfMDH in mature gametocytes or later stages of P . falciparum would establish if MDH is required for transmission of the human parasite and that the essential nature of this enzyme is merely blood stage specific . Plasmodium spp . salvage their purine requirements from the host cell , but retain the ability to synthesise pyrimidines [61] . Purine nucleosides are taken up by the parasite PfNT1 and other , as yet , unidentified AMP transporters [62] after they are delivered to the parasitophorous vacuole via the action of erythrocyte nucleoside transporters [51 , 63] and a non-selective transport process [61 , 64] . In contrast , while other Apicomplexans ( i . e . Cryptosporidium spp . , Toxoplasma spp . ) retain the capacity to salvage pyrimidines [16] , Plasmodium spp . are thought to lack enzymes required for host pyrimidine salvage [44] . Although , Plasmodium proteins have been implicated in transporting some pyrimidine precursors [65 , 66] , presumably due to very limited availability of pyrimidines in the host cell , Plasmodium parasites have been thought to be completely dependent on de novo pyrimidine synthesis for growth in asexual stages [51] . The survival of both oprt-and ompdc- mutants could be the result of two possibilities that are not mutually exclusive . The first possibility is that the mutants directly utilize reticulocyte pools of pyrimidines which are nonetheless limiting leading to a reduction in number of merozoites produced . Alternatively , mutant parasites could synthesize orotate which is secreted into the host cytoplasm and converted to UMP by host UMP synthase before being salvaged . Both outcomes require transport of nucleosides or nucleotides from the host cytoplasm to the parasite and how this is achieved is not clear . Both pyrimidine biosynthesis mutants survive only in the youngest reticulocytes which might reflect either adequacy of supply of host UMP ( or derivatives ) or the capacity of the youngest reticulocytes to convert parasite-derived orotate . Indeed enzymes involved in the later stages of pyrimidine biosynthesis , nucleoside diphosphate kinase B , CTP synthase and ribonucleotide reductase large subunit have been identified in rodent and human erythrocytes [57 , 58] . The possibility that host pyrimidine enzymes may have redundant functions with the parasite enzymes catalysing late steps in pyrimidine biosynthesis is supported by the apparent essentiality of the P . berghei genes encoding the first four steps of pyrimidine biosynthesis . A simplified illustration of life cycle stages of P . berghei development showing the characteristics of mutant parasites at various points in the life cycle is shown in Fig G in S1 Text . The REP metabolome also explains other species-specific differences between P . berghei and P . falciparum . Glutathione biosynthesis occurs in erythrocytes [67] and the enzymes for this pathway have been shown to be present in both human [57] and rodent [58] erythrocytes . Plasmodium employs its own glutathione redox system [68] to counter oxidative stress ( Fig F-A in S1 Text ) . Both ɣ-glutamylcysteine synthetase ( ɣ-gcs ) and glutathione synthetase ( gs ) are essential for parasite survival in P . falciparum [69] yet ɣ-gcs and glutathione reductase ( gr ) can be deleted in P . berghei and intra-erythrocytic asexual growth is unaffected although mosquito stage development is arrested at the oocyst stage [70 , 71] . The REP and wtEP metabolomes demonstrated that the levels of glutathione synthesis intermediates were higher in reticulocytes than in mature erythrocytes ( Fig F-B in S1 Text ) providing a mechanistic explanation for the normal growth of P . berghei ɣ-gcs and gr mutant asexual stages in reticulocytes . Also , the inhibitor of ɣ-gcs , buthionine sulphoximine ( BSO ) inhibits P . falciparum growth with an IC50 value of ~60 μM [69] yet concentrations as high as 500 μM BSO in vitro had no inhibitory effect on P . berghei parasites in in vitro cultures ( Fig F-C in S1 Text ) consistent with the reticulocyte mediated rescue of chemical disruption of the glutathione synthesis pathway in P . berghei , although it has not been investigated whether there is a difference in BSO sensitivity against the mouse or P . berghei ɣ-gcs compared to P . falciparum or human enzymes . Enzymes involved in Plasmodium intermediary carbon metabolism [12 , 42] and pyrimidine biosynthesis [51] are considered attractive targets for drug development . The metabolome surveys and drug inhibition data presented here suggest that caution should be used before extrapolating conclusions regarding gene essentiality in reticulocyte preferent parasites such as P . berghei as part of any drug discovery pathway that has been based initially upon screens in mature erythrocytes . Bioavailability in mouse models and/or drug penetration into the reticulocyte and difference in target enzyme structures between species have been proposed as reasons for the relative ineffectiveness of drugs when tested in vivo using P . berghei [52 , 53] . An alternative view is that the reticulocyte metabolome ( at least in part ) provides a reservoir of metabolites downstream of the point of action of a drug rendering the drug less effective . This has a number of consequences:
Rodent reticulocyte enrichment was done in rats by administering phenylhydrazine-HCl dissolved in 0 . 9% NaCl ( w/v ) at a dose of 100 mg/kg body weight and collecting reticulocyte enriched peripheral blood on day 5 post injection . Metabolite extraction was done as using chloroform/methanol/water ( 1:3:1 v/v ) and samples were analysed using LC-MS and GC-MS . See S1 supplementary materials and methods for details . CD34+ cells obtained from blood from human volunteers were cultured in a three-stage protocol based on the methods of [41] . Cultured reticulocytes and mature erythrocytes from matching donors were used for metabolite extraction with chloroform/methanol/water ( 1:3:1 v/v ) and samples were analyzed using LC-MS . See S1 supplementary materials and methods for details . Infection of laboratory mice , asexual culture of P . berghei stages and generation of knockout parasites was done as before [73] . Asexual growth competition assay was done by mixing wt and mutant parasites expressing different fluorescent markers and injecting them intravenously into recipient mice and monitoring the growth of the two populations by flow cytometry as done before [74] . Lethality of mutant P . berghei parasites was checked by injecting infected RBCs ( 104 ) into C57/B6 mice and monitoring parasitaemia , disease pathology and mortality over 21 days . Gametocyte conversion was monitored by flow cytometry in mutants generated in parent line ( 820cl1m1cl1 ) expressing GFP in male gametocytes and RFP in female gametocytes [75] . DNA quantification during exflagellation was also monitored by flow cytometry in mutant P . berghei parasites . Development of ookinetes in wild type , mutants and sexual crosses was observed in standard in vitro cultures maintained at 21°C . Mosquito transmission experiments were done in 5–8 days old mosquitoes used for infected blood feeds at 21°C and monitored for oocyst and sporozoite development . See S1 supplementary materials and methods for details . Inhibitors were used to perform in vitro drug susceptibility tests in standard cultures of synchronized P . berghei and P . falciparum blood stages . For testing P . berghei inhibiton , inhibitors were used at increasing concentrations to culture ring stage P . berghei for 24 hours and parasite development to schizont stage was analyzed by flow cytometry after staining iRBCs with DNA-specific dye Hoechst-33258 . P . falciparum 3D7 strain was used for determining IC50 values of inhibitors in in vitro cultures by measuring 3H-Hypoxanthine incorporation in the presence of inhibitors in increasing concentrations . See S1 supplementary materials and methods for details . All animal work was approved by the University of Glasgow’s Animal Welfare and Ethical Review Body and by the UK’s Home Office ( PPL 60/4443 ) . The animal care and use protocol complied with the UK Animals ( Scientific Procedures ) Act 1986 as amended in 2012 and with European Directive 2010/63/EU on the Protection of Animals Used for Scientific Purposes . Blood from human volunteers was supplied by the Australian Red Cross Blood Service and experiments were approved by the Walter and Eliza Hall Institute Human Research Ethics Committee , Australia . As a part of standard Australian Red Cross Blood Service practice , blood was collected from healthy donors who were informed about this study and potential risks to them and gave written consent when they donated blood . | Malaria , caused by the Apicomplexan parasites Plasmodium spp . , is a deadly disease which poses a huge health and economic burden over many populations in the world , mostly in sub-Saharan Africa and Asia . To design new intervention strategies and to improve upon existing drugs against malaria , it is useful to understand the biochemistry of the Plasmodium parasite and its metabolic interplay with the host . Some species of Plasmodium such as P . vivax grow exclusively in reticulocytes ( immature erythrocytes ) whereas others e . g . P . falciparum will also readily multiply in mature erythrocytes . We asked the questions , do these two classes of host cell offer different resources for parasite survival and could these resources influence antimalarial drug efficacy ? We used metabolomics to compare rodent reticulocytes and mature erythrocytes and identified that the metabolome of the former is more diverse and enriched . Gene disruption in the reticulocyte preferring rodent malaria parasite P . berghei was used to demonstrate that Plasmodium can utilise the elements of the metabolic reserves of reticulocytes that mature erythrocytes cannot provide . Our data suggests that the availability of the reticulocyte metabolome might reduce or block the efficacy of antimalarial drugs that target parasite metabolism and drugs tested against P . falciparum might have significantly reduced activity against P . vivax . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Host Reticulocytes Provide Metabolic Reservoirs That Can Be Exploited by Malaria Parasites |
The genetic cause for approximately 80% of familial breast cancer patients is unknown . Here , by sequencing the entire exomes of nine early-onset familial breast cancer patients without BRCA1/2 mutations ( diagnosed with breast cancer at or before the age of 35 ) we found that two index cases carried a potentially deleterious mutation in the RECQL gene ( RecQ helicase-like; chr12p12 ) . Recent studies suggested that RECQL is involved in DNA double-strand break repair and it plays an important role in the maintenance of genomic stability . Therefore , we further screened the RECQL gene in an additional 439 unrelated familial breast cancer patients . In total , we found three nonsense mutations leading to a truncated protein of RECQL ( p . L128X , p . W172X , and p . Q266X ) , one mutation affecting mRNA splicing ( c . 395-2A>G ) , and five missense mutations disrupting the helicase activity of RECQL ( p . A195S , p . R215Q , p . R455C , p . M458K , and p . T562I ) , as evaluated through an in vitro helicase assay . Taken together , 9 out of 448 BRCA-negative familial breast cancer patients carried a pathogenic mutation of the RECQL gene compared with one of the 1 , 588 controls ( P = 9 . 14×10-6 ) . Our findings suggest that RECQL is a potential breast cancer susceptibility gene and that mutations in this gene contribute to familial breast cancer development .
Breast cancer is the most common malignancy disease in women world wide . Among these , approximately 10% of breast cancer patients have a family history of breast cancer ( referred as familial breast cancer ) , but only 10–15% of familial breast cancer is owing to germline mutations in one of the two high penetrance breast cancer susceptibility genes—BRCA1 and BRCA2[1] . Additionally , mutations in the moderate breast cancer susceptibility genes , such as PALB2 , ATM , CHEK2 , BRIP1 and RAD51C , contribute to 5% of familial breast cancers[2–6] . Therefore , the genetic causes of approximately 70–80% familial breast cancer remain to be discovered . Recently , next-generation sequencing assay provides a new platform to find cancer susceptibility genes . To find potential breast cancer susceptibility genes , we performed whole-exome sequencing in nine early-onset familial breast cancer patients who do not carry a germline mutation in the BRCA1/2 genes; all nine cases were diagnosed with breast cancer at or before the age of 35 , the index case had at least one first-degree relative affected with breast cancer . Based on the data of whole-exome sequencing of the nine cases , we further screened the potential gene in an additional 439 unrelated familial breast cancer patients without BRCA1/2 mutations and performed in vitro functional analyses to evaluate whether the mutations disrupt the function of the potential gene . Finally , our results indicate that RECQL ( RecQ helicase-like ) is a novel breast cancer susceptibility gene .
The detail information of the nine unrelated early-onset familial breast cancer patients who were subjected to whole-exome sequencing is presented in S1 Table . Approximately 46 , 000 variants were identified by whole-exome sequencing in each sample . We filtered the data for genes that harbored novel , heterozygous rare variants that were truncating mutations or splice-site variants . We further retained only genes that contain different variants shared in two or more cases ( S2 Table ) . As a result , there were three genes that fit these criteria and were validated by Sanger sequencing: TTLL2 , VSIG2 and RECQL ( S3 Table ) . Of these genes , RECQL is of great interest because it is involved in DNA repair process . The two mutations in the RECQL gene were found in two index cases through the whole-exome assay ( S1A and S1B Fig ) , one index case carried a nonsense mutation in exon 4 of the RECQL gene ( encoding L128X ) , leading to a premature stop codon; another index case carried a non-synonymous mutation ( R539P ) that was predicted to affect the function of RECQL . To determine the germline mutations in the RECQL gene in an additional cohort of familial breast cancer patients , we screened the entire coding region of RECQL gene using Sanger sequencing assays in 439 familial breast cancer patients without BRCA1/2 mutations . In total , we found 15 germline variants in the RECQL gene in the 448 familial breast cancer patients ( including the nine index cases for whole-exome sequencing ) ( Table 1 and S1 Fig ) . These variants contained three nonsense mutations ( L128X , W172X and Q266X ) , two potential splice-site mutations ( c . 395-2 A>G and c . 868-12_868-11del ) and ten missense mutations ( M1T , S63P , A195S , R215Q , N363S , R455C , M458K , R539P , H461R , and T562I ) ( Table 1 ) . We further tested the germline mutations in the RECQL gene in 1 , 588 healthy controls , and three missense mutations ( M1T , N363S and H461R ) were found in the controls ( Table 1 ) . The three missense mutations ( M1T , N363S and H461R ) were present in both familial breast cancer patients and controls . Among these mutations , the mutation frequency of M1T ( rs146924988 ) and N363S ( rs138663409 ) was similar between familial breast cancer patients and controls . Thus , these two variants are more likely to be neutral ( Table 1 ) . The three nonsense mutations ( L128X , W172X and Q266X ) lead to premature protein termination , and therefore , we considered these to be clearly pathogenic . The two splice-site mutations were predicted to affect the splice site . RNA from peripheral blood samples was isolated from the index case who carried the 395-2A>G mutation . The RT-PCR pattern of the RECQL gene from the 395-2A>G mutation carrier revealed a reduced expression of the normal transcript and one new truncated product compared to the control . Further sequencing confirmed that the 395–2 A>G mutation resulted in exon 5 skipping in the abnormal RT-PCR product ( Fig 1A ) . This product of the abnormal transcript disrupts the helicase domain of RECQL and leads to a premature stop signal ( G132fs* ) ; therefore , the 395-2A>G mutation is pathogenic . The pedigree of the index case who carried the RECQL 395-2A>G mutation is presented in Fig 1B . Both the index and the twin , who were negative for BRCA1/2 mutations , carried the mutation; they had the disease at the ages of 40 and 43 , respectively . Additionally , at least ten individuals in this family had breast/ovarian cancer , lung cancer , cervical cancer , or peritoneal cancer ( Fig 1B ) . Unfortunately , blood samples were only available for the twin; thus , it was not possible to perform the segregation analysis in this family . Another index carried the c . 868-12_868-11del mutation; however , RT-PCR analysis revealed that this mutation did not affect the splicing of RECQL ( S2 Fig ) . Thus , the c . 868-12_868-11del mutation was neutral . To investigate whether the remaining eight missense mutations ( S63P , A195S , R215Q , R455C , M458K , H461R , R539P , and T562I ) influenced the efficacy of RECQL helicase activity , we first examined the structures of the RECQL protein ( RCSB PDB , 2V1X and 2WWY ) ( S3 Fig ) . On the basis of the crystal structure , T562 is located in a β-hairpin , which is required for DNA unwinding [7 , 8]; A195 is involved in dimer interaction[8]; R215 is located near the ADP-binding pocket and is expected to weaken ATP hydrolysis[9]; the conserved residues R455 and M458 are located in the zinc binding subdomain , which is important for whole-protein stability[10] ( S3 Fig ) . Next , we performed helicase activity assays using GST fusion proteins in vitro[11] . The K119A mutation served as a negative control that is reported to affect the helicase activity of RECQL[12] . By assessing the helicase ability to unwind forked DNA substrates , single-strand DNA was essentially undetectable with four mutations , R215Q , R455C , M458K , and T562I , suggesting that these four mutations completely disrupted the helicase activity ( Fig 2A and 2B ) . Additionally , we found that the A195S mutant lost approximately 83 . 4% of the helicase activity compared to the wild-type of RECQL . The remaining three missense mutations ( S63P , H461R and R539P ) showed similarly effective helicase activities compared to the wild-type of RECQL ( Fig 2A and 2B ) . Thus , the mutations of S63P , H461R and R539P were neutral . Taken together , five missense mutations ( R215Q , R455C , M458K , T562I and A195S ) were pathogenic . In total , nine germline mutations in the RECQL gene in the nine familial breast cancer patients were identified as pathogenic , including three nonsense mutations ( L128X , W172X and Q266X ) , one splice-site mutation ( 395-2A>G ) and five missense mutations ( A195S , R215Q , R455C , M458K and T562 ) ( Table 1 and Fig 3 ) . The pedigrees of the nine families are present in Fig 1B and S4 Fig . The average number of the breast cancer cases in the nine families were 2 . 8 cases/per family , with a mean age of onset of 47 . 8 years . The overall frequency of the RECQL germline mutations in the 448 familial breast cancer patients without BRCA1/2mutations was 2 . 0% ( 9 of 448 ) . One pathogenic mutation R455C was found in the 1 , 588 controls . The prevalence of the RECQL germline mutations was significantly higher in familial breast cancer patients than in the controls ( 9/448 vs . 1/1 , 588; the Fisher exact test , P = 9 . 14×10–6 ) . We then analyzed the clinical information of the nine cases with RECQL pathogenic mutations ( S4 Table ) . The mean age at diagnosis of breast cancer in the nine cases with RECQL mutation was younger than in those without RECQL mutation ( 45 . 1vs . 51 . 3 years; P = 0 . 12 ) . The hormone receptor and HER2 status were available for all the nine cases: eight and five were positive for the estrogen and progesterone receptors ( ER and PR ) , respectively . Seven cases had a HER2 negative tumor , and two had a positive tumor ( S4 Table ) . These results indicated that RECQL-associated breast cancer was similar to those of BRCA2-associated breast cancer . To test whether loss of heterozygosity ( LOH ) of RECQL occurred in the nine index cases who carried a pathogenic mutation , we performed the LOH assay in five index cases in which matched fresh tumor tissues and blood samples were available . As a result , no LOH was observed in the five cases ( S5 Fig and Table 1 ) .
The RECQL gene is located on chromosome 12p12 and encodes a protein of 649 amino-acids . It contains two important domains , the helicase domain ( residues 63–418 ) and the RecQ carboxy-terminal ( RQC ) domain ( residues 419–592 ) [7] . These domains are highly conserved in the RecQ family and are essential for helicase activity . Five RecQ helicase proteins ( named RECQL , BLM , WRN , RECQL4 and RECQL5 ) in humans are highly conserved and are considered to be genome caretakers that suppress neoplastic transformation[13] . Although no hereditary disease has been linked with the RECQL gene to date , mutations in three of five RecQ genes , BLM , WRN and RECQ4 , lead to Bloom , Werner , and Rothmund-Thomson syndromes , respectively , and are associated with cancer predisposition and/or premature aging . One recent study suggested that germline mutations in the BLM gene cause susceptibility to breast cancer , although the mutations are quite rare[14] . Another study indicated that RECQL5 polymorphisms are associated with an increased breast cancer risk in Chinese population[15] . Increasing evidence suggests that RECQL is involved in DNA double-strand break repair through the homologous recombination ( HR ) pathway[16] . RECQL-deficient cells or knockout mice exhibited chromosomal instability , sensitivity to ionizing radiation , and increased DNA damage , suggesting that RECQL plays an important role in the maintenance of genomic stability[17 , 18] . In this study , nine patients carried a pathogenic mutation in the RECQL gene . All of the pathogenic mutations are mapped to the above mentioned domains . The three nonsense mutations ( L128X , W172X and Q266X ) and one splice-site mutation ( 395–2 A>G ) resulted in a premature truncated protein and lost the helicase activity; the five missense mutations ( A195S , R215Q , R455C , M458K and T562I ) were also confirmed to disrupt the function of helicase activity through a functional analysis . Therefore , mutations in the RECQL gene can lead to breast cancer tumorigenesis . In addition , no LOH was found in the RECQL mutation carriers , suggesting that RECQL-associated tumorigenesis may be through RECQL haploinsufficiency . Here , we are the first to report that RECQL is a potential breast cancer susceptibility gene and that germline mutations in the RECQL gene are associated with predisposition to breast cancer . The 2 . 0% pathogenic mutation rate of the RECQL gene in familial breast cancer patients is remarkable and may be suitable for screening the mutations in BRCA1/2- negative breast cancer patients .
A total of 514 familial breast cancer patients were treated at the Breast Center , Peking University Cancer Hospital from 2003 to 2011 . Among these , 448 index cases were negative for BRCA1/2 germline mutations . To maximize the chance of identifying a novel breast cancer susceptibility gene , 9 unrelated early-onset familial breast cancer patients were selected from the pool of 448 cases and were subjected to whole-exome sequencing ( S1 Table ) . These nine cases were diagnosed with breast cancer at or before the age of 35 and had at least one first-degree relative affected with breast cancer . The remaining 439 unrelated familial breast cancer patients were used to screen the potential susceptibility gene . A total of 1 , 588 unrelated healthy women served as controls . Approximately 95% of familial breast cancer cases and controls are ethnic Chinese Han and reside in the northern region of China . The healthy controls were age-matched to cases . Genomic DNA was extracted from peripheral blood using standard protocols for all study subjects . This study was reviewed and approved by the Ethics Committee of Peking University Cancer Hospital ( project No . 2011KT12 ) . Informed written consent was obtained from all participants . Three micrograms of genomic DNA extracted from each blood sample was enriched for exonic regions using the SureSelect Biotinylated RNA Library ( BAITS ) . The sequences of captured libraries were generated as 90-bp pair-end reads on an Illumina Hiseq2000 . Exome capture and sequencing resulted in a minimum coverage of 10× for at least 91 . 3% of the capture target regions , and whole exomes were sequenced to an average mapped coverage of 108× . Sequencing reads were mapped to the reference GRCh37/hg19 human genome assembly using the Burrows-Wheeler Aligner ( BWA ) [19] . Further processing , including duplicate removal , local realignment and base quality recalibration , was performed using Picard and GATK . Single nucleotide variants ( SNVs ) and indels were detected by SOAPsnp[20]and SAMtools , respectively . Then , filters were applied to obtain variant results of higher confidence . We then used ANNOVAR[21]to perform annotation and classification . The variant collection was excluded from positions found in the dbSNP 132 and 1000 Genomes databases . Only genes that harbored heterozygous variants that were truncating mutations or splice-site variants were selected for further analyses . Candidate disease-causing variants were filtered for variants affecting the same gene in at least two samples using a straightforward criterion ( S2 Table ) . SNV and indel data were analyzed separately . After the filtering process , the variants in the RECQL gene and other potential genes found in the whole-exome sequencing assay were tested by Sanger sequencing using standard methods ( S3 Table ) . We designed a set of 14 pairs of primers ( S5 Table ) to screen the entire coding regions of the RECQL gene in an additional 439 unrelated familial breast cancer patients and 1 , 588 healthy controls . The purified products were then sequenced on an ABI 3730 automated sequencer ( Applied Biosystems ) . All mutations were confirmed in duplicate . Total RNA was isolated from the blood samples carrying the RECQL c . 395-2A>G mutation . Then , 2 . 3 μg of total RNA was transcribed to cDNA by the Superscript II Reverse Transcriptase ( Invitrogen ) using a random primer . PCR was then carried out using the primer pair spanning exons 3–7 of RECQL transcript 001 ( ENST00000444129 ) . The PCR products were separated on a 2% agarose gel by electrophoresis . The DNA fragments were re-amplified with 30 cycles , visualized and directly sequenced after gel extraction . cDNA from the blood samples of a healthy volunteer were used as the control . The analysis of c . 868-12_868-11del mutation was carried out using the primer pair spanning exons 6–11 ( S5 Table ) . Multiple-sequence alignments were generated for homologous RECQL protein sequences using Clustal Omega . Jalview was used to visualize and format the alignment . Mutations in human RECQL were visualized using PyMOL on the crystal structure of the human RECQL protein ( RCSB PDB , 2V1X ) and the structure of this helicase in complex with a DNA substrate ( RCSB PDB , 2WWY ) . cDNA encoding full-length RECQL was cloned into the pGEX-4T-1 plasmid to be expressed as a GST fusion protein . Eight missense mutations ( S63P , A195S , R215Q , R455C , M458K , H461R , R539P and T562I ) were introduced by site-directed mutagenesis and confirmed by sequencing . A helicase-defective mutant K119A was produced as well and served as a negative control . Recombinant proteins were purified from BL21 ( DE3 ) -RIL cells ( Stratagene ) using glutathione Sepharose columns ( GE Healthcare ) in a buffer containing 50mM Tris-HCl ( pH 7 . 0 ) , 250mM NaCl , 5mM β-mercaptoethanol and concentrated to 0 . 2 mg/ml in another buffer containing 50mM Tris-HCl ( pH 7 . 0 ) , 250mM NaCl , 5mMβ-mercaptoethanol and 20% glycerol . The proteins were assayed for helicase activity , detected by the displacement of a 32P-5’labeled 45mer oligonucleotide from a 44mer/45mer partial DNA duplex as previously described [12] . The unwinding of the substrate DNA was detected by autoradiography and quantified by Quantity One ( Bio-Rad Laboratories , Inc . ) . The helicase data represent the mean of three independent experiments with the mean ± S . D . indicated by error bars . Of the nine cases who carried a deleterious mutation in the RECQL gene , the fresh tumor tissues and blood samples were available for five cases , LOH analysis of the RECQL locus was carried out for these five cases . We specifically amplified the tumor DNA fragments using the same PCR conditions that we applied for germline DNA . Then , we directly sequenced them with an ABI 3730 automated sequencer ( Applied Biosystems ) and compared them to sequencing results of heterozygous germline DNA ( S5 Fig ) . The following GenBank reference sequences were used for variant annotation: RECQL , NM_002907 , TTLL2 , NM_031949 andVSIG2 , NM_014312 . 1000 Genomes Project , http://browser . 1000genomes . org/; dbSNP database , http://www . ncbi . nlm . nih . gov/projects/SNP/; Ensembl , http://www . ensembl . org/index . html; Picard , http://picard . sourceforge . net; GATK , http://www . broadinstitute . org/gatk/;SAMtools , http://samtools . sourceforge . net/;Clustal Omega , http://www . ebi . ac . uk/Tools/msa/clustalo/; Jalview , http://www . jalview . org/; PyMOL , http://www . pymol . org/; The Protein Data Bank , http://www . rcsb . org/pdb/ . | In this study , we aimed to find novel breast cancer susceptibility genes by whole-exome sequencing in nine early-onset familial breast cancer patients without BRCA1/2 mutations . We found that two index cases carried a potentially deleterious mutation in the RECQL gene ( RecQ helicase-like ) . We further screened the RECQL gene in an additional 439 unrelated familial breast cancer patients . In total , we found nine index cases carried a pathogenic mutation in the RECQL gene among the 448 BRCA-negative familial breast cancer patients . The RECQL is one of five RecQ helicase proteins ( named RECQL , BLM , WRN , RECQL4 and RECQL5 ) . RECQL is considered to be genome caretaker , and mutations in three of five RecQ genes , BLM , WRN and RECQ4 are associated with cancer predisposition and/or premature aging . Here , we are the first to report that mutations in the RECQL gene are associated with predisposition to breast cancer and this finding may have potential clinical implications and raise research questions about RECQL . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Mutations in RECQL Gene Are Associated with Predisposition to Breast Cancer |
Environmental stress during early development in animals can have profound effects on adult phenotypes via programmed changes in gene expression . Using the nematode C . elegans , we demonstrated previously that adults retain a cellular memory of their developmental experience that is manifested by differences in gene expression and life history traits; however , the sophistication of this system in response to different environmental stresses , and how it dictates phenotypic plasticity in adults that contribute to increased fitness in response to distinct environmental challenges , was unknown . Using transcriptional profiling , we show here that C . elegans adults indeed retain distinct cellular memories of different environmental conditions . We identified approximately 500 genes in adults that entered dauer due to starvation that exhibit significant opposite ( “seesaw” ) transcriptional phenotypes compared to adults that entered dauer due to crowding , and are distinct from animals that bypassed dauer . Moreover , we show that two-thirds of the genes in the genome experience a 2-fold or greater seesaw trend in gene expression , and based upon the direction of change , are enriched in large , tightly linked regions on different chromosomes . Importantly , these transcriptional programs correspond to significant changes in brood size depending on the experienced stress . In addition , we demonstrate that while the observed seesaw gene expression changes occur in both somatic and germline tissue , only starvation-induced changes require a functional GLP-4 protein necessary for germline development , and both programs require the Argonaute CSR-1 . Thus , our results suggest that signaling between the soma and the germ line can generate phenotypic plasticity as a result of early environmental experience , and likely contribute to increased fitness in adverse conditions and the evolution of the C . elegans genome .
Phenotypic plasticity in response to environmental stress is a critical component of organismal fitness . Environmentally-induced phenotypic variation is thought to result , in part , from programmed changes in gene expression inherited through cell divisions or transgenerationally via epigenetic mechanisms such as DNA methylation , histone modifications , and non-coding RNAs [1] . For example , in nematodes , Drosophila , and humans , nutritional status during early development can modulate the longevity of subsequent generations in the absence of the original stimulus [2–4] . In C . elegans , the observed increased longevity is dependent upon the inheritance of starvation-induced non-coding small RNAs , likely imported into the germ line from the soma [3] . While this example is intriguing , an important and unresolved question is to what extent does environmentally-induced phenotypic plasticity mediated by changes in epigenetic marks result in adaptive variation of traits that is favored by natural selection [5 , 6] . The free-living nematode C . elegans is an excellent model system to investigate the molecular mechanisms regulating environmentally induced phenotypic plasticity because its developmental trajectory is dependent on the environmental conditions experienced early in life . If conditions are favorable , C . elegans undergo continuous development consisting of four larval stages ( L1-L4 ) followed by reproductive adulthood [7] . When faced with environmental stress ( e . g . over-population , low food supply , or elevated temperatures ) , L1 larvae initiate an alternative diapause stage named dauer . Dauers are developmentally arrested , non-feeding , non-aging larva that exit diapause only if environmental conditions are favorable [8] . Natural populations of C . elegans experience a “boom and bust” reproductive strategy whereby they exist primarily as dauers for stress resistance and geographic dispersal but reproduce rapidly when food is available [9] . A potential prerequisite for the evolution of mechanisms that modulate adult phenotypes in response to specific stressors is that early environmental conditions are predictive of future conditions . The sensitivity of this developmental system to different dauer-inducing conditions , and how this may contribute to distinct phenotypic trajectories , is unknown . Our previous work has shown that postdauer ( PD ) adults that experienced crowding early in development exhibit changes in gene expression , genome-wide chromatin states , small RNA populations , and life history traits compared to isogenic animals that experienced continuous development ( controls , CON ) [10–12] . Thus , numerous molecular cues may have the potential to propagate information regarding early-life experiences to modulate adult developmental outcomes . Over a century ago , August Weismann proposed in the germ-plasm theory of heredity that only germ cells , and not somatic cells , could pass heredity information [13] . Increasingly , studies have challenged the “Weismann Barrier” by demonstrating that the passage of non-coding RNAs between somatic tissues and the germ line can result in transgenerational inheritance [14] . For example , a recent report showed that C . elegans double-stranded RNA ( dsRNA ) generated in neurons and transported to the germ line resulted in transgenerational silencing that is dependent upon the main systemic RNA interference ( RNAi ) effector , SID-1 ( WBGene00004795 ) [15] . Our previous work has shown that RNAi-pathways are required in different subsets of neurons for dauer formation in response to distinct environmental stresses , as well as the resulting reproductive plasticity observed between control and postdauer adults [11 , 16] . These observations raised the intriguing possibility that postdauer animals that experienced different early life stresses retain distinct molecular signatures mediated by non-coding RNA signals . Here , we investigate the effect that early environmental history has on modulating phenotypic plasticity in adults . We show that postdauer adults exposed to starvation ( Stv ) early in life exhibit distinct gene expression profiles and reproductive phenotypes when compared to postdauer adults that experienced crowding or high pheromone ( Phe ) . These differences are highlighted by a set of “seesaw” genes that are oppositely regulated between postdauer and control adults depending on the experienced environmental stress . In addition , we provide evidence that the significant seesaw pattern of gene expression is due to RNAi-dependent fluctuations of gene expression across whole chromosomes , resulting in transcriptome-wide seesaw trends of gene expression changes in response to stress . Moreover , our results indicate that the distinct gene expression profiles in starvation versus pheromone conditions are dependent on germline- or somatic-generated signals , respectively , that potentially move between tissue types and mediate changes in brood size . Our results suggest a model where crosstalk between the soma and the germ line governs the mRNA transcriptome and reproductive plasticity following distinct environmental histories . Finally , we provide evidence that the relationship between early life environmental stress and distinct postdauer phenotypes has contributed to the evolution of C . elegans genome organization .
To test the hypothesis that postdauer adult phenotypes are dependent upon the dauer-inducing stress experienced early in development , we conducted RNA-Seq using wild-type ( WT ) control ( CONStv ) and postdauer adults that passed through the dauer stage as a result of having experienced starvation ( PDStv ) ( S1A–S1D and S2 Figs ) . Genes with significant changes in mRNA levels due to starvation-induced passage through the dauer stage were identified by comparing CONStv and PDStv libraries and subjected to a false discovery rate ( FDR ) p-value correction of less than 5% . We identified 1 , 121 and 551 genes that exhibited significant up- and downregulation , respectively , in wild-type PDStv compared to wild-type CONStv ( WTStv ) ( Fig 1A and S1 Table ) . To determine if PDStv adults are distinct from PDPhe , we repeated our previous experiment [10] and conducted RNA-Seq on CONPhe and PDPhe populations grown in parallel to the starvation samples . We identified 441 and 560 genes that were significantly up- or downregulated , respectively , in wild-type PDPhe compared to wild-type CONPhe ( WTPhe ) ( Fig 1B and S1 Table ) . Next , we compared the transcriptional changes between WTStv and WTPhe to determine if different dauer-inducing stresses result in distinct transcriptional memories . A comparison of the 1 , 121 upregulated genes from WTStv to the 441 upregulated genes identified in WTPhe identified only two genes ( y38f1a . 6 ( WBGene00012608 ) and t05b11 . 3 ( WBGene00020246 ) ) in common . Likewise , when the 551 WTStv downregulated genes were compared to the 560 WTPhe downregulated genes , there was no commonality ( S3A Fig ) . However , we found that 249 ( 56% ) of the 441 WTPhe upregulated genes were downregulated in WTStv ( WTPhe up::WTStv down ) . Similarly , 263 ( 47% ) of the 560 downregulated WTPhe genes were upregulated in WTStv ( WTPhe down::WTStv up ) ( Figs 1A–1C and S3B and S2 Table ) . This observation indicates that not only do PDStv and PDPhe adults have distinct transcriptional profiles , but also that a subset of genes is oppositely regulated based on the experienced dauer-inducing stress . Due to their propensity for being up- or downregulated in postdauers compared to controls depending on the animals’ environmental history , we refer to these 512 genes as the “seesaw” genes . Using random simulations , we determined that the observed numbers of seesaw genes in WTPhe and WTStv gene sets are significantly greater than expected by chance ( 22 . 69-fold increase , p < 0 . 0001 and 9 . 38-fold increase , p < 0 . 0001 , respectively ) . To characterize the seesaw genes , we determined their expression patterns using several curation methods . First , we determined the enrichment of germline expressed seesaw genes by comparing their mRNA levels in wild-type animals to temperature-sensitive glp-4 ( bn2 ) ( WBGene00006936 ) mutants [17] . At the restrictive temperature , glp-4 ( bn2 ) animals are deficient in germline stem cell proliferation , resulting in the lack of a functional germ line and sterility [18] . The analysis revealed that the expression level of WTPhe up::WTStv down genes was substantially decreased in glp-4 ( bn2 ) compared to wild-type , indicating that these genes have germline enriched expression ( Fig 1D ) . In contrast , WTPhe down::WTStv up genes exhibited comparable expression levels in wild-type and glp-4 germline-deficient worms ( Fig 1E ) , suggesting that expression of these genes is enriched in somatic tissue . These trends in expression were further supported by examining the overlap of seesaw genes with tissue-enriched gene lists [19] , which revealed that 73% of WTPhe up::WTStv down seesaw genes were germline-enriched ( S3C and S3D Fig and S3 Table ) . Lastly , we used the gene function descriptions in WormBase ( WS253; [20] ) consisting of published observations , GO annotations , tissue expression data , modENCODE [21] , and Ensembl [22] information to further curate seesaw gene predicted function and spatial expression . Using these combined curation sources , we found an over-representation of genes associated with reproduction and embryonic development amongst the WTPhe up::WTStv down seesaw genes [23 , 24] ( Fig 1F and S4 Table ) . In contrast , the expression of WTPhe down::WTStv up seesaw genes are enriched in the intestine and nervous system , and have putative functions associated with innate immune response or anti-microbial defense [23–26] ( Fig 1G and S5 Table ) . Together , these results indicate that C . elegans animals maintain a cellular memory of their early life experience through the expression of sets of genes that are sensitive to environmental history and are distinct in their expression profiles and functional composition . CSR-1 ( WBGene00017641 ) is a C . elegans Argonaute protein that protects germline-expressed “self” transcripts from RNAi silencing through the organization of active chromatin domains and promotion of sense-oriented RNA polymerase II transcription genome-wide [27–32] . We previously showed that the CSR-1 RNAi pathway is required in early larval stages for dauer formation in response to starvation and high pheromone conditions and contributes to stable PD/CON changes in the chromatin state and gene expression for a subset of genes in adults [11 , 16] . In addition , we found 95% of the WTPhe up::WTStv down seesaw genes overlapped with a previously identified list of genes targeted by CSR-1 in the germ line [29] , suggesting the possibility that CSR-1 may play a prominent role in the regulation of postdauer transcriptional memory as a consequence of environmental history . In contrast , only 1 . 1% of genes in the WTPhe down::WTStv up dataset have been identified as CSR-1 targets . We therefore examined if the loss of CSR-1 affected the transcriptional changes observed in WTStv and WTPhe . Since csr-1 null mutants are sterile , we used a csr-1 hypomorph where sterility is partially rescued with a germline specific transgene [29] , and performed RNA-Seq on csr-1 control adults and starvation- and pheromone-induced csr-1 postdauer adults ( S1E–S1H and S2 Figs ) . Comparison of PD/CON gene expression levels for starvation and pheromone conditions ( csr-1Stv and csr-1Phe ) revealed that 48 genes were significantly upregulated and 224 genes were downregulated in csr-1Stv , while only 3 genes were significantly upregulated and 87 genes were downregulated in csr-1Phe ( Fig 2 and S1 Table ) . In addition , only 7 genes ( abu-14 ( WBGene00004174 ) , c47f8 . 7 ( WBGene00008163 ) , cut-2 ( WBGene00009983 ) , f53a9 . 8 ( WBGene00018731 ) , r02f11 . 1 ( WBGene00019839 ) , tts-1 ( WBGene00006650 ) , and tts-2 ( WBGene00006651 ) ) exhibited csr-1Phe down::csr-1Stv up seesaw patterns of gene expression , amongst which only one ( f53a9 . 8 ( WBGene00018731 ) ) was also found to seesaw in the WTPhe down::WTStv up dataset . None of the genes exhibited a csr-1Phe up::csr-1Stv down seesaw expression pattern ( Figs 2 and S3E , S3F and S2 Table ) . These results indicate that a functional CSR-1 RNAi pathway is required for the transcriptional memory of developmental history in starvation and high pheromone dauer-inducing conditions , including for the majority of WTPhe down::WTStv up genes that have not been previously identified as CSR-1 targets . We sought to verify the RNA-Seq results using qRT-PCR on a subset of genes from biologically independent samples of wild-type and csr-1 hypomorph strains . First , we examined the mRNA levels of 12 germline-specific , CSR-1-targeted genes that exhibited the WTPhe up::WTStv down seesaw pattern . Seven ( cye-1 ( WBGene00000871 ) , f45f2 . 10 ( WBGene00018482 ) , isw-1 ( WBGene00002169 ) , ifg-1 ( WBGene00002066 ) , cbd-1 ( WBGene00010351 ) , daz-1 ( WBGene00000935 ) , and lin-41 ( WBGene00003026 ) ) of the 12 genes ( 58% ) were validated in wild-type animals ( Figs 3 , S4C and S4E ) ; however , none of the wild-type seesaw patterns were validated in the csr-1 hypomorph ( Figs 3 and S5C ) . For the validated genes , the abrogation of seesaw gene expression in the csr-1 hypomorph was due to multiple effects on gene expression levels in postdauers and/or controls , including: PD/CON direction of gene expression change inverting in both conditions ( 29% ) ; both conditions exhibiting similar direction of change to either WTPhe ( 29% ) or WTStv ( 29% ) ; or one or both conditions no longer exhibiting a significant change in PD/CON mRNA levels ( 14% ) ( S4C and S5C Figs ) . Next , we sought to validate the mRNA levels of 12 soma-enriched , non-CSR-1-targeted genes that exhibited the WTPhe down::WTStv up seesaw patterns . We validated 10 ( ins-19 ( WBGene00002102 ) , mtl-1 ( WBGene00003473 ) , f55b11 . 4 ( WBGene00010086 ) , ttr-5 ( WBGene0000804 ) , fmi-1 ( WBGene00001475 ) , hsp-16 . 41 ( WBGene00002018 ) , f53a9 . 8 ( WBGene00018731 ) , r12e2 . 15 ( WBGene00020040 ) , spp-2 ( WBGene00004987 ) , and y51f10 . 7 ( WBGene00021768 ) ) out of the 12 genes ( 83% ) in wild-type samples ( Figs 4 , S4B and S4E ) . Consistent with our transcriptome data , the seesaw gene expression pattern for a majority of these genes were also dependent on CSR-1 , despite not being previously identified as targets of CSR-1 nor being germline-enriched . Only 2 genes , hsp-16 . 41 and r12e2 . 15 , retained a significant seesaw pattern in the csr-1 hypomorph; however , in both cases , the change in expression between csr-1Phe and csr-1Stv is opposite to the change observed in wild-type samples ( S4B and S5B Figs ) . For the remainder of the validated genes , they showed similar disruptions in expression in csr-1 hypomorph compared to wild-type as the germline-enriched genes: the PD/CON direction of gene expression inverting in both conditions ( 30% ) ; both conditions exhibited trends in gene expression similar to either WTPhe ( 20% ) or WTStv ( 10% ) ; or that one or both conditions no longer exhibited a significant change in PD/CON mRNA levels ( 20% ) . These results further confirm that CSR-1 plays a crucial role in mediating the PD/CON gene expression changes based on environmental history , regardless of whether the gene is a known CSR-1 target or not . Thus far , our implication of the CSR-1 RNAi pathway in the regulation of soma-enriched , non-CSR-1 target seesaw genes suggests the possibility that signals transported between cell and tissue types could contribute to the transcriptional memory of environmental history . In C . elegans , systemic RNAi spreads dsRNA throughout the animal and requires its main effector , the dsRNA importer , SID-1 [33 , 34] . To ascertain whether the transport of dsRNA is a mechanism eliciting the starvation- and pheromone-induced seesaw effect , we measured the PD/CON mRNA levels of genes in the sid-1 ( qt9 ) null mutant . Similar to the csr-1 hypomorph , both germline- and soma-enriched genes failed to exhibit a seesaw pattern in sid-1 adults ( Figs 3 , 4 , S4 and S7 ) . When we examined whether sid-1 was required for the seesaw gene expression for a specific dauer-inducing condition , we found that 47% of the validated genes exhibited the opposite direction of change in sid-1 compared to wild-type for the pheromone condition , compared to 20% showing a similar effect in starvation condition . Thus , SID-1 primarily contributes to seesaw gene expression profiles in somatic and germline tissues due to early life history of the pheromone condition . The observation that signals from the germ line can mediate somatic gene expression levels to affect adult lifespan is well-established in C . elegans , Drosophila , and mammals [35] . Since we have shown that the systemic RNAi effector , SID-1 , is playing a role in the transcriptional memory of environmental history , we interrogated whether signals exported from the germ line are necessary for the seesaw pattern in adult somatic tissue . To examine the potential role of the germ line in modulating the gene expression changes due to environmental history , we performed qRT-PCR to measure PD/CON mRNA levels for germline- and soma-enriched seesaw genes in a strain carrying the glp-4 ( bn2 ) allele , which lacks a functional germ line at the restrictive temperature [18] . As expected in animals lacking a germ line , the pheromone- and starvation-induced seesaw expression of the 12 germline-enriched genes was abolished in glp-4 ( bn2 ) adults grown at the restrictive temperature ( Figs 3 and S6C ) . Similarly , all but one ( f55b11 . 4 ) of the 13 soma-enriched genes also showed an elimination of the seesaw pattern in glp-4 ( bn2 ) adults ( S6B Fig ) . These results indicate that the seesaw pattern of gene expression , including genes that have enriched expression in the soma , requires a functional germ line . Again , to examine whether the germ line is required for the regulation of gene expression in specific dauer-inducing conditions , we examined whether PD/CON ratio of mRNA levels were affected for the pheromone or starvation conditions in glp-4 animals . For both germline and soma-enriched genes , we observed that 71% of the validated genes exhibited the opposite direction of change in expression in glp-4 compared to wild-type for the starvation condition , while only 24% exhibited this effect for the pheromone condition . Thus , a functional germ line is paramount for the programmed change in PD/CON mRNA levels in both the soma and germ line as a result of early life starvation . Furthermore , since SID-1 is not required for starvation-induced expression changes , these results indicate that the germline-dependent signal regulating somatic gene expression is not dsRNA . The significant excess of genes exhibiting seesaw patterns of differential expression and the ability of CSR-1 to modulate expression of non-target genes led us to investigate whether the inverse expression response to distinct dauer-inducing stresses may be a genome-wide phenomenon . This analysis revealed that a large proportion of genes whose PD/CON mRNA levels were not significantly seesawing by our original , stricter criteria still exhibited opposite changes in PD/CON mRNA levels with respect to the dauer-inducing stress . A significant inverse correlation in gene expression change was observed for 67 . 1% of genes ( 12 , 454 out of 18 , 570 genes sampled in both experiments ) in response to two dauer triggers ( R2 = 0 . 167 , p < 0 . 0001 ) ( Fig 5A , Q1 and Q3 ) . Consistent with our previous analysis , most CSR-1 targets ( 74 . 3% ) exhibited the WTPhe up::WTStv down pattern of expression ( Fig 5A , Q1 ) . Moreover , when we compared the transcriptome of csr-1Phe to csr-1Stv , we observed a pattern distinct from wild-type , particularly for genes that upregulated in WTPhe dataset ( Fig 5A and 5B , Q1 and Q4 ) . This analysis suggests that a majority of the genes in the genome , in addition to our identified set of seesaw genes , are subject to trends of CSR-1 dependent differential regulation in response to environmental history . A recent report found that genes that were similarly downregulated via RNAi-mediated chromatin remodeling in S . pombe during quiescence are located in clusters throughout the genome [36] . In C . elegans , thousands of protein-coding , germline-expressed genes are physically clustered in euchromatic domains that are established and maintained by the CSR-1 pathway [29 , 31 , 32] . To assess whether the physical location of C . elegans genes correlates with the observed transcriptome-wide seesaw patterns of gene regulation , we examined the location of CSR-1 target genes throughout the genome . First , examination of the chromosomal distribution of CSR-1 targets identified a highly significant enrichment on chromosomes I and III and paucity on chromosome V and sex chromosome X ( Fig 5C ) . Second , a striking bias that parallels the distribution of CSR-1 targets was observed for the genomic distribution of genes based on their response to the starvation condition . Genes that were downregulated in WTStv ( Q1 and Q2 ) were overrepresented on chromosomes I and III , and genes that are upregulated in WTStv ( Q3 and Q4 ) are overrepresented on chromosomes V and X ( Fig 5A and 5C ) . Expression patterns of genes in the pheromone condition did not correlate with their distribution across chromosomes beyond the seesaw relationship in expression responses observed in Q1 and Q3 ( Fig 5A ) . Thus , our analysis shows that genes with similar trends in expression patterns in starvation condition are non-randomly distributed across chromosomes , with CSR-1 targets overrepresented on the same chromosomes as genes that are downregulated in WTStv . In light of the marked chromosomal bias in expression patterns of genes in response to environmental history , we next investigated whether genes with a specific expression pattern exhibited spatial organization within chromosomes . Low recombination in the center of C . elegans chromosomes has resulted in extensive linkage disequilibrium and the operation of selection at the level of large haplotype blocks [37] , possibly facilitating the distribution of similarly regulated genes in clusters . A sliding window approach was used to assess the distribution of inversely regulated genes ( Q1 and Q3 ) exhibiting at least 2-fold change in PD/CON mRNA levels for pheromone and starvation conditions . This analysis confirmed a general enrichment of Q1 genes ( red lines ) in central regions of chromosomes I and III , and Q3 genes ( blue lines ) in the central regions of chromosomes II , V , and X ( Fig 5D ) . Further , to define the location of CSR-1 target genes relative to each other , we employed a genome-wide clustering algorithm using 4 , 191 CSR-1 targets identified in the germ line [29] in order to delineate CSR-1 “clusters . ” We found that 73% of CSR-1 targets mapped to 507 clusters ranging from 3 to 77 genes , with most ( 78% ) being between 3 to 10 genes long ( S6 Table ) . The number of clusters ( 1 . 33x; p < 0 . 0001 ) and the number of clustered CSR-1 targets ( 1 . 73x; p < 0 . 0001 ) both exceeded neutral expectations as defined by randomized gene order simulations . As would be expected , there was a highly significant enrichment of CSR-1 clusters on chromosomes I and III ( p = 0 . 0002 ) . Since germline-expressed genes and CSR-1 targets are enriched in operons [38 , 39] , we also examined the expression patterns of a defined set of C . elegans operons [40] . Interestingly , while 359 of the 901 known operons exhibit a WTPhe up::WTStv down directional change in gene expression , only 91 of the 512 ( 18% ) significant seesaw genes reside in operons , suggesting that operons alone do not account for our overarching genomic trends in gene expression . Thus , the co-localization of genes exhibiting similar seesaw trends in expression to particular chromosomes indicates that many of these loci would be simultaneously captured via genetic hitchhiking in the repeated selective sweeps that have shaped the C . elegans genome . We next questioned whether the significant seesaw changes in mRNA levels due to environmental history result in phenotypic consequences in adult animals . Due to the overlap between germline-specific genes and WTPhe up::WTStv down seesaw genes ( Figs 1D , 1F and S3D ) , we hypothesized that a reproductive phenotype could be an outcome of an animal’s environmental history . To determine whether seesaw gene expression affected the number of progeny produced by wild-type postdauer hermaphrodites , we quantified the brood size of control and postdauer adults that experienced either pheromone or starvation . Consistent with our previous reports , wild-type PDPhe had an increased brood size compared to wild-type CONPhe ( Fig 6A; [10 , 11] ) . In contrast , wild-type PDStv had a reduced brood size compared to wild-type CONStv ( Fig 6B ) . This indicates that the gene expression changes resulting from environmental history have significant consequences with respect to the fitness of C . elegans animals . Since we observed that seesaw changes in gene expression are dependent on mechanisms involving CSR-1 , GLP-4 , and SID-1 functions ( Figs 3 , 4 , S4 , S5 , S6 and S7 ) , we asked whether the same mechanisms affected the fecundity differences observed in pheromone- or starvation-induced wild-type postdauer and control adults . Even with a reduced brood size , the number of surviving progeny produced by the csr-1 hypomorph remained slightly higher in PDPhe compared to CONPhe ( Fig 6A ) . However , the decrease in brood size between postdauer and control adults was abrogated in csr-1Stv ( Fig 6B ) . We also observed that glp-4 ( bn2 ) adults grown at the permissive temperature no longer exhibited a decrease in postdauer brood size in the starvation condition ( Fig 6B ) , but continued to show an increased number of progeny in the pheromone condition ( Fig 6A ) . Recently , GLP-4 was shown to be expressed in the intestine and somatic gonad in addition to the germ line [41]; thus , our observed glp-4 PD/CON brood size phenotype at the permissive temperature suggests that GLP-4 function in the intestine or somatic gonad is contributing to the regulation of the starvation program ( see Discussion ) . In contrast , the brood size results of the sid-1 ( qt9 ) strain were opposite to those of the glp-4 ( bn2 ) strain . The increase in PDPhe/CONPhe brood size observed for wild-type adults was abolished in sid-1 ( qt9 ) adults ( Fig 6A ) ; however , the decrease in PDStv/CONStv brood size was also observed in sid-1 adults ( Fig 6B ) . Together , these results are in accordance with our gene expression analyses where changes in mRNA levels and the resulting phenotypic plasticity due to pheromone conditions are a result of SID-1 function in the soma , whereas starvation-induced changes are mediated by unknown signals from the germ line . Next , we sought to further characterize the developmental differences in the germ line that could result in altered PD/CON brood sizes . Since reproduction in self-fertilizing C . elegans hermaphrodites is sperm-limited [42] , we asked whether the fecundity differences in PD/CON was associated with changes in mRNA levels of genes regulating germ line mitotic proliferation and the onset of meiosis during spermatogenesis . Hermaphrodites possess two gonad arms , each of which is capped by a distal tip cell ( DTC ) that maintains the germline stem cell niche through GLP-1/Notch ( WBGene00001609 ) signaling ( mitotic zone , MZ ) [43 , 44] . As cells divide in the mitotic proliferative zone , the most proximal cells begin to express the RNA binding protein , GLD-1 ( WBGene00001595 ) , which promotes entry into meiosis ( transition zone , TZ ) [45 , 46] . Since hermaphrodites produce all their sperm during the larval L4 stage , we hypothesized that modulation of these genes as a result of environmental history could potentially alter the number of sperm in hermaphrodite animals . To test our hypothesis , we first examined the expression of WTPhe up::WTStv down seesaw gene , gld-1 , using a gld-1::gfp transgene expressed in the germ line . To compare developmentally synchronized animals , we examined GFP levels of postdauer and control animals that experienced either pheromone or starvation conditions and exhibited the vulva morphology characteristic of L3 , L4 . 1 , and L4 . 4 larval animals [47 , 48] , at which times the mitotic and transition zones are evident [43] . Although we were unable to validate the seesaw changes in gld-1 expression using qRT-PCR ( S4C Fig ) , we observed that GFP levels in the germ line were significantly increased in PDPhe larva compared to CONPhe larva at all stages , consistent with our RNA-Seq results ( S8A and S8B Fig and S11 Table ) . In contrast , we detected no significant change in GFP levels in PDStv compared to CONStv in animals exhibiting L3 vulval morphology , but a surprising increase in GFP levels in PDStv compared to CONStv for the L4 . 1 and L4 . 1 stages similar to the pheromone condition ( S8A and S8C Fig and S11 Table ) . In Notch signaling mutants , GLD-1 levels remain stable in the transition zone [49] , and high levels of GLD-1 are sufficient to drive germline stem cells into meiosis , even in the presence of Notch signaling [50] . Furthermore , we observed that the area of the gonad arms was also significantly different due to environmental history , such that postdauer gonad arm area was increased or decreased compared to controls in pheromone and starvation conditions , respectively , for all larval stages ( S8B and S8C Fig and S11 Table ) . Together , these results suggest that the changes in GLD-1 levels we observed in Phe larva likely reflects alterations in the numbers of cells entering meiosis and not changes in GLD-1 expression in individual cells . To further test our hypothesis , we DAPI-stained larva and counted the number of cell rows per gonad arm in control and postdauer animals that exhibited the characteristic L3 , L4 . 1 , and L4 . 4 vulva morphology and experienced either pheromone or starvation conditions ( S9 and S10 Figs ) [48 , 51] . If the sperm to oocyte developmental switch remains constant between postdauer and control animals [52] , we would predict that postdauer larva that experienced pheromone or starvation conditions to begin germline proliferation earlier or later than control larva , respectively , resulting in altered numbers of sperm available for self-fertilization in adult hermaphrodites . Indeed , we observed a significant increase for PDPhe/CONPhe total cell rows and decrease for PDStv/CONStv total cell rows for all developmental stages examined ( Fig 6C and 6D and S12 Table ) . Interestingly , at the L3 stage , we observed a similar number of cell rows in the mitotic zone , but different numbers of cell rows in the transition zone for postdauers compared to controls in both conditions , suggesting that proliferation begins earlier or later in postdauer animals that experienced pheromone or starvation conditions , respectively ( Fig 6C and 6D and S12 Table ) . As the animals aged and germ lines expanded , we observed that the numbers of cell rows in a particular region of the gonad , such as the transition zone or pachytene zone , varied between postdauer and control animals differently depending on the stage ( L4 . 1 or L4 . 4 ) and environmental condition ( pheromone or starvation ) . This result likely reflects the different mechanisms regulating pheromone and starvation gene expression changes . As an additional control , we also counted the number of cells in the spermatheca [53] , which is a part of the somatic gonad , and found that the number of cells are similar for all the L4 . 4 populations as expected ( Fig 6C and 6D and S12 Table ) . This result indicates that germline development , and the onset of germline proliferation , can be uncoupled from somatic gonad development , resulting in postdauer germ lines that are “older” or “younger” compared to their control counterparts . Based on our RNA-Seq data in adults ( Fig 1A and 1B ) , these developmental trends seem to persist from the L3 stage into adulthood to result in the WTPhe up::WTStv changes in expression of germline-enriched genes . Together , these results are consistent with the model that the onset of germline proliferation during L3 larval stages is determined by environmental and developmental history , resulting in altered sperm number and brood size in adults . Furthermore , we sought to identify genes in addition to gld-1 that may contribute to altered germline development and spermatogenesis due to environmental experience . Using sperm transcriptome and proteome datasets [54] , we found significant overlaps between the sperm transcriptome and proteome with the WTPhe up:: WTStv down ( p < 0 . 0001 and p = 0 . 002 , respectively; two-tailed Fisher’s exact test ) and WTPhe down:: WTStv up ( p < 0 . 0001 and p < 0 . 0001 , respectively; two-tailed Fisher’s exact test ) seesaw genes . GO term analyses revealed significant functional distinctions between the genes encoding protein components of sperm ( “sperm genes” ) overlapping with the two classes of seesaw genes . Sperm genes associated with the WTPhe up:: WTStv down seesaw genes were enriched for reproduction , genitalia development , oogenesis , and spermatogenesis , while the sperm genes overlapping with the WTPhe down:: WTStv up seesaw genes were devoid of reproduction-related functions and were instead richly affiliated with the cuticle and collagen ( S7 Table ) . Additional experimentation will be required to determine if the changes in sperm genes are causal to , or result from , changes in germline development in larva . In sum , these results indicate that the fecundity differences resulting from distinct life histories may have a direct relationship with the differential expression of sperm-related genes with diverging functionalities . Taken together , we posit a model whereby different “programs” regulate global changes in gene expression leading to distinct reproductive phenotypes . In animals that experience early life high pheromone condition , the changes in PD/CON mRNA levels in the germ line and soma are dependent on SID-1 function in the soma . In contrast , in postdauer animals that experience early life starvation condition , the observed gene expression changes in the germline and the soma are not dependent on SID-1 , but are instead dependent on an unidentified signal ( s ) from a functional germ line . Our data indicate that these two programs maintain a functional balance within the animal , such that when one program is disrupted by mutation ( e . g . glp-4 ( bn2 ) mutant ) , the animal exhibits the gene expression and reproductive phenotype of the alternate program ( Figs 3 , 4 , 6 , S4 , S5 , S6 and S7 ) . In addition , both programs are dependent on the CSR-1 RNAi pathway for these chromosomally-regulated gene expression differences to result in reproductive plasticity of adult animals ( Fig 7 ) . Moreover , we provide evidence that genes with similar trends in expression levels in response to pheromone and starvation conditions are co-localized on specific chromosomes , and may have contributed to the evolution of the C . elegans genome .
Our results support a model where at least two distinct life history programs , the pheromone and starvation programs ( Fig 7A ) , are orchestrated in postdauers through the exchange of signals between the germ line and the soma ( Fig 7B ) . An intriguing question is what the candidate somatic and germline signals regulating the PD/CON programs might be . In the pheromone program , the requirement for SID-1 strongly implicates dsRNAs as mediators of gene expression changes in the soma and the germ line . Endogenous dsRNAs could potentially be processed into , or stimulate the production of , siRNAs that associate with CSR-1 in specific tissue types , analogous to how exogenous dsRNA results in target specific silencing by Argonautes such as NRDE-3 ( WBGene00019862 ) [59] . Since one of the proposed functions of CSR-1 is the establishment and maintenance of euchromatic chromatin states associated with target genes , and we have shown that CSR-1 targets exhibiting similar gene expression patterns cluster at the chromosomal level ( Fig 5D ) , it is reasonable to propose that CSR-1 would also modify chromatin states due to pheromone and starvation in a tissue-specific manner . Indeed , we previously showed that changes in the levels of two histone modifications associated with euchromatin in PDPhe/CONPhe are dependent on CSR-1 [11] . These altered chromatin states could potentially spread along chromosomal regions including non-CSR-1 target genes , resulting chromosomal-biased changes in gene expression genome-wide ( Fig 5 ) . In contrast , the starvation program is independent of SID-1 but dependent on the germ line ( Fig 7B ) . One candidate mechanism for regulating the starvation-induced cellular memory is endocrine signaling , which plays a key role in lifespan and stress response in worms , flies , and mammals [60] . Effectors of endocrine signaling in C . elegans include the conserved FOXO transcription factor DAF-16 ( WBGene00000912 ) and the nuclear hormone receptor DAF-12 ( WBGene00000908 ) , both of which function downstream of the insulin/IGF-1 signaling ( IIS ) pathway [61] . In C . elegans , Drosophila , and mice , removal or alteration of the germ line results in somatic aging phenotypes [60] . Nematodes with defective gonads exhibit extended lifespans that are dependent on DAF-16 activity in the intestine , as well as DAF-12 activity in the somatic gonad [60 , 62] , suggesting that the regulation of a somatic phenotype is mediated by germline signaling . Recent reports provide further evidence of such crosstalk: the chemotaxis response to the odorant diacetyl in C . elegans is dependent on DAF-16 and germline proliferation [63]; cold tolerance in worms is dependent on a feedback mechanism involving IIS signaling , temperature-sensing neurons , the intestine , and sperm cells [64] while a thermosensory neuronal circuit promotes longevity at warm temperatures by engaging endocrine signaling [65]; and the transgenerational lifespan extension in a mutant strain with defective chromatin remodeling is dependent on a germ line to soma signaling mechanism modulated by DAF-12 [66] . Therefore , endocrine signaling is an attractive contender with which distinct life history trajectories program a postdauer animal in response to early life starvation . Physical co-localization of genes with correlated expression patterns is widespread in eukaryotes [67] and is pronounced amongst genes expressed in the testis [68] and those encoding protein components of sperm [69] . Consistent with the efficient co-regulation of neighboring genes , we observed an excess of clustered genes in the central portions of chromosomes I , II , III , V , and X that exhibited WTPhe up::WTStv down and WTPhe down::WTStv up trends in gene expression changes ( Fig 5A and 5D ) . Given the breadth of syntenic conservation between C . elegans and C . briggsae [70] , it would appear likely that this genomic architecture arose in a common ancestor and may have contributed to reproductive fitness in what was likely to be an out-crossing species . Comparative genomic analyses will thus be essential in reconstructing the evolutionary history of PD reproductive investments using C . briggsae , which also shares a conserved CSR-1 RNAi pathway [38] . However , it is also noteworthy that the non-random organization of differentially expressed genes identified in this study overlaps substantively with regions of the C . elegans genome that experience limited recombination and large-scale selective sweeps [37] . Population genetic models predict that recurrent selective sweeps in linked regions would favor the establishment of adaptive “supergenes” over biologically realistic time scales [71 , 72]; thus , we speculate that this process may have contributed to the expansion of regions harboring genes with coherent reproductive functions during C . elegans evolution [37 , 73 , 74] .
The nematode strains are Bristol wild type strain N2 , WM193 csr-1 ( tm892 ) IV; neIs20 [pie-1::3xFLAG::csr-1 + unc-119 ( + ) ] , SS104 glp-4 ( bn2 ) I , HC196 sid-1 ( qt9 ) V , and BS1080 ozIs5 [gld-1::gfp/flag , pMMO16 ( unc-119 ( + ) ) ] I . Worms were maintained using standard methods on Nematode Growth Medium ( NGM ) plates seeded with Escherichia coli OP50 at 20°C or at 15°C ( for SS104 glp-4 ( bn2 ) I ) [75] . To collect PDPhe , we used an egg white plate procedure described previously [76] . To obtain PDStv , well-fed worms were transferred to seeded NGM plates and monitored until the E . coli OP50 food was depleted and the plates were populated with dauers . See Supplemental Experimental Procedures for details . Ten L4 larvae were singled onto seeded NGM plates and transferred daily onto fresh plates until egg laying ceased . Only surviving progeny were counted . At least three independent biological replicates were conducted . Statistical significance was determined using GraphPad Prism v . 7 . Total RNA extraction was done using TRIzol Reagent ( Life Technologies ) . Two biological independent RNA-Seq libraries for a strain and a condition were prepared using the NEBNext mRNA Library Prep Master Mix Set for Illumina ( NEB ) . Data analysis was conducted on the CLC Genomics Workbench v . 8 . 5 ( Qiagen ) with differential expression calculated using EdgeR [77] . Detailed procedures are described in Supplemental Experimental Procedures . GO terms were analyzed using DAVID ( Database for Annotation , Visualization and Integrated Discovery ) v . 6 . 7 [78] . The expected frequency of seesaw genes was investigated using a randomized approach whereby sets of genes , equal in number to the observed sets of differentially expressed genes under each condition , were selected from the whole genome set , without replacement , and their overlap assessed . The expected number of overlapping “seesaw” genes and the significance of the observed numbers were directly determined from the simulated distributions , based on 10 , 000 simulations . qRT-PCR was done using samples collected from three independent biological replicates . S8 Table lists the primer sequences . Statistical analysis was done using GraphPad Prism v . 7 . Detailed procedures can be found in the Supplemental Experimental Procedures . The Global Landscape Clustering ( GLC ) algorithm ( Borziak et al . , manuscript in preparation ) was used to identify maximal sets co-localized genes that share a specific attribute , such as being CSR-1 targets . See Supplemental Experimental Procedures for a detailed description . Significance of clustering was directly determined from the simulated distributions , based on 10 , 000 simulations . Seesaw gene enrichment sliding window analyses were conducted on the EdgeR generated fold changes using gene position information based on the WormBase ( WS235 ) annotations , using only coding transcripts . Sliding windows of 2 . 5 megabase pairs with 500 kilobase pair intervals were used . Seesaw genes were defined as those showing the inverse directional change of at least 2-fold , regardless of significance . Enrichment was calculated against the genome-wide average . Genome-wide gene expression graphs were generated using the log2 transformed EdgeR fold change values . Significance of correlation between experiments was calculated based on the Pearson's correlation between samples , where degrees of freedom equals ( the number of genes in the plot– 1 ) . Enrichment of directional change across chromosomes was calculated using χ2 test with Yates' correction against the remaining chromosomes . Detailed collection methods for BS1080 PD and CON larva in Phe and Stv conditions are described in S1 Text . For the Phe conditions , control and dauer larvae were collected using water or crude pheromone , respectively , on dauer formation plates as previously described [79; 80] . Images of larva exhibiting vulva morphology characteristic of the L3 , L4 . 1 , and L4 . 4 larval stages were analyzed for GLD-1::GFP expression and gonad area using ImageJ software ( NIH ) . These same worms were next used for germ cell row counts using a standard whole worm DAPI staining protocol [81] . The size of mitotic zone , transition zone , pachytene zone was determined based on the germ cell nuclear morphology [82] . Statistical significance between CON and PD samples was determined using Student’s t-test . The accession number for the high-throughput sequencing data reported in this study is GSE92954 . | Environmental stress during early development in animals can have profound effects on adult behavior and physiology due to programmed changes in gene expression . However , whether different stresses result in distinct changes in traits that allow stressed animals to better survive and reproduce in future adverse conditions is largely unknown . Using the animal model system , C . elegans , we show that adults that experienced starvation exhibit opposite ( “seesaw” ) genome-wide gene expression changes compared to adults that experienced crowding , and are distinct from animals that experienced favorable conditions . Genes that are similarly up- or downregulated due to either starvation or crowding are located in clusters on the same chromosomes . Importantly , these gene expression changes of differently-stressed animals result in corresponding changes in progeny number , a life history trait of evolutionary significance . These distinct gene expression programs require different signaling pathways that communicate across somatic and germline tissue types . Thus , different environmental stresses experienced early in development induce distinct signaling mechanisms to result in changes in gene expression and reproduction in adults , and likely contribute to increased survival in future adverse conditions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"gonads",
"caenorhabditis",
"gene",
"regulation",
"animals",
"animal",
"models",
"germ",
"cells",
"developmental",
"biology",
"caenorhabditis",
"elegans",
"model",
"organisms",
"experimental",
"organism",
"systems",
"sperm",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"gene",
"expression",
"life",
"cycles",
"biochemistry",
"eukaryota",
"anatomy",
"cell",
"biology",
"phenotypes",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"pheromones",
"cellular",
"types",
"larvae",
"organisms",
"genital",
"anatomy"
] | 2018 | Early experiences mediate distinct adult gene expression and reproductive programs in Caenorhabditis elegans |
Recent epidemics of Zika virus ( ZIKV ) in the Pacific and the Americas have highlighted its potential as an emerging pathogen of global importance . Both Aedes ( Ae . ) aegypti and Ae . albopictus are known to transmit ZIKV but variable vector competence has been observed between mosquito populations from different geographical regions and different virus strains . Since Australia remains at risk of ZIKV introduction , we evaluated the vector competence of local Ae . aegypti and Ae . albopictus for a Brazilian epidemic ZIKV strain . In addition , we evaluated the impact of daily temperature fluctuations around a mean of 28°C on ZIKV transmission and extrinsic incubation period . Mosquitoes were orally challenged with a Brazilian ZIKV strain ( 8 . 8 log CCID50/ml ) and maintained at either 28°C constant or fluctuating temperature conditions . At 3 , 7 and 14 days post-infection ( dpi ) , ZIKV RNA copies were quantified in mosquito bodies , as well as wings and legs , using qRT-PCR , while virus antigen in saliva ( a proxy for transmission ) was detected using a cell culture ELISA . Despite high body and disseminated infection rates in both vectors , the transmission rates of ZIKV in saliva of Ae . aegypti ( 50–60% ) were significantly higher than in Ae . albopictus ( 10% ) at 14 dpi . Both species supported a high viral load in bodies , with no significant differences between constant and fluctuating temperature conditions . However , a significant difference in viral load in wings and legs between species was observed , with higher titres in Ae . aegypti maintained at constant temperature conditions . For ZIKV transmission to occur in Ae . aegypti , a disseminated virus load threshold of 7 . 59 log10 copies had to be reached . Australian Ae . aegypti are better able to transmit a Brazilian ZIKV strain than Ae . albopictus . The results are in agreement with the global consensus that Ae . aegypti is the major vector of ZIKV .
Over the past decade , Zika virus ( ZIKV ) has caused unprecedented epidemics in the Western Pacific and the Americas . ZIKV is a mosquito-borne , single-stranded RNA virus that belongs to the Flavivirus genus within the Flaviviridae family [1] . First discovered in Uganda in 1947 [2] , ZIKV spread from equatorial Africa into Asia in 1960 , producing two main genotypes , the African and Asian lineages [3 , 4] . Major epidemics of ZIKV have occurred on Yap Island , Federated State of Micronesia [5 , 6] , French Polynesia [7] , some islands in the south and south-west Pacific region [8–11] , and most recently Latin America [12–15] . Although 80% of ZIKV infections remain asymptomatic or cause a mild febrile illness [5 , 16] , recent epidemics have seen more severe disease manifestations , such as microcephaly and central nervous malformations in neonates [17 , 18] , and Guillain-Barré syndrome in adults [19 , 20] . Although ZIKV can be transmitted sexually [21] , through blood transfusion [22] , and from mother-to-child [23] , humans are primarily infected through the bite of infected Aedes ( Ae . ) mosquito species [24–28] . In Africa , where it was first isolated from Ae . africanus [29] , ZIKV is mainly transmitted by sylvatic Aedes mosquitoes ( Ae . furcifer , Ae . luteocephalus , Ae . taylori , Ae . opok , Ae dalzieli ) [24 , 30] . Initial evidence for human infections implicated Ae . aegypti in the urban transmission of ZIKV in Africa [26 , 31 , 32] . In Asia [4 , 33 , 34] and the Americas [35–37] , Ae . aegypti is considered the main vector for human ZIKV transmission . Although Ae . hensilli was suspected to be responsible for ZIKV transmission during the Yap outbreak [27] , Ae . aegypti and Ae . polynesiensis were the main vectors in the French Polynesian outbreak [28] . Recent evidence of vertical transmission of ZIKV in field-collected eggs of Ae . aegypti from Brazil suggests that , in endemic areas , virus may also be maintained in drought resistant eggs [38] . Ae . albopictus is an invasive vector which has colonized most of the tropics and subtropics , as well as more temperate regions of the United States and Europe [39] . The high vectorial capacity of Ae . albopictus for various arboviruses [40–42] places any area colonized by this species at risk of local ZIKV transmission [43 , 44] . Considerable variation in ZIKV vector competence , similar to that reported for DENV [45–47] , has been observed in both Ae . aegypti and Ae . albopictus from across the globe [25 , 48–54] . The transmission efficiency of ZIKV is governed by interactions between mosquito strain [25 , 53] and virus genotype/strain [45 , 53 , 55–57] . This variability underscores the importance of evaluating the vector competence of local mosquito populations for ZIKV . Australia remains at risk of ZIKV introduction due to its close proximity to the Western Pacific , the presence of competent strains of Ae . aegypti in Queensland [58 , 59] and Ae . albopictus in the Torres Strait [48 , 60] , and favourable climatic conditions for transmission [61] . Despite 51 reports of imported cases of ZIKV since 2014 ( Queensland Government , Australia , accessed 8 October 2018 ) , Australia has not yet reported autochthonous transmission . Previous studies have reported the vector competence of Australian Ae . aegypti for African , Cambodian and Western Pacific strains [48 , 58 , 59] and Ae . albopictus ( Torres Strait islands ) for Cambodian ZIKV [48 , 58 , 59] . These studies demonstrated that Australian mosquito strains can be infected and transmit ZIKV; however , large heterogeneity has been observed in the susceptibility of mosquitoes to infection , which may be associated with the origin of the virus strains . There have been no investigations of the vector competence of Australian strains to isolates of ZIKV from South America , despite the continent recording the largest epidemics with a high prevalence of the most severe ZIKV disease manifestations [62] . To assess the public health risk imposed by ZIKV to Australia , we determined the vector competence of local populations of Ae . aegypti and Ae . albopictus ( Torres Strait Islands ) for a strain of ZIKV isolated from a febrile patient from Paraiba state , at the centre of the 2015/2016 Brazil epidemic . In addition to maintaining infected mosquitoes under a standard constant temperature regime , we also used a fluctuating diurnal temperature range ( DTR ) . Our study indicates that Ae . aegypti has higher relative vector competence than Ae . albopictus , which may be mediated by a salivary gland barrier to virus transmission in Ae . albopictus for this ZIKV strain .
Ae . aegypti eggs were obtained from a colony established from Wolbachia-free eggs collected from Innisfail , north Queensland , in April 2016 and subsequently maintained in the QIMR Berghofer insectary at 27°C , 70% relative humidity [RH] and 12:12 h lighting with 30 min crepuscular periods . Ae . albopictus eggs were obtained from a colony established from eggs collected on Hammond Island , Torres Strait , Australia , in July 2014 and subsequently maintained in the QIMR Berghofer insectary . Eggs of both colonies were hatched and larvae were reared at a density of 400 individuals in 3 L of rainwater . Larvae were provided ground TetraMin Tropical Flakes fish food ( Tetra , Melle , Germany ) ad libitum . Pupae were transferred to a container of rainwater inside a 30 × 30 × 30 cm cage ( BugDorm , MegaView Science Education Services Co . , Taichung , Taiwan ) for adult emergence . Adult mosquitoes were provided with 10% sucrose solution on cotton wool pledgets . The Brazilian ZIKV strain KU365780 [63] used in this study was isolated from a febrile patient in Joao Pessoa City , Paraiba State , Brazil , 18-05-2015 ( provided by the Evandro Chagas Institute , Brazil ) . Viruses were propagated in C6/36 Ae . albopictus cells , maintained at 28°C in RPMI-1640 medium ( Sigma Life Sciences , USA ) supplemented with 10% fetal bovine serum . Following three passages in C6/36 cells , virus stocks were concentrated using Ultracel-100k filters ( Amicon , Tullagreen , Cork Ireland ) and frozen once at -80°C until further use . Virus stocks were titrated using a Cell Culture Enzyme-linked Immunosorbant Assay ( CCELISA ) based on the method of Broom et al . [64] . Ten-fold serial dilutions of virus stocks were inoculated on C6/36 ( Ae . albopictus ) cells grown in RPMI 1640 supplemented with L-glutamine , 5% heat denatured FBS , 1% penicillin/streptomycin ( Gibco Life Technologies , USA ) and maintained at 28°C , 5% CO2 for 5 days . Monolayers were incubated at 28°C , 5% CO2 for 5 days , and cells fixed at -20°C for 1 h in 80% acetone/20% phosphate-buffered saline ( PBS ) . Plates were air-dried and antigen was detected using a 4G4 anti-Flavivirus NS1 monoclonal antibody hybridoma supernatant ( 1:40 in PBS-Tween ) , Horseradish peroxidase ( HRP- ) conjugated goat anti-mouse polyclonal antibody ( DAKO , Carpinteria , CA , USA ) ( 1:2000 in PBS-Tween ) , and 3 , 3′ , 5 , 5′-Tetramethylbenzidine ( TMB ) Liquid Substrate System for Membranes ( Sigma-Aldrich , St . Louis , MO , USA ) . Staining was observed using an inverted microscope , and cell monolayers that stained blue were scored positive for infection . The 50% cell culture infectious dose ( CCID50 ) was determined from titration end points as previously described [65] and expressed as the CCID50/ml in C6/36 cells . Mosquito infection with ZIKV occurred in a Biosafety Level 3 insectary at QIMR Berghofer . An artificial membrane feeding apparatus , fitted with a porcine intestinal membrane , was used to orally challenge adult females ( 3–5 day old ) with a mixture of defibrinated sheep blood ( Life Technologies , Mulgrave , VIC , Australia ) and virus at a final concentration of 8 . 8 log CCID50/ml ( C6/36 cells ) for 1 h . Following ZIKV inoculation , mosquitoes were maintained in environmental growth chambers ( Panasonic ) , with either a constant temperature program set to 28°C or a fluctuating ( cyclical ) temperature program ( 24 . 5–32°C ) around a mean of 28°C [66] ( Fig 1 ) . The temperature treatments are referred to here as “constant” and “fluctuating” , respectively . For both treatments , RH was set to 75% and a 12:12 h day:night lighting program was used . Twenty mosquitoes were harvested at each of three time points ( 3 , 7 and 14 days ) post blood feeding by anaesthetizing the insects with CO2 on ice before removing legs and wings . Mosquitoes were gently transferred by their antennae to a glass plate and immobilized on double-sided sticky tape . Saliva was collected by placing the proboscis of each mosquito into a 200 μl pipette tip containing 10 μl of 10% FBS and 10% sugar solution . The insertion of the proboscis into the salivation solution was performed under a dissecting scope and peristaltic movement of the abdomen observed to indicate salivation . Mosquitoes were allowed to salivate for 20 min , after which the contents of the pipette tip were then expelled into a microtube and preserved at -80°C . Nucleic acids were extracted from individual mosquito bodies or body parts using the High Pure Viral Nucleic Acid Kit ( Roche Diagnostics , Mannheim , Germany ) according to the manufacturer’s protocol . Briefly , 200 μl Binding Buffer/poly A solution was added to each 2 ml screw cap vial containing the individual mosquito body or body parts . The samples were homogenized by shaking the tubes , containing zirconium silica glass beads ( Daintree Scientific , St Helens , TAS , Australia ) , using a MiniBeadbeater-96 ( Biospec Products , Bartlesville , Oklahoma , USA ) for 90 s . Following the addition of 50 μl of Proteinase K , nucleic acids were extracted as per the manufacturer’s instructions , and eluted in 50 μl of RNAse/DNase-free Ultrapure water ( Invitrogen ) . Samples were frozen at -80°C until further analysis . To determine the absolute number of ZIKV copies in each mosquito body or body part , a control plasmid , containing a cloned copy of the targeted ZIKV gene fragment ( nucleotides 835 to 911 , Genbank accession number EU545988 ) , was constructed . Viral RNA was extracted using the QIAamp Viral RNA Mini Kit ( Qiagen , Germany ) , and cDNA synthesized using the SuperScript III Reverse Transcriptase kit ( Invitrogen , Thermo Fisher Scientific , USA ) according to the manufacturer’s protocol . The targeted ZIKV fragment was amplified using CloneAmp HiFi PCR Premix ( Takara , Clontech Laboratories , USA ) , and cloned into the pUC19 plasmid vector ( Genscript , New Jersey , United States ) using the In-Fusion Cloning Kit ( Takara , Clontech Laboratories , USA ) as described by the manufacturer . The presence of the insert DNA was confirmed by nucleotide sequencing . For quantitative RT-PCR analysis , the plasmid was linearized by EcoRI ( Promega , USA ) and purified using the Nucleospin Gel and PCR clean-up kit ( Macherey-Nagel , Germany ) . The concentration and purity of the linearized plasmid DNA was determined using the NanoDrop Lite spectrophotometer ( Thermo Fisher Scientific , USA ) . The plasmid copy number was calculated based on the measured DNA concentration and its molecular weight . Plasmid DNA concentrations were confirmed prior to the preparation of a 10-fold serial dilution from 3×107 to 3×102 copies/μl and run in parallel with the samples in all qRT-PCRs . ZIKV RNA from mosquitoes was amplified by one-step qRT-PCR using the TaqMan Fast Virus 1-Step Master Mix ( Applied Biosystems , USA ) according to the manufacturer’s protocol , in a Rotor-Gene 6000 Version 1 . 7 . 87 system ( Corbett Research , NSW , Australia ) . Primers and probe used in this study have previously been described [6] and were synthesized at Macrogen , Korea . The probe was labelled with FAM and BHQ1 at the 5′ and 3′ ends , respectively . The 20 μl reaction mixture consisted of 1 μl extracted sample , 4 × TaqMan Fast Virus 1-Step Master Mix , 400 nM of each primer , 250 nM of probe and Ultrapure water ( Invitrogen , Thermo Fisher Scientific , USA ) . Reactions were run in triplicate , and a 10-fold serial dilution of linearized control plasmid DNA ( 3×107 to 3×102 copies/μl ) , as well as negative controls ( without template ) , were included in each run . The following thermal profile was used: a single cycle of reverse transcription for 5 min at 50°C , reverse transcriptase inactivation and DNA polymerase activation at 95°C for 20 s followed by 40 cycles of 95°C for 3 s and 60°C for 30 s ( annealing-extension step ) . Data were analysed and quantified using the Rotor-Gene 6000 Version 1 . 7 . 87 software ( Corbett Research , NSW , Australia ) . To calculate the total number of ZIKV RNA copies present in each mosquito body or body part , the measured ZIKV RNA copy numbers in 1μl were multiplied by the elution volume ( i . e . , 50 μl ) . Samples were scored positive for virus if ZIKV amplification occurred in at least two technical replicates and the number of copies was above the limit of detection of the standard curve . Samples in which ZIKV failed to amplify were classified as negative . The presence of mosquito nucleic acid in negative samples was verified by amplification of the housekeeping genes RpS17 ( Ae . aegypti; Genbank accession number AY927787 . 2 ) or RpS7 ( Ae . albopictus; Genbank accession number XM_019671546 ) . qRT-PCR for house-keeping genes were performed using the SuperScript III SYBR Green One-Step qRT-PCR kit ( Invitrogen , Life Technologies , USA ) as per manufacturer’s recommendations . The reactions were performed in a 10 μL total volume containing SuperScript III RT/Platinum Taq Mix , 2 × SYBR Green Reaction Mix , 200 nM of each RpS17/RpS7 primer ( RpS17 F: 5′-TCCGTGGTATCTCCATCAAGCT-3′ , R: 5′-CACTTCCGGCACGTAGTTGTC-3′; RpS7 F: 5’-CTCTGACCGCTGTGTACGAT-3’ , R: 5’-CAATGGTGGTCTGCTGGTTC-3’ ) , 1 μL of extracted sample and Ultrapure water ( Invitrogen , Life Technologies , USA ) . Cycling was performed using the Rotor-Gene 6000 system ( Corbett Research , NSW , Australia ) with 1 cycle at 50°C for 5 min and 95°C for 2 min , followed by 40 amplification cycles of 95°C for 15 s , 60°C for 30 s and 72°C for 20 s . Melt curve analysis was performed to analyse the specificity of the reaction . The presence of infectious virus in blood meals and in mosquito saliva samples was determined using CCELISA as described above , with the following modifications . Blood meals were titrated by 10-fold serial dilution on C6/36 ( Ae . albopictus ) cells grown in RPMI 1640 supplemented with L-glutamine , 5% heat denatured FBS , 1% penicillin/streptomycin ( Gibco Life Technologies , USA ) and maintained at 28°C , 5% CO2 for 5 days . Mosquito saliva samples were diluted 1:25 in the media described above , supplemented with 0 . 1% Gibco Amphotericin B ( ThermoFisher Scientific , Waltham , MA USA ) , and used to inoculate duplicate monolayers of C6/36 cells ( ~90% confluent ) . Samples were then fixed and stained as described above . The legs and wings were removed from mosquitoes and the remaining body was fixed in 4% paraformaldehyde and 0 . 5% Triton X overnight before mosquitoes were transferred to 70% ethanol . Mosquitoes were dehydrated and embedded in paraffin using standard procedures . Paraffin sections ( 3–4 μM ) were fixed to Menzel Superfrost Plus glass histology slides ( Menzel-Gläser , Braunschweig , Germany ) and air-dried overnight at 37°C . The sections were dewaxed and rehydrated in a descending alcohol series to water , and antigen retrieval was performed in Dako Target Retrieval solution ( pH 9 . 0 ) at 100°C for 20 min using a Biocare Medical decloaking chamber . On completion of the cooling cycle , slides were cooled for a further 20 min before being washed in three changes of Tris-buffered saline containing 0 . 025% Tween 20 ( TBS-Tween ) . The sections were incubated in Background Sniper solution ( Biocare Medical , Walnut Creek , CA , USA ) plus 1% BSA for 15 min to inhibit nonspecific antibody binding . Excess Background Sniper was removed and slides transferred to an opaque humidified chamber for subsequent incubation steps . Sections were incubated in 4G4 anti-Flavivirus NS1 monoclonal antibody hybridoma supernatant ( undiluted ) overnight at room temperature in a humidified chamber , washed three times in TBS-Tween , and incubated for 1 h in AlexaFluor-488 conjugated donkey anti-mouse antibody , diluted 1:300 in TBS-Tween . After washing three times in TBS-Tween , sections were counterstained with the fluorescent DNA stain 4' , 6-diamidino-2-phenylindole ( DAPI ) for 5 min , washed as above and mounted with coverslips using Dako fluorescent mounting medium . Slides were scanned using an Aperio ScanScope Fl slide scanner ( Aperio Techologies , Vista , CA , USA ) at a magnification of 20× . Quantitative image analysis was performed as previously described [67] . Percentage infection ( number of positive bodies/total tested ) , dissemination ( number of positive leg/wing samples per total tested ) , and transmission ( number of positive saliva samples/total tested ) were calculated at each dpi for each species under fluctuating and constant temperature regimes . Significant differences between percentages were detected using Chi-Square tests . The median and interquartile range ( IQR ) values were calculated using GraphPad Prism Version 7 . 00 ( GraphPad Software , La Jolla , California USA , 2008 ) . Log-transformed virus titres in mosquitoes with infected bodies and wings and legs were analysed using two-way Analysis of Variance ( ANOVAs ) as a function of temperature , species , and their interactions , in IBM SPSS Statistics software version 23 . 0 . Differences were considered statistically significant at p < 0 . 05 . Receiver Operator Characteristic ( ROC ) curve analysis was performed for both species to predict threshold disseminated titre at which saliva infection was likely to occur . ROC curve analyses were performed using the pROC package in R version 1 . 12 . 1 ( May 2018 ) [68] , with samples pooled across days and temperatures for each mosquito species to ensure maximum predictive power . The ZIKV staining density ( ratio of ZIKV/DNA positive pixel area ) within defined tissues in histological sections was analysed by two-way ANOVA as a function of temperature , days post infection and their interaction using GraphPad Prism Version 8 . 02 . Post hoc comparisons of the main effects of days post infection were performed using Sidak’s multiple comparison test .
High body infection percentages ( number of positive bodies/total mosquitoes tested ) were observed for both mosquito species under constant and fluctuating temperature conditions , at all the time points tested ( Table 1 ) . The body infection percentage in Ae . aegypti were 80% ( constant ) and 75% ( fluctuating ) at 3 dpi , 65% ( constant ) and 70% ( fluctuating ) at 7 dpi , and 70% ( constant and fluctuating ) at 14 dpi ( Table 1 ) . Compared to Ae . aegypti , higher body infection percentages were observed in the Ae . albopictus temperature groups at all time points ( Table 1 ) . Infection percentages in Ae . albopictus bodies reached 95% ( constant ) and 85% ( fluctuating ) at 3 dpi , 90% ( constant and fluctuating ) at 7 dpi , and 80% ( constant ) and 100% ( fluctuating ) at 14 dpi ( Table 1 ) . Disseminated infection percentages in Ae . aegypti increased from 10% ( constant and fluctuating ) at 3 dpi , to 60% ( constant ) and 70% ( fluctuating ) at 7 dpi , and remained at 70% ( constant and fluctuating ) thereafter ( Table 1 ) . Disseminated infection percentages in Ae . albopictus were 15% ( constant ) and 0% ( fluctuating ) at 3 dpi , 70% ( constant ) and 60% ( fluctuating ) at 7dpi , and 45% ( constant ) and 100% ( fluctuating ) at 14 dpi , with significant differences at this time interval ( Table 1 ) . We also found a significant difference in dissemination percentages between the vector species for the fluctuating temperature regime ( Table 1 ) . At early time points , ZIKV was either generally not detectable in saliva , or transmission percentages were too low to be detected with our sample size . None of the Ae . aegypti in the fluctuating temperature group were infectious at 3 dpi; however , in the constant temperature group , ZIKV was detected in the saliva of a single Ae . aegypti mosquito ( 5% transmission ) ( Table 1 ) . ZIKV was not detected in the saliva of Ae . albopictus at 3 dpi ( Table 1 ) . At day 7 dpi , no Ae . aegypti saliva samples were found to contain infectious ZIKV . At the same time point , ZIKV was first detected in the saliva of Ae . albopictus mosquitoes maintained at constant temperature ( 10% transmission ) , but not in the fluctuating temperature group . The ZIKV transmission percentages of Ae . aegypti were significantly higher than in Ae . albopictus at 14 dpi , for both temperature conditions ( Table 1 ) . Whereas transmission percentages of 60% ( constant ) and 50% ( fluctuating ) were observed for Ae . aegypti at 14 dpi , only 10% ( constant and fluctuating ) of Ae . albopictus had infectious saliva at this time point ( Table 1 ) . Both species exhibited high viral loads ( >107 copies/body ) in bodies from 7 dpi in constant and fluctuating temperature groups , which remained at high levels until 14 dpi ( Fig 2 , S1 Table ) . No significant differences were observed in viral copy number in bodies between constant and fluctuating temperature regimes ( p > 0 . 05 ) . Overall , we found no significant effect of temperature ( p = 0 . 718 ) , species ( p = 0 . 107 ) , or an interaction between these two factors ( p = 0 . 411 ) on viral load in mosquito bodies . We did find a significant effect of day post-infection ( p < 0 . 0005 ) on virus loads , consistent with the observed increase in body titre across the time points in both species ( Fig 2 , S1 Table ) . ZIKV RNA was detected in the wings and legs of Ae . aegypti constant and fluctuating temperature groups at 3 dpi , albeit in only a very few mosquitoes ( Fig 3 , S2 Table ) . The median number of RNA genome copies in the wings and legs of both Ae . aegypti temperature groups increased from 3 dpi and reached its highest level at 14 dpi ( >107 copies/mosquito wings and legs ) ( Fig 3 , S2 Table ) . At early time points ( 3 dpi ) , levels of ZIKV RNA were ~104 copies/ wings and legs in Ae . albopictus mosquitoes maintained at constant temperature . Thereafter , ZIKV RNA levels in Ae . albopictus marginally increased in both the constant and fluctuating temperature groups until 14 dpi ( Fig 3 , 7 and 14 dpi , S2 Table ) . In contrast to Ae . aegypti , a significantly lower disseminated viral load ( p < 0 . 05 ) was observed in the wings and legs of Ae . albopictus mosquitoes at day 14 ( Fig 3 , 14 dpi , S2 Table ) . Overall , a significant difference in the disseminated viral load was observed between species ( Fig 3 , S2 Table ) . Significant effects in disseminated titre due to day ( p < 0 . 001 ) , species ( p = 0 . 001 ) and temperature ( p = 0 . 032 ) were identified . The results suggested that exposure to constant versus fluctuating temperature does influence viral disseminated titre , although these effects were only observed at days 3 and 14 post-infection ( Fig 3 , S2 Table ) . Furthermore , there was a statistically significant interaction between species and dpi ( p < 0 . 001 ) , indicating that disseminated titres differed between species within each of the days post-infection sampled here . To visualize ZIKV distribution in Ae . aegypti tissues over time , we performed immunofluorescent antibody staining using a monoclonal antibody recognising Flavivirus NS1 proteins ( Figs 4A and 5 ) . We quantified ZIKV staining density ( Fig 4B ) through image analysis of the relative staining area of ZIKV to DNA for individual organs/tissues ( Fig 4C–4E ) . The ZIKV staining density in mosquito midguts was visible from 3 dpi ( Fig 4B ) . ZIKV staining was detectable in tissue/organs surrounding midguts ( “body” samples ) from 7 dpi . It was detected in the heads of a majority of mosquitoes from 10 dpi . Fewer salivary glands than other organs/tissues could be observed within the mid-sagittal mosquito sections , however , staining of the salivary glands that were observed indicated that a small proportion were infected by 7 dpi . By 10 dpi , all salivary glands had detectable ZIKV staining . An analysis of the staining density within the tissue/organs as a function of dpi and temperature regime found that , for all tissues , the effect of time post infection on ZIKV staining density was highly significant , whereas the effect of temperature regime was not significant ( S3 Table ) . Interactions between time and temperature were non-significant in all cases . Post hoc comparisons revealed that significant increases in staining density occurred between 7 and 14 dpi for all tissues ( Fig 4B ) . ZIKV was also detected and quantified within the ovaries of Ae . aegypti mosquitoes , which showed a significant increase in staining density between 5 and 14 d ( Fig 6A and S3 Table ) . Staining was limited to the follicular epithelium surrounding oocytes ( Fig 6B ) . We found that a disseminated titre of 7 . 50 log10 genome copies per mosquito wings/ legs ( sensitivity of 0 . 943; 95% CI: 0 . 857–1 . 000 ) was required to predict successful infection of mosquito saliva in Ae . aegypti . Surprisingly , a lower threshold titre of 6 . 52 log10 ( sensitivity of 0 . 922; 95% CI: 0 . 843–0 . 980 ) was necessary in Ae . albopictus to obtain ZIKV infection in saliva , in the proportions ( 2/20 each at 7 and 14 dpi in constant and 2/20 at 14 dpi in fluctuating temperature regimes ) of mosquitoes that were able to transmit the virus .
Our study demonstrates that Ae . aegypti populations from north Queensland are susceptible to a Brazilian epidemic ZIKV strain from Asian lineage , and able to transmit ZIKV from 10 dpi . We also show that Torres Strait Ae . albopictus could be infected in high percentages , but only 10% could transmit virus by 14 days . Our results suggest that a high threshold titre of disseminated infection in Ae . aegypti was required to overcome the salivary barrier and allow transmission . A recent report suggested that a threshold viral load of at least 105 . 1 TCID50 equivalents/mL in the legs and wings of Australian Ae . aegypti mosquitoes had to be reached for transmission of the prototype African strain of ZIKV to occur [58] . The infectious titre of disseminated virus could therefore be a significant predictor of virus detection in saliva . A similar correlation between disseminated virus titre and transmission rate has been reported for ZIKV [69] and DENV-1 [70] , with high dissemination titres resulting in increased transmissibility by Ae . aegypti . In Australia , variable vector competence of Ae . aegypti populations from north Queensland for Zika has been reported [48 , 58 , 59] . Those populations were shown to be competent vectors for the African lineage of ZIKV [58] , but relatively inefficient vectors of a Western Pacific ZIKV strain belonging to the Asian ZIKV lineage [59] . Our data suggest that Ae . aegypti from northern Queensland in Australia may be less susceptible to Asian ZIKV strains than to the prototype African strain [58] . Our findings are supported by the results of oral challenges of Australian Ae . aegypti with a strain of ZIKV from the Western Pacific [59] in which infection and transmission rates were 40 and 37% respectively , using a similar virus titre to that employed here ( 8 . 5 and 8 . 8 log CCID50/ml , for the Western Pacific and Brazilian strains , respectively ) . It should be noted that the titres used in both studies are higher than those expected in typical human viremias [6 , 36] . However , oral challenge of Australian Ae . aegypti with a lower titre of the Western Pacific ZIKV strain ( 5 . 6 log CCID50 ) resulted in only 3% of mosquitoes becoming infected [59] . Our study suggests that both high viremias and high disseminated threshold titres are required in order to obtain successful infection of Ae . aegypti and allow viral transmission to occur . Although Ae . aegypti could transmit ZIKV at moderate efficiency following challenge with a high titre , we have shown that under similar conditions , the transmission capability of Torres Strait Ae . albopictus was only 10% . Ae . albopictus is therefore less likely to participate in local transmission cycles than Ae . Aegypti in Australia . A higher transmission rate ( 87% ) of a Cambodian ZIKV strain ( Asian lineage ) has been reported for Ae . aegypti from Cairns in north Queensland [48] . Our data suggest the vector competence of Australian Ae . aegypti mosquitoes could depend on the geographical origin of populations and the virus strain/genotype , although differences between experiments will also contribute to the variation . The importance of investigating vector/virus strain interactions was recently demonstrated for a strain of Ae . aegypti from New Caledonia [69] . Infection , dissemination and transmission rates were significantly lower for recently isolated ZIKV strains from Africa and Asian lineages , compared with older African lineage isolates . In compatible combinations , ZIKV transmission occurred as early as 6 dpi [69] . Such genotype × genotype interactions have also been reported for DENV transmission [71] . Our study is in agreement with proportions of mosquitoes able to transmit ZIKV at 14 dpi reported for American Ae . aegypti challenged with Brazilian ( 75% ) , Puerto Rican ( 65% ) , and Malaysian ( 53% ) ZIKV strains [72] . Similar to a study of French Polynesian Ae . aegypti [28] , we found a significant increase in ZIKV transmission percentages from early time points ( 3 and 7 dpi ) to 14 dpi . Similar transmission patterns have also been observed for other commonly investigated flaviviruses , i . e . dengue [73] . Our results from immunofluorescence analysis indicate that ZIKV transmission in Ae . aegypti potentially occurs from 10 dpi , similar to populations from the Island of Madeira in Portugal that were infectious at 9 days following an oral challenge with a New Caledonian ZIKV strain [49] . In contrast , Ae . aegypti mosquitoes from Singapore were able to transmit an Ugandan ZIKV strain as early as 5 dpi [74] . Compared to Ae . aegypti , Ae . albopictus mosquitoes were poor vectors for the Brazilian strain of ZIKV . The ZIKV transmission percentages observed in our study are similar to those reported for French and Italian Ae . albopictus mosquitoes challenged with ZIKV from the Asian genotype [49 , 54] . However , the infection ( 10–18% ) and disseminated infection ( 10–29% ) rates reported in these studies were much lower than those observed in our study . Our results are strikingly different from a vector competence study in Singapore reporting that all Ae . albopictus mosquitoes challenged with an Ugandan ZIKV strain were infectious by 14 dpi [51] . Ae . albopictus populations from the Australian Torres Strait Islands have previously been shown to be highly susceptible to a Cambodian ZIKV strain , with a high prevalence ( >75% ) of virus in saliva at day 14 post-infection [48] . This suggests that the transmission of ZIKV in this population of Ae . albopictus is highly virus strain-dependent , as previously reported for American Ae . albopictus populations [57] . A specific vector/virus combination may therefore be more efficient at transmitting ZIKV than another . The extrinsic incubation period , which is the time between oral infection and presence of virus in the saliva of vectors , is a major determinant of transmission efficiency [75] . We established the kinetics of ZIKV infection , dissemination and transmission in Ae . aegypti by measuring viral RNA in mosquito tissues and live virus in saliva and mosquito organs and tissues and measured viral RNA in Ae . albopictus tissues and live virus in saliva . Our findings support an extrinsic incubation period ( EIP ) of approximately 10 days in Ae . aegypti under the conditions tested . We found that there were dose-dependent thresholds for infection of salivary glands in both species . Surprisingly , despite the lower transmission percentages observed for Ae . albopictus compared to Ae . aegypti , the estimated threshold for transmission was also lower . The result suggests factors other than disseminated viral titre may be responsible for the transmission percentages observed in Ae . albopictus . Possible explanations for the lower ZIKV transmission percentages at 14 dpi for Ae . albopictus , compared to Ae . aegypti , is that EIP is longer in Ae . albopictus , and/or may be modulated significantly by temperature . This has important public health implications for preparedness , and efficient implementation of mosquito control efforts . A recent study reported that the administration of a second , non-infectious blood meal significantly shortened the EIP of ZIKV-infected Ae . aegypti and Ae . albopictus by enhancing virus escape from the mosquito midgut [76] . Ae . albopictus may therefore be more competent for ZIKV transmission under field conditions of frequent feeding , suggesting the risk of an outbreak mediated by this vector may be higher than is indicated by our data . Whether this holds true for Australian Ae . aegypti and Ae . albopictus remains to be determined . Last , we observed ZIKV staining in mosquito ovarian tissue , limited to the follicular epithelium surrounding developing eggs . This may indicate a potential route of infection leading to vertical transmission , which has been observed recently from field specimens collected as larvae [38] . Although most vector competence studies only take mean temperature values into account , recent evidence for DENV shows that diurnal temperature range ( DTR ) plays an important role in influencing the behaviour of Ae . aegypti [73 , 77] . The DTR mimics more realistic field conditions , which could provide more accurate predictive disease outbreak models [73 , 77 , 78] . Taking into account the daily temperature fluctuation recorded during the summer months in Cairns Australia , we tested the effect of temperature fluctuations on Ae . aegypti and Ae . albopictus vector competence for ZIKV . Fluctuating temperature significantly affected viral dissemination to wings and legs rather than viral titre in bodies . Our findings suggest using a DTR that mimics field conditions is needed to better understand infection dynamics within mosquito hosts . This study has demonstrated that north Queensland Ae . aegypti are more competent for a Brazilian strain of ZIKV than Ae . albopictus , confirming that Ae . aegypti is the primary vector of Asian lineage ZIKV . The risk of emergence of ZIKV in Australia is potentially high due to the presence of competent mosquito vectors , climatic conditions suitable for transmission , imported cases , and a large naïve population for ZIKV . However , our data were obtained under high-titre challenge conditions and should be viewed in the context of a recent study that shows low competence of north Queensland Ae . aegypti under more typical viremic titres [59] . We also need to add the caveat that our estimates of vector competence were derived from a single experimental replicate . Additional replicates may yield different estimates due to stochastic variance inherent in vector competence experiments . In the absence of an effective vaccine and as ZIKV transmission is primarily vector-borne , mosquito control is likely to be the most effective preventative measure . In this regard , the use of the endosymbiotic bacterium Wolbachia pipientis has shown potential for the biocontrol of ZIKV [79] and other human pathogenic flaviviruses and alphaviruses [80 , 81] . Large field releases in north Queensland of novel Wolbachia-transinfected Ae . aegypti mosquitoes , refractory to infection by a range of arboviruses [79–81] , have shown the ability to drive Wolbachia into wild populations [82] . Our data could be beneficial for modelling likely ZIKV transmission dynamics in north Queensland and addressing emerging ZIKV threats to Australia . | Zika virus ( ZIKV ) is a mosquito-borne pathogen that generally causes a mild febrile illness but mostly remains asymptomatic in 50–80% of infections . Infection during pregnancy can cause congenital malformations , notably microcephaly . In adults , it can cause Guillain-Barré syndrome . The recent ZIKV epidemic in the Americas has been linked to the urban vector Aedes aegypti . The presence of the species in Australia makes the region vulnerable to emerging mosquito-borne viruses . A mosquito’s competence to transmit a pathogen will depend on both the virus and vector strains . Here , we determine the vector competence of Australian Ae . aegypti and Ae . albopictus mosquitoes for a ZIKV epidemic strain , originating from the epicentre of the Brazilian outbreak , under constant and fluctuating temperatures that simulate field environments in Australia . Our results demonstrate that , although both species were susceptible to ZIKV infection , Ae . aegypti is more likely to transmit virus . Our results may aid in the formulation of public health strategies to mitigate the threat of ZIKV . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"nuclear",
"staining",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"geographical",
"locations",
"microbiology",
"australia",
"saliva",
"animals",
"viruses",
"rna",
"viruses",
"insect",
"vectors",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"infectious",
"diseases",
"staining",
"south",
"america",
"aedes",
"aegypti",
"medical",
"microbiology",
"microbial",
"pathogens",
"exocrine",
"glands",
"disease",
"vectors",
"insects",
"brazil",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"dapi",
"staining",
"eukaryota",
"anatomy",
"flaviviruses",
"oceania",
"viral",
"pathogens",
"physiology",
"salivary",
"glands",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"organisms",
"zika",
"virus"
] | 2019 | Vector competence of Australian Aedes aegypti and Aedes albopictus for an epidemic strain of Zika virus |
Recently , most onchocerciasis control programs have begun to focus on elimination . Developing an effective elimination strategy relies upon accurately mapping the extent of endemic foci . In areas of Africa that suffer from a lack of infrastructure and/or political instability , developing such accurate maps has been difficult . Onchocerciasis foci are localized near breeding sites for the black fly vectors of the infection . The goal of this study was to conduct ground validation studies to evaluate the sensitivity and specificity of a remote sensing model developed to predict S . damnosum s . l . breeding sites . Remote sensing images from Togo were analyzed to identify areas containing signature characteristics of S . damnosum s . l . breeding habitat . All 30 sites with the spectral signature were found to contain S . damnosum larvae , while 0/52 other sites judged as likely to contain larvae were found to contain larvae . The model was then used to predict breeding sites in Northern Uganda . This area is hyper-endemic for onchocerciasis , but political instability had precluded mass distribution of ivermectin until 2009 . Ground validation revealed that 23/25 sites with the signature contained S . damnosum larvae , while 8/10 sites examined lacking the signature were larvae free . Sites predicted to have larvae contained significantly more larvae than those that lacked the signature . This study suggests that a signature extracted from remote sensing images may be used to predict the location of S . damnosum s . l . breeding sites with a high degree of accuracy . This method should be of assistance in predicting communities at risk for onchocerciasis in areas of Africa where ground-based epidemiological surveys are difficult to implement .
Onchocerciasis , or river blindness , has historically been one of the most important causes of blindness worldwide [1] , [2] . The disease is caused by the filarial parasite Onchocerca volvulus . It is estimated that 37 million individuals worldwide are at risk for O . volvulus infection , with most residing in rural Africa [3] . In Africa , the parasite is primarily transmitted by black flies of the Simulium damnosum sensu lato species complex , which develop as larvae in fast running rivers and streams [3] . Transmission is most intense in and around river basins , rendering many such areas uninhabitable [4] . Unfortunately , areas bordering the river basins contain much of the fertile land found in sub-Saharan African savanna ecosystems . By preventing the agricultural use of the most fertile lands , onchocerciasis has had a significant negative impact on the economic growth of many of the poorest countries of Africa [5] . Treatment and control of onchocerciasis as a public health problem was revolutionized by the discovery that Mectizan ( ivermectin ) had a potent effect on the larval stages of O . volvulus and by the subsequent offer of Merck , Sharpe and Dohme to donate Mectizan free of charge for the treatment of onchocerciasis for as long as needed . This generous donation resulted in the establishment of two major international programs , the African Programme for Onchocerciasis Control , ( APOC ) and the Onchocerciasis Elimination Program of the Americas ( OEPA ) , whose goals were to control onchocerciasis as a public health problem in Africa ( APOC ) , or to eliminate the parasite from the Americas ( OEPA ) . Both programs utilize community-wide mass distribution of Mectizan as their primary strategy of meeting these goals . OEPA has been successful in interrupting transmission in the majority of onchocerciasis foci in South and Central America [6] . It was initially believed that Mectizan distribution alone could not successfully eliminate onchocerciasis in Africa , due to the widespread distribution of the infection and the intensity of transmission [4] . However , recent data have suggested that this is not the case , and that long-term community wide distribution of Mectizan may be capable of eliminating onchocerciasis in at least some foci in Africa [7]–[10] . This discovery has resulted in a re-focusing of the international community from an emphasis on control of onchocerciasis in Africa towards an emphasis upon possible elimination [11] , [12] . However , elimination of onchocerciasis through community-wide Mectizan distribution is logistically difficult , as treatment must be carried out at least annually for many years [13] . Thus , if elimination efforts are to be expanded in Africa , accurate delineation of endemic communities is necessary . Currently , endemic communities are identified through ground-based epidemiological surveys . These can be difficult to conduct in remote and conflict-ridden regions of Africa such as in Southern Sudan and the Democratic Republic of Congo . Thus , methods to identify at-risk communities that cannot be easily reached by ground-based epidemiological surveys are urgently needed . A common characteristic of all of the members of the S . damnosum s . l . species complex is that the immature stages require fast flowing well oxygenated water for development [14] . This means that these flies are quite localized within lotic ecosystems in sub-Saharan Africa . Furthermore , the geographic extent and intensity of human infection is limited by the fact that these insect vectors arise from restricted riverine habitats i . e . , fast flowing water . Thus , unlike malaria and many other tropical diseases , the distance that an adult black fly can disperse from its breeding site in search of a blood meal delineates the geographic distribution of onchocerciasis . Simulium damnosum s . l . generally travels no further than 12 km from its river source in search of a blood meal [15] and intense transmission of the parasite , resulting in hyper-endemic prevalence levels of disease are generally confined to communities located within 10 km of a river [16] , [17] . For these reasons , the common name of “river blindness” has been used to describe onchocerciasis [18] . The ability to identify S . damnosum s . l . aquatic larval sites using remote sensing data would be an effective method to delineate areas most at risk for onchocerciasis . The overall goal of this study was to conduct validation studies to evaluate the sensitivity and specificity of a remote sensing model that identifies S . damnosum s . l . aquatic larval habitats in Africa .
Studies in West Africa were conducted along the Sarakawa River in northern Togo . This area has three distinct seasons: warm and dry ( November–March ) , hot and dry ( March–May ) , and hot and wet ( June–October ) . Annual rainfall varies from about 250 mm to 1 , 000 mm . The terrain is mostly flat with undulating plains and hills . Most of the study site region lies on a savanna plateau , with fields , brush , and scattered trees . The geological history of the site is marked by Precambrian volcanic activity . The study site has four ecological zones or landscapes including: 1 ) savanna with sparse tree cover; 2 ) savannas with forested cover , 3 ) grassy savannas and 4 ) savannas in temporarily flooded riverine habitats with both forested and grassy adjacent areas . The area is endemic for the savanna-dwelling sibling species of S . damnosum s . l . ( S . damnosum s . s . and S . sirbanum ) which together represent the two major onchocerciasis vectors in sub-Saharan Africa [18] . QuickBird sub-meter satellite data obtained from Digital Globe Inc . , ( Longmont , CO , USA ) were used for this study . The satellite image and data of the study site were acquired on July 15 , 2010 , roughly at the mid-point of the rainy season . The satellite data contained 25 km2 of the land area in the study site . The QuickBird image data were delivered as pan-sharpened composite products in infrared colors . The clearest cloud-free images available of the contiguous sub-areas of the study site along the river and tributaries were used to identify land cover and other spatial features associated with S . damnosum s . l . aquatic habitats . The QuickBird imagery was classified using the Iterative Self-Organizing Data Analysis Technique ( ISODATA ) unsupervised routine in ERDAS Imagine v . 8 . 7 ( ERDAS , Inc . , Atlanta , Georgia ) , as previously described [19] . The identification of a spectral signature characteristic of S . damnosum s . l . positive aquatic sites has been previously described [19] . This spectral signature is characteristic of the habitat features found at known positive sites , which include fast flowing water passing over a substrate of Precambrian rock . The model developed to predict S . damnosum s . l . aquatic habitats based on this signature was designated the black rock-rapid ( BRR ) model . The spectral signature found to be characteristic of the habitat features that formed the basis of the BRR model is shown in Figure 1 . The waveband composition data of the signature was 34% red , 11% blue and 55% green . To develop the BRR model , individual pixel ( 0 . 6 m2 per pixel ) spectral reflectance estimates in the QuickBird images were extracted from georeferenced validated S . damnosum s . l . aquatic habitats using a Li-Strahler geometric-optical model , as previously described [20] . This procedure allowed for the creation of a spectral signature of a unit of habitat . The model used three scene components: sunlit canopy ( C ) , sunlit background ( G ) and shadow ( T ) generated from the QuickBird image , to determine the sub-pixel endmember spectra associated with the known habitats . The C , G and T component classes were estimated using the ENVI software package ( Exelis Visual Information Solutions , Boulder , CO ) which employs an object-based classification algorithm [21] . Non-parametric estimators from the endmember spectra and the geometric-optical model were then used to construct a Boolean model that generated a robust spectral signature reference in an ArcGIS database specific for the verified S . damnosum s . l . habitats . These unique identifiers of aquatic habitat spectral signatures were then used to predict larval habitats along unsurveyed rivers in both Togo and Uganda . A detailed protocol describing how to utilize the spectral signature to identify S . damnosum aquatic habitat is presented supplemental file S1 . To assess the BRR model's ability to predict riverine larval habitat sites that may become temporarily active under varying flow or flooding conditions , a second model was developed . The strategic approach taken was to overlay a Digital Elevation Model with signals characteristic of Precambrian rock plus white water , or Precambrian rock alone . In order to accomplish this , we used PCI Geomatics software ( PCI Geomatics , Toronto , Canada ) , which supported an automatic overlay of an interpolated wet and dry Precambrian rock signature along the river course . This analysis revealed the locations of both active habitats ( i . e . those with water flowing over Precambrian rock ) as well as sites that might become active under increased flow or flooding conditions . The Digital Elevation Model is a simple tool to locate differences in elevation that would show areas where such fast flowing water could occur during different river flow conditions . PCI Geomatics Orthoengine software ( PCI Geomatics , Toronto , CA ) was used to generate a Digital Elevation Model from the QuickBird images . To accomplish this , we used the ArcGIS image server extension supported by version 10 . 3 of Geomatica Orthoengine for constructing a Toutin's rigorous model , with multiple Rational Polynomial Coefficients ( RPC ) as previously described [20] . The RPC method used a model developed by Digital Globe , Inc . that approximates a 3-D physical sensor for the QuickBird satellite data . Since bias or error may still have existed in the RPCs representing the interpolated endmember of the signatures , the results were post-processed with a polynomial adjustment product to correct for these biases . As a first step in assessing the ability of the BRR model to predict the presence of S . damnosum s . l . breeding sites , images were collected from the Sarakawa river basin in Northern Togo , which is endemic for the savanna dwelling sibling species of S . damnosum s . l . ( S . damnosum sensu strictu and S . sirbanum ) . These two species together represent the most important onchocerciasis vectors in the savanna regions of sub-Saharan Africa . These images were analyzed to predict potential S . damnosum s . l . larval habitats along a 12 km stretch of the river course using the BRR model . The model predictions were then validated by a ground-based validation team , which traversed the 12 km of river by boat . The team included an experienced entomologist employed by the Onchocerciasis Control Program of Burkina Faso . The team visited all of the sites predicted to contain larvae . Sampling was conducted at each site to determine if larvae were present and to estimate the density of the population , if larvae were found . The team also visited and sampled every other site identified by the entomologist as potential S . damnosum s . l . aquatic habitat . Validation studies carried out in Northern Uganda along the Achwa River ( Figure 2 ) . This area had long suffered from political instability promulgated by the Lord's Resistance Army . As a result , little was known concerning the current prevalence of onchocerciasis in this area . This area presented an ideal opportunity to test the performance of the BRR model in an area where little was known concerning the biology and riverine distribution of black flies and the epidemiology of the disease . The validation studies conducted along the Achwa River employed a similar procedure to that utilized in Northern Togo , with the exception that the sites were visited by walking along the river banks rather than by boating the river course .
Of the 30 sites along the Sarakawa River in Northern Togo predicted to be larval habitats by the BRR model , all ( 100% ) were found to contain S . damnosum s . l . larvae . In contrast , none of the 52 sites not predicted by the BRR model , but deemed to be potential habitat by the entomologist accompanying the verification team contained S . damnosum s . l . larvae ( Figure 3 ) . Together , these data suggested that the BRR model exhibited a sensitivity and specificity approaching 100% for the prediction of S . damnosum s . l . riverine larval sites in Togo . Water levels in the rivers of West Africa fluctuate substantially between the rainy and dry seasons , potentially producing seasonally active breeding sites . A second model was developed to predict such seasonally active sites . There was a complete correspondence between the sites predicted by the the BRR model and this second model ( Figure 4 ) , suggesting that the BRR model had identified all active and potentially active breeding sites in the study area . S . damnosum s . s . and S . sirbanum are found in savanna ecosystems throughout most of sub-Saharan Africa . To test the generality of the BRR model , it was applied to predict S . damnosum s . l . riverine sites in Northern Uganda . A total of 25 potential S . damnosum s . l . larval breeding sites were predicted ( Figure 5 ) . Of the 25 sites predicted to be suitable S . damnosum s . l . aquatic habitats by the BRR model , 23 ( 92%; 95% CI 81–100% ) were found to contain S . damnosum s . l . larvae . In contrast , just 2/10 ( 20%; 95% CI 0–45% ) sites examined which were not predicted to represent S . damnosum s . l . aquatic habitat by the model were found to contain larvae . The BRR model thus exhibited a sensitivity of 80% and a specificity of 92% when applied in Uganda , a performance that was statistically significant ( p<0 . 0001; Fisher's Exact test ) . The two sites that were not predicted by the model which nonetheless were found to contain larvae consisted of low hanging streamside vegetation immersed in fast flowing water ( Figure 6 ) . The mean number of larvae found at the sites predicted by the BRR model ( 21 . 91 ) was significantly greater that the mean number of larvae at the sites consisting of immersed overhanging vegetation ( 4 . 0; p<0 . 001 , Mann Whitney U test ) .
It has long been recognized that members of the Simulium damnosum s . l . sibling species complex require quite specific ecological conditions for oviposition and for immature development . This species complex requires fast flowing , well oxygenated water in which to develop [14] . This forces the immature stages to localize in areas along rivers and streams where these conditions exist . This key characteristic formed the basis for the vector control activities that underpinned the former Onchocerciasis Control Programme in West Africa ( OCP ) . The OCP's primary strategy was to employ targeted insecticide treatment of S . damnosum s . l . aquatic breeding sites to reduce the vector black-fly populations . This strategy was successful in dramatically reducing ocular onchocerciasis in an area encompassing 11 countries [22] . The studies reported here suggest that a remote sensing model based upon the identification of habitat features that are characteristic of these riverine sites was capable of predicting the locations of S . damnosum s . l . breeding sites with a high degree of accuracy . The ability to remotely predict the location of S . damnosum s . l . breeding sites will be extremely useful in precisely mapping onchocerciasis transmission foci as the current African onchocerciasis control programs move into the era of elimination [11] , [12] . In particular , this model may help provide precise mapping of S . damnosum s . l . riverine foci in the eastern Democratic Republic of Congo and Southern Sudan , both of which currently suffer from political instability and a lack of infrastructure . Similarly , the ability to predict and precisely locate riverine breeding sites using remote sensing satellite data will provide specific location-based data for mapping the extent of transmission zones in areas that abut international borders , where it is difficult to establish cross border collaborations to conduct ground based studies and coordinated control initiatives . However , while the BRR model may prove to be a useful tool in delineating communities at risk for onchocerciasis and therefore potentially eligible for Mectizan distribution , effectively delivering treatment to such isolated communities will remain a challenge . The model may also be useful in guiding localized vector control measures that may in some cases could be used to supplement Mectizan distribution . Such measures may need to be considered in areas where biting densities are high [23] , [24] , or in areas that are co-endemic with the ocular parasite Loa loa , as treatment with Mectizan in such co-endemic areas is complicated by the risk of severe adverse events ( SAEs ) to treatment [25] , [26] . The existence of SAEs has limited the use of Mectizan by community directed treatment programs in some areas where onchocerciasis and loasis are co-endemic [27] . The extent of Loasis has previously been mapped using a rapid epidemiologic assessment method ( RAPLOA ) [28] . It may be possible to employ the BRR model to predict areas likely to be endemic for onchocerciasis in areas previously mapped using the RAPLOA procedure . Combining these maps could identify communities at risk for co-endemicity for both diseases , thereby permitting more complete geographic coverage of the Mectizan community distribution programs while simultaneously minimizing the risk of L . loa associated SAEs in the treated areas . Because there are wide variations in river flow between the wet and dry seasons , it was possible that the BRR model would not detect breeding sites that were seasonally active . However , a test of this hypothesis utilizing a secondary model based solely upon the presence of pre- Cambrian rock ( wet or dry ) in areas with sufficient elevation change to generate fast flowing water ( whether or not water was currently present ) did not predict the presence of any additional potentially seasonally active breeding sites . The most likely explanation for this is that the images analyzed by both models were taken in July at the mid-point of the rainy season and all potential breeding sites were active . This suggests that the BRR model may be most accurate when used to analyze images taken during the rainy season . However , because the model predictions were based upon high resolution images ( 0 . 6 m2 ) it is also possible that at this degree of resolution the signature characteristic of fast flowing water will be present at all sites during all seasons . Despite this , it is likely that the absolute amount of suitable habitat present at each site will vary according to seasonal changes in water flow . In this regard , it would be of interest to determine if the actual productivity of a breeding site might be predictable based upon the amount of habitat pixels detected by the BRR model . Furthermore , application of the model to images taken at the end of the rainy season or at the beginning of the dry season might be used to predict the most productive breeding sites , permitting a more efficient targeting of localized vector control measures , if such measures become part of the elimination strategic plan in some areas . Studies investigating these possibilities are underway . The current cost of QuickBird imagery ( ca . $17 USD per km2 ) will prove quite expensive if this model is to be applied for predicting S . damnosum s . l . breeding habitats on a large scale . However , it may be possible to reduce this cost substantially . First , the number of images necessary might be reduced by limiting the QuickBird image acquisition to river and streambeds , which can be initially identified using free or less expensive data . Alternatively , now that a signature characteristic of S . damnosum s . l . larval habitats has been identified with the high-resolution imagery , it may be possible to extract this signature from lower resolution data . It may also be possible to use features extracted from free or less expensive remote sensing data ( e . g . vegetation cover , elevation change and river gradient ) to pinpoint the regions along a river likely to contain suitable habitats , thereby narrowing down those areas for which QuickBird images need to be obtained . Studies exploring these possibilities are currently underway . In the validation studies conducted in Uganda , the BRR model did not identify a two breeding sites that were characterized by low hanging riverbank vegetation immersed in fast flowing water . Apparently these sites do not exhibit the spectral signature detected by the BRR model . However , the success of the BRR model suggests that a similar approach might be taken to identify such sites , which would be characterized by immersed vegetation in an area of fast flowing water . Combining this model with the BRR model predictions might be successful in improving the sensitivity of the remote sensing prediction of S . damnosum s . l . breeding sites . The BRR model was developed to predict the riverine locations of the savanna dwelling sibling species of S . damnosum s . l . , S . damnosum s . s . and S . sirbanum . While these sibling species represent the most important and most widely distributed onchocerciasis vectors in Africa , the model may not be universally applicable to predict the riverine sites of all O . volvulus black fly vectors in Africa . For example , as the current model relies on spectral signatures collected from the visible range , it requires a clear line of sight or relatively large open areas ( >5 m ) to identify these sites using sub-meter satellite data . Thus , the model will have limited applicability along the coastal rain forested areas of sub-Saharan Africa , where overhanging vegetation often obscures productive riverine sites . Similarly , the current model will not be applicable to the East African vector Simulium neavei . Simulium neavei has a unique life cycle in which the larvae are phoretic on the bodies of freshwater crabs ( Potamonautes loveni ) found in the streams of Western Kenya and Eastern Uganda [29] . Additional studies will be required to determine if similar remote sensing models might be developed to locate and predict larval breeding habitats areas inhabited by these species . | Onchocerciasis , or river blindness , represents a major cause of socio-economic disruption throughout much of Africa . The discovery that Mectizan ( ivermectin ) was effective in treating onchocerciasis , together with the decision of Merck Sharpe and Dohme to donate this drug for the treatment of this disease , revolutionized onchocerciasis control efforts . But until recently , it was thought that ivermectin alone was not sufficient to eliminate this scourge . However , recent studies have suggested that long-term ivermectin treatment can eliminate onchocerciasis in some foci in Africa . This has revolutionized the thinking of the international community , turning its attention from an emphasis on control to an emphasis on elimination . For an elimination program to be successful , it is necessary to accurately map all at-risk communities . In the case of onchocerciasis , this is commonly done by epidemiological surveys . Here we report the validation of a remote sensing method to identify areas endemic for onchocerciasis . This method is based upon predicting the sites where the black fly vector for the parasite breeds . The method was found to be very accurate at predicting black fly breeding sites in West Africa and in northwest Uganda . The method should be useful in assisting in mapping at-risk communities in areas of Africa where ground-based surveys are difficult or impossible to implement . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"infectious",
"diseases",
"global",
"health",
"neglected",
"tropical",
"diseases",
"onchocerciasis",
"infectious",
"disease",
"control"
] | 2013 | Validation of a Remote Sensing Model to Identify Simulium damnosum s.l. Breeding Sites in Sub-Saharan Africa |
The Spumaretrovirinae , or foamyviruses ( FVs ) are complex retroviruses that infect many species of monkey and ape . Although FV infection is apparently benign , trans-species zoonosis is commonplace and has resulted in the isolation of the Prototypic Foamy Virus ( PFV ) from human sources and the potential for germ-line transmission . Despite little sequence homology , FV and orthoretroviral Gag proteins perform equivalent functions , including genome packaging , virion assembly , trafficking and membrane targeting . In addition , PFV Gag interacts with the FV Envelope ( Env ) protein to facilitate budding of infectious particles . Presently , there is a paucity of structural information with regards FVs and it is unclear how disparate FV and orthoretroviral Gag molecules share the same function . Therefore , in order to probe the functional overlap of FV and orthoretroviral Gag and learn more about FV egress and replication we have undertaken a structural , biophysical and virological study of PFV-Gag . We present the crystal structure of a dimeric amino terminal domain from PFV , Gag-NtD , both free and in complex with the leader peptide of PFV Env . The structure comprises a head domain together with a coiled coil that forms the dimer interface and despite the shared function it is entirely unrelated to either the capsid or matrix of Gag from other retroviruses . Furthermore , we present structural , biochemical and virological data that reveal the molecular details of the essential Gag-Env interaction and in addition we also examine the specificity of Trim5α restriction of PFV . These data provide the first information with regards to FV structural proteins and suggest a model for convergent evolution of gag genes where structurally unrelated molecules have become functionally equivalent .
Spuma- or foamy viruses ( FVs ) are complex retroviruses and constitute the only members of the Spumaretrovirinae subfamily within the Retroviridae family . They have been isolated from a variety of primate hosts [1] , [2] , [3] , [4] as well as from cats [5] , [6] , [7] , cattle [8] , horses [9] and sheep [10] . Endogenous FVs have also been described in sloth [11] , aye-aye [12] and coelacanth [13] . Prototypic foamy virus ( PFV ) is a FV isolated from human sources [14] , [15] . The PFV genome is highly similar to that of isolates of simian foamy virus from chimpanzee ( SFVcpz ) and so infection in humans is believed to have arisen through a zoonotic transmission [16] , [17] , [18] . Nevertheless , even though FVs are endemic within non-human primates and display a broad host range , human-to-human transmission of PFV has never been detected . Moreover , although in cell culture FV infection causes pronounced cytopathic effects [19] infection in humans is apparently asymptomatic [20] , [21] , [22] making their usage as vectors for gene therapy an attractive proposition [23] . FVs share many similarities with other retroviruses in respect of their genome organisation and life cycle . However , they vary from the Orthoretrovirinae in a number of important ways . These include the timing of reverse transcription that occurs in virus producer cells rather than newly infected cells [24] , [25] and the absence of a Gag-Pol fusion protein [26] [27] . In addition , the Gag protein remains largely unprocessed in FVs [28] whereas within the Orthoretrovirinae processing of the Gag polyprotein represents a critical step in viral maturation , producing the internal structural proteins Matrix ( MA ) , Capsid ( CA ) and Nucleocapsid ( NC ) found in mature virions . Furthermore , FV Gag lacks the Major Homology Region ( MHR ) and Cys-His boxes found in orthoretroviral CA and NC , respectively . Also unique to FVs is a requirement for the interaction of the Gag protein with the viral envelope protein ( Env ) in order to bud from the producer cell [29] , [30] , [31] . Nevertheless , despite these profound dissimilarities , the Gag protein contains the cytoplasmic targeting and retention signal ( CTRS ) [32] , [33] , [34] , essential for both FV and betaretrovirus replication . Moreover , in all retroviral subfamilies Gag carries out the same functional roles including assembly , nucleic acid packaging , transport to and budding through the cytoplasmic membrane of the producer cell as well as trafficking through the cytoplasm of the target cell and uncoating . Similarly , FV Gag also contains the determinants for restriction by Trim5α [35] , [36] that in orthoretroviruses are the residues displayed on the assembled CA lattice [37] . To date , high resolution X-ray and/or NMR structures have been reported for MA , CA and NC components of Gag from numerous retroviruses [38] , [39] , [40] , [41] , [42] , [43] , [44] , [45] , [46] , [47] . However , structural information with regard to the Gag of FVs has remained elusive and is vital for any detailed understanding of how FV Gag fulfils its many functions . Here we report the crystal structure an amino terminal domain from the Gag of PFV ( PFV-Gag-NtD ) , provide the molecular details of the interaction of this domain with the N-terminal leader sequence from the PFV Envelope ( PFV-Env ) and demonstrate that the PFV-Gag-NtD is also the target for Trim5α restriction factors . Our data reveal that the FV Gag is unique and structurally unrelated to the Gag protein of other retroviruses . Nevertheless , the Gag-NtD has functional properties associated with both the MA and CA proteins of the orthoretroviruses . These findings have important implications for the evolution of FVs and the mechanism of virus restriction by Trim5α .
PFV Gag is a 648 polypeptide and the major FV structural protein in the assembled virion . Bioinformatic analysis of the primary sequence and that of related FVs suggested that the N-terminal 179 residues of PFV-Gag comprised a stable domain ( PFV-Gag-NtD ) . This fragment was expressed in E . coli and subsequently the crystal structure determined by SAD methods and refined at a resolution of 2 . 4 Å . The final Rwork/Rfree are 17 . 2% and 23 . 0% respectively . Details of the structure solution and refinement are presented in Table 1 . In the crystal , the asymmetric unit comprises a dimer of the protein with residues 9–179 clearly visible in the electron density map for both monomers . The residues preceding Glu9 along with the N-terminal His-tag are not visible and presumably disordered in the crystal . The structure of PFV-Gag-NtD dimer , Figure 1A , comprises a mixed alpha-beta fold dominated by a large central coiled-coil , resembling a two-bladed propeller . The N-terminal region of the protein forms a head domain containing a central 4-stranded β-sheet together with two helices , α1 and α2 , that pack against one side of the sheet forming a tight hydrophobic core . The loop between strands β3 and β4 crosses to the opposing side of the sheet where helix α2 leads into a region lacking secondary structure that precedes three further short helices α3 , α4 and α5 . Helix α5 is immediately followed by α6 , a long helix ( 58 Å ) comprising residues Arg140-Ser179 that forms the coiled coil domain making the majority of interactions between the two monomers . The observation of this unusual arrangement prompted us to examine the structural relatedness of the PFV-Gag-NtD with the Gag-derived proteins from other retroviruses and those of hepadnavirus . However , similarity searches undertaken using the DALI [48] and SSM [49] search engines revealed no significant homology between the PFV-Gag-NtD and the retroviral MA or CA . In fact , no homology was detected with any structure deposited in the PDB making the foamy virus Gag-NtD at present unique . The PFV-Gag-NtD dimer interface buries approximately 1700 Å2 of the monomer surface area . The large central coiled-coil formed by helix α6 comprises the majority of this interface , supplemented by residues in helices α4 and α5 and the adjoining loop . The coiled-coil contains three regions of leucine zipper at residues Leu143/147 , Leu160/161 and Leu171/175 . Additionally , a highly synergistic hydrogen-bonding network centred on residue Glu154 is located between two of the zipper regions . Here , the Glu154 sidechain forms hydrogen bonds with the sidechains of Glu154* , Gln150* and Tyr127* of the opposing monomer . Gln150 makes further hydrogen bonds with Ser151* that in turn is hydrogen bonded to the mainchain carbonyl of Val130 , Figure 1B . The loop between helix α4 and α5 runs alongside this region also making several interactions . In addition , at the amino terminus of the coiled-coil , Arg140 makes bifurcated hydrogen bonds with the backbone hydroxyl of Met138 and the sidechain hydroxyl of Glu144 as well as further hydrogen bonds with the backbone of Ala136 in helix α4 and the side chain of Asp141* in the opposing monomer . This extensive network of intermolecular protein-protein interactions and large molecular interface of 1700 Å2 is nearly twice that of the HIV-1 CA-CtD dimer interface , 920 Å2 [50] , [51] and suggests that Gag-NtD of PFV forms a tightly associated dimer . Moreover , sequence alignment with the N-terminal region of Gag from other primate foamy viruses , Figure 1C , reveals strong sequence conservation in loops and secondary structure elements in the head domains together with several buried hydrophobic residues in the coiled coil indicating that a conserved dimeric Gag N-terminal domain is likely a feature of the primate foamy viruses . Given the unexpected nature of the dimer observed in the crystal structure , the conformation and self-association properties of the Gag-NtD from PFV and from the non-primate feline foamy virus ( FFV ) were examined using a variety of solution hydrodynamic methods . Initial assessment by Size Exclusion Chromatography coupled Multi-Angle Laser Light Scattering ( SEC-MALLS ) , over a range of protein concentration ( 12-1 . 5 mg/ml ) , yielded invariant solution molecular weights of 40 . 0 kDa and 34 . 0 kDa for PFV- and FFV-Gag-NtD respectively , Figure 2A . By comparison , the sequence-derived molecular weights are 22 . 8 kDa and 19 . 0 kDa . Given these values together with the lack of a concentration dependency of the molecular weight it is apparent that along with PFV , the Gag-NtD from FFV also forms strong dimers in solution . To confirm the oligomeric state , velocity ( SV-AUC ) and equilibrium ( SE-AUC ) analytical ultracentrifugation of PFV- and FFV-Gag-NtD was undertaken . A summary of the experimental parameters , molecular weights derived from the data and statistics relating to the quality of fits are shown in Table 2 . Analysis of the sedimentation velocity data for PFV-Gag-NtD revealed no concentration dependency of the sedimentation coefficient ( S20 , w = 3 . 08 ) over the range measured , Figure 2B . Similar data were obtained for FFV-Gag-NtD ( S20 , w = 2 . 72 ) indicating both proteins are single stable species . The molecular weights derived from either C ( S ) or discrete component analysis were 47 kD and 36 kD respectively , Table 2 , consistent with a PFV- and FFV-Gag-NtD dimer . The frictional ratios ( f/fo ) obtained from the analysis , 1 . 4–1 . 5 , also suggest both dimers have a similar elongated conformation . Multispeed sedimentation equilibrium studies at varying initial protein concentration were also carried out and typical equilibrium distributions for PFV- and FFV-Gag-NtD from individual multispeed experiments are presented in Figure 2C . Analysis of individual gradient profiles showed no concentration dependency of the molecular weight and so data were fit globally with a single ideal molecular species model , producing weight averaged molecular weights of 44 kDa and 33 . 7 kDa for PFV- and FFV-Gag-NtD respectively . These data confirm that formation of stable dimeric structures is a common property shared among the Gag proteins of divergent FVs and N-terminal domain mediated dimerisation is likely an important component of FV assembly . The interaction of foamy virus Gag and Env proteins is a requirement for successful budding and the production of infectious particles [52] . Mutations in either Gag-NtD or the N-terminal leader peptide region of Env ( Env-LP ) have been shown to block viral egress [31] , [53] , [54] . To better understand this interaction and shed light on how FV Gags recruit Env , we examined the interaction of the PFV-Gag-NtD with the PFV-Env-LP using SV-AUC . Sedimentation data were recorded for Gag-NtD and for equimolar mixtures of Gag-NtD with either of two Env leader peptides , residues 5–18 or 1–20 , Figure 3A . The data were fitted using the C ( S ) distribution of sedimentation coefficients and the integrated absorbance of the fast moving Gag-NtD 3S component then quantified . In samples containing peptide-protein mixtures a small increase in the apparent sedimentation coefficient of the 3S boundary is apparent , accompanied by an increase in the integrated absorbance , Figure 3B . This shift and absorbance increase results from association of the strongly absorbing Env peptides with the PFV-Gag-NtD ( ε280 = 11 , 400 M−1 cm−1 ) and simple quantitation of the absorbance change reports the proportion of peptide bound and association constant for the interaction ( see methods ) . In this way an equilibrium association constant of 2 . 0×104 M−1 for the Gag-NtD interaction with Env residues 5–18 ( Env5–18 ) and 1 . 3×105 M−1 for the interaction with Env residues 1–20 ( Env1–20 ) was determined . To confirm this observation the interaction of Env1–20 with PFV-Gag-NtD was examined using isothermal titration calorimetry ( ITC ) . The results presented in Figure 3C reveal a 1∶1 stoichiometry where each monomer of the PFV-Gag-NtD binds a single Env peptide with an equilibrium association constant of 1 . 5×105 M−1 consistent with the SV-AUC experiments . The structure of the PFV-Gag-NtD bound to the Env1–20 leader peptide was determined by molecular replacement and refined at a resolution of 2 . 9 Å with a final Rwork/Rfree of 22 . 6% and 27 . 1% respectively , Table 1 . The asymmetric unit comprises two dimers of the complex with residues 9–179 of the Gag-NtD clearly visible in the electron density map for two of the four monomers and residues 9–170 in the two remaining protomers . Four helical Env peptides are also present , bound at the periphery of each head domain close to α1 and the associated α1-β1 loop of the Gag-NtD monomers , Figure 4A . Largely , the conformation of the Gag-NtD head and stalk domains are the same as in the free structure ( RMSD of 0 . 4 Å between all equivalent Cα atoms ) excepting some small differences in the conformation of the β3–β4 loop , Supplementary Figure S1 . However , in the bound structure the α1-β1 loop around the highly conserved residue Pro30 undergoes a concerted 2 . 5 Å shift , Figure 4B and Supplementary Figure S1 . Comparison of surface hydrophobicity profiles of the free and bound structures Figure 4B , reveals that this movement opens the Env binding site exposing a deep apolar pocket to accommodate the hydrophobic side chains from the Env peptide . In the complex residues Met1Env to Thr6Env of Env constitute an extended N-terminal region and Leu7Env to Met16Env form the hydrophobic α-helix bound to Gag . Hydrogen bonding between the sidechains of Thr6Env in the N-terminal region and Gln9Env on the amino-terminal turn of the Env helix provides stabilising interactions that maintain the helical conformation of the Env , Figure 4C . Inspection of Gag-Env interface reveals a network of hydrophobic interactions with the apolar and aromatic sidechains of Leu7Env , Trp10Env and Trp13Env on one face of the Env helix packing against the Val14 , Leu17 , Val18 and Leu21 sidechains on α1 of Gag , Figure 4C . In particular , the side chain of Leu7Env is seated in the apolar pocket in the Gag-NtD were it makes hydrophobic interactions with the aliphatic side chains of both Leu17 and Leu21 . Val14 packs against the ring of Trp10Env that also makes a hydrogen bonding interaction between the indole Nε proton and the carbonyl of Leu66 in the β2–β3 loop . This hydrophobic interface is accompanied by a number of polar contacts between the backbone of residues Ala2Env , Pro3Env and Met5Env in the Env N-terminal extended region with the sidechains of Asn63 and Gln59 in the β2–β3 loop and the mainchain of His32 and Pro30 in the α1-β1 loop . The bound conformation is further stabilised by an accompanying helix capping interaction between the Asn29 sidechain and the N-terminal turn of the Env helix . In order to probe the importance of the interactions in the Gag-Env interface observed in the crystal structure a series of serine and asparagine substitutions were introduced at Val14 , Leu17 , Val18 and Leu21 to make the Env binding site progressively polar . In addition , in order to examine the contribution of the α1-β1 loop to the Env interaction a conservative Asn29 to Gln substitution was also introduced . The affinity of binding of these Gag-NtD mutants to Env-LP was examined using the sedimentation velocity assay , Figure 5A–E and Supplementary Figure S2 . In all cases , the single polar substitutions introduced into the Env binding site reduced the affinity of the Gag-Env interaction . The decrease varied from 5 – 2 fold in the order Leu21>Val14>Leu17≈Val18 identifying these residues as being required for the Gag-Env interaction . Double substitutions decreased the affinity even further with the Val14/Leu21 to serine having the greatest effect , resulting in around a twenty-fold reduction in binding , Figure 5F . Moreover , the triple substitution where Val14 , Val18 and Leu21 were all substituted by serine reduced binding to an undetectable level , Supplementary Figure S2 . The conservative change Asn29 to glutamine has little effect on Env binding perhaps reflecting the importance of the backbone movement around Pro30 rather than sidechain interactions for Env-binding at this position . It has been shown previously that mutation of Leu17 in PFV-Gag-NtD gives rise to viral defects and negatively affects viral egress . Substitution by alanine has only minor effects on Env incorporation and particle release but progeny particles show a severe reduction of the infectivity . In contrast , serine substitution results in a loss of viral budding capacity [53] . To assess in vivo effects of serine substitution at Leu17 and at other positions in the Gag-Env interface the Leu17 , Val14 and Leu21 to Ser mutations that disrupt the Gag-Env interaction in vitro were introduced and transfected cells assayed for particle production as well as Env/Gag incorporation and viral infectivity , Figure 6 . In these in vivo experiments , the greatest effects were seen with Leu17 and Leu17/Leu21 mutant viruses that show greatly reduced levels of Gag released into the supernatant compared to wild type . By contrast , only a small reduction in Gag release was observed in the Val14 virus and in the Leu21 virus the amount of Gag is comparable to wt , Figure 6A . Examination of Env production and processing in the producer cells reveals it is unaffected by any of the mutations , Figure 6B . However , Env incorporation into virions is greatly reduced in both the Leu17 and Leu17/Leu21 particles , moderately reduced in the Val14 virus and that near wt levels are present in Leu21 particles . These results are mirrored when particle release was quantified , Figure 6C , where Leu21 particle production is only slightly reduced , Val14 is reduced around 3-fold , Leu17 around 20-fold and in the double substitution no particles are detectable in the cell supernatant . Where particles were released they were tested for infectivity relative to wt , Figure 6D . Although the Val14 mutant showed greater defects in viral Env and Gag incorporation the viruses were only around 6-fold reduced in infectivity whereas viruses with the Leu21 substitution showed around a 300 fold reduction in infectivity and no infectivity ( >100 , 00-fold reduced ) was detectable for the Leu17 mutant . Taking these data together it is apparent that the Leu17 mutation has the least effect on in vitro Env binding but causes very large defects in PFV virion production with little , if any , incorporation of viral proteins into particles . The Leu21 substitution weakens the in vitro Gag-Env interaction more , has little effect on particle production but the resulting viruses are poorly infectious and viruses with the Val14 substitution display intermediate effects having both reduced particle production and reduced infectivity . Previous experiments have demonstrated that Gag from PFV and the closely related SFVmac contain the target for Trim5α restriction . Moreover , PFV and SFVmac display a differential susceptibility to restriction mediated by the B30 . 2 domain of Brown capuchin Trim5α ( bc-T5α ) that is effective only against SFVmac and not PFV [36] . Based on sequence alignment , chimeras were prepared to more precisely map the target of Trim5α restriction in FV Gag . These included PSG-4 and SPG-4 , that swap the N-terminal ∼300 residues between PFV and SFVmac Gag and two further chimeras , one where the N-terminal 186 residues of SFVmac Gag was replaced by the N-terminal 195 residues of PFV Gag ( PSG-5 ) and a second where the N-terminal 195 residues of PFV Gag was replaced with the N-terminal 186 residues of SFVmac Gag ( SPG-5 ) . The results of bc-T5α restriction assays of parent and chimeric PFV and SFVmac viruses are summarised in Figure 7A , and detailed in Supplementary Figure S3 . These data confirm that PFV is resistant to bc-T5α restriction and that SFVmac is susceptible , and that sensitivity maps the N-terminal 300 amino acids of Gag , Figure 7A ( PSG-4 and SPG-4 ) . More importantly , these data also reveal that transfer of the N-terminal 186 residues of SFVmac to PFV ( SPG-5 ) now renders the virus susceptible to restriction by bc-T5α . Conversely , transfer of N-terminal 195 residues of PFV to SFVmac ( PSG-5 ) results in reduced sensitivity to bc-T5α restriction demonstrating that at least one determinant of restriction in primate FVs is contained within the Gag-NtD . Since NtD s of PFV and SFVmac Gag share a high degree of sequence similarity , the conserved residues along with those involved in the dimer interface were mapped onto the PFV-Gag-NtD structure , Figure 7B . Examination of this combined pattern of sequence conservation and surface accessibility reveals a large patch of surface exposed non-conserved residues on the upper surface of the molecule spanning from the β2–β3 loop across the outer surface of α2 and into the α2–α3 loop . The distribution of non-conserved residues over the top surface of the molecule is reminiscent of the distribution of residues that constitute the restriction factor binding sites in the N-terminal domain of the capsid of conventional retroviruses [55] . This suggests that the mode of foamy virus restriction by Trim5α is likely to be the same as in orthoretroviruses . In order to test this notion , mutations were introduced into the RING , B-Box and coiled coil domains of bc-T5α and the restriction of PFV and SFVmac by these impaired factors assayed . These data , summarised in Table 3 and detailed in Supplementary Figure S4 , show that disruption of the individual RING and B-Box domains or deletion of the coiled-coil region completely abolishes bc-T5α restriction of SFVmac and does not alter PFV susceptibility . Taken together with data demonstrating that the B30 . 2 domain of Trim5α mediates the Gag specificity of restriction [36] this demonstrates that FV restriction is reliant on the same functional regions required for orthoretrovirus restriction and likely occurs by the same mechanism .
Based upon both the functional similarities and positioning within PFV Gag it might be expected that the Gag-NtD would display a strong structural similarity with MA of orthoretroviruses . However , following extensive searching of the Protein Database ( PDB ) no such similarity was apparent and in fact no structures related to PFV-Gag-NtD were found at all . Like FV-Gag-NtD , the orthoretroviral MA protein is required for targeting Gag to the membrane and for viral budding . This is accomplished through a combination of a highly basic region ( HBR ) and in some subfamilies a myristoyl group located at the N-terminus of MA [56] , [57] , [58] . However , although the MA functional properties are conserved , neither of these motifs is present in the PFV-Gag-NtD . Further , the structure of MA is highly conserved amongst retroviruses , consisting of a four α-helix globular core and an associated fifth helix [59] , [60] , [61] , [62] , [63] , [64] , [65] . By comparison , our data reveals the PFV-Gag-NtD to be entirely unrelated comprising a mixed α/β protein with head and stalk domains . The dimeric organisation of FV-Gag-NtD is also not a conserved feature of orthoretroviral MAs . In HIV , myristoyl-MA promotes assembly and budding directly at the plasma membrane ( PM ) [56] and although it is unclear what the MA oligomerisation state is within immature and mature virions trimeric assemblies have been reported in vitro [61] , [63] . In the betaretroviruses that like FVs assemble intracellularly at the pericentriolar region [32] , [33] , only weak self-association of MA has been demonstrated [66] . By contrast , in the delta-retrovirus HTLV-1 the presence of stable disulphide linked dimers of Gag and MA in both immature particles and mature virions has been observed [67] . Thus , although FV-Gag-NtD and orthoretroviral MA have membrane-targeting roles in the late part of the viral life cycle , the differences in structure and organisation suggests the existence of different evolutionary pathways . Evidence for this notion also comes from sequence comparisons of the predicted Gag protein from FVs ranging from primate to sloth revealing they all share the same motifs and that they are unrelated to orthoretroviral Gag [11] . This implies there is one evolutionary pathway for the FVs with a single Gag protein and another for the orthoretroviruses in which the Gag precursor protein undergoes significant processing . Moreover , based on the observation of endogenous foamy virus in coelacanths , this divergence occurred more than 400 million years ago [13] . Foamy virus replication also has similarities with that of hepadnaviruses , including reverse transcription in the producer cell and an infectious DNA genome in the virion [24] , [25] . As there is no apparent structural homology with orthoretroviral Gag one possibility is that FV Gag may be related to a hepadnavirus structural protein . Inspection of capsid protein of hepadnavirus B ( Hep-B ) [68] reveals that Hep-B CA is an all-helical protein with a prominent 4-helix bundle making up the interface between CA dimers . This arrangement is reminiscent of the coiled-coil dimer interface of the PFV-Gag , However , in Hep-B the 4-helix bundle forms “spikes” that protrude from the exterior of the capsid shell . Given the arrangement of FV Gag with the N-terminal MA layer found at the greatest radius and the more C-terminal regions of Gag projecting to the virion interior [29] it seems unlikely that FV Gag is related to hepadnavirus CA . This further supports the notion that FV Gag-NtD is the product of convergent evolution that has driven the formation of a unique structure with properties of orthoretroviral MA and CA . The cytoplasmic targeting and retention signal ( CTRS ) found in the MA of betaretroviruses and in the Gag-NtD of FVs , promotes assembly in the pericentriolar region of the cell [32] , [33] , [34] . The consensus sequence in betaretroviruses spans residues Pro43 to Gly60 in MA of the archetypal betaretrovirus Mason-Pfizer monkey virus ( MPMV ) [69] , [70] . Within this sequence the majority of residues , Pro43 to Ile53 , constitute the loop that links helix α2 to helix α3 of MA whilst the remainder make up the first two turns of α3 [65] . In FVs the proposed CTRS constitutes residues 43 to 60 of the PFV-Gag-NtD [34] where residues Leu40 to Arg50 form the loop that links β1 to β2 and the remainder make up the β2 strand . Although the betaretroviral and FV CTRSs appear largely dissimilar , one common feature of both is a double aromatic motif G43WWGQ47 in PFV and P43WFPQ47 in MPMV . In both cases the sequences are located in the loop regions of the CTRS and comprise a structural motif consisting of a tight turn and a surface exposed aromatic and glutamine side chain , Figure 8 . In MPMV , mutation of the CTRS causes Gag to traffic as a monomer to the plasma membrane where assembly and production of infectious virus still occurs [70] . By comparison , absence of a functional CTRS in FVs completely abrogates assembly and whilst addition of a myristoylation signal to PFV facilitates Gag trafficking to the plasma membrane , infectious particles are not produced [54] , [71] . The severest effects on capsid formation and particle production were observed when alanine substitution mutations were introduced at Trp45 or Arg50 in the CTRS of PFV [53] . However , examination of the Gag-NtD structure now reveals that although Trp45 and Arg50 are part of the CTRS both are actually deeply buried in the core of the head domain . Arg50 also forms a number of important hydrogen bonds with neighbouring residues stabilising the interaction of the head domain with helix α5 immediately preceding the coiled-coil . Therefore , in these instances the severe mutational effects associated with alanine substitution can be likely attributed to destabilisation and/or misfolding of the Gag-NtD . However , mutation of the surface exposed Trp44 in the double aromatic motif does allow particle assembly but with a large reduction in both particle export and infectivity ( ∼105 fold ) [53] . In this case , given the exposure of the Trp44 sidechain , Figure 8 , the lack of particle egress might be attributed directly to loss of a di-hydrophobic motif dependent CTRS function causing mislocalisation or incorrect trafficking of assembled virions . FV egress requires interaction between the Gag and Env proteins to ensure correct membrane trafficking and viral budding . It seems likely that FV Gag becomes associated with Env through interaction with Env leader peptide ( Env-LP ) displayed on the cytosolic side of the ER and Trans-Golgi network ( TGN ) after core assembly at the pericentriolar region [31] , [33] . Env then directs the intracellular transport of the assembled particles to enable mature viruses to bud at the PM or sometimes into intracellular vacuoles . This interaction guarantees Env incorporation into virions and the loss of either interacting domain ( Gag-NtD or Env-LP ) results in the intracellular stranding of assembled FV capsids [30] , [52] . Mutations in the Env binding site of PFV-Gag-NtD have been shown to affect viral assembly , egress and infectivity . Of note is Leu17 that when substituted by serine results in loss of virus production , Figure 6A . However , our in vitro binding data , Figure 5 , reveal only modest reductions in affinity ( 2–5 fold ) when single serine substitutions are introduced into the Env binding site suggesting that the Leu17 to serine mutation may have effects prior to Gag-Env association . This notion is further supported by the fact that the mutant displays a phenotype similar to that of the Trp45 and Arg50 alanine mutations that disrupt the CTRS [53] . Examination of virion production and Env incorporation in Val14 and Leu21 serine substitution mutants reveals reduced levels in Val14 serine mutant but near wild type amounts in Leu21 particles . The small effects on virus production observed with the Leu21 and Val14 mutations also correlate well with the modest reductions in KA observed with the single-site mutations . This likely reflects the situation that recruitment of Env by a preassembled FV core rather than by Gag monomers is subject to the avidity effects of having many Gag binding sites arrayed on the core surface . Therefore , even under conditions of reduced binding the cores can still recruit enough Env to bud efficiently . However , whilst the effects on particle number and Gag-Env interaction are small , the Val14 and Leu21 serine mutations result in reduced infectivity , similar to when Leu17 is replaced by alanine [53] , suggesting that disruption of the Gag-Env interaction may also be detrimental for post-entry events in the target cells . In the structure , residues 7–16 of the Env leader peptide comprise the amphipathic α-helix bound in the Env binding site of the Gag-NtD and residues 1–6 provide intramolecular hydrogen bonding that stabilises the helical conformation . The affinity of the interaction , 1 . 5×105 M−1 , is comparable with the value of 0 . 65×105 M−1 reported for the interaction of residues 1–30 of the FFV Env leader peptide with the equivalent Gag-NtD [29] . Therefore , the hydrophobic interface observed in the structure likely represents the complete interaction between the leader peptide and FV Gag . The apolar character of the Env binding site is largely conserved among primate FVs although there is significant variation in the primary sequences of the α1 helix , Figure 1C . By contrast , the sequences of the N-terminal 13 residues of the Env leader peptide are largely invariant giving rise to the conserved motif [M-A-P-P-M- ( T/S/N ) -L- ( E/Q ) -Q-W-Φ-Φ-W] where Φ denotes a residue with a hydrophobic side chain . Our binding data show that removal of the first 4 residues ( MAPP ) along with Ala19 and His20 , not visible in the crystal structure , results in a significant reduction in Gag-Env binding , Figure 3 . It has also been demonstrated previously that the N-terminal four residues as well as the conserved tryptophan residues Trp10ENV and Trp13ENV , are essential for PFV egress [31] . Moreover , mutation of the equivalent conserved tryptophans in FFV greatly reduces the Gag-Env interaction in vitro [29] . The necessity for the N-terminal five residues is now apparent from the Gag-Env complex structure as many of the residues in the N-terminal extended region make polar contacts with Gag but also make intramolecular interactions with the Env helix to stabilise the conformation that binds to the Gag . The importance of the tryptophans is also apparent as they form part of the hydrophobic interface with Gag . Given the degree of conservation in the N-terminal of Env it is likely that this mode of interaction is a common feature of the Env-LP interaction with the Gag-NtD in other FVs . In orthoretroviruses , the viral core is enclosed by a hexameric lattice of CA assembled through combined homotypic and heterotypic interactions mediated by the amino-terminal ( CA-NtD ) and carboxy-terminal ( CA-CtD ) domains of CA [39] , [51] , [72] , [73] . In FVs , the structural organisation of the core is less characterised but two regions of FV-Gag required for assembly have been identified . Reminiscent of orthoretroviral CA-NtD and CA-CtD , the first corresponds to the Gag-NtD coiled-coil dimer defined in our structural studies [74] ( Figure 1 ) and the other found in the central region of FV-Gag ( Gag-CtD ) includes a conserved YXXLGL assembly motif [75] . In all likelihood the interior structural organisation of the FV virion is also formed by combinatorial heterotypic and homotypic protein-protein interactions mediated by these assembly domains , although the requirement for other regions of Gag , not yet identified , cannot be excluded . A further functional similarity of FV Gag-NtD and orthoretroviral CA-NtD is that both appear to be the target of Trim5α , mediated restriction Figure 7 , [36] , [37] and in orthoretroviruses , it is proposed that underlying hexagonal pattern of the assembled CA is recognised by a complementary hexagonal assembly of Trim5α in order to initiate the restriction process [76] , [77] . Presently , the overall arrangement of the Gag protein in an assembled FV is unknown but since the same species dependent Trim5α restriction of PFV and other FVs is apparent [35] , [36] the requirement for a lattice structure that arrays FV-Gag-NtD on the exterior of the FV core might also be expected . One possibility is that FV-Gag-NtD dimerisation combined with FV-Gag-CtD interactions generates a higher-order hexagonal Gag assembly targeted by Trim5α factors . However , given the obligate nature of the FV Gag-NtD dimer together with its organisation , dimensions and lack of structural homology with orthoretroviral CA it is difficult to envisage how a hexagonal assembly of equivalent spacing to that of the orthoretroviruses might be present in the FV particle . These observations raise the question of whether Trim5α might target other regular , or even irregular , molecular arrangements in addition to the hexagonal assemblies . Current models rely on a rather rigid overlapping of the orthoretroviral CA and Trim5α supramolecular assemblies . The inclusion of FVs in the cadre of Trim5α targets suggests there is potential flexibility in the pattern recognition receptor activity of Trim5α . Determining how this is accomplished awaits further structural and microscopic studies of the FV virion .
Human HT1080 [78] and 293T [79] cells were maintained in Dulbecco modified Eagle medium supplemented with 10% foetal calf serum and 1% penicillin and streptomycin . Restriction factors were delivered into cells using Moloney MLV ( MoMLV ) -based vectors produced by transfection of 293T cells . MoMLV-based delivery vectors were made by co-transfection of VSVG , pHIT60 , and pLgatewayIRESEYFP containing the restriction gene . FVs were produced by a four-plasmid PFV vector co-transfection system [80] , [81] in which pciSFV-1env ( providing Env ) , pcziPol PFV vector ( providing Pol ) , pMD9 ( a minimal vector genome with an EGFP marker gene ) , and a Gag-expressing construct were co-transfected . FV vector supernatants were harvested 48 h post-transfection , aliquoted , and stored at −80°C until further use . Subsequently , individual vector supernatant aliquots were pre-titrated on HT1080 cells using the EGFP marker gene and flow cytometric analysis . For the two-colour restriction assay described below , FV vector supernatants were then used at dilutions that resulted in 3 to 40% EGFP-positive HT1080 cells . A chimeric TRIM5α with the RBCC domain of human TRIM5α and the PRYSPRY domain of brown capuchin , referred to here as capuchin TRIM5α because the PRYSPRY domain determines restriction specificity , has been described previously [82] . A series of mutants of this factor in RING ( C15A/C18A ) , B-box 2 ( C95A/H98A , W115E and E118K/R119K ) and coiled-coil ( delta 130–231 ) were prepared by site directed mutagenesis . Preparation of the PFV and SFV packaging plasmids , PFV pcziGag4 ( PGWT ) and SFVmac pcziSG ( SGWT ) respectively , as well as chimeric PFV/SFV Gag packaging constructs , PSG-4 and SPG-4 , has been described previously [36] . The novel chimeric constructs PSG-5 and SPG-5 were generated by recombinant overlap PCR starting with PGWT and SGWT . PSG-5 contains amino acids 1–195 of PFV and 187–647 of SFV while SPG-5 encodes amino acids 1–186 of SFV and 196–648 of PFV . PFV Gag point mutants were generated in context of a the original PFV Gag packaging construct pcziGag4 [80] ( L17S/L21S ) , or a C-terminally HA-tagged variant thereof , pcziPG CLHH ( V14S , L17S , L21S ) . Restriction was determined by our previously described two-colour fluorescence activated cell sorter ( FACS ) assay [83] . Briefly , HT1080 cells were transduced with the MLV-based pLgatewayIRESEYFP retroviral vector carrying the restriction gene and an EYFP marker gene 2 days prior to challenging with FVs carrying the EGFP marker . The percentage of YFP positive cells ( i . e . restriction factor-positive cells ) that were EGFP positive ( i . e . FV infected ) was then determined by FACS . This was compared to the percentage of FV-infected cells ( EGFP positive ) in cells that did not express the restriction factor ( EYFP negative ) . A ratio that was less than 0 . 3 was taken to represent restriction , while a ratio greater than 0 . 7 indicated the absence of restriction . Cell culture supernatants containing recombinant viral particles were generated as described previously [84] . Briefly , 293T cells were co-transfected in 10 cm dishes with a Gag expression plasmid ( pcziGag4 or PG mutants thereof , as indicated ) , Env ( pcoPE ) , Pol ( pcoPP ) , and the transfer vector ( puc2MD9 ) at a ratio of 16∶1∶2∶16 using Polyethyleneimine ( PEI ) reagent and 16 µg DNA total . At 48 h post transfection ( p . t . ) cell-free viral vector supernatant was harvested using 0 . 45 µm sterile filters . For transduction efficiency analysis 2×104 HT1080 cells were plated in 12-well plates 24 h before infection . The target cells were incubated with 1 ml of plain cell-free viral supernatant or serial dilutions thereof for four to six hours . Determination of the percentage of eGFP-expressing cells was performed 72 h after infection by flow cytometry analysis and used for titre determination as previously described [85] . All transduction experiments were repeated at least three times . To compare the infectivity in repetitive experiments the titre obtained for wild type supernatants in individual experiments was set to an arbitrary value of 100% . The other values were then normalized as percentage of the wild type value . Viral protein expression in transfected cells and particle-associated protein composition was examined by Western blot analysis . Preparation of cell lysates from one transfected 10-cm cell culture dish was performed by incubation with 0 . 6 ml lysis buffer for 20 min at 4°C followed by centrifugation through a QIAshredder ( Qiagen ) . All protein samples were mixed with equal volumes of 2×PPPC ( 100 mM Tris-HCl; pH 6 . 8 , 24% glycerol , 8% SDS , 0 . 2% Bromophenol blue , 2% ß-mercaptoethanol ) prior to separation by SDS-PAGE using 7 . 5% polyacrylamide gels . Viral particles were concentrated from cell-free supernatant of transfected 293T cells by ultracentrifugation through a 20% sucrose cushion at 4°C and 25 , 000 rpm for 3 h in an SW32 rotor . The viral pellet was resuspended in phosphate-buffered saline ( PBS ) . Immunoblotting using polyclonal antisera specific for PFV Gag [86] or PFV Env leader peptide [75] was performed as previously described [31] . The chemiluminescence signal was digitally recorded using a LAS3000 imager and quantified using ImageGauge in the linear-range of the sample signal intensities as described previously [87] . The DNA sequences coding for PFV-Gag residues 1–179 ( PFV-Gag-NtD ) and FFV residues 1–154 ( FFV-Gag-NtD ) were amplified by PCR from template plasmids pcziGag4 and pcDWF003 containing the PFV and FFV Gag genes respectively . PCR products were inserted into a pET47b expression vector ( Novagen ) using ligation independent cloning in order to produce N-terminal His-tag fusions with 3C protease cleavage sites . The correct sequence of expression constructs was verified by automated DNA sequencing ( Beckman Coulter Genomics ) . His-tagged PFV- and FFV-Gag-NtD were expressed in the E . coli strain Rosetta 2 ( DE3 ) and purified using Ni-NTA affinity ( Qiagen ) and size exclusion chromatography on Superdex 200 ( GE healthcare ) . Selenium was incorporated into PFV-Gag-NtD by replacement of methionine with seleno-methionine in defined culture medium and by inhibition of methionine biosynthesis just prior to IPTG induction [88] . Verification of the processed N-terminal methionine , correct molecular mass and degree of selenium incorporation was obtained by electrospray ionisation mass-spectrometry . Peptides comprising residues 1–20 and 5–18 from the PFV-Env leader region were purchased HPLC purified from Pepceuticals Ltd . Size exclusion chromatography coupled multi-angle laser light scattering ( SEC-MALLS ) was used to determine the molar mass of FFV- and PFV-Gag-Ntd . Samples ranging from 1 . 5 to 12 . 0 mgml−1 were applied in a volume of 100 µl to a Superdex 200 10/300 GL column equilibrated in 20 mM Tris-HCl , 150 mM NaCl and 0 . 5 mM TCEP , pH 8 . 0 , at a flow rate of 0 . 5 ml/min . The scattered light intensity and the protein concentration of the column eluate were recorded using a DAWN-HELEOS laser photometer and OPTILAB-rEX differential refractometer respectively . The weight-averaged molecular mass of material contained in chromatographic peaks was determined from the combined data from both detectors using the ASTRA software version 6 . 0 . 3 ( Wyatt Technology Corp . , Santa Barbara , CA , USA ) . Sedimentation velocity experiments were performed in a Beckman Optima Xl-I analytical ultracentrifuge using conventional aluminium double sector centrepieces and sapphire windows . Solvent density and the protein partial specific volumes were determined as described [89] . Prior to centrifugation , samples were prepared by exhaustive dialysis against the buffer blank solution , 20 mM Tris-HCl , 150 mM NaCl and 0 . 5 mM TCEP , pH 7 . 5 . Centrifugation was performed at 50 , 000 rpm and 293 K in an An50-Ti rotor . Interference data were acquired at time intervals of 180 sec at varying sample concentration ( 0 . 5–2 . 5 mg/ml ) . Data recorded from moving boundaries was analysed using both a discrete species model and in terms of the size distribution functions C ( S ) using the program SEDFIT [90] , [91] , [92] . For analysis of Env peptide binding , sedimentation velocity experiments were conducted in 3 mm pathlength centrepieces using equimolar mixtures ( 75 µM ) of PFV-Gag-NtD and Env peptides . In these experiments , radial absorbance scans at 280 nm were also recorded along with the interference data . Sedimentation equilibrium experiments were performed in a Beckman Optima XL-I analytical ultracentrifuge using charcoal filled Epon six-channel centrepieces in an An-50 Ti rotor . Prior to centrifugation , samples were dialyzed exhaustively against the buffer blank , 20 mM Tris-HCl , pH 7 . 5 , 150 mM NaCl , 0 . 5 mM TCEP . After centrifugation for 18 h , interference data was collected 2-h intervals until no further change in the profiles was observed . The rotor speed was then increased and the procedure repeated . Data were collected on samples of different concentrations of FFV- and PFV-Gag-NtD ( 14–100 µM ) at three speeds and the program SEDPHAT [93] , [94] was used to determine weight-averaged molecular masses by nonlinear fitting of individual multi-speed equilibrium profiles ( A versus r ) to a single-species ideal solution model . Inspection of these data revealed that the molecular masses showed no significant concentration dependency and so global fitting incorporating the data from multiple speeds and multiple sample concentrations was applied to extract a final weight-averaged molecular mass . Data were analysed using a general binding expression , Eq . 1 . This expression relates the association constant Ka to the fraction of bound peptide θ ( θ = [PL]/[Lt] ) in terms of the total concentrations of peptide [Lt] and protein [Pt] and is a modification of the formulae employed in [95] , [96] . ( 1 ) In sedimentation velocity experiments θ was determined from the integrated absorbance of the 3S species in the C ( S ) function that best fits the sedimentation data . As equimolar ratios of peptide and protein were employed ( [Lt] = [Pt] ) Eq . 1 can be simplified and equilibrium association constants determined from Eq . 2 . ( 2 ) ITC was carried out using an ITC-200 calorimeter ( MicroCal ) . Briefly , PFV-Gag-NtD was prepared by were dialysis against 25 mM Na-phosphate pH 6 . 55 , 100 mM NaCl , 0 . 5 mM TCEP . A typical experiment involved 20 injections of 1 mM Env peptide in the injection syringe into 50 µM PFV-Gag-NtD in the sample cell . Data was analysed using the Origin-based software provided by the manufacturers . PFV-Gag-NtD was crystallised using hanging drop vapour diffusion . Typically , A 10 mg/ml solution of PFV-Gag-NtD in 150 mM NaCl , 5% glycerol , 10 mM Tris-HCl , pH 8 . 0 was mixed with an equal volume of crystallisation solution containing 16% PEG 6000 ( w/v ) , 12% ethylene glycol , 0 . 03 M MgCl2 hexahydrate and suspended over a reservoir of the crystallisation solution . Crystals appeared within 14 days and were transferred into fresh crystallisation solution supplemented with 20% glycerol and flash-frozen in liquid nitrogen . The crystals belong to the space group P21 with one copy of the PFV dimer in the asymmetric unit . Seleno-methionine derived protein was crystallized under the same conditions . Crystals of the PFV-Gag-NtD-Env complex were also grown by vapour diffusion by mixing 500 µM 1∶1 complex in 150 mM NaCl , 5% glycerol , 10 mM Tris-HCl , pH 8 . 0 with an equal volume a crystallisation solution containing 10% PEG 4000 ( w/v ) , 20% glycerol , 0 . 03 M MgCl2 , 0 . 03 M CaCl2 , 0 . 1 M Tris-Bicine pH 8 . 5 . Crystals appeared within 2 days and were harvested into fresh crystallisation solution supplemented with 20% glycerol and flash-frozen in liquid nitrogen prior to data collection . Crystals of the complex also belong to the space group P21 but with two copies of the PFV dimer-peptide complex in the asymmetric unit . The structure of PFV-Gag-NtD was solved by single wavelength anomalous diffraction ( SAD ) using a dataset recorded at 0 . 9791 Å at 100 K on beamline I03 at the Diamond Light Source ( Didcot , UK ) using crystals of the seleno-methionine substituted protein . Data was processed using the HKL program package [97] and 13 selenium atoms were located by SAD methods in PHENIX [98] . Further density modification in PHENIX resulted in a figure of merit of 0 . 79 and a map of sufficient quality for a near complete model to be built using Arp/Warp [99] . The model was completed by iterative rounds of refinement and model building in PHENIX and COOT [100] . TLS groups were included in final round of refinement as determined by TLSMD [101] . The structure was refined to a final Rwork/Rfree of 17 . 2/23 . 0 respectively and has good geometry with 98 . 8% of residues in the preferred region of the Ramachandran plot , only 1 . 2% in the additionally allowed region and no outliers . Details of crystal parameters and data refinement statistics are presented in Table 1 . Data for the PFV-Gag-NtD-Env complex was collected at 100 K on beamline I03 and processed and scaled in space group P21 using XDS/XSCALE [102] . The structure was solved by molecular replacement using Phaser [103] with the Gag-NtD dimer used as a search model to locate the two copies of the complex in the asymmetric unit . The model was completed by iterative rounds of TLS based refinement and model building using Refmac5 [104] and COOT . TLS groups were defined using TLSMD . The structure was refined to a final Rwork/Rfree of 22 . 6/27 . 1 in which 98 . 8% of residues lie within preferred regions of the Ramachandran plot and the remaining 1 . 2% residues lie within the additionally allowed region . The crystal and refinement parameters are given in Table 1 . The coordinates and structure factors of PFV-Gag-NtD and PFV-Gag-NtD-Env complex have been deposited in the Protein Data Bank under accession numbers 4JNH and 4JMR respectively . | Foamyviruses ( FVs ) or spuma-retroviruses derive their name from the cytopathic effects they cause in cell culture . By contrast , infection in humans is benign and FVs have entered the human population through zoonosis from apes resulting in the emergence of Prototypic Foamyvirus ( PFV ) . Like all retroviruses FVs contain gag , pol and env structural genes and replicate through reverse-transcription and host genome integration . Gag , the major structural protein , is required for genome packaging , virion assembly , trafficking and egress . However , although functionally equivalent , FV and orthoretroviral Gag share little sequence homology and it is unclear how they perform the same function . Therefore , to understand more about the relationship between FV and orthoretroviral replication we have carried out structural/virological studies of PFV Gag . We present the structure of Gag-NtD , a unique domain found only in FV Gag and show that despite functional equivalence , Gag-NtD is entirely structurally unrelated to orthoretroviral Gag . We also provide the molecular details of an essential interaction between Gag-NtD and the FV Envelope and demonstrate that Gag-NtD contains the determinants of Trim5α restriction . Our findings are discussed in terms of evolutionary convergence of retroviruses and the implications of alternative arrangements of Gag on pattern recognition by viral restriction factors . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biomacromolecule-ligand",
"interactions",
"medicine",
"protein",
"interactions",
"viral",
"transmission",
"and",
"infection",
"viral",
"classification",
"macromolecular",
"assemblies",
"host-pathogen",
"interaction",
"microbiology",
"viral",
"structure",
"protein",
"structure",
"infectious",
"diseases",
"zoonoses",
"viral",
"core",
"proteins",
"structural",
"proteins",
"biology",
"vectors",
"and",
"hosts",
"viral",
"evolution",
"biochemistry",
"viral",
"envelope",
"transmembrane",
"proteins",
"virology",
"evolutionary",
"biology",
"evolutionary",
"processes"
] | 2013 | A Unique Spumavirus Gag N-terminal Domain with Functional Properties of Orthoretroviral Matrix and Capsid |
Infection of host cells by Toxoplasma gondii is an active process , which is regulated by secretion of microneme ( MICs ) and rhoptry proteins ( ROPs and RONs ) from specialized organelles in the apical pole of the parasite . MIC1 , MIC4 and MIC6 assemble into an adhesin complex secreted on the parasite surface that functions to promote infection competency . MIC1 and MIC4 are known to bind terminal sialic acid residues and galactose residues , respectively and to induce IL-12 production from splenocytes . Here we show that rMIC1- and rMIC4-stimulated dendritic cells and macrophages produce proinflammatory cytokines , and they do so by engaging TLR2 and TLR4 . This process depends on sugar recognition , since point mutations in the carbohydrate-recognition domains ( CRD ) of rMIC1 and rMIC4 inhibit innate immune cells activation . HEK cells transfected with TLR2 glycomutants were selectively unresponsive to MICs . Following in vitro infection , parasites lacking MIC1 or MIC4 , as well as expressing MIC proteins with point mutations in their CRD , failed to induce wild-type ( WT ) levels of IL-12 secretion by innate immune cells . However , only MIC1 was shown to impact systemic levels of IL-12 and IFN-γ in vivo . Together , our data show that MIC1 and MIC4 interact physically with TLR2 and TLR4 N-glycans to trigger IL-12 responses , and MIC1 is playing a significant role in vivo by altering T . gondii infection competency and murine pathogenesis .
Toxoplasma gondii is a coccidian parasite belonging to the phylum Apicomplexa and is the causative agent of toxoplasmosis . This protozoan parasite infects a variety of vertebrate hosts , including humans with about one-third of the global population being chronically infected [1] . Toxoplasmosis can be fatal in immunocompromised individuals or when contracted congenitally [1] , and is considered the second leading cause of death from foodborne illnesses in the United States [2] . T . gondii invades host cells through an active process that relies on the parasite actinomyosin system , concomitantly with the release of microneme proteins ( MICs ) and rhoptry neck proteins ( RONs ) from specialized organelles in the apical pole of the parasite [3] . These proteins are secreted by tachyzoites [4 , 5] and form complexes composed of soluble and transmembrane proteins . Some of the MICs act as adhesins , interacting tightly with host cell-membrane glycoproteins and receptors , and are involved in the formation of the moving junction [6] . This sequence of events ensures tachyzoite gliding motility , migration through host cells , invasion and egress from infected cells [4 , 7] . Among the released proteins , MIC1 , MIC4 , and MIC6 form a complex that , together with other T . gondii proteins , plays a role in the adhesion and invasion of host cells [8 , 9] , contributing to the virulence of the parasite [10 , 11] . Several studies have shown that host-cell invasion by apicomplexan parasites such as T . gondii involves carbohydrate recognition [12–15] . Interestingly , MIC1 and MIC4 have lectin domains [11 , 16–18] that recognize oligosaccharides with sialic acid and D-galactose in the terminal position , respectively . Importantly , the parasite’s Lac+ subcomplex , consisting of MIC1 and MIC4 , induces adherent spleen cells to release IL-12 [17] , a cytokine critical for the protective response of the host to T . gondii infection [19] . In addition , immunization with this native subcomplex , or with recombinant MIC1 ( rMIC1 ) and MIC4 ( rMIC4 ) , protects mice against experimental toxoplasmosis [20 , 21] . The induction of IL-12 is typically due to detection of the pathogen by innate immunity receptors , including members of the Toll-like receptor ( TLR ) family , whose stimulation involves MyD88 activation and priming of Th1 responses , which protects the host against T . gondii [19 , 22] . It is also known that dysregulated expression of IL-12 and IFN-γ during acute toxoplasmosis can drive a lethal immune response , in which mice succumb to infection by severe immunopathology , the result of insufficient levels of IL-10 and/or a collapse in the regulatory CD4+Foxp3+ T cell population [23 , 24] . Interestingly , regarding the innate immune receptors associated with IL-12 response during several infections , the extracellular leucine-rich repeat domains of TLR2 and TLR4 contain four and nine N-glycans , respectively [25] . Therefore , we hypothesized that MIC1 and MIC4 bind TLR2 and TLR4 N-glycans on antigen-presenting cells ( APCs ) and , through this interaction , trigger immune cell activation and IL-12 production . To investigate this possibility , we assayed the ability of rMIC1 and rMIC4 to bind and activate TLR2 and TLR4 . Using several strategies , we demonstrated that TLR2 and TLR4 are indeed critical targets for both MIC1 and MIC4 . These parasite and host cell structures establish lectin-carbohydrate interactions that contribute to the induction of IL-12 production by innate immune cells , and we show here that the MIC1 lectin promotes T . gondii infection competency and regulates parasite virulence during in vivo infection .
The native MIC1/4 subcomplex purified from soluble T . gondii antigens has lectin properties , so we investigated whether their recombinant counterparts retained the sugar-binding specificity . The glycoarray analysis revealed the interactions of: i ) the Lac+ subcomplex with glycans containing terminal α ( 2–3 ) -sialyl and β ( 1–4 ) - or β ( 1–3 ) -galactose; ii ) , rMIC1 with α ( 2–3 ) -sialyl residues linked to β-galactosides; and iii ) of rMIC4 with oligosaccharides with terminal β ( 1–4 ) - or β ( 1–3 ) -galactose ( Fig 1A ) . The combined specificities of the individual recombinant proteins correspond to the dual sugar specificity of the Lac+ fraction , demonstrating that the sugar-recognition properties of the recombinant proteins are consistent with those of the native ones . Based on the sugar recognition selectivity of rMIC1 and rMIC4 , we tested two oligosaccharides ( α ( 2–3 ) -sialyllactose and lacto-N-biose ) for their ability to inhibit the interaction of the MICs with the glycoproteins fetuin and asialofetuin [26] . Sialyllactose inhibited the binding of rMIC1 to fetuin , and lacto-N-biose inhibited the binding of rMIC4 to asialofetuin ( Fig 1B ) . To ratify the carbohydrate recognition activity of rMIC1 and rMIC4 , we generated point mutations into the carbohydrate recognition domains ( CRDs ) of the rMICs to abolish their lectin properties [11 , 18 , 27] . These mutated forms , i . e . rMIC1-T126A/T220A and rMIC4-K469M , lost their capacity to bind to fetuin and asialofetuin , respectively ( Fig 1B ) , having absorbance as low as that in the presence of the specific sugars . Thus , our results indicate that rMIC1 and rMIC4 maintained their lectin properties , and that the CRD function can be blocked either by competition with specific sugars or by targeted mutations . We have previously demonstrated that the native Lac+ subcomplex stimulates murine adherent spleen cells to produce proinflammatory cytokines [20] . We evaluated whether recombinant MIC1 and MIC4 retained this property and exerted it on BMDCs and BMDMs . BMDCs ( Fig 2A–2D ) and BMDMs ( Fig 2E–2H ) produced high levels of the proinflammatory cytokines IL-12 ( Fig 2A and 2E ) , TNF-α ( Fig 2B and 2F ) , and IL-6 ( Fig 2C and 2G ) . This was not attributable to residual LPS contamination as the recombinant protein assays were done in the presence of polymyxin B , and LPS levels were less than 0 . 5ng/ml [see Materials and Methods section] . Although conventional CD4+ Th1 cells are known to be the major producers of IL-10 during murine T . gondii infection [28] , we also found that rMIC1 and rMIC4 induced the production of this cytokine by BMDCs ( Fig 2D ) and BMDMs ( Fig 2H ) . We verified that the two recombinant proteins induced the production of similar levels of IL-12 , TNF-α , and IL-6 by both BMDCs ( Fig 2A–2C ) and BMDMs ( Fig 2E–2G ) . Both MICs induced the production of similar levels of IL-10 in BMDCs ( Fig 2D ) ; however , BMDMs produced significantly higher levels of IL-10 when stimulated with rMIC1 than when stimulated with rMIC4 ( Fig 2H ) . These cytokine levels were similar to those induced by the TLR4 agonist LPS . Thus , recombinant MIC1 and MIC4 induce a proinflammatory response in innate immune cells , which is consistent with the results obtained for the native Lac+ subcomplex [20] . To investigate the mechanisms through which T . gondii MIC1 and MIC4 stimulate innate immune cells to produce cytokines , we assessed whether these MICs can activate specific TLRs . To this end , BMDMs from WT , MyD88-/- , TRIF-/- , TLR2-/- , TLR4-/- , or TLR2/4 DKO mice , as well as HEK293T cells transfected with TLR2 or TLR4 , were cultured in the presence or absence of rMIC1 and rMIC4 for 48 hours . The production of IL-12 by BMDMs ( Fig 3A–3I ) and IL-8 by HEK cells ( Fig 3J and 3K ) were used as an indicator of cell activation . IL-12 production by BMDMs from MyD88-/- , TRIF-/- , TLR2-/- , and TLR4-/- mice was lower than that of BMDMs from WT mice ( Fig 3A–3D ) ; no IL-12 was detected in cultures of TLR2/4 DKO mice cells stimulated with either rMIC1 or rMIC4 ( Fig 3E ) . These results show that TLR2 and TLR4 are both relevant for the activation of macrophages induced by rMIC1 and rMIC4 . The residual cytokine production observed in macrophages from TLR2-/- or MyD88-/- mice may be the result of activation of TLR4 ( Fig 3A and 3C ) , and vice versa; e . g . , the residual IL-12 levels produced by macrophages from TLR4-/- mice may be the result of TLR2 activation . The finding that MICs fail to induce IL-12 production in DKO mice BMDMs suggests that cell activation triggered by T . gondii MIC1 or MIC4 does not require the participation of other innate immunity receptors beyond TLR2 and TLR4 . Nevertheless , because parasite components such as DNA or profilin engage TLR9 , TLR11 , and TLR12 to produce IL-12 in macrophages [19 , 22 , 29] , we investigated the involvement of these receptors , as well as TLR3 and TLR5 , in the response to rMIC1 or rMIC4 . BMDMs from TLR3-/- , TLR5-/- , TLR9-/- , and TLR11/12 DKO mice stimulated with rMIC1 or rMIC4 produced similar levels of IL-12 as cells from WT ( Fig 3F–3I ) , indicating that the activation triggered by rMIC1 or rMIC4 does not depend on these receptors . Additionally , stimulation of HEK cells transfected with human TLR2 ( Fig 3J ) or TLR4 ( Fig 3K ) with optimal concentrations of rMIC1 ( S1A and S1C Fig ) and rMIC4 ( S1B and S1D Fig ) induced IL-8 production at levels that were higher than those detected in the absence of stimuli ( medium ) , and similar to those induced by the positive controls . Finally , by means of a pull-down experiment , we demonstrated a physical interaction between rMIC1 and TLR2 or TLR4 and between rMIC4 and TLR2 or TLR4 ( Fig 3L ) . We hypothesized that in order to trigger cell activation , rMIC1 and rMIC4 CRDs target oligosaccharides of the ectodomains of TLR2 ( four N-linked glycans ) [25] and TLR4 ( nine N-linked glycans ) [30] . This hypothesis was tested by stimulating BMDCs ( Fig 4A ) and BMDMs ( Fig 4B ) from WT mice with intact rMIC1 and rMIC4 or with the mutated forms of these microneme proteins , namely rMIC1-T126A/T220A and rMIC4-K469M , which lack carbohydrate binding activity [11 , 18 , 27] . IL-12 levels in culture supernatants were lower upon stimulation with rMIC1-T126A/T220A or rMIC4-K469M , showing that WT induction of cell activation requires intact rMIC1 and rMIC4 CRDs . The same microneme proteins were used to stimulate TLR2-transfected HEK293T cells ( Fig 4C ) , and similarly , lower IL-8 production was obtained in response to mutated rMIC1 or rMIC4 compared to that seen in response to intact proteins . These observations demonstrated that rMIC1 and rMIC4 CRDs are also necessary for inducing HEK cell activation . We used an additional strategy to examine the ability of rMIC1 and rMIC4 to bind to TLR2 N-glycans . In this approach , HEK cells transfected with the fully N-glycosylated TLR2 ectodomain or with the TLR2 glycomutants [25] were stimulated with a control agonist ( FSL-1 ) or with rMIC1 or rMIC4 . HEK cells transfected with any TLR2 form , except those expressing totally unglycosylated TLR2 ( mutant Δ1 , 2 , 3 , 4 ) , were able to respond to FSL-1 ( Fig 4D ) , a finding that is consistent with the previous report that the Δ1 , 2 , 3 , 4 mutant is not secreted by HEK293T cells [25] . Cells transfected with TLR2 lacking only the first or the third N-glycan ( mutant Δ1; Δ3 ) responded to all stimuli . The response to the rMIC1 stimulus was significantly reduced in cells transfected with five different TLR2 mutants , lacking some combination of the second , third , and fourth N-glycans ( Fig 4D ) . Moreover , rMIC4 stimulated IL-8 production was significantly reduced in cells transfected with the mutants lacking some combination of the third and fourth N-glycans ( Fig 4D ) . These results indicate that T . gondii MIC1 and MIC4 use their CRDs to induce TLR2- and TLR4-mediated cell activation . Among the TLR2 N-glycans , the rMIC1 CRD likely targets the second , third , and fourth glycan , whereas the rMIC4 CRD targets only the third and fourth . Additionally , our findings suggested that TLR2 and TLR4 activation is required to enhance the production of IL-12 by APCs following rMIC stimulation . Because IL-12 production is induced by rMICs that engage TLR2 and TLR4 N-glycans expressed on innate immune cells , we investigated whether such production is impaired when APCs are infected with T . gondii lacking MIC1 and/or MIC4 proteins , as well as complemented strains expressing mutant versions of these proteins that fail to bind TLR2 or TLR4 carbohydrates . We generated Δmic1 and Δmic4 strains in an RH strain expressing GFP and Luciferase using CRISPR/Cas9 to replace the endogenous MIC gene with the drug-selectable marker HPT ( HXGPRT–hypoxanthine-xanthine-guanine phosphoribosyl transferase ) ( Fig 5A and 5B ) . We then complemented MIC deficient parasites with mutated versions expressing an HA-tag , thus generating the Δmic1::MIC1-T126A/T220AHA or Δmic4::MIC4-K469MHA strains ( Fig 5A ) that expressed endogenous levels of MIC1 and MIC4 as confirmed by Western Blotting ( Fig 5C ) . IL-12 secretion by BMDCs and BMDMs infected with WT , Δmic1 , Δmic1::MIC1-T126A/T220A , Δmic4 and Δmic4::K469M parasites was assessed at 24 hours post infection . All mutant strains ( Δmic1 , Δmic1::MIC1-T126A/T220A , Δmic4 and Δmic4::K469M ) induced lower IL-12 secretion by BMDCs ( Fig 5D ) and BMDMs ( Fig 5E ) compared to that induced by WT parasites , indicating that engagement of TLR2 and TLR4 cell surface receptors by the MIC lectin-specific activity led to an early release of IL-12 . Using flow cytometry , we confirmed that parasites deficient in MIC1or MIC4 , or mutated in their carbohydrate recognition domain resulted in lower intracellular IL-12 production than WT infected BMDCs ( Fig 5F–5H ) . Interestingly , the Toxo+ BMDCs presented the same level of intracellular IL-12 , independent of the T . gondii strain infected ( Fig 5F and 5H ) . Whereas the Toxo- BMDCs produced less IL-12 when they were infected with knockout or CRD-mutated T . gondii compared to WT-infected cells ( Fig 5G and 5H ) . Taken altogether , these results indicate that MIC1 and MIC4 induce IL-12 production in innate immune cells during in vitro T . gondii infection . It is known that other parasite factors act as IL-12 inducers , such as profilin , which is a TLR11 and TLR12 agonist [29 , 31] , or GRA7 [32] , GRA15 [33] , and GRA24 [34] , which directly trigger intracellular signalling pathways in a TLR-independent manner , and these likely account for the majority of IL-12 released after 24 hours of intracellular infection . Given the importance of MIC1 and MIC4 as lectins that engage TLR2 and TLR4 N-glycans to induce increased levels of IL-12 release during T . gondii in vitro infection , we investigated the biological relevance of these proteins during in vivo infection . Mice were injected with 50 tachyzoites of RH WT , Δmic1 , Δmic1::MIC1-T126A/T220A , Δmic4 and Δmic4::MIC4-K469M strains into the peritoneum of CD-1 outbred mice , a lethal dose that causes acute mortality . The survival curve showed that parasites deficient in MIC1 ( Δmic1 group ) or mutated to remove MIC1 lectin binding activity ( Δmic1::MIC1-T126A/T220A group ) were less virulent , resulting in a slight , but significant ( p = 0 . 0017 ) increase in mouse survival ( 12 days post-infection ) compared to WT infected mice that all succumbed to infection by day 10 ( Fig 6A ) . This was not the result of a difference in parasite load , which was equivalent across all T . gondii-infected mice at Day 5 ( Fig 6D and 6I ) . Whereas , the absence of the MIC4 gene or MIC4 lectin activity did not change the survival curve ( Fig 6E ) indicating that MIC4 is less relevant than MIC1 during in vivo infection . Acute mortality in CD-1 mice infected with Type I T . gondii is related to the induction of a cytokine storm , mediated by high levels of IFN-γ production . Thus , we measured systemic levels of IFN-γ and IL-12 in mice infected with WT , Δmic1 , Δmic1::MIC1-T126A/T220A , Δmic4 and Δmic4::MIC4-K469M strains . According to Kugler et al . ( 2013 ) , the peak of systemic IL-12p40 and IFN-γ during ME49-T . gondii infection is between days 5–6 post-infection , therefore , we measured these cytokines in the serum of CD-1-infected mice at day 5 . Mice infected with Δmic1 or Δmic1::MIC1-T126A/T220A strains had 3–5 fold lower systemic levels of IL-12 ( Fig 6B; p = 0 . 016 ) and IFN-γ ( Fig 6C; p ≤0 . 0002 ) than WT infected mice . In contrast , mice infected with parasites lacking the MIC4 gene , or those expressing the mutant version of MIC4 showed no difference in IL-12 ( Fig 6F ) or IFN-γ ( Fig 6G ) compared to WT infected mice . Hence , only MIC1 altered systemic levels of key cytokines induced during T . gondii in vivo infection , and mice survived longer with lower systemic levels of cytokines typically associated with acute mortality . To formally show that MIC1 alters systemic levels of pro-inflammatory cytokines associated with acute mortality , we complemented Δmic1 parasites at the endogenous locus with a Type I allele of MIC1 expressing an HA tag ( MIC1HA ) . Western blotting for either MIC1 or HA expression showed WT levels of MIC1 expression in the complemented parasites Δmic1::MIC1HA ( Fig 7A ) . The complemented strain restored WT virulence kinetics during in vivo infection and all mice died acutely , in contrast to Δmic1 or Δmic1::MIC1-T126A/T220A parasites , that had a slight , but significant delay in their acute mortality kinetics ( Fig 7B; p = 0 . 0082 ) . Systemic levels of IFN-γ ( Fig 7C ) and parasite load ( Fig 7D and 7E ) from mice infected with the complemented strain were indistinguishable from WT . To better resolve the apparent difference in acute mortality , parasites were injected into the right footpad to monitor mouse weight loss and survival kinetics [35] . Mice infected locally in the footpad with Δmic1 survived significantly longer , or did not die ( Fig 7G; p = 0 . 0002 ) , and lost less weight during acute infection ( Fig 7F ) than those infected with WT or Δmic1::MIC1 complemented parasites . Further , mice infected with Δmic1::MIC1-T126A/T220A parasites that fail to bind TLR2 and TLR4 N-glycans in vivo also lost less weight and survived significantly longer than WT or Δmic1::MIC1 complemented parasites ( Fig 7F and 7G ) . In conclusion , our results suggest that MIC1 operates in two distinct ways; as an adhesin protein that promotes parasite infection competency , and as a lectin that engages TLR N-glycans to induce a stronger proinflammatory immune response , one that is unregulated and results in acute mortality upon RH infection of CD-1 mice .
In this study , we report a new function for MIC1 and MIC4 , two T . gondii microneme proteins involved in the host-parasite relationship . We show that rMIC1 and rMIC4 , by interacting directly with N-glycans of TLR2 and TLR4 , trigger a noncanonical carbohydrate recognition-dependent activation of innate immune cells . This results in IL-12 secretion and the production of IFN-γ , a pivotal cytokine that mediates parasite clearance and the development of a protective T cell response [19 , 22] , but in some cases promotes a dysregulated cytokine storm and acute mortality , as seen during RH infection of CD-1 mice [36] . This MIC-TLR activation event explains , at least in part , the resistance conferred by rMIC1 and rMIC4 administration against experimental toxoplasmosis [20 , 21] . T . gondii tachyzoites express microneme proteins either on their surface or secrete them in their soluble form . These proteins may form complexes , such as those of MIC1 , MIC4 , and MIC6 ( MIC1/4/6 ) , in which MIC6 is a transmembrane protein that anchors the two soluble molecules MIC1 and MIC4 [8] . Genetic disruption of each one of these three genes does not interfere with parasite survival [8] nor its interaction with , and attachment to , host cells [10]; however , MIC1 has been shown to play a role in invasion and contributes to virulence in mice [10] . We previously isolated soluble MIC1/4 , a lactose-binding complex from soluble T . gondii antigens ( STAg ) [17] , and its lectin activity was confirmed by the ability of MIC1 to bind sialic acid [9] and MIC4 to β-galactose [18] . We also reported that MIC1/4 stimulates adherent splenic murine cells to produce IL-12 at levels as high as those induced by STAg [20] . Recently , it was also demonstrated that MIC1 , MIC4 and MIC6 are capable of inducing IFN-γ production from memory T cells in mice chronically infected with T . gondii [37] . Our data herein shows that MIC1/4 binds to and activates TLRs via a novel lectin-carbohydrate interaction , rather than by its cognate receptor-ligand binding groove , establishing precisely how the interactions of microneme protein ( s ) with defined glycosylated receptor ( s ) expressed on the host cell surface are capable of altering innate priming of the immune system . To formally demonstrate the MIC1/MIC4 binding to glycosylated TLR cell surface receptors we generated recombinant forms of MIC1 and MIC4 , which retained their specific sialic acid- and β-galactose-binding properties as indicated by the results of their binding to fetuin and asialofetuin as well as the glycoarray assay . Both recombinant MIC1 and MIC4 triggered the production of proinflammatory and anti-inflammatory cytokines in DCs and macrophages via their specific recognition of TLR2 and TLR4 N-glycans , as well as by signaling through MyD88 and , partially , TRIF . Importantly , our results establish how binding of rMIC1 and rMIC4 to specific N-glycans present on TLR2 and TLR4 induces cell activation through this novel lectin-carbohydrate interaction . The ligands for MIC1 and MIC4 , α2-3-sialyllactosamine and β1-3- or β1-4-galactosamine , respectively , are terminal N-glycan residues found on a wide-spectrum of mammalian cell surface-associated glycoconjugates . Thus , it is possible that additional lectin-carbohydrate interactions may exist between MIC1/4 and other cell surface receptors beyond TLR2 and TLR4 . Such interactions likely evolved to facilitate adhesion and promote the infection competency of a wide-variety of host cells infected by T . gondii , further underscoring how these proteins exist as important virulence factors [10] beyond immune priming . However , it is the immunostimulatory capacity of rMIC1 and rMIC4 to target N-glycans on the ectodomains of TLR2 and TLR4 that likely rationalizes how these microneme proteins function as a double-edged sword during T . gondii infection . Mice infected by Type I strains die acutely due to a failure to regulate the cytokine storm induced by high levels of IL-12 and IFN-γ[38 , 39] . In this study , T . gondii Type I strains engineered to be deficient in MIC1 or defective in binding TLR2/4 N-glycans lost less weight , survived significantly longer , and produced less IL-12 and IFN-γ . Future studies that test whether the immunostimulatory effect of MIC1/4 alters the pathogenesis and cyst burden of Type II strains of T . gondii should be pursued to formally demonstrate that Type II parasites rely on MIC1/4 induction of Th1-biased cytokines in order to limit tachyzoite proliferation and induce a life-long persistent bradyzoite infection . Several pathogens are known to synthesize lectins , which are most frequently reported to interact with glycoconjugates on host cells to promote adherence , invasion , and colonization of tissues [40–43] . Nonetheless , there are currently only a few examples of lectins from pathogens that recognize sugar moieties present in TLRs and induce IL-12 production by innate immune cells . Paracoccin , a GlcNAc-binding lectin from the human pathogen Paracoccidioides brasiliensis , induces macrophage polarization towards the M1 phenotype [44] and the production of inflammatory cytokines through its interaction with TLR2 N-glycans [45] . Furthermore , the galactose-adherence lectin from Entamoeba histolytica activates TLR2 and induces IL-12 production [46] . In addition , the mammalian soluble lectin SP-A , found in lung alveoli , interacts with the TLR2 ectodomain [47] . The occurrence of cell activation and IL-12 production as a consequence of the recognition of TLR N-glycans has also been demonstrated using plant lectins with different sugar-binding specificities [48 , 49] . The binding of MIC1 and MIC4 , as well as the lectins above , to TLR2 and TLR4 may be associated with the position of the specific sugar residue present on the receptor’s N-glycan structure . Since the N-glycan structures of TLR2 and TLR4 are still unknown , we assume that the targeted MIC1 and MIC4 residues , e . g . sialic acid α2-3-linked to galactose β1-3- and β1-4-galactosamines , are appropriately placed in the receptors’ oligosaccharides to allow the recognition phenomenon and trigger the activation of innate immune responses . Several T . gondii proteins have previously been shown to activate innate immune cells in a TLR-dependent manner , but independent of sugar recognition . This is the case for profilin ( TgPRF ) , which is essential for the parasite’s gliding motility based on actin polymerization; it is recognized by TLR11 [29] and TLR12 [31 , 50] . In addition , T . gondii-derived glycosylphosphatidylinositol anchors activate TLR2 and TLR4 [51] , and parasite RNA and DNA are ligands for TLR7 and TLR9 , respectively [19 , 22 , 50] . The stimulation of all of these TLRs culminate in MyD88 activation which results in IL-12 production [19 , 22] . Several other T . gondii secreted effector proteins regulate the production of proinflammatory cytokines such as IL-12 , independent of TLRs . For example , the dense granule protein 7 ( GRA7 ) induces MyD88-dependent NF-kB activation , which facilitates IL-12 , TNF-α , and IL-6 production [32] . MIC3 is reported to induce TNF-α secretion and macrophage M1 polarization [52] , whereas GRA15 expressed by Type II strains activates NF-kB , promoting the release of IL-12 [33] , and GRA24 triggers the autophosphorylation of p38 MAP kinase and proinflammatory cytokine and chemokine secretion [34] . In contrast , TgIST interferes with IFN-γ induction by actively inhibiting STAT1-dependent proinflammatory gene expression indicating that the parasite is capable of both activating as well as inhibiting effector arms of the host immune response to impact its pathogenesis in vivo [53] . Thus , multiple secretory effector proteins of T . gondii , including MIC1 and MIC4 , appear to work in tandem to ultimately promote protective immunity by either inducing or dampening the production of proinflammatory cytokines , the timing of which is central to controlling both the parasite’s proliferation during the acute phase of infection and the induction of an effective immune response capable of establishing a chronic infection [19] . Our results regarding soluble MIC1 and MIC4 confirmed our hypothesis that these two effector proteins induce the innate immune response against T . gondii through TLR2- and TLR4-dependent pathways . This is consistent with previous studies that highlight the importance of TLR signaling , as well as the MyD88 adapter molecule , as essential for conferring resistance to T . gondii infection [29 , 51 , 54 , 55] . In addition , we show that both MIC1 and MIC4 on the parasite surface contribute to the secretion of IL-12 by macrophages and DCs during in vitro infection , but only MIC1 plays a significant role during in vivo infection , demonstrated by its ability to promote a dysregulated induction of systemic levels of IFN-γ and a proinflammatory cytokine storm that leads to acute mortality during murine infection .
All experiments were conducted in accordance to the Brazilian Federal Law 11 , 794/2008 establishing procedures for the scientific use of animals , and State Law establishing the Animal Protection Code of the State of Sao Paulo . All efforts were made to minimize suffering , and the animal experiments were approved by the Ethics Committee on Animal Experimentation ( Comissão de Ética em Experimentação Animal—CETEA ) of the Ribeirao Preto Medical School , University of Sao Paulo ( protocol number 065/2012 ) , following the guidelines of the National Council for Control of Animal Experimentation ( Conselho Nacional de Controle de Experimentação Animal—CONCEA ) . The lactose-eluted ( Lac+ ) fraction was obtained as previously reported [17 , 21] . Briefly , the total soluble tachyzoite antigen ( STAg ) fraction was loaded into a lactose column ( Sigma-Aldrich , St . Louis , MO ) and equilibrated with PBS containing 0 . 5 M NaCl . The material adsorbed to the resin was eluted with 0 . 1 M lactose in equilibrating buffer and dialyzed against ultrapure water . The obtained fraction was denoted as Lac+ and confirmed to contain MIC1 and MIC4 . For the recombinant proteins , rMIC1 and rMIC4 sequences were amplified from cDNA of the T . gondii strain ME49 with a 6-histidine tag added on the N-terminal , cloned into pDEST17 vector ( Gateway Cloning , Thermo Fisher Scientific Inc . , Grand Island , NY ) , and used to transform DH5α E . coli chemically competent cells for ampicillin expression selection , as described before [21] . The plasmids with rMIC1-T126A/T220A and rMIC4-K469M were synthesized by GenScript ( New Jersey , US ) using a pET28a vector , and the MIC sequences carrying the mutations were cloned between the NdeI and BamH I sites . All plasmids extracted from DH5α E . coli were transformed in E . coli BL21-DE3 chemically competent cells to produce recombinant proteins that were then purified from inclusion bodies and refolded by gradient dialysis , as described previously for rMIC1 and rMIC4 wild type forms [21] . Endotoxin concentrations were measured in all protein samples using the Limulus Amebocyte Lysate Kit–QCL-1000 ( Lonza , Basel , Switzerland ) . The rMIC1 , rMIC1-T126A/T220A , rMIC4 and rMIC4-K469M contained 7 . 2 , 3 . 2 , 3 . 5 and 1 . 1 EU endotoxin/μg of protein , respectively . Endotoxin was removed by passing over two polymyxin-B columns ( Affi-Prep Polymyxin Resin; Bio-Rad , Hercules , CA ) . Additionally , prior to all in vitro cell-stimulation assays , the proteins samples were incubated with 50 μg/mL of polymyxin B sulphate salt ( Sigma-Aldrich , St . Louis , MO ) for 30 min at 37°C to remove possible residual LPS . The carbohydrate-binding profile of microneme proteins was determined by Core H ( Consortium for Functional Glycomics , Emory University , Atlanta , GA ) , using a printed glycan microarray , as described previously [56] . Briefly , rMIC1-Fc , rMIC4-Fc , and Lac+-Fc in binding buffer ( 1% BSA , 150 mM NaCl , 2 mM CaCl2 , 2 mM MgCl2 , 0 . 05% ( w/v ) Tween 20 , and 20 mM Tris-HCl , pH 7 . 4 ) were applied onto a covalently printed glycan array and incubated for 1 hour at 25°C , followed by incubation with Alexa Fluor 488-conjugate ( Invitrogen , Thermo Fisher Scientific Inc . , Grand Island , NY ) . Slides were scanned , and the average signal intensity was calculated . The common features of glycans with stronger binding are depicted in Fig 1A . The average signal intensity detected for all of the glycans was calculated and set as the baseline . Ninety-six-well microplates were coated with 1 μg/well of fetuin or asialofetuin , glycoproteins diluted in 50 μL of carbonate buffer ( pH 9 . 6 ) per well , followed by overnight incubation at 4°C . Recombinant MIC1 or MIC4 proteins ( both wild type ( WT ) and mutated forms ) , previously incubated or not with their corresponding sugars , i . e . α ( 2–3 ) -sialyllactose for MIC1 and lacto-N-biose for MIC4 ( V-lab , Dextra , LA , UK ) , were added into coated wells and incubated for 2 h at 25°C . After washing with PBS , T . gondii-infected mouse serum ( 1:50 ) was used as the source of the primary antibody . The assay was then developed with anti-mouse peroxidase-conjugated secondary antibody , and the absorbance was measured at 450 nm in a microplate-scanning spectrophotometer ( Power Wave-X; BioTek Instruments , Inc . , Winooski , VT ) . Female C57BL/6 ( WT ) , MyD88-/- , TRIF-/- , TLR2-/- , TLR3-/- , TLR4-/- , double knockout ( DKO ) TLR2-/-/TLR4-/- , TLR5-/- , and TLR9-/- mice ( all from the C57BL/6 background ) , 8 to 12 weeks of age , were acquired from the University of Sao Paulo—Ribeirao Preto campus animal facility , Ribeirao Preto , Sao Paulo , Brazil , and housed in the animal facility of the Department of Cell and Molecular Biology—Ribeirão Preto Medical School , under specific pathogen-free conditions . The TLR11-/-/TLR12-/- DKO mice were maintained at American Association of Laboratory Animal Care-accredited animal facilities at NIAID/NIH . For the in vivo infections , female CD-1 outbred mice , 6 weeks of age were acquired from Charles River Laboratories , Germantown , MD , USA . A clonal isolate of the T . gondii RH-Δku80/Δhpt strain was used to generate the GFP/Luciferase strain , which was the recipient strain to generate the single-knockout parasites . The GFP/Luc sequence was inserted into the UPRT locus of Toxoplasma by double crossover homologous recombination using CRISPR/Cas-based genome editing and selected for FUDR resistance to facilitate the targeted GFP/Luc gene cassette knock-in . The MIC1 and MIC4 genes were replaced by the drug-selectable marker hpt ( hxgprt—hypoxanthine-xanthine-guanine phosphoribosyl transferase ) flanked by LoxP sites . For all gene deletions , 30 μg of guide RNA was transfected along with 15 μg of a repair oligo . Parasites were transfected and selected as previously described [57 , 58] . For the MIC gene complementation , the sequence was amplified from RH genomic DNA with the addition of one copy of HA-tag sequence ( TACCCATACGATGTTCCAGATTACGCT ) before the stop codon , and cloned into pCR2 . 1-TOPO vector , followed by site-directed mutagenesis using the Q-5 kit ( New England Biolabs ) in order to generate point mutations into MIC1 ( MIC1-T126A/T220A ) and MIC4 ( MIC4-K469M ) sequences . For transfections , 30 μg of guide RNA was transfected along with 20 μg of linearized pTOPO vector containing the MIC mutated sequences . Strains were maintained in human foreskin fibroblast ( HFF ) cells grown in Dulbecco’s modified Eagle’s medium ( DMEM ) supplemented with 10% heat-inactivated foetal bovine serum ( FBS ) , 0 . 25 mM gentamicin , 10 U/mL penicillin , and 10 μg/mL streptomycin ( Gibco , Thermo Fisher Scientific Inc . , Grand Island , NY ) . Bone marrows of WT , MyD88-/- , TRIF-/- , TLR2-/- , TLR3-/- , TLR4-/- , DKO TLR2-/-/TLR4-/- , TLR5-/- , TLR9-/- , and DKO TLR11-/-/TLR12-/- mice were harvested from femurs and hind leg bones . Cells were washed with RPMI medium and resuspended in RPMI medium with 10% FBS , 10 U/mL penicillin , and 10 μg/mL streptomycin ( Gibco ) . For dendritic cell ( DC ) differentiation , we added 10 ng/mL of recombinant murine GM-CSF ( Prepotech , Rocky Hill , NJ ) , and 10 ng/mL murine recombinant IL-4 ( eBioscience , San Diego , CA ) ; for macrophage differentiation , 30% of L929 conditioned medium was added to RPMI medium with 10% FBS . The cells were cultured in 100 × 20 mm dish plates ( Costar; Corning Inc . , Corning , NY ) , supplemented with respective conditioned media at days 3 and 6 for DCs , and at day 4 for macrophages . DCs were incubated for 8–9 days and macrophages for 7 days; the cells were then harvested and plated into 24-well plates at 5 × 105 cells/well for protein stimulations or T . gondii infections , followed by ELISA . Cell purity was analyzed by flow cytometry . Eighty-five percent of differentiated dentritic cells were CD11b+/CD11c+ , while 94% of differentiated macrophages were CD11b+ . Human embryonic kidney 293T ( HEK293T ) cells , originally acquired from American Tissue Culture Collection ( ATCC , Rockville , MD ) , were used as an expression tool [59] for TLR2 and TLR4 [45 , 60] . The cells grown in DMEM supplemented with 10% FBS ( Gibco ) , and were seeded at 3 . 5 × 105 cells/mL in 96-well plates ( 3 . 5 × 104 cells/well ) 24 h before transfection . Then , HEK293T cells were transiently transfected ( 70–80% confluence ) with human TLR2 plasmids as described previously [25] or with CD14 , CD36 , MD-2 and TLR4 [61] using Lipofectamine 2000 ( Invitrogen ) with 60 ng of NF-κB Luc , an NF-κB reporter plasmid , and 0 . 5 ng of Renilla luciferase plasmid , together with 60 ng of each gene of single and multiple glycosylation mutants and of TLR2 WT genes [25] . After 24 h of transfection , the cells were stimulated overnight with positive controls: P3C ( Pam3CSK4; EMC Microcollections , Tübingen , Germany ) , fibroblast stimulating ligand-1 ( FSL-1; EMC Microcollections ) , or LPS Ultrapure ( standard LPS , E . coli 0111:B4; Sigma-Aldrich ) ; or with the negative control for cell stimulation ( the medium ) . Cells transfected with empty vectors , incubated either with the medium or with agonists ( FSL-1 or P3C ) , were also assayed; negative results were required for each system included in the study . IL-8 was detected in the culture supernatants . The absence of Mycoplasma contamination in the cell culture was certified by indirect fluorescence staining as described previously [62] . The quantification of human IL-8 and mouse IL-12p40 , IL-6 , TNF-α , and IL-10 in the supernatant of the cultures was performed by ELISA , following the manufacturer’s instructions ( OptEIA set; BD Biosciences , San Jose , CA ) . Human and murine recombinant cytokines were used to generate standard curves and determine cytokine concentrations . The absorbance was read at 450 nm using the Power Wave-X spectrophotometer ( BioTek Instruments ) . The pcDNA4/TO-FLAG plasmid was kindly provided by Dr . Dario Simões Zamboni . The pcDNA4-FLAG-TLR2 and pcDNA4-FLAG-TLR4 plasmids were constructed as follows . RNA from a P388D1 cell line ( ATCC , Rockville , MD ) was extracted and converted to cDNA with Maxima H Minus Reverse Transcriptase ( Thermo-Fisher Scientific , Waltham , MA USA ) and oligo ( dT ) . TLR2 and TLR4 were amplified from total cDNA from murine macrophages by using Phusion High-Fidelity DNA Polymerase and the phosphorylated primers TLR2_F: ATGCTACGAGCTCTTTGGCTCTTCTGG , TLR2_R: CTAGGACTTTATTGCAGTTCTCAGATTTACCCAAAAC , TLR4_F: TGCTTAGGATCCATGATGCCTCCCTGGCTCCTG and TLR4_R: TGCTTAGCGGCCGCTCAGGTCCAAGTTGCCGTTTCTTG . The fragments were isolated from 1% agarose/Tris-acetate-ethylenediaminetetraacetic acid gel , purified with GeneJET Gel Extraction Kit ( Thermo-Fisher Scientific ) , and inserted into the pcDNA4/TO-FLAG vector by using the restriction enzymes sites for NotI and XbaI ( Thermo-Fisher Scientific ) for TLR2 , and BamHI and NotI ( Thermo-Fisher Scientific ) for TLR4 . Ligation reactions were performed by using a 3:1 insert/vector ratio with T4 DNA Ligase ( Thermo-Fisher Scientific ) and transformed into chemically competent Escherichia coli DH5α cells . Proper transformants were isolated from LB agar medium plates under ampicillin selection ( 100 μg/mL ) and analyzed by PCR , restriction fragment analysis , and DNA sequencing . All reactions were performed according to the manufacturer’s instructions . We used the lysate of HEK293T cells transfected ( 70–80% confluence ) with plasmids containing TLR2-FLAG or TLR4-FLAG . After 24 h of transfection , the HEK cells were lysed with a non-denaturing lysis buffer ( 20 mM Tris , pH 8 . 0 , 137 mM NaCl , and 2 mM EDTA ) supplemented with a protease inhibitor ( Roche , Basel , Switzerland ) . After 10 min of incubation on ice , the lysate was subjected to centrifugation ( 16 , 000 g , at 4°C for 5 min ) . The protein content in the supernatant was quantified by the BCA method , aliquoted , and stored at -80°C . For the pull-down assay , 100 μg of the lysate from TLR2-FLAG- or TLR4-FLAG-transfected HEK cells were incubated with 10 μg of rMIC1 or rMIC4 overnight at 4°C . Since these proteins had a histidine tag , the samples were purified on nickel-affinity resin ( Ni Sepharose High Performance; GE Healthcare , Little Chalfont , UK ) after incubation for 30 min at 25°C and centrifugation of the fraction bound to nickel to pull down the TgMIC-His that physically interacted with TLR-FLAG ( 16 , 000 g , 4°C , 5 min ) . After washing with PBS , the samples were resuspended in 100 μL of SDS loading dye with 5 μL of 2-mercaptoethanol , heated for 5 min at 95°C , and 25 μL of total volume was run on 10% SDS-PAGE . After transferring to a nitrocellulose membrane ( Millipore , Billerica , MA ) , immunoblotting was performed by following the manufacturer’s protocol . First , the membrane was incubated with anti-FLAG monoclonal antibodies ( 1:2 , 000 ) ( Clone G10 , ab45766 , Sigma-Aldrich ) to detect the presence of TLR2 or TLR4 . The same membrane was then subjected to secondary probing and was developed with anti-TgMIC1 ( IgY; 1:20 , 000 ) or anti-TgMIC4 ( IgY; 1:8 , 000 ) polyclonal antibodies and followed by incubation with secondary polyclonal anti-chicken IgY-HRP ( 1:4 , 000 ) ( A9046 , Sigma-Aldrich ) to confirm the presence of rMIC1 and rMIC4 . Bone marrow-derived dendritic cells ( BMDCs ) and bone marrow-derived macrophages ( BMDMs ) were infected with WT ( Δku80/Δhpt ) , Δmic1 , Δmic1::MIC1-T126A/T220A , Δmic4 and Δmic4::K469M strains ( Type I , RH background ) recovered from T25 flasks with HFF cell cultures . The T25 flasks were washed with RPMI medium to completely remove parasites , and the collected material was centrifuged for 5 min at 50 g to remove HFF cell debris . The resulting pellet was discarded , and the supernatant containing the parasites was centrifuged for 10 min at 1 , 000 g and resuspended in RPMI medium for counting and concentration adjustments . BMDCs and BMDMs were dispensed in 24-well plates at 5 × 105 cells/well ( in RPMI medium supplemented with 10% FBS ) , followed by infection with 3 parasites per cell ( multiplicity of infection , MOI 3 ) . Then , the plate was centrifuged for 3 min at 200 g to synchronize the contact between cells and parasites and incubated at 37°C . The supernatants were collected at 24 hous after infection for quantification of IL-12p40 . Six-week-old female CD-1 outbred mice were infected by intraperitoneal injection with 50 tachyzoites of RH engineered strains diluted in 500 μl of phosphate-buffered saline . The mice were weighed daily and survival was evaluated . Bioluminescent detection of firefly luciferase activity was performed at day 5 post-infection using an IVIS BLI system from Xenogen to monitor parasite burden . Mice were injected with 3 milligrams ( 200 μl ) of D-luciferin ( PerkinElmer ) substrate , and after 5 minutes the mice were imaged for 300 seconds to detect the photons emitted . The data were plotted and analysed using GraphPad Prism 7 . 0 software ( GraphPad , La Jolla , CA ) . Statistical significance of the obtained results was calculated using analysis of variance ( One-way ANOVA ) followed by Bonferroni's multiple comparisons test . Differences were considered significant when the P value was <0 . 05 . | Toxoplasmosis is caused by the protozoan Toxoplasma gondii , belonging to the Apicomplexa phylum . This phylum comprises important parasites able to infect a broad diversity of animals , including humans . A particularity of T . gondii is its ability to invade virtually any nucleated cell of all warm-blooded animals through an active process , which depends on the secretion of adhesin proteins . These proteins are discharged by specialized organelles localized in the parasite apical region , and termed micronemes and rhoptries . We show in this study that two microneme proteins from T . gondii utilize their adhesion activity to stimulate innate immunity . These microneme proteins , denoted MIC1 and MIC4 , recognize specific sugars on receptors expressed on the surface of mammalian immune cells . This binding activates these innate immune cells to secrete cytokines , which promotes efficient host defense mechanisms against the parasite and regulate their pathogenesis . This activity promotes a chronic infection by controlling parasite replication during acute infection . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"blood",
"cells",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"cytokines",
"immune",
"cells",
"toxoplasma",
"gondii",
"immunology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"developmental",
"biology",
"protozoans",
"toxoplasma",
"molecular",
"development",
"immune",
"system",
"proteins",
"white",
"blood",
"cells",
"animal",
"cells",
"proteins",
"protozoan",
"infections",
"immune",
"system",
"toll-like",
"receptors",
"biochemistry",
"signal",
"transduction",
"eukaryota",
"lectins",
"cell",
"biology",
"physiology",
"biology",
"and",
"life",
"sciences",
"immune",
"receptors",
"cellular",
"types",
"macrophages",
"organisms"
] | 2019 | The lectin-specific activity of Toxoplasma gondii microneme proteins 1 and 4 binds Toll-like receptor 2 and 4 N-glycans to regulate innate immune priming |
The transcription factor CONSTANS ( CO ) is a central component that promotes Arabidopsis flowering under long-day conditions ( LDs ) . Here , we show that the microRNA319-regulated TEOSINTE BRANCHED/CYCLOIDEA/PCF ( TCP ) transcription factors promote photoperiodic flowering through binding to the CO promoter and activating its transcription . Meanwhile , these TCPs directly interact with the flowering activators FLOWERING BHLH ( FBHs ) , but not the flowering repressors CYCLING DOF FACTORs ( CDFs ) , to additively activate CO expression . Furthermore , both the TCPs and FBHs physically interact with the flowering time regulator PHYTOCHROME AND FLOWERING TIME 1 ( PFT1 ) to facilitate CO transcription . Our findings provide evidence that a set of transcriptional activators act directly and additively at the CO promoter to promote CO transcription , and establish a molecular mechanism underlying the regulation of photoperiodic flowering time in Arabidopsis .
Flowering is a transition from the vegetative to the reproductive phase in the plant life cycle , which is crucial for successful reproduction . Genetic approaches in the model plant Arabidopsis , in which flowering is often promoted under long-day ( LD ) but is delayed during short-day ( SD ) conditions , reveal that CONSTANS ( CO ) plays crucial roles in photoperiod monitoring and flowering time determination [1–3] . In Arabidopsis , CO encodes a B-box-type zinc finger transcriptional activator [4] . The co mutant lines flower late under LDs , whereas the plants overexpressing CO display early flowering phenotype in both LDs and SDs [4 , 5] . Under LDs , CO displays a biphasic diurnal expression pattern that its transcript levels first rise at the late afternoon to form a small peak in the light period , and a second peak appears during the midnight [5] . Several studies have revealed that the CO protein stabilization is tightly controlled in a light-dependent manner by a number of factors , such as phytochrome A ( PHYA ) , cryptochrome 2 ( CRY2 ) and FKF1 ( FLAVIN-BINDING , KELCHREPEAT , F-BOX1 ) and CONSTITUTIVE PHOTOMORPHOGENIC 1 ( COP1 ) [6–10] . Therefore , the induction of CO mRNA levels at dusk under LDs but not the peak expression at night is essential for the CO protein accumulation and subsequent photoperiodic flowering promotion . To date , several components have been identified to precisely regulate the diurnal transcription of CO in Arabidopsis . The transcription factors CYCLING DOF FACTORs ( CDF1-5 ) are the well characterized repressors of CO transcription [11 , 12] . However , as the repressors , CDFs could not fully explain the remarkable up-regulation of CO transcript levels at dusk . The four basic helix-loop-helix-type ( bHLH ) transcription factors FLOWERING BHLH 1 ( FBH1 ) , FBH2 , FBH3 , and FBH4 have been identified as the CO transcriptional activators that preferentially bind to the E-box cis-elements of the CO promoter in the afternoon to induce the expression of CO [13] , proposing a complicated temporal interplay among repressors and activators in restricting the CO transcription . However , unlike CDFs , FBHs do not show robust daily oscillation at either mRNA or protein levels , implying that their time-dependent binding preference on CO promoter is potentially affected by some other unidentified regulators or co-activators [13] . In addition to the transcription factors , PHYTOCHROME AND FLOWERING TIME 1 ( PFT1 ) , encoding the Mediator complex subunit 25 ( MED25 ) in Arabidopsis , was reported to genetically act upstream of CO and promote flowering [14 , 15] . However , the molecular mechanisms about how PFT1 relies on the information from light signals to control flowering time through affecting CO transcript levels remain obscure . The plant-specific TEOSINTE BRANCHED1/CYCLOIDEA/PCF ( TCP ) family transcription factors contain a conserved non-canonical bHLH domain , which mediates DNA binding or interactions with other proteins [16] . In Arabidopsis , the jaw-D mutants , in which microRNA319 ( miR319 ) is over accumulated and five class II TCP genes including TCP2 , TCP3 , TCP4 , TCP10 , and TCP24 are down-regulated , show delayed flowering phenotype [17–19] . However , the functional mode and action mechanism of these TCPs transcription factors in regulation of Arabidopsis flowering time remain unclear . In this study , we demonstrate that the miR319-regulated TCPs function as direct transcriptional activators of the photoperiodic flowering regulator CO to promote Arabidopsis flowering under the inductive photoperiod . Furthermore , these TCPs transcription factors physically interact with the flowering activators FBHs . Meanwhile , we found that these TCPs and FBHs transcription factors directly interact with the flowering time regulator PFT1 to facilitate CO transcription , and this conclusion is further supported by the observation that PFT1 proteins are exclusively enriched in the TCP- and FBH-binding regions of CO promoter under LDs . Thus , we uncover a transcriptional activation complex for direct activation of CO transcription to promote Arabidopsis photoperiodic flowering .
Previous studies have shown that the Arabidopsis tcp4 and jaw-D mutants displayed delayed flowering under LDs [17–19] . To further evaluate the potential role of the miR319-regulated TCPs transcription factors in the photoperiodic flowering pathway , we examined the flowering time phenotype of jaw-D mutant lines under both LD ( 16 h light/8 h dark ) and SD ( 8 h light/16 h dark ) conditions . As expected , the jaw-D plants displayed an obvious late-flowering phenotype compared with wild type ( WT ) Columbia-0 ( Col-0 ) under inductive LDs ( Fig 1A and 1B ) , but flowered normally under the non-inductive SDs ( S1 Fig ) . Next , we analyzed the expression patterns of miR319 together with the five miR319-regulated TCPs , including TCP2 , TCP3 , TCP4 , TCP10 and TCP24 , in both wild type and jaw-D mutant plants . In consistent with previous reports [17 , 19] , we also showed that the five miR319-regulated TCPs were all significantly down-regulated in jaw-D plants compared with WT , coupling with the elevated miR319 levels ( Fig 1C ) . Intriguingly , the mRNA levels of TCP2 , TCP4 and TCP24 exhibited similar diurnal expression patterns ( Fig 1C ) . Accordingly , we assumed that miR319-regulated TCPs may be involved in the Arabidopsis photoperiodic flowering pathway . To further confirm this hypothesis , we generated transgenic plants carrying a 35S:TCP4 construct [17] ( S2 Fig ) , and observed a significantly early flowering phenotype in the 35S:TCP4 transgenic plants under both LDs and SDs ( Fig 1A and 1B; S1 Fig ) . In summary , we concluded that the miR319-regulated TCPs are positive regulators of Arabidopsis photoperiodic flowering . The delayed flowering phenotype of jaw-D plants under LDs led us to examine whether the CO expression is altered in the jaw-D mutant line . Expectedly , the transcript levels of CO were notably reduced in the jaw-D line during the time periods of CO mRNA peaks [Zeitgeber time ( ZT ) 12–16 and ZT 20–24] , as compared with Col-0 seedlings ( Fig 2A , left panel ) . On the contrary , the CO mRNA levels in the 35S:TCP4 line were obviously increased compared with those in WT ( Fig 2A , right panel ) . Notably , we observed a significant up-regulation of CO in the 35S:TCP4 seedlings at dusk during the light phase ( ZT 12–16 ) ( Fig 2A , right panel ) . As is well known , CO directly activates the expression of its downstream targeting flowering-time gene FLOWERING LOCUS T ( FT ) . Thus , we assumed that the FT transcription might be also affected in the jaw-D and 35S:TCP4 plants . As expected , FT expression was partially compromised in jaw-D mutant , but significantly up-regulated in the 35S:TCP4 transgenic line at different time points ( S3 Fig ) , consistent with the altered CO levels in these lines . Together , these results imply that the miR319-regulated TCPs might play positive regulatory roles in activating CO transcription . To investigate the functional relationship between TCPs and CO in vivo , we tested for genetic interactions between these genes . We crossed the CO-overexpressing transgenic line 35S:CO into the jaw-D background to generate the jaw-D/35S:CO plant and examined its flowering time phenotype in LDs . As expected , the 35S:CO and jaw-D plants displayed early-flowering and late-flowering phenotypes , respectively ( Fig 2B and 2C ) . However , all of the jaw-D/35S:CO seedlings flowered early resembling the flowering time phenotype of 35S:CO plants ( Fig 2B and 2C ) , indicating that overexpression of CO can rescue the late-flowering phenotype of jaw-D . Meanwhile , we introduced 35S:TCP4 into the co-9 mutant background [20] to generate the co-9/35S:TCP4 line . The flowering time analysis in LDs revealed that , similar to co-9 , all the co-9/35S:TCP4 seedlings exhibited late-flowering phenotype compared to Col-0 ( Fig 2D and 2E ) , suggesting that the early flowering phenotype caused by 35S:TCP4 is largely dependent on the function of CO . Together , these data suggest that the miR319-regulated TCPs may function genetically upstream of CO to promote photoperiodic flowering . As plant-specific transcription factors , the miR319-regulated TCPs predominantly bind to the common TCP-binding motifs [TBM , GGACC ( A/C ) ] to regulate the expression of target genes [19] . We screened the CO promoter sequence ( 2-kb ) , and identified five putative TBM sequences . Two of the TBMs are adjacent to the CO transcriptional start site ( named as TBM 1 and TBM 2 ) with positions of -263/-257 and -324/-318 , while the other three located at -1341/-1335 ( named as TBM 3 ) , -1371/-1365 ( named as TBM 4 ) and -1484/-1478 ( named as TBM 5 ) , respectively ( Fig 3A , upper panel ) . To investigate the association of TCP4 with the CO promoter in vivo , we performed chromatin immunoprecipitation ( ChIP ) assay . Considering that TCP4 was predominantly expressed in the vascular tissues of leaves ( S4 Fig ) , we used the leaf-expressed and viable BLSpro:rTCP4-GFP transgenic plants ( labeled as rTCP4-GFP in this study; rTCP4 represents miR319-cleavage-resistant TCP4; S5 Fig ) [21 , 22] for the ChIP assay . Here , we designed eight specific amplicons ( represented by P1-P8 in Fig 3A , upper panel ) which covered the 2-kb region of CO promoter . As a result , the TCP4-GFP enrichments were specifically observed at the P8 and P2/P3/P4 regions of CO promoter in the rTCP4-GFP ChIP samples ( rTCP4-GFP + αGFP in Fig 3A ) , compared with the negative controls ( WT + αGFP and rTCP4-GFP - αGFP in Fig 3A ) , with the highest level at the P8 region ( Fig 3A ) . Consistently , P8 and P3 span the CO promoter regions where the TCP-binding motifs are located ( TBM 1/2 in P8 , and TBM 3/4/5 in P3 in Fig 3A ) , indicating that TCP4 is specifically associated with the TBM cis-elements on the CO promoter region in vivo . This binding specificity was further supported by the yeast one-hybrid ( Y1H ) and electrophoretic mobility shift ( EMSA ) assays , in which TCP4 exclusively associated with the P8 and P3 fragments , but not the TBM cis-elements mutated mP8 and mP3 [the TBM cis-elements GGACC ( C/A ) were replaced by AAAAAA] ( Fig 3B and 3C; S1 and S2 Tables ) . Together , these data strongly demonstrate that TCP4 directly binds to the TBM cis-elements in the CO promoter region both in vitro and in vivo . Next , to evaluate the direct regulation of TCP4 as well as other miR319-regulated TCPs on CO expression , we performed transient transcriptional activity assays in Nicotiana benthamiana leaves using the CO promoter ( 2-kb ) fused with the LUC gene that encodes firefly luciferase as a reporter [23] . Results showed that the LUC signals were significantly elevated by co-expression of the rTCPs , including rTCP2 , rTCP3 , rTCP4 , rTCP10 and rTCP24 ( Fig 3D and 3E; S5 Fig ) , supporting the hypothesis that the miR319-regulated TCPs are direct transcriptional activators of CO transcription . This conclusion was further confirmed in Arabidopsis by generating the β-estradiol-inducible pERGW-rTCP4 transgenic plants ( Fig 3F ) . In the presence of the chemical inducer β-estradiol , the pERGW-rTCP4 seedlings showed efficient induced expression of TCP4 ( Fig 3F , left panel ) . Most importantly , the CO transcript levels were also significantly up-regulated following the β-estradiol treatment ( Fig 3F , right panel ) , confirming the direct activation of CO expression by TCP4 . Because the FBHs and CDFs transcription factors have been shown to separately act as activators and repressors of CO transcription [11–13 , 24] , we asked whether the miR319-regulated TCPs physically interact with these transcription factors . Indeed , yeast two-hybrid ( Y2H ) assays revealed an obvious interaction between TCP4 and FBH1 ( Fig 4A ) , and this interaction was further confirmed by LUC complementation imaging ( LCI ) assay in N . benthamiana ( Fig 4B , upper panel ) . However , no interaction was detected between TCP4 and CDF1 ( Fig 4A; Fig 4B , lower panel ) , suggesting that the interaction between TCP4 and FBH1 may be specifically occurred with biological significance . To further confirm the interaction between TCP4 and FBH1 in vivo , we crossed FBH1-GFP with TCP4-Myc to generate the FBH1-GFP/TCP4-Myc double transgenic Arabidopsis plant , and conducted co-immunoprecipitation ( Co-IP ) assay . Confidently , the interaction signal was exclusively observed in FBH1-GFP/TCP4-Myc plant , but not in FBH1-GFP or TCP4-Myc control samples ( Fig 4C ) , further supporting the physical interaction between TCP4 and FBH1 . Moreover , we conducted Förster resonance energy transfer ( FRET ) assays using Arabidopsis protoplast cells . Here , we employed a quantitative non-invasive fluorescence lifetime imaging ( FLIM ) approach to detect FRET efficiency [25] . In this assay , two tested proteins are separately fused with cyan fluorescent protein ( CFP ) and yellow fluorescent protein ( YFP ) to generate the donors and acceptors , and the lifetime ( described as τ ) of the donor fluorescence ( CFP ) is measured in the presence or absence of the acceptor proteins ( represented by τDA and τD , respectively ) . If there is a physical interaction between the donor and acceptor , the lifetime τDA will be considerably shorter than τD . First , we fused TCP4 with CFP to generate the donor , and FBH1 and CDF1 with YFP as the acceptors ( Fig 4D–4F ) . The combination of TCP4-CFP/YFP was also designed to represent the negative control in the absence of acceptor ( τD , Fig 4D–4F ) . Confocal microscope detection suggested that all the indicated proteins were properly accumulated in the protoplast cells , and the fluorescent signals of TCP4-CFP , FBH1-YFP and CDF1-YFP fusion proteins were all exclusively observed and perfectly merged in the nuclei ( Fig 4D ) , indicating similar subcellular localization of TCP4 , FBH1 and CDF1 . Subsequently , we measured the CFP lifetime in the combination samples . As expected , the average lifetime of CFP in the TCP4-CFP/FBH1-YFP co-expression cells was 0 . 99 ± 0 . 19 ns ( τDA; mean ± SD , n = 17 nuclei ) , which was remarkably ( P < 0 . 01; Student’s t test ) shorter than the 2 . 59 ± 0 . 39 ns ( τD; mean ± SD , n = 7 nuclei ) determined in the negative control TCP4-CFP/YFP ( Fig 4E and 4F ) , suggesting a strong physical interaction between TCP4 and FBH1 . However , in the TCP4-CFP/CDF1-YFP co-expression samples , the average lifetime of CFP was 2 . 48 ± 0 . 16 ns ( τDA; mean ± SD , n = 15 nuclei ) , very similar to that of the negative control , illustrating that TCP4 does not interact with CDF1 ( Fig 4E and 4F ) . Further assays by Y2H in yeast and LCI in N . benthamiana revealed that TCP4 also interacts with FBH2 , FBH3 and FBH4 ( S6A and S6B Fig ) ; meanwhile , FBH1 could also interact with other miR319-regulated TCPs such as TCP2 , TCP3 , TCP10 and TCP24 ( S6C and S6D Fig ) . These results strongly suggest the functional conservation and redundancy among the miR319-regulated TCPs as well as the FBH homologs . Our simultaneous analyses using the CDF homologs CDF1 , CDF2 and CDF3 revealed that these CDFs failed to interact with all the tested TCPs ( TCP2 , TCP3 , TCP4 , TCP10 and TCP24 ) and FBHs ( FBH1 , FBH2 , FBH3 and FBH4 ) ( S7 Fig ) , indicating that the CO transcription repressors CDFs may function independently of the TCPs and FBHs activators . To define the interaction domains between the TCPs and FBHs , we used TCP4 and FBH1 as the representatives in our analyses . First , we generated the different truncated forms of TCP4 and FBH1 ( NT , amino terminal; MD , middle domain; CT , carboxyl terminal; S8A Fig ) . The transcriptional activation activity assay in yeast cells demonstrated that the MD in TCP4 as well as the NT and MD in FBH1 are the functional transcriptional activation domains ( S8B Fig ) . Interestingly , our LCI assays using different truncated forms of TCP4 and FBH1 revealed that the MD of TCP4 and the NT/MD of FBH1 are exactly the domains for their interaction ( S8C and S8D Fig ) . Based on the above analyses , we confidently uncovered the coupling of transcriptional activation domains and their interaction parts between TCP4 and FBH1 . The above conclusion led us to further evaluate the biological significance of the physical interaction between TCP4 and FBH1 on their transcriptional activation activities . To this end , we carried out transient transcriptional activity assays in N . benthamiana . As the activators of CO transcription , both TCP4 and FBH1 significantly led to obvious induction in the luminescence intensity of COpro:LUC by about 6- to 10-fold changes compared with the empty vector control sample ( Fig 4G and 4H; combinations 2 and 3 ) . Most importantly , we observed a more significant up-regulation of LUC reporter activity in the TCP4 and FBH1 co-expressed sample ( Fig 4G and 4H; combination 4 ) , in which the luminescence intensity was almost 20 times higher than that in the negative control ( Fig 4G and 4H; combination 1 ) , confidently suggesting an additive effect of TCP4 and FBH1 on CO transcription activation . Meanwhile , our qRT-PCR assays revealed that TCP4 and FBH1 were similarly expressed in different infiltrated samples ( Fig 4I ) . Together , these data imply that TCP4 and FBH1 might function additively to regulate the expression of CO in Arabidopsis . PFT1 , encoding the Arabidopsis Mediator complex subunit 25 ( MED25 ) which usually acts as a transcriptional co-activator , was previously described as an essential regulator of flowering time [14 , 15] . These observations promoted us to test whether there may be a functional relationship between PFT1 and the CO activators TCPs and/or FBHs . Indeed , we found that both TCP4 and FBH1 could interact with PFT1 , according to the Y2H assay in yeast as well as the LCI and Co-IP assays in N . benthamiana ( Fig 5A–5C ) . Next , we conducted FLIM-FRET to further confirm these physical interactions in Arabidopsis protoplast cells . Similarly , confocal microscope detection confirmed that the TCP4-CFP , FBH1-YFP and PFT1-YFP/CFP fusion proteins were all properly accumulated , and the fluorescent signals of these fusion proteins were exclusively merged in the nuclei ( Fig 5D ) . Subsequently , an average lifetime of 0 . 90 ± 0 . 08 ns ( τDA; mean ± SD , n = 18 nuclei ) was determined in the TCP4-CFP/PFT1-YFP co-expression samples ( Fig 5E and 5F ) , which was significantly ( P < 0 . 01; Student’s t test ) shorter than that of 2 . 59 ± 0 . 39 ns ( τD ) in the negative control ( TCP4-CFP/YFP as shown in Fig 4E and 4F ) ; meanwhile , the average lifetime of CFP in the PFT1-CFP/FBH1-YFP samples was only 0 . 95 ± 0 . 28 ns ( τDA; mean ± SD , n = 9 nuclei ) , remarkably ( P < 0 . 01; Student’s t test ) shorter than the 2 . 80 ± 0 . 28 ns ( τD; mean ± SD , n = 6 nuclei ) in the negative control PFT1-CFP/YFP ( Fig 5E and 5F ) . These data strongly demonstrate that PFT1 directly interacts with TCP4 and FBH1 in Arabidopsis . Certainly , our extended assays further confirmed that PFT1 also interacts with other miR319-regulated TCPs ( including TCP2 , TCP3 , TCP10 and TCP24 ) and FBH1 homologs ( FBH2 , FBH3 and FBH4 ) ( S9 Fig ) . The above finding that PFT1 physically interacts with TCP4 promoted us to ask whether PFT1 is functionally involved in the TCP4-regulated photoperiodic flowering pathway . To address this question , we crossed the 35S:TCP4 transgenic plants with the pft1-2 mutant line to generate the 35S:TCP4/pft1-2 plant , and examined its flowering time phenotype in LDs . Our results showed that pft1-2 exhibited a late-flowering phenotype compared to the WT Col-0 ( Fig 5G and 5H ) , which is consistent with the previous studies [14 , 15 , 26] . More importantly , even though 35S:TCP4 could trigger early flowering ( Fig 1A and 1B; Fig 5G and 5H ) , its promotional effect on flowering was largely compromised in the pft1-2 background ( Fig 5G and 5H ) . These results imply that PFT1 is genetically required for the role of TCP4 in promoting flowering . We further hypothesized that PFT1 might be required for the full functions of miR319-regulated TCPs in the activation of CO transcription . To test this idea , we performed the transient activation activity assays in N . benthamiana ( Fig 6A ) . Consistent with the above results , the expression of TCP4 alone led to about 10-fold up-regulation of the COpro:LUC reporter activity ( Fig 3D and 3E , combination 4; Fig 6A , combination 3 ) ; whereas , PFT1 failed to elevate the COpro:LUC reporter activity ( Fig 6A , combination 2 ) , indicating that PFT1 alone is not able to activate CO transcription . Interestingly , when we co-expressed PFT1 and TCP4 , an obvious additive effect was observed as the luminescence intensities increased by almost 30 folds ( Fig 6A , combination 4 ) , compared with the empty vector control ( Fig 6A , combination 1 ) , significantly higher than that in the TCP4 single-expression samples ( Fig 6A , combination 3 ) , suggesting that PFT1 potentially facilitates the transcriptional activation activity of TCP4 on CO transcription . Parallel experiments showed that PFT1 could also dramatically enhance the transcriptional activation activity of FBH1 in promoting CO transcription ( Fig 6A , combinations 5–8 ) . Our qRT-PCR assays revealed that TCP4 , FBH1 and PFT1 were all similarly expressed in different infiltrated samples ( Fig 6B ) . Thus , we concluded that PFT1 may function as co-activator of both TCPs and FBHs in the activation of CO transcription . To well understand the action mechanism by which PFT1 regulates flowering time , we first analyzed the time-course expression pattern of PFT1 under LDs . Interestingly , similar to CO [2 , 27] , PFT1 displayed a diurnal rhythmic expression pattern , with a small elevation of PFT1 mRNA levels in the afternoon ( ZT 8–12 ) and an obvious peak at the midnight ( ZT 20–24 ) ( Fig 6C ) . In consistence with the previous report [14] , our data confirmed that the CO expression in the pft1-2 mutant line was significantly compromised compared with that in the WT Col-0 control ( Fig 6D ) , indicating an essential role of PFT1 in facilitating CO transcription . Based on our findings that PFT1 physically interacts with the CO activators TCPs and FBHs ( Fig 4 ) , we were interested to test whether PFT1 proteins are enriched in the CO promoter regions in vivo . To this end , we conducted the ChIP assays using the 35S:PFT1-GFP transgenic Arabidopsis plants [15] . As expected , the PFT1-GFP enrichments were remarkably detected at the P8 and P3 regions of CO promoter with a maximum enrichment at P8 ( PFT1-GFP + αGFP in Fig 6E ) , compared with the negative controls ( WT + αGFP and PFT1-GFP - αGFP in Fig 6E ) . Significantly , the enrichment tendency of PFT1 at the different CO promoter regions well correlated with that of TCP4 ( Fig 3A ) , and was also very similar to that of FBH1 as shown in a previous study [13] . Taken together , we propose that PFT1 is probably enriched in the CO promoter regions to act synergistically with TCPs and FBHs to facilitate CO transcription .
In this study , we showed that the miR319-regulated TCPs interact with the flowering time regulators FBHs and PFT1 to activate CO transcription and promote Arabidopsis photoperiodic flowering . Previous observations suggested that the miR319-regulated TCPs transcription factors may be involved in regulation of Arabidopsis flowering time [17–19] . However , the functional mode and action mechanism of these TCPs transcription factors in regulation of Arabidopsis flowering time remain unclear . In this study , we show that down-regulation of the miR319-regulated TCPs in the jaw-D mutant plants causes late flowering phenotype in LDs , but not in SDs ( Fig 1A and 1B; S1 Fig ) , demonstrating that the miR319-regulated TCPs modulate flowering time through regulating the photoperiodic flowering pathway in Arabidopsis . In support of this view , the expression of CO , a central component of the photoperiodic flowering pathway in Arabidopsis , were significantly reduced in the jaw-D mutant plants in amplitude under LDs ( Fig 2A ) , while up-regulated by both constitutive and inducible overexpression of TCP4 ( Figs 2A and 3F ) . Further , we showed that TCP4 can bind to the TBM cis-elements of the CO promoter and all the miR319-regulated TCPs directly activate CO transcription ( Fig 3A–3E ) . Based on these findings , we conclude that the miR319-regulated TCPs may act as positive regulators of photoperiodic flowering through direct activation of CO transcription in Arabidopsis . Nevertheless , the in planta interplay between the miR319-regulated TCP transcription factors and CO promoter still needs to be intensively analyzed in the future , considering that the non-native promoter used for driving TCP4 expression in this study might cause a non-physiological effect for TCP4 . Therefore , it should be intriguing to uncover the dynamic enrichment pattern of each member of these TCP proteins on the CO promoter , which will be useful for better understanding the contribution of these TCPs to the daily CO oscillation . FBHs act as CO transcription activators in regulating flowering time [13] . Our findings that TCPs physically interact with FBHs provide a novel mechanism for the regulation of CO transcription in the photoperiodic flowering pathway ( Fig 4; S6 and S8 Figs ) . The previous study showed that the E-box cis-elements contained in the -509/-196 region of CO promoter are essential for FBH1 binding as well as FBH1-dependent gene activation [13] . Coincidently , our ChIP assays revealed a preferred binding fragment of CO promoter by TCP4 containing the TBM cis-elements in the -348/-155 region ( P8 in Fig 3A ) , which is adjacent to and partially overlapped with the FBH1 binding region . The spatial proximity of the DNA-binding sites to some extent causes the possibility of direct interaction between TCP4 and FBH1 . However , our further assays revealed that TCP4 and FBH1 interact with each other through their transcriptional activation domains ( S8 Fig ) , not through their DNA-binding domains ( i . e . the bHLH domains that located in the N- or C-terminals of TCP4 and FBH1 , respectively , as shown in S8A Fig ) , suggesting the physical interaction between TCP4 and FBH1 might facilitate their transcriptional activation activities on CO transcription . Indeed , an additive effect of TCP4 and FBH1 in activating CO transcription was obviously observed in our analyses ( Fig 4G and 4H ) , implying a potential interplay among the TCPs and FBHs transcription factors . However , it should be noticed that TCP4 and FBH1 themselves could , at least in part , activate the transcription of CO ( Fig 4G and 4H ) . Thus , the additive effect of TCP4 and FBH1 might be attributed to more abundant activators enriched on the CO promoter and/or their cooperation upon the co-expression of these two transcription factors . Here , we assume that the miR319-regulated TCPs and FBHs might function cooperatively and/or independently to activate the CO expression in certain situations . However , it is eagerly needed to explore the genetic interaction between the TCPs and FBHs regarding the regulation of CO expression in vivo in the future . In this study , we confirmed that both TCPs and FBHs physically interact with the transcriptional co-activator PFT1 ( Fig 5 and S9 Fig ) . Although PFT1 , encoding the Mediator subunit 25 in Arabidopsis , was initially identified as a positive regulator of flowering time more than ten years ago [14] , the molecular mechanisms of its action in regulation of flowering time remain obscure to date . Mediator is a multiprotein complex that promotes transcription by recruiting the RNA polymerase II ( RNAPII ) to the promoter regions upon the physical interaction with specific DNA-bound transcription factors [28 , 29] . Our observations reinforce that co-expression of PFT1/MED25 with TCP4 or FBH1 additively elevated the CO transcription levels ( Fig 6A ) , while the loss-of-function of PFT1 leads to an obvious reduction of CO mRNA levels ( Fig 6D ) . It is noteworthy , in our assays , that PFT1 failed to promote CO transcription in the absence of TCP4 or FBH1 ( Fig 6A ) , implying the essential roles of TCPs and FBHs for the function of PFT1 in activating CO transcription . This hypothesis was further supported by our ChIP assay results that the PFT1 proteins were enriched with the peaks in the CO promoter regions near the TCP4- and FBH1-binding sites ( Figs 3A and 6E ) . Collectively , our results suggest that PFT1 potentially acts as positive regulator of CO transcription . Based on our findings , we proposed a working model on the control of photoperiodic flowering time ( Fig 6F ) . Briefly , the miR319-regulated TCPs and FBHs directly bind to the adjacent regions of CO promoter in the wild-type Arabidopsis plants; they physically interact with each other through their transcriptional activation domains to activate CO transcription through direct interaction with PFT1 , and consequently promote flowering under LDs ( Fig 6F , upper panel ) . By contrast , in the jaw-D mutant plants , the association of TCPs with CO promoter is drastically blocked due to the overdose of miR319 and consequent decrease of TCP proteins , leading to down-regulation of CO transcription during the peak expression time , which as a result causes delayed flowering ( Fig 6F , lower panel ) .
The transgenic and mutant lines used in this study were previously described: jaw-1D [17]; co-9 [20]; BLSpro:rTCP4-GFP [21]; COpro:GUS [30]; 35S:CO [7]; pft1-2 [26]; 35S:PFT1-GFP [15] . Arabidopsis thaliana were grown under LD ( 16-h-light/8-h-dark ) or SD ( 8-h-light/16-h-dark ) conditions at 22°C . Time-course analyses were performed on 12-d-old seedlings grown on half-strength Murashige and Skoog medium . Nicotiana benthamiana was grown in a greenhouse at 22°C with a 16-h-light/8-h-dark cycle . Analyses of flowering time were performed as previously described [15] . Flowering time was recorded from at least 15 plants per genotype that were grown in soil under either LDs or SDs , and was scored as the number of days from germination to the first appearance of buds at the apex ( days to bolting ) . The rosette leaf number was counted after the main stem has bolted 1 cm . For Gateway cloning , all the gene sequences were cloned into the pQBV3 or pENTRY vectors ( Gateway ) and subsequently introduced into certain destination vectors following the Gateway technology ( Invitrogen ) . For ligase dependent cloning , the endonuclease digested vectors and PCR fragments were separately purified by PCR cleanup kit ( Axygen , AP-PCR-250 ) , and ligated at 16°C with T4 DNA ligase ( New England Biolabs , M2020 ) . For ligase-independent ligation , the ligation free cloning mastermix ( abm ) was used following the application handbook . For generation of miR319-cleavage-resistant forms of TCPs ( rTCPs ) as well as the TBM cis-elements mutated CO promoter fragments , the one-step site-directed mutagenesis strategy was performed . The wild type TCP ( TCP2 , TCP3 , TCP4 , TCP10 and TCP24 ) sequences or the CO promoter fragments were first connected into the pQBV3 entry vector , and then the mutations on miR319-target sites or TBM cis-elements were introduced by the specifically designed primers , as described previously [31] . The PCR amplifications were carried out by pre-heating at 94°C for 3 min , 16 cycles of 94°C for 1 min , 55°C for 1 min and 68°C for 7 min , followed by incubation at 68°C for 1 h . The PCR products were purified by PCR cleanup kit ( Axygen , AP-PCR-250 ) , and digested by 1 μl of DpnI ( New England Biolabs , R0176L ) . The obtained products were transformed into Escherichia coli competent cells for sequencing . The primer sequences used for site-directed mutagenesis are listed in S5 Table . For the generation of 35S:TCP4 lines , the expression vector 35S:TCP4 [17] was transformed into the Agrobacterium strain GV3101 ( pMP90 ) . For the construction of TCP4pro:GUS transgenic lines , the 5’ upstream region of the TCP4 sequence ( -2000/-1 ) was amplified from Col-0 genomic DNA , and cloned into the binary vector pMDC162 vector to generate TCP4pro:GUS expression construct . For the generation of β-estradiol-inducible pERGW-rTCP4 transgenic plants , the construct pQBV3-rTCP4 containing the miR319-cleavage-resistant TCP4 was introduced into the destination vector pERGW to fuse with a β-estradiol-inducible promoter following the Gateway cloning strategy . All of the binary vectors were introduced into the wild type Col-0 plants by Agrobacterium-mediated transformation to generate transgenic plants [20] . More details of the DNA constructs are listed in S6 Table . jaw-D/35S:CO , co-9/35S:TCP4 , FBH1-GFP/TCP4-Myc and 35S:TCP4/pft1-2 were generated by genetic crossing . The 12-d-old Arabidopsis seedlings grown on half-strength Murashige and Skoog medium were collected at the indicated time after the onset of light . Total RNA was extracted using Trizol ( Invitrogen ) reagent . About 2 μg of total RNA and Moloney murine leukemia virus reverse transcriptase ( M-MLV; Promega ) were further used for reverse transcription . The cDNA was diluted to 100 μL with water in a 1:5 ratio , and 2 μL of the diluted cDNA was used for quantitative reverse transcription-polymerase chain reaction ( qRT-PCR ) with the SYBR Premix Ex Taq ( Perfect Real Time; TaKaRa ) . The reverse transcription of mature miRNAs was performed as described previously [32] , by using specifically designed stem-loop primers . qRT-PCR was performed using the following program: 120 sec at 95°C , 45 cycles of 10 sec at 95°C , and 1 min at 60°C . ACTIN7 and U6 expression levels were used as the internal controls for coding genes and miRNAs , respectively . All the experiments were performed independently three times . All the primers used for real time quantification are listed in S3 Table . For yeast one-hybrid assay , the pHIS2 derivatives were co-transformed with GAL4-AD constructs harboring the P8 and P3 sequences , or mP8 and mP3 fragments which contain mutated TBM cis-elements [GGACC ( C/A ) were replaced by AAAAAA] ( see S1 Table ) into the yeast ( Saccharomyces cerevisiae ) strain AH109 . The transformed cells were first grown on synthetic dextrose medium lacking Leu and Trp ( SD-L/W ) and then were transferred to the synthetic dextrose medium lacking Leu , Trp and His ( SD-L/W/H ) medium supplemented with 3-amino-1 , 2 , 4-triazole ( 3-AT ) for selection . For yeast two-hybrid assay , the GAL4-AD and GAL4-BD derivatives were co-transformed into the yeast strain AH109 , and grown on SD-L/W . Further , the yeast cells were screened on the SD-L/W/H or SD-L/W/H/A media with 3-AT . For transcriptional activation activity analysis , the GAL4-BD derivatives were separately transformed into yeast strain AH109 , and the transformed yeast cells were grown and selected on SD-L and SD-L/H/A media , respectively . Each experiment was independently repeated for three times with similar results . The LCI assays for the protein interaction detection was performed in N . benthamiana leaves as described previously [33] . Briefly , the full-length or truncated forms of the genes were separately fused with the N- and C-terminal parts of the luciferase reporter gene LUC . Agrobacteria cells harboring the nLUC and cLUC derivative constructs were co-infiltrated into N . benthamiana leaves , and the LUC activities were analyzed 48 h after infiltration using NightSHADE LB 985 ( Berthold ) . In each analysis , five independent N . benthamiana leaves were infiltrated and analyzed , and totally three biological replications were performed with similar results . The transcriptional activity assays were performed in N . benthamiana leaves as previously described [23] . The 2-kb CO promoter sequence was amplified from Col-0 genome DNA , and fused with the luciferase reporter gene LUC through Gateway reactions ( Invitrogen ) into the plant binary vector pGWB35 [34] to generate the reporter construct COpro:LUC . For the construction of the effectors , the full-length coding sequences of indicated genes were amplified and cloned into the plant binary vector pGWB17 [34] . The reporter and effector constructs were separately introduced into Agrobacterium strain GV3101 ( pMP90 ) , to carry out the co-infiltration in N . benthamiana leaves . LUC activities were observed and quantified 48 h after infiltration using NightSHADE LB 985 ( Berthold ) . In each experiment , 10 independent N . benthamiana leaves were infiltrated and analyzed , and totally three biological replications were performed with quantification . The co-immunoprecipitation ( Co-IP ) analyses were performed using transgenic Arabidopsis seedlings or Agrobacterium-infiltrated N . benthamiana leaves . The indicated genes were combined into pGWB5 and pGWB17 vectors to produce the GFP- or Myc-fused constructs , and transformed into Agrobacterium strain GV3101 . For Agrobacterium-mediated transient expression , the Agrobacteria harboring indicated derivatives were co-infiltrated into the N . benthamiana leaves , and the samples were collected 48 h post infiltration . The total proteins were extracted using the lysis buffer ( 50 mM Tris-HCl at pH 7 . 5 , 150 mM NaCl , 5 mM EDTA at pH 8 . 0 , 0 . 1% Triton X-100 , 0 . 2% NP-40 ) with freshly added PMSF ( phenylmethylsulphonyl fluoride , 10 mM ) and protease inhibitor cocktail ( Roche , 11873580001 ) . Anti-GFP- and Anti-Myc-conjugated agarose beads ( MBL , M047-8 and D153-8 ) were used for the immunoprecipitation . In western blotting , anti-GFP ( 1:2000; Roche , 11814460001 ) , anti-Myc ( 1:5000; Roche , 11667149001 ) and anti-mouse IgG ( 1:75000 , Sigma , A9044-2ML ) antibodies were used for the detection of GFP- and Myc-tagged proteins . Totally three independent biological replicates were performed . For FLIM-FRET assays , the indicated proteins were separately fused with cyan fluorescent protein ( CFP ) and yellow fluorescent protein ( YFP ) to generate the donors and receptors . The FLIM-FRET experiments were performed as described previously [25] with some modifications . Briefly , the donors and receptors were co-expressed in Arabidopsis protoplast cells . After 24 h incubation , the CFP and YFP fluorescence signals were imaged with the confocal microscope ( Carl Zeiss , LSM880 ) . Förster resonance energy transfer was measured by fluorescence lifetime imaging using the PicoHarp 300 time-correlated single-photon counting ( TCSPC ) module ( PicoQuant , Germany ) . A pulsed laser ( 405 nm ) tuned at 84MHz was used to excite the CFP . The emission from 450 to 490 nm was collected by the detector in 512×512 pixel format . The acquired FLIM decay curve from regions of interest ( ROI ) was fitted by two-exponential theoretical models using SymPhoTime 64 software , and the mean CFP lifetimes were calculated as the mean values of the fit function and analyzed using SymPhoTime 64 software . In each analysis , 6 independent nuclei were quantitatively analyzed , and totally three independent replications were performed with similar results . Arabidopsis seedlings were grown on half-strength Murashige and Skoog medium in 16-h-light/8-h-dark LDs for 14 days , and the samples ( 2 to 3 grams ) were collected at ZT 14 . The ChIP assays were carried out following the procedure described previously [23] . The ChIP assays were separately performed with ( + αGFP ) or without ( - αGFP ) the anti-GFP antibody . Finally , the GFP-specific enrichments of the fragments from CO promoter were analyzed by qPCR , and the enrichment fold of a certain fragment was calculated by normalizing to the amount of ACTIN7 promoter enriched in the same sample . Anti-GFP-CHIP grade antibody ( Abcam , ab290 ) and protein G plus agarose ( Santa cruz , sc-2002 ) were used for the immunoprecipitation . The enrichment of DNA fragments was determined by qPCR with specific primers , as shown in S4 Table . Three biological replicates were performed . The maltose binding protein ( MBP ) tagged TCP4 protein was expressed in E . coli strain Transetta-DE3 ( Transgen biotech , CD801 ) , and purified using the amylose resin ( New England Biolabs , E8021V ) following the manual . The CO promoter probes containing the TBM cis-element were synthesized and labeled with digoxigenin-11-ddUTP at the 3’ end by using DIG gel shift kit ( Roche , 03353591910 ) . Unlabeled dimerized oligo-nucleotides of CO promoter fragments containing the wild type or mutated TBM cis-elements were generated as the competitors . EMSAs were performed as previously described [33] . Competition for TCP4 binding was performed with 125× cold probes containing TBM cis-elements [GGACC ( C/A ) ] or mutated TBM cis-elements ( AAAAAA ) . Sequences of probes and competitors are shown in S2 Table . Three biological replicates were performed . Sequence data from this study can be found in the Arabidopsis Genome Initiative database under the following accession numbers: CO ( At5g15840 ) , TCP2 ( At4g18390 ) , TCP3 ( At1g53230 ) , TCP4 ( At3g15030 ) , TCP10 ( At2g31070 ) , TCP24 ( At1g30210 ) , FT ( AT1G65480 ) , FBH1 ( At1g35460 ) , FBH2 ( At4g09180 ) , FBH3 ( At1g51140 ) , FBH4 ( At2g42280 ) , CDF1 ( At5g62430 ) , CDF2 ( At5g39660 ) , CDF3 ( At3g47500 ) , PFT1 ( At1g25540 ) , ACTIN7 ( At5g09810 ) and U6 ( At3g14735 ) . | Plants monitor day-length changes ( photoperiod ) throughout the year to precisely align their flowering time , which is crucial for successful reproduction . In Arabidopsis , some components , such as CONSTANS ( CO ) , have been proved to play central roles in promoting the photoperiodic flowering under long-day conditions ( LDs ) . In this study , we demonstrate that the microRNA319-regulated TEOSINTE BRANCHED/CYCLOIDEA/PCF ( TCP ) transcription factors directly bind to the CO promoter . Meanwhile , these TCPs physically interact with the flowering activators FLOWERING BHLH ( FBHs ) and the flowering regulator PHYTOCHROME AND FLOWERING TIME 1 ( PFT1 ) to form a complex to activate CO transcription and promote photoperiodic flowering under LDs . Our results emphasize the importance of miR319-regulated TCPs in regulating plant flowering time . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"biotechnology",
"plant",
"anatomy",
"gene",
"regulation",
"regulatory",
"proteins",
"brassica",
"dna-binding",
"proteins",
"dna",
"transcription",
"plant",
"science",
"model",
"organisms",
"luminescent",
"proteins",
"genetically",
"modified",
"plants",
"yellow",
"fluorescent",
"protein",
"experimental",
"organism",
"systems",
"transcription",
"factors",
"seedlings",
"plants",
"flowering",
"plants",
"genetic",
"engineering",
"research",
"and",
"analysis",
"methods",
"arabidopsis",
"thaliana",
"genetically",
"modified",
"organisms",
"proteins",
"gene",
"expression",
"leaves",
"biochemistry",
"plant",
"and",
"algal",
"models",
"genetics",
"biology",
"and",
"life",
"sciences",
"plant",
"biotechnology",
"organisms"
] | 2017 | MicroRNA319-regulated TCPs interact with FBHs and PFT1 to activate CO transcription and control flowering time in Arabidopsis |
Dengue ( DEN ) and yellow fever ( YF ) are re-emerging in East Africa , with contributing drivers to this trend being unplanned urbanization and increasingly adaptable anthropophilic Aedes ( Stegomyia ) vectors . Entomological risk assessment of these diseases remains scarce for much of East Africa and Kenya even in the dengue fever-prone urban coastal areas . Focusing on major cities of Kenya , we compared DEN and YF risk in Kilifi County ( DEN-outbreak-prone ) , and Kisumu and Nairobi Counties ( no documented DEN outbreaks ) . We surveyed water-holding containers for mosquito immature ( larvae/pupae ) indoors and outdoors from selected houses during the long rains , short rains and dry seasons ( 100 houses/season ) in each County from October 2014-June 2016 . House index ( HI ) , Breteau index ( BI ) and Container index ( CI ) estimates based on Aedes ( Stegomyia ) immature infestations were compared by city and season . Aedes aegypti and Aedes bromeliae were the main Stegomyia species with significantly more positive houses outdoors ( 212 ) than indoors ( 88 ) ( n = 900 ) ( χ2 = 60 . 52 , P < 0 . 0001 ) . Overall , Ae . aegypti estimates of HI ( 17 . 3 vs 11 . 3 ) and BI ( 81 . 6 vs 87 . 7 ) were higher in Kilifi and Kisumu , respectively , than in Nairobi ( HI , 0 . 3; BI , 13 ) . However , CI was highest in Kisumu ( 33 . 1 ) , followed by Kilifi ( 15 . 1 ) then Nairobi ( 5 . 1 ) . Aedes bromeliae indices were highest in Kilifi , followed by Kisumu , then Nairobi with HI ( 4 . 3 , 0 . 3 , 0 ) ; BI ( 21 . 3 , 7 , 0 . 7 ) and CI ( 3 . 3 , 3 . 3 , 0 . 3 ) , at the respective sites . HI and BI for both species were highest in the long rains , compared to the short rains and dry seasons . We found strong positive correlations between the BI and CI , and BI and HI for Ae . aegypti , with the most productive container types being jerricans , drums , used/discarded containers and tyres . On the basis of established vector index thresholds , our findings suggest low-to-medium risk levels for urban YF and high DEN risk for Kilifi and Kisumu , whereas for Nairobi YF risk was low while DEN risk levels were low-to-medium . The study provides a baseline for future vector studies needed to further characterise the observed differential risk patterns by vector potential evaluation . Identified productive containers should be made the focus of community-based targeted vector control programs .
Dengue ( DEN ) and yellow fever ( YF ) are re-emerging diseases of public health importance caused by arboviral pathogens [1–4] . Both diseases share a common ecological niche including non-human primates as reservoir hosts and are vectored primarily by Aedes ( Stegomyia ) species [5] . Dengue fever is caused by one of the four serotypes of the dengue virus ( DENV 1–4 ) with about 390 million infections reported worldwide each year , 16% of which are from Africa [6 , 7] . Additionally , an estimated 900 million people are living in YF endemic areas with about 90% of the global infections reported from Africa [8 , 9] . The rapid geographic spread of these diseases in recent times in Africa and especially in East Africa represents a worrying new trend with occurrence of major epidemics affecting urban human populations [10 , 11] . This is exemplified by recent DEN outbreaks in Somalia 2011 , 2013 [12] , Tanzania 2013 , 2014 [4 , 13] , Sudan 2010 , 2015 [14 , 15] and various parts of Kenya 2011 , 2013 , 2015 [1 , 2] . An outbreak of YF was reported in Kenya in 1992–93 [16] , in Sudan 2003 , 2005 , 2012 [17–19] and neighboring Uganda 2011 , 2016 [20 , 21] . Despite the fact that the last YF outbreak in Kenya occurred over two decades ago , the country is still classified among countries with medium to high risk of YF transmission in Africa [22] , and a number of YF cases have recently been imported from Angola where there was an ongoing outbreak [21] . There are currently no antiviral drugs available for either DEN or YF . However , there is a safe efficacious vaccine against YF , and a new , partially approved vaccine for DEN , for use only in geographical settings where epidemiological data indicate a high burden of the disease [23] . Unfortunately , the costs and availability of these vaccines have proved to be challenging for effective disease prevention . While the recent DEN and YF outbreaks in Africa have attracted renewed public health and research attention , effective monitoring and risk assessment for their occurrence remains limited . Dengue virus ( DENV ) is known to be transmitted primarily by Aedes furcifer in Africa and Ae . aegypti aegypti in Asia and the Americas [5] . Aedes aegypti aegypti is highly anthropophilic and its larvae develop mostly in artificial containers in and around human habitations , compared to the more sylvatic Ae . aegypti formosus subspecies which develop mostly in tree holes hence linking the emergence of DEN in tropical urban areas to Ae . aegypti aegypti [24 , 25] . Although the role of Ae . aegypti in the transmission of yellow fever virus ( YFV ) in East Africa is poorly understood , it plays an important role in YFV transmission in West Africa , driving human-to-human transmission and resulting in dreaded urban outbreaks [26 , 27] . Yellow fever outbreaks in East and Central Africa have so far been associated with Ae . bromeliae , a member of the Ae . simpsoni species complex [28–30] . Aedes bromeliae is a peri-domestic mosquito species capable of biting humans and monkeys , thereby driving small scale outbreaks in rural populations , with potential to move virus across species from primates to humans [5] . Other species such as Ae . africanus and Ae . luteocephalus , feed on forest monkeys and sustain the sylvatic cycle of YF [31] . Although Ae . albopictus a secondary DEN vector is not known to be present in Kenya , Ae . aegypti and Ae . bromeliae are present in the major cities [32] , hence the need to assess the risk of arboviral disease emergence associated with these vectors . Risk assessment through surveillance of abundance and distribution of Aedes mosquitoes , which are key players in transmission of the pathogens that cause these diseases is critical . This largely relies on estimation of traditional Stegomyia indices ( House Index-HI , Container Index-CI and Breteau Index-BI ) of immature mosquito populations in households [33–36] . Estimation of such indices may be of operational value and can facilitate the determination of local vector densities and measurement of the potential impact of container-specific vector control interventions such as systematically eliminating or treating larval habitats with chemicals [37] . Surprisingly , estimations of these indices as a means of assessing risk of DEN and YF in Kenya are scarce and/or exclusive to Ae . aegypti in outbreak situations [31] . Moreover , similar investigations on other Stegomyia species such as Ae . bromeliae are completely lacking , in spite of its’ potential role in YFV transmission in Africa [5] . Unplanned urbanization remains an important risk factor that has contributed to the resurgence of these diseases by providing abundant larval habitats from water-retaining waste products and storage facilities in the presence of susceptible human populations [38–40] . A better epidemiologic understanding of entomological thresholds relating to risk can help to prevent a severe outbreak in urban settings . Potential exists for emergence of these diseases , especially YF from proximal sylvan areas , and subsequent introduction into urban areas where dense susceptible populations and competent domestic vectors abound [41] , as demonstrated by the recent YF outbreak in Angola and the Democratic Republic of Congo [11 , 21] . To assess the potential risk of urban transmission of these diseases we estimated HI , CI and BI in the three major cities of Kenya , namely Kilifi ( DEN-prone ) and Kisumu and Nairobi ( DEN-free ) in the light of known differential outbreak reports of DEN . These cities , which serve as major tourism , trade and shipping hubs for much of eastern Africa , have high levels of human population movement and potential for heightened risk of importation of viruses . We also investigated possible seasonal patterns and associated risk indices for Ae . aegypti and Ae . bromeliae , as the two vector species implicated in disease transmission in East Africa , inclusive of Kenya . We further characterized the most productive container types based on the number of immature mosquitoes surveyed , reared to adults , and identified; information , which can be used to guide targeted source reduction/control operations .
The study was carried out on the outskirts of the major cities of Kenya; Nairobi and Kisumu ( with no history of DEN outbreak ) and Mombasa ( DEN endemic and outbreak prone ) . While the phenomenon of DEN expansion is associated with urban human settlement , incidence of the disease in rural areas is also on the rise and is sometimes even higher than in urban and semi-urban areas/communities [40 , 42 , 43] . Therefore , our study targeted the cities , where we specifically selected sites in peri-urban suburbs around the main cities , Githogoro ( Nairobi County ) , Kisumu ( Kisumu County ) and Rabai ( suburb within Kilifi County , at the outskirts of Mombasa city ) , mainly for logistical reasons , including ease of access to homesteads and households . Githogoro is located about 13 . 1 km from the Central Business District ( CBD ) on the outskirts of Nairobi ( 01°17'S 36°48'E ) , the largest city and capital of Kenya ( Fig 1 ) . Nairobi has a total surface area of 696 km2 , a population of 3 . 1 million people [44] , and is situated at an altitude of 1 , 661 m above sea level ( asl ) . Githogoro is an urban informal settlement with most of the houses made of iron sheeting and consisting of a single room . A few houses have more than one room and some yard space . In Kisumu ( 00°03′S 34°45′E ) , the study sites included Nyalenda B , Kanyakwar and Kajulu located on the outskirts of Kisumu CBD at a distance of approximately 6 . 5 km , 5 . 8 km and 27 . 8 km , respectively . Kisumu is the third largest city in Kenya and the second most important city after Kampala in the greater Lake Victoria basin ( Fig 1 ) . It has a human population of >400 , 000 [44] and is situated at an altitude of 1 , 131 m asl . The houses in this area mostly have cemented walls and roofs made of iron sheeting . Water storage in containers is a common practice by the communities . The study sites included Bengo , Changombe , Kibarani , and Mbarakani , in Rabai , which is located on the outskirts of Mombasa , though administratively it belongs to Kilifi County ( Fig 1 ) . Rabai is situated about 24 . 5km to the north-west of Mombasa CBD , the second largest city in Kenya , which is situated on an island ( 4°03'S 39°40'E ) . Mombasa has a total surface area of 294 . 7 km2 , a population of 1 . 2 million people [44] and is situated at an attitude of 50 m asl . The houses in Rabai have walls that are either cemented , made of stones , or mud . The roofing system consists of iron sheeting or grass thatch . Water storage in containers is an equally common practice in these communities . All three-study cities generally experience two rainy seasons , the long rains season ( April-June ) and the short rains season ( October-December ) , interspersed by two dry seasons ( January-March and July-September ) . We conducted a cross-sectional survey of water holding containers situated both indoors and outdoors for presence of immature mosquito stages ( larvae at all instars and pupae ) . The inspections and entomological surveys were conducted by a team of four trained personnel in houses that were selected at random for the initial survey . An interval of one house was applied during the first sampling and unique numbers assigned to each house for ease of identification in subsequent surveys during the next season . In cases where a house could not be sampled in subsequent surveys , either due to absence of the inhabitants or the owners declining entry , it was substituted for the next closest available house . Each survey was conducted over five consecutive days and 100 houses from the selected sites were targeted , within each of the three main urban areas ( Nairobi , Kilifi , Kisumu ) . Repeat sampling of the same 100 houses / city was conducted for the dry season ( July-September 2015 in Nairobi; January-March 2016 in Kilifi and Kisumu ) and for the long rains ( April-June 2015 in Kilifi , and Kisumu; April-June 2016 in Nairobi ) and short rains ( October-December 2014 in Kilifi , and Kisumu , October-December 2015 in Nairobi ) seasons . As such , there was a total of three sampling occasions ( with 100 houses being sampled per study city and per season , corresponding to 900 sampling points ) , for the survey conducted from October 2014 to June 2016 . Sampling in Nairobi was limited to Githogoro , whereas in Kilifi ( Rabai ) and Kisumu , operational surveys were conducted to reflect the proportionate size of each site in terms of the number of houses present . These sites were Bengo , Kibarani , Changombe and Mbarakani in Kilifi and Kajulu , Kanyakwar and Nyalenda B in Kisumu . The survey of immature stages of Aedes Stegomyia mosquito species targeted artificial water-holding containers ( indoors and outdoors ) of any size and natural breeding sites ( tree holes , banana axils , flower axils and colocasia ) in peri-domestic areas of selected houses . Sampling was carried out using standardized sampling tools based on the type of water holding container encountered [45] . For small discarded containers ( mostly found around the house , holding water which is not for household use ) , the water was emptied into a white tray and a plastic Pasteur pipette was used to collect the immatures . Jerrican ( small plastic containers , 5-40L holding water for household use ) surveys entailed pouring the water through a sieve into a bowl with a good contrast and collecting all immatures from the sieve with an aspirator . In large containers such as metal and plastic drums ( 50-210L containers used to store water for household use ) , the immatures were collected using ladles and aspirators when less than 20 were present or by emptying the water through a sieve when there were more than 20 . Ladles , aspirators and pipettes were used to collect immatures from tyres as well as from tree holes and leaf axils . Flashlights were used where necessary . We captured information on each container sampled including: indoor or outdoor , natural or artificial , and the capacity of the container ( >70L , 20L-70L , <20L ) . Immatures collected from containers were placed in whirlpaks ( Nasco , FortAtkinson , WI ) labeled with the pertinent information and transported to the field laboratory . Larval samples were placed in individual rearing trays for each container types . All pupae collected for the separate container types were transferred to individual adult cages . Larvae were fed fish food ( Tetramin ) daily and the trays were inspected twice a day and pupae transferred to adult cages as well . This was done until all collected larvae/pupae had emerged to adults . During rearing , male and female Aedes mosquitoes were left together in a cage ( small plastic buckets covered with fine netting materials and secured with rubber bands ) and supplied with a 6% glucose solution on cotton wool . At the end of each sampling session , all adults were knocked down using triethylamine , placed in cryotubes and preserved in liquid nitrogen for transportation to the laboratory at the International Centre of Insect Physiology and Ecology in Nairobi . In the laboratory the resulting adult mosquitoes were morphologically identified using available taxonomic keys [46–48] and counted and data on the species and number collected from the different container types were captured in Excel . A container was considered positive when at least one Ae . aegypti or Ae . bromeliae larva or pupa was found [45] , and a house positive if at least one container type indoor was found infested with Ae . aegypti and/or Ae . bromeliae larvae . We estimated the classical Stegomyia indices: HI ( percentage of houses infested with Ae . aegypti or bromeliae immatures ) , CI ( percentage of water-holding containers infested with Ae . aegypti or bromeliae immatures ) , and BI [number of Ae . aegypti or bromeliae positive containers ( indoor and outdoor ) per 100 houses inspected] . We tested for significance of area/site and for seasonal effects in the patterns of observed indices ( BI , HI , CI ) using analysis of variance ( ANOVA ) followed by mean separation using the Tukey test ( P = 0 . 05 ) . Data for the different seasons were also pooled in each area to estimate the overall Stegomyia indices , and similarly compared for the different seasons and areas . Correlation analysis was performed to test for significant correlations between the indices- BI , HI , and CI . The density of Ae . aegypti ( total number of mosquitoes collected per total number of positive containers ) indoors and outdoors was established and the difference compared within each area using a t-test . The inspected containers were further categorized into 9 types based on similarity in certain features ( e . g . size , natural or artificial , etc ) . The productivity of each of these container types was calculated per season and area as the percentage of the total number of immatures ( larvae or pupae ) determined by the adults reared from the container types ( Productivity = 100 x ( total number of immatures ) / number of positive containers ) . We also applied ANOVA to test for significant differences in the proportion of positive containers ( positivity ) and compared the productivity among the container types after angular transformation . Container positivity for the different seasons was compared within an area using the Chi-Square test . All analyses were carried out in R version 3 . 3 . 1 [49] at α = 0 . 05 level of significance . Based on estimated indices we classified the areas/sites in terms of epidemic risk levels for YF or DEN as low , medium or high with reference to established epidemic thresholds [50 , 51] . HI values for Ae . aegypti and Ae . bromeliae were used to estimate risk of YFV transmission for the individual species with values of HI > 35% , BI > 50 and CI > 20% considered as high risk of urban transmission of YFV; HI < 4% BI < 5 and CI < 3% considered as unlikely or low risk of the disease transmission [50] . Similarly , the Pan American Health Organization ( PAHO ) has established threshold levels for dengue transmission based on HI for Ae . aegypti with low being an HI < 0 . 1% , medium an HI 0 . 1%–5% and high an HI > 5% [51] . We sought permission from household heads through oral informed consent to allow water-holding containers in their residences to be surveyed . Household survey of mosquitoes was carried out with ethical approval from Kenya Medical Research Institute Scientific and Ethics Review Unit ( KEMRI-SERU ) ( Project Number SERU 2787 ) .
A total of 11 , 695 mosquitoes were reared from the larvae and pupae collected from water holding containers , both indoors and outdoors , from all sites and cities . These included Ae . aegypti ( 63 . 5% ) , Ae . bromeliae ( 2 . 9% ) , Eretmapodite chrysogaster ( 1 . 9% ) and Culex spp . ( 31 . 53% ) . Aedes metallicus , other Aedes species ( Ae . tricholabis , Ae . durbanensis ) together with Aedeomyia furfurea , Uranotaenia spp , Anopheles gambiae s . l and Toxorhynchites spp . each comprised 0 . 1% or less of the total collection ( Table 1 ) . Focusing on our species of interest , a total of 7 , 424 Ae . aegypti were collected from all sites comprising 3 , 342 ( 45 . 0% ) from Kilifi , 3 , 733 ( 50 . 3% ) from Kisumu and 349 ( 4 . 7% ) from Nairobi with an overall higher proportion ( 76% ) being collected outdoors than indoors ( 24% ) . The Ae . aegypti densities recorded indoors and outdoors were not significantly different in the DEN-outbreak prone county of Kilifi ( n = 17 . 5 indoors , n = 15 . 4 outdoors , P = 0 . 7 ) . In contrast , counties of Kisumu ( n = 8 . 3 indoors , n = 16 . 8 outdoors , P = 0 . 036 ) and Nairobi ( n = 0 . 7 indoors , n = 14 . 7 outdoors , P = 0 . 048 ) ( with no documented records of DEN outbreaks ) had significantly higher densities of Ae . aegypti outdoors compared to indoors ( Fig 2 ) . Similarly , a total of 335 Ae . bromeliae were collected mainly outdoors ( 92% ) . The highest proportion was sampled in Kilifi ( 63% , n = 211 ) , followed by Kisumu ( 32 . 8% , n = 110 ) and then Nairobi ( 4 . 2% , n = 14 ) ( Table 1 ) . The rainy seasons recorded the highest proportions of Ae . aegypti in all three areas evaluated in this study . In Kilifi , long rains constituted 1 , 648 ( 49 . 3% ) of the total Ae . aegypti collected , followed by short rains 1 , 172 ( 35 . 1% ) with the lowest 522 ( 15 . 6% ) observed during the dry season . An analogous pattern was found in Kisumu and Nairobi . In Kisumu , the long rains , short rains and dry season each accounted for 1 , 470 ( 39 . 4% ) , 1 , 441 ( 38 . 6% ) and 822 ( 22 . 0% ) of the total Ae . aegypti sampled . Surprisingly , collection of Ae . aegypti in Nairobi was highest during the short rains 152 ( 43 . 6% ) , followed by the long rains 143 ( 41% ) and then the dry season at 54 ( 15 . 4% ) . However , the seasonal difference observed between long and short rains in Nairobi was not statistically significant ( χ2 = 0 . 38 , P = 0 . 5 ) . Relative to Ae . aegypti , very low numbers of Ae . bromeliae were encountered from containers during our study . However , a seasonal pattern of abundance , with the highest proportion collected during one of the rainy seasons , was observed at all the areas . In Kilifi , Ae . bromeliae collected during the long rains , short rains and dry seasons made up 52 . 9% , 45 . 1% and 1 . 9% , respectively , of the total collection . However , in Kisumu the highest proportion was recorded in the short rains ( 70 . 9% ) , while the long rains and dry seasons recorded 10% and 19 . 1% respectively of the total collection . In Nairobi , there was no record of Ae . bromeliae in the short rains and dry seasons , and this mosquito species was only recorded in the long rains . In terms of occurrence in container types , Ae . aegypti was mostly encountered in artificial containers such as jerricans , drums , tyres and other discarded containers at all the sites . However , to a lesser extent Ae . aegypti was found in natural container types such as tree holes and leaf axils in Kilifi and Kisumu ( Table 2 ) . Natural breeding sites like leaf axils were the most productive site for Ae . bromeliae at all the sites ( Table 3 ) . In fact , Ae . bromeliae was not found breeding in artificial containers in Nairobi , although to a minor extent it bred in artificial containers such as Jerricans and other discarded containers ( Table 3 ) in Kilifi and Kisumu , mostly co-habiting with Ae . aegypti . There was no significant difference in Ae . aegypti immature productivity by season or area . However , the contribution of container types to productivity of this species varied significantly ( Df = 9 , F = 6 . 41 P < 0 . 0001 ) . Significant differences were mostly observed between drums and animal drinking containers ( P = 0 . 0008 ) , drums and basins ( P = 0 . 01 ) , drums and natural breeding sites ( P = 0 . 002 ) , jerricans and animal drinking containers ( P = 0 . 01 ) , jerricans and natural breeding sites ( P = 0 . 02 ) , tyres and animal drinking containers ( P = 0 . 013 ) and between tyres and natural breeding sites ( P = 0 . 022 ) . Overall , in Kilifi , the most productive container types were jerricans ( 36 . 3% ) in the long rains , discarded containers ( 34 . 7% ) in the short rains , and drums ( 49 . 0% ) in the dry season ( Table 4 ) . Similarly in Kisumu , the most productive container types were the jerricans ( 29 . 5% ) in the long rains , drums ( 24 . 5% ) and discarded containers ( 24 . 1% ) in the short rains and drums in the dry ( 38 . 1% ) season ( Table 4 ) . In Nairobi , drums ( 32 . 9% ) were the most productive container types in the long rains , tyres ( 84 . 9% ) in the short rains , and tanks ( 63 . 0% ) in the dry season ( Table 4 ) . The most productive containers for Ae . bromeliae in Kilifi and Kisumu were discarded containers and natural breeding sites , while in Nairobi natural breeding sites were the most productive breeding sites ( Table 5 ) . Based on the number of each container types surveyed and the number positive , we found significant differences in container positivity between the areas ( Df = 2 , F = 9 . 6 , P = 0 . 0002 ) and seasons ( Df = 2 , F = 84 . 26 , P = 0 . 018 ) . Significant differences existed in the container type positivity between Kilifi and Kisumu [95% CI , ( 0 . 329 , 26 . 392 ) , P = 0 . 043] , Kisumu and Nairobi [95% CI , ( -37 . 214 , -11 . 152 ) , P < 0 . 0001] , but not between Kilifi and Nairobi . Generally , animal drinking containers and tyres were the most positive containers in Kilifi , tanks and discarded containers in Kisumu , and tyres and tanks in Nairobi . Similarly , container positivity was significantly different between the long rains and dry seasons [95% CI , ( 2 . 393 , 28 . 456 ) , P = 0 . 016] , long and short rains [95% CI , ( -27 . 122 , -1 . 059 ) , P = 0 . 03] , but not between the short rains and dry season . The proportion of positive containers was significantly different for all three seasons in Kilifi ( χ2 = 119 . 0 , P < 0 . 0001 ) and Nairobi ( χ2 = 31 . 7 , P < 0 . 0001 ) but not in Kisumu ( χ2 = 4 . 45 , P < 0 . 1078 ) . Tyres were the most positive containers both in the long and short rains in Kilifi while drums were the most positive containers in the dry season . In Kisumu , tanks constituted the most positive containers in the long rains , basins in the short rains and drums in the dry season . In Nairobi , discarded containers ranked as the highest positive containers in the long rains , tyres in the short rains and tanks in the dry season . The overall Ae . aegypti CI was higher during the long rains followed by dry season and then short rains in Kilifi . In Kisumu , CI was higher in the dry season , followed by the long rains and then short rains , while in Nairobi , CI was higher in the long rains followed by short rains and then dry season ( Fig 3A ) . The seasonal differences observed in all three cities were not significant ( P = 0 . 14 ) . However , the observed CI values were significantly different among the different cities ( Df = 2 , F = 16 . 69 , P = 0 . 012 ) , with differences recorded between Kilifi and Kisumu [95% CI , ( 0 . 483 , 35 . 450 ) , P = 0 . 046] , Kisumu and Nairobi [95% CI , ( -45 . 45 , -10 . 48 ) , P = 0 . 01] , but not between Kilifi and Nairobi . CI was equally significantly different even at smaller scale among the sites ( Df = 5 , F = 3 . 133 , P = 0 . 037 ) . Overall , CI was highest in Kanyarkwar ( Kisumu ) and lowest in Kibarani ( Kilifi ) . The overall Ae . aegypti HI was highest in the long rains ( 24% , 15% and 0% ) , compared to the short rains ( 20% , 12% and 0% ) and dry season ( 8% , 7% and 1% ) respectively in Kilifi , Kisumu , and Nairobi ( Fig 3B ) . Our analysis showed that overall HI values varied significantly in the different cities ( Df = 2 , F = 11 . 24 , P = 0 . 023 ) with among area differences recorded between Kilifi and Nairobi [95% CI , ( -29 . 96 , -4 . 04 ) , P = 0 . 02] , but not between Kilifi and Kisumu or Kisumu and Nairobi . Also , the overall HI was highest in Kanyarkwar ( Kisumu ) and lowest in Githogoro ( Nairobi ) . Overall BI for Ae . aegypti varied significantly across the seasons ( P = 0 . 044 ) , with highest values observed in the long rains ( 141 , 134 and 28 ) , compared to the short rains ( 82 , 83 and 7 ) and dry season ( 22 , 46 and 7 ) in Kilifi , Kisumu and Nairobi , respectively ( Fig 3C ) . Also , significant variation in the overall BI values was evident between areas ( BI: Df = 2 , F = 8 . 68 , P = 0 . 035 ) and seasons ( Df = 2 , F = 7 . 52 , P = 0 . 044 ) . Among-area differences were observed between Kisumu and Nairobi [95% CI , ( -145 . 66 , -3 . 68 ) , P = 0 . 043] , but not between Kilifi and Kisumu or Kilifi and Nairobi . Likewise significant seasonal differences in BI values occurred between the long rains and dry seasons [95% CI , ( 6 . 01 , 147 . 99 ) , P = 0 . 0386] , but not between the long and short rains , or the short rains and dry seasons in all three areas . Similarly , the overall BI was highest in Kanyarkwar ( Kisumu ) and lowest in Githogoro ( Nairobi ) . Based on HI values estimated for Ae . aegypti in reference to threshold levels for DEN transmission ( low HI < 0 . 1% , medium HI 0 . 1%–5% and high HI > 5% ) established by PAHO [51] , both Kilifi and Kisumu were classified as being at high-risk for DEN transmission in all three seasons , while Nairobi was classified as being at low risk in both the long and short rains and at medium risk in the dry season ( Table 6 ) . Even small-scale differences in DEN risk across sites among the major areas Kilifi and Kisumu were evident , highest in Kanyakwar ( Kisumu ) and Mbarakani ( Kilifi ) ( Table 6 ) . Similarly , with reference to the WHO threshold levels for urban YFV transmission ( low HI < 4% , Medium 4%-35% and high HI > 35% ) , our risk level values for Ae . aegypti , show that Kilifi and Kisumu could be classified as being at medium-risk of an urban YF epidemic in all three seasons based on estimated HI values , and Nairobi at low risk in all three seasons ( Table 7 ) . We found no significant difference in overall index values ( CI , HI and BI ) for Ae . bromeliae ( Fig 3D , 3E and 3F ) , among the three areas in the different seasons ( P > 0 . 05 ) . However , based on the HI estimated for this species , compared to the established threshold levels for urban YFV transmission [50] and assuming that Ae . bromeliae could transmit YFV , only Kilifi could be classified as being at medium risk during the long rains but at low risk in the short rains and dry seasons . Both Kisumu and Nairobi can be classified as being at low risk levels of transmission in all three seasons ( Table 8 ) . Equally strong positive correlations were recorded between the BI and HI ( R2 = 0 . 887 , P = 0 . 001 ) as well as the BI and CI ( R2 = 0 . 721 , P = 0 . 028 ) ( Table 9 ) .
Aedes aegypti and Ae . bromeliae were the major Stegomyia species recorded at all sites/cities , justifying estimation of indices for the two species considering their potential roles in DENV and YFV transmission [26 , 27 , 29 , 30] . Our findings support the sympatric existence of both species in these growing urban ecologies in Kenya . Although particular container types were more likely to be positive than others , it was noteworthy that these were not necessarily the most productive , suggesting that positivity did not always translate to productivity . Aedes aegypti in all three areas were mostly found breeding in jerricans , drums ( which were particularly productive in all seasons ) , tyres , and discarded containers . This was equally observed in an earlier study in Mombasa city , during entomologic investigations of a recent DEN outbreak [2] . These containers could be targeted at the community level through awareness creation and public health education for the control of Ae . aegypti mosquitoes . In this way , the local inhabitants can help reduce Ae . aegypti larval sites by reducing these containers in and near their homes or by properly covering them to prevent gravid females from laying their eggs in them [37] . Observations from this study show that Ae . aegypti is also capable of developing in natural sites especially in the water holding axils of banana plants . Aedes aegypti breeding in banana and colocasia plants have also been reported by Philbert and Ijumba ( 2013 ) in a study on the preferred breeding habitats of Ae . aegypti in Tanzania [52] . This adaptation should be monitored as it will take away any gains made from targeting control of breeding in artificial water holding containers . Immature stages of Ae . bromeliae , a species which is known to preferentially breed in phytotelmata , the water-holding axils of plants [53] , were also found developing in artificial containers indoors and outdoors in this study . Its ability to develop in artificial containers both indoors and outdoors has also been reported in another study in coastal Kenya [54] . Both Ae . aegypti and Ae . bromeliae were also found co-developing in several artificial and natural breeding sites . Utilization of artificial breeding sites may be an indication that Ae . bromeliae is increasingly adapting to the urban environment , bringing it closer to human hosts and increasing the risk of transmission of a range of the arboviruses that cause human disease , including YFV . Risk values for both Ae . aegypti and Ae . bromeliae were different not only between areas and seasons , but we found finer scale differences between the sites , suggesting spatio-temporal variation with non-uniform risk even within the same general ecology . Although water storage in containers is a common practice in these cities during the rainy and dry seasons , DEN outbreaks that have occurred in Mombasa have mostly been associated with the long and short rains [2] . The estimated HI and BI for Ae . aegypti both showed the same seasonal pattern in all three areas . The strong correlations between the traditional Stegomyia indices observed in this study , clearly indicates that they are all important in determining risk of transmission . It will also be important to investigate how the Stegomyia indices correlate with the observed DEN cases , especially in the coastal site of Kilifi County . Estimated risk values suggested that both Kilifi and Kisumu were at high risk of DEN transmission while Nairobi was at low risk . Based on our findings , risk of DEN in Kilifi is high especially during the long rains ( April-June ) and short rains ( November- December ) . This correlates with reports of DEN outbreaks in coastal Kenya , with outbreak peaks during the long and short rains in the 2013/2014 outbreaks [1 , 2] . High indices were also recorded in Mombasa city during this outbreak [2] , with HI values comparable to that reported for Kilifi and Kisumu in our study . High indices have also been recorded in neighboring countries of Ethiopia [55] and Tanzania [56] , which are prone to DEN outbreaks . Low indices were recorded in Nairobi , and this may partially explain the absence of reports of epidemic DEN in this part of the country , in spite of people arriving with infection from endemic areas during outbreaks [57] . Surprisingly , this study recorded high DEN risk indices in Kisumu yet there has been no reported outbreak in the region . This finding suggests that the mere presence of high abundance of Ae . aegypti as observed in Kisumu , may not be sufficient in estimating the risk of DEN transmission and that other factors should be considered including susceptibility of the Ae . aegypti population to the DENV , as well as their feeding behavior . All of these can affect vectorial capacity as has been demonstrated for Ae . albopictus [58] . We also observed significantly higher numbers of Ae . aegypti immatures outdoors compared to indoors in Kisumu and Nairobi . There is reason to believe that immatures will eventually emerge to adults posing biting risk to humans both indoors and outdoors in Kilifi compared to the outdoor risk in Kisumu and Nairobi , thereby leading to an increased risk of exposure to DEN transmission . This differential proximity of Ae . aegypti to human dwelling/activity may be a contributing factor to the differential epidemiology and outbreak pattern of DEN in the different cities . Earlier studies on the ecology of Ae . aegypti in the Kenyan coast suggested that the larvae of the domestic form Ae . aegypti aegypti develops indoors as opposed to the sylvatic form Ae . aegypti formosus which develops outdoors especially in forest tree holes and a polymorphic population which develops either indoors or outdoors in tree holes , steps cut into coconut palm trees , discarded tires , or tins [24] . Based on our observation , it is likely that the vector population in Kisumu and Nairobi is predominantly Ae . aegypti formosus , which has been described in other studies as a less efficient DEN vector when compared to Ae . aegypti aegypti [59 , 60] . A study to correlate the indoor vs outdoor larval habitats to possible genetic diversity among the species and susceptibility to DEN viruses is warranted . Aside from the aforementioned biological factors which can impact occurrence of DEN outbreaks , temperature is by far the most important climatic variable that can modulate this pattern [61] and should also be considered . Generally , the different study areas have different average monthly temperatures , 22°C to 28°C in Nairobi , 28°C to 30°C in Kisumu and 27°C to 31°C in the coastal area of Kenya where DEN is endemic . We are not sure how well the observed differences in the risk indices relate to the prevailing environmental temperature among the different areas . Higher temperatures have been shown to increase the ability of Ae . aegypti to transmit DENV by reducing the extrinsic incubation period [62–64] . However , it is important to note that the diurnal temperature fluctuations may be more important in modulating the transmission dynamics . This study only inferred risk from infestation patterns of Ae . aegypti . How these risks relate to actual prevalence in the human population is deserving of further consideration . There is evidence to suggest that some silent DEN transmission goes unreported in Kisumu , as a serological survey carried out by Blaylock et al . ( 2011 ) in this part of the country reported DEN seroprevalence levels of 1 . 1% . This value is similar to that reported by Morrill et al . ( 1991 ) for DEN in the coastal area of Kenya during non-epidemic periods [65] . Dengue is known to manifest clinically like malaria and diagnostic tools for DEN detection are unavailable in most health centers in the East African region , including Kenya [57] . It is therefore very important to confirm undiagnosed malaria cases , as it is possible some of these cases may actually be DEN . Generally , the risk of an urban YF epidemic occurring in Kenya based on vector abundance data observed in this study was classified as low to medium , with the risk due to Ae . aegypti being higher as compared to Ae . bromeliae . However , the role of Ae . aegypti in the transmission of YFV in East Africa has not been fully evaluated and in the documented outbreak that occurred in Kenya in 1992/93 , it was observed that this was driven by sylvatic vectors mainly Ae . africanus and Ae . keniensis and that Ae . aegypti was not at all associated with the outbreak [31] . Aedes bromeliae has also been described as a YFV vector in this region , as it was the principal vector in the largest YF outbreak that occurred in Omo River in Ethiopia [29] , as well as in outbreaks in Uganda [30] . Aedes simpsoni is a complex of at least three sister species of which only Ae . bromeliae has been implicated as a YFV vector [66] . To understand better the risk due to this species , it will be important to differentiate the sub-species occurring in these urban areas in parallel with vector competence status , which was outside the scope of this study . In Kilifi and Kisumu the high abundance of Ae . aegypti especially in the rainy season is considered sufficient to allow YFV transmission in association with other YFV vectors species such as Ae . bromeliae , Aedes metallicus and Er . chrysogaster found at some of the sites . However , their ability to act as efficient YFV vectors in urban areas in Kenya needs to be evaluated as data on their vectorial capacity is completely lacking . It is important to note that high numbers of Ae . bromeliae were recorded in our study area in Kilifi , and that clarification of the role of this species in the transmission of endemic arboviruses , such as DENV and chikungunya virus is needed , as it may be acting as a potential secondary vector . In conclusion , Ae . aegypti remains the only known DEN vector in Kenya with sufficient abundance in the major cities to sustain transmission . It is highly abundant and the risk values are indicative of high risk of DEN transmission in Kilifi and Kisumu . The key containers that are utilized by this species for oviposition are water storage containers that can be effectively targeted to reduce vector numbers and , consequently , the risk of virus transmission through community mobilization and public health education . The oviposition site preference , indoor vs outdoor containers , between the study areas is suggestive of behavioral and/or genetic variation occurring in the different vector populations , calling for further studies . Overall , our findings provide a baseline for future studies to understand further the observed differential risk patterns especially with respect to the vectorial capacity of the different populations of Ae . aegypti and Ae . bromeliae for DENV and YFV transmission . | Despite the growing problem of dengue ( DEN ) and yellow fever ( YF ) evidenced from recent outbreaks in East Africa , risk assessment for their transmission and establishment through surveys of populations of the Aedes mosquito vectors , remain scarce . By estimating standard indices for the potential vectors , Aedes aegypti and Aedes bromeliae we partly could deduce the risk of transmission of these diseases in three major cities of Kenya , namely Kilifi ( DEN-prone ) and Kisumu and Nairobi ( without any DEN outbreak reports ) . When compared to established threshold risk levels by WHO and PAHO , our findings suggest low-to-medium risk of urban YF , and high risk of DEN transmission for Kilifi and Kisumu but not Nairobi ( low risk level for YF and low-to-medium risk for DEN ) . The observed seasonal risk patterns , higher Aedes infestation outdoors than indoors and productive container types ( jerricans , drums , discarded containers and tyres ) , provide insights into the disease epidemiology and are valuable for targeted vector control , respectively . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"plant",
"anatomy",
"medicine",
"and",
"health",
"sciences",
"geographical",
"locations",
"tropical",
"diseases",
"animals",
"developmental",
"biology",
"plant",
"science",
"neglected",
"tropical",
"diseases",
"insect",
"vectors",
"africa",
"infectious",
"diseases",
"geography",
"aedes",
"aegypti",
"dengue",
"fever",
"life",
"cycles",
"leaves",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"kenya",
"urban",
"areas",
"earth",
"sciences",
"geographic",
"areas",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"viral",
"diseases",
"larvae",
"organisms"
] | 2017 | Assessment of risk of dengue and yellow fever virus transmission in three major Kenyan cities based on Stegomyia indices |
An explosive outbreak of dengue fever occurred in Guangdong Province , China in 2014 . A community-based integrated intervention was applied to control this outbreak in the capital city Guangzhou , where dengue epidemic was mainly caused by imported cases . We used a time series generalized additive model based on meteorological factors to assess the effectiveness of this intervention . The results showed that there was significant reduction in mosquito density following the intervention , and there was a 70 . 47% ( 95% confidence interval: 66 . 07% , 74 . 88% ) reduction in the reported dengue cases compared with the predicted cases after 12 days since the beginning of the intervention , we estimated that a total of 23 , 302 dengue cases were prevented . This study suggests that an integrated dengue intervention program has significant effects to control a dengue outbreak in areas where dengue epidemic was mainly caused by imported dengue cases .
Dengue fever , a mosquito-borne viral disease caused by any of the four dengue virus serotypes , is regarded as one of the most important arboviral diseases globally [1] . At present , no specific antiviral treatment or vaccine against dengue fever is available . It was estimated that about 2 , 500 million people live in areas at the threat of dengue infection worldwide [2] . Currently , Dengue fever distributes in most tropical countries of the South Pacific , Asia , the Caribbean , the Americas , and Africa; the importance of dengue to public health is growing rapidly due to its geographical expansion probably resulting from population growing , increasing population movement , environmental change , particularly climate change [3 , 4] . Guangdong Province has the highest dengue infestation level in mainland China [5] . In 1978 , dengue fever re-emerged in Guangdong Province after disappearing for about 30 years in mainland China . Since then , it occurred in Guangdong almost every year [6 , 7] . The dengue cases in this province were mainly caused by imported cases from surrounding dengue endemic countries and areas [8] . An unprecedented explosive outbreak of dengue fever occurred in Guangdong Province in 2014; the case number was more than 10 times of the total number in previous 10 years with 6 deaths [9] . During the initial stage before September 23 , only routine control measures were implemented around the outbreak sites , which focused on insecticide space-spraying . However , since September 23 , the increasing intensity and spatial expansion prompted the government to take a comprehensive integrated community-based control strategy [7] . A comparison of the two strategies was illustrated in S1 Table . Briefly , compared with the routine control measures , the integrated community-based control strategy is to mobilize all community partners to participate in the dengue control activity with the highest administrative leadership and support , and to monitor the mosquito density in all public places and to control the density at a safe level . Evaluation of the effectiveness of the intervention measures , referred to as accountability research , has been increasingly viewed as a necessary component of responsible governmental policy [10] . This study aims to assess the effectiveness of the comprehensive and intensified control measures based on community in Guangzhou , the capital city of Guangdong Province , China .
Guangzhou , the capital city of Guangdong Province , is situated in the southern China . It has an area of 7434 km2 and about 12 . 7 million inhabitants . The climate is subtropical humid , with an average annual temperature of 21 . 9°C , the highest mean temperature ( 33 . 0–34 . 9°C ) is observed between July and August and the lowest mean temperature ( 6 . 5–12 . 1°C ) between January and February , the annual average rainfall of 1500 to 2000 mm . This city has short , mild , dry winters and long , hot , wet summers . The present study was reviewed and approved by the Medical Ethics Committee of Guangdong Provincial Centre for Disease Control and Prevention . All the participants' medical data were anonymized , as we only used the daily number of dengue fever cases for this study . Integrated community-based control strategy required every community to be involved in the dengue fever control activity under the leadership of the government; a multi-sectoral collaboration mechanism was established; the health authority was responsible for the technical organization and inspection . Here , a community is defined as a residential unit situated in a given geographical area with an administrative organization , the geographic size may vary greatly . The integrated intervention measures included larval breeding eradication , killing adult mosquitoes with pesticides , public health education and community involvement , as well as rigorous administrative leadership . Taking Guangzhou as an example , a set of special financial support and resources was allocated for this intervention , it was estimated that a total of 3 . 3 million people have participated in this activity and 272 million RMB were spent to purchase pesticides and related instruments . This rigorous strategy was organized by the provincial and municipal governments , which convened the Special Dengue Control Committee for oversight , with technical advice and training . The Committee team and relevant provincial and municipal professionals met with health department staff to propose and discuss the strategy and to gain initial consent . After approval had been given , the horizontal component was implemented by the health personnel of each community , park , school teachers , etc; community members were also mobilized to do household cleaning , particularly water container management . The district and community health officials were responsible for coordination of different sectors , inspection , evaluation , summary of the field activity , and delivery of health education messages to the public . Each community was obliged to establish a dengue control team to implement standard dengue control activities: entomological surveillance and breeding source reduction through periodic inspection of houses , larviciding of various containers , adult mosquito density control , communication and education on dengue prevention , and enforcement of mosquito control legislation . All the public places , including hospitals , schools , parks , public squares and tourist sites , were requested to do mosquito density survey and report the survey results to the local health department every day . The school teachers and students were also required to participate in the clean-up campaigns , the children were educated to provide knowledge and control approaches to their family members , and participation in dengue or project-oriented plays , songs , quiz , and so on . Dengue fever has been a legally notifiable communicable disease in China since 1989 . Daily records of dengue cases between 2006 and 2014 in Guangzhou were obtained from the China National Notifiable Disease Reporting System . The information included age , sex , occupation , date of symptom onset , whether the diagnosis was clinical or confirmed by laboratory test , etc . In the study area , Aedes albopictus is the dominant transmission vector [11] . Herein , Breteau Index ( BI ) , one of the accepted indicators for Aedes density [12] , was collected from various districts in Guangzhou . From each district , three streets were selected as the BI monitoring points and water containers and mosquito larvae were checked from 100 houses through a weekly survey; the houses for the survey remained consistent during the study period . BI is calculated according to the number of positive containers per 100 houses inspected . Daily meteorological data , including mean temperature , relative humidity from 2006 to 2014 , were retrieved from the China Meteorological Data Sharing Service System ( http://cdc . cma . gov . cn/index . jsp ) . To evaluate the effectiveness of this integrated intervention program , we collected daily data on dengue fever and meteorological variables ( daily mean temperature ( °C ) and relative humidity ( % ) ) in Guangzhou for the period January 1 , 2006 through December 31 , 2014 . A generalized additive model with a quasi-Poisson link function to account for over-dispersion in daily dengue cases was utilized to establish the predicting model . In the model , the daily number of dengue cases was treated as dependent variable , and daily meteorological variables with certain lag days , temporal trend and public holidays ( PH ) were used as explanatory variables . Public holidays was defined as the holidays and weekend when people don't need to work . To control for the non-linear relationship between the explanatory variables and dengue fever , we used a smoothing function based on penalized splines for temporal trend and meteorological factors [13] , the degrees of freedom of the smoothing function were selected based on previous studies [14] . For example , we initially applied 7 df per year for time trends to filter out the information at time scales of longer than two months , 6 df for mean temperature , and 3 df for relative humidity . The model can be specified as: log[ E ( Yt ) ] =α+β1*AR ( dengue , 1 ) + s ( t , df=7/year ) + s ( Tempn , df =6 ) + s ( Humidityn , df=3 ) +β2*PH where E ( Yt ) is the expected number of dengue cases on day t , AR ( dengue , 1 ) is the term of auto-regression of dengue cases of previous day , α is the model intercept , s ( ) indicates a smoother based on penalized splines , df is the degree of freedom , t represents time to adjust for long-term trend and seasonality , Tempn is the mean temperature on a lag of n days , Humidityn presents the relative humidity on a lag of n days , the lag days ranged from 14 to 30 days according to the transmission pattern of this disease [15]; PH represents a binary variable for the public holidays , β is the regression coefficient . The model specification in terms of lag days of meteorological variables and degree of freedom for smoothing function was determined using the R square ( R2 ) criteria , the higher of the R2 value , the better model fit . We established the model using the data from January 1 , 2006 through September 23 , 2014 , which was then used to predict the dengue epidemic during August 25 to 31 December , 2014 . The reduction rate was calculated by comparing the predicted dengue cases with the actual cases . The sensitivity of the effect estimates was assessed in terms of the degrees of freedom of the smoothing function of temporal trends ( 5 , 6 and 8 per year ) and meteorological variables , including mean temperature ( df = 4 , 5 and 7 ) and relative humidity ( df = 4–6 ) . We also fitted the model using observation data of different cut-off time points ( for example , from January 1 , 2006 through September 5 , 2014 , and through September 15 , 2014 ) . We also applied an SEIR ( Susceptible , Exposed , Infected and Removed ) model to do the analysis to check the robustness of the effect estimate , the details of the method were shown in the supplementary materials .
During the study period , a total of 39 , 214 dengue cases were reported in Guangzhou . Among them , 36 , 837 were notified in 2014 with an incidence of 290 per 100 , 000 population , accounting for 93 . 94% of the total cases during the study period . There were slightly more female cases with male-to-female sex ratio of 0 . 96:1 ( 19248:19966 ) . From 2006 to 2014 , there were three epidemic years in which the number of annual dengue cases reached more than 700 ( i . e . , 774 cases in 2006 , 1 , 268 cases in 2013 , and 36 , 837 cases in 2014 ) . The incidence of dengue fever presented an obvious seasonal pattern with higher rate occurring from June to November ( Fig 1 ) . Fig 2 shows the temporal trend of Breteau Index during the period of June 2014 through December 2014 . The average index was 10 . 88 before the integrated intervention , after which , the index decreased to an average of 2 . 11 . An exponential decay model suggested that the Breteau index decreased in an exponential function ( alpha = 633 . 332 , gamma = -0 . 039 ) after the intervention . The model with moving average of 7–28 lag days’ meteorological variables and df of 7 for temporal trend , 6 for temperature and 3 for relative humidity was found to have the best model fit with R2 being 99 . 4% ( internal validity ) . During the outbreak in 2014 , the epidemic peak was observed on October 1 with 1 , 596 dengue cases , after that the epidemic began to decrease; while our predicted epidemic peak was estimated on October 5 with 1 , 810 dengue cases ( Fig 3 ) . The comparison of the epidemic curves in 2006–2013 and 2014 also supported that the epidemic peak in 2014 was relatively earlier than that in 2006–2013 . The reduction rate of daily dengue cases for October 5 to November 26 was 70 . 47% ( 95% confidence interval ( CI ) : 66 . 06% , 74 . 88% ) , about 23 , 302 dengue cases have been prevented . When we used alternative time period to examine the effect estimates , the reduction rate was 61 . 10% ( 95% CI: 57 . 26% , 70 . 95% ) , and about 22 , 641 dengue cases might have been prevented for the period of October 1 to December 31 , and 75 . 71% ( 95% CI: 73 . 26% , 78 . 17% ) , and the prevented dengue cases were 21 , 539 . Sensitivity analyses found that the effect estimates were relatively robust to the degrees of freedom of smoothing adjustment for temporal trend and weather variables ( S2 Table ) . For example , when we used 6 of df for temporal adjustment , the reduction rate was 70 . 43% ( 95% CI: 66 . 02% , 74 . 84% ) and an estimated 23 , 244 dengue cases might have been prevented by the intervention . And when using data of different time periods to fit the model , we also obtained similar effect estimates , for example , when we used data from January 1 , 2006 through September 15 , 2014 to construct the model , the reduction rate was 70 . 18% ( 95% CI: 65 . 67% , 74 . 70% ) , and the prevented dengue cases were 22 , 348 . The SEIR model also yielded a consistent result with that of main model ( as shown in S1 Fig ) , which estimated that 25 , 532 dengue cases might have been prevented .
For the past few decades , vector control methods to reduce mosquito breeding sites and density remained the mainstay of prevention and control of dengue fever [16] . This approach , however , is usually of questionable efficacy and is often inefficient due to absence of active community involvement [17] . Alternative approaches emerged in recent years , including genetically-modified mosquitoes , biological control methods ( such as Wolbachia ) , anti-viral drugs and vaccines [18] . The present study indicated that the comprehensive and intensified dengue intervention strategy based on community participation was effective to rapidly reduce the mosquito density and curtail the dengue outbreak in an area where dengue epidemic was mainly caused by imported cases . Routine dengue control measures rely mainly on vector control and generally consists of source reduction , larviciding and/or insecticide space-spraying [19] . However , the vector control strategy usually lacks effectiveness and sustainability [20] , while community involvement and enhanced government leadership strategies have been proved to be successful in a few countries , which included systematic human resources and prevention facility investment and scientific allocation of these resources [21] . Historically , there have been only a few examples of successful dengue prevention through vector control [22–24] . The first one was the highly successful , vertically structured paramilitary hemispheric eradication campaign directed by the Pan American Sanitary Board from 1946 to 1970 [22] . The second was also a rigorous , top-down , military-like vector control program in Cuba , which was based on intensive insecticidal treatment followed by larval habitat management in 1981 [23] . The third successful program was reported in Singapore [24] . Therefore , assessing some new dengue control strategies is very important to provide potentially high-impact interventions for resource-poor countries where dengue is a major public-health problem . The unsuccessful vector control strategy in some countries might be due to that the vector control strategy was unsustainable or lack of sufficient community participation , but did not necessarily mean that vector control measures were unable to reduce transmission . Actually , the experiences from Vietnam have showed integrated vector control strategy based on community involvement was effective in prevention and control of dengue fever epidemic [25] . Although the integrated intervention strategy has proven to be successful in Guangzhou , one challenge is how to keep the program sustainable by activating each component of the integrated community-based strategy in the provincial dengue contingency plan when an outbreak is first suspected . The strategy should also be considered by other countries and areas with similar situation as Guangzhou , such as the dengue outbreak was caused by imported dengue cases , and similar hierarchical structure to adopt the strategy . The hierarchical structure and social-economic status in different countries could affect successful adoption of the strategy; the key components rely on motivating community and individual engagement , which highly depends on their perception of the severity of the disease , and willingness to take responsibility . Among them , barriers to sustaining dengue vector control actions are significant and include , among others , onerous behaviors that must be carried out on a weekly basis , and which may not be perceived as beneficial given the time needed to carry them out; other barrier included some breeding sources of mosquitoes in the community that are not amenable to individual control and thus serve as a reminder to the community that their neighborhood lacks good quality public services; ineffective vector control strategies due to budget and personnel limitations . Along with the increasing importance of dengue fever , the authors predict that the comprehensive dengue intervention model , or modifications of it , will become increasingly important . In summary , this study suggests that a comprehensive and enhanced dengue intervention strategy based on community engagement has significant effect to control a dengue outbreak in areas where the dengue epidemic was mainly caused by imported cases . | Dengue fever , caused by any of the four dengue virus serotypes , is regarded as one of the most important arboviral diseases globally . Guangdong Province in south China has the highest dengue infestation level in mainland China . An explosive outbreak occurred in this province in 2014 , with a total of 36 , 837 cases and 6 deaths being notified . A community-based integrated intervention program was implemented to control this outbreak in Guangzhou , the capital city of the province , where dengue epidemic was mainly caused by imported cases . It was estimated that a total of 3 . 3 million people and 272 million RMB were invested in this intervention . This study used a time series generalized additive model based on meteorological factors to evaluate the effectiveness of this intervention program . The analysis showed that there was significant reduction in mosquito density following the intervention ( Breteau Index ( BI ) reduced from 10 . 88 to 2 . 11 ) , and there was a 70 . 47% ( 95% confidence interval: 66 . 07% , 74 . 88% ) reduction in the reported dengue cases compared with the predicted cases after 12 days since the beginning of the intervention . A total of 23 , 302 dengue cases were prevented due to the community-based intervention . This study suggests that an integrated dengue intervention program is effective to control a dengue outbreak in areas where dengue epidemic was mainly caused by imported dengue cases . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"china",
"atmospheric",
"science",
"tropical",
"diseases",
"geographical",
"locations",
"animals",
"health",
"care",
"mathematics",
"statistics",
"(mathematics)",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"humidity",
"insect",
"vectors",
"infectious",
"diseases",
"dengue",
"fever",
"epidemiology",
"disease",
"vectors",
"insects",
"arthropoda",
"people",
"and",
"places",
"mosquitoes",
"asia",
"meteorology",
"confidence",
"intervals",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"physical",
"sciences",
"community-based",
"intervention",
"health",
"education",
"and",
"awareness",
"organisms"
] | 2016 | Community Involvement in Dengue Outbreak Control: An Integrated Rigorous Intervention Strategy |
Reliable signal transmission constitutes a key requirement for neural circuit function . The propagation of synchronous pulse packets through recurrent circuits is hypothesized to be one robust form of signal transmission and has been extensively studied in computational and theoretical works . Yet , although external or internally generated oscillations are ubiquitous across neural systems , their influence on such signal propagation is unclear . Here we systematically investigate the impact of oscillations on propagating synchrony . We find that for standard , additive couplings and a net excitatory effect of oscillations , robust propagation of synchrony is enabled in less prominent feed-forward structures than in systems without oscillations . In the presence of non-additive coupling ( as mediated by fast dendritic spikes ) , even balanced oscillatory inputs may enable robust propagation . Here , emerging resonances create complex locking patterns between oscillations and spike synchrony . Interestingly , these resonances make the circuits capable of selecting specific pathways for signal transmission . Oscillations may thus promote reliable transmission and , in co-action with dendritic nonlinearities , provide a mechanism for information processing by selectively gating and routing of signals . Our results are of particular interest for the interpretation of sharp wave/ripple complexes in the hippocampus , where previously learned spike patterns are replayed in conjunction with global high-frequency oscillations . We suggest that the oscillations may serve to stabilize the replay .
The ground state of cortical networks is characterized by irregular and asynchronous spiking activity [1]–[4] and its dynamics are highly sensitive to perturbations , e . g . , missing or additional spikes [2] , [3] , [5]–[8] . Yet , reliable transmission of information in the presence of such perturbations is assumed to be essential for neural computation . A common hypothesis states that such transmission might be achieved by propagating signals along subnetworks ( layers ) connected in a feed-forward manner . Indeed , propagation of synchronous and rate signals in feed-forward networks ( FFNs ) has been demonstrated in vitro [9]–[11] and recent experiments suggest that , e . g . , the generation of bird-songs relies on activity propagation in feed-forward structures [12] . Moreover , sequential replay observed in hippocampal and neocortical areas also suggest such an underlying feed-forward structure [13]–[18] . Layered feed-forward networks that support propagation of synchrony are termed synfire chains [19]–[23] . The propagated signal is a synchronous pulse-packet [21] , [24] , i . e . , a fraction of synchronously active neurons of one layer which induces synchronous activity in the following , postsynaptic , layer and so on . Robust signal transmission in synfire chains embedded in larger recurrent networks is usually obtained by an increased connectivity ( compared to the embedding network ) between the neurons of successive layers of the FFN [25]–[27] . Alternatively , increased synaptic efficiencies [28] , or the combination of enhanced synaptic weights and non-additive coupling ( mediated by fast dendritic spikes , cf . [29] ) can enable such a robust propagation [30] , [31] . A hallmark of cortical dynamics is the presence of oscillations of various frequencies . A plethora of experimental studies links oscillations in , e . g . , delta- ( Hz ) , gamma- ( Hz ) , fast-gamma-band ( Hz ) or the high-frequency range of up to Hz ( “ripples” ) , to attentional states , sensory stimulus selection , ongoing information and memory processing [32]–[40] . In this article we investigate how background oscillations influence the transmission of synchronous activity in feed-forward networks . More precisely , we consider sparse feed-forward structures that emerge as part of a random network and that exhibit moderately enhanced synaptic efficiencies ( cf . also [30] , [41] ) . In particular , the feed-forward structures considered are too weak ( in the sense of connectivity and coupling strength ) to propagate synchronous signals on top of asynchronous background activity . However , we demonstrate that additional oscillatory input , excitatory and inhibitory spike trains generated by an external oscillating neuronal population , can enable robust propagation of synchrony . We consider both conventional additive couplings , mediated by transient conductance changes on the dendritic input site , and non-additive couplings that take nonlinear processing of inputs by fast dendritic spikes ( e . g . , [29] , [42]–[44] ) into account . These dendritic spikes are evoked by highly synchronous inputs ( i . e . , inputs arriving within a time window of less than a few milliseconds ) and cause strong , rapid depolarization in the soma of the postsynaptic neuron , exceeding the depolarization expected from additive processing of inputs . Thereby they may foster directed [30] , [31] and undirected [45] propagation of synchrony . We show that for additively coupled networks , external oscillations support propagation of synchrony only if the ( average ) excitatory input exceeds the inhibitory input . This exceeding causes a net depolarization of the neurons which in turn enables propagation of synchrony . However , there is no resonance between the propagating synchronous signal and the oscillatory stimulation , and temporally distributed external inputs would have the same effect . In contrast , for non-additively coupled networks the sensitivity of dendritic spike elicitation to synchronous inputs yields resonances to oscillations , i . e . , there is a specific stimulation frequency range which enables propagation of synchrony . Dendritic spikes are not suppressed by inhibition [46] such that they support synchrony propagation also if the inputs are balanced , i . e . , if the ( average ) inhibitory input equals ( or even exceeds ) the excitatory input . Interestingly , the existence of resonance frequencies provides the possibility to guide synchronous activity along different pathways with distinct resonance frequencies . The mechanism of oscillation-induced signal transmission is robust against changes of the system properties . In particular , networks with peaked and with broad delay distributions exhibit qualitatively similar transmission dynamics . Further , we identify the hippocampus as a core candidate region for oscillation-induced signal transmission as in the hippocampus both high-frequency oscillations and replay of spike patterns are simultaneously observed in experiments .
As a starting point , we investigate isolated FFNs and briefly describe the mechanism underlying propagation of synchrony in networks with and without dendritic nonlinearities . A detailed description of the neuron and network setup is provided in the Methods Section . Each neuron of the FFN receives much more input from the external homogeneous background than from the preceding layer . Therefore , in the absence of synchrony , the FFN's dynamics in the ground state is mainly determined by this external background input , and the neurons of the FFN fire asynchronously with a low rate . However , exciting a fraction of neurons of the first layer of the FFN to spike synchronously causes a synchronous input to the second layer , a fraction of which subsequently spikes synchronously . This process continues from layer to layer and thus can induce persistent propagation of synchrony . One can derive an iterated map ( cf . also [30] , [31] ) that specifies the average number of neurons which spike synchronously , i . e . , within a certain time interval , given that in the preceding layer neurons have spiked synchronously . We denote the probability for a neuron in the asynchronous ground state to spike within a time interval of milliseconds after receiving an input of strength by . Say that in some layer , neurons spike synchronously , then each neuron of the following layer will receive some number of synchronous inputs of strength . As each of the spikes sent is received by every neuron of the postsynaptic layer with probability , follows a binomial distribution , , such that on average ( 1 ) neurons spike within a time interval of milliseconds . We assess the temporal development of the size of the synchronous pulse in every layer by considering the average number of neurons spiking synchronously in layer as a function of the average number synchronous spiking neurons in the preceding layer . Thus , replacing by and by in Equation ( 1 ) we obtain the map ( 2 ) where is the continuous interpolation of the right hand-side of Equation ( 1 ) for continuous . The fixed points of the map ( 2 ) determine the stability region for the propagation of synchrony ( cf . Fig . 2 ) . For small coupling strength , there is only one fixed point and any synchrony propagation will extinguish within few layers ( cf . also Fig . 1b , e ) . For sufficiently large layer size and coupling strengths , stable propagation of synchrony can be achieved , the size and temporal spread of the synchronous pulse are stable throughout the layers ( for an extensive analysis see [31] ) : This is due to the appearance of two additional fixed points , ( unstable ) and ( stable ) , which emerge via a tangent bifurcation in the map ( 2 ) upon increasing . A synchronous pulse will propagate with a typical group size . In a given network , persistent propagation is possible if the connection strengths are larger than some critical value . We denote the critical connection strength , i . e . , the bifurcation point at which the fixed points and emerge , by for FFNs with linear dendrites and by for FFNs with nonlinear dendritic interactions . Stable propagation of synchrony occurs with a certain propagation frequency , which is defined as the inverse of the average time interval between two consecutive synchronous pulses . is governed by ( i ) the synaptic delay and ( ii ) the the spike latency , i . e . , the average time that an arriving input needs to trigger a spike in the postsynaptic neuron ( if it does so ) . The synaptic delay is fixed for a given setup , but in general depends on the strength of the input and thereby on the connection strength . For networks with linear dendrites , decreases with increasing input strength ( cf . Fig . 3a ) : The increase of the input causes a steeper and steeper rise of the evoked postsynaptic potential , and therefore reduces the ( average ) time the neuron needs to reach the threshold . In contrast , is constant for networks with nonlinear dendritic interactions: The spiking of the neuron is triggered by the additional current pulse mimicking the dendritic spike . This current pulse ( and with it the resulting depolarization ) is independent of the actual input strength ( see also Methods Section ) , and the rise of the postsynaptic potential is so steep that is practically constant for . We note that for large input the spike latency for neurons with nonlinear dendritic interaction is larger than for neurons without: The latency between dendritic stimulation and the onset of the somatic response to the dendritic spike can be estimated to ms [29] , [50] , and is therefore delayed compared to the onset of the somatic response to the linear ( electrically ) transmitted signal . As a consequence of the constancy of the latency , for FFNs with non-additive couplings the propagation frequency depends only weakly on the connection strength . If a propagation of synchrony is enabled for , this propagation occurs with a certain ‘natural’ propagation frequency . In contrast to linearly coupled FFNs , the propagation frequency remains approximately constant for connection strengths above the critical connection strength , ( Fig . 3b ) . For connection strengths satisfying the propagation frequency jumps: If is increased above for some , a smaller number of spikes can trigger a dendritic spike , i . e . , a reduced fraction of the synchronous pulse packet is sufficient to trigger dendritic spikes , such that the neurons in each layer tend to spike earlier . This shortens the ( average ) responding time to the synchronous pulse packet and the propagation frequency increases . We remark that for large connection strengths , the FFN enters a pathological state of activity: Neurons of one particular layer share inputs from the preceding layer and this causes correlations in their spiking activity . If the single connections become stronger ( i . e . , only a few inputs are needed to generate a dendritic spike and a somatic output spike ) also these correlations become stronger . They may accumulate over the layers of the FFN and lead to spontaneous synchronous spiking activity propagating along the later layers of the FFN [47]–[49] . Thus , there exist cutoff-connection strengths and for networks with linear and nonlinear dendritic interactions , above which the global spiking activity is characterized by network oscillations and a meaningful propagation of synchronous activity is not possible anymore . Whereas signal transmission is possible in FFNs with and without dendritic nonlinearities , the underlying mechanism is different: In linearly coupled networks transmission is achieved by eliciting somatic spikes directly , thus also asynchronous inputs and depolarizing constant external currents may contribute to spike propagation . In nonlinearly coupled networks transmission is mediated by dendritic spikes ( all-or-none events ) , and therefore only highly synchronized spiking input contributes . Depending on the coupling strength FFNs may or may not be capable of propagating synchronous signals . But how do external oscillations influence the propagation of synchrony ? Do systems with and without dendritic nonlinearities exhibit qualitatively the same behavior ? To answer this question , we first consider isolated FFNs , which receive balanced oscillatory stimulation with frequencies equal to the propagation frequencies observed for the onset of propagation of synchrony in unstimulated FFNs . Thus we expect the stimulation to be in resonance with the propagating synchronous pulse in the FFN . The impact of different stimulation frequencies and the possibility of complex locking patterns between oscillations and propagating synchrony is investigated in the following Sections . How does the amplitude of the oscillatory input as controlled by the number of oscillating ( virtual ) neurons influences signal propagation ? For networks with additive couplings we find that the critical connection strength ( i . e . , the minimal connection strength which enables propagation of synchrony ) increases with increasing oscillation amplitude ( details on the setup of the oscillatory input are provided in the Methods Section ) as illustrated in Fig . 4a , c: The additional input is balanced , so that the mean input to each neuron is constant ( for all ) , but both the mean excitatory and inhibitory conductances are increased . In this high-conductance state the effective membrane time constant decreases and consequently the amplitude and the width of postsynaptic potentials decrease [51] , [52] . In other words , the additional inputs arising from oscillations decrease the excitability of the neurons . Thus , stronger inputs ( in terms of conductances ) are needed to generate the same depolarization as in networks without external oscillations and the critical connectivity , , increases . This is the same phenomenon that hinders synfire-explosions [26] , [53] in networks with conductance-based synapses as described in [27] . In contrast , in networks with non-additive couplings , the critical connection strength decreases with increasing oscillation amplitude ( see Fig . 4b , d ) . In such networks the propagation of synchrony is mainly mediated by dendritic spikes . Dendritic spikes are elicited if the excitatory input on a dendrite within a certain time-window , , is larger than the dendritic threshold . Inhibition fails to suppress dendritic spikes [46] and thus its increase does not hinder signal propagation . If the frequency of network oscillations is in the range of the natural propagation frequency , and the oscillations are in phase with the propagating signal , the synchronous pulse from the preceding group arrives at each layer synchronously with the oscillatory inputs . Thus , less input from the preceding layer is needed to reach the dendritic threshold . Taken together , by effectively lowering the dendritic threshold the external inputs reduce the critical connectivity . In Fig . 4b , d we show that this reduction can yield propagation of synchrony at drastically reduced synaptic efficiencies within the FFN; in the given example the critical connection strength is reduced by a factor of two to three ( from nS to nS ) . The downside of the robustness of dendritic spikes to inhibition is that even balanced oscillations may cause pathological activity if oscillation amplitude becomes too strong: With increasing amplitude the neurons of the FFN become more and more sensitive to inputs from the previous layer . Thus , similar to the regime of overly strong feed-forward connections , correlations in their spiking activity accumulate along the layers of the FFN [47]–[49] and induce spontaneous propagation of synchrony ( gray areas in Fig . 4 ) . Like balanced oscillations also unbalanced oscillations may be expected to alter the propagation efficiencies of FFNs: The average external excitatory input is larger or smaller than the inhibitory input , and thus the average ground state membrane potential of the neurons is shifted which influences the neurons' excitability . As we show below this shift clearly influences propagation of synchrony in additively coupled networks , but has only a weak effect in non-additively coupled systems . For a given excitatory coupling strength we denote the corresponding balanced inhibitory coupling strength by ( 3 ) where is chosen such that the peaks of the excitatory and inhibitory postsynaptic potentials are of equal amplitude when the input is received at resting potential ( cf . also Methods Section ) . We consider isolated FFNs stimulated by oscillations as in the previous section , but we decrease or increase the strength of the inhibitory inputs by a factor compared to the balanced regime , i . e . , ( 4 ) For additively coupled networks and such input indeed promotes synchrony propagation ( cf . Fig . 5a , red lines ) : The oscillatory input depolarizes the neurons of the FFN and thus less synaptic input is needed to elicit a somatic spike; the critical connectivity decreases . At the same time , the increased excitability of the neurons lowers the threshold for pathological activity , . Likewise , for the neurons are hyperpolarized by the oscillatory input which impedes the generation of somatic spikes; the critical connectivity increases ( cf . Fig . 5a , blue lines ) . In contrast , in non-additively coupled networks , the critical connectivity is largely unaffected by changing the balance of inhibition and excitation ( cf . Fig . 5b ) . Here , propagation of synchrony is mediated mainly by dendritic spikes , and their generation is not influenced by inhibition . Pathological activity is induced if correlations in spontaneous spiking activity accumulate over the layers . Because inhibition reduces the overall spiking activity ( and also the probability that a dendritic spike triggers a somatic one ) , with increasing ( and thus increasing inhibition ) the pathological threshold increases . We note that although unbalanced oscillations may promote propagation of synchrony in additively coupled networks , the mechanism underlying this support differs from propagation of synchrony in non-additively coupled networks . The effect is attributed to the increase of the ( average ) ground state membrane potential and , as we demonstrate below could as well be obtained by additional constant ( over time ) input currents with the same strength as the mean input due to the oscillations . Oscillations may support propagation of synchrony ( if in resonance ) , but how does their actual impact depends on system features such as frequency and amplitude of external oscillations ? In the following , we investigate which frequency ranges support or hinder synchrony propagation . In particular , we show that networks with non-additive coupling exhibit resonance to stimulations where the frequency is rationally related to the natural propagation frequency . In networks with additive couplings , we do not find such a resonance effect , even if the stimulation is unbalanced and therefore supports signal propagation . First , we consider networks with linear couplings: As pointed out in the previous section , balanced oscillatory inputs decrease the excitability of the neurons of the FFN . Thereby it decreases the capability of the network to propagate synchronous signals for all stimulation frequencies . With increasing , the total number of input spikes per unit time increases and the effective time constant decreases further such that the propagation becomes more and more difficult . Fig . 6a illustrates that the presence of balanced oscillations indeed inhibits synchrony propagation increasingly , the stronger and the more prominent the oscillations are ( i . e . , larger and ) . The support of signal transmission by unbalanced input ( cf . Fig . 5 ) is caused by an increase of the ground state's membrane potential . With increasing and this depolarization increases ( increased net excitation ) and thus facilitates synchrony propagation more and more . Likewise , the propagation frequency increases until the stimulation gets too strong and the system enters a pathological activity state . We do not observe resonance to the oscillatory stimulation , and the promotion of propagation of synchrony can equally well be obtained by an additional constant ( over time ) excitatory input which is proportional to the stimulation frequency ( cf . Fig . 6b ) . In contrast , networks with non-additive couplings show resonance , and even balanced oscillations enable propagation of synchrony for configurations where signal propagation fails for homogeneous external background ( i . e . , in the absence of external oscillations , cf . Fig . 4b , d ) . For stimulation frequencies , we observe a locking of the propagating signal to the external stimulus: The input from a preceding layer is not sufficient to excite sufficiently many neurons to spike synchronously and to enable persistent propagation . It can , however , take place if there is additional input . An oscillatory external input then influences the timing of the propagating pulse-packet and the propagation frequency locks to the stimulation frequency ( cf . Fig . 6c , d ) . With changing , we observe multiple resonance peaks for setups where the ratio of and is rational , for some small integers . The arrival of the input from every th external oscillation coincides with the arrival of the synchronous pulse from the preceding layer at every th group . Examples are shown in Fig . 7 for frequency ratios ( the propagation at every third layer is supported by the external input ) , ( the propagation at every second layer is supported by the external input from every third oscillation ) and ( every second oscillatory input coincides with the arrival of the synchronous pulse from the preceding layer ) . We remark that the sub-harmonic resonances are less prominent than the main resonance frequency , however , they can nonetheless enable oscillation-induced signal transmission even in systems where the oscillation frequency is small compared to the natural propagation frequency ( cf . for example Fig . 7a ) . Near the resonance frequencies the propagation frequency locks to the stimulation frequency ( cf . Fig . 6c gray areas ) . If the stimulation frequency increases above the resonance frequencies , synchrony propagation breaks down: Due to non-zero synaptic delay , initiation time of a dendritic spike and rise-time of the excitatory postsynaptic potential , there is a minimal time interval a signal needs to propagate from one layer to another . Thus , if the external stimulation frequency becomes too large , the inputs from the preceding layer arrive too late , i . e . , outside the dendritic integration window , and therefore the additional inputs do not support propagation of synchrony as described above . We only observe frequency lockings for small integers . The number counts the ( minimal ) number of layers a signal has to propagate in the absence of external simulations as the propagation of synchrony is supported by the oscillatory input only for every th layer . For large , however , the signal either propagates even in the absence of additional inputs ( i . e . , there is no need for supporting the signal propagation ) or it has decayed after layers and cannot be stabilized by external inputs . Large imply high stimulation frequencies , and with increasing stimulation frequency the external input becomes more and more stationary in the sense that additional ( oscillatory ) inputs are delivered to the neurons of the FFN all the time . A propagation of synchrony may be enabled , but the signal does not lock to the stimulation frequency anymore ( cf . Fig . 6c ) . Above we demonstrated how oscillations can support signal transmission in FFNs with homogenous delays . As shown in the subsequent sections , we observe the same resonance phenomena equally prominent in FFNs with distributed delays , even if the delay distributions are broad . We also remark that we can describe the emergence of oscillation supported propagation of synchrony using methods introduced in [30] , [31] . In Supporting Material S1 Text we provide a simplified , analytically tractable model by describing the dynamics in terms of probabilistic threshold units . In particular , we derive an analytical expression for the minimal amplitude of the oscillatory input , , for which robust signal propagation is possible and compare the analytical predictions with numerical simulations ( cf . Supporting Material S2 Text ) . Networks with non-additive coupling exhibit resonance to oscillatory signals and this provides the possibility of specifically activating FFNs with different resonance frequencies . As we demonstrate below such resonant signal transmission establishes a mechanism to read out signals encoded in the structure of a recurrent network . In how far do the results for pure feed-forward structures without recurrent connectivity can be generalized to recurrent systems as relevant for biological neural circuits ? The main difference between isolated FFNs and recurrent FFNs is the emergence of a projection of the synchronous activity to all neurons of the network , not only to the neurons of the layer following the currently active one . For additively coupled networks this projection ( similar to balanced oscillatory input ) shifts the range of coupling strengths ( 5 ) for which a persistent propagation of synchrony is possible to larger connection strengths . The length of the interval , however , is unchanged ( for details see Supporting Material Text S2 ) . For non-additively coupled networks , the critical connectivity is largely unaffected , but with more and more prominent recurrent connections the pathological threshold decreases . For moderate recurrent connection strengths propagation of synchrony can be induced by oscillations also in recurrent networks without causing pathological activity; though if the connections are too large activity might spread not only from one layer to the next , but might propagate over the whole network ( ‘synfire explosion’ activity , [25] , [26] , [53] ) . We investigate and discuss such recurrent systems in detail in Supporting Material S2 Text . The main resonance frequency in non-additively coupled FFNs is given by the natural propagation frequency . This frequency , however , is determined by the average time an arriving synchronous input at a given layer needs to trigger a somatic spike and the average synaptic delay , ( 6 ) We illustrate this dependency in Fig . 8a indicating the resonance peaks for different . Here , the coupling delays between neurons of successive layers are drawn uniformly from an interval of length centered at , ( 7 ) With increasing , the natural propagation frequency and thus the resonance peaks are shifted to smaller frequencies . The width of the resonance peak is determined by the temporal spread of the propagating synchronous pulse itself , the temporal spread of the oscillatory inputs ( ; cf . also Supporting Material S1 Text ) and the width of the dendritic integration window . In particular , the width of the resonance peaks increases with increasing as shown in Fig . 8a . The existence of separated resonance peaks provides the possibility to specifically activate different signal transmission routes by oscillations of suitable frequencies . As a simple example consider a recurrent network containing two FFNs ( cf . Fig . 8b , c ) . The coupling delays between neurons of successive layers of the first FFN are centered at ms , the coupling delays between neurons of successive layers of the second FFN are centered at . As before , the feed-forward couplings in both FFNs are too weak to enable a robust propagation of synchrony in the absence of external oscillations ( cf . Fig . 8b ) . Yet , external oscillations fitting to the resonance frequencies of the FFNs may enable robust propagation in one of the FFNs without activating the other . The close-up view in Fig . 8c shows that indeed the propagation in both FFNs occur with different propagating frequencies . So far we considered networks with homogeneous or narrow delay distribution . However , heterogeneous delays provide a desynchronizing force to propagating synchronous signals . Here , we investigate the robustness of oscillation-induced signal propagation with respect to heterogeneous coupling delays . We show that even in networks with broad delay distributions external oscillations support signal propagation . Starting with a homogenous delay distribution , i . e . , all delays , we study broadened ones in the following . More precisely , we draw the the conduction delays from a log-normal distribution , i . e . , the probability density function is given by [54] ( 8 ) where and are the parameters of the probability distribution . To keep the resonance frequencies comparable , we keep the mode , i . e . , the maximum of the probability distribution ( 8 ) , constant with increasing distribution width parameter . For given and , the parameter of the probability distribution ( 8 ) is given by [54] ( 9 ) and thus Equation ( 8 ) reads ( 10 ) The standard deviation of the distribution is given by ( 11 ) In Fig . 9a we show the log-normal distribution for fixed and different and . We consider signal propagation in FFNs in the presence of balanced oscillations . Synchronous pulses may propagate along the layers , if the summed input from the external oscillation and the previous layer is strong enough to excite sufficiently many neurons to spike . However , the dendritic integration window is small , typically in the order of a few milliseconds . Only inputs arriving simultaneously within this time interval can contribute to the generation of dendritic spikes , and thus may elicit subsequent somatic spikes . By increasing the width of the delay distribution , the arrival times of the inputs from the previous layer become more and more distributed . Consequently , the number of spikes arriving simultaneously with the external spikes , i . e . , within a time interval centered at the expected arrival times of the external synchronous pulses , decreases — thus , the effective number of inputs decreases ( cf . also Supporting Material Text S1 ) . However , this decrease might be compensated by , e . g . , larger layer sizes . As an example , we illustrate in Fig . 9B that an FFN with a layer size of neurons ( green line ) can tolerate heterogeneous delay distributions with a standard deviation up to ms ( same network setup as in Fig . 4b , d and 6c , d ) . In a similar network with increased layer sizes an oscillation-induced propagation of synchrony is possible for substantially broader delay distributions ( e . g . , for up to ms ) . Oscillation-induced signal transmission can take place if the total expected input within the relevant time window is sufficiently large . Therefore both the width of the delay distribution and the layer size influence the width of the resonance peaks . With increasing the arrival times of the spikes from the previous layer become more distributed , and the total number of spikes within a time interval decreases . We illustrate the effect of an increasing width of delay distribution in Fig . 10a: Starting with a setup where a synchronous signal can propagate over all layers for homogeneous coupling even in the absence of external oscillations , an increase of the width of the delay distribution results in the formation of resonance peaks . The arriving inputs become more and more distributed and therefore signal propagation is only possible if the input from the previous layer is supported by external oscillations . If the delay distribution becomes broader , the frequency bands which enable oscillation-induced signal transmission become narrower , and eventually for sufficiently large a robust signal transmission is not possible anymore . Similarly , for a given width of the delay distribution an increase of the layer size may enable oscillation-induced signal transmission and cause the formation of resonance peaks ( cf . Fig . 10b ) . With increasing the total number of potential inputs from the previous layer ( and thus also the number of potential inputs within the relevant time window of length ) increases . If this number becomes sufficiently large , robust propagation of synchronous pulses is enabled . We conclude that oscillation-induced signal propagation in FFNs is possible even if the delay distribution is broad , and that heterogeneities in the delays can be compensated by increased layer sizes . We remark that heterogeneous weights ( in contrast to heterogeneous delays ) do not constitute a desynchronizing force in networks with nonlinear dendritic interactions: The spike latency ( and thus the propagation frequency ) is only weakly affected by the coupling strength ( cf . Fig . 3 ) . Thus , if the input is sufficient to elicit dendritic spikes , the timing of the consecutive somatic spike ( if triggered ) does not depend on the input strength from the previous layer or the external input - only the timing of presynaptic inputs is important . We have demonstrated that oscillation-induced signal transmission is present in systems with heterogeneous coupling delays . In the previous section we studied the influence of inhomogeneities in a rather general setting . In this section we consider a specific example: We employ a delay distribution as expected for subnetworks in the hippocampus . In this area , spike patterns generated during exploration are replayed during sleep [14] , [15] , [55] , [56] , accompanied by high-frequency network oscillations [57]–[59] . The replay has been hypothesized to be realized by local feed-forward structures [13] , [14] , [16] , [60] , possible supported by dendritic sodium spikes [30] , [31] , [41] which have been prominently found in the hippocampus [29] , [43] , [46] , [61] . In the following we show that oscillations in hippocampal-like network structures indeed support signal transmission . Importantly , the expected resonance frequencies quantitatively agree with the oscillation frequencies observed in neurophysiological experiments . We assume that the delays are a function of the distance between the presynaptic and the postsynaptic neuron . Further , we take variations of the dendritic conduction time into account . In general , the total conduction delay can be decomposed into two contributions , ( 12 ) ( i ) the axonal delay , i . e . , the time interval between the presynaptic spike and the onset of the synaptic transmission , and ( ii ) the dendritic delay , i . e . , the time delay between the onset of the synaptic transmission and the onset of the postsynaptic ( somatic ) response . The axonal conduction delays are proportional to the distance between the presynaptic and postsynaptic neuron . For simplicity , we assume that the neurons are distributed uniformly on a quadratic patch with side length . Thus , ( 13 ) where is the axonal conduction velocity and and are the absolut distances between two neurons in the horizontal and vertical direction with the probability density function ( 14 ) for . The dendritic conduction delays are drawn uniformly from the interval ( 15 ) and account for the variability in the distance between the synaptic contact sites and the soma . As an example , we consider the recurrent connections in the hippocampal region CA1 . Here , the range of local axonal interconnections is estimated to be in the order of m [62] , [63]; in some direction connections extending over m or more are found [62]–[65] . The axonal conduction delay in the hippocampus is measured to be in the range of mms [66] , [67] , for numerical simulation we assume m/ms in the middle of the biologically plausible parameter range . Further , we assume the variation in the dendritic conduction delays to be in the interval in agreement with experimental data [68]–[70] . In Fig . 11a we show the resulting probability density functions for different patch sizes . With increasing side length the probability distributions become broader and the peak of the distribution is shifted to larger delays . As shown in Fig . 11b synchronous pulses may propagate in FFNs in the presence of external oscillations . We observe resonances as before ( cf . Fig . 6 ) , and the resonance frequencies are shifted to smaller frequencies with increasing patch size . Interestingly , the oscillation frequencies accompanying replay in the hippocampus are larger in area CA1 than in the more globally connected region CA3 [57] , [71] , [72] . We hypothesize that the existence of long range connection in CA3 ( and therefore an effectively increased patch size ) cause lower resonance frequencies . The observed oscillations during replay might therefore be optimized for the specific region and support the replay of spike patterns encoded in weak FFNs . The oscillations themselves might be generated by global network oscillations based on dendritic spikes [50] , by highly connected nodes ( so-called hub-neurons ) which are a prominent feature of networks with broad degree distribution [41] , by interneuron network oscillations [57] , [73] , [74] , or by avalanches of spikes propagating in a network of axons coupled by axo-axonal gap junctions [75]–[77] .
Reliable and controlled transmission of signals is considered essential for computation in cortical networks . Propagation of synchronous activity along layered feed-forward networks may be one important way to realize such transmission [19] , [21] , [23] . Starting with a random recurrent network , feed-forward structures are assumed to be formed in a “training phase” previous to the recall of the learned sequences by , e . g . , spike time dependent plasticity [78]–[80] . Moreover , propagating synchronous pulses are a candidate for generating precisely timed spike patterns in the millisecond range as observed in various neurophysiological studies ( e . g . , [81]–[84] ) . Robust propagation , however , typically requires a highly prominent feed-forward anatomy , either in the sense of densely connected layers of neurons [25]–[27] or strongly increased connection strengths between neurons of successive layers ( compared to remaining connections of the network ) [28] . Such prominent structures are experimentally not observed . In previous articles we have shown that fast dendritic spikes can support signal transmission in the form of propagation of synchrony [30] , [31] . They specifically amplify activity that is synchronous , and thus enable a robust propagation in networks with moderate feed-forward anatomy . In this article we demonstrated that the presence of background oscillations can relax this requirement even further by supporting the propagating signal by additional external inputs . These additional inputs excite the neurons of the network ( including the current target layer of the propagating synchronous pulse ) and therefore enable a robust propagation with less inputs from the preceding layer . As a consequence robust signal transmission may emerge in networks with weaker couplings between the layers of the feed-forward network . Such weaker structures , where the differences between feed-forward connections and remaining recurrent couplings are smaller , can be formed faster by synaptic plasticity ( assuming a constant plasticity rate ) , i . e . , the process of creating ( and reconfiguring ) information pathways is simplified . Alternatively , the background oscillations can enable robust signal transmission in feed-forward networks with reduced layer size ( while keeping the coupling strengths fixed ) . We may expect that this leads to an increase in the storage capacity of recurrent networks , because a reduced number of “memory-encoding” neurons is required for reliable signal propagation . We remark that the mechanism of oscillation-induced signal transmission is related to the idea of “communication through coherence” [33] , where the information flow between neural groups is enabled by coherent rhythmic modulation in the neural excitability in the different sub-networks . Similarly , in our approach the oscillatory input excites the neurons ( and , even more importantly , the non-linear dendrites of the neurons ) of the local network , and therefore acts as a “clock” enabling the successful propagation of synchronous pulses in the local network . Experimental data suggests that there is a balance between excitatory and inhibitory input to single neurons in cortical networks during spontaneous and sensory-evoked activity [85]–[87] . We therefore considered external oscillatory input which is composed of an excitatory as well as an inhibitory component . We find that for additively coupled networks , only unbalanced external inputs that cause a net depolarization , support propagation of synchrony . Further , this support does not depend on the oscillatory nature of the input and could equally well be established by a temporally constant input current with the strength of the temporal mean input . In contrast , for networks with non-additive couplings the ratio of the excitatory and inhibitory input is less important . In these networks propagation of synchrony is mainly mediated by dendritic spikes , which are elicited if the excitatory input within a short time interval exceeds the dendritic threshold [29] , [42]–[44] . Further , inhibition fails to suppress the generation of such dendritic spikes [46] and thus even inputs with a net hyperpolarizating effect support signal propagation . Due to the short dendritic integration window the timing of the external input is important , and thus only oscillatory inputs of a suitable frequency range can facilitate the propagation of synchrony . Whenever the ratio of the stimulating frequency and the “natural” propagation frequency of the feed-forward network is rational , resonances and locking patterns emerge . The resonance frequencies themselves are determined by the average conduction delays between the neurons of the FFN . This provides a mechanism to selectively activate different signaling pathways by oscillations of suitable frequency . If either the synaptic couplings or the oscillatory inputs are too strong , synchronous activity may spread over the entire network , generating a large scale synchronous population burst and a subsequent phase of refractoriness . The occurrence of such pathological activity states which transiently silences the network can terminate the induced propagating signal and therefore hinder signal transmission . These observations may be relevant for understanding the neurological implications of epileptic-like seizures . For clarity of presentation , we first demonstrated the effect of oscillation-induced propagation of synchrony for FFNs with homogenous or relatively narrow delay distributions . In biological neural circuits , the distribution of delays might be substantially broader . One might expect that this may blur out signals and hinder their reliable transmission . However , in networks with nonlinear dendrites , for the generation of dendritic spikes ( and consecutive somatic spikes ) inputs from both , the previous layer and the oscillatory network are needed . Therefore , broad delay distributions only decrease the “effective” layer size , i . e . the fraction of inputs from the previous layer which can arrive within the relevant time interval to support spike generation . As a consequence FFNs with broad delay distribution require a moderately increased layer size , but the general mechanism of oscillation-induced signal transmission is unchanged . In this article we considered oscillatory input arriving from an external source . For clarity , we separated the local ( signal processing ) network and the oscillation-generating network to study the impact of oscillations . We note that there are no conceptual differences if we consider networks , in which such oscillations arise from the embedding network itself . For example , we have recently shown that in networks with a broad distribution of synaptic connections moderate network oscillations which are suited to support signal transmission naturally emerge [41]: So-called hub-neurons ( higher than average connected neurons ) can echo the propagating synchronous signal , start to oscillate and therefore provide an oscillatory , supporting feedback . As another example intrinsic network oscillations can emerge due to recurrent inhibition or the excitatory-inhibitory loop [88] , [89] . Oscillation-supported signal transmission might also arise from network intrinsic responsivity modulations such as sub-threshold membrane potential oscillations in resonator-type neurons [90] , if they are synchronized and sufficiently strong to depolarize the dendritic compartments in a rhythmic way . Furthermore , network level resonances [91] may support propagation of synchrony . Dendritic spikes are prominently found in , e . g . , the hippocampus ( cf . [29] , [43] , [46] , [61] and others ) . In this cortical area spike patterns observed during spatial exploration are replayed during sleep or resting phases ( e . g . , [14] , [15] , [55] , [56] ) . Interestingly , this replay is accompanied by high-frequency oscillations in the range of up to Hz [57] , [58] , [71] . We estimate the distribution of conductance delays for recurrent connections in the hippocampal areas CA1/CA3 , and show that the expected resonance frequencies for the support of synchrony propagation agree quantitatively with the frequencies observed in neurophysiological experiments . This suggests that the high-frequency oscillations may contribute to the stabilization of the replay of spike patterns in the hippocampus . Our choice of parameters , including that of ( average ) conduction delays , is guided by neurophysiological observations in the hippocampus . However , in other cortical systems substantially larger delays have been reported ( see , e . g . , [92] for an overview ) . Because the natural propagation frequency decreases with increasing conduction delays , this suggests that the mechanism of oscillation-induced signal transmission is not restricted to high-frequency oscillations as present in the hippocampus . Furthermore , oscillations can stabilize signal transmission for stimulation frequencies where the ratio of natural propagation frequency and stimulation frequency is rational . Therefore oscillation-induced signal transmission can be enabled by stimulation with frequencies substantially smaller than the natural propagation frequency . For example , only every second or third synchronous pulse might be supported by the oscillatory input ( cf . Fig . 7 ) . The widths of these sub-harmonic resonances are smaller than the main resonance peak ( around ) , however , we have shown that they can enable oscillation-induced signal transmission even if the oscillation frequencies are small compared to the natural propagation frequency . Finally , we emphasize that the occurrence of the identified mechanism of signal transmission by oscillation-induced propagation of synchrony need not be restricted to information processing in neural networks . In Supporting Material S1 Text , we derive a simplified , analytically tractable model describing the network activity in terms of probabilistic threshold units . Its analysis reveals that the main prerequisite for oscillation-induced signal transmission is the threshold-like processing of inputs of the single elements in the network . We may therefore expect that the mechanism also plays a role in other networks of sharply nonlinear threshold units . Networks of such units describe a variety of real-world phenomena , like the transmission of rate activities in neural networks ( McCullogh-Pitts model , e . g . , [59] , [93] ) , ( failure ) cascades in social , supply or communication networks ( e . g . , [94] , [95] ) , or signaling in gene and protein networks ( threshold Boolean networks , e . g . , [96] .
We consider networks of neurons of the integrate-and-fire type [97] . Single neurons interact by sending and receiving action potentials ( spikes ) . The state of neuron is described by its membrane potential and its temporal dynamics are determined by ( 16 ) where is the membrane capacity , is the leak conductance and is the equilibrium potential . and are currents arising from excitatory and inhibitory inputs , respectively . Whenever the membrane potential exceeds the spiking threshold at some time , a spike is sent to the post-synaptic neurons , where it arrives after a delay time . The sending neuron's potential is reset to , and the neuron is refractory for a time period , i . e . , for . Simulation results were obtained using the simulation software NEST [98] . The effects of the synaptic inputs on postsynaptic neurons are modeled by transient conductance changes . Denoting the reversal potentials of excitatory and inhibitory currents by and , the input currents to neuron arising from synaptic inputs from other neurons of the network are given by ( 17 ) ( 18 ) and are linear superpositions of single responses , ( 19 ) ( 20 ) where and denote the excitatory and inhibitory coupling strength from neuron to neuron and is the th spiking time of neuron . and specify the time course of the synaptic conductance change given by the difference of two exponentials [97] with time constants and , ( 21 ) for describing the effect of an excitatory and inhibitory input , respectively , that is received at time . The normalization constant ( 22 ) is chosen such that the peak conductance . Throughout this article , we denote the strength of a synaptic connection by the value of the peak conductance , i . e . , a single input of strength causes a conductance change . We note that , to keep the model as simple as possible , we did not incorporate any saturation in the linear model . This is in contrast to the model with nonlinear dendrites ( see below ) , since a dendrite becomes refractory after generation of a dendritic spike . Besides linear summation of inputs ( as described above ) , we consider nonlinear amplification of synchronous inputs mediated by fast dendritic spikes . These have been found in single neuron experiments ( e . g . , [29] , [42]–[44] ) and introduced in recent models of neural networks [30] , [41] , [45] , [50] , [99] . The amplification is based upon dendritic action potentials which generate a strong depolarization in the soma . Here , three properties are of particular interest: ( i ) The amplification is very sensitive to input synchrony ( relevant time window milliseconds ) , ( ii ) the peak of the depolarization in the postsynaptic neuron ( pEPSP ) is reached a certain time interval after stimulation with only sub-millisecond jitter and ( iii ) with increasing stimulation strength the amplitude of the pEPSP saturates . We model the contribution of such dendritic spikes to the neuronal input as follows ( see also [30] , [50] ) : We augment the neurons with an additional nonlinear dendrite . Inputs that arrive at the linear dendrite are processed as described above . Inputs on the nonlinear dendrite also cause a conductance change as described above , but additional depolarizations of the membrane potential mimicking the effect of a dendritic spike may be generated . If the total excitatory input to a nonlinear dendrite within a time interval exceeds a certain threshold , a current pulse is initiated which takes effect on the membrane potential after a delay time . To account for the experimentally observed saturation of the somatic depolarization caused by dendritic spikes we limit the maximal conductance change within a time interval to and model the current pulse in a phenomenological approach such that the depolarization caused by a suprathreshold input , , resembles the characteristics and time course of the depolarization observed in single neuron experiments ( cf . [29] ) . More precisely , the current pulse is described by the sum of three exponential functions , ( 23 ) with positive prefactors , , and decay time constants , and which are chosen such that the somatic depolarization fits experimental data . After initiation of such a current pulse the ( nonlinear ) dendrite becomes refractory for a time period and does not transmit spikes within the refractory time period . This refractoriness yields the experimentally observed saturation for inputs exceeding the dendritic threshold . We note that for the generation of a dendritic spike only the excitatory inputs are considered . Consequently , in accordance with recent experimental findings , inhibition fails to suppress fast dendritic sodium spikes . However , the probability that a somatic spike is initiated by a dendritic one might be reduced by hyperpolarization of the soma [46] ( cf . also [41] ) . We investigate sparsely , randomly connected recurrent networks and study the propagation of synchrony in naturally occurring feed-forward subnetworks ( FFNs ) . “Naturally occurring” here means that the feed-forward structures are present as part of a recurrent network and are not generated by , e . g . , adding feed-forward connections . However , they are highlighted by moderately increased excitatory connections . We denote the total number of neurons in the recurrent network by . The network itself constitutes an Erdös-Rényi random graph: A directed excitatory synaptic connection between any pair of neurons exists with probability . Inhibition in recurrent networks is usually assumed to be mediated by a population of inhibitory neurons ( interneurons ) . Spiking of excitatory neurons causes a response of inhibitory neurons which in turn project an inhibitory input to the excitatory neurons . Here , we simplify this inhibitory feed-back mechanism and assume that the spiking of neurons , additionally to the excitatory input on the postsynaptic neurons , have an inhibitory effect: An inhibitory connection between any pair of neurons exists with probability . We remark that there might exist an inhibitory and excitatory connection between two neurons . However , these cases are rare due to the sparsity of the considered networks ( typically ) . The simplification of the inhibitory feed-back loop eases the analytical treatment , but is not crucial for the effect of oscillation induced propagation of synchrony as discussed later on ( cf . also [41] ) . For clarity of presentation coupling strengths are assumed homogeneous; excitatory connections have strength , the strength of inhibitory connections is denoted by . We choose the ratio between inhibitory and excitatory connection strengths , , such that the peaks of single excitatory and inhibitory postsynaptic potentials measured at resting membrane potential are of equal amplitude . We define FFNs by assigning neurons randomly to groups of neurons each , where each neuron belongs to one group at most . These groups constitute the layers of the FFN . By construction , the connectivity between neurons of successive groups of the FFN statistically equals the overall connectivity . To enable propagation of synchrony , we increase the strengths of the already existing excitatory connections between neurons of successive layers; this connection strength is denoted by . For clarity of presentation , in the first part of the article we investigate the influence of oscillations on propagating synchrony in isolated FFNs . Here , only excitatory connections between neurons of successive layers are present , i . e . , , but . However , recurrent connections ( ) do not change the results qualitatively . We comprehensively study recurrent FFNs and discuss differences to isolated FFNs in Supporting Material S2 Text . We evaluate up to which layer a synchronous pulse propagates in the FFN by considering the signal-to-noise ratio ( SNR ) : After a synchronous pulse is initiated in the first layer ( ) at time , we determine for the following layers , , how many neurons have spiked within a time window of length lagging behind the synchronous pulse in the previous layer ( centered at ) by a temporal shift . The temporal shift is chosen after simulation such that the number of spikes ( 24 ) becomes maximal . Here are the indices of neurons of group , is the th firing time of neuron , and denotes the characteristic function . Starting with , the following are determined by first evaluating according to Equation ( 24 ) , and then defining as the mean of all spikes contained in the interval . Further , we determine the noise level in each layer by measuring the probability to find spikes from neurons of group in a time window during a control time interval in which no synchronous activity is induced ( an external oscillatory input is present , if applicable ) . The noise level is then given by the minimal value satisfying ( 25 ) with constant . Finally , we denote the propagation up to the th layer as successful if the SNR is larger than a constant , ( 26 ) This means , in particular , that we can distinguish the background ( spontaneous ) activity from the transmitted signal in all layers . In the ground state of balanced networks [2] , [3] single neurons fire irregularly and their spiking activity is approximately described by Poissonian spike trains [4] , [100] , [101] . In addition to inputs from the recurrent network each neuron receives inputs from remote networks , and we emulate this influence by independent excitatory and inhibitory ( Poissonian ) spike trains . We denote the rates by and and the strength of single spikes ( peak conductances ) by and , respectively . Similarly to the recurrent connections , we assume the external input to be balanced , such that the total input is balanced as well . As a consequence , the neurons are in a fluctuation-driven regime , and in the absence of synchrony the neurons spike asynchronously and irregularly and their output spike trains resemble Poissonian spike trains themselves . In this article we study the impact of neuronal oscillations on the ability of recurrent networks to propagate synchronous signals . Oscillatory input may arise from oscillations in other circuits or within the local network itself . To systematically investigate the influence of oscillations on synchrony propagation in a controlled way , we emulate such oscillations by excitatory and inhibitory inputs generated by a ‘virtual’ population of neurons that spike with a mean frequency . Within each oscillation period , spike times are drawn from a Gaussian distribution centered at ( for the th oscillation , ) with standard deviation . Each of these spikes causes an excitatory input of strength with probability and an inhibitory input of strength with probability to each neuron of the recurrent network ( cf . Fig . 12 ) . Here and in the following the term “balanced oscillations” refers to oscillatory input for which excitatory inputs and inhibitory inputs cause postsynaptic potentials of equal amplitude if the average excitatory inputs exceed the inhibitory inputs or vice versa , we denote such inputs as “unbalanced oscillations” . Whereas unbalanced oscillations induce a net depolarization or hyperpolarization of the neurons in the network , balanced oscillations maintain the balance between excitation and inhibition , and are thus expected to change the average membrane potential in the ground state only weakly . However , they may influence the effective time constant of the neurons as discussed in the Results Section ( cf . also [51] , [52] ) . The aim of the article is to understand the influence of the oscillatory nature of the input on propagating synchrony , and resonances between signal propagation and input oscillations . We discuss balanced oscillations and unbalanced oscillations separately . Throughout the article ( for simplicity ) we consider a homogeneous neuron population . The single neuron parameters are pF , mV , mV , nS , mV and ms [72] , [102] for all . The time constants of the excitatory conductances ( AMPA ) are ms and ms [103] , [104] . For simplicity we assume the same time constants for inhibitory conductances ( GABAA ) , ms and ms . The reversal potentials are mV and mV [72] , [97] . To obtain balanced recurrent ( and external oscillatory ) inputs , the ratio between excitatory and inhibitory couplings is chosen such that the peaks of single excitatory and inhibitory postsynaptic potentials equal each other when the inputs are received at resting membrane potential , i . e . , ( 27 ) for standard neuron parameters . We consider sparsely connected networks ( standard connection probability ) with homogenous coupling delays ms in the first part of the article , and with heterogeneous coupling delay distribution in the second part . For the latter , the underlying distribution and parameters are stated in the corresponding sections . Each neuron receives excitatory and inhibitory Poissonian spike trains with rates kHz . Single inputs have strength nS and nS , respectively . The parameters of the dendritic spike current are chosen according to single neuron experiments [29] , [42]–[44] , nS , nA , nA , nA , ms , ms , ms and ms ( cf . also , [30] , [50] ) . The standard value for the length of the dendritic integration window is ms; in the last part of the article it is varied as indicated . For the detection of propagating synchronous signals , we considered time windows of length ms , and considered time lags between successive synchronous pulses up to ms . The noise level is determined during an observation interval ms , we further set the constant for defining the chance level to and require a minimal SNR of . | Rhythmic activity in the brain is ubiquitous , its functions are debated . Here we show that it may contribute to the reliable transmission of information within brain areas . We find that its effect is particularly strong if we take nonlinear coupling into account . This experimentally found neuronal property implies that inputs which arrive nearly simultaneously can have a much stronger impact than expected from the sum of their individuals strengths . In such systems , rhythmic activity supports information transmission even if its positive and negative part exactly cancels all the time . Further , the information transmission can adapt to the oscillation frequency to optimally benefit from it . Finally , we show that rhythms with different frequencies may enable or disable communication channels , and are thus suitable for the steering of information flow . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"computational",
"neuroscience",
"neuroscience",
"biology",
"and",
"life",
"sciences",
"computational",
"biology"
] | 2014 | Oscillation-Induced Signal Transmission and Gating in Neural Circuits |
Endothelial cell ( EC ) plasticity in pathological settings has recently been recognized as a driver of disease progression . Endothelial-to-mesenchymal transition ( EndMT ) , in which ECs acquire mesenchymal properties , has been described for a wide range of pathologies , including cancer . However , the mechanism regulating EndMT in the tumor microenvironment and the contribution of EndMT in tumor progression are not fully understood . Here , we found that combined knockdown of two ETS family transcription factors , ERG and FLI1 , induces EndMT coupled with dynamic epigenetic changes in ECs . Genome-wide analyses revealed that ERG and FLI1 are critical transcriptional activators for EC-specific genes , among which microRNA-126 partially contributes to blocking the induction of EndMT . Moreover , we demonstrated that ERG and FLI1 expression is downregulated in ECs within tumors by soluble factors enriched in the tumor microenvironment . These data provide new insight into the mechanism of EndMT , functions of ERG and FLI1 in ECs , and EC behavior in pathological conditions .
Reciprocal interactions between tumor cells and stromal cells have a profound impact on tumor progression . Among tumor stromal cells , endothelial cells ( ECs ) and blood vessels are important components , as they supply nutrients and oxygen , and act as an entrance into systemic circulation , leading to metastatic organs . However , the behavior of ECs within tumors is not fully understood . Endothelial-to-mesenchymal transition ( EndMT ) is a phenotypic conversion process in which ECs lose their specific characteristics and obtain mesenchymal properties . Whereas EndMT is physiologically induced during embryonic heart development [1 , 2] , it is also pathologically induced in a wide range of diseases associated with fibrosis and vasculopathy [3 , 4] . In cancer pathology , a landmark study by Zeisberg et al . showed that EndMT is a potential source of cancer-associated fibroblasts ( CAFs ) , which are well-known tumor stromal cells typically recognized to have a pro-tumor role [5 , 6] . However , the mechanism of EndMT induction in the tumor microenvironment and the impact of EndMT on tumor progression remain unclear . Our group previously reported that knockdown of GATA2 , a transcription factor ( TF ) essential for EC differentiation and function , induces EndMT-like conversion in ECs , suggesting that dysregulation of EC-related TFs can trigger EndMT [7] . Interestingly , ECs have no single master regulator , and this role appears to be shared by a variety of EC-related TFs , including the ETS , GATA , SOX , and FOX families [8] . A strong candidate for a pioneer TF in ECs , which appears at the earliest stage of cell fate determination and binds closed chromatin to form lineage-specific epigenetic conditions , is ETV2 , a member of the ETS family . However , ETV2 is expressed transiently during EC differentiation and is not detected in mature ECs [9 , 10] . Thus , ETS factors specifically expressed in mature ECs , ERG and FLI1 , may be especially important among EC-related TFs . This is supported by recent data showing that constitutive expression of ERG and FLI1 with transient expression of ETV2 directly reprograms amniotic cells to mature ECs [11] . The ETS family is a member of TFs with a well-conserved DNA binding domain named the ETS domain , which typically recognizes the consensus sequence 5′-GGAA-3′ [12] . Among them , ETS-related gene ( ERG ) and Friend leukemia integration 1 ( FLI1 ) are specifically and highly expressed in ECs [13] . Notably , ERG and FLI1 show ~70% overall amino acid sequence similarity and only 2 mismatches within ~80 amino-acid stretch in the ETS domain . ERG- and FLI1-knockout mice commonly show embryonic lethality at E10 . 5–11 . 5 and severe hemorrhage due to defective angiogenesis [14–17] . In addition , double knockdown of erg and fli1 in zebrafish showed more severe vascular defects compared to individual knockdown , indicating that these TFs have synergistic roles [18] . In support of this , several groups demonstrated that ERG promotes the expression of some EC-specific genes such as CDH5 , HDAC6 , CLDN5 , ENG , VWF , and RHOJ to maintain EC function , while ENG is also under the control of FLI1 [19–25] . Additionally , a recent paper reported that ERG controls the TGFβ/SMAD signaling pathway to protect ECs from EndMT [26] . Although the emerging roles of ERG and FLI1 have been recognized in ECs , the functions of these TFs have not been thoroughly analyzed using genome-wide approaches . In this study , we conducted a comprehensive microarray and chromatin immunoprecipitation-sequencing ( ChIP-seq ) analysis to characterize the ERG- and FLI1-mediated transcriptional regulation in ECs . Our results indicate that combined downregulation of ERG and FLI1 expression leads to EndMT associated with dynamic changes in transcriptome and epigenome . We identified microRNA-126 , which is specifically expressed in ECs , as the key downstream target of ERG and FLI1 to regulate EndMT . Furthermore , we also show that ERG and FLI1 expression is downregulated in tumor tissues by soluble factors . These findings might provide new insight into EC phenotypic changes mediated by the loss of ERG/FLI1 in the pathological environment .
Considering previous findings that constitutive ERG and FLI1 expression with transient ETV2 expression directly reprograms somatic cells into ECs , we assessed how the ablation of ERG and/or FLI1 affects mature EC phenotype [11] . We knocked down these TFs either alone or in combination using two independent sets of siRNAs in primary cultured human umbilical vein ECs ( HUVECs ) ( Fig 1A and 1B and S1A Fig ) . Consistent with the previous report that depletion of ERG expression leads to EndMT [26] , knockdown of ERG alone upregulated a mesenchymal marker ( TAGLN ) and an epithelial-to-mesenchymal transition ( EMT ) /EndMT driver gene ( SNAI2 ) . In contrast , knockdown of FLI1 alone did not upregulate these EndMT marker expression . Interestingly , combined knockdown of ERG and FLI1 using 2 sets of siRNAs effectively led to decreased expression of endothelial markers ( CDH5 , PECAM1 ) and increased expression of EndMT markers ( ACTA2 , TAGLN , COL1A1 , and SNAI2 ) consistently ( Fig 1C and S1B Fig ) . EndMT-like conversion was also observed at the protein level , with an indication that the conversion process requires 7 days of culture to be completed ( Fig 1D ) . Along with marker expression changes , HUVECs morphologically changed from an EC-specific cobblestone-like shape into a mesenchymal-like spindle shape ( S1C Fig ) . Additionally , combined knockdown of ERG and FLI1 led to a defect in EC function as indicated by the loss of tube formation ability ( S1D Fig ) . Given that reduced expression of ERG triggers apoptosis [20] , we evaluated the apoptosis level by cleaved caspase-3 immuno-detection after siRNA treatment targeting ERG and FLI1 individually and together . Combined treatment of siERG and siFLI1 as well as siERG treatment alone significantly induced caspase-3 cleavage ( S1E Fig ) . However , cleaved caspase-3 comprised a minor fraction , indicating that apoptotic cell death does not mainly affect the cell phenotype as a whole in our experiments . To confirm the induction of EndMT in a genome-wide manner , transcriptomic changes were analyzed in HUVECs treated with siERG , siFLI1 , or both , for 3 or 7 days using a gene expression microarray ( Fig 1E ) . We identified 1 , 190 differentially expressed genes , which were classified into 3 clusters: cluster 1 , genes driven by ERG and/or FLI1; cluster 2 , genes repressed by ERG alone; and cluster 3 , genes synergistically repressed by ERG and FLI1 . Cluster 3 was further classified into two sub-clusters: cluster 3–1 , genes upregulated 3 days after siRNA treatment; and cluster 3–2 , genes upregulated 7 days after siRNA treatment . Cluster 1 includes EC specific genes coding for ROBO4 , vWF , Apelin , SOX18 , and Claudin-5 ( Fig 1E ) . Gene ontology ( GO ) analysis confirmed that this cluster is highly related to vascular functions including ‘vascular development’ and ‘angiogenesis’ ( Fig 1F ) . Importantly , cluster 1 genes were reduced further by siERG than by siFLI1 , but significantly further reduced by siERG and siFLI1 in combination compared to siERG alone ( S2A Fig ) . Cluster 2 includes genes encoding IL-8 and ICAM1 ( Fig 1E ) , consistent with the previous reports showing that these genes are upregulated by ERG knockdown [27 , 28] . GO analysis showed an enrichment of genes related to mitosis and cell cycle in cluster 2 ( S2C Fig ) . In contrast , cluster 3 is comprised of mesenchymal-related genes encoding mesenchymal markers ( S100A4/FSP-1 , αSMA , SM-22α , and N-Cadherin ) , extracellular matrix ( ECM ) ( collagen , versican , and fibronectin ) , an EMT/EndMT driver ( SLUG ) , and inflammatory cytokines and chemokines ( e . g . IL-1β , CXCL10 , and CXCL11 ) . GO analysis revealed an enrichment of inflammation-related terms , such as ‘defense response’ and ‘inflammatory response’ in this cluster ( Fig 1E and 1F ) . As for cluster 3 genes , while siERG caused a greater change in expression level than siFLI1 , siFLI1 in combination with siERG dramatically upregulated gene expression more than either one individually ( S2A and S2B Fig ) . These results clearly indicate that FLI1 , as well as ERG , have a non-negligible contribution to mesenchymal conversion . The heatmap also shows diverse patterns of ups and downs in gene expression across 8 experimental conditions , some of which are obscure in the current visualization method ( Fig 1E ) . To make the individual functions of ERG and FLI1 more clearly , we performed hierarchical clustering again by using only day 3 datasets , which reflect the direct responses of ERG/FLI1 knockdown . Consequently , we detected 8 major regulation patterns ( I–VIII ) and other minor patterns ( S3 Fig ) . The list of genes in each pattern is also shown ( S1 Table ) . Note that Z-scores do not coincide with the violin plots in S2A Fig , and each pattern does not necessarily correspond to a certain cluster in Fig 1E , since the expression values are re-normalized within day 3 datasets and newly clustered without day 7 datasets . As for gene sets in which ERG and FLI1 cooperatively represses expression ( I–III ) , ERG predominantly represses expression in the majority of cases ( I ) , while FLI1 can also have a predominant role in repressing specific genes ( II ) . ERG and FLI1 almost redundantly repress certain genes ( III ) . As for gene sets in which ERG individually represses expression ( IV and V ) , an additional knockdown of FLI1 partially ( IV ) or completely ( V ) counteracts increased expression by ERG knockdown . This implies that FLI1 ( V ) , along with other TFs ( IV ) , promotes the expression of specific gene sets in the absence of ERG . As for gene sets in which ERG and FLI1 cooperatively promote expression ( VII and VIII ) , again , ERG predominantly promotes expression in the majority of cases ( VII ) and FLI1 can also have a predominant role in repressing specific genes ( VIII ) . Interestingly , ERG and FLI1 have opposing roles in regulating VI genes; ERG promotes and FLI1 represses expression . Next , we evaluated the mechanism of EndMT induction via downregulation of ERG and FLI1 expression by analyzing the functions of these TFs in ECs . We screened the genome-wide binding regions of ERG and FLI1 in HUVECs by ChIP-seq analysis . Prior to performing the ChIP assay , specificities were evaluated by immunoblot analysis to examine the possible cross-reactivity of anti-ERG and anti-FLI1 antibodies given the high structural similarity of these TFs ( S4A Fig ) . Although the commercially available antibodies tested showed some cross-reactivity , antibodies that dominantly detected specific targets were used for the ChIP assay . ChIP-seq revealed 77 , 467 and 47 , 002 peaks in ERG and FLI1 , respectively ( Fig 2A ) . The peaks of ERG and FLI1 highly overlapped with each other , reflecting the structural similarity of these TFs ( Fig 2A ) . Reproducibility of the ChIP-seq was confirmed by the similarity between two biological replicates as shown in S4B Fig . The peaks of ERG and FLI1 were distributed as shown in Fig 2B and S4C Fig . Motif analysis of ERG and FLI1 showed that the ETS- and AP-1-binding motifs are ranked first and second , respectively ( Fig 2C ) . This is consistent with several reports illustrating the coordinating activity and physical binding of ERG/FLI1 and AP-1 [29 , 30] . GO analysis showed that the peaks of ERG and FLI1 are highly enriched in the proximal region of genes associated with vascular function , such as ‘blood vessel morphogenesis’ and ‘angiogenesis’ , indicating that these proteins are essential TFs in ECs , as expected ( Fig 2D ) . To further dissect the functions of ERG and FLI1 in ECs , we obtained additional ChIP-seq data for H3K4me3 and H3K27Ac , which are major histone modifications that mark promoters and/or enhancers , in HUVECs treated with siControl and siERG+siFLI1 . Reproducibility of the ChIP-seq data was confirmed using two biological replicates ( S5A Fig ) . From a macroscopic perspective , the peaks commonly bound by ERG and FLI1 highly overlapped with H3K27Ac , indicating transcriptionally active regulatory regions ( Fig 3A ) . Additionally , this H3K27Ac was significantly lost by siERG+siFLI1 treatment ( Fig 3A ) . In contrast , the regions bound by ERG/FLI1 and H3K27me3 , which marks the transcriptionally repressive state , were mutually exclusive ( S5B and S5C Fig ) . These results indicate that ERG and FLI1 bind genomic regions permissive to TF binding and are associated with gene activation rather than repression . Because the peaks of ERG and FLI1 were enriched in the proximal regions of genes associated with EC function ( Fig 2D ) , we predicted that these TFs may activate the transcription of a wide range of EC-specific genes . Indeed , ChIP-seq data showed that ERG and FLI1 bound to the upstream regions of various EC-specific genes ( Fig 3B and S6A Fig ) . Additionally , these peaks highly overlapped with the H3K27Ac peaks , which were lost with siERG+siFLI1 treatment ( Fig 3B and S6A Fig ) ; these results are consistent with the dagger-marked regions in Fig 3A . Loss of H3K27Ac was accompanied by decreased gene expression ( S2B Fig ) . These data clearly showed that ERG and FLI1 directly bind the enhancer/promoter regions of various EC-specific genes and promote their transcription . This accounts for one side of the mechanism of EndMT induction; downregulation of ERG and FLI1 expression leads to a significant decrease in EC-specific gene expression , and thus to the loss of endothelial properties . In addition to changes in histone modification in the regulatory regions of EC-specific genes , we also observed that H3K4me3 and/or H3K27Ac marks increased in the upstream regions of mesenchymal-related genes accompanied by increased gene expression after knockdown of ERG and FLI1 in HUVECs ( S2B and S6B Figs ) . Interestingly , these regions were not necessarily bound by ERG and FLI1 . This observation encouraged us to perform a comprehensive analysis on ChIP-seq and gene expression microarray data . Fig 3C shows the local profiles of ChIP-seq signals observed in the whole genomic region which are classified into 15 classes ( left panel ) , and classes which frequently appear in the regulatory regions of the genes in each microarray cluster ( Fig 1E ) ( right panel ) . For example , genomic regions grouped into class 5 are commonly bound by ERG and FLI1 . These regions are marked with H3K27Ac , which are lost upon siERG+siFLI1 treatment . This class is strongly correlated with cluster 1 in the gene expression microarray ( P = 1 . 81E-18 ) , confirming that ERG and FLI1 directly regulate the expression of cluster 1 genes which are highly associated with EC-specific genes ( Fig 1F ) . Fig 3C also illustrates that EndMT mediated by ERG/FLI1 loss induces genome-wide changes of histone modifications , even in regions not bound by ERG and FLI1 . Particularly , class 14 ( P = 1 . 36E-23 ) , 10 ( P = 2 . 58E-23 ) , and 11 ( P = 7 . 88E-17 ) , which are highly correlated with the regulatory regions of cluster 3 genes , shows very low or no ERG/FLI1 binding , while levels of H3K4me3 , H3K27Ac , or both , are increased after siERG+siFLI1 treatment . Taken together , ERG and FLI1 prevent ECs from EndMT by directly promoting EC-specific genes and indirectly repressing EndMT-promoting genes via epigenetic regulation . Interestingly , a detailed motif analysis on Fig 3C showed that the GGAA repeat sequence was more highly enriched in the ERG+FLI1+ enhancer ( class 4 and 5 ) than in the ERG+FLI1- enhancer ( class 7 ) ( Fig 3D ) . These results suggest that ERG and FLI1 may interact with each other and cooperatively activate certain EC-specific genes . Therefore , we investigated whether ERG and FLI1 have physical contact by performing a co-immunoprecipitation assay for Myc-tagged ERG and Flag-tagged FLI1 . As shown in Fig 3E , when precipitated using anti-Myc antibody , co-immunoprecipitated FLI1 was detected and vice versa . Furthermore , ERG and FLI1 synergistically drive the activity of VWF promoter , an EC-specific gene ( Fig 3F ) . These results indicate that ERG and FLI1 form a complex , and cooperatively regulate the expression of a set of EC-specific genes . To identify the critical downstream target of ERG/FLI1 with EndMT-inhibiting function , we globally screened genes bound and transcriptionally activated by ERG and FLI1 using ChIP-seq and microarray data . Genes that meet the following three criteria in ChIP-seq were listed: 1 ) the upstream region has ERG/FLI1-binding peaks , 2 ) ERG/FLI1 peaks overlap with H3K27Ac peaks , and 3 ) the H3K27Ac peaks are reduced by siERG+siFLI1 . Moreover , genes that meet the following two criteria in microarrays were listed: 1 ) siERG+siFLI1 reduces expression by >70% , and 2 ) siERG+siFLI1 reduces expression more than siERG or siFLI1 alone . Finally , 293 candidate genes were commonly listed ( S7A Fig and S2 Table ) . Through all screenings , SMAD1 was a possible candidate . During manuscript preparation , it was reported that ERG controls the TGFβ/SMAD signaling pathway to block EndMT by promoting SMAD1 expression and inhibiting DNA binding of SMAD3 in , for example , CNN1 and TGFB2 promoters [26] . Consistent with this finding , our ChIP-seq and microarray data clearly showed that ERG and FLI1 bound the SMAD1 promoter , and combined knockdown of ERG/FLI1 reduced SMAD1 expression ( S8A Fig ) . Moreover , ERG and FLI1 bound promoter regions of CNN1 and TGFB2 , and combined knockdown of ERG/FLI1 increased their expression ( S8B Fig ) . Taken together , these data suggest that FLI1 as well as ERG can modify the SMAD pathway to protect ECs from EndMT . Subsequently , to identify a new molecule , we collected information on microRNA , as microRNAs are a well-recognized regulator of EMT [31] . Among the microRNAs related to EMT/EndMT , we searched for the downstream target of ERG/FLI1 ( S7B Fig ) . As a result , our ChIP-seq data indicated that microRNA-126 ( miR-126 ) is the most promising direct target because ERG and FLI1 bind the enhancer/promoter regions of miR-126 , and these regions overlap with H3K27Ac , which was significantly lost after siERG+siFLI1 treatment ( Fig 4A ) . Moreover , miR-126 expression was significantly decreased by combined knockdown of ERG and FLI1 , indicating that these TFs directly promote the expression of miR-126 ( Fig 4B ) . In contrast , we failed to detect ERG/FLI1 peaks and/or significant changes in histone modification enrichment in the other EMT/EndMT-related microRNAs such as the let-7 family and the miR-200 family after knockdown of ERG and FLI1 ( S9 Fig ) . MiR-126 is an EC-specific microRNA located in intron 7 of EGFL7 , which is also an EC-specific gene . Interestingly , MirDIP microRNA target prediction database [32] indicated that mesenchymal-related genes involving TAGLN , COL1A1 , and SNAI2 are potential targets of miR-126 ( S3 Table ) . Moreover , a recent study showed that miR-126 blocks TGFβ-induced EndMT by targeting PIK3R2 mRNA [33 , 34] . Thus , we investigated whether restoration of miR-126 counteracts endothelial/mesenchymal marker expression changes through downregulation of ERG and FLI1 expression . As shown in Fig 4C and 4D , transfection with a miR-126 mimic partially counteracted decreased expression of CDH5 and PECAM1 . A miR-126 mimic also counteracted increased expression of ACTA2 , COL1A1 , and SNAI2 , but not TAGLN . In contrast , a miR-126 inhibitor induced partial EndMT in HUVECs , indicated by the downregulation of CDH5 expression and the upregulation of TAGLN , COL1A1 , and SNAI2 expression ( S10 Fig ) . Taken together , these data suggest that EndMT mediated by the loss of ERG/FLI1 is at least in part based on the reduced expression of miR-126 under the direct control of these TFs . Because EndMT is known to be induced in the tumor microenvironment [5] , we investigated the expression of ERG and FLI1 in ECs in tumor tissues by immunofluorescent staining . Note that the anti-ERG antibody used for immunostaining differed from that used for ChIP-seq and shows cross-reactivity with FLI1 as described in the manufacturer’s datasheet ( S4A Fig ) . First , we observed the expression of ERG and FLI1 in normal aorta and skin ECs . Consistent with the fact that ERG has been widely recognized as an EC marker in immunostaining assays and previous reports showing strong detection of FLI1 in ECs [35] , strong and homogenous expression of these TFs were observed in CD31+ ECs ( Fig 5A ) . In contrast , within tumors formed after subcutaneous injection of B16F10 melanoma cells , some ECs showed reduced expression of ERG and FLI1 ( Fig 5B and 5C ) . Similar results were observed in tumor tissues formed by E0771 cells ( intra-fat pad ) and 3LL cells ( subcutaneous ) ( S11A and S11B Fig ) . These data suggest that EndMT mediated by ERG/FLI1 loss is induced in the tumor microenvironment in vivo . We then evaluated the cause of downregulation of ERG and FLI1 expression in ECs within tumors . Previous studies reported that the extracellular environment including soluble factors represented by inflammatory cytokines , hypoxia , and high glucose trigger EndMT under pathological conditions [4 , 36] . Among these , we investigated whether soluble factors and hypoxia reduce the expression of ERG and FLI1 , as aberrant soluble factor profiles and hypoxia are key characteristics of the tumor microenvironment . First , to examine the effect of soluble factors , we evaluated expression changes of ERG and FLI1 in HUVECs treated with culture media conditioned with intratumoral whole cell populations . As a result , ERG and FLI1 expression was significantly decreased after a 4-hour conditioned media treatment by all three implanted tumors investigated , indicating that expression of these TFs is at least partially downregulated by soluble factors enriched in the tumor microenvironment ( Fig 6A ) . In addition , the 24-hour treatment with B16F10 tumor tissue-conditioned media induced the expression of mesenchymal markers in HUVECs while the 4-hour treatment did not ( Fig 6B ) . We also found that various inflammatory cytokines , particularly TNFα , IL-1β and IFNγ downregulated the expression of ERG and/or FLI1 in HUVECs ( S12A Fig ) . In contrast , cobalt chloride , which is known to induce hypoxia by activating HIF-1 , did not downregulate the expression of ERG or FLI1 ( S12B Fig ) . These results suggest that soluble factors play a key role in promoting EndMT by repressing ERG/FLI1 expression in the tumor environment . To assess the relevance between EndMT mediated by loss of ERG/FLI1 and cancer progression , we analyzed Kaplan-Meier plots obtained from the PrognoScan database ( http://www . abren . net/PrognoScan/ ) . In consideration of the technical limitations of PrognoScan-based Kaplan-Meier plots , which were constructed based on the transcriptome of whole tumor tissues , we set two criteria as follows: ( 1 ) ERG is used as a prognostic marker because its expression is highly limited to ECs , while FLI1 is expressed in other cell populations such as myeloid cells [13]; ( 2 ) the probe set 213541_s_at is used because it is validated in the current study . Under these conditions , we found that lower expression of ERG was significantly related to poor prognosis in melanoma ( overall survival ) , breast cancer ( disease-specific survival ) , and lung cancer ( overall survival ) ( S13A and S13B Fig ) . These data support the idea that EndMT mediated by loss of ERG/FLI1 promotes tumor progression .
In the current study , we found that the expression of ERG and FLI1 in tumor ECs is downregulated because of soluble factors enriched in the tumor microenvironment . Reduced expression of ERG and FLI1 resulted in the loss of a broad range of EC-specific genes under the direct control of these TFs , leading to the loss of endothelial characteristics . Further , among the EC-specific genes transcriptionally activated by ERG and FLI1 , we found that miR-126 partially blocks EndMT , the loss of which triggers EndMT ( Fig 7 ) . Additionally , we gained genome-wide insight into the functions of ERG and FLI1 , which have been recently recognized as essential TFs in EC differentiation and function [37] . ERG directly regulates several EC-specific genes to maintain EC function [19–26] . Additionally , reduced expression of ERG leads to upregulation of CXCL8 , ICAM1 , and VCAM1 , which are representative genes in inflammatory ECs , suggesting that ERG is a “gatekeeper” against inflammatory phenotypes [27 , 28 , 38] . In contrast to ERG , the function of FLI1 is less-characterized in ECs; its ablation results in upregulation of CTSL and CXCL6 and downregulation of CXCL5 and CCN1 in the context of systemic sclerosis [39–42] . Our ChIP-seq and microarray analysis found that ERG and FLI1 bind enhancer/promoter regions and directly regulate various EC-specific genes , some of which are first characterized by our genome-wide study . Moreover , our study indicated the synergistic role of ERG and FLI1; they can interact physically ( Fig 3E ) , and synergistically transactivate VWF promoter activity ( Fig 3F ) . Notably , ERG also interacts with other molecules such as NFκB and SMAD3 to regulate gene transcription [26 , 28 , 38] . Thus , the molecular interaction networks of ERG and FLI1 need to be further analyzed to clarify the whole mechanism of mesenchymal transition . ERG and FLI1 , individually or in combination , have diverse modes of gene regulation ( S3 Fig ) . In support of the idea that ERG and FLI1 have a physical interaction , some regulation patterns indicate that these TFs mutually support each other’s function ( I , II , III , VII , and VIII in S3 Fig ) . In this case , ERG usually has a predominant role in regulating gene expression ( I , VII ) . Moreover , ERG and FLI1 may change a function depending on whether they form a complex or not; FLI1 can drive a set of genes only in the absence of ERG ( IV and V in S3 Fig ) . Interestingly , pattern V gene sets include EC inflammatory genes ( ICAM1 and CXCL8 ) , raising the hypothesis that FLI1 triggers an inflammatory state in response to the loss of ERG as an “emergency signal” . In contrast , ERG and FLI1 occasionally have opposing roles; ERG promotes , but FLI1 represses the expression of pattern VI gene sets . This raises another possible hypothesis that FLI1 can fine-tune ERG-driven upregulated expression in regulating a subset of genes through direct contact . In contrast to the combinatorial effect of siERG and siFLI1 , detailed results induced by siERG or siFLI1 alone are inconsistent between figures and with the literature . For example , ACTA2 is downregulated by siERG in S1B Fig , but upregulated in Fig 1C , S2B Fig , and the literature [26] . In addition , TAGLN is upregulated further by siERG alone than by siERG+siFLI1 in Fig 1C , which is inconsistent with the other figures . These inconsistencies are possibly due to the difference in experimental techniques ( qPCR and microarray ) , materials ( siRNAs ) , and primary HUVEC lots and passage numbers . Further specialized analysis will be needed to precisely determine the individual role of ERG and FLI1 in regulating a certain gene . Transcriptional activities of ERG and FLI1 can also be regulated in a post-translational manner . For example , ERG is activated by phosphorylation at serine 96 , 215 , and 276 in ECs [43] . In contrast , the transcriptional activity of FLI1 is disrupted by phosphorylation at threonine 312 and subsequent acetylation at lysine 380 via the non-canonical TGFβ signaling pathway [44] . These data indicate that EndMT can be triggered by attenuated functions of ERG and FLI1 as well as by reduced mRNA expression of these TFs . Through genome-wide target screenings , we focused on miR-126 as a major EndMT-inhibiting factor under the direct control of ERG and FLI1 ( Fig 4 ) . MiR-126 is an EC-specific microRNA that maintains EC function by reinforcing VEGFR signaling [45 , 46] . In support of our finding , miR-126 knockout mice show phenotypes similar to Erg and Fli1 knockout mice , suggesting that miR-126 is a pivotal downstream molecule of ERG and FLI1 [14–17 , 45] . It has been previously reported that miR-126 is transcriptionally activated by ETS1 and ETS2 [45 , 47] , but neither ERG nor FLI1 can activate a proximal promoter of miR-126 [47] . In the current study , we found that these TFs bind not only to a proximal promoter but also to distal regulatory regions of miR-126 . Coupled with histone modification changes by ERG/FLI1 knockdown , we conclude that ERG/FLI1 can promote miR-126 expression . MiR-126 has been shown to target PIK3R2 mRNA , which codes for a negative regulator of the PI3K/Akt signaling pathway , to protect ECs from EndMT in the context of TGFβ1-induced EndMT . Loss of miR-126 results in diminished activation of PI3K/Akt signaling and subsequent nuclear translocation of FOXO3a , which cooperates with SMAD3/4 to activate EndMT program genes [33 , 34] . We therefore assessed the contribution of the PIK3R2-PI3K/Akt-FOXO3a axis to EndMT mediated by the loss of ERG/FLI1 . However , we failed to find significant upregulation of PIK3R2 in HUVECs treated with siERG+siFLI1 , despite clear downregulation of miR-126 ( S14A Fig ) . This observation raises a possibility that another downstream pathway works in EndMT mediated by ERG/FLI1 loss . SNAI2 and TWIST2 , well-characterized EMT/EndMT regulators , are significantly upregulated upon combined knockdown of ERG and FLI1 , but SNAI2 or TWIST2 knockdown could not counteract the endothelial/mesenchymal marker expression changes through the downregulation of ERG and FLI1 expression ( S14B and S14C Fig ) . Although ERG directly binds and transcriptionally activates SNAI2 in the context of endocardial-to-mesenchymal transition during heart development [14] , we observed upregulation of SNAI2 expression after siERG treatment , probably due to the difference between ECs and endocardial cells , or postnatal and embryonic stages . Another well-known EndMT inducer , TGFβ2 , was upregulated by suppression of ERG and FLI1 in HUVECs ( S8 Fig ) . Thus , TGFβ2 and the downstream SMAD signaling pathway may enhance mesenchymal transition in an autocrine manner , which needs to be further investigated . Previous studies showed that ERG or FLI1 expression was downregulated in atherosclerosis , systemic sclerosis , pulmonary arterial hypertension , and liver fibrosis [26 , 38 , 48 , 49] . In these reports , ablation of ERG and/or FLI1 in ECs resulted in upregulation of cytokine and chemokine expression and showed inflammatory phenotypes , consistent with our study ( Fig 3E and 3F ) . Here , we additionally show that ERG and FLI1 expression is also downregulated in cancer pathology . Given that inflammation is a well-known pro-tumor factor [50] , ECs that have undergone EndMT mediated by ERG/FLI1 loss may promote tumor progression in an inflammation-dependent manner . It has been shown that TNFα and IL-1β downregulate ERG expression , and IFNγ downregulates FLI1 expression [26 , 27 , 38 , 51] . In the current study , we found that soluble factors enriched in tumor tissues are responsible for reduced expression of ERG and FLI1 ( Fig 6 ) . In the tumor microenvironment , ERG and FLI1 can be downregulated via the mixture of cytokines from many sources such as infiltrating immune cells ( e . g . neutrophils , macrophages , and NK cells ) , ECs , CAFs , and tumor parenchyma . In addition , TNFα , IL-1β , IFNγ , and TGFβ have been shown to induce EndMT [36 , 52 , 53] . Some studies reported that the combination of inflammatory cytokines synergistically induce EndMT , which may be attributed to the synergistic downregulation of ERG and FLI1 expression [52 , 53] . Taken together , these data suggest that multiple cytokine dynamics rather than a single cytokine-mediated signaling would lead to EndMT via ERG and FLI1 reduction in the tumor microenvironment . In conclusion , we identify ERG and FLI1 as critical regulators of EndMT in ECs . ERG and FLI1 , individually or in combination , directly induce EC-specific genes and indirectly repress mesenchymal genes by epigenetic regulation . Importantly , ERG and FLI1 cooperatively regulate a set of EC-specific genes , indicating that both TFs are important for EC function . Our work delineates the role of ERG and FLI1 in ECs , and suggests that maintaining the expression of these TFs may be possible therapeutic options for various EndMT-related diseases including cancer .
The Animal Care and Use Committee of the University of Tokyo and The Animal Care and Use Committee of Kumamoto University School of Medicine approved the study ( A29-070R2 ) . The work was conducted according to guidelines issued by the Center for Animal Resources and Development of Japan . The guideline totally following the international animal research rule with 3R ( Replacement , Reduction and Refinement ) . Human umbilical vein endothelial cells ( HUVECs ) were purchased from Lonza ( Basel , Switzerland ) . HUVECs were cultured in EGM-2 ( CC-3162 , Lonza ) supplemented with 5% fetal bovine serum ( FBS ) in a humidified atmosphere of 5% CO2 at 37°C . B16F10 cells , E0771 cells , and 3LL-luc cells ( 3LL cells which constitutively expresses luciferase gene ) were cultured in DMEM ( D5796; Sigma , St . Louis , MO , USA ) supplemented with 10% FBS . 3LL-luc and E0771 were kind gifts from Dr . Yoshihiro Hayakawa ( Toyama University , Japan ) and Dr . Robin Anderson ( Peter MacCallum Cancer Centre , Australia ) , respectively . HUVECs were treated with siRNAs ( 4 nM ) , miRNA mimics ( 30 nM ) , or miRNA inhibitor ( 100nM ) using RNAiMAX Transfection Reagent ( 13778075 , Thermo Fisher Scientific , Waltham , MA , USA ) following the manufacturer’s instructions . In the control experiment , comparable concentrations of control siRNAs/miRNAs/miRNA inhibitor produced by the same manufacturer were transfected . A list of siRNAs and miRNAs is shown in S4 Table . HUVECs treated with siRNAs for 3 days were harvested with a cell scraper . Whole cell lysates were prepared using lysis buffer ( 1% NP-40 , 10% glycerol , 137 mM NaCl , 20 mM Tris-HCl , 1 . 5 mM MgCl2 , and 1 mM EDTA ) containing cOmplete protease inhibitor cocktail ( 11873580001 , Roche , Basel , Switzerland ) . The protein concentration of the whole cell lysate was quantified with a Pierce BCA Protein Assay Kit ( 23227 , Thermo Fisher Scientific ) . After preparing samples using Sample Buffer Solution with Reducing Reagent ( 09499–14 , Nacalai , Kyoto , Japan ) , 50 μg of protein was separated by SDS-PAGE and transferred to PVDF membranes ( 10600023 , GE Healthcare , Little Chalfont , UK ) . Immunoblots were blocked with 5% skimmed milk/TBST ( used also for antibody diluent below ) for 1 hour and subsequently incubated with anti-ERG ( 1:1000; ab136152 , Abcam , Cambridge , UK ) , anti-ERG ( 1:1000; ab92513 , Abcam ) , anti-FLI1 ( 1:5000; ab15289 , Abcam ) , and anti-Caspase3 ( 1:1000; 9662 , Cell Signaling Technology , Danvers , MA , USA ) overnight at 4°C . After the blots were washed with TBST , they were incubated with HRP-conjugated anti-mouse IgG ( 1:80 , 000; A9044 , Sigma ) or anti-rabbit IgG ( 1:80 , 000; A9169 , Sigma ) overnight at 4°C . After washing with TBST , chemiluminescent signals on the blots were detected using Chemi-Lumi One Super ( 02230 , Nacalai ) on an ImageQuant LAS 4000 mini ( GE Healthcare ) . For loading controls , the blots were stripped with WB Stripping Solution ( 05364–55 , Nacalai ) and reprobed with an antibody against β-actin ( 1:2000; A1978 , Sigma ) . Cells were detached using 0 . 2% EDTA/PBS . For VE-cadherin and CD31 staining , cells were blocked with 2% bovine serum albumins ( BSA ) ( 019–23293 , Wako , Osaka , Japan ) /PBS for 30 minutes . For αSMA and collagen type I staining , the cells were fixed in 4% paraformaldehyde ( 09154–85 , Nacalai ) for 10 minutes at room temperature , and blocked/permeabilized with 2% BSA/0 . 1% Triton X-100/PBS for 30 minutes . The cells were incubated with primary and subsequently secondary antibodies in 2% BSA/PBS for 1 hour at 4°C . Samples were analyzed with Guava easyCyte ( Millipore , Billerica , MA , USA ) . Antibodies used were Cy3-conjugated anti-αSMA ( 1:500; C6198 , Sigma ) , PE-conjugated anti-CD31 ( 1:50; 303105 , Biolegend , San Diego , CA , USA ) , anti-collagen type I ( 1:80; AB758 , Millipore ) and Alexa Fluor 647-conjugated anti-goat IgG ( 1:200; A-21447 , Thermo Fisher Scientific ) , and Alexa Fluor 647-conjugated anti-VE-cadherin ( 1:50; 561567 , BD Biosciences , San Jose , CA , USA ) . HUVECs were seeded into 35-mm dishes . After siRNA treatment , images were captured at hourly intervals for 72 hours with BioStudio ( Nikon , Tokyo , Japan ) . The mixture of 320 μL of Atelo Collagen ( IAC-30 , Koken , Tokyo , Japan ) , 40 μL of 10× MEM ( 1430030 , Thermo Fisher Scientific ) , and 40 μL of 10× neutralization buffer was plated in a 24-well plate and incubated for 1 hour at 37°C . The 10× neutralization buffer contained 0 . 1 M HEPES ( 17557–94 , Nacalai ) and 0 . 1 M NaHCO3 . HUVECs treated with siRNAs for 7 days were seeded onto the solidified gel and incubated overnight . After the medium was removed , 250 μL of collagen gel mixture was layered on the cells and incubated for 1 hour at 37°C . Finally , 500 μL of EGM-2 supplemented with 5% FBS and 50 ng/mL VEGF ( 223–01311 , Wako ) were added . After 48-hour incubation , tube formation was observed with a bright field microscope . ChIP-seq was performed as described previously [37 , 55] . Cells were fixed with 1% formaldehyde ( 061–00416 , Wako ) for 10 minutes or 0 . 5% formaldehyde for 5 minutes at room temperature , and then added at a concentration of 200 μM of glycine to stop the reaction . Cells were harvested with a cell scraper . Chromatin was sheared to 150–1000 bp with a SONIFIER 250 ( Branson , Danbury , CT , USA ) or DNA Shearing system S200 ( Covaris , Woburn , MA , USA ) . Immunoprecipitation was performed with anti-ERG ( ab136152 , Abcam ) , anti-FLI1 ( ab15289 , Abcam; sc-356 , Santa Cruz Biotechnology , Dallas , TX , USA ) , anti-H3K4me3 ( MABI0304 , MBL , Nagoya , Japan ) , and anti-H3K27Ac ( MABI0309 , MBL ) which were bound to Dynabeads M-280 Sheep Anti-Mouse IgG ( 11201D , Themo Fisher Scientific ) , Dynabeads M-280 Sheep Anti-Rabbit IgG ( 11203D , Themo Fisher Scientific ) , or Dynabeads Protein G ( 10004D , Themo Fisher Scientific ) . After the beads were washed , resuspended in elution buffer ( 1% SDS , 0 . 1 M NaHCO3 ) containing 1 mg/mL Pronase ( 10165921001 , Roche ) for at least 2 hours at 42°C and subsequently at 65°C overnight . DNA was purified with a QIAquick PCR Purification Kit ( 28106 , QIAGEN , Hilden , Germany ) following the manufacturer’s instructions , and quantified with a Qubit 3 . 0 Fluormeter ( Q33216 , Thermo Fisher Scientific ) and Qubit dsDNA HS Assay Kit ( Q32851 , Thermo Fisher Scientific ) . The sequence library was prepared from 2 . 5 ng DNA with a KAPA Hyper Prep Kit for illumina following the manufacturer’s protocol ( KK8502 , Kapa Biosystems , Wilmington , MA , USA ) and sequenced with a Genome Analyzer IIx ( Illumina , San Diego , CA , USA ) or HiSeq 2000 ( Illumina ) . Protocols for each antibody are summarized in S6 Table . Isolated human VWF promoter ( -2182/+1475 ) -luc [66] was transiently co-transfected with pCI-Erg , pCI-Fli1 , or both into Cos-7 cells . Two days later , luciferase activities were calculated using the Dual-Luciferase assay kit ( Promega , Madison , WI , USA ) as described previously [67] . Cos-7 cells were transfected with Myc-tagged Erg ( pEF6-Erg ) and Flag-tagged Fli1 ( pFlag-CMV2-Fli1 ) . After 24-hour incubation , cells were harvested and lysed with NP40 lysis buffer ( 0 . 5% NP-40 , 50 mM Tris-HCl , 50 mM NaCl , 1 mM EDTA ) containing complete protease inhibitor cocktail . After centrifuging at top speed for 10min , supernatant was precleared using ProteinG sepharose beads ( 71-7083-00 AI , GE healthcare ) , and then immunoprecipitation was performed with anti-Myc ( 5 μg; sc-40 , Santa Cruz Biotechnology ) or anti-Flag ( 5 μg; F1804 , Sigma ) overnight . After incubating with ProteinG beads for 2 hours , beads were washed with NP40 lysis buffer four times , boiled with Sample Buffer Solution with Reducing Reagent ( 09499–14 , Nacalai ) , and then subjected to immunoblot analysis with anti-Myc-Tag ( 1:1000; 2278 , Cell Signaling Technology ) or anti-Flag ( 1:2000;14793 , Cell Signaling Technology ) . C57BL6/N mice were purchased from Japan SLC ( Shizuoka , Japan ) . All animals were housed under a 12-hour dark-light cycle at 22 ± 1°C with ad libitum food and water . The Animal Care and Use Committee of the University of Tokyo and The Animal Care and Use Committee of Kumamoto University School of Medicine approved the protocols for animal experiments . Male and female 5–8-week-old mice were used for the experiments . B16F10 cells ( 1 × 106 cells ) and 3LL-luc cells ( 1 × 106 cells ) were inoculated subcutaneously into the right flank of syngeneic C57BL/6N mice . E0771 cells ( 2 × 105 cells ) were inoculated into the mammary fat-pad of syngeneic female C57BL/6N mice . Seven days ( B16F10 ) or 10 days ( 3LL-luc , E0771 ) after tumor inoculation , tumor tissues were used for immunofluorescent staining and collection of tumor tissue-conditioned media . Mice were sacrificed with CO2 and perfused with 10 mL of 4% paraformaldehyde ( 09154–85 , Nacalai ) . Tissues were harvested and fixed again with 4% paraformaldehyde for 2 hours at 4°C , followed by immersion in 30% sucrose overnight at 4°C . Tissues were embedded in OCT Compound ( 4583 , Sakura Finetek Japan , Tokyo , Japan ) and sectioned at a thickness of 10–20 μm with a Microm HM550 Cryostat ( Thermo Fisher Scientific ) . The sections were blocked/permeabilized in 10% FBS/0 . 5% Triton X-100/PBS for 1 hour . The sections were incubated with anti-ERG ( 1:100; ab92513 , Abcam ) , anti-FLI1 ( 1:100; ab15289 , Abcam ) , and anti-CD31 ( 1:100; 550274 , BD Biosciences ) overnight at 4°C . The sections were then incubated with biotin-conjugated anti-rabbit IgG ( 1:1000; BA-1000 , Vector Laboratories , Burlingame , CA , USA ) overnight at 4°C . Finally , the sections were incubated with Alexa Fluor 488-conjugated streptavidin ( S-32354 , Thermo Fisher Scientific ) and Alex Fluor 594-conjugated anti-rat IgG ( A-21209 , Thermo Fisher Scientific ) overnight at 4°C . After the sections were incubated with TO-PRO-3 ( 1:500; T3650 , Thermo Fisher Scientific ) or DAPI ( 1:500; 342–07431 , Dojindo , Kumamoto , Japan ) for 20 minutes at room temperature , they were mounted with FluorSave Reagent ( 345789 , Millipore ) . Samples were observed with a confocal laser microscope Fluoview FV-1000 ( Olympus , Tokyo , Japan ) . Kaplan-Meier plots were obtained from the PrognoScan database [68] . Datasets listed in S13A Fig ( Probe set ID: 213541_s_at ) were used , and Kaplan-Meier plots were reconstituted based on the raw data table . The accession number for the gene expression microarray and ChIP-seq reported in this paper is GEO: GSE109696 . Other relevant data are within the paper and its Supporting Information files . Numerical data underlying all graphs in this manuscript is shown in S7 Table . Data were analyzed by two-tailed unpaired Student’s t-test , non-parametric Mann-Whitney U test , or one-way ANOVA followed by Scheffe’s test . P-values < 0 . 05 were considered significant . | Differentiated cells possess unique characteristics to maintain vital activities . However , cells occasionally show abnormal behavior in pathological settings due to dysregulated gene expression . Endothelial-to-mesenchymal transition ( EndMT ) is a phenomenon in which endothelial cells lose their characteristics and acquire mesenchymal-like properties . Although EndMT is observed in various diseases including cancer , and augments fibrosis and vascular defects , the mechanism of EndMT induction is not fully understood . Here , we show that EndMT is triggered via reduced expression of ERG and FLI1 , which have recently been recognized as pivotal transcription factors in endothelial cells ( ECs ) . Mechanistically , ERG and FLI1 activate EC-specific genes and repress mesenchymal-like genes via epigenetic regulation to prevent EndMT . Furthermore , we demonstrate that microRNA-126 , which is specifically expressed in ECs , is the key downstream target of ERG/FLI1 for regulating EndMT . Finally , we show that ERG and FLI1 expression is decreased in ECs within tumors , suggesting that EndMT is induced in the tumor microenvironment . Collectively , these findings indicate that loss of ERG and FLI1 leads to the aberrant behavior of ECs in pathological conditions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"gene",
"regulation",
"natural",
"antisense",
"transcripts",
"cancer",
"treatment",
"immunology",
"histone",
"modification",
"oncology",
"micrornas",
"signs",
"and",
"symptoms",
"bioassays",
"and",
"physiological",
"analysis",
"epigenetics",
"chromatin",
"research",
"and",
"analysis",
"methods",
"small",
"interfering",
"rnas",
"inflammation",
"chromosome",
"biology",
"gene",
"expression",
"chromatin",
"modification",
"immune",
"response",
"microarrays",
"biochemistry",
"rna",
"diagnostic",
"medicine",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"non-coding",
"rna"
] | 2018 | Downregulation of ERG and FLI1 expression in endothelial cells triggers endothelial-to-mesenchymal transition |
Identification of genomic regions that are identical by descent ( IBD ) has proven useful for human genetic studies where analyses have led to the discovery of familial relatedness and fine-mapping of disease critical regions . Unfortunately however , IBD analyses have been underutilized in analysis of other organisms , including human pathogens . This is in part due to the lack of statistical methodologies for non-diploid genomes in addition to the added complexity of multiclonal infections . As such , we have developed an IBD methodology , called isoRelate , for analysis of haploid recombining microorganisms in the presence of multiclonal infections . Using the inferred IBD status at genomic locations , we have also developed a novel statistic for identifying loci under positive selection and propose relatedness networks as a means of exploring shared haplotypes within populations . We evaluate the performance of our methodologies for detecting IBD and selection , including comparisons with existing tools , then perform an exploratory analysis of whole genome sequencing data from a global Plasmodium falciparum dataset of more than 2500 genomes . This analysis identifies Southeast Asia as having many highly related isolates , possibly as a result of both reduced transmission from intensified control efforts and population bottlenecks following the emergence of antimalarial drug resistance . Many signals of selection are also identified , most of which overlap genes that are known to be associated with drug resistance , in addition to two novel signals observed in multiple countries that have yet to be explored in detail . Additionally , we investigate relatedness networks over the selected loci and determine that one of these sweeps has spread between continents while the other has arisen independently in different countries . IBD analysis of microorganisms using isoRelate can be used for exploring population structure , positive selection and haplotype distributions , and will be a valuable tool for monitoring disease control and elimination efforts of many diseases .
Two alleles are identical by state ( IBS ) if they have the same nucleotide sequence . These alleles can be further classified as identical by descent ( IBD ) if they have been inherited from a common ancestor [1] . While a genomic region that is IBD must also be IBS , the converse of this statement is not true . It therefore follows that individuals who share a genomic region IBD are in fact related . For closely related individuals , these regions tend to be large and frequently distributed across the genome . However , as individuals become more distantly related , recombination breaks down IBD regions over time such that they become smaller , less frequently distributed and may disappear altogether [1] . For extremely distant relatives , small IBD segments will persist which are the result of non-random allele associations or linkage disequilibrium ( LD ) [2] . Such ancient IBD is not the focus of this article , instead we are concerned with IBD that has been inherited from a recent common ancestor , within 25 generations . Human genetic studies have greatly benefited from identification of IBD regions , with applications including disease mapping [3] , discovery of familial relatedness [4] and determining loci under selection [5 , 6] . With considerable work focusing on human studies , much of the statistical framework underpinning IBD algorithms has been tailored to diploid genomes , making them unsuitable for analysis of non-diploid organisms [7] . In particular , IBD analysis of microorganisms that cause disease , such the malaria causing parasite , Plasmodium , and bacterium Staphylococcus aureus , are not feasible with the current methodologies due to the haploid nature of their genomes and the presence of multiple strains in an infection . IBD analysis would be invaluable for the study of these , and other diseases , as it can be used to infer fine-scale population structure [8 , 9] , investigate transmission dynamics [8 , 9] and identify loci under selection that may be associated with antimicrobial resistance . The main challenge for IBD analysis of microorganisms is the presence of multiple infections , where the number of strains in an infection is termed the multiplicity of infection ( MOI ) , or alternatively the complexity of infection ( COI ) . For a haploid organism like Plasmodium , the genomic data extracted from an infection with MOI = 1 is trivially phased . This makes analysis of such isolates relatively straightforward . However , when MOI > 1 the genomic data will appear as heterozygous . In this instance , such isolates are typically excluded from population genomic analyses as statistical methods are not well equipped to deal with this added complexity [10–12] . The first probabilistic model for identifying IBD between pairs of haploid genomes was introduced by Daniels et al . [9] , who implemented a hidden Markov model ( HMM ) for IBD detection in Plasmodium . This model has since been used for in-depth analyses of population structure and disease transmission in malaria [8 , 9] , and was recently made available as the tool hmmIBD [13] . However , as it is only applicable to haploid genomes , it is limited to MIOI = 1 isolates only . Recently , we developed a similar HMM in the package XIBD [7] , for detecting IBD on the human X chromosome . Since the X chromosome is haploid in males and diploid in females , XIBD requires three separate models to account for the difference in ploidy between male and female pairs . Here , we make use of these models in our latest tool isoRelate , which is a freely available R package that performs IBD mapping on recombining haploid species , that also allows for multiple infections . Our model uses unphased genotype data from biallelic single nucleotide polymorphisms ( SNPs ) , which can be obtained from either array data or sequencing data that randomly samples SNP variation throughout the genome . The use of biallelic SNPs means that at most 2 alleles can be shared IBD between any two isolates with MOI > 1 . As such , IBD is likely to be inferred between the dominant two clones in an infection . However , IBD can be inferred between minor clones given their relative contribution to the infection is high enough to be captured by genotyping algorithms . isoRelate also offers a number of useful functions for downstream analyses following the detection of IBD segments , including identification of loci under selection using a novel statistic based on IBD inference , and is currently the only tool with such exploratory features . We perform extensive simulation analyses to assess the performance of isoRelate when detecting IBD segments in the presence of multiclonal infections , in addition to comparisons of our proposed selection statistic with several existing methods and their ability to detect complex patterns of positive selection . Furthermore , we demonstrate the value of IBD analysis with isoRelate by analyzing whole genome sequencing ( WGS ) data for a previously published global Plasmodium falciparum dataset of 2 , 550 isolates [14] . We use isoRelate to explore the population structure of P . falciparum in different geographical regions and investigate the distribution of shared haplotypes over positively selected regions using relatedness networks implemented in isoRelate .
We performed a simulation study to assess the power and accuracy of isoRelate in detecting IBD segments using sequencing data for P . falciparum when the number of clones in an infection , and their respective frequencies , varies . In particular , we assessed the performance of isoRelate when isolates have MOI = 1 , MOI = 2 ( relative clonal frequencies: 50:50 , 75:25 and 90:10 ) and MOI = 3 ( relative clonal frequencies: 34:33:33 , 50:30:20 , 70:20:10 ) . We arbitrarily selected chromosome 12 ( Pf3D7_12_v3 ) for IBD inference , and simulated sequencing data for pairs of isolates separated from 1 to 25 generations ( siblings to 24th cousins ) , where isolates separated by 25 generations are likely to have on average IBD segments of length 2cM , which is the smallest length that an IBD segment is detected with high power by most IBD algorithms for human genome analyses [1] . For each of the 25 generations , we simulated 200 haploid pairs of related isolates , mimicking MOI = 1 infections , totalling 10 , 000 simulated isolates . Similarly , 10 , 000 MOI = 2 and MOI = 3 isolates were each simulated such that only one clone in the mixed infection had relatedness included in its genome , where this clone was randomly assigned as the major , minor or middle ( for MOI = 3 ) clone in the isolate , with respect to clonal frequency ( see Material and Methods for more details on the simulation process ) . The results from this analysis are displayed in Fig 1 , where we define power as the average proportion of a segment that is detected as a function of the size of the true IBD segment , and accuracy as the probability that at least 50% of a detected segment is true as a function of the reported size of the detected segment . Naturally , isoRelate has the greatest ability to detect IBD segments when there are fewer clones in the isolate . It is also capable of detecting IBD in multiclonal infections , however as the number of clones increases and the major clone’s frequency decreases , the power and accuracy of isoRelate also decreases . Additionally , isoRelate is able to detect IBD in the minor clone when it contributes to more than 20% of the infection . Overall , isoRelate has the greatest power to detect IBD segments that are 4cM or larger in P . falciparum . This corresponds to detecting relatedness between clones separated by up to 13 generations ( or 25 meioses ) . Additionally , if IBD segments are detected that are 1 . 5cM or longer , then there is at least an 80% chance that they will be real . We believe that the allele frequency spectrum of P . falciparum , which is heavily skewed to the right ( S1A Fig ) , reduces the performance of isoRelate as most SNPs have the reference allele resulting in little genetic variation between isolates . To test this , we performed a second simulation whereby the allele frequency spectrum was generated to follow a uniform distribution ( S1B Fig ) . Here , isoRelate performs exceptionally well , even in the presence of mixed infections ( Fig 2 ) . In particular , IBD segments as small as 2cM are detected with high power and accuracy . These results suggest that the allele frequency spectrum of the species under evaluation will impact of the ability of isoRelate to detect IBD segments , with increasingly skewed distributions resulting in reduced IBD performance . We also validated our methodology by applying isoRelate to the MalariaGEN Pf3k genetic cross dataset [15] to detect known recombination events . This dataset contains the parents and offspring of three P . falciparum strain crosses; 3D7 x HB3 , 7G8 x GB4 , and HB3 x Dd2 . There are 21 , 40 and 37 isolates for the three crosses respectively , and 11 , 612 SNPs , 10 , 903 SNPs and 10 , 637 SNPs remaining following filtering procedures ( S1 Table ) . We combined the results for all three crosses and found that isoRelate detected 98% of all reported IBD segments , with an average concordance between inferred and reported segments of 99% . Additionally , isoRelate detected segments with 99% accuracy . We did not detect IBD between any of the founders . This is expected given the documented origins of these three strains , which were derived from very different geographic regions [16] . False negatives , where IBD was not inferred between parents and offspring , were observed predominantly in genomic regions located between recombination events . Moreover , identical segment boundaries were detected between all replicate isolates . We developed a selection statistic based on inferred IBD to assess the significance of excess IBD sharing indicative of positive selection . Briefly , we transformed a binary IBD matrix to account for variations in relatedness between isolates and SNP allele frequencies , then performed normalization allowing us to calculate–log10 p-values for each SNP . We assessed the performance of our proposed selection statistic on SNP data simulated from an evolutionary model for P . falciparum under three scenarios of positive selection; hard selective sweep , soft selective sweep ( i . e . recurrent variants ) and selection on standing variation . For each selective sweep , selection coefficients of s = 0 . 01 , s = 0 . 1 and s = 0 . 5 were examined , where selection on standing variation was introduced to existing alleles with population allele frequencies of either f = 0 . 01 , f = 0 . 05 or f = 0 . 1 , while hard sweeps and soft sweeps were introduced to new alleles . Sweeps were randomly inserted along a 2 . 27 Mb region , which is approximately the size of P . falciparum chromosome 12 . Ten replicate simulations were performed for each combination of selection parameters , resulting in a total of 150 simulated datasets . 200 haplotypes were sampled at 50 , 100 , 200 and 500 generations following the introduced sweeps ( see Material and Methods for more details on the simulation process ) . We compared the selection signatures generated by isoRelate to those detected by the integrated haplotype score ( iHS ) [17] and haploPS [18] . iHS makes use of the extended haplotype homozygosity ( EHH ) test , which calculates the probability that two randomly selected chromosomes have identical haplotypes adjoining an identical core haplotype [11 , 17 , 19] . In contrast , haploPS identifies positive selection by comparing the lengths of identified haplotypes with other haplotypes genome-wide at similar frequencies . Both iHS and haploPS require knowledge of haplotype phase , therefore we performed initial comparisons of isoRelate , iHS and haploPS using only isolates with MOI = 1 as haplotype phase is known . A second analysis was performed allowing isolates to have MOI > 1 ( S2 Table ) . isoRelate and iHS produce selection statistics that follow known distributions . We thus generated Q-Q plots for SNP specific test statistics for both of these methods ( S2 and S3 Figs ) . We calculated the power and accuracy of isoRelate , iHS and haploPS in detecting these sweeps , where power was defined as the proportion of sweeps that were detected within 50kb of the selected SNP , and accuracy was defined as the proportion of detected sweeps that were within 50kb of the selected SNP . The results from the analysis of MOI > 1 isolates are comparable to the analysis of MOI = 1 isolates ( S4 and S5 Figs ) , thus we describe the results from the analysis of MOI = 1 isolates only . No method is able to detect a sweep with a selection coefficient of s = 0 . 01 with high power and accuracy , regardless of the type of sweep ( S4 Fig ) . Sweeps with selection coefficients of s = 0 . 1 and s = 0 . 5 are more readily identified . For analysis of hard sweeps , haploPS outperforms isoRelate and iHS , particularly as the selection coefficient increases ( Fig 3 , S4 Fig ) . Specifically , haploPS is able to detect a hard sweep with selection coefficient s ≥ 0 . 1 at least 500 generations after its introduction while isoRelate and iHS are limited to less than 200 generations . Soft selective sweeps and selection on standing variation are less readily identified than hard selective sweeps , particularly as the initial allele frequency f increases . Such complex sweeps are limited to detection within 200 generations of the initial pressure by all methods . Hughes and Verra [20] used three generations per year as a conservative estimate of the average generation time in P . falciparum . Given this , all methods should be able to detect complex sweeps that occurred up to approximately 66 years ago , depending on the selection coefficient , which is within the timeframe of reported antimalarial drug resistance [21] . More generally , haploPS has the greatest power to detect sweeps of all types , however this comes at the cost of more falsely detected sweeps resulting in reduced accuracy . In contrast , both isoRelate and iHS detect sweeps with high accuracy . This suggests that a combination of tools would be useful for inferring positive selection , where consensus sweeps would be a good indication of true selection . To demonstrate the ability of isoRelate to investigate a haploid species with well-characterized selection signals , we performed IBD mapping of 2 , 550 P . falciparum isolates from 14 countries across Africa , Southeast Asia and Papua New Guinea as part of the MalariaGEN Pf3K dataset . The samples in this dataset were collected during the years 2001 to 2014 ( S3 Table ) and details of the collection process and sequencing protocols have been described elsewhere [14 , 16] . We define within-country analyses as all pairwise IBD comparisons between isolates from the same country ( 14 analyses in total ) while between-country analyses as all pairwise-country comparisons ( 91 analyses in total ) where pairs of isolates contain one isolate from each country . 2 , 377 isolates remained after filtering , with 994 isolates ( 42% ) classified as having multiple infections ( S3 and S4 Tables ) . The mean number of SNPs remaining post filtering for within-country analyses was 31 , 018 SNPs , with the least number of SNPs in Papua New Guinea isolates ( 18 , 270 SNPs ) and the largest number of SNPs in Guinea ( Africa ) isolates ( 44 , 528 SNPs ) ( S3 Table ) . SNPs for between-country analyses were selected if they were present in both countries and if their minor allele frequencies differed by less than 30% . This criterion resulted in the inclusion of at least 75% of SNPs present in both populations , where on average 12 , 271 SNPs remained per analysis . The smallest number of SNPs was in the analysis between Mali and Papua New Guinea ( 1 , 945 SNPs ) , while the largest number of SNPs was in the analysis between Guinea and Malawi ( 29 , 138 SNPs ) ( S5 Table ) . These highly varying numbers of informative SNPs largely reflect geographical isolation and population structure [22 , 23] , but are also influenced by the quality of the WGS data , with poorer quality sequencing leading to fewer SNPs . Analyses with so few SNPs , such as Mali and Papua New Guinea , are less likely to detect selection signatures since smaller IBD segments will fail to be detected , however are still useful for identifying closely related isolates that are expected to share large IBD segments over many SNPs . We calculated the proportion of pairs IBD at each SNP and investigated the distributions of these statistics across the genome ( Fig 4 , S6 Table , S6 Fig ) . We identified higher levels of relatedness in Southeast Asia than in Africa or in Papua New Guinea , with isolates from Cambodia displaying the highest average sharing across the genome ( 5% , calculated as the mean proportion of pairs IBD genome-wide ) , reflecting high background relatedness . The Cambodian dataset consists of isolates collected from four study locations; therefore we stratified the relatedness proportions by study location to identify sites with extremely high amounts of relatedness . We detected high relatedness between 87% ( 2 , 890/3 , 321 ) of pairs from the Pailin Province of Cambodia , with on average 29% of pairs IBD per SNP ( S7 and S8 Tables , S7 and S8 Figs ) . This reflects an extremely high number of closely-related isolates , i . e . clones and siblings . Isolates from Pailin make up 16% of the Cambodian dataset and inflate the overall signal seen in Cambodia . We also detected high amounts of relatedness , including many clonal isolates , in the Thai Province of Sisakhet , which borders Cambodia , reflecting similar transmission dynamics between regions in close proximity . Relatedness-networks can be created using clustering techniques to identify groups of isolates sharing a common haplotype . We constructed a relatedness-network to investigate clusters of isolates sharing near-identical genomes , reflecting identical infections or ‘duplicate’ samples ( Fig 5 ) . Southeast Asia has a number of large clusters containing highly related isolates with the five largest clusters belonging to Cambodia , containing between 12 and 68 isolates , indicative of clonal expansions . The largest cluster contains mostly isolates from the Pursat Province of Cambodia , however the remaining isolates are from the Pailin Province and the Ratanakiri Province of Cambodia , suggesting common haplotypes between western and eastern Cambodia ( S9 Fig ) . In contrast , we did not find any isolates within Guinea or Mali to be highly related , nor did we find isolates from different countries to be highly related ( S9 Table , S10 and S11 Figs ) . The genome-wide distributions of the proportion of pairs IBD can identify genomic regions with high amounts of sharing that may be under positive selection . This has been previously demonstrated for IBD studies in human populations [5 , 6] . To assess the significance of our selection signals , based on the composite IBD , we calculate the genome-wide distributions of the–log10 p-values for within-country analyses ( Fig 6 ) and report the top five signals of selection for each country in S10 Table . Q-Q plots are displayed in S12 Fig . We observe signals of selection over several known P . falciparum antimalarial drug resistance genes such as Pfcrt ( chloroquine resistance transporter ) and Pfdhfr ( dihydrofolate reductase ) in addition to several regions suspected of being associated with antimalarial drug resistance ( chr6:1 , 102 , 005–1 , 283 , 312; chr12:700 , 000–1 , 100 , 000 ) . Many of these signals also show substantial continent and/or country variation . Below , we examine the selection signals overlapping two known P . falciparum antimalarial resistance genes , Pfcrt and Pfk13 ( kelch 13 ) , as well as the signals seen on chromosome 6 and chromosome 12 , to demonstrate the interpretive possibilities of IBD signals obtained with isoRelate . We explored selection signatures to determine if the haplotypes under selection at multiple loci were jointly inherited in some pairs of isolates . Specifically , we investigated whether haplotypes associated with antimalarial drug resistance at two loci were jointly inherited . We investigated the P . falciparum multidrug resistance gene 1 ( Pfmdr1 ) , located on chromosome 5: 957 , 890–962 , 149 , which has been associated with chloroquine resistance and amodiaquine resistance when the Pfmdr1 N86Y mutation is present along with the Pfcrt K76T mutation [38] . Fig 11 displays genome-wide selection signals in Ghana , stratified by pairs that are IBD over Pfmdr1 and pairs that are not IBD over Pfmdr1 . A significant signal of selection is observed over Pfcrt in both stratified groups , suggesting Pfcrt is under selection jointly with Pfmdr1 as well as independently of Pfmdr1 . Of the isolate pairs that are IBD over Pfmdr1 , 13% are also IBD over Pfcrt while 6% are IBD over Pfcrt and carry both the N86Y mutation and the K76T mutation . The median proportion of genome inferred IBD between these pairs is 1% , alleviating concerns that joint inheritance of both variants is due to highly related pairs . A smaller signal is observed over Pfmspdbl2 in both groups . Increased copy number of Pfmspdbl2 has been associated with decreased sensitivity to halofantrine , mefloquine and lumefantrine [39 , 40] and we find copy number alterations to be present also . In particular , copy number alterations of Pfmspdbl2 are present in isolates carrying the N86Y variant that are IBD over both Pfmdr1 and Pfmspdbl2 . Increased relatedness over Pfmspdbl2 in isolates that share haplotypes over Pfmdr1 suggests that these isolates may be multidrug resistant .
We have presented here a method that identifies recent IBD sharing between pairs of haploid microorganisms in the presence of multiclonal infections . We explored the power and accuracy of our method , isoRelate , in two comprehensive simulation studies , where we investigated our ability to detect IBD segments of various sizes in the presence of multiple infections as well as our ability to detect complex patterns of positive selection . Here we showed that IBD segments of 2cM or larger are detectable with isoRelate , however as MOI increases and the dominant clonal proportion decreases , the power to detect IBD segments naturally decreases . This is due to added heterozygosity in the isolate and largely reflects the ability of the genotyping algorithm to capture multiple haplotypes . Additionally , for species like P . falciparum , where the allele frequency spectrum is heavily skewed , IBD performance is compromised , however isoRelate is still powerful for detecting segments of 4cM or larger . When assessing the performance of isoRelate at detecting complex patterns of positive selection , sweeps were detectable up to 200 generations after their introduction . Given that isoRelate is designed to infer recent common ancestry , it follows that more recent signals of positive selection are identified . However , sweeps were only detected in our simulations when the selection coefficient was sufficiently large ( s ≥ 0 . 1 ) , which is true for assessments of iHS and haploPS also . Unlike iHS and haploPS , our method accounts for the amount of relatedness between isolates and does not require phased data . In the Pf3k dataset analysed here , more than 40% of isolates were multiclonal and would typically be excluded from analysis with both iHS and haploPS , whilst isoRelate was able to use all isolates . Including isolates with MOI > 1 in analyses can provide useful insights as to the spread of haplotypes between geographical regions . For example , an isolate with MOI = 2 may consist of two genetically distinct haplotypes that each originate from different villages . This could occur if the infected individual travels between villages , potentially introducing new haplotypes into the exiting parasite populations . Haplotype spread such as this can be visualized using relatedness networks and one instance of this can be seen in the Cambodian dataset ( S9 Fig ) . Here , an isolate with MOI > 1 from the Pailin Province is highly related to two , otherwise unrelated , clusters of isolates from the Pursat Province and Pailin Province of Cambodia . While the ability to include isolates with MOI > 1 is informative , analyses are generally more powerful with MOI = 1 isolates . To this end , Zhu et al . [41] has recently developed a statistical framework , DEploid , for deconvolving multiclonal isolates which would enable analysis of individual clones within an isolate . However , DEploid requires a reference panel of single-clone isolates as a proxy for the population of interest and has not yet been tested on large cohorts like the Pf3k dataset . Similarly , one limitation of isoRelate is that HMMs are computationally intensive algorithms , where the computational time increases linearly with the number of SNPs and quadratically with the number of isolates . An IBD analysis of isolates from Malawi ( 357 isolates; 40 , 225 SNPs; 63 , 546 pairwise analyses ) takes approximately 8 hours on a single-core processor , while the analysis of isolates from Guinea ( 100 isolates; 44 , 528 SNPs; 4 , 950 pairwise analyses ) takes less than 1 hour . Additionally , the computational time for MOI > 1 isolates is longer than for MOI = 1 isolates as the observation state space is larger , resulting in more genotypic combinations to account for . However , isoRelate allows for parallelization of analyses on multicore processors , which will considerably reduce the computation time . The ability of isoRelate to detect IBD segments also depends on the quality of the data . Most genetic datasets will contain a small number of genotyping errors and missing genotype calls , which can result in incorrect IBD inference and/or reduced performance . However , IBD inference has been shown to be robust to both missing data [6] and genotyping errors [42] , within reason . Furthermore , IBD analyses require several criteria to be met . This includes the availability of a good quality reference genome and the fact that the organism must recombine as one of its main sources of genetic variation . As such these methods do not appear to be applicable to Mycobacterium tuberculosis for example , but should work , at least theoretically , with any other organism that shares these criteria with P . falciparum . Amongst these are P . vivax [43] and some species of Staphylococcus [44] . Moreover , isoRelate can be applied to any dense genomic data that produces SNP genotypes , which includes WGS , RNA sequencing and SNP arrays . Additional downstream analyses can be performed with IBD estimates , whereby IBD patterns could be tested for associations with important epidemiological variables such as occupation and exposure to mosquitoes . This could be performed in a multivariate normal modeling framework such as that employed by the SOLAR package [45] , where the IBD of the human host is replaced with the IBD of the sampled isolates from the host . Furthermore , IBD mapping has the potential to track emerging drug resistance and , for diseases that experience relapse infections such as malaria caused by Plasmodium vivax , may be able to distinguish between new or relapsing infections in drug efficacy and cohort studies , though these applications have yet to be explored .
We applied the Fws metric to within-country SNP sets to determine isolates that had multiple infections [16] . An isolate was classified as having multiple infections if Fws< 0 . 95 . For each country PED and MAP files for downstream analysis were extracted using moimix [50] . Heterozygous SNP calls were retained for isolates assigned as having MOI greater than 1 , otherwise heterozygous SNPs were set to having a missing value at those SNPs to signify the likelihood of a genotyping error . We extend a first order hidden Markov model ( HMM ) that detects IBD segments between pairs of human samples to allow detection of IBD between pairs of non-human , haploid samples [7] . The assumption of a first order HMM is unlikely to hold in the presence of dense datasets containing linkage disequilibrium , however we do not consider this to be an issue with P . falciparum due to the short LD segments in its genome [23 , 51] . Furthermore , false positive IBD segments due to LD tend to be much smaller than true IBD segments and are filtered out with length-based filtering criterion . Genotype calls are used to determine the number of alleles shared IBD at each SNP between a pair of isolates . The potential number of shared alleles at a SNP defines the state space in the HMM and is dependent on the MOI of the pair under consideration . An isolate with MOI = 1 consists of a single strain and is analyzed as if it were haploid; thus sharing either 0 or 1 allele IBD with any other isolate . An isolate with MOI > 1 consists of multiple genetically distinct ( and possibly related ) strains , and is considered diploid; sharing 0 , 1 or at most 2 alleles IBD with other isolates . Here , the ability of our model to detect IBD in the minor clone of an isolate with MOI > 2 will depend on the clonal frequency and the genotyping algorithms ability to capture SNP variation at that frequency . Initial probabilities , emission probabilities and transition probabilities are calculated as in Henden et al . [7] and are described in the S1 Methods . We model an error rate in the calculation of the emission probabilities , which could reflect either a genotyping error or a mutation , where a larger error rate is likely to result in more IBD detected . Both the initial probabilities and the emission probabilities require population allele frequencies . For the simulation study , whereby the performance of isoRelate is assessed , we compute the allele frequencies for each dataset of MOI separately . Additionally , for the analysis of the Pf3k dataset , we compute these frequencies for each country separately . This is necessary due to the highly divergent sets of SNPs observed in P . falciparum globally [52] . To perform IBD analyses between isolate from different countries , SNPs were included in the analysis if the population allele frequencies between the pair of countries differed by less than 0 . 3 . A MAF concordance threshold of 0 . 3 was arbitrarily used in the analysis as this threshold resulted in the inclusion of at least 75% of SNPs present in both populations , for all pairwise-population analyses . Population allele frequencies for the combined countries were then calculated using all isolates from pairs of countries being examined . SNPs with MAF less than 1% were removed from the analysis along with SNPs with missing genotype data for more than 10% of isolates . Similarly , isolates with missing genotype data for more than 10% of SNPs were removed and a genotyping error rate of 1% was included in the model . S3 and S5 Tables give the number of isolates and SNPs before and after filtering for each country and pairwise-country dataset . IBD segments are reported based on the results from the Viterbi algorithm [53] and segments that contain less than 20 SNPs or have lengths less than 50 , 000bp are excluded , as they are likely to represent distant population sharing that is not relevant to recent selection . In the Pf3k analysis , IBD analyses were performed between all pairs of isolates that remained once filtering procedures had been applied . The algorithm has been developed as an R package , isoRelate , and can be downloaded from https://github . com/bahlolab/isoRelate . Using normalisation procedures previously applied in algorithms such as EIGENSTRAT [54] we derived a test statistic that approximately followed a normal distribution and which could thus be interpreted probabilistically using distributional assumptions , rather than resorting to computationally demanding permutation tests . To calculate the test statistic we first created a matrix of binary IBD status with rows corresponding to SNPs and columns corresponding to isolate pairs . For each column , we subtract the column mean from all rows to account for the amount of relatedness between each pair . Following this we subtract the row mean from each row and divide by the square root of pi ( 1-pi ) , where pi is the population allele frequency of SNP i . This adjusts for differences in SNP allele frequencies , which can affect the ability to detect IBD . Next we calculate row sums and divide these values by the square root of the number of pairs . These summary statistics are normalized genome-wise by binning all SNPs into 100 equally sized bins partitioned on allele frequencies and then we subtracted the mean and divided by the standard deviation of all values within each bin . Negative z-scores are difficult to interpret when investigating positive selection; therefore we square the z-scores such that the new summary statistics follow a chi-squared distribution with 1 degree of freedom . This produces a set of genome wide test statistics ( XiR , s ) , where XiR , s is the chisquare distributed test statistic for IBD sharing from isoRelate at SNP s . We calculate p-values for ( XiR , s ) , after which we perform a–log10 transformation of the p-values to produce our final summary statistics , used to investigate the significance of selection signatures . Finally , a 5% genome-wide significance threshold was used to assess evidence of positive selection . We performed a standard analysis of selection signals using the scikit-allel v0 . 201 . 1 package in Python 2 . 7 [55 , 56] . To compute selection statistics on simulated data , we calculated the integrated haplotype score ( iHS ) for SNPs passing a MAF filter of 1% [17] . We note that SNPs were removed from analysis if they were not in a core region of the genome as defined by Miles et al . [15] . We report the iHS if the EHH decays to 0 . 05 before reaching the final SNP examined within a maximum gap distance of 2 Mb spanning the EHH region , otherwise iHS was set to missing . To standardize iHS we binned all SNPs into 100 equally sized bins partitioned on allele frequencies and then subtracted the mean and divided by the standard deviation of iHS within that bin . We computed log10 p-values using the normalized iHS from a standard normal distribution . To detect selection using haploPS [18] , SNPs passing a MAF filter of 1% that were in core regions of the genome were analysed . We first calculated the adjusted haploPS score for haplotypes identified at core frequencies of 5% to 95% in increments of 5% . This score is calculated by comparing the lengths of the identified haplotypes to the lengths of other haplotypes that are present as similar frequencies in the dataset . Regions were considered to be under positive selection if the adjusted haplotype score was less than 0 . 05 . Since haplotypes are identified across multiple core frequencies , similar regions of positive selection are detected across these frequencies . We stacked the significant haplotypes around each SNP , identified across the different core frequencies , and calculated the number of significant haplotypes that overlap each SNP . Regions that have undergone strong positive selection in the form of a hard sweep will typically be inferred as positively selected across multiple core frequencies , therefore the number of significant haplotypes that overlap each SNP within these regions should be larger than those in regions that have not undergone selection . Since a large number of analyses were carried out ( 10 replications for each of the 15 scenarios of sweeps , with haplotypes sampled at 4 time points following selection ) , results were summarised as follows . For isoRelate and iHS , we calculated the genetic distance between the SNP with the largest–log10 p-value and the selected allele . While for haploPS we calculated the distance between the selected allele and the SNP with the most number of significant haplotypes inferred across the core frequencies . Boxplots were created for each combination of scenarios from the 10 replications . Boxplots centered around zero with a small interquartile range are indicative of a sweep being consistently detected , and a method performing well . To examine the haplotype sharing between isolates within and between countries , both as genome-wide averages and at a regional level , we generated relatedness networks using the R package igraph [57] . Each node in the network represents a unique isolate and an edge is drawn between two nodes if the isolates are IBD anywhere within the interval for the regional investigations ( Figs 4–7 ) and if the isolates share more than 90% of their genome IBD for the genome-wide analyses ( Fig 2 ) . Isolates with MOI = 1 are represented by circle nodes while isolates with MOI > 1 are represented by squares . Node colors are unique for isolates from different countries . To investigate multidrug resistance we extract all pairs who are IBD over a drug resistant gene of interest . Here a pair is classified as IBD if they have an IBD segment that partially or completely overlaps the specified interval . From this subset of pairs we calculate our selection signal as per usual and investigate the distribution of these statistics across the genome . All selection signatures that reach significance provide evidence of co-inheritance and thus mutual-selection in these pairs . Therefore we examine joint selection of an antimalarial drug resistant gene with other drug resistant genes for evidence of multidrug resistance . | There are growing concerns over the emergence of antimicrobial drug resistance , which threatens the efficacy of treatments for infectious diseases such as malaria . As such , it is important to understand the dynamics of resistance by investigating population structure , natural selection and disease transmission in microorganisms . The study of disease dynamics has been hampered by the lack of suitable statistical models for analysis of isolates containing multiple infections . We introduce a statistical model that uses population genomic data to identify genomic regions ( loci ) that are inherited from a common ancestor , in the presence of multiple infections . We demonstrate its potential for biological discovery using a global Plasmodium falciparum dataset . We identify low genetic diversity in isolates from Southeast Asia , possibly from clonal expansion following intensified control efforts after the emergence of artemisinin resistance . We also identify loci under positive selection , most of which contain genes that have been associated with antimalarial drug resistance . We discover two loci under strong selection in multiple countries throughout Southeast Asia and Africa where the selection pressure is currently unknown . We find that the selection pressure at one of these loci has originated from gene flow , while the other loci has originated from multiple independent events . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"parasite",
"groups",
"plasmodium",
"drugs",
"microbiology",
"parasitology",
"genetic",
"mapping",
"antimalarials",
"apicomplexa",
"genome",
"analysis",
"pharmacology",
"molecular",
"genetics",
"genomic",
"signal",
"processing",
"antimicrobial",
"resistance",
"genomics",
"molecular",
"biology",
"signal",
"transduction",
"haplotypes",
"cell",
"biology",
"heredity",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"genomics",
"statistics",
"cell",
"signaling"
] | 2018 | Identity-by-descent analyses for measuring population dynamics and selection in recombining pathogens |
Mouse APOBEC3 ( mA3 ) is a cytidine deaminase with antiviral activity . mA3 is linked to the Rfv3 virus resistance factor , a gene responsible for recovery from infection by Friend murine leukemia virus , and mA3 allelic variants differ in their ability to restrict mouse mammary tumor virus . We sequenced mA3 genes from 38 inbred strains and wild mouse species , and compared the mouse sequence and predicted structure with human APOBEC3G ( hA3G ) . An inserted sequence was identified in the virus restrictive C57BL strain allele that disrupts a splice donor site . This insertion represents the long terminal repeat of the xenotropic mouse gammaretrovirus , and was acquired in Eurasian mice that harbor xenotropic retrovirus . This viral regulatory sequence does not alter splicing but is associated with elevated mA3 expression levels in spleens of laboratory and wild-derived mice . Analysis of Mus mA3 coding sequences produced evidence of positive selection and identified 10 codons with very high posterior probabilities of having evolved under positive selection . Six of these codons lie in two clusters in the N-terminal catalytically active cytidine deaminase domain ( CDA ) , and 5 of those 6 codons are polymorphic in Rfv3 virus restrictive and nonrestrictive mice and align with hA3G CDA codons that are critical for deaminase activity . Homology models of mA3 indicate that the two selected codon clusters specify residues that are opposite each other along the predicted CDA active site groove , and that one cluster corresponds to an hAPOBEC substrate recognition loop . Substitutions at these clustered mA3 codons alter antiviral activity . This analysis suggests that mA3 has been under positive selection throughout Mus evolution , and identified an inserted retroviral regulatory sequence associated with enhanced expression in virus resistant mice and specific residues that modulate antiviral activity .
Species susceptible to infectious retroviruses have evolved numerous constitutively expressed antiviral factors that target various stages of the retroviral life cycle . The factors responsible for this intrinsic immunity include 3 that act at post-entry stages of virus replication: Fv1 , APOBEC3 and TRIM5α . Fv1 was discovered in mice , [1] and only mice carry Fv1 [2] , [3] . TRIM5α was initially identified in primates as an anti-HIV-1 restriction factor [4] , [5] , and while mice carry TRIM5α related sequences [6] , no mouse orthologue with virus restriction activity has been identified . Active APOBEC3 genes , on the other hand , are found in various species including human and mouse , and mouse and human APOBEC3 have antiviral activity against multiple retroviruses [reviewed in 7] . The APOBEC3 editing enzyme is incorporated into budding virions . During reverse transcription in subsequently infected cells , the virion-associated APOBEC3 catalyzes C-to-U deamination , resulting in G-to-A mutations in the viral DNA [8] . The increased mutational load has a major impact on viral fitness , and there is also some evidence that APOBEC3 antiviral activity is enhanced by additional deamination-independent mechanisms that act before proviral integration [9] , [10] . APOBEC3 was initially described in primates , and human APOBEC3 paralogues responsible for resistance are present as a cluster of 7 genes on chromosome 22 , the most extensively studied of which is APOBEC3G ( hA3G ) . HIV-1 can avoid inhibition by hA3G through the action of one of its viral accessory proteins , Vif ( viral infectivity factor ) , that prevents incorporation of hA3G into the virion [11] . The antiviral activity of hA3G can be observed with Vif-negative HIV-1 and SIV lentiviruses as well as other retroviruses such as equine infectious anemia virus ( EIAV ) and mouse leukemia viruses ( MLVs ) . In the mouse , there is only a single APOBEC3 copy ( mA3 ) on chromosome 15 . Several observations indicate that mA3 functions in antiviral defense: mA3 inhibits infection by several viruses including HIV-1 and mouse retroviruses such as mouse mammary tumor virus ( MMTV ) , intracisternal A-particles ( IAPs ) and MusD endogenous retroviruses [12]–[14]; mA3 knockout mice are more susceptible to MMTV infection and tumorigenesis [15]; endogenous retroviruses ( ERVs ) of MLV in the sequenced Mus genome show modifications consistent with APOBEC3 activity [16] . Two recent studies proposed that mA3 is responsible for the Friend virus resistance factor Rfv3 [10] , [17] . Rfv3 is one of several host resistance factors that , like Fv1 , were discovered in studies with the pathogenic Friend MLV ( FrMLV ) [18] . Rfv3 was identified as a non-major histocompatibility complex gene that influences the duration of viremia , partly through its effects on the production of virus-neutralizing antibodies [19] . The prototype Rfv3 virus restrictive strain is C57BL , and BALB/c is the prototype non-restrictive strain . The Rfv3 gene map location on chromosome 15 [20] has now been linked to the locus of Apobec3 [10] , [17] . That mA3 is responsible for Rfv3 resistance is supported by the observations that mA3 of C57BL restricts FrMLV replication and FrMLV-induced disease more effectively than BALB/c mA3 , and that genetic inactivation of mA3 generates an FrMLV susceptible phenotype [10] , [17] . It has also been shown that the C57BL mA3 allelic variant is more effective than the BALB/c allele in restricting MMTV [21] . The mA3 genes in prototype Rfv3 restrictive and nonrestrictive strains differ in protein sequence , splicing pattern , and expression level , and all three of these factors may contribute to resistance [10] , [17] , [21] . Few strains and Mus species have been characterized for these differences [21] , so we sequenced mA3 genes from various inbred strains and wild mice representative of the major taxonomic groups of Mus . In this paper , we demonstrate that an MLV long terminal repeat ( LTR ) disrupts a splice donor site in the mA3 of C57BL and other strains and species and is associated with altered expression levels , we demonstrate strong positive selection of this gene in Mus that involves sites that distinguish the mA3 genes of Rfv3 virus resistant and susceptible mice , we use homology modeling to position the positively selected residues in two clusters on opposite sides of the putative active site groove , and we describe the antiviral activity of mA3 genes carrying mutations at these sites .
Analysis of the antiviral activities of chimeric and wild type C57BL and BALB/c mA3s by Takeda and colleagues [10] indicated that the mA3 anti-FrMLV activity resides in the N-terminal half of the C57BL protein . This 194 amino acid residue segment contains the active Z2-type cytidine deaminase region ( CDA ) [22] , [23] , and the translated protein sequences of restrictive C57BL and nonrestrictive BALB/c prototypes differ from one another in this region at nine residues [10] . To determine the distribution of the restrictive variant among mice and to identify novel variants , we sequenced segments of mA3 containing these 9 residues from inbred strains and wild-derived mice representing different taxa and/or mice trapped in different geographic locations ( Table S1 ) ( Figure 1A ) . In the course of this analysis , we identified a 531 bp sequence inserted into the intron of mA3 of some laboratory strains between exons 2 and 3 ( Figure 1A , 1B ) . The insertion was sequenced and identified as an intact retroviral LTR ( Figure 1C ) . This LTR is 96 . 6% identical to the LTR of the xenotropic gammaretrovirus ( X-MLV ) NZB-IU-6 , an MLV isolated from NZB strain mice [24] , [25] . The mA3 LTR insert shows the expected direct repeats characteristic of retroviral insertions , CAT and TATA boxes , and a comparable enhancer region . The LTR is inserted in an antisense orientation , and the site of insertion is the splice donor site at the end of exon 2 ( Figure 1D ) . Part of the splice donor site contributes to the direct repeat flanking the insertion . The insertion alters the last base of the splice donor site , a position that is not highly constrained in the consensus sequence . We screened 32 laboratory mouse strains for presence of this LTR insertion by PCR ( Figure 2 ) . The insertion was identified in 6 strains , including C57BL and the 3 related strains NZB , NZL and NZO . The LTR was absent from other NZB-related strains , from other strains in the C57/C58 series and was also absent from 21 strains from other families of inbred strains . The sequences of exons 2–4 of 13 strains were compared , and the only strains identified as having the C57BL/6 coding sequence , NZB and RF , also carried the LTR insertion . ( Figure 2 ) . The common inbred strains of mice are a mosaic of Eastern European M . m . musculus , Western European M . m . domesticus and Asian M . m . castaneus [26] , [27] . Therefore we looked for the sequence polymorphisms associated with the C57BL allele and for the MLV LTR in M . musculus subspecies from breeding stocks established from mice trapped in Old World sites where these commensal ( house mouse ) subspecies originated , and from M . musculus mice trapped in the Americas where they had been introduced from Europe and Asia ( Figure 2 ) . Two wild-derived mice from the Delmarva ( Delaware-Maryland-Virginia ) Peninsula , CL and LEWES , had this LTR along with the C57BL mA3 coding sequence . PCR fragments diagnostic of the LTR insert were also found in other Maryland mice as well as in two mice trapped in California , one of three M . m . castaneus breeding lines , and three of four lines developed from mice trapped in the former Czechoslovakia . The LTRs sequenced in 4 laboratory strains and 5 wild-derived mice were 99% identical to one another , and the mA3 genes of the LTR+ wild mice had several substitutions compared to the C57BL gene . Thus , the LTR was acquired in Eurasian species , and these LTR modified mA3 genes continued to accumulate mutations after this insertion event . Previous reports had determined that mA3 mRNAs can lack exon 5 [10] , [13] , [14] , and that BALB/c mA3 can also lack exon 2 [10] . We examined 31 mA3 mRNAs from cultured cells or tissues of 24 different inbred strains and wild-derived M . musculus mice for these splice variants by RT-PCR ( Figure 3 ) . mA3 mRNAs isolated from different tissues of the same mouse produced the same pattern of PCR products . Eleven of these 24 mice carry the LTR ( Figure 3B ) , and all 11 mice produced a single PCR product of the size expected for a spliced message lacking exon 5 ( Figure 3A ) . Among the 13 LTR-free mice , two , M . m . molossinus and the LTR− inbred MOLD/RkJ line of this subspecies , produced this same single isoform , while the other 11 LTR− mice additionally produced an exon5+ message that in 10 mice was significantly more abundant than the Δexon5 isoform ( Figure 3A ) . Both sequenced BALB 3T3 mA3s lacked exons 2 and 5 , and a third barely detectable smaller PCR product was observed in BALB 3T3 and other LTR− mice of the size consistent with the absence of exons 2 and 5 ( Figure 3A ) . The distribution of the MLV LTR among these mice suggests that the LTR was inserted into the mA3 variant that produces the Δexon5 isoform . Previous reports had noted that mA3 expression level is significantly higher in C57BL mouse tissues ( LTR+ ) than in BALB/c ( LTR− ) [10] , [21] . We isolated total RNA from the spleens of 11 mice that had been typed for the LTR and for mA3 splicing patterns . Included were mice from 2 breeding lines of M . m . molossinus , the inbred MOLD/RkJ strain and a mouse from a random bred colony , both of which are LTR− and produce the Δexon5 isoform ( Figure 3B ) . Quantitative real-time PCR analysis showed that the 7 LTR+ mice produced 4–20 fold higher levels of mA3 mRNA than did the 4 LTR− mice , including the two M . m . molossinus mice ( Figure 3C ) . These data demonstrate a correlation between the LTR and expression level but not splicing pattern . We used sequenced segments of mA3 from 4 inbred strains and 21 wild-derived mouse species and subspecies for phylogenetic analysis . The sequences were used to construct phylogenies , and were analyzed with the PAML suite of programs [28] for evidence of adaptive evolution and to identify possible sites of positive selection . Two sets of DNA sequences were analyzed separately: exons 2–4 amplified from genomic DNA or RNA and a set of 8 near full length DNAs generated by RT-PCR ( Text S1 , S2 ) . The sequences in the smaller dataset of 8 DNAs do not include the extreme 5′ and 3′ends of the gene or exon 5 which was absent from all but 3 of the 8 sequenced mRNAs . The sequences were used to construct neighbor-joining trees ( based on Kimura 2-parameter distances ) for the near full-length sequences ( Figure S1A ) and for the 2–4 exon set ( Figure 4A ) . Modifications to the trees were made based on generally accepted phylogenetic trees [29] , [30] . The data-based and taxonomy-based trees were both used for PAML analysis and produced nearly identical statistics ( Tables S2 , S3 ) . Values of dN/dS along each tree branch were calculated using the free-ratio model of PAML . A dN/dS value >1 suggests that positive selection has acted along that lineage . Several branches of the trees show evidence of positive selection with dN/dS>1 , or , when dS = 0 , by the identification of 4 or more replacement substitutions . Likelihood ratio tests indicate that mA3 has a significant probability of having experienced positive selection , and this was the case for all codon frequency models , and for both datasets ( Figures 4B and S1B , Tables S2 , S3 ) . The Bayes empirical Bayes calculation of posterior probabilities in PAML identified specific mA3 codon positions as having significant probability of positive selection . In the separate analyses of the two datasets , we identified 20 codons as being under positive selection with high posterior probability P>0 . 95 , and 10 of these 20 codons were under very strong positive selection with P>0 . 99 ( Tables 1 , S2 , S3 ) . Sixteen of these 20 codons are in exons 2–4 . Analysis of the smaller set of 8 near full-length genes identified a subset of the positively selected codons identified by analysis of exons 2–4 . The full-length sequence analysis also identified 5 additional codons under positive selection with P>0 . 95 that were not identified in the exon 2–4 analysis: one codon , 142 , in exon 3 of the active CDA and four codons , 201 , 273 , 316 and 371 , in the inactive C-terminal CDA ( Tables 1 , S3 ) . There are 15 mA3 codons that specify different amino acids in virus restrictive C57BL and sensitive BALB/c mice . Eleven of these codons were found to be under positive selection ( P>0 . 95 ) , and 5 of the codons under very strong positive selection ( P>0 . 99 ) mapped to two clusters in the active CDA ( Figure 5 ) . Because this type of analysis is designed to identify sites involved in diversifying selection ( antagonistic interactions with pathogens being a prime example ) , our results indicate that most of the residues that distinguish C57BL and BALB/c mice identify key sites likely to be involved in genetic conflicts . These results also suggest that mA3 has had a defensive role that predates development of the laboratory strains and involves species in all 4 Mus subgenera . Homology models for the C57BL N-terminal active CDA sequences were chosen from the LOMETS homology modeling program based on templates that had the highest sequence identity . The search identified several templates with highest confidence , crystal structures determined for the catalytic domain of hA3G ( PDB ID 3IR2 ) [31] and ( PDB ID 3IQS , 3E1U ) [32] . The hA3G-3IR2 template model , based on the active hA3G C-terminal Z1 deaminase domain , was chosen for detailed analysis because it provides more coverage of the N-terminal Z2 domain of the mouse sequence [23] , [31] , and because it was the top LOMETS solution overall . The C57BL mA3 CDA sequence has 36 . 4% identity to the hA3G CDA ( Figure 6A ) . Superposition of the hA3G-3IR2 crystal structure and the mouse homology model show they share the 5 stranded β-sheet core surrounded by 6 α-helices that is common to known deaminase structures , along with a conservation of active-site loops involved in substrate binding ( Figure 6B ) . The sidechain conformations of the C57BL residues involved in coordinating Zn are identical to their counterparts in the hA3G structure ( Figure 6A , 6B ) . The overall fold between the human and mouse structures is nearly the same with the RMSD ( root mean square deviation ) between backbone atoms of the C57BL mA3 model and the human structure being 0 . 56Å . The RMSD between all atoms for the mouse model and the human structure is 0 . 94Å . Mutagenesis , NMR DNA titration data and structural analysis of hA3G-3E1U and the NMR structure hA3G-2JYW have identified key residues important in deaminase activity and formation of the substrate groove [32]–[34] . Among these key hA3G sites are the catalytic E259 , 3 hydrophobic residues and 10 critical residues of which 9 are charged , all of which are within and brimming the groove and all of which are needed for deaminase activity ( Figure 6A ) . N244 and R256 are associated with active center loop 3 ( AC loop 3 ) , R213 and R215 are present in active center loop 1 ( AC loop 1 ) , residue R313 resides on the floor of the groove and D316 , D317 , R320 face the substrate groove at or near the end of helix 4 . The most obvious difference between hA3G-3IR2 and the mouse model in these functionally important sites is the presence of an 8 residue deletion in the AC loop 3 of the mouse model . hA3G AC loop 3 is an unstructured loop , and the deletion of the majority of the residues in the mouse AC loop 3 suggests they play no critical role; the mouse AC loop 3 structure , however , does conserve the two residues found at the hA3G loop base , N244 and R256 , known to be critical for deamination [32] , and it is likely that these mouse residues , N66 and I70 , serve similar functions in mA3 . In contrast to this difference in AC loop 3 , the functionally important AC loop 1 and helix 4 residues in hA3G are retained in mA3 , and closely align with the two clusters of residues in mA3 shown here to be under positive selection ( Figure 6A , 6B , 6C ) . On the other side of the substrate groove from selected AC loop 1 residues 34–38 is the region encompassing residues 134–139 in C57BL ( and the corresponding region in hA3G ) ; these residues are at the end of helix 4 with some residues participating in the α-helix and the rest as a loop . A solvent accessible surface representation of the mA3 structure indicates the position of the predicted substrate groove , and suggests the location of the two clusters of positively selected residues on opposite sides of this substrate groove ( Figure 6D ) . The residues at the end of helix 4 and the residues in the 34–38 cluster on the other side of the mA3 groove likely serve steric roles in maintaining groove structure and likely also have functional roles based on charge and hydrophobicity that govern substrate interactions . 293T cells were cotransfected with the pLRB302 Friend virus clone and mA3 clones to assess the relative antiviral activities of 4 mA3 clones: the wild type Rfv3 virus resistant C57BL mA3 [13] and three clones with mutations that introduced residues of the Rfv3 virus sensitive BALB/c: M1 ( G34R , K37I , G38D ) , M2 ( V134I , Q135R , T139N ) , M3 ( all 6 substitutions ) ( Figure 7 ) . Cells and virus-containing supernatants were harvested 48 hours after transfection . Cells were analyzed by immunoblotting for mA3 expression , and infectious virus in the supernatants was quantitated by the XC overlay test . For each of the transfected mA3 clones , infectious virus titers decreased in a dose dependent manner relative to increasing expression of mA3 ( data not shown ) . The wild type C57BL mA3 and the BALB-like M3 mutant both showed antiviral activity , but the antiviral activity of M3 was reduced relative to wild type mA3 ( Figure 7 ) . The M1 mutant mA3 was found to reduce the infectivity of Friend virus as effectively as wild type C57BL mA3 , whereas M2 more closely resembled M3 in antiviral activity suggesting that substitutions in the 134–139 cluster are particularly important for anti-FrMLV activity .
This analysis indicates that mA3 has been involved in genetic conflicts through Mus evolution . This gene shows strong positive selection marked by an increase in replacement versus synonymous substitutions . Six of the 10 codons that evolved under strongest positive selection are in two clusters in the N-terminal catalytically active CDA . Five of these 6 codons specify different amino acids in MLV and MMTV restrictive and nonrestrictive mouse strains , and mutational analysis suggests these residues contribute to antiviral activity . We also demonstrate that the antiviral allelic variant has acquired a retroviral LTR insertion , the presence of which is associated with elevated mA3 expression levels in the spleens of inbred and wild-derived mice . Retroviral insertions can be important functional components of the host genome , and can clearly affect host gene expression . Examination of spontaneous mutations in the mouse suggested that 10–12% of all mutations are due to ERV insertions [35] . Like the mA3 LTR , most of these mutant-associated ERVs are in reverse orientation in introns , and the responsible mutational mechanisms include two of relevance here: aberrant splicing and enhanced transcription driven by the ERV LTR . While the mA3 LTR is inserted at a splice donor site , it does not alter splicing of the associated intron , and although all mice carrying this LTR produce the same Δexon5 mA3 isoform , the absence of this LTR in at least one mouse species producing that isoform ( M . m . molossinus ) suggests that the LTR was acquired by mice already preferentially producing this splice variant . As for LTR-driven altered expression levels , two of three previous studies that compared mA3 RNA levels in virus-resistant and susceptible strains reported that mA3 expression levels are significantly higher in mice carrying the LTR+ C57BL allele compared to LTR− BALB/c [21] , [17] , [10] . Our analysis of mA3 expression levels shows a correlation between the presence of the LTR and elevated expression in a variety of inbred strains and mouse species . Because enhancer activation of cellular genes by viral LTRs can occur with insertions in either orientation and at considerable distance from the cellular promoter , it is thus possible that the enhancer of this inserted LTR sequence drives the elevated expression observed in the LTR+ mice . This elevated expression in conjunction with altered splicing may together have contributed to the evolution of the antiviral C57BL mA3 . It has been suggested that the Δexon5 isoform has enhanced antiviral activity due to its resistance to the viral protease [36]; elevated expression of this variant due to subsequent LTR insertion would further boost the survival value of this factor . It is particularly intriguing that this X-MLV LTR sequence is found in NZB and CZECH mice and one breeding line of M . m . castaneus . These mice are unusual among laboratory strains and wild mice in that they harbor highly active X-MLV ERVs producing infectious virus , and such active ERV expression increases the likelihood of insertional mutagenesis . NZB mice are characterized by lifelong viremia with X-MLVs [37] . M . m . castaneus and CZECH mice are among wild mouse Eurasian populations with highest copy number of X-MLV ERVs [38] , and we have isolated infectious X-MLV-related virus from both of these wild mice [39] , [40] . If in fact the inserted MLV LTR causes elevated mA3 expression , then this would provide another instance of an ERV sequence that is co-opted by the virus-infected host for an antiviral function , other examples in the mouse being Fv1 , Fv4 , and Rmcf [41] . In addition to differences in splicing and expression levels , mA3 genes of virus resistant and sensitive mice differ in protein sequence . Our phylogenetic analysis showed that most of these polymorphic sites are under strong positive selection . The alignment of these sites with functionally important residues in the hA3G C-terminal active CDA suggests they serve similar roles in the mouse and that therefore , this function has been important during Mus evolution . That this evolutionarily important function is related to mA3 deaminase activity is supported by the observation that the great majority of these selected residues are in the N-terminal half of mA3 which encodes the active Z2 CDA [22] and that antiviral activity resides in the first 194 amino acids ( exons 1–4 ) [10] . In the predicted mA3 structure , these positively selected residues are positioned in one of two loops assigned functional importance in hA3G , AC loop 1 and a cluster of residues facing AC loop 1 on the other side of the putative substrate groove [32]–[34] . The charged and hydrophobic residues in these regions are positioned to maintain structural integrity of the groove and to interact with one another and the nucleic acid substrate in a way that could contribute to substrate specificity . Three positively selected residues , G34 , K37 and G38 , in the mA3 AC loop 1 sequence KNLGYAKGRKD are most likely responsible for providing conformational freedom ( in the case of the G34 and G38 ) and for interacting favorably with the phosphate backbone ( in the case of K37 ) . The electrostatic contributions of K37 along with K40 and D41 probably play an important role in determining substrate affinity and specificity while Y35 is in a position to stack with a nucleotide base . The analogous sequence in hA3G is NNEPWVRGRHE ( 207–217 ) with R213 , H216 and E217 positioned to interact electrostatically with a phosphate backbone and W211 able to stack with a nucleotide base . R39 ( mA3 ) and R215 ( hA3G ) are positioned similarly in that the residue provides an elaborate H-bonding network defining the shape of AC loop 1 [32] . Five positively selected residues ( V134 , Q135 , D136 , E138 and T139 ) lie in a region that comprises the end of helix 4 and an adjacent loop that define the side of the substrate binding groove opposite of AC loop 1 ( mA3 sequence YNVQDPET ) . Close inspection of this region in the mouse model reveals that the sidechain of D136 is in a position to H-bond with T139 maintaining the helical nature of helix 4 despite the presence of P137 . This has the result of allowing Q135 to form the top-side of the groove allowing V134 , N133 and Y132 to form the side of the groove with Y132 in position to stack with a nucleotide base . Y132 is invariant in our mouse sequences along with nearby W102 which defines the floor of the groove . The homologous segment of human APOBECs has now been implicated in the distinctive substrate preferences among AID/APOBEC family members which target cytosine within different sequence motifs . A recognition loop responsible for these preferences ( hA3G sequence IYDDQGRCQ ) lies between the β4 strand and the α4 helix ( Figure 6A , residues 314–322 ) [42] . That this highly variable region controls substrate preferences is also supported by mutational analysis [32] , [43] . Alignment of the active CDAs of hA3G and mA3 indicates that this loop overlaps the 134–139 cluster of positively selected residues in mA3 . This suggests that genetic conflicts between host and pathogen in this case produced positive selection that may be driven , not by protein-protein interactions , but by the interaction of mA3 and varying ssDNA substrates , a suggestion that is also consistent with the finding that the efficiency of substrate deamination is sensitive to ssDNA secondary structure [44] . Mutational analysis of 6 codons in the two clusters under positive selection showed that introduction of BALB/c residues , particularly in the 134–139 cluster , reduced antiviral activity against Friend MLV . Further studies may determine if the differences associated with overexpressed mA3 in transiently transfected cells have physiological relevance , and whether substitutions at these sites similarly affect restriction of other retroviruses . It has been reported that mA3 shows stronger antiviral activity against HIV-1 than against MLV [45] , suggesting that the genetic conflicts responsible for positive selection during Mus evolution may have resulted from interactions with pathogens unrelated to the FrMLV used here . Previous phylogenetic analysis of hA3G had identified 21 sites under very strong positive selection , 9 of which are in the active CDA [46] . One of these sites , R213 , aligns with one of the clusters of residues ( positions 34–38 ) under strong selection in mA3; however , the analysis of hA3G did not identify selection in the region aligning with the second cluster under strong selection in mA3 ( positions 134–139 ) , although this segment is a substrate recognition loop that is highly variable among members of the AID/APOBEC family [42] . The additional sites identified to be under positive selection in the hA3G active CDA have no positively selected counterparts in mA3 . Among these additional sites in hA3G , two , H248 and K249 , lie in AC loop 3 [46] . Mutagenesis and analysis of hA3G structure have implicated this loop in antiviral deamination [32] , but much of AC loop 3 is deleted in the mouse , leaving only the key residues at the base of this loop that align with critical residues N244 and R256 . The residues at these sites are invariant in our mA3 sequences suggesting their evolution is under purifying selection . The differences in AC loop 3 between hA3G and mA3 and the fact that different residues are under selection in hA3G and mA3 suggests there may be functional differences between these proteins . Our analysis of the full-length mA3 sequences also identified four sites under positive selection in the C-terminal half of the protein ( Tables 1 , S1 ) that carries the Z3 CDA that has been determined to be inactive [22] . It is not clear what role these residues serve . An antiviral role for the C-terminal half of mA3 is suggested by the observation that that the conserved glutamates in the N-terminal Z2 domain and the C-terminal Z3 domain of mA3 are both required for antiviral activity against HIV-1 [45] . Other evidence suggests that the inactive CDA is involved in virus encapsidation [47] . We note that alignment of the mouse Z2 and Z3 CDA regions shows that one of the two selected Z3 codons , P316 , aligns with the 134–139 selected cluster of codons in Z2 , VQDPET . Another selected codon in the Z3 CDA , T273 , aligns with an hA3G segment with two codons under selection in primates [46] . This suggests the possibility that this Z3 CDA may have had deaminase activity in some branches of the Mus lineage . Further analysis of the C57BL and BALB/c mA3 genes should shed light on the functional roles of the polymorphic residues in the two groove-associated clusters . The information from additional phylogenetic , structural , and functional comparisons will help describe the range of antiviral activity and evolutionary history of this gene . We are currently analyzing additional mA3 mutants for antiviral activity , and using molecular dynamics simulations to describe the structural implications of specific substitutions .
DNA and RNA were isolated from animals and cell lines developed from laboratory mouse strains and from wild mice and wild mouse-derived breeding colonies ( Table S1 ) . Many wild-derived mice were obtained from M . Potter ( NCI , Bethesda , MD ) . SAMP8 mice were provided by R . Carp ( New York State Institute for Basic Research in Developmental Disabilities , Staten Island , NY ) , SIM . S mice were obtained from E . Boyce ( Memorial Sloan-Kettering Cancer Center , NY ) , and mice trapped in California were provided by S . Rasheed ( University of Southern California , Los Angeles ) . Mice or DNA samples of M . spretus ( SPRET/EiJ ) , M . m . castaneus ( CAST/EiJ ) , various inbred lines derived from M . m . molossinus , PERA , PERC , PWD , and the inbred strains listed in Figure 2 were obtained from The Jackson Laboratory ( Bar Harbor , ME ) . A set of African pygmy mouse DNA samples was obtained from Y . Cole and P . D'Eustachio ( Depts . Biochemistry and Medicine , NYU , New York ) ; these mice had been classed into 4 species of subgenus Nannomys mice on the basis of skeletal features by J . T . Marshall ( Smithsonian Natural History Museum , Washington , DC ) . A sample of M . m . macedonicus DNA was provided by R . Elliott ( Roswell Park , Buffalo ) . Cell lines used as DNA and RNA sources included NZB-Q and M . fragilicauda cells obtained from J . Hartley ( NIAID , Bethesda , MD ) , cells from some wild mouse species obtained from J . Rodgers ( Baylor College of Medicine , Houston , TX ) , and NIH 3T3 , M . dunni [48] , SC-1 [49] , A9 ( C3H/He ) [50] , and CMT93 ( C57BL ) ( ATCC CCL-223 ) . All studies in which animals are involved were performed in accordance with the guidelines of the Committee on the Care and Use of Laboratory Animals under an NIAID-approved animal study protocol [51] , and all studies and procedures were reviewed and approved by the Institutional Animal Care and Use Committee of the NIH . APOBEC3 segments were amplified from mouse genomic DNAs or RNAs using primers designed from coding , flanking or intron sequences based on the C57BL genomic sequence ( GenBank No . NT_03921 ) ( Figure 1A ) . Exon 2 was amplified using forward intron primer a: 5′-CTCCTCTCCCTCTGTCTTCCT and reverse primer b: 5′-GGATTCAAGGTATGAGCCACCATGC . Exons 3 and 4 were amplified using primer c: 5′-GCTTCAACAGGGCTCAGAGTGC and primer d: 5′-AGGTTTGGGAGGAGGGAGAAC . Reverse transcription PCR ( RT-PCR ) was used to amplify near full-length APOBEC3 from total RNA using primer e in exon 1 ( 5′-GGACCATTCTGTCTGGGATGCAGCCATCG ) and primer f in exon 9 ( 5′-GACATCGGGGGACCAAGCTGTAGGTTTCC ) and a shorter RT-PCR fragment was generated using primer a and primer g ( 5′-GGTTGTAAAACTGCGAGTAAAATTCC ) . The larger RT-PCR product contained 1083 bp of the full-length 1287 bp mA3 sequence . Most of these products lacked the 99 bp exon 5 , and the aligned sequences lack 72 bp at the 5′ end and 33 bp at the 3′ end of the gene . PCR products were sequenced directly in some cases , and in others fragments were first cloned into pCR2 . 1-TOPO ( Invitrogen , Carlsbad , CA ) before sequencing ( Text S1 , S2 ) . Total RNAs from mouse spleens were isolated using Trizol ( Invitrogen ) . Reverse transcription was carried out at 50°C for 1 hour using 2 µg of total RNA in the presence of Oligo ( dT ) primer ( Ambion , Austin , TX ) and SuperScript III ( Invitrogen ) . After reverse transcription , the reaction mixtures were diluted to 1000 µl with DEPC-water . 1 µl of the diluted cDNA were added to a 15 µl PCR reaction mix containing 0 . 4 µl of 10 µM primers and 2× SYBR Green PCR mix ( Applied Biosystems , Foster City , CA ) . APOBEC3 transcripts were amplified using primers 5′-GACCATTCTGTCTGGGATGCA and 5′-TTCTAGTCACTTCATAGCACA . β-actin was also measured using primers ( 5′- GTGGGGCGCCCCAGGCACCA; 5′- CTCCTTAATGTCACGCACGATTTC ) as a normalization control . Amplification was done under the condition of 15 s at 95°C and 1 min at 60°C for 50 cycles in a 7300 Real Time PCR System ( Applied Biosystems ) . HA-tagged mA3 [13] was obtained from the NIH AIDS Research and Reference Reagent Program ( Germantown , MD ) ( catalog no . 10021 ) and mutagenized using the QuikChange mutagenesis kit ( Stratagene , La Jolla , CA ) to introduce substitutions at 6 codons . M1 ( G34R , K37I , G38D ) was generated using primer 5′-CCACTTTAAGAACCTACGCTATGCCATTGATCGGAAAGATACCTTC and its reverse complement . M2 ( V134I , Q135R , T139N ) was generated using primer 5′-GCTCCCGCCTCTACAACATCCGAGACCCAGAAAATCAGCAGAATCTTTGC and its reverse complement . M3 , containing mutations at all 6 codons , was generated by mutating M1 with the primers designed for M2 . Mutations were confirmed by sequencing . Attempts to generate stable transfectants of various mouse cells expressing these mA3 variants were not successful . Human 293T cells were co-transfected with 3–4 µg of the pLRB302 clone of Friend MLV [52] obtained from L . Evans ( RML , NIAID , Hamilton , MT ) , and 0 . 5 or 1 . 0 µg mA3 . At 48 hours after transfection , the culture supernatant was collected and virus infectivity was measured by the XC overlay test [53] . In this test , subconfluent cultures of NIH 3T3 cells were infected with virus dilutions , irradiated 4 days later and overlaid with rat XC cells . Infectivity was determined as plaque-forming units per ml of culture fluid . Infectivity was normalized against reverse transcriptase activity [54] or virus-associated capsid protein in pelleted virus . After electrophoresis on 12 . 5% SDS-polyacrylamide genes and transfer to polyvinylidene difluoride membranes , capsid protein was detected using polyclonal goat anti-Rauscher MLV p30 antiserum ( Viromed Biosafety Laboratories ( NCI/BCB Repository ) , Camden , NJ ) and horseradish peroxidase conjugated rabbit anti-goat antibody ( Invitrogen catalog # R21459 ) . The transfected 293T cells were lysed and tested for mA3 expression by western immunoblot analysis . Cell lysates were subjected to electrophoresis and western blots were probed with a monoclonal antibody against HA , HA-7 ( Sigma catalog #H-3663 ) and a monoclonal anti-tubulin antibody ( Sigma #T-9026 ) . DNA sequences were aligned using MUSCLE [55] and improved manually . Two phylogenies were produced , one for the full-length sequences and one for the exon 2–4 sequences . In all cases the Kimura 2-parameter distance-based neighbor-joining phylogenies for each set returned by PHYLIP ( version 3 . 68 ) [56] ) were corrected for closer correspondence to the consensus Mus phylogeny [29] , [30] . The trees were corrected to make the Nannomys species a monophyletic group and to place M . spretus basal to the M . musculus node . The codeml program of the PAML4 package [28] was used for maximum likelihood analysis of codon evolution [57] . Both lineage-specific and codon-specific analyses were performed . In the lineage-specific selection analyses , the free ratio model ( codon model = 1 ) was used to calculate branch-specific rates of dN/dS . In this model each branch is assumed to have a specific dN/dS ratio . The likelihood of the phylogeny under this model was tested against the likelihood of the phylogeny under the model of one uniform dN/dS ratio across all branches ( codon model 0 ) using a likelihood ratio test ( LRT ) . The significance of the LRT value was assessed using a chi-squared distribution with 49 degrees of freedom for the exon 2–4 sequence analysis and 12 degrees of freedom for the full-length sequence analysis . Selection acting on Apobec3 codons was analyzed using two models of equilibrium codon frequencies and four models of codon selection . The two codon frequency models used were the F3x4 model ( codon frequencies estimated from the nucleotide frequencies in the data at each codon site ) and the F61 ( Codon Table ) model ( frequencies of each of the 61 non-stop codons estimated from the data ) . The codon selection models were two neutral/negative selection models ( M1 and M7 ) which were compared against corresponding positive selection models which included a category for dN/dS>1 ( M2 and M8 , respectively ) . The significance of this additional codon selection category was assessed using LRTs of the phylogeny likelihoods under the neutral and positive selection models . Significance of the test statistics was calculated using a chi-squared distribution with two degrees of freedom . The Bayes empirical Bayes algorithm [58] was used to calculate the posterior probability of individual codons experiencing dN/dS>1 . The C57BL mouse mA3 sequence ( GenBank No . NM_030255 ) was submitted to the LOMETS program [59] . A model constructed using a template with the highest sequence identity was chosen from the top ten solutions ranked by a combination of highest sequence identity , most coverage , Z-score and overall confidence . The model was generated using Modeller v4 [60] and energy optimized in SYBYL7 . 3 using the AMBER7 ff99 forcefield with AMBER7 ff99 atom types and charges with the Powell method to a termination gradient of 0 . 05 kcal/mol·Å . The model was examined using Procheck [61] to detect any bad geometries . mA3 exon 2–4 sequences were given GenBank Accession Nos . GQ901957–GQ901974 . Near full length sequences were given GenBank Nos . GQ871500–506 . | APOBEC3 ( mA3 ) is a cytidine deaminase with antiretroviral activity . Genetic variants of mA3 are associated with the restriction factor Rfv3 ( recovery from Friend leukemia virus ) and with resistance to mouse mammary tumor virus . We sequenced mA3 from laboratory strains and wild mouse species to examine its evolution . We discovered that the mA3 allele in virus resistant mice is disrupted by insertion of the regulatory sequences of a mouse leukemia virus , and this insertion is associated with enhanced mA3 expression . We also subjected the Mus mA3 protein coding sequences to statistical analysis to determine if specific sites are subject to strong positive selection , that is , show an increased number of amino acid replacement mutations . We identified 10 such sites , most of which distinguish the mA3 genes of Rfv3 virus restrictive and nonrestrictive mice . Six of these sites are in two clusters that , in human APOBEC3G , are important for function . We generated a structural model of mA3 , positioned these clusters opposite each other along the putative mA3 active site groove , and demonstrated that substitutions at these sites alter antiviral activity . Thus , mA3 has been involved in genetic conflicts throughout mouse evolution , and we identify an inserted regulatory sequence and two codon clusters that contribute to mA3 antiviral function . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"virology/mechanisms",
"of",
"resistance",
"and",
"susceptibility,",
"including",
"host",
"genetics",
"virology/host",
"antiviral",
"responses",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics"
] | 2010 | Adaptive Evolution of Mus Apobec3 Includes Retroviral Insertion and Positive Selection at Two Clusters of Residues Flanking the Substrate Groove |
Viral infections during pregnancy are a significant cause of infant morbidity and mortality . Of these , rubella virus infection is a well-substantiated example that leads to miscarriages or severe fetal defects . However , structural information about the rubella virus has been lacking due to the pleomorphic nature of the virions . Here we report a helical structure of rubella virions using cryo-electron tomography . Sub-tomogram averaging of the surface spikes established the relative positions of the viral glycoproteins , which differed from the earlier icosahedral models of the virus . Tomographic analyses of in vitro assembled nucleocapsids and virions provide a template for viral assembly . Comparisons of immature and mature virions show large rearrangements in the glycoproteins that may be essential for forming the infectious virions . These results present the first known example of a helical membrane-enveloped virus , while also providing a structural basis for its assembly and maturation pathway .
Rubella virus is an airborne human pathogen that causes a contagious disease with measles-like symptoms in children and adults . Rubella infection in pregnant women can lead to fetal death or severe life-long disabilities such as mental retardation , deafness , cataracts and heart defects in the new-born infants , collectively referred to as congenital rubella syndrome [1] . Despite the availability of an effective vaccine for rubella since the 1960s , the virus is still a global health concern with over 100 , 000 babies born with congenital rubella syndrome every year [2] . Rubella virus is an enveloped , positive-stranded RNA virus with virions ranging from 500 Å to 900 Å in diameter [3] . The virions form a variety of shapes that range from nearly spherical to elongated tube-like structures [3 , 4] . The structural components of rubella virus are comprised of three proteins , glycoproteins E1 ( 58kDa ) , E2 ( 42–47kDa ) , and the capsid protein ( 31kDa ) [5 , 6] . The glycoproteins are type I transmembrane proteins [5 , 7] that form heterodimers on the virion surface [8] . E1 is responsible for recognition and attachment to cellular receptors [9] . It is also involved in membrane fusion in the presence of low pH and calcium ions [10–12] . E2 is required for efficient folding and transportation of E1 through cellular compartments [7 , 8] . Of the two glycoproteins , only the structure of the E1 ectodomain in its trimeric , post-fusion conformation is known [13] . The rubella E1 ectodomain has an elongated structure , similar to the post-fusion conformations of the alphavirus E1 and flavivirus E glycoproteins [13] . The E1 ectodomain has three domains DI , DII and DIII . DII contains two conserved hydrophobic fusion loops at one end of the rubella E1 structure , whereas DI and DIII form the other end [13] . The third structural protein of rubella virus is the capsid protein , which interacts via its amino-terminal domain with the viral RNA genome to form the inner nucleocapsid [6 , 14] . The capsid protein exists as a homodimer [8] and is attached to the viral membrane through the E2 signal peptide [15] . The atomic structure of the carboxy-terminal domain of the capsid protein is known [16] and presumably forms the structural framework of the viral nucleocapsid . The rubella virus structural proteins assemble on and bud from Golgi membranes [1 , 4 , 17] in host cells . The newly budded virions form uniformly dense particles within the Golgi and are referred to as ‘immature’ virions [18] . These immature virions undergo structural reorganization during their passage through the Golgi complex to form mature virions that have a double shell-like architecture . The mature virions are then secreted into the extracellular environment [3 , 16 , 18] . Based on their similar genome organizations , rubella virus and alphaviruses are classified as the only members of the Togavirus family [1 , 19 , 20] . However , these two groups of viruses have little sequence similarity [21] and differ in their assembly as well as maturation strategies [1 , 18] . The alphaviruses include well-known viruses such as Chikungunya virus , Sindbis virus and Semliki Forest virus . They have been studied extensively and their structure determinations have been facilitated by the icosahedral nature of the virions [22–24] . Structural studies have helped in characterizing alphaviruses when complexed with neutralizing antibodies [25–30] and antiviral compounds [31–33] . However , structural information on rubella virus has been minimal owing to the pleomorphic nature of the virus . In this report , we have used a combination of cryo-electron tomography , sub-tomogram averaging and in vitro assembly studies to elucidate the three-dimensional structure and assembly of rubella virions . We have further taken advantage of the direct detector technology to achieve close to nanometer resolution for the sub-tomogram average of the asymmetric and relatively small rubella virus glycoprotein spikes ( ~100kDa molecular weight ) . Additionally , we have isolated immature rubella virions that give insights into the different maturation states of rubella virus . Collectively , these analyses suggest a distinctive assembly and maturation pathway for rubella virus that might be critical for the teratogenic pathogenicity of this medically significant virus .
Samples of mature rubella virus were prepared for cryo-electron tomography as described previously [3] . About 35 tomograms were collected using the Leginon software [34 , 35] and reconstructed using the IMOD software package [36] ( Fig 1A , Materials and methods ) . Distributions of size , shape and morphology of the observed rubella virions ( Fig 1A ) were similar to previous reports [3 , 16] . The virions have an outer shell , about 90–130 Å thick , that includes the glycoproteins and virus membrane . The inner virion shell consists of the capsid protein and the viral genome . The membrane and nucleocapsid are separated on average by about 70 Å [3] . Thin strips of density run across this gap , providing a continuity between the inner nucleocapsid shell and the outer glycoprotein shell [16] ( Fig 1B ) . Variable arrays of surface glycoprotein “spikes” project outwards from the rubella virus membrane [3 , 16] ( Fig 1B ) . However , the resolution limits of earlier tomographic studies did not allow characterization of the spatial arrangement of the individual glycoproteins , which can now be delineated with the current tomographic data in this study . The glycoproteins form rows on the virion surface with a separation of 65 Å to 90 Å between rows ( Fig 1C ) . The average separation between glycoprotein spikes is 50 to 55 Å along each row ( Fig 1C ) . The defining characteristic of the rubella virus surface is the tendency to form sets of four to six parallel rows of glycoproteins . In elongated virions , the glycoprotein rows wrap around the virion in a six-start helical pattern that terminate near the ends of the virion ( Fig 2A and S1 Movie ) . Irregular shaped virions that are large and partially tubular also have short regions of helically wrapped glycoprotein rows ( Fig 2B and S2 Movie ) . Small and pseudo-spherical virions appear to be composed of two or three different sets of glycoprotein rows that merge to form a closed shell ( Fig 2C and S3 Movie ) . The helical pitch values are different between individual rubella virions ( Fig 2 ) . This irregular helical nature of the glycoproteins on the surface of differently shaped rubella virions was not previously known . The rubella glycoproteins ( E1 and E2 ) are present as heterodimeric complexes or spikes on the surface of the virus [3 , 8] . However , no direct data was previously available on the relative positions of the rubella surface glycoproteins . To address this lack of information a total of 240 glycoprotein rows containing similarly spaced adjacent spikes from different virions were selected and averaged using the PEET sub-tomogram averaging software [37 , 38] ( Materials and methods ) . The resulting averaged map has significant density in only about one half of the volume that represents the base of the spike ( S1 Fig ) . This suggests that the glycoprotein complexes are positioned similarly along a surface row but the external ends of the glycoproteins have different conformations . Consequently , individual glycoprotein spike volumes that appear predominantly straight were picked manually using the IMOD software [36] . The selected sub-volumes were split into two datasets for independent processing and subjected to sub-tomogram averaging procedures ( Materials and methods , S2 Fig ) . The averaged glycoprotein spike structure consists of a broad base below a narrower stalk ( Fig 3A ) and has an estimated resolution of 11 . 0 Å ( at 0 . 143 FSC cut-off ) and 14 . 9 Å ( at 0 . 5 FSC cut-off ) ( S3 Fig ) . The top ( external ) part of the averaged glycoprotein spike has weaker density compared to its base , indicating that there is more variability in the region distal from the membrane . The atomic structure of the E1 ectodomain ( without it’s stem region ( PDB ID: 4ADG ) [13] ) was fitted into the averaged density using the EMfit program [39] ( Fig 3B ) . A complete three-dimensional search of all possible angles produced three top fits . However , all fits resulted in similar orientations with the long direction of the known E1 structure roughly perpendicular to the viral membrane ( S1 Table ) . The volume of the averaged spike was 133 Å3 . From this , the volume occupied by the fitted E1 ectodomain ( PDB ID: 4ADG ) was subtracted , leaving a residual volume of 52 Å3 at the base of the spike . The residual volume can accommodate a protein of roughly 33kDa molecular weight . This is larger than the molecular mass of the E2 ectodomain polypeptide ( 25kDa ) but is smaller than the estimated mass of the glycosylated E2 ectodomain ( 37kDa to 42kDa ) . Taking into consideration that the glycosylation on the E2 ectodomain is known to be heterogeneous [5 , 40] , it is likely that the densities corresponding to the varied carbohydrate moieties on E2 were canceled out during sub-tomogram averaging . Thus , the residual volume at the base of the averaged rubella glycoprotein spike would be sufficient to accommodate the ectodomain of E2 glycoprotein ( Fig 3B ) . The arrangement of glycoprotein spikes on rubella virions was further examined by placing the sub-tomogram averaged spike , back into each of the original tomographic positions that had been used to obtain the averaged spike structure . The relative orientations of the long axes of the glycoprotein spikes to the viral membrane varied from 30° to 90° . Along glycoprotein rows , adjacent spikes were slightly rotated with respect to each other , along an axis perpendicular to the plane of the membrane . This analysis showed that along unbroken glycoprotein rows in rubella virions , the glycoprotein complexes are similarly oriented with respect to each other , such that E2 would most likely be located between adjacent E1 positions ( Fig 3C and 3D ) . The internal nucleocapsid shell of rubella virus consists of the capsid protein and the viral RNA genome . The nucleocapsid surface follows the contour of the viral membrane [3] . In previous tomographic studies [16] , the individual subunits of the nucleocapsid could not be resolved and appeared merely as rows running approximately perpendicular to the glycoprotein rows on the virion surface . In the current tomograms , the nucleocapsid layer appears as globular subunits arranged in a grid-like pattern ( Fig 4A ) . The viral genome is closely associated with the capsid protein leaving a sparsely populated region in the center of the virions . The nucleocapsid layer is more variable than the glycoprotein arrangement on the virion surface , though this could be due to interference from density associated with the viral RNA . The spacing between capsid units in the viral nucleocapsid varies from 40 Å to 70 Å . However , in well-resolved and ordered regions of the virus the nucleocapsid subunits are positioned roughly underneath the surface glycoprotein heterodimers , with each nucleocapsid unit associated with one glycoprotein spike ( Fig 4B and 4C , S4 Fig ) . To differentiate between the viral genome and the capsid protein in the viral nucleocapsid , recombinant rubella virus capsid protein molecules were produced in vitro to form nucleocapsid cores ( Materials and methods ) . These in vitro assembled rubella virus nucleocapsid cores have a smooth exterior and are hollow . They display a variety of shapes and sizes with diameters ranging from 400 Å to 900 Å ( Fig 5A ) , illustrating that the capsid protein by itself has an inherent tendency to form pleomorphic particles . An agarose gel assay using the purified , recombinant nucleocapsid cores demonstrated that these particles contain nucleic acids of different sizes ( S5 Fig ) . Treatment with benzonase nuclease during cell lysis greatly reduced the yield of the nucleocapsid cores . However , benzonase treatment of purified nucleocapsid cores does not affect the integrity of the cores , but only removes its nucleic acid content ( S5 Fig ) . Thus , benzonase treated nucleocapsid cores were used to remove any interference from nucleic acids for the tomographic studies . Purification of the recombinantly produced capsid protein also resulted in the isolation of a series of intermediate assembled complexes . The smallest identifiable complex consisted of units that appeared to have 4-fold symmetry ( Fig 5A ) . Other larger complexes consisted of linear rows of these tetramers ( Fig 5B ) . Cryo-electron tomograms of the in vitro assembled nucleocapsid cores were collected and processed in a similar manner to the infectious virus samples ( Materials and methods ) . Tomogram sections of the nucleocapsid cores show a tetramer-like pattern as seen in the assembly intermediates ( Fig 5C and 5D ) . The tomograms also show that these tetrameric arrays have occasional discontinuities that might be necessary to form three-dimensional , closed nucleocapsid core particles ( Fig 5D ) . The arrangement of the capsid subunits in the nucleocapsid cores agrees with the pattern observed in the viral nucleocapsids ( Fig 4B and 4C ) . This implies that the nucleocapsid in rubella virions is composed of a pseudo-tetrameric arrangement of capsid proteins in contact with the viral genome . These observations also confirm that the bulk of the capsid protein lies in the inner shell of the virion , clarifying previous estimates of the capsid protein location [3 , 16] . The cryo-electron density from about 20 isolated tetrameric units seen in the tomograms of the in vitro assembled nucleocapsid cores were re-oriented to a common orientation and averaged ( Materials and methods ) . This showed that each monomer of the tetrameric unit has a dumbbell-shaped structure ( Fig 5E ) . The rubella capsid protein exists as a functional dimer . The C-terminal domain of the capsid protein structure consists of approximately 150 amino acids with 27 amino acids at the C-terminus being disordered in the crystal structure ( PDB: 4HBE ) [16] . The N-terminal domain consists of approximately 127 amino acids whose atomic structure is unknown . The structure of the C-terminal domain of the capsid protein dimer was fitted into one lobe of the dumbbell shaped averaged density of the in vitro assembled nucleocapsid tetramers using the EMfit program [39] ( Fig 5F , S1 Table ) . The two top fitting orientations for the capsid protein in the averaged density are 180° apart , relative to each other , along an axis perpendicular to the two-fold axis of the capsid dimer ( S6 Fig , S1 Table ) . These best fitting conformations of the capsid protein are similar to the orientations of the capsid protein expected to be facing the viral membrane [16] . As the N- and C-terminal domains of rubella capsid protein are similar in size , the remaining density of the averaged dumbbell shaped unit must be the location of the N-terminal domain of the capsid protein . Using these fitting results , it can be deduced that the disordered residues at the C-terminal end of the capsid protein , not seen in its crystal structure [16] , likely correspond to the thin strips of density visible in the cross-section of rubella virion tomograms ( Fig 1B ) , linking the inner nucleocapsid shell to the membrane anchored E2 signal peptide in the viral membrane . The dumbbell shape of the capsid protein also accounts for the double-layer appearance seen in the in vitro nucleocapsids ( Fig 5C and 5D ) . In the virions , there are long pieces of density , which correspond to the viral RNA , that are closely associated to the N-terminal region of the capsid proteins . Thus , the double layer characteristic of the nucleocapsid is not as obvious in the virions as it is in the in vitro nucleocapsid cores . Vero cells infected with rubella virus were lysed at 22 hours’ post-infection to release intracellular virion particles . Further virus purification was carried out using the cell lysates ( Materials and methods ) . Cryo-electron microscopy of the purified , immature virus particles shows that the immature virions are uniformly dense and variable in size ( Fig 6A ) corroborating earlier descriptions of immature rubella virions [18] . The immature virions have a smooth exterior with no prominent features ( Fig 6A ) , which implies that during the early stages of virion budding and transport inside host cells , rubella E1 lies close to the virion surface instead of protruding out from the surface as in extracellular mature virions ( Fig 6B and 6C ) . The uniform dense nature of the immature virions also suggests that in the initial immature state , the glycoproteins and membrane layer are more closely interacting with the nucleocapsid layer than they are in the mature virions . Hence , the glycoprotein and nucleocapsid layers assemble probably into a more compact arrangement in the initial immature form , with the glycoproteins in register with the capsid proteins . Loss of order could be a product of structural reorganization that occurs during virion maturation .
The rubella virus structure had been expected to have T = 3 , icosahedral symmetry [1 , 8 , 41] . Instead , the structure of rubella virus , as shown here , has an irregular helical organization of its surface glycoproteins and a pseudo-tetrameric inner nucleocapsid arrangement . The glycoprotein arrangement in rubella virions is unique , as other known membrane enveloped viruses exhibit helical structures only in their inner nucleoprotein complex or in their matrix protein layer , such as in paramyxoviruses [42] , marburgviruses [43] , and influenza-A viruses [44] . Thus , rubella virus is the only known example of a helical surface structure associated with a membrane enveloped virus . The relative positions of the rubella glycoproteins , with an extended E1 conformation and with E2 at the base of the spike complex , is different from the glycoprotein heterodimer conformation observed in alphaviruses . However , this structural placement of rubella glycoproteins is consistent with protease studies [10 , 45] and immunological reactivity studies on rubella virus [46 , 47] , which indicate that the E1 molecule is more exposed and accessible than E2 on the virion’s surface . E1 is the primary target for neutralizing antibodies against rubella virus . The common antibody binding region on E1 ( between residues 202 to 285 ) [48–52] is exposed to the surrounding environment in the spike density , given the orientation of E1 as determined by the fitting results . Though the E1 crystal structure used in this study is similar to the post-fusion E1 and E glycoprotein structures of alpha- and flaviviruses , the rubella E1 structure was determined under neutral pH and in the absence of detergents [13] , unlike in the case of the post-fusion structures of alphavirus E1 [53] and flavivirus E glycoproteins [54] . Moreover , the translation of domain III between the pre- and post-fusion conformations in alpha- and flaviviruses , is possibly a consequence of the large conformational change of the glycoproteins from being tangential to being almost perpendicular to the viral membrane . Under these considerations , the proposed movement of domain III in alpha- and flaviviruses might not be directly applicable to rubella virus . Even with a possible translational movement of domain III , the position of the volume assigned to rubella E2 would not change significantly . Furthermore , expression of rubella E1 and E2 ectodomains together results in secretion of the E1 ectodomain alone [13] , indicating that the E2 ectodomain has a higher affinity to membranes than E1 . Thus , these observations are consistent with the placement of E2 close to the base of the rubella glycoprotein spike , rendering it relatively inaccessible compared to E1 . The molecular weight of rubella virus E2 is only about one half of E2 in alphaviruses . There is also no significant sequence similarity between the E2 proteins of rubella and alphaviruses . Hence , the structure of E2 in these viruses is probably quite different . In alphaviruses , the glycoprotein E1-E2 heterodimers lie close to the membrane surface in their pre-fusion state with the E1 fusion loop physically masked by the neighboring E2 molecule . In the rubella glycoprotein complex , the E1 protein extends out from the virion surface and does not appear to require E2 for shielding its two fusion loops . The rubella virus E1 crystal structure [13] , which was determined in the absence of detergent and at neutral pH , also presents an extended E1 conformation with fusion loops exposed but not positioned for membrane insertion . The observed variability in the glycoprotein spikes suggest that the tips of the E1 structures are flexible , which would help E1 to avoid non-productive membrane interactions at neutral pH . This might also help explain the requirement for calcium ions to stabilize the rubella E1 fusion loop conformations for membrane insertion at low pH [12 , 13] . Difference in the glycoprotein spike orientations between the immature and mature virion forms could also be to protect the fusion loops on the E1 glycoprotein structure from undergoing unproductive membrane interactions in the low pH environment of the cellular Golgi network . Subsequent virion maturation in the Golgi complex would then be necessary for reorganization of the viral envelope to yield fusion competent , extracellular , ‘spiky’ mature virus particles . A similar strategy of surface glycoprotein reorganization during virion maturation occurs also in bunyaviruses [55] and flaviviruses [56] . In addition to the glycoprotein arrangement , the association of protein tetramers to form enclosed shells as seen in the rubella virus nucleocapsids is also unusual and not observed in other known virus structures . Rubella capsid protein molecules expressed in bacterial cells , spontaneously form nucleocapsid cores in the presence of cellular nucleic acids . This indicates that the rubella capsid protein does not need the rubella genome for nucleocapsid formation , similar to previous observations [57] , but can associate with random nucleic acids to form nucleocapsids . In the context of virus assembly , this implies that the binding of the capsid protein to the viral RNA is not entirely sequence specific . This suggests that the virus machinery uses other methods , such as an abundance of viral RNA relative to cellular RNA at the virus budding site [58] , for efficient packaging of the viral genome into budding virions . Thus , rubella virus appears to share certain common characteristics , not only with alphaviruses but also with other arbovirus genera , such as flaviviruses and bunyaviruses . However , in addition to these features that suggest similarities to arboviruses , rubella virus has also evolved some unique traits such as an exposed E1 fusion loop conformation . Together , these observations point to a more complex structural evolution in rubella virus than previously assumed .
Mature rubella virus ( M33 strain ) was cultured in Vero cells ( ATCC CL-81 ) and purified according to previously published protocols [3] . For purification of intracellular immature rubella virions , a modified protocol similar to purification of intracellular retroviruses [59] was followed . Briefly , rubella virus ( M33 strain ) infected Vero cells were lysed 22 hours’ post-infection by suspending the cells in a hypotonic buffer containing 20mM Tris pH 8 . 0 , 10mM NaCl plus 15mM MgCl2 and then homogenizing the suspension in a Dounce homogenizer . The cell lysates were centrifuged at 3000×g for 5min to pellet cell debris . The supernatant was further centrifuged at 20 , 000×g for 20 min . The supernatant was again collected and incubated with 5mM EDTA for 10min followed by addition of 10μg/ml of RNase A . After 15 minutes , the supernatant buffer concentration was adjusted to contain 120mM NaCl . The supernatant containing the immature virions was further purified similar to the mature virus purification protocol [3] . A discrete band in the expected virus density range was observed after density gradient ultracentrifugation , which was then extracted for cryo-EM analysis . Mock-infected Vero cells were lysed and treated in the exact same manner as the rubella virus infected Vero cells during immature virion purification . No bands were seen in the density gradient after the final gradient ultracentrifugation step in the mock-infected control . The rubella virus capsid gene has two methionines in the first 10 nucleotides of its sequence ( at positions 1 and 9 ) . To improve bacterial expression , the nucleotide sequence corresponding to amino acids 1 to 8 was removed . Rubella virus ( M33 strain ) capsid protein ( RVC ) gene sequence corresponding to amino acids 9–277 was cloned into a pTXB1 plasmid ( New England Biolabs ) . The RVC-9-277-pTXB1 construct was transformed into E . coli BL21 ( DE3 ) cells ( Novagen ) . Cells were grown at 37°C and induced with isopropyl thiogalactoside ( IPTG ) when the O . D reached 0 . 8 . The induced cultures were grown at 25°C for 5 hours and pelleted . Cell pellets were re-suspended in buffer containing 50mM Tris ( pH 7 . 2 ) and 500mM NaCl . The cells were lysed using sonication and the target protein purified using a chitin column ( New England Biolabs ) . The purified RVC-9-277 protein was applied to a Sepahcryl S-400 column ( GE Healthcare ) to fractionate the sample according to different apparent molecular weights . Fractions corresponding to assembled nucleocapsids and smaller capsid protein complexes were pooled and concentrated separately . The nucleocapsid core samples were treated with 0 . 001% benzonase and the buffer was exchanged before sample preparation for cryo-electron tomography . For cryo-electron tomography , the purified samples were mixed with 4X concentrated 10nm BSA gold solution ( Aurion , Wageningen , Netherlands ) in a 15:1 volume ratio . 3μl of this mixture was applied to Quantifoil R1 . 2/1 . 3 grids ( Electron Microscopy Sciences ) , blotted and plunge frozen in liquid ethane . Tilt-series data was collected using a Titan Krios ( operated at 300kV ) and a Gatan K2 Summit detector . The Leginon software [34 , 35] was used to collect data at appropriate grid positions . A magnification of 11000X was used with a pixel size of 1 . 32Å . Tilt-series were collected from -60° to +60° in 1 . 5° steps with a defocus range of 4–5μm for the virus samples and 5–6μm for the in vitro assembled nucleocapsid core samples . The total dose applied to each sample was between 80–90 e-/Å2 [34 , 35] . The tilt-series were aligned using gold fiducial markers and reconstructed using the IMOD software [36] . Ctfplotter program from the IMOD suite [36] was used to estimate the defocus of each individual tilt-image in the tilt-series . CTF correction and phase-flipping of the aligned tilt-series was then carried out as part of the IMOD tomographic reconstruction procedure . Though accurate resolution estimation is challenging for tomographic reconstructions , the resolution of the virus tomograms was estimated to be better than 50 Å because it was possible to see separation between the two layers of the viral membrane in directions perpendicular to the electron beam . For negative staining , 1% solution of ammonium molybdate was added to the sample on a grid , blotted , washed twice with distilled water and then air dried after blotting off the excess liquid . The grids were analyzed using a CM200 ( FEI ) microscope with a 1k×1k CCD camera . Sub-volumes of virions were extracted from the cryo-electron tomograms using IMOD [36] . The representation of the virion density was inverted using the EMAN2 suite [60] . Using the UCSF Chimera software [61] , the virion density maps were low pass filtered using a gaussian filter to 75 Å resolution , followed by removal of dis-continuous densities using the ‘hide dust’ feature . The virions were then rotated around different orthogonal axes for making a movie using Chimera [61] . Snapshots from these movies in color were used to make Fig 2A–2C . For sub-tomogram averaging of individual glycoprotein spikes that were elongated and predominantly straight , 7290 sub-volume positions were picked on the surface of virions in all directions from 8X binned tomogram data using IMOD [36] . Each glycoprotein spike was identified using two points . The first point was placed close to the spike end which is distal to the membrane and the second point to the spike end closer to the membrane . Only spikes that looked predominantly straight were picked from the tomograms , such that a line connecting the two points on the spike , passed essentially through the center of the spike stalks . The program ‘stalkInit’ in the PEET suite of programs was then used to convert these points into corresponding ‘motive lists’ with position co-ordinates for each spike as input into the PEET suite . Rotation angles were calculated for each spike as the angles between the lines indicating the spikes and the tomogram ‘y’ axis . Subsequent sub-tomogram extraction and averaging procedures were carried out with the PEET software [37 , 38] using CTF corrected , 4X binned data . As the spikes on the virion surface are densely packed and touch each other at the base , a tight cylindrical mask with a radius of 6 pixels ( 31 . 6 Å ) and a soft Gaussian fall-off of 6 pixels was applied to the particles such that only one spike was visible in the extracted sub-volumes . Before starting the alignment and averaging procedure , all the particles were rotated to align the long axis of the spikes ( given by the two identification points ) in one common direction . Individual particles were checked to make sure that all the spikes were oriented in the same direction with no inversion of the spikes’ orientations . The search parameters allowed for a complete 360° search about the long axis of the spikes , but restricted the search in the other two directions to ±30° with 0° being along the long direction of the spikes . The initial alignment iterations were coarse searches followed by finer interval searches in later iterations . Missing wedge compensation feature within the PEET software [37 , 38] was applied during alignment and averaging of the sub-tomogram volumes . Initially , five spikes that looked properly formed from different tomograms were selected and independently used to align 256 sub-volumes to calculate initial models . The model which best resembled the individual spike conformations was selected for further steps . The remaining sub-volumes were split into two halves and processed independently . The initial model was used as the reference model for aligning the individual spikes at the first iteration . For subsequent iterations , the reference model was updated by averaging spikes with the highest correlation coefficients to the reference model in the previous iteration , representing 2/3rd of all the spikes . Sub-volumes were aligned to the updated reference model at each iteration . After the final iteration , the determined relative position and angles for each sub-volume was used to average all the sub-tomograms together to give an averaged 4X binned map of the glycoprotein spike . The determined relative orientations and positions were then applied to 2X binned data and the final sub-tomogram average map was calculated using 2X binned data with a pixel size of 2 . 64 Å . The reference models at every alternate iteration along with the final averaged map are shown in S2 Fig . The final 2X binned averages from the two half-sets were applied with a soft-edged mask that covered only the glycoprotein spike in order to exclude the surrounding , weak membrane densities . A Fourier-Shell Correlation ( FSC ) curve was subsequently calculated between the two masked averages using the EMAN2 software suite [60] . The sub-tomogram averaged map was subsequently low pass filtered to 8 Å and then sharpened with an ad hoc B-factor of—375 Å2 to increase the impact of higher resolution terms in the map . This sharpened map was used for fitting of the E1 crystal structure . For sub-tomogram averaging of rows of glycoproteins ( as opposed to individual spikes ) , 240 rows containing four spikes each , with approximately similar spacing between the spikes , were picked using IMOD [36] from 8X binned data . The aligning and averaging procedures were performed with the PEET software [37 , 38] using 4X binned data . Four different rows were used as independent starting models . The sub-volumes were aligned against the starting models for iterative refinements and averaged . All the independent averaged results had very similar final density . Sub-tomogram averaging of the recombinantly produced capsid tetramers were also performed in a similar way to the above described routine for glycoprotein rows . Twenty tetramer sub-volumes were picked from 8X binned tomograms of nucleocapsid cores . Alignment and averaging was performed using 4X binned data . | Rubella virus ( RV ) causes serious fetal defects when contracted during pregnancy . Despite its medical importance , due to the irregular shapes and different sizes of the virions , the RV structure has remained unknown . Using cryo-electron tomography , we have determined the RV structure , which shows a unique , helical outer surface . Subsequent local averaging of the RV surface spikes has established the conformations of its immunogenic glycoproteins . In vitro assembly studies on the virus capsid protein have provided insights into the interactions necessary for virus assembly . Comparisons between mature and immature RV show large conformational changes in the virion structure that are essential for virus maturation . These results help to gain a structural understanding of RV pathogenicity , which may also be relevant to other teratogenic viruses . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"rubella",
"virus",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"pathogens",
"microbiology",
"nucleocapsids",
"viral",
"structure",
"viruses",
"rna",
"viruses",
"glycoproteins",
"infectious",
"diseases",
"medical",
"microbiology",
"microbial",
"pathogens",
"viral",
"packaging",
"viral",
"replication",
"virions",
"biochemistry",
"virology",
"virus",
"glycoproteins",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"rubella",
"glycobiology",
"organisms"
] | 2017 | Assembly, maturation and three-dimensional helical structure of the teratogenic rubella virus |
Alphaviruses are a group of widely distributed human and animal pathogens . It is well established that their replication is sensitive to type I IFN treatment , but the mechanism of IFN inhibitory function remains poorly understood . Using a new experimental system , we demonstrate that in the presence of IFN-β , activation of interferon-stimulated genes ( ISGs ) does not interfere with either attachment of alphavirus virions to the cells , or their entry and nucleocapsid disassembly . However , it strongly affects translation of the virion-delivered virus-specific RNAs . One of the ISG products , IFIT1 protein , plays a major role in this translation block , although an IFIT1-independent mechanism is also involved . The 5’UTRs of the alphavirus genomes were found to differ significantly in their ability to drive translation in the presence of increased concentration of IFIT1 . Prior studies have shown that adaptation of naturally circulating alphaviruses to replication in tissue culture results in accumulation of mutations in the 5’UTR , which increase the efficiency of the promoter located in the 5’end of the genome . Here , we show that these mutations also decrease resistance of viral RNA to IFIT1-induced translation inhibition . In the presence of higher levels of IFIT1 , alphaviruses with wt 5’UTRs became potent inducers of type I IFN , suggesting a new mechanism of type I IFN induction . We applied this knowledge of IFIT1 interaction with alphaviruses to develop new attenuated variants of Venezuelan equine encephalitis and chikungunya viruses that are more sensitive to the antiviral effects of IFIT1 , and thus could serve as novel vaccine candidates .
The Alphavirus genus in the Togaviridae family contains 30 species and has a worldwide distribution [1] . Most alphaviruses are transmitted by mosquito vectors between amplifying vertebrate hosts [2] . In vertebrates , alphaviruses cause acute infections , characterized by high titer viremia that allows infection of mosquitoes during blood feeding . A number of alphaviruses , including Venezuelan ( VEEV ) and eastern ( EEEV ) equine encephalitis viruses , and chikungunya virus ( CHIKV ) , are globally important , emerging public health threats . These viruses can cause epidemics of severe meningoencephalitis with frequent lethal outcomes , or polyarthritis with excruciating and chronic joint pain [3] . Over the last ten years , there have been multiple outbreaks of CHIKV infection with millions of people infected [4–7] , including an ongoing epidemic in the Americas with more than 40 countries affected and over 1 . 1 million suspected cases . Epizootic strains of VEEV and EEEV are almost universally lethal for equids [8] . In addition , the latter viruses can be transmitted efficiently by aerosol [9] , are highly stable in lyophilized form , and were developed previously as biological warfare agents [9] . In spite of their public health threat , the pathogenesis of alphaviruses on the molecular and cellular levels remains poorly understood , and no approved vaccines or therapies exist for any of them . Alphavirus infections are sensitive to type I IFN both in vivo and in vitro [10–12] . Mice defective in IFN signaling succumb to most alphavirus infections within a few days [13] . IFN treatment induces a large set of IFN-stimulated genes ( ISGs ) , whose protein products prevent infection with many pathogens , including alphaviruses [14–17] . Although hundreds of ISGs have been described , only few have had their antiviral functions unambiguously defined , particularly for alphaviruses [18–24] . The involvement of so many cellular genes implies that the antiviral response might be highly redundant against a given pathogen . Such redundancy would reduce the possibility of selection of virus mutants resistant to an ISG product . However , it is also plausible that only a subset of ISGs is essential for protection against a specific pathogen or groups of related pathogens . Thus , identification of pathogen-specific ISGs might lead to development of targeted therapeutics lacking the therapy-limiting side effects of IFNs . Previous studies have shown that translation of alphavirus-specific , but not cellular RNAs , is inhibited in IFN-β-treated cells [25–28] . However , these studies were based on either using replication-competent viral RNAs or in vitro-synthesized RNAs , delivered into the cells by transfection procedures that do not mimic the natural infectious process . To overcome such limitations , we designed a new experimental system , which allows selective investigation of the effect of ISG activation on the very early steps of alphavirus infection , prior to viral RNA replication . To achieve this , we exploited the unique characteristics of the recently discovered Eilat ( EILV ) alphavirus [29 , 30] . Similar to other members of the genus , the EILV genome is an 11 . 5 kb positive polarity RNA , which is packaged into icosahedral nucleocapsids , surrounded by lipid envelopes with imbedded trimeric glycoprotein spikes [30] . Upon delivery into the cytoplasm , the genomic RNA serves as a template for translation of viral nonstructural proteins , nsP1-4 , which mediate replication of the genomic RNA and transcription of the subgenomic ( SG ) RNA [1] . The latter RNA functions as a template for synthesis of viral structural proteins . EILV retains the critical characteristics of other alphaviruses in terms of its genomic coding strategy , virion structure and nucleotide sequences of the RNA promoters [30] . The amino acid sequences of the replication enzymes and structural proteins also demonstrate high degrees of conservation with other alphaviruses . However , importantly , while the nsP1-4 complexes actively replicate viral RNA in mosquito cells , they are nonfunctional in vertebrate cell lines [29] . We have modified the EILV genome and converted this virus into an efficient system for delivery of nonreplicating RNAs of interest into vertebrate cells via the natural alphavirus-specific entry pathway . Using EIL/VEEV chimeric viruses and other experimental approaches , we studied the components of early , type I IFN-induced cell resistance to alphaviruses and the mechanisms employed by alphaviruses to interfere with the inhibitory effect of type I IFN . Our studies demonstrated the following: ( i ) type I IFN pretreatment blocks translation of alphavirus RNAs delivered in infectious virions , but does not affect virus attachment , entry or nucleocapsid disassembly; ( ii ) the IFN-induced inhibition of translation relies on both IFIT1-dependent and-independent mechanisms; ( iii ) the 5'UTRs derived from genomes of different alphaviruses differ in their ability to drive translation of the downstream genes after type I IFN pretreatment or in the presence of IFIT1; ( iv ) the efficiency of replication of the corresponding viruses in IFIT1-producing cells correlates directly with translation sensitivity of their genomes to IFIT1; ( v ) the presence of IFIT1 makes wt , but not tissue culture-adapted , alphaviruses strong inducers of type I IFN; and ( vi ) by modifying sequences in the 5'UTRs , we have generated alphaviruses with graded levels of sensitivity to IFIT1 , an approach that can be applied to vaccine development . Collectively , our results indicate that IFIT1 can act both as an antiviral effector molecule and an inducer of innate immunity against alphaviruses . Our data also provide a mechanistic explanation for alphavirus resistance and sensitivity to IFIT1 and adaptation of natural isolates to cell culture .
One of the distinguishing features of the newly discovered EILV alphavirus is its very efficient replication and plaque formation in Aedes albopictus and other mosquito cells yet inability to replicate in cells of vertebrate origin [29] . This cell-restricted phenotype is determined both by structural proteins , which are not able to promote EILV entry into vertebrate cells , and the nonstructural proteins , which appear to lack appropriate interaction with key cellular factors required for RNA replication [29] . We have exploited EILV’s inability to replicate RNA in vertebrate cells to engineer a set of chimeric viruses , in which the EILV structural genes were replaced by those derived from the attenuated VEEV strain TC-83 . VEEV TC-83 demonstrates efficient replication in a wide variety of insect and vertebrate cells , and the designed chimeras were expected to infect vertebrate cells but to fail to subsequently replicate their RNA . The EIL/VEEV recombinant virus encoded the EILV replication machinery , and all of the EILV-specific cis-acting RNA promoter elements , including the 5' and 3'-UTRs of the viral genome and SG RNA , as well as the 51-nt conserved sequence element ( CSE ) and SG promoter [1] , driving expression of VEEV structural protein genes ( Fig 1A ) . Other chimeras , EIL/nLuc/VEEV and EIL/GFP/VEEV , had additional SG promoters to control the expression of nanoLuc ( nLuc ) and GFP , respectively ( Fig 1A ) . In C7/10 mosquito cells , within 24 h post of electroporation with in vitro-synthesized RNAs , chimeric viruses were released into the supernatant attaining titers exceeding 5x108 PFU/ml . Almost all of the cells transfected with EIL/GFP/VEEV RNA were GFP-positive within 8 h post electroporation , indicating that no additional adaptive mutations were required for RNA replication in mosquito cells . The generated chimeras were capable of efficient replication in mosquito C7/10 cells ( Fig 1B ) , but no virus growth was detected in vertebrate NIH 3T3 ( Fig 1B ) , BHK-21 , or Vero cells . In contrast to recombinant chimeras , the control VEEV TC-83 productively replicated in both vertebrate and mosquito cell lines ( S1 Fig ) . A lack of infectious virus release does not conclusively rule out the possibility of viral RNA replication . Thus , we next assessed levels of EIL/VEEV- and VEEV TC-83-specific RNAs in the NIH 3T3 cells at different times post infection by RT-qPCR using primers specific to the VEEV E2 structural gene ( S1 Fig ) . By 24 h post infection , we detected more than a 10 , 000-fold increase in the concentration of VEEV TC-83-specific RNAs , compared to input RNA present in the viral particles adsorbed during inoculation . Within the same time frame , EIL/VEEV-infected cells demonstrated a 100-fold decrease in RNA concentration . This confirms that EILV-specific non-structural proteins fail to support significant levels of transcription and replication in NIH 3T3 cells . Analysis of the composition of viral particles produced by the chimeric viruses revealed an important new feature: in mosquito cells , these viruses package both genomic and SG RNAs ( Fig 1C and 1D ) . Both RT-qPCR and electrophoretic analysis of metabolically [3H]uridine-labeled RNAs demonstrated large amounts of SG RNAs in the viral particles , which were harvested from C7/10 cells well before the development of noticeable cytopathic effects ( CPE ) and further purified by ultracentrifugation through 25% sucrose . This was the result either of the inefficiency of the VEEV capsid protein binding to the heterologous EILV-specific RNA packaging signal ( PS ) [31] or as described for Aura virus [32] , a natural ability of EILV to nonspecifically package the abundant SG RNAs . Regardless , the released virions could be potentially exploited as delivery vehicles for both genomic and SG RNAs . Indeed , infection of NIH 3T3 and other vertebrate cells with mosquito cell-derived , chimeric EIL/GFP/VEEV viruses resulted in reporter GFP gene or VEEV TC-83 structural protein expression despite an absence of viral RNA replication or de novo transcription of the SG RNA ( S2 Fig ) . To analyze expression of another reporter , nLuc , from the delivered SG RNA , mosquito cell-derived EIL/nLuc/VEEV chimeric virus was purified to homogeneity by continuous sucrose gradient ultracentrifugation . Then , NIH 3T3 cells were infected at 4°C and further incubated at 37°C . nLuc was expressed efficiently within the first 4 h post infection , but after this , an increase in Luc activity was no longer detected ( Fig 2A and 2B ) , which suggested a lack of de novo nLuc RNA synthesis . Luciferase expression was not detected in the infected cells treated with the protein synthesis inhibitors puromycin and cycloheximide . In contrast , EIL/nLuc/VEEV-infected C7/10 cells demonstrated a continuous increase in nLuc expression for at least 24 h ( S3 Fig ) , reflecting the reporter SG RNA synthesis . The lack of RNA replication of EIL/VEEV-based viruses in vertebrate cells makes them of little value for studying alphavirus pathogenesis and virus-host interactions . However , these viruses represent a unique experimental tool , which allows designing and packaging into infectious virions a very wide variety of mRNA ( SG RNAs ) , and studying the effect of IFN pretreatment on the early steps of alphavirus infection in a variety of experimental conditions in the absence of viral RNA synthesis/replication . Expression of heterologous genes encoded by SG RNAs of chimeric viruses was sensitive to IFN-β pretreatment . The pre-treated cells produced no GFP ( S2 Fig ) or nLuc ( see below ) upon infection with EIL/GFP/VEEV and EIL/nLuc/VEEV , respectively . The inhibition of nLuc expression depended on both concentration of IFN-β and the time of pretreatment relative to infection ( Fig 2C and 2D ) . Pretreatment with IFN-β at a concentration of 1 IU/ml for 20 h inhibited nLuc expression by 10-fold , and concentrations above 100 IU/ml reduced it to background levels ( Fig 2C ) . Four-hour-long incubation of cells with IFN-β at a concentration of 100 IU/ml ( Fig 2D ) , also decreased nLuc expression by 10-fold , and no expression above the background level was detected after 16 h of cell pre-incubation with IFN-β . In these experiments , the noticeably high background levels of nLuc activity were the result of packaging of nLuc protein into viral particles in C7/10 cells and then delivery together with viral RNAs ( see the next section ) . The experiments with EIL/VEEV-based RNA delivery demonstrated strong quantitative differences with data derived from the RNA transfection experiments [25] . Type I IFN pretreatment inhibited expression of the RNA-encoded genes not within 10 fold , but by more than two orders of magnitude . However , regulation of expression of mRNAs delivered by viral particles may be not exclusively determined by translation . The early events in alphavirus infection , such as binding to cell surface receptors , endocytosis , fusion with the endosomal membrane , and intracellular transport and the release of nucleocapsid into the cytoplasm followed by its disassembly can all affect viral gene expression . Moreover , VEEV glycoproteins reportedly contribute to the development of virus resistance to type I IFN [33] . Thus , possible effect ( s ) of type I IFN treatment on other early steps of EIL/VEEV infection had to be investigated . The potential effects of type I IFN pretreatment on virus attachment to plasma membrane were studied by two different approaches . First , we infected NIH 3T3 cells , pre-treated with IFN-β , and mock-treated , with equal numbers of PFUs of VEEV virions at 4°C for 1 h . Then cells were fixed and immunostained using VEEV-specific Abs without permeabilization of the plasma membrane . The three-dimensional ( 3D ) images of multiple randomly selected cells were acquired on a confocal microscope ( Fig 3A ) and used to quantify the numbers of cell-bound virions ( Fig 3B ) ( see Materials and Methods for details ) . No statistically significant difference ( P > 0 . 1 ) was detected between the IFN-β- and mock-treated cells ( Fig 3B ) . The second approach took advantage of the ability of VEEV and other alphavirus virions to nonspecifically package detectable levels of other cytoplasmic proteins and even larger protein complexes , such as ribosomes [34] , into released infectious virions . Virions released from EIL/nLuc/VEEV-infected C7/10 cells were purified to homogeneity by density gradient ultracentrifugation . The residual nLuc in these purified samples co-pelleted with viral particles , with no Luc activity remaining in the supernatants , strongly suggesting that nLuc was associated with purified virions . Then , equal amounts of the purified EIL/nLuc/VEEV virions were adsorbed to IFN-β- and mock-treated NIH 3T3 cells at 4°C , and after extensive washing with PBS , cells were lysed and assessed for nLuc activity . No measurable difference was found between the samples ( Fig 3C ) . To avoid reaching saturation conditions in the above experiments , the amount of nLuc-containing virus used for binding was optimized to remain in the linear range ( S4 Fig ) . As a control , we used soluble nLuc-containing , flow-through media , which passed centrifugal filters used for virus concentration . Despite this fraction having a few orders of magnitude higher original nLuc concentration than the fraction of purified viral particles , the amount of cell-binding nLuc in these media samples was 500-fold lower ( Fig 3C ) . For analysis of virus entry and disassembly , NIH 3T3 cells were treated with 100 IU/ml of IFN-β for 20 h or mock-treated . Cells then were infected with EIL/nLuc/VEEV at 4°C , and further incubated for 1 h at 37°C in the presence of cycloheximide to allow virus entry and disassembly , but not translation of the proteins from the delivered RNAs . Staining of fixed and permeabilized cells with a VEEV capsid-specific mAb demonstrated that in both IFN-β- and mock-treated cells , capsid was present in the cytoplasm ( Fig 3D ) with characteristic accumulation in the nuclear pores [35] . Its distribution and concentrations were indistinguishable in IFN-β- and mock-treated cells . The mAb used was specific to the very amino terminal fragment of capsid and did not interact with assembled nucleocapsids . Thus , the detected staining reflected nucleocapsid disassembly . Taken together , these results strongly suggested that IFN-β treatment does not affect virus entry or disassembly , and thus , inhibition of translation of the incoming mRNAs appears to be a dominant mechanism regulated by IFN-β . These results are generally consistent with the data from RNA transfection-based experiments [25] , but strongly underscore the effect of IFN-β pretreatment on translation of alphavirus-specific RNAs . The additional nLuc SG RNA in the EIL/nLuc/VEEV genome is dispensable for virus replication , and thus , such reporter SG RNA can be modified without affecting synthesis of proteins involved in viral genome replication and packaging in C7/10 cells . Accordingly , we modified the 5'-end of the nLuc-encoding reporter SG RNA to mimic the 5'UTRs of the genomes of different New and Old World alphaviruses ( Figs 4A and S5 ) . To preserve characteristic stem-loop structures that are predicted in the 5'UTRs , we used computer m-Fold predictions and cloned longer ( ~200 nt ) fragments upstream of the nLuc ORF . These fragments contained the natural initiating AUG and encoded the amino terminal fragments of the corresponding nsP1 proteins ( nsP1Δ ) , fused in frame with nLuc ( Fig 4A ) . The additional engineered control constructs included those containing a 5’UTR of β-globin mRNA ( EIL/5'βglo-nLuc/VEEV ) , which was additionally modified to lack a stable stem-loop structure , and EMCV IRES ( EIL/5'EIL-IRES-nLuc/VEEV ) ( Fig 4B ) . The latter sequence was cloned between the EILV SG RNA-specific 5’UTR and nLuc to drive cap-independent translation . All of the recombinant viruses were rescued from the constructs after RNA transfection into C7/10 cells . As described above ( Fig 1C and 1D ) , the released viral particles contained high concentrations of nLuc-encoding reporter SG RNA with engineered 5’UTRs . Next , murine fibroblast cells either were treated with different concentrations of IFN-β or mock-treated , and then infected at the same MOI with the designed recombinant viruses . Translation of the delivered reporter SG RNAs was evaluated at 4 h post infection . This experimental format provided the most robust comparison of the effect of type I IFN pretreatment on translation efficiency . The results were highly reproducible and some are presented in Fig 4C . The following patterns were observed: Collectively , the data suggest that the 5'UTRs of alphaviruses vary in their ability to promote translation in the presence of ISGs induced by IFN-β . However , we cannot rule out an unlikely possibility that the applied IFN-β pretreatment might also differentially affect stability of delivered SG RNAs with different 5’ termini . The features that distinguish alphavirus genomic and SG RNAs from cellular mRNAs are the absence of 2'-O methylation of the penultimate nucleotide of the cap [cap ( 0 ) ] [38] and the presence of a stable stem-loop secondary structure at the 5' termini of viral genomic and SG RNAs [1] . As prior studies reported that IFIT1 functions as a key inhibitor of translation of cap ( 0 ) -containing mRNAs [39] , we experimentally tested the roles of different IFIT protein family members in the development of the type I IFN-induced translation inhibition of alphavirus-specific RNAs . Murine embryonic fibroblasts ( MEFs ) derived from wt , IFIT1-/- , IFIT2-/- , or IFIT locus-/- mice were treated with different concentrations of IFN-β for 20 h and infected with EIL/nLuc/VEEV . The activity of nLuc was assessed at 4 h post infection and normalized to that detected in mock-treated cells . Compared to wt MEFs , the IFIT1-/-counterparts demonstrated markedly less inhibition of translation of the delivered SG nLuc RNA in response to IFN-β pretreatment ( Fig 5A ) . The deletion of the entire locus ( IFIT1 , IFIT2 , and IFIT3 ) did not cause much additional change: IFIT locus-/- and IFIT1-/- MEFs exhibited essentially the same downregulation of nLuc translation . IFIT2-/- MEFs demonstrated an IFN-β-dependent decrease in nLuc expression similar to that detected in wt MEFs ( Fig 5A ) . Thus , the experiments suggest little if any antiviral role for IFIT2 and IFIT3 in this context , despite IFIT2 was readily detectable in the wt and IFIT1-/- MEFs by Western blot , and its expression was strongly induced by IFN-β treatment . Importantly , in contrast to wt MEFs , the IFIT1-/- MEFs did not discriminate between the wt VEEV 3908- and attenuated VEEV TC-83-derived 5’UTRs ( Fig 5B ) . After IFN-β pretreatment , inhibition of translation of the SG RNAs containing either 5’UTR was equally inefficient . Thus , IFIT1 appears to be the primary regulator of translation of the incoming viral RNA among IFIT family members . In the used experimental system , other IFITs did not play detectable role in translation regulation . However , other IFN-inducible factors , which remain to be characterized , account for an additional , 10-fold or more inhibition of translation of alphavirus-specific RNAs upon their natural delivery in viral particles ( Fig 5 ) . They need to be identified and further characterized . As IFIT1 is induced both by IFN-dependent ( canonical ) and IFN-independent ( noncanonical , e . g . , IRF3-dependent ) mechanisms [40] , we analyzed the contributions of these pathways to IFIT1 induction . Both components were clearly detected in microarray-based experiments [17] with IFN-β-treated and noncytopathic VEEV ( VEEV/GFP/C1 ) -infected cells ( Fig 6 ) . The latter virus encodes a mutated capsid protein , which is incapable of inhibiting cellular transcription ( Fig 6A ) [12] . IFIT1 gene expression was strongly induced by the canonical , IFN-β-dependent , pathway: cells treated with IFN-β for 24 h demonstrated a ~75-fold increase in the presence of IFIT1-specific mRNA ( Fig 6B ) . Within 24 h after the cessation of IFN-β-treatment , the levels of IFIT1-specific mRNA returned to baseline . The IFN-independent induction pathway was evident in IFN-α/βR-/- MEFs , which lack the ability to respond to type I IFN treatment . In these cells , VEEV/GFP/C1 developed a persistent infection ( Fig 6C , top panel ) , characterized by an increase in IFIT1 mRNA expression in the absence of type I IFN signaling ( Fig 6C , bottom panel ) . During the persistent stage of virus replication , cells demonstrated 5-10-fold higher than baseline expression levels of IFIT1 , which apparently played a role in replication control , but was insufficient for virus clearance . In comparison , wt cells infected with the mutant VEEV showed clearance at days 5–8 post infection ( Fig 6D , top panel ) . This clearance correlated with higher levels of IFIT1 mRNA accumulation , which were even greater than those detected in IFN-β-treated cells ( Fig 6D , bottom panel ) , indicating a synergistic or additive effects of virus replication and autocrine effects of type I IFN on induction of IFIT1 . The previously described virus reactivation [12 , 17] , occurring at day 9–10 post infection ( Fig 6D ) , was also accompanied with an increase in IFIT1 expression . This apparently contributed to the control of virus replication at lower levels . Our data suggested that IFIT1 had a critical role in inhibiting alphavirus infection and promoting virus clearance from infected cells , and that the antiviral effect may be determined by the level of gene activation/IFIT1 protein expression ( Fig 6 ) . To further test this hypothesis , we generated IFIT1-expressing stable cell lines in wt , IFIT locus-/- , or IFIT1-/- MEFs . Regardless of the cell type used , ectopic expression of IFIT1 negatively affected spread of both SINV/GFP and VEEV/GFP ( Fig 7A ) , derived from attenuated SINV strain Toto1101 and VEEV TC-83 , respectively , and the efficiency of formation of GFP-positive foci ( Fig 7B ) . In all of the generated IFIT1-expressing bulk cell lines , individual cells demonstrated different levels of GFP expression upon infection with these viruses , additionally suggesting the dependence of virus replication on the level of IFIT1 expression . The selection of IFIT1 KI wt MEFs ( IFIT1 KI MEFs ) clones indeed revealed different levels of IFIT1-specific RNA and corresponding protein . For the next series of experiments , we used clones demonstrating IFIT1 expression levels that were similar or lower than that detected in MEFs treated with 100 IU/ml of murine IFN-β for 20 h ( Figs 7C and S6 ) . The levels of VEEV TC-83 replication in the cloned cell lines correlated directly with the efficiency of IFIT1 expression . Levels of IFIT1 expression similar to that in the IFN-treated cells completely blocked replication of VEEV TC-83 ( Fig 7D ) . Clones with lower concentrations of IFIT1 supported virus replication , albeit the titers of released viruses were 3 to 7 orders of magnitude below those detected in wt MEFs . Next , we tested whether the differences in translation of SG RNAs containing 5’UTRs derived from different alphavirus genomes ( Fig 4 ) correlate with resistance of virus replication to IFIT1 expression . The IFIT1 KI/3 cell line ( IFIT1 ectopically expressed in wt MEFs ) , which demonstrated the highest resistance to TC-83-based VEEV/GFP ( Fig 7C and 7D ) , and parental wt MEFs were infected with a panel of alphaviruses , which included wtVEEV 3908 , VEEV/GFP ( VEEV TC-83 vaccine strain-based virus ) , NA EEEV , SINV/GFP ( Toto1101 laboratory strain-based virus ) , AR SINV/GFP ( SINV/GFP with 5’UTR derived from natural isolate of SINV AR339 ) , SFV , CHIKV/GFP and unrelated viruses ( EMCV and VSV ) at the same MOI . Titers of the released viruses were determined at 24 h post infection , and CPE was evaluated ( Fig 8A ) . Consistent with the translation experiments , among the wt alphavirus isolates , NA EEEV replicated poorly in MEFs that ectopically expressed IFIT1 . Other alphaviruses with wt 5’UTRs ( wtVEEV 3908 , AR SINV/GFP , SFV and CHIKV/GFP ) demonstrated greater inherent resistance to IFIT1 and higher levels of replication . SFV was the least inhibited by IFIT1 in terms of infectious virus release and CPE development in IFIT1 KI/3 cells . Importantly , tissue culture-adapted variants of VEEV ( VEEV/GFP ) and SINV ( SINV/GFP ) , containing known point mutations in their 5'UTRs resulting from passaging of wt viruses in cultured cells , were sensitized to the antiviral effects of IFIT1 . The control , unrelated viruses EMCV ( picornavirus ) and VSV ( rhabdovirus ) replicated equally well in parental and IFIT1 KI/3 MEFs , confirming that IFIT1 does not substantially inhibit viruses that utilize IRES-dependent translation or have cap 1 structures on their mRNA [41 , 42] , and that ectopic expression of IFIT1 does not non-specifically inhibit virus replication . Even though some of the wt alphaviruses , particularly SFV , showed inherent resistance to IFIT1 , they did not form plaques on IFIT1 KI/3 MEFs ( Fig 8A ) . This decreased ability to spread and cause cell death suggested a possible second cell defense mechanism associated with IFIT1 expression . Indeed , IFIT1 KI/3 MEFs released high levels of IFN-β upon infection with viruses having wt 5’UTRs ( Fig 8B ) . Remarkably , the amounts of IFN-β induced by these viruses were far greater in IFIT1-expressing cells than in wt MEFs despite the orders of magnitude lower yield of virus release . Viruses having 5’UTRs derived from tissue culture-adapted variants VEEV TC-83 and SINV Toto1101 , demonstrated the opposite pattern of IFN-β induction: low , but detectable , levels of IFN-β in wt MEFs and undetectable levels of IFN release in IFIT1 KI/3 MEFs . Some of the variants with wt 5’UTRs ( AR SINV/GFP and CHIKV/GFP ) were designed to express GFP from a second SG promoter . In wt MEFs , they produced GFP at levels readily detectable by fluorescence microscopy within 2 to 3 h post infection . However in IFIT1 KI/3 MEFs , the production of GFP was delayed by 8–10 h , suggesting markedly lower levels of virus replication . Thus , one explanation for the higher levels of IFN-β detected in IFIT1 KI/3 MEFs is that lower levels of virus replication in IFIT1-expressing cells , compared to wt MEFs , renders wt viruses less capable of interfering with cellular transcription and induction of IFN-β . Consequently , in the plaque assay the released type I IFN acted to protect uninfected cells from the slowly released viruses , which inhibited plaque formation . The experiments evaluating the role of 5’UTRs on RNA translation ( Fig 4 ) suggested that the VEEV TC-83-specific 5'-terminus makes the RNA more sensitive to type I IFN pretreatment than wt VEEV 5’UTR . However , the TC-83-specific 5'UTR is predicted to retain a considerable stem-loop structure ( Fig 9A ) that may weakly antagonize IFIT1 , as the modified β-globin 5’UTR lacking secondary structure at the 5' end demonstrated even greater sensitivity to IFN-β pretreatment ( Fig 4C ) . Based on these data and the results of our previous studies of the 5'-promoter structure and function [43–45] , we designed an artificial 5’UTR in the VEEV TC-83 background ( Fig 9A ) . This sequence was predicted ( i ) to remain functional as a promoter for RNA synthesis in mammalian cells due to the presence of multiple AU repeats , but ( ii ) to be a less efficient promoter in insect cells due to its lack of a 5'-terminal stem-loop structure , and ( iii ) to be more sensitive to IFIT1-specific inhibition because of its long 5'-terminal RNA fragment , which is not involved in base pairing [46] . The engineered construct 5’mutVEEV was viable and , based on infectious center assay , did not require additional adaptive mutations for replication . The rescued 5'mutVEEV replicated in wt MEFs with efficiency similar to that of the parental VEEV TC-83 ( Fig 9A ) . However , in IFIT1 KI/1 MEFs expressing the lowest level of IFIT1 , titers of released 5'mutVEEV were 20 to 200-fold lower than those of VEEV TC-83 . Accordingly , in contrast to the parental virus , the designed mutant did not cause noticeable CPE in IFIT1 KI/1 cells ( Fig 9B ) . An additional characteristic of this mutant , as predicted , was less efficient replication in mosquito cells ( Fig 9C ) . To extend these results , another set of 5'UTR mutants was designed based on the CHIKV/GFP ( 181/25 strain ) virus . Since knowledge of the CHIKV 5’UTR promoter is less advanced compared to SINV and VEEV , we introduced only 1 or 2 point mutations ( CHIKVmut1/GFP and CHIKVmut2/GFP , respectively ) , which destabilized the base of the predicted 5'-stem , but were not expected to affect the overall predicted secondary structure ( Fig 9D ) . Both viruses were viable , and became sensitive to IFIT1 . These CHIKV mutants replicated to the same level as parental virus in wt MEFs ( Fig 9C ) , but did not replicate productively in either IFIT1 KI/2 or IFIT1 KI/3 MEFs: no GFP expression was detected , and virus titers remained at the level of detection , 3–5 orders of magnitude below the corresponding titers of parental CHIKV/GFP 181/25 . Thus , a knowledge of both promoter structure and the antiviral function of IFIT1 enabled us to design increasingly attenuated variants of CHIKV and VEEV . To confirm additional attenuation , we infected s . c . 6-day-old NIH Swiss mice with parent VEEV TC-83 and the 5'mutVEEV . All of the TC-83-infected mice failed to gain weight and succumbed to the infection within 6 days ( Fig 9E and 9F ) . In comparison , only few of the mice infected with the same or higher doses of 5'mutVEEV died within 14 days of infection . 5'mutVEEV-infected mice still developed disease , which was detectable by their slower growth compared to those sham-infected , but most of them recovered .
To date , type I IFNs remain the most potent , broad-spectrum antiviral agents . IFN treatment of cells induces a large set of ISGs , which prevent infection with many viral pathogens , including alphaviruses . The list of currently known ISGs includes more than 300 members , but the exact mechanisms of function in inhibition of virus replication have been identified for only a small subset [18–24] . The detailed mechanistic investigation of the functions of individual ISGs is complicated by the difficulty of dissecting particular processes in virus replication independently of one another . The recently discovered and characterized EILV alphavirus provided us with an opportunity to evaluate the effects of type I IFN pretreatment on the very early steps of the alphavirus replication cycle and to study them independently of viral RNA replication . Chimeras of EILV and VEEV cannot replicate in vertebrate cells , but in mosquito cells they package the engineered nLuc- or GFP-encoding SG RNAs into VEEV protein-encased particles , which then efficiently deliver them into vertebrate cells ( Fig 1 ) . Thus , such chimeras represent an efficient means of delivery of engineered SG RNAs into many cell types , and most importantly , such RNAs are delivered via the natural virus entry pathway . In these SG RNAs , 5’UTRs can be designed to completely mimic the 5'-ends of alphavirus genomes and then tested for promoting translation of reporter genes under different experimental conditions ( Fig 4 ) . Using this experimental system , we found that IFN-β pretreatment could block alphavirus infection before the beginning of RNA replication . IFN-β did not affect the steps of virus entry and virion disassembly ( Fig 3 ) , but as suggested previously [25] , strongly inhibited translation of the incoming alphavirus RNA . The inhibitory effect depended on the time of IFN pretreatment and IFN-β concentration ( Fig 2 ) . The efficiency of translation of the engineered SG RNAs in the IFN-β-treated cells was also found to strongly depend on the origin of their 5’UTR ( Fig 4 ) . Translation of the RNAs containing 5’UTRs from the genomes of natural isolates of the Old World alphaviruses , SINV , CHIKV and SFV , in particular , was more IFN-resistant than from templates having non-alphaviral 5’UTRs . Among the New World alphaviruses , the 5’UTRs of NA or SA EEEVs were more sensitive to type I IFN , in terms of their ability to promote RNA translation , than 5’UTR derived from a wt epizootic VEEV genome . The previously described mutation of nt 3 in the VEEV TC-83 5’UTR made translation of this control template less resistant to IFN-β pretreatment , and this effect correlated with the previously described IFN-sensitive phenotype of the attenuated vaccine strain of VEEV [37] . The experimental data also suggested the existence of at least two mechanisms by which IFN-specifically inhibited translation of the alphavirus templates . One was determined by IFIT1 expression , and the second occurred through an IFIT1-independent pathway . The latter mechanism affected translation of the delivered capped RNAs in IFIT1-/- MEFs , and downregulated translation of the EMCV IRES- rather than cap-containing templates in wt MEFs ( Figs 4 and 5 ) . It may be based on PKR/eIF2a- and/or PARP-mediated mechanism of translation regulation [22 , 47] . IFIT1 is a member of a family of IFN-induced proteins with tetratricopeptide repeats [21] . Four IFIT family members have been characterized extensively in humans ( IFIT1 ( also known as ISG56 ) , IFIT2 ( ISG54 ) , IFIT3 ( ISG60 ) and IFIT5 ( ISG58 ) ) and three members are expressed in mice: Ifit1 ( Isg56 ) , Ifit2 ( Isg54 ) , Ifit3 ( Isg49 ) . IFIT1 has been described as a protein that binds viral RNA cap structures lacking 2'-O methylation , which results in inhibition of binding of eukaryotic translation initiation factors and ultimately , downregulation of virus replication [41 , 46 , 48 , 49] . Recent studies indicate that IFIT1 primarily interferes with the interaction of eIF4E with the cap structure [50] . Subsequently , it was shown that secondary structural motifs at the 5'-end of wt VEEV genomic RNA interfere more efficiently with IFIT1 binding than the mutation-containing 5'UTR of the attenuated VEEV TC-83 strain [39] . In our study , the previously identified difference in IFIT1 binding to wt and TC-83-specific 5’UTRs strongly correlated with IFN-β-induced translation inhibition of the RNA templates having the same 5'-termini . This difference in translation inhibition was no longer detectable in IFIT1-/- MEFs ( Fig 5 ) , suggesting that type I IFN-induced expression of IFIT1 has a critical role in translational block of alphavirus infections and can discriminate between wt and mutated 5’UTRs . The results in IFIT1-transfected IFIT-locus-/- MEFs ( Fig 7 ) suggest that IFIT1 can function independently of other IFIT proteins to attenuate translation of alphavirus RNAs . Thus , other IFIT members have at best subordinate roles in inhibiting the translation of alphavirus RNAs . This inhibitory effect varies with the ability of IFIT1 to bind to RNA displaying 5'-ppp moieties , where a complex of IFIT1 , IFIT2 , and IFIT3 putatively is required [51] . IFIT1 expression in MEFs is induced by both the type I IFN-dependent pathway and replication of alphavirus-specific RNAs ( Fig 6 ) . Resistance of translation of the mRNA templates containing 5’UTRs derived from different alphaviruses in IFN-β-treated cells correlated with the efficiency of replication of the corresponding viruses in the presence of higher levels of IFIT1 ( Figs 4 and 8 ) . CHIKV , SINV , SFV and VEEV with natural wt 5’UTRs replicated in IFIT1-expressing cell lines more efficiently than other alphaviruses ( Fig 8 ) . At a high MOI , SFV , the most resistant alphavirus , also induced detectable CPE . However , the direct interference with translation of alphavirus genomes appears to be not the only mechanism by which IFIT1 affects virus replication . Spread of IFIT1-resistant viruses with wild type 5’UTRs was inhibited by a second mechanism: in the presence of higher levels of IFIT1 , these replicating viruses became potent inducers of type I IFN ( Fig 8 ) . In IFIT1-expressing cells , low titers of released viruses and very slow rates of accumulation of GFP indicated inefficient replication and expression of the encoded proteins . Capsid and nsP2 proteins of the New World and the Old World alphaviruses , respectively , function stoichiometrically to inhibit cellular transcription [35 , 52] . Thus , in the presence of higher levels of IFIT1 , the inefficiently replicating wt viruses likely were sensed by cellular pattern recognition receptors , such as RIG-I and MDA5 . However , slow accumulation of viral proteins with transcription inhibitory functions ( nsP2 or capsid ) rendered them less capable of interfering with type I IFN activation . The released IFN-β activated the antiviral state in as yet uninfected cells and also probably had an autocrine effect on already established virus replication . Very inefficient replication of attenuated strains of VEEV ( TC-83 ) and SINV ( Toto1101 ) in the same IFIT1 KI/3 MEFs made the latter viruses , in turn , incapable of IFN-β induction at all ( Fig 8 ) . Computer predictions and biochemical data indicate that a distinguishing characteristic of the 5’UTRs in alphavirus genomes is the presence of short stem-loop structures at the 5'-termini . A stem-loop structure at the 5'-end of VEEV has been confirmed by enzymatic analysis [53] and by nuclear magnetic resonance imaging [39 , 53] . The nucleotide sequences and computer-predicted stem-loop structures differ among alphavirus species , but demonstrate a high degree of conservation between geographically distant isolates . Thus , as shown for VEEV and SINV [39] , these secondary structures likely regulate the efficiency of interaction with IFIT1 with the 5'-terminal cap ( 0 ) and ultimately , the efficiency of translation of viral genomes in type I IFN-activated cells . Based on the accumulated experimental data , the molecular mechanism of this regulation of IFIT1-cap ( 0 ) interaction appears to be straightforward . To make alphaviruses capable of replication and IFIT1-resistant , the first two nucleotides need to be AU , and to be followed by a G-C-rich stem . Mutation of nt3 , G to A , in the case of VEEV for example , or the presence of the wt AUA in both NA and SA EEEVs either makes first nucleotides unpaired ( VEEV TC-83 ) or has strong negative effect on the stability of the hairpin base ( EEEV ) . This in turn increases the efficiency of IFIT1 binding and reduces the viral resistance to IFIT1-mediated inhibition of translation . In contrast to other alphaviruses , EEEV is a poor inducer of type I IFN in vivo [10 , 54] . Consequently , in the absence of type I IFN-induced IFIT1 induction , there is little selective pressure for EEEV to evolve an IFIT1 binding-resistant stem-loop in the 5’UTR . The only alphavirus examined , which does not follow this rule , is SINV . Its 5'UTR starts with AUU , but the two uridine residues are predicted to form an RNA stem , with the GG nucleotides that follow downstream . This presence of U instead of more standard A in position 3 also might make binding of the first 3 nt to IFIT1 less efficient . Destabilization of following two G-C pairs by G5A mutation in Toto1101 affects the stability of the entire stem , promotes IFIT1-RNA interaction and makes SINV more sensitive to type I IFN [39] and the antiviral effects of IFIT1 . The existence of a simple means to reduce sensitivity to IFIT1 by increasing the stem stability raises the question as to why all alphaviruses have not evolved greater resistance to this ISG product . Indeed , we readily selected a VEEV mutant capable of more efficient replication in the presence of IFIT1 . However , it should be noted that the IFIT1 protein is only one of many members of the highly redundant system of the ISG products . Its contribution to development of IFN-induced antiviral state is limited , and expression of other ISGs has deleterious effect on alphavirus replication; thus , selection of mutants with a maximal resistance level to IFIT1 appears not to be sufficiently beneficial for virus replication . There are also alternative explanations for why there is a limit to IFIT1 resistance in the 5' end . ( A ) Mutations in the 5'UTR that enhance the amount of double-stranded RNA character could sensitize viruses to other pathogen recognition receptors ( e . g . RLR or TLR ) or antiviral ISG products . ( B ) The 5'-terminal sequences in alphavirus genomes have other key functions including ( i ) facilitating translation of the nsPs , ( ii ) encoding promoter elements for synthesis of the negative strand of the viral genome , and ( iii ) their complementary sequence in the negative-strand RNA intermediate encodes promoters for synthesis of the positive strand [43–45 , 53] . Thus , higher resistance to IFIT1 , determined by the stability of stem-loop structures in the 5’UTRs , may be not beneficial for virus replication in both vertebrate hosts and/or mosquito vectors . The increase in resistance of alphaviruses to IFIT1 may have great cost in terms of reduction of promoter function and efficiency of viral replication . Importantly , the 5'-terminus of EILV genome is also predicted to form stable stem-loop ( S5 Fig ) , but this virus does not replicate in vertebrates and thus , never interacts with IFIT1 . However , the stem-loop structure of the EILV 5’UTR is similar to that of other alphaviruses , additionally suggesting its importance for promoter and perhaps other as yet undiscovered functions . One of them might be inhibition of cap ( 0 ) interaction with possible mosquito IFIT1 ortholog , which stays to be discovered . Passaging of the alphaviruses in cultured cells , traditionally used to adapt them for more efficient replication , represents an interesting selection system . During serial passaging of alphaviruses , such as the 83 passages performed for development of the vaccine strain VEEV TC-83 [55 , 56] , resistance to IFIT1 no longer plays a significant role . As a result , VEEV TC-83 evolved an adaptive mutation in the 5’UTR [57 , 58] , which increased the rates of replication in cultured cells [53] . Thus , passaging led to enrichment of virus populations with variants adapted not only to binding to the heparan sulfate-containing moieties on the cell surface [59] , but also to more efficient spread and growth to higher titers . This mutation in the 5’UTR , however , made VEEV less resistant to type I IFN [37] . Apparently , similar selection occurred in the process of development of the laboratory strain of SINV Toto1101 [60] . The accumulated and detailed knowledge about alphavirus promoter structures and functions [43–45 , 53] , and the new data from this and other studies regarding the resistance of alphaviruses to IFIT1 , provide an opportunity to design new types of attenuated viruses ( Fig 9 ) . For instance , the engineered 5’mutVEEV replicated efficiently in mammalian , but not mosquito , cells and was more sensitive to the antiviral effects of IFIT1 . This mutant virus was more attenuated than VEEV TC-83 , which at least partially retains the secondary structure of the VEEV-specific 5’UTR . It also demonstrated higher level of attenuation in 6-day-old mice . Similarly , the re-engineered CHIKV 181/25 variants with mutated 5’UTR exhibited higher sensitivity to IFIT1-mediated inhibition of replication . In summary , the results of this study demonstrate that ( i ) type I IFN pretreatment blocks translation of the alphavirus RNAs delivered in infectious virions , but not other early events in alphavirus infection . ( ii ) The IFN-induced inhibition of translation is determined by both IFIT1-dependent and IFIT1-independent mechanisms . ( iii ) Alphaviruses vary in their ability to replicate in the presence of IFIT1 , and this difference is determined by the structures of the 5’UTRs in their genomes . ( iv ) The presence of IFIT1 at higher levels makes IFIT1-resistant wt alphaviruses more potent inducers of type I IFN . Thus , IFIT1 acts as both an antiviral effector molecule and inducer of innate immunity against alphaviruses . ( v ) Other members of IFIT family have subordinate roles in the inhibition of alphavirus replication . ( vi ) Our data provide a plausible explanation of the mechanism of alphavirus attenuation during their serial passaging in vaccine development . ( vii ) A mechanistic understanding of alphavirus 5’UTR promoter and immune evasion functions allows engineering of extensive , irreversible changes into 5’ termini of alphavirus genomes . These modifications affect virus replication in mosquito , but not mammalian cells , make such variants attenuated and can be used as one of the means in vaccine development .
NIH 3T3 cells were obtained from the American Type Culture Collection ( Manassas , VA ) . They were maintained in alpha minimum essential medium ( αMEM ) supplemented with 12 . 5% fetal bovine serum ( FBS ) and vitamins at 37°C . BHK-21 cells were provided by Paul Olivo ( Washington University , St . Louis , MO ) and were cultured in αMEM supplemented with 10% FBS and vitamins . Mosquito C7/10 cells were obtained from Henry Huang ( Washington University , St . Louis , MO ) and propagated at 30°C in Dulbecco’s modified Eagle’s medium ( DMEM ) supplemented with 10% heat-inactivated FBS and 10% tryptose phosphate broth ( TPB ) . Wild type , IFIT1-/- , IFIT2-/- and IFIT-locus-/- murine embryonic fibroblasts ( MEFs ) were generated and maintained according to published protocols [61] in DMEM supplemented with 10% FBS . The IFIT-locus-/- mice used to produce MEFs were generated by Ganes Sen ( Cleveland Clinic , Cleveland OH ) , provided as a generous gift , and will be described elsewhere . The original plasmids containing the infectious cDNAs of VEEV TC-83 ( pVEEV and pVEEV/GFP ) , pVEEV 3908 , SINV Toto1101 ( pSINV ) , SINV/GFP Toto1101 ( pSINV/GFP ) , CHIKV 181/25 ( pCHIKV ) and Eilat virus ( pEILV ) were described elsewhere [12 , 30 , 62–65] . Other plasmids were designed using standard PCR-based techniques . The introduced modifications are described in the Results . Sequences of the plasmids and details of the cloning procedures can be provided upon request . Plasmids containing the complete viral genomes were purified by CsCl gradient centrifugation and then linearized using the unique restriction sites located downstream of the poly ( A ) tails in the cDNA copies of viral genomes . The corresponding RNAs were synthesized in vitro using SP6 RNA polymerase in the presence of a cap analog as previously described [66] . The yield and integrity of the transcripts were analyzed by agarose gel electrophoresis under non-denaturing conditions . The transcription mixtures were used for transfection without further RNA purification . Viruses were rescued either by electroporation of the in vitro-synthesized RNAs into BHK-21 or C7/10 cells [44] . Viruses were harvested at 24 h post transfection of BHK-21 cells or at 48 h post transfection of C7/10 cells . Virus titers were determined using a standard plaque assay on C7/10 or BHK-21 cells [44 , 67] . For some of the mutants , we assessed the infectivity of RNA by infectious center assay . In this assay , ten-fold dilutions of electroporated BHK-21 cells were seeded into 6-well Costar plates containing subconfluent , naïve BHK-21 cells . After 2 h of incubation at 37°C , media were replaced by 2 ml of MEM containing 0 . 5% of agarose and 3% FBS . Plaques were stained with crystal violet after 48 h of incubation at 37°C and infectivity was determined in PFU per μg of transfected RNA . The experiments with pathogenic alphaviruses were performed in the BSL3 facility of the UAB SEBLAB according to the IBC-approved protocols . Subconfluent C7/10 cells in 150-mm dishes were infected with EIL/VEEV or other chimeric viruses , containing nLuc or GFP genes with varying 5’UTRs under control of an additional SG promoter , at an MOI of 10 PFU/cell for 1 h at 30°C . Then cells were incubated overnight in complete media and then in serum-free VP-SF media ( Invitrogen ) for 24 to 30 h . These media were harvested before cytopathic effect ( CPE ) developed and additionally clarified by low-speed centrifugation . Viral particles were then concentrated using Amicon Ultra 100K centrifugal filters ( Millipore ) as described elsewhere [68] . In most experiments , viruses were used without further purification . In some experiments , concentrated samples were purified further by ultracentrifugation in continuous 20–50% sucrose gradients in a SW-40 rotor at 38 , 000 rpm for 3 h at 4°C . Visible bands of viral particles were collected , diluted in PBS , and concentrated by ultracentrifugation in discontinuous 25–50% sucrose gradients . Bands were collected diluted in PBS supplemented with 1% FBS , and virus titers were determined by standard plaque assay on C7/10 cells [44] . C7/10 cells were infected with EIL/GFP/VEEV at an MOI of 20 PFU/cell and incubated at 30°C for 16 h . Viral RNAs were then labeled with [3H]uridine ( 50 uCi/ml ) between 16 and 24 h post infection in the presence of 1 μg/ml of Actinomycin D . RNAs were isolated from the cells and viral particles recovered from the supernatant by ultracentrifugation through 25% sucrose , and analyzed by agarose gel electrophoresis under denaturing conditions [69] . For the analysis of intracellular concentration of virus-specific RNAs , relative concentration of IFIT1-specific mRNA , and analysis of RNA content in the released viral particles , RNA were isolated using the RNeasy minikit ( Qiagen ) . cDNA were synthesized using the QuantiTect reverse transcription kit ( Qiagen ) . Quantitative PCR was performed using the SsoFast EvaGreen Supermix ( Bio-Rad ) in a CFX96 real-time PCR detection system ( Bio-Rad ) for 40 cycles . The specificity of the quantitative PCR was confirmed by analyzing the melting temperatures of the amplified products . The efficiency of each pair of primers was determined using the standard curves obtained by performing real-time PCR on 10-fold dilutions of a control sample . For RNA samples isolated from the cells , the qPCR reactions were performed in parallel with primers specific to β-actin for normalization , and the fold difference in RNA concentration was calculated using the ΔΔCT method . Each qPCR was performed in triplicate , and the means and standard deviations were calculated . For RNAs isolated from viral particles , RT-qPCR was performed as described above using primers specific to the EILV nsP region , GFP , or VEEV E2 gene in parallel . The absolute copy number of each amplicon was determined based on a standard curve obtained using cDNA from in vitro synthesized EIL/GFP/VEEV genomic RNA , containing a known amount of RNA copies . The number of genomic RNA copies was obtained with the EILV-nsP primers . The number of SG RNA 1 copies represents the number of copies obtained determined with GFP primers after subtraction of the number of G RNA copies . Finally , the number of SG RNA 2 represents the number of copies determined using the primers specific to the structural E2 gene with subtraction of the number of copies determined by using GFP-specific primers . NIH 3T3 , MEFs or C7/10 cells were seeded into 6-well Costar plates and infected at MOIs indicated in the figure legends . After 1 h incubation at 37°C ( for NIH 3T3 and MEFs ) or 30°C ( for C7/10 ) , cells were washed twice with PBS , overlaid with 1 ml of complete media and further incubated at the cell-specific temperatures . At the indicated times post infection , media were replaced . Virus titers in the harvested samples were determined by standard plaque assay on BHK-21 or C7/10 cells , as indicated in the figure legends . The IFIT1-encoding ORF was synthesized by RT-PCR from RNA isolated from NIH 3T3 cells , which were pre-treated with 500 IU/ml of IFN-β for 24 h . It was cloned into a modified PiggyBAC plasmid under control of the CMV promoter . MEF and IFIT locus-/- MEF ( 3 x 105 cells/well ) were seeded in 6-well plates and transfected with the plasmid using the TransIT-3T3 kit according to the manufacturer’s instructions ( Mirus ) . After blasticidin selection , cells were either used as a pool or further cloned to isolate clones with different levels of IFIT1 expression . Three clones , IFIT1 KI/1 , IFIT1 KI/2 and IFIT1 KI/3 , demonstrating different levels of IFIT1 by qPCR and Western blot , were used for further experiments . NIH-3T3 cells or MEFs in the 6-well Costar plates were either treated with IFN-β or remained mock-treated as described in the figure legends . Cells were infected with EIL/5’nLuc/VEEV variants at an MOI of 10 PFU/cell . For the viruses encoding nLuc with a modified 5’UTR , in some cases , MOIs were adjusted to obtain equivalent amounts of nLuc activity in the mock-treated , infected cells . Cells were infected for 1 h ( see figure legends ) , washed with PBS and overlaid with 1 ml of complete media . At the specified time points , cells were lysed in reporter lysis buffer ( Promega ) and frozen at -80°C . After thawing , cell lysates were clarified by centrifugation for 1 min at 16 , 000 x g , and nLuc activity was determined using the Nano-Glo luciferase assay system ( Promega ) according to the manufacturer’s instructions . For all experiments , one of at least three independent and reproducible experiments is presented . The lysates of IFIT1 KI/1 , IFIT1 KI/2 , IFIT1 KI/3 and wt MEFs were treated overnight with 500 IU/ml of IFN-β , were analyzed by 10% SDS-PAGE . Proteins were transferred on nitrocellulose membrane and stained with mouse IFIT1-specific mAb ( clone ISG56/13 ) followed by AlexaFluor800-labeled secondary Abs . Fluorescence intensities were analyzed and quantified on a LI-COR imager . NIH-3T3 cells were seeded in 8-well Ibidi chambers , and either treated or mock-treated with 100 IU/ml of IFN-β for 20 h at 37°C . To allow viral particles adsorption without internalization , cells were incubated with VEEV TC-83 for 1 h at 4°C , washed with cold PBS and immediately fixed with 4% paraformaldehyde ( PFA ) . Adsorbed viral particles were stained using VEEV TC-83-specific mouse Abs ( gift of Robert Tesh , UTMB ) and AlexaFluor555-labeled secondary Abs . Cell nuclei were stained with Hoechst dye . Images were acquired on a Zeiss LSM700 confocal microscope with a 63X 1 . 4NA PlanApochromat oil objective . The 3D image stacks were further processed using Huygens Professional ( Scientific Volume Imaging , Hilversum , Netherlands ) for deconvolution , using experimental PSF , and Imaris for 3D rendering ( Bitplane AG , St . Paul , MN ) . Spot function of Imaris was used for quantitative analysis of the numbers of attached virions . NIH-3T3 cells were seeded into 8-well Ibidi chambers and either treated or mock-treated with 100 IU/ml of IFN-β for 20 h . Cells were incubated with concentrated virus for 1 h at 4°C to allow virus adsorption to the cells , then washed and incubated for 1 h to 37°C in medium supplemented with 50 μg/ml of cycloheximide to allow entry and nucleocapsid disassembly in the absence of translation and replication of the incoming viral RNAs . Cells were then fixed with 4% PFA and stained using a rat mAb specific to the amino-terminal fragment of capsid protein , which is not exposed in assembled nucleocapsids , and AlexaFluor555-labeled secondary Abs . Nuclei were stained with Hoechst dye . Images were acquired on a Zeiss LSM700 confocal microscope with a 63X 1 . 4NA PlanApochromat oil objective . Images were assembled in Imaris ( Bitplane AG , St . Paul , MN ) . Concentrations of IFN-β in the culture supernatants of infected cells were quantified using the VeriKine Mouse IFN Beta enzyme-linked immunosorbent assay ( ELISA ) kit ( PBL Interferon Source ) according to the manufacturer’s instructions . NIH 3T3 cells and IFN-α/βR-/- MEFs were infected with VEEV/GFP/C1 at an MOI of 20 PFU/cell or treated with 1000 IU/ml of IFN-β . At the indicated time points , total RNA was isolated using TRIzol according to the manufacturer’s instructions ( Invitrogen ) and additionally purified with the RNeasy minikit ( Qiagen ) . The cDNA synthesis , labeling , hybridization on Mouse Gene 1 . 0 ST Array GeneChips ( Affymetrix ) and image processing were performed at the Heflin Center Genomics Core facility ( UAB ) . Two independent RNA samples were prepared for each indicated time point for VEEV/GFP/C1-infected cells and three RNA samples for IFN-β-treated cells . The robust multichip average ( RMA ) algorithm was used to normalize the raw intensity values using the GeneSpring software program , version GX 11 . 5 ( Agilent ) . To assess the residual virulence of the designed variant of VEEV TC-83 , 5’mutVEEV , 6-day-old NIH Swiss mice ( Harlan ) were inoculated subcutaneously ( s . c . ) with 2x105 or 106 PFU of the viruses diluted in PBS in a volume of 20 μl . Animals were checked twice daily for signs of the disease or death . Weight was evaluated on a daily basis . The animal studies were carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The protocol was approved by the Institutional Animal Care and Use Committee of the University of Alabama at Birmingham ( Project Number: 09469 ) . | Alphaviruses represent a group of highly important human pathogens , which are transmitted by mosquito vectors between vertebrate hosts . Alphavirus replication in vertebrates depends on their ability to interfere with host antiviral responses on both cellular and organismal levels . The identification of cellular factors , which affect virus replication , and characterization of their functions may prove crucial for the design of new effective vaccine candidates . We have demonstrated that the protein product of one of the interferon-stimulated genes , IFIT1 , is a potent inhibitor of translation of the incoming alphavirus genomes and ultimately , virus replication . The secondary structure of the 5’untranslated regions ( 5’UTRs ) of alphavirus genomes was shown to play a critical role in alphavirus resistance to this inhibitory effect . Moreover , in IFIT1-expressing cells , wt alphaviruses exhibiting low sensitivity to IFIT1 also were found to induce high levels of type I IFN . Altogether , our data show that alphavirus 5’UTRs were evolutionarily selected to meet the requirements of both functioning as promoters for positive- and negative-strand RNA synthesis and supporting the resistance to inhibitory effects of IFIT1 . We further exploited this new knowledge to develop mutated alphaviruses , which displayed higher sensitivity to IFIT1 and more attenuated phenotypes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | IFIT1 Differentially Interferes with Translation and Replication of Alphavirus Genomes and Promotes Induction of Type I Interferon |
Mitochondria cannot form de novo but require mechanisms that mediate their inheritance to daughter cells . The parasitic protozoan Trypanosoma brucei has a single mitochondrion with a single-unit genome that is physically connected across the two mitochondrial membranes with the basal body of the flagellum . This connection , termed the tripartite attachment complex ( TAC ) , is essential for the segregation of the replicated mitochondrial genomes prior to cytokinesis . Here we identify a protein complex consisting of three integral mitochondrial outer membrane proteins—TAC60 , TAC42 and TAC40—which are essential subunits of the TAC . TAC60 contains separable mitochondrial import and TAC-sorting signals and its biogenesis depends on the main outer membrane protein translocase . TAC40 is a member of the mitochondrial porin family , whereas TAC42 represents a novel class of mitochondrial outer membrane β-barrel proteins . Consequently TAC40 and TAC42 contain C-terminal β-signals . Thus in trypanosomes the highly conserved β-barrel protein assembly machinery plays a major role in the biogenesis of its unique mitochondrial genome segregation system .
Mitochondria are a hallmark of eukaryotic cells [1] . They derive from an endosymbiotic event between an archaeal host cell and an α-proteobacterium . The bacterial symbiont was subsequently converted into an organelle . Continued evolution since the origin of the mitochondrion , approximately 1 . 5–2 billion years ago , has led to a great diversification of the organelle [2 , 3] . This is illustrated by the immense variation of the morphology and the behaviour of mitochondria in different species and the large variation in the organization and coding content of their genomes . However , faithful transmission of mitochondria and their genomes to their daughter cells is a problem essentially all eukaryotic cells need to solve [4 , 5] . In contrast to most other eukaryotes trypanosomes and their relatives have a single mitochondrion that only contains a single unit mitochondrial genome , termed kinetoplast DNA ( kDNA ) . The kDNA consists of two genetic elements the maxi- and the minicircles . The maxicircles are present in 25–50 copies and are 22 kb in length [6] . They contain a number of protein-coding genes expected to be present in the mitochondrial genome . Most of them are cryptogenes whose primary transcripts have to be edited by multiple uridine insertions and or deletions to become functional mRNAs . The minicircles are heterogenous in sequence , occur in several thousand copies and encode the guide RNAs that provide the information for RNA editing [7 , 8] . Maxi- and minicircle are highly topologically interlocked and build a large disc-shaped network that is physically linked to the basal body of the single flagellum via a structure termed tripartite attachment complex ( TAC ) . The three zones of the TAC include the unilateral filaments that connect the kDNA network to the inside of the inner membrane ( IM ) , a segment containing tightly apposed detergent-resistant differentiated IM and outer membranes ( OM ) and finally the exclusion zone filaments that build a bridge from the OM to the basal body [9–11] . Due to the single unit nature of the kDNA network , its replication and segregation need to be tightly coordinated . kDNA replication occurs at a precise stage of the cell cycle immediately before the onset of the S phase in the nucleus [12 , 13] . During replication the kDNA doubles in size forming a dumbbell-shaped network . The process has been studied in detail and involves numerous protein factors . The segregation of the replicated kDNA networks depends on an intact TAC . Thus during kDNA replication a new TAC forms that connects the basal body of the new flagella with the replicating kDNA network . Division of the replicated kDNA network finally is linked to the segregation of the old and the new basal bodies . Initially the two kDNA networks remain connected by a structure termed "nabelschnur" which becomes resolved when the distance between the two networks is larger than 1 μm . The segregation then continues and the basal bodies with the attached kDNA networks move further apart [14 , 15] . Subsequently , prior to cytokinesis , the mitochondrion is divided in two , the plane of division intersecting between the two kDNA networks [16] . Presently only a few components of the TAC have been identified . The first one was p166 [17]: it contains a single predicted transmembrane domain ( TMD ) which however is not essential for its localization making it difficult to decide whether p166 is a TAC protein of the IM or the unilateral filaments . TAC102 is a component of the unlilateral filaments [18] , whereas p196 [19] and TAC65 [20] were shown to be extramitochondrial TAC subunits that localize to the exclusion zone filaments . The same was the case for an as yet unkown antigen recognized by the monoclonal antibody Mab 22 [21] . Furthermore , TAC40 and peripheral archaic translocase of the OM 36 ( pATOM36 ) , two OM proteins that localize to the TAC , have also been described [20 , 22] . TAC40 is a β-barrel protein of the mitochondrial porin family . pATOM36 is unusual as it is localized in the TAC region but also present all over the OM . The dual localization of pATOM36 reflects its dual function in kDNA inheritance and in the biogenesis of a subset of α-helically anchored OM proteins including most subunits of the archaic translocase of the OM ( ATOM ) [20 , 23] . Complementation experiments revealed that the two functions are distinct , since the C-terminus is only essential for the biogenesis of OM proteins but not for the segregation of the kDNA . Thus , pATOM36 is an integrator of mitochondrial protein import and mitochondrial genome inheritance [20 , 24] . Unlike other OM proteins , the OM subunits of the TAC not only need to be targeted to mitochondria and inserted into OM , but they also have to be sorted to the region of the OM close to the single unit kDNA , where a new TAC is being formed . Here we have discovered two novel subunits of the TAC that are integral mitochondrial OM proteins . Moreover we provide a detailed analysis of the biogenesis pathways of the two newly discovered TAC subunits as well as of the previously characterized TAC40 . We show that one of the new subunits contains separable targeting signals for import into the organelle and for sorting to the TAC , whereas the other one defines a novel class of mitochondrial β-barrel proteins .
Recently we have shown that the mitochondrial β-barrel protein TAC40 exclusively localizes to the TAC and is essential for its function [22] . In order to identify further TAC subunits we performed SILAC-based immunoprecipitation ( IP ) experiments using 29–13 T . brucei cells and a cell line expressing in situ HA-tagged TAC40 . The two cell lines were grown in the presence of isotopically-labeled heavy or light lysine and arginine . Subsequently , identical cell numbers from both populations were mixed and whole cell lysates were prepared which were subjected to IP using anti-HA antibodies . The resulting eluates were analyzed by quantitative MS and Fig 1A shows that they contain only two proteins that were five-fold or more enriched . One was TAC40 , whose tagged variant served as a bait , the other one a 60 kDa protein , termed TAC60 ( Tb927 . 7 . 1400 ) . Subsequently a cell line allowing tetracycline-inducible expression of a Myc-tagged version of TAC60 was used to do a second set of SILAC-IPs . The most highly enriched proteins recovered in these IPs were TAC60 , TAC40 and a protein of 42 kDa , termed TAC42 ( Fig 1B ) . Finally we produced a cell line expressing an HA-tagged version of TAC42 and performed a third set of SILAC-IPs which besides TAC42 itself recovered TAC40 and TAC60 as the most enriched proteins ( Fig 1C ) . In summary these reciprocal IPs show that TAC40 , TAC60 and TAC42 form a protein complex . To find out where TAC60 is localized the cell line expressing Myc-tagged TAC60 was analyzed by immunofluorescence ( IF ) . The cells were stained with an anti Myc-antibody , with the DNA-staining agent DAPI , which labels both the nuclei and the kDNA networks , and with the monoclonal antibody YL1/2 . The latter recognizes tyrosinated α-tubulin and in T . brucei stains the basal bodies as well as the distal part of the subpellicular array of microtubules [25] . An overlay of all three signals reveals that TAC60 localizes to a dot-like structure between the basal bodies and the kDNA networks ( Fig 2A , left panel ) , as would be expected for a TAC subunit . The TAC is to a large part detergent-resistant , which allows the isolation of a fraction containing flagella whose basal bodies are still connected to the kDNA [9] . IF analysis shows that in such fractions Myc-tagged TAC60 localizes between the kDNA and the flagellum ( Fig 2A , right panel ) indicating that the protein is a structural subunit of the TAC . In order to investigate the function of TAC60 a tetracycline-inducible RNAi cell line was produced . Analysis of DAPI-stained cells in the left panel of Fig 2B shows that ablation of TAC60 in insect stage T . brucei leads to a rapid loss of the kDNA networks reaching approximately 50% after 1 . 5 days of induction . Growth on the other hand is affected after 3 days only ( Fig 2B ) . The right panel in Fig 2B demonstrates that a large majority of the 30% of cells , that have retained their kDNA after two days of RNAi induction , have greatly enlarged kDNA networks . Moreover , in a minority of cells smaller kDNA networks are observed . The massive over-replication of kDNA networks is a hallmark that distinguishes cells with a deficient TAC from cells ablated in kDNA replication [17 , 18 , 20 , 22] . As expected a transmission electron microscopy analysis shows enlarged kDNA networks , which are in part stacked on top of each other but whose ultrastructure is undisturbed ( Fig 2D ) . Mitochondrial translation and thus the kDNA as well as the TAC are essential in both the insect and bloodstream form of T . brucei [26] . In line with this ablation of TAC60 in bloodstream form cells leads to a rapid loss of the kDNA with a subsequent growth arrest ( Fig 2C , left panel ) . However , if the same experiment is done in a bloodstream form cell line of T . brucei that , due to a single point mutation in the nuclear-encoded γ-subunit of the mitochondrial ATPase , can grow in the absence of the kDNA [27] , a different result is obtained: in such a cell line TAC function is dispensable and ablation of TAC60 , while causing the loss of the kDNA , does not slow down growth ( Fig 2C , right panel ) . The set of experiments done for TAC60 ( Fig 2 ) were also used to analyze TAC42 ( Fig 3 ) . The obtained results were essentially identical for both proteins . In summary , these experiments ( Fig 2 and Fig 3 ) establish that TAC60 and TAC42 are essential novel subunits of the TAC that are not involved in any other essential functions unrelated to the kDNA . As expected tagged TAC60 co-fractionates with the mitochondrial marker ATOM40 when cells are extracted with a low concentration of digitonin ( Fig 4A , upper panel ) . Moreover , TAC60 is exclusively recovered in the pellet when a crude mitochondrial fraction is subjected to carbonate extraction at high pH indicating that TAC60 is an integral membrane protein ( Fig 4A , middle panel ) . It has previously been shown that organellar proteins that accumulate in the cytosol upon inhibition of mitochondrial protein import are rapidly degraded by the cytosolic proteasome [28 , 29] . Thus to analyze whether TAC60 localizes to the outer or the inner mitochondrial membrane we followed the fate of a tagged version of the protein in inducible ATOM40- and TbTim17-RNAi cell lines . ATOM40 and TbTim17 are core subunits of the archaic protein translocase of the OM ( ATOM ) and the protein translocase of the IM ( TIM ) , respectively [23 , 30–34] . The top panel of Fig 4B shows that the steady state levels of tagged TAC60 in whole cells rapidly decrease during ATOM40 RNAi . The same is the case for a previously characterized β-barrel protein , the voltage dependent anion channel ( VDAC ) , which first needs to be translocated into the intermembrane space ( IMS ) before it is inserted into the OM . In the TbTim17 RNAi cell line in contrast the steady state levels of TAC60 remain essentially constant . The IM protein cytochrome oxidase subunit IV ( CoxIV ) , whose import requires TbTim17 and therefore serves as a positive control , however accumulates as unprocessed precursor protein ( Fig 4B , bottom panel ) . Thus mitochondrial import of tagged TAC60 depends on ATOM40 but not on TbTim17 indicating that it is an OM protein . This is further supported by a normalized abundance profile of TAC60 over six subcellular fractions , produced in a previous proteomic analysis ( Fig 4A , bottom panel ) ( However since TAC60 was detected in only one of two experiments , it was not included in the OM proteome defined in the study ) [35] . In silico analysis of TAC60 of T . brucei using various prediction programs detects two high confidence TMDs ( 121–141 and 238–258 ) that are found in essentially all TAC60 orthologues of trypanosomatids . Moreover , HHPred analysis [36] indicates that the C-terminal 150 aa of TAC60 and its orthologues has some similarity to bacterial tRNA/rRNA methyltransferases . In order to analyze the topology of TAC60 experimentally we used the split GFP approach [37] . To that end a cell line expressing N-terminally HA-tagged GFP lacking the C-terminal β-strand ( HA-GFP1-10 ) was produced . As expected the truncated GFP fractionates with the cytosol ( Fig 4C , top right panel ) . Subsequently , two variants of TAC60 that were N- or C-terminally fused to the last β-strand of GFP ( GFP11-TAC60 , TAC60-GFP11 ) were expressed in the same cell line . The IF analysis in Fig 4C ( bottom panel ) shows a GFP signal that to a large part is in close proximity of the kDNA and thus is consistent with a TAC localization . The same signal is not seen in the absence of tetracycline , which prevents the expression of the fusion proteins . A weak background signal close to but not overlapping with the TAC is visible for both TAC60-GFP11 and GFP11-TAC60 in the absence and in the presence of tetracycline . S1A Fig shows that the signal is due to autofluorescence . In summary , these results show that both the N- and the C-termini of TAC60 face the cytosol and thus are consistent with the notion that TAC60 has two TMDs . The integral membrane subunits of the TAC not only need to be imported into mitochondria but also require sorting to the single unit TAC . In order to test whether mitochondrial targeting and subsequent sorting to the TAC require distinct signals , we expressed C-terminal tagged versions of TAC60 that were truncated either at their N- or C-termini or on both ends ( Fig 5A ) . IF analysis showed four distinct localizations of the truncated TAC60 versions ( Fig 5B , 5C , 5D and 5E , S1B Fig . ) : - The variants lacking the C-terminal 153 and 283 aa ( ΔC153/ΔC283 ) were localized to the TAC . Although some dots that do not overlap with the kDNA are also seen ( Fig 5B ) . The same was the case if the C-terminal 283 amino acid deletion was combined with N-terminal truncations of 75 and 97 aa ( ΔN75_ΔC283/ΔN97_ΔC283 ) ( Fig 5B ) . Interestingly , however expression of these two variants causes some cells to show an enlarged kDNA or kDNA loss , respectively . - The variant lacking the N-terminal 114 aa ( ΔN114 ) was also localized to the TAC ( Fig 5C ) , although in contrast to the variants described above ( Fig 5B ) the rest of the mitochondrion was also stained . Thus , only a fraction of the tagged ΔN114 variant is localized at the TAC indicating that the efficiency of TAC sorting is reduced . - The variant lacking the N-terminal 140 aa ( ΔN140 ) was mitochondrially localized ( Fig 5D ) , but not sorted to the TAC demonstrating that mitochondrial targeting and TAC-sorting are distinct events . - The variants lacking either N-terminal 233 and 257 aa ( ΔN233/ΔN257 ) or the C-terminal 320 and 408 aa ( Δ320/ΔC408 ) finally showed a diffuse cytosolic localization ( Fig 5E ) . In summary , these results define a 140 aa long segment of TAC60—aa 140–283—encompassing the IMS-exposed loop of the protein and the more C-terminal TMD as essential for mitochondrial targeting . Sorting to the TAC however requires an additional N-terminal 26 aa segment that consists essentially of the first TMD of TAC60 . However , on its own the segment only confers an incomplete TAC localization which is illustrated by the fact that a fraction of the ΔN114 variant yields an overall mitochondrial staining . The immunoblot in S2A Fig confirms that all TAC60 variants are expressed in comparable amounts . However , in many cases the anti-tag antiserum detects additional signals below or above the predicted bands . The ΔN114 and ΔN140 variants show the highest heterogeneity and besides the correctly sized protein at least four major degradation products are detected as well . These degradation products are all mitochondrially localized ( S2B Fig ) suggesting that TAC subunits that are imported into mitochondria but not sorted to the TAC are degraded . For many other variants closely spaced double bands or additional signals above the correctly sized protein are observed . Preliminary experiments indicate that at least in the case of full length TAC60 and the ΔC153 variant protein phosphatase treatment results in a more simplified pattern shifted towards a lower molecular weight range ( S2C Fig ) . Thus , the observed heterogeneity within the TAC60 variants might be caused by phosphorylation and possibly other postranslational modifications . To investigate whether the correctly localized TAC60 truncations ΔC153 , ΔC283 , ΔN75_ΔC283 , and ΔN97_ΔC283 are functional we expressed them in a TAC60-RNAi cell line that targets part of the ORF that is absent in the truncations and therefore allows complementation experiments . The results in Fig 6 demonstrate that both C-terminally truncated variants complemented growth to wild-type level when expressed in the corresponding RNAi cell line . Thus the C-terminal domain of TAC60 that shows similarity to bacterial tRNA/rRNA methyltransferases is dispensable for TAC function . The same experiments were also done for the ΔN75_ΔC283 and ΔN97_ΔC283 TAC60 variants . While both of them localize to the TAC ( Fig 5B ) they were not able to complement the growth phenotype indicating that their function is impaired ( Fig 6 ) . Thus , while the N-terminal 97 aa are dispensable for TAC60 targeting they are required for the function of the protein . Bioinformatic analysis of TAC42 does not detect any significant sequence similarity to proteins outside the kinetoplastids . As TAC60 , tagged TAC42 co-fractionates with the mitochondrial marker ATOM40 when cells are extracted with low concentration of digitonin ( Fig 7A , upper panel ) . Moreover , the normalized abundance profile of TAC42 produced in a previous proteomic analysis suggest an OM localization [35] ( as in the case of TAC60 , TAC42 was only detected in one of two experiments in this study and was therefore not included in the OM proteome ) ( Fig 7A , bottom panel ) . TAC42 is exclusively recovered in the pellet fraction in a carbonate extraction suggesting it is an integral membrane protein ( Fig 7A , middle panel ) . This is surprising since TAC42 lacks predicted TMDs and in silico analyses do not predict it to be a β-barrel membrane protein . Thus , to test whether the biogenesis of TAC42 depends on the β-barrel insertion machinery we expressed a C-terminally tagged version of TAC42 in a cell line allowing inducible ablation of Sam50 , the core subunit of the sorting and assembly machinery ( SAM ) . Interestingly , in this cell line the level of a tagged version of TAC42 decreased during RNAi ( Fig 7B ) . As expected the same was the case for the well characterized β-barrel proteins VDAC [38] and ATOM40 [28] , which serve as positive controls . The C-terminally anchored OM protein ATOM69 [23] and cytosolic translation elongation factor 1a ( EF1a ) on the other hand were not affected . A pioneering study in yeast identified a moderately conserved sequence in the last β-strand of mitochondrial β-barrel proteins that serves as a sorting signal which directs the protein to the β-barrel protein insertion machinery [39] . Fig 8A shows that the three trypanosomal β-barrel proteins TAC40 , ATOM40 and VDAC as well as TAC42 have C-termini corresponding to the β-signal consensus sequence . Thus , in order to test whether these sequences function as sorting signals we expressed tagged TAC42 and TAC40 variants containing mutated variants of the putative β-signals . In one variant , termed 1mut , the invariant glycine was mutated to alanine . In the other variant , termed 4mut , all four conserved positions were changed , the first two to alanines and the last two to serines . The digitonin extraction in the top panel of Fig 8B shows that approximately 45% of the tagged wildtype version of TAC42 is recovered in the pellet corresponding to a crude mitochondrial fraction . The fact that 55% of the tagged proteins remains in the supernatant is likely due to overexpression when compared to the endogenous protein . Interestingly , however in the case of the 1mut and 4mut versions of TAC42 only 8–11% of the proteins are recovered in the pellet fractions . The same experiments were also done with the previously characterized β-barrel protein TAC40 [22] and , as in the case of TAC42 , mitochondrial targeting of the 1mut and 4mut versions of TAC40 was dramatically impaired ( Fig 8B ) . TAC40 was also ectopically tagged but likely less overexpressed compared to TAC42 which may explain why almost all of the tagged wildtype variant of the protein is mitochondrially localized . Thus , mutating the conserved glycine or all conserved amino acids in the β-signal consensus sequence of TAC42 and TAC40 progressively reduces mitochondrial targeting of the two proteins . The β-signal mediates the interaction with the β-barrel insertion machinery [39] . Its absence is therefore expected to interfere with membrane insertion after the proteins have been translocated into the IMS . In order to test this prediction we analyzed the mitochondria-associated fractions of the tagged TAC42 and TAC40 and its corresponding 1mut and 4mut versions by carbonate extraction at high pH to determine whether the proteins have been inserted into the OM . The lower panels in Fig 8B show that the tagged wildtype TAC42 and TAC40 are recovered in the pellet fractions together with ATOM40 , which serves as marker for a correctly inserted integral membrane protein . However , membrane insertion of the 1mut variant of TAC42 is reduced by 50% and in the case of 4mut variants by 70–80% for both proteins . Thus , TAC42 and TAC40 contain a C-terminal β-signal that is essential for correct targeting and membrane insertion of the proteins into the mitochondrial OM . The β-signal is expected to be conserved in all eukaryotes [39] . In order to show direct interaction of the putative β-barrel protein TAC42 with the SAM complex we therefore performed in vitro import experiments using trypanosomal substrate proteins and isolated yeast mitochondria . Radioactive substrates , produced by in vitro translation using rabbit reticulocyte lysate , were incubated with isolated mitochondria from either wildtype S . cerevisiae or from a yeast strain carrying a deletion of the SAM complex subunit Sam37 , whose function is to promote β-barrel protein insertion into the OM by linking the SAM and the TOM complex [40 , 41] . The in vitro import reactions were analyzed by BN-PAGE . The results in Fig 9A show that the wild-type protein but not the 4mut variant of TAC42 accumulate in a time dependent manner in a complex of approximately 200 kDa . Moreover , when the wild-type TAC42 is imported into mitochondria lacking Sam37 a much smaller amount of the complex is observed . Furthermore , its molecular weight is slightly lower due to the absence of Sam37 [42] . In summary these results show that trypanosomal TAC42 directly interacts with the yeast SAM complex in a β-signal and Sam37-dependent manner indicating that TAC42 is indeed a novel kinetoplastid-specific mitochondrial β-barrel protein . Thus at least two essential OM subunits of the TAC—TAC40 and TAC42—are β-barrel proteins . In line with this , ablation of Sam50 , the core component of the SAM complex , leads to a rapid increase of cells lacking kDNA networks , whereas in the cells that have retained the kDNA it is massively overreplicated ( Fig 9B ) . Ablation of Sam50 therefore essentially reproduces the phenotypes that are observed after ablation of individual TAC subunits . However , the same is not seen in cells ablated for ATOM40 the channel subunit of the main OM protein translocase [20] . Thus these results underscore the importance of the trypanosomal SAM complex for the assembly of this unique mitochondrial DNA inheritance system .
The TAC is a single unit structure that links the kDNA to the basal body . During each cell cycle a new TAC needs to be formed to guarantee that the replicated kDNA networks are correctly segregated during the binary fission of the single mitochondrion of T . brucei . Understanding TAC biogenesis is hampered by the fact that only few of its subunits are known and that their targeting pathways have not been studied . The exclusion zone filaments , that form the bridge from the basal body to the OM , consists of cytosolic proteins which may reach the TAC region by diffusion and therefore not require specific targeting . The subunits of the unilateral filaments , however , need to be imported into the mitochondrial matrix before they can assemble to link the kDNA network with the mitochondrial IM . A similar situation is found for the membrane-embedded TAC subunits in the differentiated membranes which connect the cytosolic with the matrix-localized TAC filaments . Its subunits potentially first need to be imported and inserted into the OM and IM membranes , respectively , before they are laterally sorted to the new TAC that is being assembled . In our studies we have focused on the OM region of the differentiated membrane domain of the TAC . We have discovered two novel integral OM TAC subunits—TAC60 and TAC42 , that are required for kDNA segregation—determined their topology and deciphered their biogenesis pathways . TAC60 is essential for TAC function , contains two TMDs and its N- and C-termini both face the cytosol . In a deletion analysis we have uncoupled mitochondrial targeting from TAC sorting and identified the first TAC sorting signal . Mitochondrial targeting of TAC60 requires a 143 long region that includes the IMS-exposed loop and its more C-terminal TMD . Insertion of TAC60 into the mitochondrial OM is mediated by ATOM40 , the pore-forming subunit of the master protein translocase in the OM . However , to reach its final destination , the TAC , TAC60 requires an additional 26 aa comprising the first TMD . Whether the TAC sorting signal depends on the TAC60 mitochondrial targeting signal or whether it could in principle work on its own , provided that the first TMD of TAC60 is inserted into the OM in the correct topology , remains unknown at the moment . Moreover , presently the TAC sorting signal appears to be specific for TAC60 indicating that other TAC subunits may have different sorting signals . The essential TAC subunit TAC42 lacks predictable TMDs . Its localization to the TAC depends on both Sam50 , the core component of the SAM complex , and on the presence of a β-signal consensus sequence at its C-terminus . It has been shown in yeast that this sequence mediates the interaction of β-barrel proteins with the SAM complex [39] . Moreover , TAC42 can be inserted into the OM of isolated yeast mitochondria provided that they have a functional SAM complex and that it carries a functional β-signal . In summary these result demonstrate that TAC42 is mitochondrial β-barrel protein even though in silico analysis fails to predict so . T . brucei has six known β-barrel membrane proteins: the metabolite transporter VDAC [38] , a second VDAC-like protein of unknown function [43] , ATOM40 [28 , 44] and Sam50 [45] , the core components of the ATOM and the SAM complex , and finally TAC40 [22] and TAC42 essential subunits of the TAC . Four of them ( VDAC , the VDAC-like protein , ATOM40 , TAC40 ) belong to the mitochondrial porin protein family of whereas Sam50 is an Omp85-like protein . TAC42 is unique it is neither a mitochondrial porin nor an Omp85-like protein but defines a novel class kinetoplastid-specific β-barrel proteins essential for mitochondrial DNA inheritance . The presence of two distinct β-barrel proteins , TAC40 and TAC42 , in the TAC is striking . β-barrel proteins are exclusively found in the OMs of bacteria , mitochondria and plastids [41 , 46] . It can be speculated that their prominent presence in the TAC indicates that the ancestor of this DNA inheritance system evolved very early , at a time when the integral membrane proteins present in the OM of the mitochondrial ancestor were restricted to β-barrel proteins . Moreover , the membrane domain of the TAC shows some architectural similarity to the double membrane spanning secretion systems of gram negative bacteria [47] . Both types of structures link OM and IM of bacterial evolutionary origin . Moreover , the OM is in both cases spanned by a β-barrel type structure ( generally multimeric in the case of bacteria ) and thus requires a Sam50/BamA-type insertion system . However , whereas the bacterial secretion systems serve to export bacterial effector proteins , there is no evidence that the TAC is involved in transport processes . While many subunits of the TAC are still unknown we get a progressively more detailed picture of its OM constituents . Up to now four essential , integral OM subunits of the TAC have been characterized: the β-barrel proteins TAC40 and TAC42 , which form a complex with TAC60 , as well as the dually localized pATOM36 . This complexity is surprising since the TAC is expected to have a structural function , linking the kDNA network to the basal body of the flagellum . A single OM protein that interacts on the cytosolic side with the exclusion zone filaments and on the IMS side with an IM protein should in principle be sufficient to do this job . We would like to propose two possible explanations for this unexpected complexity . It could be that the function of the TAC goes far beyond providing a structural linkage . Being localized between the kDNA and the flagellum , the TAC would be ideally suited to serve as a signaling platform that for example may regulate and integrate kDNA replication and segregation with flagellar growth and cytokinesis . Indeed pATOM36 has already been shown to mediate both mitochondrial protein import and mitochondrial DNA inheritance [20] . Alternatively , it might be that the TAC is the product of constructive neutral evolution . This ratched-like evolutionary process provides a non-adaptive explanation why macromolecular complexes can be comprised of more subunits than their function seem to demand [48 , 49] . In the case of the TAC , a possible scenario would be that an autonomously functioning ancestral TAC subunit would fortuitously bind to another protein . Binding to this protein would not affect the function of the TAC subunit , but it would have the potential to suppress mutations , which if present in the absence of the binding partner would inactivate the TAC subunit . Should such mutations occur the TAC subunit would lose its autonomy , as its function would now depend on the other protein . Thus , constructive neutral evolution may have led to four or more OM TAC subunits even though common sense suggests a single one should be enough . The two proposed explanations are not mutually exclusive , as both may have contributed to the complex TAC architecture . In order to disentangle the two we need a more complete picture of the TAC composition and architecture as well as a detailed functional analysis of its subunits .
Transgenic procyclic cell lines are based on T . brucei 29–13 [50] and were cultured at 27°C in SDM-79 containing 10% ( v/v ) fetal calf serum ( FCS ) . Transgenic bloodstream form trypanosomes are based on the New York single marker ( NYsm ) strain or on a derivative thereof termed F1γL262P [27] . All bloodstream from cells were grown at 37°C in HMI-9 supplemented with 10% FCS ( v/v ) . Full length TAC60 ( Tb927 . 7 . 1400 ) ( Fig 1B , Fig 2A ) or deletion variants of it ( Fig 5 , Fig 6 , S1B and S2 Figs ) , termed ΔN114 ( nt 343–1662 ) , ΔN140 ( nt 421–1662 ) , ΔN233 ( nt 697–1662 ) , ΔN257 ( nt 772–1662 ) , ΔC153 ( nt 1–1200 ) , ΔC283 ( nt 1–810 ) , ΔN75_ΔC283 ( nt 226–810 ) , ΔN97_ΔC283 ( nt 292–810 ) , ΔC320 ( nt 1–696 ) , ΔC408 ( nt 1–435 ) were cloned into a modified pLew100 expression vector containing a puromycin resistance gene in which a cassette had been inserted allowing C-terminal triple Myc-tagging [51] . For the experiment shown in Fig 4A , B one allele of TAC60 was tagged in situ at the C-terminus with a triple c-Myc-epitope [51] in the background of procyclic RNAi cell lines targeting ATOM40 and TbTim17 , both of which have been described before [28 , 52] . Full length TAC42 ( Tb927 . 7 . 3060 ) ( Fig 8 ) and TAC40 ( Fig 8 ) or mutated variants of them ( Fig 8 ) , termed 1mut ( TAC40: G351A; TAC42: G382A ) or 4mut ( TAC40: R349A , G351A , L354S and V356S; TAC42: R380A , G382A , A385S and L387S ) were cloned into a modified pLew100 expression vector containing a puromycin resistance gene in which a cassette had been inserted allowing C-terminal triple Myc-tagging [51] . For the experiment shown in Fig 3A and Fig 7 one allele of TAC42 was tagged in situ at the C-terminus with a triple HA-epitope and expressed in T . brucei 29–13 cells and in a previously described RNAi cell line targeting Sam50 [52] . The in situ HA-tagged TAC40 cell line ( Fig 1A ) has been described before [22] . The RNAi in the procyclic and bloodstream form cell lines was targeted against the ORF of TAC60 ( nt 256–772 ) or TAC42 ( nt 238–710 ) , respectively . For the complementation experiments of TAC60 ( Fig 6 ) , a different RNAi cell line targeting the 3'-part of the TAC60 ORF ( nt 1220–1629 ) was established . The Split-GFP approach was used as described [53] . For the results shown in Fig 4C , the GFP 1–10 OPT ( GFP1-10 ) and the M3 strand 11 ( GFP11 ) were amplified from a pET-15b-based vector ( generous gift from Prof . Steven Boxer , University of Stanford ) . GFP 1–10 OPT was cloned into a pLew100-based expression vector containing a blasticidin resistance cassette . The resulting construct allows inducible expression of N-terminal HA-tagged cytosolic GFP1-10 ( HA-GFP1-10 ) . GFP11 was cloned into a pLew100 expression vector containing the puromycin resistance gene . The resulting construct allows C-terminal tagging of proteins with GFP11 . Subsequently , the complete ORF of TAC60 was cloned into this modified pLew100 vector , yielding the construct TAC60-GFP11 . Another pLew100 expression vector was established , which allows N-terminal tagging of proteins with GFP11 . Subsequently , the complete ORF of TAC60 was cloned into this modified pLew100 vector , yielding the construct GFP11-TAC60 . A T . brucei 29–13 cell line expressing HA-GFP1-10 was established , which subsequently was transfected with TAC60-GFP11 or GFP11-TAC60 , respectively . For visualization of tagged proteins , the respective cell lines were induced for 1 day with 1 μg/ml of tetracycline . The following polyclonal rabbit antisera directed against the indicated antigens were produced in our lab and have been used before [23 , 35] . The working dilutions for immunoblots ( IB ) and IF are indicated: VDAC ( IB 1:1 , 000 ) , ATOM40 ( IB 1:10 , 000; IF 1:1 , 000 ) , cytochrome C ( IB 1:1 , 000 ) , ATOM69 ( IB 1:50 ) , LipDH ( IB 1:10 , 000 ) and CoxIV ( IB 1:1 , 000 ) . Commercially available monoclonal antibodies were used as follows: mouse c-Myc ( Invitrogen , 132500; IB 1:2 , 000; IF 1:50 ) , mouse HA ( Enzo Life Sciences AG , CO-MMS-101 R-1000; IB 1:5 , 000; IF 1:1 , 000 ) and mouse EF1a ( Merck Millipore , Product No . 05–235; IB 1:10 , 000 ) . Monoclonal anti-tyrosinated α-tubulin antibody YL1/2 [54] ( IFA 1:500 ) produced in rat was a generous gift from Prof . Keith Gull , University of Oxford . Secondary antibodies for IB analysis were IRDye 680LT goat anti-mouse , IRDye 800CW goat anti-rabbit ( LI-COR Biosciences , 1:20 , 000 ) and horse radish peroxidase-coupled goat anti-mouse and anti-rabbit ( Sigma-Aldrich , 1:5 , 000 ) . Secondary antibodies for IF were goat anti-mouse Alexa Fluor 633 , goat anti-mouse Alexa Fluor 596 , goat anti-rabbit Alexa Fluor 488 and goat anti-rat Alexa Fluor 488 ( all from ThermoFisher Scientific , 1:1000 ) Digitonin extraction was used to generate crude mitochondrial enriched fractions [55] to demonstrate mitochondrial localization of a protein of interest . For this , 5x107 or 1x108 cells were incubated for 10 min on ice in 20 mM Tris-HCl pH 7 . 5 , 0 . 6 M sorbitol , 2 mM EDTA containing 0 . 025% ( w/v ) digitonin . After centrifugation ( 6 , 800 g , 4°C ) , the resulting mitochondria-enriched pellet was separated from the supernatant and equal cell equivalents of each fraction were subjected to SDS-PAGE and immunoblotting . Alternatively , the mitochondria-enriched pellets were used for subsequent alkaline carbonate extractions ( see below ) . A mitochondria-enriched pellet fraction obtained by digitonin extraction was resuspended in 100 mM Na2CO3 pH 11 . 5 , incubated on ice for 10 min and centrifuged ( 100 , 000 g , 4°C , 10 min ) to separate the membrane fraction from soluble proteins . Equal cell equivalents of all samples were analyzed by SDS-PAGE und immunoblotting . SILAC-IP experiments were essentially done as described [20] . T . brucei 29–13 cells , their derivatives allowing expression of in situ HA-tagged TAC40 or TAC42 , and a transgenic cell line allowing inducible expression of Myc-tagged TAC60 were used as indicated ( Fig 1 ) . Cells expressing or not expressing the tagged bait protein were grown in either light ( unlabeled ) or heavy ( 13C615N4-L-arginine; 13C615N2-lysine ) arginine- and lysine-containing SDM-80 medium containing 10–15% ( v/v ) dialyzed FCS ( BioConcept , Switzerland ) for around 10 doubling times to establish complete labeling of proteins with light or heavy amino acids . Equal numbers of cells grown in the presence of heavy or light arginine and lysine were mixed and harvested . The resulting pellets were solubilized in 20 mM Tris-HCl , pH 7 . 4 , 0 . 1 mM EDTA , 100 mM NaCl , 10% glycerol , 1 . 5% ( w/v ) digitonin and 1X Protease Inhibitor mix ( EDTA-free , Roche ) for 15 min at 4°C . The extracts were centrifuged ( 20 , 000 g , 15 min , 4°C ) and the resulting supernatants were incubated with anti-HA affinity beads ( Roche ) or anti-Myc affinity beads ( EZview red , Sigma ) , equilibrated in the same buffer as above but containing only 0 . 2% ( w/v ) of digitonin . After 2 h of incubation at 4°C , the supernatant was discarded and the beads were washed 3 times with 0 . 5 ml of the same buffer . TAC40 and TAC60 protein complexes were eluted by boiling the resin for 5 min in 60 mM Tris-HCl , pH 6 . 8 containing 0 . 1% SDS whereas TAC42 protein complexes were eluted using SDS gel loading buffer without β-mercaptoethanol . SILAC-IP experiments were performed in three biological replicates including a label-switch each . Proteins purified in TAC42 SILAC-IPs were separated on a 4–20% Mini Protean TGX SDS-PAGE gel ( BioRad ) . Afterwards , the gel was fixed with acetic acid and methanol and stained for 3 hours with fresh ammonium sulfate-based colloidal coomassie ( 10% phosphoric acid , 10% ammonium sulfate , 0 . 12% coomassie brilliant blue G , 20% methanol ) . Subsequently , the gel lanes were cut into 10 pieces per replicate , which were then destained by repetitive rounds of incubation in 10 mM NH4HCO3 and 10 mM NH4HCO3 containing 50% ethanol . When completely destained , the gel pieces were dehydrated in 100% ethanol . Finally , cysteine residues of proteins were reduced with 5 mM bond breaker solution ( Thermo Fisher Scientific ) , alkylated with 100 mM iodoacetamide in 10 mM NH4HCO3 , and proteolytically digested with trypsin ( 37°C , incubation overnight ) . Proteins purified in TAC40 and TAC60 SILAC-IPs were were reduced , alkylated and tryptically digested in-solution as described previously [29] . LC-MS analyses of peptide mixtures were performed on an LTQ Oritrap XL ( TAC40 , TAC60 ) or an Orbitrap Elite ( TAC42 ) mass spectrometer ( Thermo Fisher Scientific , Bremen , Germany ) , each directly coupled to an UltiMate 3000 RSLCnano HPLC system ( Thermo Fisher Scientific , Dreieich , Germany ) , as described before [32] . For quantitative MS data analysis , MaxQuant/Andromda was used ( version 1 . 4 . 1 . 2 for TAC40 , 1 . 5 . 1 . 0 for TAC60 , and 1 . 5 . 3 . 30 for TAC42 data; [56 , 57] ) . MS/MS data were searched against all entries for T . brucei TREU927 retrieved from the TriTryp database ( version 8 . 1 ) applying MaxQuant default parameters with the exceptions that protein identification and quantification were based on one unique peptide and one ratio count ( i . e . SILAC peptide pair ) . Mean log10 protein abundance ratios ( TAC40/42/60 versus control ) and p-value ( one-sided Student's t-test ) across at least two 2 biological replicates were determined . Proteins identified and quantified in TAC40 , TAC42 , and TAC60 SILAC IPs are listed in S1–S3 Tables . 35S-Met-labelled proteins were synthesized using the TNT T7 Quick for PCR ( Promega ) in vitro translation kit according to the instruction manual . Radiolabeled precursors proteins were incubated with isolated yeast mitochondria in import buffer ( 3% ( w/v ) BSA , 250 mM sucrose , 80 mM KCl , 5 mM MgCl2 , 5 mM L-methionine , 2 mM KH2PO4 , 10 mM MOPS-KOH , pH 7 . 2 , 2 mM NADH , 5 mM ATP , 10 mM creatine phosphate , 0 . 1 mg/ml creatine kinase ) at 25°C . Mitochondria were washed with SEM ( 250 mM sucrose , 1mM EDTA , 10mM MOPS pH 7 . 2 ) and analyzed by blue native electrophoresis ( 4–16% gradient gels ) and autoradiography . IF and northern blots were done as described [23] . IF images were acquired with a DFC360 FX monochrome camera ( Leica Microsystrems ) and a DMI6000B microscope ( Leica Microsystems ) . Image analysis was done using LAS X software ( Leica Microsystems ) , ImageJ , and Adobe Photoshop CS5 . 1 ( Adobe ) . Relative quantification of the fluorescent intensity of kDNA networks is described in [22] . Isolation of flagella was done according to [58] . Transmission electron microscopy was exactly done as described [22] . | Trypanosoma brucei and its relatives are important human and animal pathogens . Unlike most other eukaryotes trypanosomes have a single mitochondrion with a single unit mitochondrial genome , termed the kinetoplast DNA ( kDNA ) . During each cell cycle the kDNA is replicated and subsequently segregated into the two organelles that are formed during binary fission of the mitochondrion . Segregation depends on the tripartite attachment complex ( TAC ) which physically links the kDNA to the basal body of the flagellum . Thus , the TAC couples the segregation of the replicated kDNA to the segregation of the old and new flagella . We have characterized the outer membrane section of the TAC and shown that it contains a complex of three integral membrane proteins , TAC60 , TAC42 and TAC40 , each of which is essential for TAC function . Furthermore , we have identified which protein import systems are required for their biogenesis . In the case of TAC60 we demonstrate that membrane insertion and sorting to the TAC are separate processes requiring distinct cis-elements . Finally , we show that TAC42 is a novel mitochondrial beta-barrel protein whose biogenesis depends on the beta-signal in its C-terminus . Thus , TAC60 , TAC42 and TAC40 are essential trypanosomatid-specific proteins that may be exploited as drug targets . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"parasitic",
"protozoans",
"membrane",
"proteins",
"trypanosoma",
"brucei",
"protozoans",
"mitochondria",
"energy-producing",
"organelles",
"bioenergetics",
"epigenetics",
"cellular",
"structures",
"and",
"organelles",
"kinetoplasts",
"genetic",
"interference",
"gene",
"expression",
"cell",
"membranes",
"pathogen",
"motility",
"biochemistry",
"rna",
"trypanosoma",
"eukaryota",
"cell",
"biology",
"nucleic",
"acids",
"virulence",
"factors",
"integral",
"membrane",
"proteins",
"genetics",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"trypanosoma",
"brucei",
"gambiense",
"organisms",
"flagella"
] | 2017 | Biogenesis of the mitochondrial DNA inheritance machinery in the mitochondrial outer membrane of Trypanosoma brucei |
Humans and animals are capable of evaluating actions by considering their long-run future rewards through a process described using model-based reinforcement learning ( RL ) algorithms . The mechanisms by which neural circuits perform the computations prescribed by model-based RL remain largely unknown; however , multiple lines of evidence suggest that neural circuits supporting model-based behavior are structurally homologous to and overlapping with those thought to carry out model-free temporal difference ( TD ) learning . Here , we lay out a family of approaches by which model-based computation may be built upon a core of TD learning . The foundation of this framework is the successor representation , a predictive state representation that , when combined with TD learning of value predictions , can produce a subset of the behaviors associated with model-based learning , while requiring less decision-time computation than dynamic programming . Using simulations , we delineate the precise behavioral capabilities enabled by evaluating actions using this approach , and compare them to those demonstrated by biological organisms . We then introduce two new algorithms that build upon the successor representation while progressively mitigating its limitations . Because this framework can account for the full range of observed putatively model-based behaviors while still utilizing a core TD framework , we suggest that it represents a neurally plausible family of mechanisms for model-based evaluation .
Here , we briefly review the formalism of reinforcement learning in a Markov decision process ( MDP ) , which provides the foundation for our simulations ( see [25] or [26] for a fuller presentation ) . An MDP is defined by a set of states , a set of actions , a reward function R ( s , a ) over state/action pairs , and a state transition distribution , P ( s′|s , a ) , where a denotes the chosen action . States and rewards occur sequentially according to these one-step functions , driven by a series of actions; the goal is to learn to choose a probabilistic policy over actions , denoted by π , that maximizes the value function , Vπ ( s ) , defined as the expected cumulative discounted reward: Vπ ( s ) =Eπ[∑k=0∞γkRt+k|St=s] . Here , γ is a parameter controlling temporal discounting . The value function can also be defined recursively as the sum of the immediate reward of the action chosen in that state , R ( s , a ) , and the value of its successor state s’ , averaged over possible actions , a , and transitions that would occur if the agent chose according to π: Vπ ( s ) =∑aπ ( a|s ) [R ( s , a ) +∑s′P ( s′|s , a ) γVπ ( s′ ) ] ( 1 ) The value function under the optimal policy is given by: V* ( s ) =maxaE[∑k=0∞γkRt+k|St=s , At=a]=maxa[R ( s , a ) +∑s′P ( s′|s , a ) γV* ( s′ ) ] ( 2 ) Knowledge of the value function can help to guide choices . For instance , we can define the state-action value function as the value of choosing action a and following π thereafter: Qπ ( s , a ) =Eπ[∑k=0∞γkRt+k|St=s , At=a]=R ( s , a ) +∑s′P ( s′|s , a ) γVπ ( s′ ) ( 3 ) Then at any state one could choose the action that maximizes Qπ ( s , a ) . ( Formally this defines a new policy , which is as good or better than the baseline policy π; analogously , Eq 2 can be used to define the optimal state-action value function , the maximization of which selects optimal actions . ) Note that it is possible to write a recursive definition for Q in the same manner as Eq 1 , and work directly with the state-action values , rather than deriving them indirectly from V . For expositional simplicity , in this article , we work instead with V wherever possible ( mainly because this is easier to depict visually , and simplifies the notation ) , and accordingly we assume in our simulations that the agent derives Q using Eq 3 for guiding choices . To be concrete , in a spatial “gridworld” task of the sort we simulate , this amounts to computing a value function V over locations s , and using it to derive Q ( the value of actions a heading in each of the four cardinal directions ) by examining V for each adjacent state . Although this simplifies bookkeeping for this class of tasks , this is not intended as a substantive claim . Indeed , the last algorithm we propose will work directly with Q values , and the others can easily be re-expressed in this form . The problem of reinforcement learning is then reduced to learning to predict the value function Vπ ( s ) or V* ( s ) . There are two main families of approaches . Model-based algorithms learn to estimate the one-step transition and reward functions , P ( s′|s , a ) and R ( s , a ) , from which it is possible to compute V* ( or Q* ) using Eq 2 . This typically involves unrolling the recursion in Eq 2 into a series of nested sums , an algorithm known as value iteration . The alternative , model-free , approach exemplified by TD learning bypasses estimating the one-step model . Instead , it directly updates a cached estimate of the value function itself . In particular , following a transition s → s′ initiated by action a , a reward prediction error , δ , is calculated and used to update V ( s ) : δ=R ( s , a ) +γV ( s′ ) −V ( s ) ( 4 ) V ( s ) ←V ( s ) +αTDδ where αTD is a learning rate parameter . The TD update rule is derived from the recursion in Eq 1: each step iteratively pushes the left hand side of the equation , V ( s ) , closer to R ( s , a ) + γV ( s′ ) , which is a one-sample estimate of the right hand side . Finally , analogous sample-based updates may also be conducted offline ( e . g . , between steps of actual , “online” experience ) . This is a key insight of Sutton’s Dyna architecture [27] ( see also [28] ) . The approach , like TD , caches estimates of V ( s ) . Here TD learning is supplemented by additional offline updates . Specifically , samples consisting of a transition and reward triggered by a state-action ( s , a , r , s’ ) are generated either from a learned one-step model’s probability distributions P ( s′|s , a ) and R ( s , a ) , or instead simply replayed , model-free , from stored episodes of previously experienced transitions . For each sample , V ( s ) is then updated according to Eq ( 4 ) . Given sufficient iterations of sampling between steps of real experience , this approach can substitute for explicit value iteration and produce estimates at each step comparable to model-based approaches that more directly solve Eqs 1 or 2 . A further distinction , which will become important later , is that between on-policy methods , based on Eq 1 , and off-policy methods , based on Eq 2 . On-policy methods estimate a policy-dependent value function Vπ ( s ) , whereas off-policy methods directly estimate the optimal value function V* ( s ) . Typically , model-based methods are off-policy ( since having learned a one-step model it is possible to use Eq 2 to directly compute the optimal policy ) ; whereas different TD learning variants can be either on- or off-policy . Due to the similarity between the phasic responses of midbrain dopamine neurons , and the TD prediction error δ ( Eq 4 ) , it has long been suggested that this system implements TD learning [2 , 3] . More specifically ( e . g . [4]; Fig 1A ) it has been suggested that values V or Q are associated with the firing of medium spiny neurons in striatum [29 , 30] , as a function of an input state ( or state-action ) representation carried by their afferent neurons in frontal cortex , and that learning of the value function is driven by dopamine-mediated adjustment of the cortico-striatal synapses connecting these neurons . Selection among these striatal value representations would then drive action choice . Although not entirely uncontroversial , a great deal of evidence about dopamine and its targets supports this hypothesis ( see [26 , 31] for fuller reviews ) . Such theories provide a neural implementation of Thorndike’s early law of effect , the reinforcement principle according to which rewarded actions ( here , those paired with positive prediction error ) tend to be repeated [34] . However , the hypothesis that animals or humans rely exclusively on this principle to make decisions has long been known to be false , as demonstrated by a line of learning experiments whose basic logic traces to rodent spatial navigation experiments by Tolman [23] ( for modern variants , see [6 , 28 , 35–37] ) . To facilitate simulation and analysis , we here frame the logic of these experiments in terms of “grid-world” spatial MDPs . When viewed as MDPs , Tolman’s experiments can be divided into two categories , which require subjects to adjust to either of two different sorts of local changes in the underlying MDP . Experience with these changes is staged so as to reveal whether they are relying on cached values or recomputing them from a representation of the full MDP . Accordingly , revaluation tasks , such as latent learning , reward devaluation , and sensory preconditioning , examine whether animals appropriately adjust behavior following changes in R ( s , a ) , such as a newly introduced reward ( Fig 2A ) . Analogously , contingency change ( e . g . , detour or contingency degradation ) tasks examine whether animals appropriately adjust behavior following changes in P ( s′|s , a ) , such as a blocked passageway ( Fig 2B ) . Model-free RL is insensitive to these manipulations because it caches cumulative expected rewards and requires additional learning to update the stored V . Conversely , model-based RL , which uses the one-step model directly to compute V at decision time , reacts immediately and flexibly to any experience that affects it . Note that the difference in behavior on these types of tasks predicted by the algorithms is categorical , and not a question of degree or learning speed . In particular , because of the representations they learn and update , model-based algorithms can make the correct choice following these manipulations without any further retraining ( i . e . so long as they learn locally about the new contingency or value , they can immediately make appropriate choices in distal parts of the state space ) , whereas model-free algorithms cannot ( in general , they must first experience trajectories starting from the test state and leading to the state with the changed value or transition contingency ) . Animals sometimes fail to correctly update behavior following revaluations , consistent with inflexible , model-free caching schemes [38] . However , findings that in other circumstances animals can indeed flexibly adapt their behavior following such manipulations ( without any further retraining–e . g . tested on the very first trial , or without feedback ) has long been interpreted as evidence for their use of internal models , as in model-based RL or similar methods [1 , 23 , 39] . A key goal of this article is to interrogate this assumption , and to consider neural mechanisms that , despite falling short of full model-based RL , might support such behavioral flexibility . A further set of rodent lesion studies have used reward devaluation tasks to suggest that apparently model-based and model-free behaviors ( i . e . , behavior that is either flexible or insensitive following reward devaluation ) depend on dissociable sets of brain areas ( e . g . [5 , 40]; Fig 1B ) . This led to the hypothesis ( e . g . , [1 , 32 , 41] ) that these two forms of reinforcement learning depend on competing systems in the brain—the dopaminergic TD system previously described , plus a second–less clearly understood–circuit supporting model-based behavior . But how is this latter computation carried out in the brain ? A number of fairly abstract theories have been based around explicit computation of the state-action value based on some form of Eq 3 , e . g . by learning an estimate of the one-step transition function , P ( s’|s , a ) and the immediate reward function R ( s , a ) and using them iteratively to compute the future value by tree search , value iteration , or Bayesian inference [1 , 41–43] . These theories have not spoken in detail about the neural implementation of these computations , but an accompanying presumption has been that the model-based system does not rely on a dopaminergic prediction error signal . This is because the TD prediction error of Eq 4 ( for γ > 0 , which is the parameter regime needed to explain phasic dopamine’s signature responses to the anticipation as well as receipt of reward [44] ) is specifically useful for directly learning long-run cumulative values V . In contrast , the idea of model-based learning is to derive these values iteratively by stringing together short-term predictions from a learned one-step model [12 , 45] . Note that the prediction error normally thought to be reported by dopamine neurons is not appropriate here: the prediction error signal for updating the immediate reward model R ( s , a ) is like Eq 4 but with γ = 0 , which is not consistent with anticipatory phasic dopamine responses . ( However , correlates of prediction errors for γ = 0 have been observed using human fMRI [46] ) . Furthermore , the hypothesized process of adding these rewards up over anticipated trajectories at choice time , such as by value iteration or tree search , has no counterpart in model-free choice . Instead , learning from anticipatory TD errors stores complete long-run values ( e . g . , in corticostriatal synapses ) , requiring no further computation at choice time . However , neither the rodent lesion data nor another body of work studying the neural correlates of model-based learning in humans suggests such a clean differentiation between the dopaminergic-striatal circuit ( supposed to support TD ) and some other presumably non-dopaminergic substrate for model-based learning . Instead , lesions suggest each type of learning is supported by a different subregion of striatum , together with connected regions of cortex ( Fig 1B ) and basal ganglia . This suggests that putatively model-based and model-free systems may map onto adjacent but structurally parallel cortico-basal ganglionic loops [33] , thus perhaps involving analogous ( striatal ) computations operating over distinct ( cortical ) input representations [47] . Also contrary to a strict division between systems , both dorsomedial and dorsolateral striatal territories have similar interrelationships with dopamine [48] , though the involvement of their differential dopaminergic input in devaluation sensitivity has not been completely assessed [49] . Research on humans’ learning in a two-step MDP ( which has similar logic to devaluation studies ) supports the causal involvement of dopamine in model-based learning [7–10] . Furthermore , dopaminergic recordings in rodents [11] ( though see [37] ) , and neuroimaging of prediction error signals in human striatum [6] suggest that these signals integrate model-based evaluations . Altogether , the research reviewed here supports the idea that model-based evaluations are at least partly supported by the same sort of dopaminergic-striatal circuit thought to support TD learning , though perhaps operating in separate cortico-striatal loops . This suggestion , if true , provides strong hints about the neural basis of model-based behavior . However , for the reasons discussed above , this also seems puzzlingly inconsistent with the abstract , textbook [50] picture of model-based learning by Eq 3 . Several more neurally explicit theories of some aspects of model-based computation have been advanced , which go some way toward resolving this tension by incorporating a dopaminergic component . Doya [51] introduced a circuit by which projections via the cerebellum perform one step of forward state prediction , which activates a dopaminergic prediction error for the anticipated state . The candidate action can then be accepted or rejected by thresholding this anticipated prediction error against some aspiration level . It is unclear , however , how this one-step , serial approach can be generalized to tasks involving stochastic state transitions , direct comparison between multiple competing actions , or rewards accumulated over multiple steps ( as in tasks like [52] ) . A similar idea has arisen from recordings in spatial tasks , where the firing of place cells along trajectories ahead of the animal suggests a hippocampal basis for a similar ( though multi-step ) state anticipation process , potentially driving evaluation of these candidate states using learned reward values in ventral striatum [53] . It is , however , unclear how this activity fits into a larger circuit for accumulation of these evaluations and comparison between options . Finally , another candidate approach is based on the Dyna framework discussed above . In this case , model-generated experience can be played back “off-line , ” e . g . between trials or during rest . These ersatz experiences can , in turn , drive dopaminergic prediction errors and updating of striatal Q values using the same mechanisms as real experience . As noted above , given sufficient off-line replay , this can achieve the same effect as model-based planning; in particular , it can update Q values following revaluation and other manipulations [28 , 54] . However , without a more traditional “on-line” planning component , this approach degrades ( to that of basic , model-free Q learning ) when there is insufficient time or resources for off-line replay , and when truly novel situations are encountered [28] . Here we propose and analyze a different family of approaches to these problems , which relate to the above proposals in that they incorporate elements of upstream predictive input to ventral striatum , and also of a different and more-flexible approach to offline updates . The proposed approach , based on the SR , builds even more directly on the standard TD learning model of dopaminergic-striatal circuitry . The research reviewed above suggests that flexible , seemingly model-based choices may be accomplished using computations that are homologous to those used in model-free RL . How can this be ? In fact , it is known that evaluations with some features of model-based learning can result from TD learning over a different input representation . As shown by Dayan [15] , Eq 1 can reformulated as: Vπ ( s ) =∑s′Mπ ( s , s′ ) ∑aπ ( a|s′ ) R ( s′ , a ) ( 5 ) Here , Mπ is a matrix of expected cumulative discounted future state occupancies , measuring the cumulative time expected to be spent in each future state s′ , if one were to start in some state s and follow policy π ( Fig 3 ) : Mπ ( s , s′ ) =E[∑t=0∞γtI ( st=s′ ) |S0=s] , ( 6 ) where I ( ⋅ ) =1 if its argument is true and 0 otherwise . Thus , this form rearranges the expectation over future trajectories in Eq 1 by first computing expected occupancies for each state , then summing rewards obtained , via actions , in each state over these . Mπ can also be used as a set of basis functions for TD learning of values . Specifically , we represent each state using a vector of features given by the corresponding row of M ( Fig 3 ) , i . e . by the future occupancies expected for each state s′ . Then we approximate Vπ ( s ) by some weighted combination of these features: Vπ ( s ) =∑s′Mπ ( s , s′ ) w ( s′ ) ( 7 ) Comparing Eqs 5 and 7 demonstrates this approximation will be correct when the weight w ( s′ ) for each successor state corresponds to its one-step reward , averaged over actions in s′ , ∑aπ ( a|s′ ) R ( s′ , a ) . One way to learn these weights is using standard TD learning ( adapted for linear function approximation rather than the special case of a punctate state representation ) . In particular , following a transition s → s′ , each index i of w is updated: w ( i ) ←w ( i ) +αTDδMπ ( s , i ) ( 8 ) Here , δ is defined as in Eq 4 . Note that in the algorithms discussed below , the agent must estimate the successor matrix Mπ from experience . If the feature matrix Mπ were known and static , a simpler alternative to Eq ( 8 ) for w would be to learn the one-step rewards by a delta rule on the immediate reward R . Since the successor representation is just a particular case of a linear feature vector for TD learning , the advantage of learning weights by the TD rule of Eq 8 is that weights learned this way will estimate value Vπ for any feature matrix M , such as estimates of the successor matrix Mπ prior to convergence ( S1 Fig ) . Altogether , this algorithm suggests a strategy for providing different inputs into a common dopaminergic/TD learning stage to produce different sorts of value predictions ( see also [16] ) . In particular , whereas model-free valuation may arise from TD mapping of a punctate representation of the current state ( Fig 3B ) in sensory and motor cortex to values in dorsolateral striatum ( Fig 1B ) , at least some aspects of model-based valuation may arise by analogous TD mapping of the successor representation ( Fig 3C and 3D ) in prefrontal cortex or hippocampus to values in dorsomedial striatum ( Fig 1B ) . This is possible because the successor matrix M has a predictive aspect reflecting knowledge of the state transitions P ( s′|s , a ) , at least in terms of aggregate occupancy , separate from the state/action rewards R ( s , a ) . This approach may thus offer a solution to how flexible , seemingly model-based choices can be implemented , and indeed can arise from the same dopaminergic-striatal circuitry that carries out model-free TD learning . What remains to be shown is whether algorithms based on this strategy–applying the SR as input to TD learning–can produce the full range of model-based behaviors . In the remainder of this paper , we simulate the behavior of such algorithms to explore this question . To simulate learning using the SR , we need to also simulate how the successor matrix Mπ is itself produced from experience . Mπ can be defined through a recursive equation that is directly analogous to Eqs 1 and 2: Mπ ( s , : ) =1s+γ∑s′Tπ ( s , s′ ) Mπ ( s′ , : ) , ( 9 ) where 1s is the vector of all zeros except for a 1 in the sth position and Tπ is the one-step state transition matrix that is dependent on π , Tπ ( s , s′ ) = ∑aπ ( a|s ) P ( s′|s , a ) . Similar to how approaches to estimating V are derived from Eqs 1 and 2 , one could derive analogous approaches to estimating Mπ from Eq 9 . Specifically , one could utilize a “model-based” approach that would learn Tπ and use it iteratively to derive a solution for Mπ . Alternatively , a TD learning approach could be taken to learn Mπ directly , without use of a one-step model Tπ . ( This approach is analogous to model-free TD methods for learning V , though it is arguably not really model-free since Mπ is itself a sort of long-run transition model . ) This TD learning approach would cache rows of M and update them after transitioning from their corresponding states , by moving the cached row closer to a one-sample estimate of the right hand side of Eq 9 . Lastly , such TD updates could also occur offline , using simulated or previously experienced samples . This approach for learning Mπ would be comparable to the Dyna approach for learning V . The three models we consider below correspond to these three different possibilities . Finally , note that SR-based algorithms have favorable computational properties; in particular , at choice time , given Mπ ( e . g . if it is learned and cached rather than computed from a one-step model ) , SR can compute values Vπ with a single dot product ( e . g . , a single layer of a linear neural network , Eq 7 ) , analogous to model-free TD algorithms . This is in contrast to the multiple steps of iterative computation required at choice time for computing value via Eq 1 in standard model-based approaches . This comes at the cost of storing the successor matrix Mπ: if S is the number of states in the task , the SR matrix has a number of entries equal to S2 . Such entries of Mπ can be stored as the ( all-to-all ) set of weights from a single layer of a neural network mapping input states to their successor representation .
The original SR [15] ( which we call SR-TD ) constructs the future state occupancy predictions Mπ using a TD learning approach . This approach caches rows of Mπ and incrementally updates them after transitioning from their corresponding states . Specifically , following each state transition s → s′ each element of row s is updated as follows: Mπ ( s , : ) ←Mπ ( s , : ) +αSR[1s+γMπ ( s′ , : ) −Mπ ( s , : ) ] , ( 10 ) where 1s is the vector of all zeros except for a 1 in the sth position . Mπ ( s , : ) is used as input to another TD learning stage , this time to learn the weights w for predicting expected future value from the state occupancy vector . To simulate SR-TD , we have the agent learn Mπ and w in parallel , updating each ( according to Eqs 10 and 8 , respectively ) at each transition; and sample actions according to an ϵ-greedy policy ( see Methods ) . Here , we explore a novel “model-based” approach , SR-MB , for constructing the expected state occupancy vector Mπ ( s , : ) . SR-MB learns a one-step transition model , Tπ and uses it , at decision time , to derive a solution to Eq 9 . One key constraint on a model-based implementation suggested by the data is that the computation should be staged in a way consistent with the architecture suggested by Fig 1A . Specifically , the TD architecture in Fig 1A suggests that , because the states are represented in cortex ( or hippocampus ) and weights ( which capture information about rewards ) and value are represented in downstream cortico-striatal synapses and medium spiny striatal neurons , information about R ( s , a ) and V ( s ) should not be used in the online construction of states . For the SR approach , this implies that M be constructed without using direct knowledge of R ( s , a ) or V ( s ) . As we see below , this serial architecture–a cortical state-prediction stage providing input for a subcortical reward-prediction stage–if true , would impose interesting limitations on the resulting behavior . To construct Mπ ( s , : ) , SR-MB first learns the one-step state transition matrix Tπ , implemented in our simulations through separate learning of P ( s′|s , a ) as well as π ( a|s ) , the agent’s previously expressed decision policy ( see Methods ) . Prior to each decision , Tπ is used to compute a solution to Eq 9 . This solution can be expressed in either of two forms . A given row , s , of M can be computed individually as the sum of n-step transition probabilities starting from state s: Mπ ( s , : ) =1sT+γTπ ( s , : ) +γ2Tπ2 ( s , : ) +γ3Tπ3 ( s , : ) …=∑n=0∞γnTπn ( s , : ) ( 11 ) Alternatively , matrix inversion can be used to solve for the entire successor matrix at once: Mπ= ( I−γTπ ) −1 ( 12 ) To implement SR-MB , we use Eq 12 . However , this is not a mechanistic commitment of the model , since Eq 11 is equivalent . Given Mπ , SR-MB learns the reward prediction weights w and forms V and Q values in the same way as SR-TD . Note finally that this scheme is similar to solving Eq 1 for on-policy values Vπ by value iteration , except that the sums are rearranged to put state prediction upstream of reward prediction , as per Eq 7 and in line with the neural architecture of Fig 1A . The max operator in Eq 2 prevents a similar rearrangement that would allow this scheme to be used for off-policy optimal values V* ( Eq 2 ) , as discussed below . The restriction to on-policy values Vπ is the major empirical signature of this version of the model . Here we introduce a third approach towards solving Eq 9 , SR-Dyna , which can be compared to Sutton’s Dyna approach [27] for solving Eqs 1 and 2 . Akin to how Dyna replays experienced transitions offline to update estimates of V ( s ) , SR-Dyna replays experienced transitions to update the successor matrix . When this approach is combined with an ‘off-policy’ update rule , similar to Q learning , to update the successor matrix offline , it is capable of solving the off-policy planning problem . Utilizing this type of update , however , requires us to work with a state-action version of the successor representation , H , which can be used directly to form Q values [60 , 61] . The key idea here is to define future occupancy not over states but over state/action pairs , sa . Analogous to Eq 5 , Qπ can then be expressed: Qπ ( sa ) =∑s′a′H ( sa , s′a′ ) R ( s′a′ ) , ( 13 ) H is a matrix of expected cumulative discounted future state-action visitations , i . e . given that you are starting with state s and action a , the cumulative ( discounted ) expected number of times you will encounter each other state/action pair: H ( sa , s′a′ ) =E[∑t=0∞γtI ( sat=s′a′ ) |sa0=sa] . ( 14 ) H can then be used as a linear basis for learning Q ( s , a ) , using the SARSA TD algorithm to learn a weight for each column of H . In particular , when state-action s’a’ is performed after state action sa , a prediction error is calculated and used to update w: δ=R ( sa ) +γQ ( s′a′ ) −Q ( sa ) ( 15 ) w ( i ) ←w ( i ) +αTDδH ( sa , i ) , foralli Like M , H can be defined recursively: H ( s , : ) =1sa+γ∑s′Tπ ( sa , s′a′ ) H ( s′a′ , : ) ( 16 ) where Tπ is the one-step state-action transition matrix , Tπ ( sa , s′a′ ) =∑s′∑a′P ( s′|s , a ) π ( a′|s′ ) . As with SR-TD , this recursion can be used to derive a TD-like update rule by which an estimate of H can be iteratively updated: H ( sa , ⋅ ) ←H ( sa , ⋅ ) +αSR[1sa+γH ( s′a′ , ⋅ ) −H ( sa , ⋅ ) ] ( 17 ) As with SR-MB , it is also possible to derive H from Tπ ( sa , s′a′ ) using an explicit “model-based” solution analogous to Eq 9 . However , here , we investigate the approach of updating H off-line ( e . g . , between trials or during rest periods ) using replay of experienced trajectories ( e . g . [62] ) . The key assumption we make is that this off-line replay can sequentially activate both the state and reward ( cortical and basal ganglia ) stages of Fig 1A , giving rise to an off-policy update of H with respect to the policy π* that is optimal given the current rewards . By comparison , as articulated above , we assumed such policy maximization was not possible when computing the successor representation M on-line for SR-MB , since this entire computation was supposed to happen in cortex at decision time , upstream of the striatal reward learning stage . Following each transition , SR-Dyna stores the sample ( s , a , s’ ) . Then in between decisions , SR-Dyna randomly selects ( with a recency weighted bias ) k samples ( with replacement ) . For each sample , it updates H as follows: H ( sa , ⋅ ) ←H ( sa , ⋅ ) +αSR[1sa+γH ( s′a′* , ⋅ ) −H ( sa , ⋅ ) ] ( 18 ) where a′*=argmaxa′∑s′′a′′H ( s′a′ , s′′a′′ ) w ( s′′a′′ ) That is , when H updates from previously experienced samples , it performs an off-policy update using the best action it could have chosen , rather than the one it actually chose .
More specifically , our motivation to develop this approach was based on three related sets of findings in the empirical literature . The first are that lesions to dorsomedial striatum prevent animals from adjusting preferences following reward revaluation [5] . In contrast , lesions to neighboring dorsolateral striatum cause rats to maintain devaluation sensitivity , even following overtraining [40] . In the framework presented here , neurons in dorsomedial striatum could represent values derived by applying TD learning to the successor representation and neurons in dorsolateral striatum could represent values derived by applying TD to tabular representations . Lesions to dorsomedial striatum would thus force the animal to work with values in dorsolateral striatum , derived from tabular representations and thus not sensitive to devaluation . In contrast , lesions to dorsolateral striatum would cause the brain to work with values derived from the SR , which are devaluation-sensitive . The second set of findings include several reports that the phasic DA response ( or analogous prediction error related BOLD signals in humans ) tracks apparently model-based information [6 , 11] . We have focused our simulations on choice behavior , and have not presented our theories' analogous predictions about the responses of neurons , such as DA cells , thought to signal decision variables . However , whenever the SR algorithms' expectations about action values incorporate "model-based" information ( such as latent learning , Fig 4A ) neural signals related to those predictions and to prediction errors would be similarly informed . Thus the theories predict systematic expectancy-related effects in the modeled dopamine response , tracking the differences in choice preference relative to the standard “model-free” accounts , which are blind to reward contingencies in these tasks . A third distinct set of findings also speaks to a relationship between dopamine and model-based learning . These are reports that several measures of dopaminergic efficiency ( both causal and correlational ) track the degree to which human subjects engage in model-based decision strategies in both multistep reward revaluation tasks and multiplayer games [7–10 , 68] . One possibility is that these effects reflect strengthened vs . weakened phasic dopaminergic signaling , which in our model controls reward learning for SR-based “model-based” estimates in dorsomedial striatum . However , this account does not explain the specificity of these effects to measures of putative model-based ( vs . model-free ) learning . These effects may instead be related to functions of dopamine other than prediction error signaling , such as tonic dopamine’s involvement supporting working memory [69] or its hypothesized role controlling the allocation of cognitive effort [41 , 70 , 71] . The framework outlined in this paper is not the only direction toward a neurobiologically explicit theory of putatively model-based behavior , nor even the only suggestion explaining the involvement of dopamine . As discussed above and pointed out in [28] , Sutton’s original Dyna algorithm–in which experience replayed offline is used to update action values V or Q directly–offers another avenue by which seemingly model-based flexibility can be built on the foundation of the standard prediction error model of dopamine . This is a promising piece of the puzzle , but exclusive reliance on replay to underpin all behavioral flexibility seems unrealistic . Among our innovations here is to suggest that replay can also be used to learn and update a successor representation , which then confers many of the other advantages of model-based learning ( such as flexibility in the face of reward devaluation ) without the dependence on further replay to replan . Furthermore , the addition of SR to the Dyna framework explains a number of phenomena that replay , on its own , does not . For instance , given that Dyna-Q works with a single set of cached Q values , updated through both experience and replay , it is not clear how it could , on its own , explain the apparent segregation of revaluation sensitive and insensitive value estimates in dorsomedial and dorsolateral striatum [5 , 40] . Another potential solution to some of the puzzles motivating this work is that dopamine could have a role in action selection , as part of a circuit for partial model-based action evaluation [47] . According to this idea , dopamine neurons could compute a prediction error measuring the difference between the value of the current state and the future value of a predicted successor state , caused by a given candidate action . The size of this prediction error could then determine whether the action is performed . This mechanism would endow the brain with a single step of model-based prediction . However , it is not straightforward how this sort of approach could underlie model-based learning in tasks requiring more than a single step of prediction , and accordingly our simulations ( see Fig 4A and supplemental materials ) show that it cannot solve any of the revaluation tasks considered here , which all probe for deeper search through the state space . A recent study provided convincing behavioral evidence that humans sometimes simplify model-based action selection by combining just one step of state prediction with cached successor state values [72] . Yet this same study along with others [52] also provided evidence that humans can plan through more than one step and thus are not confined to this approximation . It is also not straightforward how this sort of mechanism could endow model-based predictions in cases where stochasticity requires consideration of “trees” of possible future states . Nevertheless , by elucidating a more general framework in which a predictive state representation may feed into downstream dopaminergic reward learning , we view our framework as fleshing out the spirit of this suggestion while also addressing these issues . We similarly realize other conceptual suggestions in the literature suggesting that more flexible model-based like behavior may arise not through tree-search like planning , but rather by applying model-free RL to more sophisticated state representations [73] . A more specific application of this idea , [74] demonstrated that a sophisticated representation that includes reward history can produce model-based like behavior in the two-step reward revaluation task . The successor representation adds to this work by clarifying for any task’s transition structure , the precise representation that can be used to generate model-based behavior . Relatedly , because it places model-based state prediction in the input to a standard TD learning circuit , our framework could easily be extended to include several modules with inputs corresponding to several different types or granularities of models: for instance , varying degrees of temporal abstraction corresponding to different time discount factors in Eq 6 [46 , 75] . This would parallel a number of other recent suggestions that different striatal loops model the world at different levels of hierarchical abstraction [47 , 76] , while also harmonizing the somewhat underspecified model-based evaluation process these theories assume with the predominant temporal difference account of striatal learning . Although our presentation culminated with proposing an algorithm ( SR-Dyna ) that can in principle perform equivalently to full model-based learning using value iteration , this need not be the only goal and there need not be only a single answer . The behaviors associated with model-based learning may not have unitary sources in the brain but may instead be multiply determined . All of the algorithms we have considered are viable candidate pieces of a larger set of decision systems . Notably , the experiments we have highlighted as suggesting striatal or dopaminergic involvement in “model-based” learning and inspiring the present work all use extremely shallow planning problems ( e . g . operant lever pressing , two-stimulus Pavlovian sequences , or two-step MDPs ) together with reward revaluation designs . Even SR-TD is sufficient to explain these . It may well be that planning in other tasks , like chess , or in spatial mazes , is supported by entirely different circuits that really do implement something like tree search; or that they differentially require replay , like SR-Dyna . Also , although replay-based approaches go a long way , value computation at choice time using more traditional model-based approaches is likely needed at the very least to explain the ability to evaluate truly novel options ( like the value of “tea jelly”; [77] ) using semantic knowledge . Some evidence that rodents may use more than just replay to compute values , even in spatial tasks , comes from findings that the prevalence of sharp-wave-ripples , a putative sign of replay , is inversely related to the prevalence of vicarious trial and error behaviors , a process thought to be involved in decision-time value computation , potentially by standard MB dynamic programming or alternatively SR-MB [78] . Relatedly , if the brain might cache both endpoint decision variables like Q , or their precursors like M , update either or both with off-line replay , and optionally engage in further model-based recomputation at choice time , then the arbitration or control question of how the brain prioritizes all this computation to harvest rewards most effectively and efficiently becomes substantially more complicated than previously considered . The prioritization of replay–which memories to replay when–becomes particularly important . The particular ordering and dynamics of replay are also outside our modeling here: in order to focus our investigation on the simplest behavioral predictions of SR-Dyna , we chose the simplest , naive sampling scheme in which the agent replays a single state-action-state transition with uniformly random probability . This sampling strategy is not a mechanistic commitment ( nor of course does it reflect the dynamics of realistic hippocampal replay trajectories ) , and we expect that like Dyna-Q , SR-Dyna would work even more efficiently given more sophisticated replay prioritization schemes . In this regard , we expect that the prioritization of replay , like the arbitration between model-based vs model-free tradeoffs [1 , 41 , 79] , might operate according to the principles of efficient cost-benefit management . We expect that in addition to more typical observed patterns of replay , such a scheme may be able to explain cases where the replayed sequences are not a simple reflection of the animal’s current policy [80] . The current model is robust to differences in replay but would need to be extended with a more principled and detailed replay model to address these questions . With simulations , we have presented experiments that could be used to elicit recognizable behavior form the different algorithms proposed here . Although we ruled out the simplest approach , SR-TD , due to its inflexibility , it is worth more carefully considering the evidence against it . The main counterexamples to SR-TD are transition revaluation and detour tasks . Apart from the classic work of Tolman and Honzik [57] , the original results of which are actually quite mixed ( see [81] ) , there is surprisingly little evidence to go on . A number of different studies have shown that healthy animals will normally choose the shortest alternative route after learning about a blockade preventing a previously preferred route ( e . g . [82–84] ) . However , in these studies , the animal learns about the blockade after starting from the maze starting location . Thus , unlike in our simulations in which the animal learns about the blockade in isolation , animals in these tasks would have the opportunity to learn from direct experience that maze locations leading up to the blockade are no longer followed by maze locations further along the previously preferred path . Such tasks could thus potentially be solved by SR-TD . Studies that show that animals will take a shortcut to a goal that is discovered along a preferred path present a somewhat cleaner test for SR-TD [85 , 86]; however it is often difficult to interpret a potential role of exploration or visual ( rather than cognitive map ) guidance in the resulting behavior . Work in humans , however , seems to more clearly suggest an ability to solve detour tasks without re-learning [87] . Simon and Daw [24] for instance directly assessed SR-TD’s fit to human subjects’ choice adjustments in a changing spatial maze , and found it fit poorly relative to traditional model-based learning . Overall , additional careful work that measures how animals respond to transition changes , learned in isolation , is needed . Whereas Tolman’s other early reward revaluation experiments ( latent learning ) have been conceptually replicated in many modern , non-spatial tasks like instrumental reward devaluation and sensory preconditioning , the same is not true of detours . Indeed , the modern operant task that is often presented as analogous to detours , so-called instrumental contingency degradation ( e . g . , [88] ) , is not functionally equivalent . In such tasks , the association between an action and its outcome is degraded through introduction of background rewards . However , because the information about the changed contingency is not presented separately from the rest of the experience about actions and their rewards , unlike all the other tests discussed here , contingency degradation as it has been studied in instrumental conditioning can actually be solved by a simple model-free learner that re-learns the new action values . The puzzle here is actually not how animals can solve the task , but why they should ever fail to solve it . This has thus led to a critique not of model-based but of model-free learning theories [47] . In any case , the modeling considerations proposed here suggest that more careful laboratory work on “transition revaluation” type changes to detect use of SR-TD , is warranted . Similarly , “policy revaluations” along the lines of that in Fig 5 would be useful to detect to what extent planning along the lines of SR-MB is contributing . Finally , although SR-Dyna in principle can perform model-based value computation , this depends on sufficient replay . The SR-Dyna hypothesis suggests the testable prediction that behavior should degrade to SR-TD under conditions when replay can contribute less . A number of experiments in the rodent literature have explored the behavioral deficits that result from interrupting sharp-wave ripples ( events in which hippocampal replay is known to occur ) . Such manipulations have been shown to produce behavioral deficits that are consistent with the SR-Dyna hypothesis , yet not exclusively predicted by it . For instance , two studies found that suppression of hippocampal sharp-wave ripples during rest slows down acquisition of the correct behavioral policy in spatial learning tasks in which the task environment is static [89 , 90] . These results are consistent with the notion that the purpose of replay is to provide additional experience , which is used to update some representation relevant to learning . However , these results are not specifically diagnostic of the successor matrix . For instance , preventing replay would slow down policy acquisition for Dyna-Q as well as SR-Dyna ( S2 Fig ) . Another study found a more specific effect of suppressing hippocampal sharp wave ripples during performance of a task . Here , the manipulation caused learning deficits selective for a subset of trials in which animals faced a hidden-state problem . In particular , these were trials in which animals had to choose which direction to turn depending on the events of the previous trial [91] . Such tasks constitute hidden-state problems , in that the “state” required to make the correct choice cannot be deduced entirely from an animal’s immediate sensory experience . In RL terms , to solve these problems , the animals must construct an augmented internal state , distinct from a simple representation of the immediate sensory situation [92] . One interpretation of the experimental result is that blocking replay interfered with this internal state construction process . This result resonates with our SR-Dyna proposal , which also posits that replay is involved in constructing the mapping between the sensory state and a different , augmented internal representation of it: the SR . However , our model as currently specified augments the state space with predictive features to support model-based flexibility , and does not currently address other sorts of elaborations of the state input that have been used in other work to facilitate learning in situations with hidden state or other uncertain sensory input [93–95] . Fully understanding these results therefore requires augmenting our model to address hidden state as well as state prediction . In fact , these two functions may be closely related: a number of approaches to the hidden state problem in the computational RL literature address it using predictive representations that are related to the SR [96 , 97] . In human studies , factors like the duration of off-task rest periods and presence of distractor tasks during such periods have been manipulated to extend , limit , or interfere with replay . Some evidence suggests that distractor tasks at decision time have no effect on reward revaluation [27] , consistent with SR-Dyna . Other recent work has demonstrated that humans benefit from additional pre-decision time in revaluation tasks that closely resemble “policy revaluation” [98] and that this benefit recruits a network including the prefrontal cortex and basal ganglia . Such work is consistent with the predictions of both Dyna-Q as well as SR-Dyna accounts of value updating presented here . Overall , future work will need to combine such manipulations of replay , with the three revaluation tasks imagined in this paper and demonstrate differences in the effects of manipulations on reward versus transition and policy revaluations . We have recently demonstrated , though without attempting to manipulate replay , that humans are worse at adjusting behavior following transition and policy revaluations compared to reward revaluations , suggesting that they may at least partially use either an SR-TD or SR-Dyna ( with limited sample backups ) strategy for evaluation [59] . We have suggested , on the basis of rodent lesion studies , that the SR may be encoded in parts of prefrontal cortex that project to dorosomedial striatum . However , we should note that recent work has also implicated the hippocampus as a potential site of the SR . Specifically , a state-state version of the SR can explain some properties of hippocampal place cells [18 , 99] as well as fMRI measures of the representation of visual stimuli in tasks where such stimuli are presented sequentially [100 , 101] . This work has largely built on ideas of the hippocampus in general as a site of cognitive map [102] as well as prior suggestions that hippocampal place cells may in fact encode the transition structure of the environment [103] and that such transition information may make them ideal basis functions for TD learning [104] If a state version of the SR exists in the hippocampus , we think it is reasonable that value weights would be leaned by neurons connecting the hippocampus to ventral striatum , in the same TD manner discussed in this paper . However , we also think a case can be made for the prefrontal cortex as another candidate basis for an SR . In addition to the rodent lesion evidence reviewed in the introduction of this paper , the prefrontal cortex shares many cognitive-map properties observed in the hippocampus [105] and has been suggested to be the basis of state representations for reinforcement learning [93 , 106] . A number of human studies have demonstrated the PFC’s role in the representation of prospective goals [107 , 108] . Furthermore , unlike the hippocampus , parts of the PFC appear to be involved in action representation in addition to state representation [109] , thus making it a candidate to hold a potential state-action version of the successor matrix . Importantly , these proposals are in no way mutually exclusive . For instance , recent work has demonstrated that hippocampal output is necessary for preserving PFC representations of task structure [110] . Overall , further experimental work will be required to determine whether either or indeed both these areas serves as the basis for the successor representation , and what specific roles they play in learning and representation . Finally , the SR may contribute to a number of other cognitive processes . Above we noted that there is evidence that areas of medial temporal lobe seem to encode predictive representations . In line with this , it has been noted that there is a close correspondence between the update rule used by SR-TD and update rules in the temporal context model of memory [19] . Also , recent approaches to reinforcement learning in the brain have advocated for a hierarchical approach in which punctate actions are supplemented by temporally abstract policies [111] . In this context , it has been suggested that the SR may be useful for discovering useful temporal abstractions by identifying bottlenecks in the state space that can then be used to organize states and action into a hierarchy [18 , 112] . The efficacy of the SR for model-based RL opens the possibility that the brain accomplishes planning , action chunking , and grouping episodic memories using a common mechanism . Overall , this article has laid out a family of candidate mechanistic hypotheses for explaining the full range of behaviors typically associated with model-based learning , while connecting them with the circuitry for model-free learning as currently understood . In addition to the transition and policy revaluation behavioral experiments suggested above , future neuroimaging work could seek evidence for these hypotheses . Specifically , failures to flexibly update decision policies that are caused by caching of either the successor representation ( as in SR-TD or SR-Dyna with insufficient replay ) or a decision policy ( as in SR-MB ) should be accompanied by neural markers of non-updated future state occupancy predictions . Such neural markers could be identified using representational similarity analysis ( e . g . [113] ) , cross-stimulus suppression ( e . g . [114] ) or through use of category specific , decodable , visual stimuli ( e . g . [115] ) . Similar work in experimental animals such as rodents ( e . g . [116] ) could use the full range of invasive tools to trace the inputs to dorsomedial vs . dorsolateral striatum , so as to examine the information represented there and how it changes following the various sorts of revaluation manipulations discussed here . As has been the case for model-free learning , the emergence of an increasingly clear and quantitative taxonomy of different candidate algorithms is likely to guide this work and help to elucidate the neural basis of model-based learning .
All simulations were carried out in 10x10 ( N = 100 states ) grid-worlds in which the agent could move in any of the four cardinal directions , unless a wall blocked such a movement . States with rewards contained a single action . Upon selecting that action , the agent received the reward and was taken to a terminal state . Each task was simulated with each algorithm 500 times . For each simulation , we recorded the agent’s value function at certain points . For SR-Dyna , which worked with action values rather than state values , the state value function was computed as the max action value available in that state . Figures display the median value , for each state , over the 500 runs . To determine the implied policy for the median value function , we computed , for each state , which accessible successor state had the maximum median value . | According to standard models , when confronted with a choice , animals and humans rely on two separate , distinct processes to come to a decision . One process deliberatively evaluates the consequences of each candidate action and is thought to underlie the ability to flexibly come up with novel plans . The other process gradually increases the propensity to perform behaviors that were previously successful and is thought to underlie automatically executed , habitual reflexes . Although computational principles and animal behavior support this dichotomy , at the neural level , there is little evidence supporting a clean segregation . For instance , although dopamine—famously implicated in drug addiction and Parkinson’s disease—currently only has a well-defined role in the automatic process , evidence suggests that it also plays a role in the deliberative process . In this work , we present a computational framework for resolving this mismatch . We show that the types of behaviors associated with either process could result from a common learning mechanism applied to different strategies for how populations of neurons could represent candidate actions . In addition to demonstrating that this account can produce the full range of flexible behavior observed in the empirical literature , we suggest experiments that could detect the various approaches within this framework . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"learning",
"medicine",
"and",
"health",
"sciences",
"neurochemistry",
"chemical",
"compounds",
"decision",
"making",
"applied",
"mathematics",
"dopaminergics",
"brain",
"social",
"sciences",
"neuroscience",
"organic",
"compounds",
"learning",
"and",
"memory",
"hormones",
"simulation",
"and",
"modeling",
"algorithms",
"cognitive",
"psychology",
"mathematics",
"animal",
"behavior",
"cognition",
"amines",
"neurotransmitters",
"catecholamines",
"zoology",
"dopamine",
"research",
"and",
"analysis",
"methods",
"behavior",
"neurochemicals",
"chemistry",
"neostriatum",
"biochemistry",
"psychology",
"organic",
"chemistry",
"anatomy",
"biogenic",
"amines",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"cognitive",
"science"
] | 2017 | Predictive representations can link model-based reinforcement learning to model-free mechanisms |
Recently , we reported on a new class of naphthoquinone derivatives showing a promising anti-trypanosomatid profile in cell-based experiments . The lead of this series ( B6 , 2-phenoxy-1 , 4-naphthoquinone ) showed an ED50 of 80 nM against Trypanosoma brucei rhodesiense , and a selectivity index of 74 with respect to mammalian cells . A multitarget profile for this compound is easily conceivable , because quinones , as natural products , serve plants as potent defense chemicals with an intrinsic multifunctional mechanism of action . To disclose such a multitarget profile of B6 , we exploited a chemical proteomics approach . A functionalized congener of B6 was immobilized on a solid matrix and used to isolate target proteins from Trypanosoma brucei lysates . Mass analysis delivered two enzymes , i . e . glycosomal glycerol kinase and glycosomal glyceraldehyde-3-phosphate dehydrogenase , as potential molecular targets for B6 . Both enzymes were recombinantly expressed and purified , and used for chemical validation . Indeed , B6 was able to inhibit both enzymes with IC50 values in the micromolar range . The multifunctional profile was further characterized in experiments using permeabilized Trypanosoma brucei cells and mitochondrial cell fractions . It turned out that B6 was also able to generate oxygen radicals , a mechanism that may additionally contribute to its observed potent trypanocidal activity . Overall , B6 showed a multitarget mechanism of action , which provides a molecular explanation of its promising anti-trypanosomatid activity . Furthermore , the forward chemical genetics approach here applied may be viable in the molecular characterization of novel multitarget ligands .
Among the tropical diseases , there are maladies whose etiological agents belong to the Trypanosomatidae family of the Protista , order Kinetoplastea , that are responsible for infections concentrated in the poorest , mainly rural areas of the planet , and that are grouped under the name of “most neglected diseases” [1] . In particular , parasites of the genus Trypanosoma are responsible for Chagas' disease in Latin America and sleeping sickness in sub-Saharan Africa [2]–[5] . Because of their occurrence in low-income and middle-income countries , these diseases do not have high visibility in Western societies , although sleeping sickness is among the neglected tropical diseases with the highest rates of death [6] . Vaccine development has been hampered by either the high degree of antigenic variation as exhibited by the bloodstream dwelling African trypanosome , Trypanosoma brucei , and the localization of the American trypanosome , Trypanosoma cruzi , within cells of the human host , despite a successful experimental oral vaccine based on attenuated T . cruzi has been reported [7] . In this context , chemotherapy still represents a viable option for treatment of these infections [8] . However , the majority of the currently available drugs are decades old ( some back to 1920 ) and have , unfortunately , many limitations , including high toxicity and the emergence of drug resistance . The latter issue has called for designing innovative approaches to drug discovery for infections by trypanosomes [9] , [10] . A major role in this respect is played by combination therapy , which has been shown to be a possible strategy for both preventing and overcoming chemotherapy-induced resistance [11] . A logical alternative to combination therapy is the development of drugs able to hit multiple targets [12] , [13] . Such multitarget compounds are single chemical entities that can provide the same pharmacological profile as drug combinations , but potentially with fewer side effects . In fact , when two or more drugs are administered as a combination , there is a possibility that the drugs may interact with each other ( drug-drug interaction ) . This interaction could increase or decrease the effective concentration of one of the drugs or , more frequently , could even enhance the adverse effects . Indeed , single multitarget compounds have a much simpler pharmacokinetic profile than combination therapy , also prevent possible side effects due to drug-drug interactions , greatly simplify the therapeutic regimen , with positive consequences for patient adherence and caregiver compliance , and finally an overall improved selectivity . Furthermore , the easier and cheaper manufacturing and formulation of a single active pharmaceutical ingredient would make multitarget drugs inherently more cost-effective and widely accessible than combinations [14] . It should be mentioned that if there is any synergism or additive effect among the targets , then the effective dosage of a multitarget drug is most likely lower than that of a single-target drug . When lowering the therapeutic dose , however , it will be crucial to find a balance between decreasing the dose to avoid side effects and keeping it sufficiently high to prevent the development of resistance . On these premises , it has been proposed that against trypanosomatid-borne diseases such compounds may prove more efficacious , tolerable , and affordable than the available arsenal of drugs [12] , [15] . Naphthoquinone and other quinone derivatives have been reported as one of the major natural product classes with significant activity against Trypanosoma [16]–[18] . For instance , lapachol exhibits a marked anti-trypanosomal profile , while displaying no serious toxic effects in humans [19] . In view of the well-known biological properties of this class of compounds , it is highly possible that naphthoquinones exert their anti-trypanosomatid profile by means of a multitarget mechanism . Indeed , a multitarget profile for this class of compounds is easily conceivable , because quinones , like many other natural products , provide plants with potent defense chemicals with an intrinsic multifunctional mechanism of action [20] . Furthermore , it can be hypothesized that in addition to a possible target-related mechanism , the general free-radical-generation mechanism of quinones – probably also at the basis of their general cytotoxicity – may contribute to multitarget profile of these molecules [21] [22] . Indeed , it has been reported in the literature that parasitic protists are particularly sensitive to oxidative stress [23] . In this field , we have recently reported on the preparation of a focused library of 16 compounds based on the 1 , 4-naphthoquinone and 1 , 4-anthraquinone natural occurring scaffolds [24] . From this small compound collection , several molecules were active against Trypanosoma and Leishmania at low concentrations . Some of the derivatives exhibited potency in the nanomolar range , with one ( 2-phenoxy-1 , 4-naphthoquinone , B6 in Figure 1 ) displaying an ED50 value of 80 nM against Trypanosoma brucei rhodesiense , as assessed in experiments using in vitro cultured parasites . It also showed a selectivity index ( ratio of the compound's ED50 values on mammalian cell lines and trypanosomes ) of 74 [24] , which is very close to the specifications required by WHO/TDR to be considered an anti-trypanosomatid hit [25] . However , the molecular mechanism and the target ( s ) responsible for the biological profile of this class of compounds have remained undisclosed . Here , by means of a chemical proteomics approach , we aimed at identifying the putative molecular target ( s ) of B6 . In particular , using its immobilized derivative 1 ( Fig . 1 ) , we isolated its targets from parasite extracts . Then , by means of biochemical experiments , we verified the ability of the molecule to bind to recombinant forms of the identified targets . In light of the general property of naphthoquinones to generate free radicals , we finally analyzed oxygen consumption in permeabilized trypanosomes and production of reactive oxygen species ( ROS ) in trypanosome mitochondrial cell fractions . This allowed us to elucidate additional B6 potential mechanisms of action , as chemical proteomics is not suited to identify non-protein targets .
Before covalently attaching B6 to the solid support through a linker , we analyzed which site ( s ) of the molecule were more appropriate for linking purposes . To this end we synthesized derivatives 1–3 ( Fig . 1 ) , which carry in different positions amino or hydroxyl groups that can be easily exploited as anchor points . The synthesis was carried out according to the procedure reported for B6 [24] , and which relies on the substitution of a 2-bromonaphthoquinone with the corresponding phenol . Synthesis of 2- ( 4-amino-phenoxy ) -[1] , [4]naphthoquinone ( 1 ) . To a stirred solution of 4-amino-phenol ( 0 . 46 g , 4 . 2 mmol ) in 90 ml dimethylformamide ( DMF ) potassium carbonate ( 1 . 70 g , 12 . 3 mmol ) was added . After stirring for 1 h at room temperature , 2-bromo-[1] , [4]-naphthoquinone ( 1 . 0 g , 4 . 2 mmol ) was added . After stirring for further 3 h , the reaction was diluted with water and ice ( 500 ml ) and the resulted brownish solid was collected by filtration to give 0 . 49 g of crude 1 , which was crystallized by EtOH/water ( 40% yield ) . IR ( Nujol ) 3421 , 3382 , 1680 , 1635 , 1609 , 1508 , 1204 , 980 , 718 cm−1; 1H NMR ( 300 MHz , CDCl3 ) δ 8 . 25-8 . 22 ( m , 1H ) , 8 . 12-8 . 09 ( m , 1H ) , 7 . 80-7 . 78 ( m , 2H ) , 6 . 95 ( d , 2H , J = 8 . 5 Hz ) , 6 . 76 ( d , 2H , J = 8 . 5 Hz ) , 6 . 04 ( s , 1H ) , 3 . 77 ( br s , 2H , exchangeable with D2O ) ; HRMS ( ES ) m/z calculated for C16H11NO3Na 288 . 0637 , found 288 . 0639 [M++Na+] . Synthesis of 5-hydroxy-2-phenoxy-[1] , [4]naphthoquinone ( 2 ) . It was synthesized in 33% yield from phenol and 2-bromo-5-hydroxy-[1] , [4]-naphthoquinone , following the procedure reported for 1 and crystallization from methylene chloride/petroleum ether . 1H-NMR ( CDCl3 , 400 MHz ) : δ 5 . 88 ( s , 1H ) , δ 7 . 13 ( d , J = 8 . 4 , 2H ) , δ 7 . 26-7 . 35 ( m , 2H ) , δ 7 . 47 ( t , J = 7 . 6 , 2H ) , δ 7 . 61 ( t , J = 8 . 0 , 1H ) , δ 7 . 74 ( d , J = 7 . 2 , 1H ) , δ 12 . 10 ( s , exch , 1H ) ; 13C-NMR ( CDCl3 , 400 MHz ) : δ 113 . 17 , 114 . 40 , 119 . 91 , 121 . 22 , 125 . 56 , 127 . 03 , 130 . 70 , 131 . 30 , 135 . 90 , 152 . 77 , 161 . 30 , 161 . 34 , 179 . 43 , 191 . 07; MS ( ESI+ ) m/z:289 [M++Na+] . Synthesis of 8-hydroxy-2-phenoxy-[1] , [4]naphthoquinone ( 3 ) . It was synthesized in 35% yield from phenol and 2-bromo-8-hydroxy-[1] , [4]-naphthoquinone , following the procedure reported for 1 and crystallization from methylene chloride/petroleum ether . 1H-NMR ( CDCl3 , 400 MHz ) : δ 5 . 94 ( s , 1H ) , δ 7 . 13 ( d , J = 8 . 8 , 2H ) , δ 7 . 26-7 . 34 ( m , 2H ) , δ 7 . 47 ( t , J = 7 . 6 , 2H ) , δ 7 . 59-7 . 67 ( m , 2H ) , δ 11 . 79 ( s , exch , 1H ) ; 13C-NMR ( CDCl3 , 400 MHz ) : δ 114 . 18 , 119 . 22 , 121 . 22 , 124 . 30 , 126 . 98 , 130 . 70 , 132 . 20 , 137 . 48 , 152 . 77 , 160 . 38 , 162 . 25 , 184 . 25; MS ( ESI+ ) m/z:289 [M++Na+] . The synthesized compounds were then tested against T . b . rhodesiense parasites as previously reported [24] , and their activities are shown in Figure 1 along with that of B6 . Because of its superior trypanocidal activity , 1 was selected for further immobilization studies . For identification of the B6 targets , T . b . rhodesiense was used . The parasites were isolated from blood of an infected mouse ( received from the Swiss Tropical and Health Institute , Basel , Switzerland ) . Parasite lysates were prepared as described previously [26] . Briefly , 5×108 cells were lysed in 200 µl lysis buffer consisting of 160 µl PBS ( 38 mM Na2HPO4 , 2 mM NaH2PO4 , 29 mM NaCl , pH 8 . 0 ) containing 44 mM glucose and 40 µl of lysis buffer concentrate ( 100 mM HEPES , pH 7 . 5 , 750 mM NaCl , 5% Triton-X-100 , 10 mM tris ( 2-carboxyethyl ) phosphine ( TCEP ) , 50% glycerol , and 1 µl/ml protease cocktail ( Sigma ) . After short periods of sonication the mixture was centrifuged for 10 min at 14 , 000 g , the supernatant was recovered and stored in aliquots of 100 µl at −80°C , until needed . The protein concentration , determined using the Bradford dye assay with BSA as reference protein , was found to be 8 . 4 mg/ml . 1 was immobilized to epoxy-activated Sepharose 6B using a modified protocol described earlier [26] . To this end , swollen and thoroughly washed matrix was resuspended in two volumes of 35 mM ligand dissolved in a solution containing 50% dioxane/50% H2O ( solution 1 ) . Coupling was performed for 48 h at 40°C using 40 µl of swollen resin mixed with 80 µl of ligand solution . After coupling , the matrices were centrifuged for 1 min at 3000 g , and then washed 4–5 times with 10 volumes ( 400 µl ) of solution 1 . Afterwards , remaining free epoxides were reacted by adding 1 ml of a solution containing 50% dioxane and 50% acetic acid ( pH 3 . 2 ) and incubation for 20 h at 40°C . After incubation , the samples were centrifuged at 3000 g for 2 min and the supernatants collected . The resin was finally washed in five rounds , each with 10 volumes of solution 1 . Again all supernatants were collected . Supernatants of all steps were collected in a 25 . 0 ml volumetric flask and used to determine the amount of ligand washed off . Direct absorbance of the scans of the immobilized ligand on the matrix resuspended in 50% glycerol solution ( v/v ) clearly confirmed successful coupling ( data not shown ) . The amount of inhibitor bound to the matrix was determined by back calculation of the amount of compound applied and amount recovered . Routinely 2–3 µmol/ml of compound were bound . A control matrix was prepared without ligand and treated as described above . Before incubation with the lysate the matrices were washed twice with water and then equilibrated in the lysate buffer by washing each resin three times with the lysate buffer . The lysate ( 100 µl; diluted to 2 . 1 mg/ml protein using PBS ) was incubated with the ligand-bound matrix for 2 h at 4°C while mixing at 700 rpm . Then , the matrix was washed 5 times with the lysate buffer while mixing for 90 sec at 1400 rpm . The control matrix was incubated with the same amount of lysate and treated equally . Finally , both matrices were washed with a solution containing 5 mM HEPES ( pH 7 . 0 ) and sent to the Functional Genomics Center of Zurich for protein identification directly from the matrices . To this end , resins were re-suspended in 50 µl trypsin solution ( 10 ng/µl trypsin in 10 mM Tris-HCl , 2 mM CaCl2 , pH 8 . 2 ) and incubated at 37°C overnight . Supernatants were separated and beads extracted twice with 5% formic acid in 10% acetonitrile . All three supernatants of the corresponding matrices were combined , dried , and then dissolved in 25 µl 0 . 1% formic acid . 2 µl were injected via an autosampler and run with two different gradients ( A and B ) for LC/ESI/MS/MS-QTOF analysis . Database searches were performed by using the ProteinLynx Global Server ( Swiss Prot , all species ) and Mascot ( Swiss Prot , excluding the major eukaryotic species from the search ) search programs . Only hits with enough independent peptide spectra to give a probability of over 95% were considered for the study . The expression construct has been obtained by subcloning the corresponding GK gene ( isolated from T . b . brucei strain Lister 427 ) from the formerly designed pET15-TbGK plasmid [27] into vector pET28 to create the final expression plasmid pET28-TbGK ( unpublished data ) . TbGK was then expressed in E . coli BL21 ( DE3 ) -CodonPlus-RIL grown in 1 liter LB medium supplemented with kanamycin ( 50 µg/ml ) and chloramphenicol ( 34 µg/ml ) for 16 h at 37°C . Expression was induced by adding 1 mM isopropyl β-d-thiogalactanopyranoside ( IPTG ) , and growth was continued for another 2 h at 37°C . Bacteria were harvested ( 20 min , 4°C , 5000 rpm ) , resuspended in a mixture of 96% buffer A ( 20 mM Tris-HCl , 200 mM NaCl , pH 7 . 6 ) and 4% buffer B ( 20 mM Tris-HCl , 200 mM NaCl , 500 mM imidazole , pH 7 . 6 ) , and supplemented with a small amount of DNase . The suspension was passed twice through a French Press , centrifuged ( 20 min , 4°C , 9800 rpm ) , and the clarified crude extract was applied onto a 5 ml Ni-chelating column . TbGK was eluted using a 10 column volumes linear imidazole-gradient using buffer A and buffer B . Eluted fractions were analyzed by SDS-PAGE . Fractions containing TbGK were combined , supplemented with 2 mM CaCl2 and 10 U/ml thrombin to cleave off the histidine tag . After overnight incubation at 16°C , the digested protein was concentrated and desalted using a Superdex 75 column equilibrated with buffer A . Protein concentration was determined at 280 nm ( extinction coefficient = 81650 M−1 cm−1 ) . The protein was diluted to a concentration of 2 mg/ml and stored at 4°C . The TbGK activity was measured using a modified version of a continuous enzyme-coupled spectrophotometric assay developed previously [28] . Briefly , the ATP consumption associated with glycerol phosphorylation was coupled to the oxidation of NADH via the coupled pyruvate kinase/lactate dehydrogenase enzyme pair . The assays were performed in 1 . 0 ml triethanolamine/HCl buffer ( 0 . 1 M triethanolamine , 2 . 5 mM MgSO4 , 10 mM KCl , 5 mM glycerol , 0 . 1 mM ATP , 2 . 2 mM phosphoenolpyruvate , 0 . 1% DMSO , pH 8 . 0 ) , at 25°C in the presence of 5 U pyruvate kinase , 3 U lactate dehydrogenase , and 150 ng TbGK . The concentration of B6 was varied in the range of 1 nM to 10 µM . The reaction was started by adding 0 . 42 mM NADH . The IC50 value of B6 was calculated as the mean of three independent experiments . T . b . brucei glyceraldehyde-3-phosphate dehydrogenase ( TbGAPDH ) was expressed using expression vector pET28a carrying the GAPDH gene which had been subcloned from a former expression clone ( T . b . brucei strain Lister 427 ) based on pET3a [29] using NdeI and BamHI restriction enzymes ( unpublished results ) . A culture of E . coli strain BL21 ( DE3 ) harboring an expression plasmid with TbGAPDH was grown at 37°C in 100 ml of LB medium . When the OD at 600 nm was between 0 . 5 and 0 . 8 , expression was induced by addition of 1 mM IPTG . Growth was continued overnight . TbGAPDH was purified essentially as above reported for TbGK with the only difference that the clarified crude extract was applied onto a Talon resin column ( Talon , Clontech ) , and that the protein of interest was eluted with 200 mM imidazole . TbGAPDH was identified by SDS-PAGE and Coomassie blue staining . The kinetics of the TbGAPDH reaction was monitored using a continuous enzyme-coupled spectrophotometric assay , as previously described [29] . In brief , TbGAPDH activity was measured following the NADH oxidation at 340 nm , in a coupled assay with 3-phosphoglycerate kinase and using a Jasco V-550 spectrophotometer . All measurements were performed at 25°C in 0 . 01 M triethanolamine , pH 7 . 6 , 1 . 7 mM ATP , 1 mM EDTA , 100 mM KCl , 5 mM MgSO4 , 1 . 7 mM NaHCO3 , 25 µg 3-phosphoglycerate kinase . The IC50 values were determined in a final volume of 1 ml in the presence of 120 ng TbGAPDH and 5 . 6 mM 3-phosphoglycerate ( 3-PGA ) , while varying the B6 concentration after a pre-incubation of enzyme plus inhibitor for 10 min . The reaction was started by the addition of 0 . 2 mM NADH . Each point of the curve was measured in triplicates and the value for IC50 was estimated from graphically plotted dose-response curves . The type of inhibition was determined with respect to NADH and 3-PGA in consideration of Michaelis-Menten steady-state conditions . To investigate the inhibition mechanism with respect to NADH , TbGAPDH was incubated for 10 min at room temperature with different inhibitor concentrations ( 0–15 µM ) in a total volume of 1 ml . The reaction was initiated by the addition of 5 . 6 mM 3-PGA and NADH ( ranging from 5 µM to 200 µM ) . The inhibition mechanism with respect to 3-PGA was determined in a similar way . To this end , TbGAPDH was incubated with varying inhibitor concentrations ( 10 nM–100 µM ) . The reaction was initiated by the addition of 0 . 2 mM NADH and 3-PGA ( ranging from 50 µM to 5 . 6 mM ) . The α and Ki values were obtained from Dixon and secondary plots . The reported values represent the mean of two independent experiments . For a comprehensive analysis of the effect of simultaneous inhibition of GAPDH and GK on the glycolytic flux by B6 , we used a mathematical model . Modeling was done with the T . brucei glycolysis model [30] in the open-source software package PySCeS [31] with Python 2 . 6 . In the model version used here , the equations for the cytosolic and glycosomal adenylate kinases were replaced with mass action kinetics consistent with the equilibrium constant of 0 . 442 mM [32] , the glycerol-3-phosphate oxidase ( GPO ) has access to the glycosomal pools of glycerol-3-phosphate ( G3P ) and dihydroxyacetone phosphate and phosphoglycerate mutase has access to the glycosomal pool of 3-PGA . These adjustments hardly affected control distribution and the glycolytic flux in this version of the model is equally sensitive to glycerol inhibition at anaerobic conditions as the model described in [30] . To simulate titration of a non-competitive inhibition , the Vmax of GAPDH and/or GK was multiplied by , using a Ki for GAPDH of 4 µM ( based on measurements in this paper: 3 . 6 or 5 . 0 µM depending on the substrate that was varied ) and for GK 0 . 90 µM ( as reported in this paper ) . To simulate anaerobic conditions , the Vmax of the short mitochondrial respiratory chain ( in the model referred to as GPO , consisting of a mitochondrial FAD-dependent glycerol-3-phosphate dehydrogenase , ubiquinone and the alternative oxidase ( TAO ) ) was set to zero . Bloodstream- and procyclic-form T . b . brucei cells ( strain Lister 427 , cell line 449 ) were cultured up to the exponential growth phase and homogenates were obtained by grinding the washed parasite pellets with silicon-carbide abrasive grain ( mesh 300 ) in disruption buffer containing 250 mM sucrose , 25 mM Tris-HCl and l mM EDTA , pH 7 . 8 ( STE buffer ) . Differential centrifugation [33] was performed as follows: the suspension was taken up in another 3 ml STE buffer and centrifuged at 30 g for 3 min . The supernatant , representing the cell homogenate , was centrifuged at 1 , 500 g for 10 min giving the nuclear fraction . The post-nuclear supernatant was then centrifuged at 5 , 000 g for 10 min giving the large-granular ( mitochondria-enriched ) fraction as pellet . This fraction was resuspended in 300 µl STE buffer . Oxygen consumption was monitored in a thermostated vessel at 37°C with a Clark electrode ( oxygraph ) . Measurements in permeabilized bloodstream-form T . brucei were done in buffer 1 ( 96 . 9 mM NaCl , 3 . 1 mM KCl , 5 mM MgCl2 , 2 mM Na2HPO4 , 90 mM Tris at pH 7 . 5 ) . Batches of 2×107 cells were pelleted , washed in buffer 1 and incubated for 5′ on ice in 1 ml of buffer 1 containing 1 ng digitonin to permeabilize the cells . Subsequently the permeabilized cells were pelleted and washed twice with buffer 1 without digitonin after which the pellet was taken up in 1 ml buffer 1 and transferred to the oxygraph . Substrates and inhibitors were added as described in the text . 2 mM salicylhydroxamic acid ( SHAM ) was used to completely inhibit the T . b . brucei alternative oxidase ( TAO ) that is part of the GPO . B6 and SHAM were dissolved in DMSO . The DMSO concentration used in these oxygen consumption measurements was less than 0 . 7% . The method used to measure H2O2 production in T . b . brucei mitochondria is based on the fluorogenic probe 2′ , 7′-dichlorodihydrofluorescein diacetate ( H2DCFDA ) which emits an intense green fluorescence only after deacetylation and subsequent oxidation . ROS production by the mitochondrial fraction of bloodstream form T . b . brucei was measured in a 96-well microtiter plate using a fluorescence plate reader ( Victor Wallace multiplate reader ) . In each well , 0 . 25 mg mitochondrial protein/ml and 5 µM H2DCFDA to a final volume of 0 . 2 ml with 10 mM Tris-HCl , 50 mM KCl , 1 mM EDTA , pH 7 . 5 were present . The reaction , performed at 25°C , was started by the addition of 10 mM G3P , in the presence and absence of 10 µM B6 . We used bloodstream and procyclic forms of T . b . brucei strain Lister 427 , cell lines 449 [34] , constitutively expressing the E . coli tetracycline ( Tet ) repressor gene via the chromosomally integrated plasmid pHD449 that also confers phleomycin resistance . Bloodstream forms were cultured in HMI-9 medium containing 10% heat-inactivated foetal calf serum ( Invitrogen ) and 0 . 18 µg/ml phleomycin ( Cayla ) at 37°C under water-saturated air with 5% CO2 . Procyclic trypanosomes were grown in SDM79 medium [35] supplemented with 15% foetal calf serum and 0 . 5 µg/ml phleomycin at 28°C under water-saturated air with 5% CO2 . The glucose-depleted SDM-80 medium , first employed by Lamour et al . [36] , was supplemented with 5 µg/ml hemin , 9% ( v/v ) dialyzed “glucose free” heat-inactivated foetal calf serum ( Sigma ) and 1% ( v/v ) of normal heat-inactivated foetal calf serum ( Invitrogen ) . T . b . rhodesiense in vitro growth inhibition activity of 1–3 ( Fig . 1 ) was assessed using bloodstream form of STIB 900 strain , following the procedure reported in Orhan et al . [37] . The anti-trypanosomal activity tests were also performed on T . b . brucei according to the “Long Incubation Low Inoculation Test” ( LILIT ) [38] , [39] using cultured bloodstream and procyclic forms of T . b . brucei strain Lister 427 , cell line 449 [34] . It should be noted that T . b . rhodesiense , used in the initial inhibition activity assays , and T . b . brucei , used in all molecular and biochemical studies and in some growth inhibition assays performed at a later stage , are different subspecies that at the molecular level are almost identical . The essentially only difference is that the former subspecies is human pathogenic , due to resistance to a lytic factor present in human serum , whereas the latter subspecies is susceptible to the lytic factor . In addition , variable expression levels of proteins may give rise to differences in drug susceptibilities between subspecies . Docking of B6 was carried out using the crystal structure of TbGAPDH ( PDBid: 2X0N ) [40] . The binding pocket was defined as 10 Å from Cys166 . Tautomeric states of histidines and the positions of asparagine and glutamine side chain amidic groups were optimized to improve the hydrogen bonding pattern . Polar hydrogen atoms were also optimized . The adopted force field was a modified version of the ECEPP/3 force field [41] . B6 was assigned the MMFF force field atom types and charges [42] . Docking simulations were carried out by means of the Biased Probability Monte Carlo stochastic optimizer as implemented in ICM [43] , [44] . The molecular conformation of the system was described by means of internal coordinate variables . The other binding site residues were represented by pre-calculated 0 . 5 Å spacing potential grid maps , representing van der Waals potentials for hydrogens and heavy atoms , electrostatics , hydrophobicity , and hydrogen bonding , respectively . The van der Waals interactions were described by a smoother form of the 6–12 Lennard-Jones potential with the repulsive contribution capped at a cutoff value of 4 kcal/mol . Poses from Monte Carlo sampling were rescored by means of the standard ICM empirical scoring function [45] .
A typical target isolation project begins with structure-activity relationship ( SAR ) studies , in which various portions of the small molecule of interest are modified to determine which one ( s ) can be used as points of attachment to a solid matrix [46] . It is important to note that several small molecules that have no sites appropriate for modification are not suited for affinity-based target isolation [47] . In our case , we successfully modified B6 to accommodate functional groups through which it could be covalently linked to the resin . We investigated three positions for the introduction of reactive amino or hydroxyl groups: one on the phenoxy moiety ( compound 1 ) and two on the napthoquinone portion ( compounds 2 and 3 ) ( Fig . 1 ) . Subsequent biological studies showed that the chemical modifications performed on B6 resulting in 1 and 3 did not have a dramatic impact on the observed anti-Trypanosoma profile . Conversely , 2 showed a significant decrease of the trypanocidal activity ( Fig . 1 ) . In particular , as 1 was only 7 times less active than B6 , we assumed that 1 might retain B6-like binding properties . Thus , we next performed affinity chromatography and target identification studies using 1 . A T . b . rhodesiense lysate was prepared as described in the Methods section , and 1 was immobilized on an affinity chromatography column . After LC separation using two gradients and subsequent mass spectrometry and database searches , four trypanosome proteins were identified with a probability of more than 95% , and which were found only on the coupled matrix but were absent in the control experiment ( Table S1 ) : i . e . , ( i ) glycosomal glycerol kinase ( TbGK; SwissProt accession code Q9NJP9 ) , ( ii ) tubulin beta chain ( P04107 ) , ( iii ) tubulin alpha chain ( P04106 ) , and ( iv ) glycosomal glyceraldehyde-3-phosphate dehydrogenase ( TbGAPDH; P22512 ) ( Table S1 ) . Two proteins , TbGK and TbGAPDH , were selected for further studies as putative targets of B6 , while tubulins were excluded because of the fact that they are well known to interfere with affinity chromatography studies due to their high abundance [48] . Indeed , unspecific association of T . b . brucei tubulin with affinity matrix has been observed in a similar affinity work by Mercer et al . [49] . For the chemical validation of both TbGK and TbGAPDH as putative targets of B6 , the proteins were recombinantly expressed and purified to near-homogeneity . After purification of TbGK the SDS-PAGE analysis revealed a band of apparent high purity and with a subunit molecular weight corresponding to that of TbGK ( 56 kDa ) ( Fig . 2A ) . The purified protein indeed possessed TbGK activity , which was inhibited by B6 with an IC50 value of 0 . 90±0 . 30 µM ( Fig . 3A ) . Expression and purification of TbGAPDH afforded pure and active enzyme exhibiting the expected subunit weight of 42 kDa ( Fig . 2B ) . The inhibition assay showed that B6 was also a good inhibitor of this enzyme , with an IC50 value of 7 . 25±1 . 62 µM ( Fig . 3B ) , strongly indicating that TbGAPDH interaction with immobilized compound 1 indeed was specific , and that TbGAPDH was not retained due to its well-known tendency to be isolated by non-specific association with chromatography resins [26] . These experiments confirmed the chemical proteomics results . In fact , both TbGK and TbGAPDH could be inhibited by B6 , and therefore both represent possible molecular targets of this naphthoquinone derivative . To investigate the metabolic effect of combined and separate non-competitive inhibition of GAPDH and/or GK , we performed simulations with the bloodstream-form T . brucei glycolysis model , based on the Lister 427 strain [30] . Inhibition of GAPDH and GK together has a strong effect on the ATP production flux ( Fig . 4A ) . Earlier results showed that 20–40% inhibition of glycolytic ATP production flux will result in 50% inhibition of the growth rate [50] . In the simulations , 40% inhibition of the ATP production flux is reached at an inhibitor concentration of 7 µM . This is only 3 . 5-fold lower than the experimentally determined ED50 of 24 . 8 µM for B6 on cultured bloodstream-form Lister 427 trypanosomes ( see Methods and below for details ) . Modeling the inhibition of GAPDH or GK separately for aerobic glycolysis showed that GAPDH inhibition alone could be sufficient for the full effect ( Fig . 4A ) . This can be expected from the fact that under aerobic conditions , there is only a small flux to glycerol . However , under anaerobic conditions , glucose is broken down to equimolar amounts of pyruvate and glycerol [50] . Therefore , we also did simulations under anaerobic conditions . In this context , B6 had an even a stronger effect ( Fig . 4B ) , but now GK inhibition alone would give an equally strong effect on the ATP production flux as the combined inhibition of GAPDH and GK . We conclude , based on the modeling , that inhibition of GK is therefore only important when the parasite experiences periods of low oxygen tension . Under aerobic conditions prevailing in most part of the blood circulation system , the non-competitive inhibition of GAPDH should be sufficient for maximal effect of B6 . Therefore , we further focused on the effect of B6 on GAPDH . Kinetic studies were performed to investigate the mechanism by which B6 inhibits TbGAPDH . Figure 5 shows TbGAPDH activity with respect to NADH ( Fig . 5A ) or 3-PGA ( Fig . 5B ) as substrate and at increasing concentrations of inhibitor B6 . Double reciprocal plots showed B6 to act as a non-competitive inhibitor: the fitted Ki was similar when NADH or 3-PGA was varied ( Ki of 3 . 60±0 . 57 µM for variation of NADH and Ki of 4 . 99±1 . 70 µM for variation of 3-PGA , exhibiting an α value of 0 . 6 and 0 . 4 , respectively ) . The factor α describes the effect of the inhibitor on the affinity of the substrate toward the enzyme and the effect of the substrate on the inhibitor affinity for the enzyme [51] . We measured the IC50 value of B6 after a pre-incubation of TbGAPDH with the inhibitor for 10 min and then diluted them 10-fold to the final concentration for the assay . After 1∶10 dilution , the compound was able to inhibit TbGAPDH with an IC50 value close to that obtained under standard assay conditions ( IC50 = 9 . 98±3 . 53 µM with dilution and IC50 = 7 . 25±1 . 62 µM without dilution; see Fig . 3B ) . The long pre-incubation time necessary for any B6 effect and the observation that IC50 value is not significantly affected by dilution suggested a possible tight or covalent binding inhibition mechanism for the compound [52] . A possible explanation for this behavior could be related to the cysteine trap mechanism of naphthoquinones previously reported for similar compounds [53] . In particular , we might hypothesize that B6 could trap Cys166 , which has been shown to play a crucial role in GAPDH's catalytic activity [54] . However , we cannot rule out that based on the observed non-competitive inhibition kinetics , B6 binds GAPDH outside the catalytic pocket . To investigate the possible covalent bond interaction , we tried to detect the B6/GAPDH complex by ESI-TOF and MALDI-TOF mass spectroscopies . However , this peptide fragment was not ionized enough to be detected by mass spectrometry under our test conditions ( data not shown ) . Since mass-spectroscopic analysis was not helpful for clarification of the covalent bonding interaction between Cys166 and B6 , we next exploited molecular docking to predict the plausible binding mode of B6 in the GAPDH active site . A docking simulation was performed using B6 and the available three-dimensional structure of TbGAPDH . The docking results ( Fig . 6A ) clearly showed that B6 could be suitably placed in the TbGAPDH active site to undergo a nucleophilic attack from the Cys166 side chain , and consequently form a covalent adduct , which can be responsible for the overall inhibition mechanism of B6 . In Figure 6B , two possible mechanisms by which the covalent adduct could be formed are shown . The Cys166 thiol undergoes 1 , 4-Michael addition to B6 to form the corresponding thioether-substituted hydroquinone . Alternatively , a reaction at C3 with the phenate displacement as leaving group and formation of the substitution product may take place . Both of these mechanisms have already been reported for phenoxybenzoquinone derivatives as VEGFR inhibitors [53] . Tests of growth inhibition by B6 were performed on procyclic trypanosomes cultured in medium containing glucose ( SDM79 medium ) , resulting in glycolytic activity for the cells' main energy supply , or without glucose ( SDM80 medium ) , where the proline is the main substrate for free energy production through mitochondrial metabolism . Under these latter conditions , the glycosomal GAPDH would still be needed for gluconeogenesis , to synthesize glucose 6-phosphate for production of glucoconjugates , but the gluconeogenic flux is assumed to be much lower than the glycolytic flux when glucose is present . It was observed that cells grown under the condition of active glucose metabolism were somewhat more susceptible for inhibition by B6 ( ED50 = 0 . 25±0 . 09 µM ) than the cells relying on proline metabolism as their main source of free energy ( ED50 = 0 . 67±0 . 16 µM ) . This is in support of our observation that TbGAPDH , a crucial enzyme of the trypanosomal glycolytic pathway , may be an intracellular target of B6 . Although this difference in ED50 values is relatively small , it may still be significant . In addition , the difference between ED50 and IC50 values ( see previous paragraph ) , may suggest that there should be another target , which plays an important role in causing B6's trypanocidal activity . Naphthoquinones have been shown to be able to generate free radicals ( mainly ROS ) at the mitochondrial level [55] . To determine whether B6 could exert its trypanocidal action also through this mechanism , respiration was monitored in permeabilized T . b . brucei bloodstream-form cells in the presence of B6 and/or SHAM ( Fig . 7 ) . SHAM is an inhibitor of TAO , which acts as the final electron acceptor in bloodstream form T . b . brucei . The differences in absolute oxygen consumption rate with 10 mM G3P as electron donor between individual experiments are probably due to different extents of permeabilization , and hence we also included the results in percentages ( Fig . 7B–D , lower panels ) . As expected , 2 mM SHAM inhibited respiration completely ( Fig . 7A–B ) . When B6 was added to permeabilized T . b . brucei respiring on G3P , the oxygen consumption rate increased . This happened both in the absence and presence of SHAM ( Fig . 7A–C ) , showing that B6 is not just relieving SHAM inhibition . The oxygen consumption in the presence of SHAM plus B6 was the same irrespective of whether B6 was added after or together with SHAM ( compare Fig . 7B and D ) . Clearly , B6 resulted in non-TAO mediated oxygen consumption . This finding might be explained by taking into account the capability of B6 to react with molecular oxygen , likely as a consequence of ROS production . Addition of an extra 2–4 mM SHAM did not relieve this B6-mediated oxygen consumption ( Fig . 7B–D ) . To prove the capability of B6 to generate ROS , we measured the production of radicals during respiration in a mitochondrial fraction of T . b . brucei ( bloodstream form ) in the presence and absence of B6 ( Fig . 8 ) . Indeed , in this experiment , production of ROS , due to the reactivity of B6 when it was reduced by the mitochondrial FAD-dependent glycerol-3-phosphate dehydrogenase , was detected . After 40 min , B6 , in the presence of G3P , was able to generate radicals in a significant way compared to the control experiments . B6 caused a considerable production of ROS during the respiration of T . b . brucei bloodstream-form mitochondria , but not in bovine heart submitochondrial particles ( Table S2 and Fig . S1 ) . To provide support for the notion that the trypanocidal effect of B6 could in part be attributed to the generation of toxic ROS , we made use of an available bloodstream form T . b . brucei cell line in which glucose-6-phosphate dehydrogenase ( G6PDH ) expression can be knocked down by RNA interference ( RNAi ) [56] . Previously , it has been shown that this cell line grows equally well as wild-type cells under normal ( reducing ) growth conditions , but it is highly susceptible to ROS when the G6PDH expression is partially knocked down , an effect attributed to decreased NADPH production [57] . Indeed , administration of B6 to cells induced for decreased expression of G6PDH by RNAi showed that these cells are much more susceptible . The ED50 for the induced RNAi cell line is 0 . 25 µM , whereas for the non-induced RNAi cell line , it was 10 . 8 µM . This clearly showed that B6 could increase mitochondrial ROS production as a further molecular mechanism at the basis of its trypanocidal profile .
Neglected tropical diseases are a huge health emergency , which requires remarkable efforts in the search for novel drug candidates to combat them . In fact , the drug discovery pipeline in the field of neglected tropical infectious diseases is almost dry , and fast technologies should be exploited to identify novel classes of potential drug candidates . A possible integrated strategy could be the use of parallel synthesis to generate libraries of small organic molecules combined with fast phenotypic assays that allow testing hundreds of compounds in a reasonable amount of time . This strategy could rapidly provide new hit candidates that can be further progressed to the hit-to-lead and lead optimization steps of the drug discovery process . Thus , it is currently considered equally productive as the target-based approach [58] and it also show higher strengths when searching for multitarget ligands [13] . However , to rationally carry out the two latter steps , information on the molecular target ( s ) , and possibly on the hit-target binding mode can be of paramount importance , as any modification on the hit scaffold can be rationally guided by computational and biophysical/biochemical methods . Therefore , once a hit compound has been identified by means of cell-based experiments , it is fundamental to try identifying the potential target ( s ) by means of bioanalytical approaches . In this respect , chemical proteomics has emerged as a promising method to fish out targets from cell lysates using affinity chromatography [26] . Indeed , chemical proteomics has been shown to be highly suitable in complementing cell-based experiments , and to provide fundamental information for accelerating , in a rational manner , the progress of new classes of compounds through the drug discovery process [59]–[61] . It should be noted that such an approach can be particularly suited for compounds that form covalent adducts or bind tightly to protein targets . In addition , non-protein targets ( e . g . radical oxygen production , DNA , etc . ) can be missed by using this approach . Here , we have reported on the application of chemical proteomics techniques aiming at the identification of the potential molecular targets of a new class of naphthoquinone derivatives . They have recently been characterized , by means of cell-based experiments , to act as trypanocidals , but the molecular target ( s ) of this class of compounds remained undisclosed . We have here identified two potential targets ( TbGAPDH and TbGK ) of B6 , a representative member of this class of compounds . Subsequent biochemical assays have clearly demonstrated that B6 is able to inhibit both targets with IC50 values in the micromolar range . However , mathematical modeling has shown that GK inhibition contributed to B6 trypanocidal activity only under low oxygen conditions . In fact , this enzyme under most physiological conditions does not play an essential role in the trypanosome's metabolism , and is thus considered as a sub-optimal drug target , which may become important when inhibited in conjunction with other enzymes ( i . e . TAO ) [62] . Conversely , GAPDH is a vital parasitic enzyme and a well-validated molecular target for antiparasitic drug discovery [62] . To fully account for the nanomolar profile of B6 as assessed by phenotypic cell-based assay , other mechanisms had to be evoked . In light of this , and based on a vast literature reporting that quinones and naphthoquinones can interact with the mitochondrial respiratory chain [55] , [63]–[65] , we have performed further experiments using trypanosomal isolated mitochondria and permeabilized cells . In this way , we could demonstrate that B6 was also able to interfere with the respiratory chain by generating ROS , and supporting the likelihood that B6 interacts with additional targets located in T . brucei's mitochondrion . We also observed an ED50 of B6 for bloodstream-form T . b . brucei strain Lister 427 of 24 . 8 µM , whereas the ED50 of procyclics of the same strain grown with glucose as free-energy source is 0 . 25 µM . The fact that glucose-grown procyclic trypanosomes are 100-fold more susceptible than glucose-grown bloodstream-form cells strongly suggests that in procyclic trypanosomes inhibition of the glycosomal GAPDH is not the dominant cause of death . The nature of the dominant target of B6 in the procyclics , other than GAPDH , remains to be determined . An alternative possibility is that the bloodstream-form , whose natural habitat is the human body , expresses a more elaborate system to deal with ROS generated by the host than the procyclics that live in the insect's midgut . Consequently , if B6 generates important ROS in cells of both life-cycle stages , the cultured bloodstream-forms may also be better equipped to deal with it than the cultured procyclic cells . In addition , ROS production might be higher in procyclics , as respiratory chain complexes I and III , which are well-known sites of ROS production , are not expressed in the bloodstream forms of T . brucei . As for GK , its inhibition likely only marginally contributed to the final trypanocidal activity of our compound . However , glycerol may be an additional substrate in the blood and , although of much less use than glucose ( lower concentration , and lower ATP yield ) , glycerol consumption may relieve the effect of glycolysis inhibition . In this respect , inhibition of GK by B6 may have relevance in vivo as it might prevent temporal rescue of the parasite by glycerol utilization or under transient anaerobic condition . In conclusion , naphtoquinones like B6 can be considered a promising class of natural-like multitarget compounds against T . brucei , which warrants further studies to definitely elucidate its multitarget profile against parasitic protists . | The multitarget approach can represent a promising strategy for the discovery of innovative drug candidates against neglected tropical diseases . However , multitarget drug discovery can be very demanding , because of the highly time-consuming step related to the fine balancing of the biological activities against selected targets . An innovative workflow for discovering multitarget drugs can be envisioned: i ) design and synthesis of natural-like compounds; ii ) test them using phenotypic cell-based assays; iii ) fishing potential targets by means of chemical proteomics . This workflow might rapidly provide new hit candidates that can be further progressed to the hit-to-lead and lead optimization steps of the drug discovery process . The two latter steps can benefit from information on the molecular target ( s ) , which may be identified by chemical proteomics . Herein , we report on the elucidation of the mode of action of a new series of anti-trypanosomal naphthoquinone compounds , previously tested using cell-based assays , by means of chemical proteomics , classical biochemistry , molecular and system biology . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"biochemistry",
"medicinal",
"chemistry",
"chemistry",
"biology",
"computational",
"biology"
] | 2013 | Naphthoquinone Derivatives Exert Their Antitrypanosomal Activity via a Multi-Target Mechanism |
Dengue control and prevention rely heavily on insecticide-based interventions . However , insecticide resistance in the dengue vector Aedes aegypti , threatens the continued effectiveness of these tools . The molecular basis of the resistance remains uncharacterised in many endemic countries including Malaysia , preventing the design of evidence-based resistance management . Here , we investigated the underlying molecular basis of multiple insecticide resistance in Ae . aegypti populations across Malaysia detecting the major genes driving the metabolic resistance . Genome-wide microarray-based transcription analysis was carried out to detect the genes associated with metabolic resistance in these populations . Comparisons of the susceptible New Orleans strain to three non-exposed multiple insecticide resistant field strains; Penang , Kuala Lumpur and Kota Bharu detected 2605 , 1480 and 425 differentially expressed transcripts respectively ( fold-change>2 and p-value ≤ 0 . 05 ) . 204 genes were commonly over-expressed with monooxygenase P450 genes ( CYP9J27 , CYP6CB1 , CYP9J26 and CYP9M4 ) consistently the most up-regulated detoxification genes in all populations , indicating that they possibly play an important role in the resistance . In addition , glutathione S-transferases , carboxylesterases and other gene families commonly associated with insecticide resistance were also over-expressed . Gene Ontology ( GO ) enrichment analysis indicated an over-representation of GO terms linked to resistance such as monooxygenases , carboxylesterases , glutathione S-transferases and heme-binding . Polymorphism analysis of CYP9J27 sequences revealed a high level of polymorphism ( except in Joho Bharu ) , suggesting a limited directional selection on this gene . In silico analysis of CYP9J27 activity through modelling and docking simulations suggested that this gene is involved in the multiple resistance in Malaysian populations as it is predicted to metabolise pyrethroids , DDT and bendiocarb . The predominant over-expression of cytochrome P450s suggests that synergist-based ( PBO ) control tools could be utilised to improve control of this major dengue vector across Malaysia .
The mosquito Aedes aegypti Linnaeus , 1762 ( Diptera: Culicidae ) is the most important vector of dengue , Zika , yellow fever [1 , 2] and chikungunya [3 , 4] viruses to human throughout the tropical world . This domestic mosquito usually bites during daylight , feeding predominantly on humans , mating and resting indoors and breeding in man-made containers in and around human dwellings [5] . Control of Ae . aegypti relies on reducing breeding sites and insecticide-based interventions such as treatment of breeding sites , Ultra Low Volume ( ULV ) space sprays , fogging and thermal spraying [6] . Unfortunately , many vector control programs are threatened by the development of insecticide resistance in Ae . aegypti [7 , 8] . Resistance to multiple insecticides such as pyrethroids , dichlorodiphenyltrichloroethane ( DDT ) , bendiocarb and organophosphates has been reported in Ae . aegypti [9–12] . Resistance to pyrethroid insecticides , the primary insecticide class used against adult mosquitoes is particularly worrying in the context of the re-emergence of dengue and other arboviruses worldwide , including Zika virus [13 , 14] . The elucidation of the mechanisms of insecticide resistance may aid in the in the design of suitable resistance management strategies to prolong the effectiveness of the existing insecticide-based control tools . The two main causes of insecticide resistance are alterations in the target sites and increase in the rate of insecticide metabolism [15] . Target site resistance is caused by mutations in target genes such as the voltage gated sodium channel ( VGSC ) which causes knockdown ( kdr ) resistance , mutations in the acetylcholinesterase ( Ace-1 ) gene and GABA receptors [15 , 16] . The most important target site resistance for mosquitoes is kdr as it confers resistance to both pyrethroids and DDT . kdr occurs as a result of a change in the affinity of the insecticides to their binding sites , because of mutations in the sodium channel [17] . Several kdr mutations have been identified in Ae . aegypti , and the association between the V1016G/I and the F1534C mutations and pyrethroid resistance has been established [18–20] . In Malaysia , a recent study revealed that the frequency of the 1534C resistant mutation ranges from 40 to 80% whereas the 1016G mutation is found at around 20 to 39% [10] . A significant correlation was also established between F1534C genotypes and pyrethroid resistance [10] in Malaysian mosquitoes , whereas no significant correlation was found for the V1016G mutation . However , an additive effect to pyrethroid resistance was observed when both 1534C and 1016G were present [10] . Another key resistance mechanism is metabolic resistance through up-regulation of detoxification genes . The three main enzymes families responsible for insecticide resistance in mosquitoes are the monooxygenases ( cytochrome P450s ) , glutathione S-transferases ( GSTs ) and carboxylesterases ( COEs ) [21 , 22] . Metabolic resistance can occur as a result of point-mutations affecting protein activity ( e . g . change in binding affinity or an altered substrate specificity ) [23–25] or via mutations in cis/trans regulatory loci of these three enzyme families [15] . In the case of cytochrome P450s elevated expression of several genes from this family has previously been shown to be primarily responsible for resistance towards pyrethroids , carbamates and organophosphates [7 , 15 , 26] . Multiple and widespread resistance to insecticides was recently reported in Ae . aegypti populations across Malaysia [10] . Although the F1534C kdr mutation was shown to play some role in the case of pyrethroids and DDT resistance , PBO synergist assays suggested that metabolic resistance mechanisms play an important role in the resistance patterns . Furthermore , it remains unknown whether differences observed in resistance profiles notably in Kuala Lumpur ( KL ) , where high and multiple resistance was observed , is supported by differences in underlying molecular basis of the resistance . Therefore , to aid the design of suitable resistance management strategies for control of Ae . aegypti in Malaysia , we investigated the molecular basis of multiple insecticide resistance using a microarray-based genome-wide transcriptional analysis of various populations of this species across Malaysia . This study detected key resistance genes and revealed that metabolic resistance is primarily driven by the cytochrome P450 gene family .
An institutional clearance for this study , including the sampling of mosquito , was granted by the Ministry of Health , Malaysia . Ae . aegypti mosquitoes were collected using ovitraps in July and August 2010 in four states ( Fig 1A ) , namely: Penang ( Northwest ) , Kota Bharu ( Northeast ) , Kuala Lumpur ( Central ) and Johor Bharu ( South ) in Malaysia as previously described [10] . Old tyres , flower pots , tree holes and containers that held water were also inspected for larvae . The larvae were reared at the Vector Control Research Unit ( VCRU ) , Universiti Sains Malaysia , in Penang ( Malaysia ) as recently described [10] . Adult Ae . aegypti were induced to lay eggs on filter papers that were later dried and shipped to the Liverpool School of Tropical Medicine ( LSTM ) under the LSTM import license from DEFRA . The egg batches were then allowed to hatch in the insectary in water supplemented with hay infusion solution . Larvae were reared as above and the adults fed with 10% sucrose , and kept at a room temperature of 27 ± 2°C with relative humidity of 70 ± 10% . The bioassays were performed in the same conditions with F1 generation or subsequent F2 generations . Analysis revealed that all four populations were resistant to permethrin and to deltamethrin . The highest resistance levels to both insecticides were observed in Kuala Lumpur with nearly all mosquitoes surviving the 1 h exposure . However , in Kota Bharu , the high permethrin resistance ( 10% mortality ) contrasted with only a moderate resistance to deltamethrin ( 82% mortality ) . These populations were also resistant to DDT with the highest resistance level recorded again in Kuala Lumpur with no mortality after 1 h exposure . Widespread resistance is also observed against the carbamate bendiocarb except in Kota Bharu where 91% mortality was observed in females . Full susceptibility was observed for the organophosphate malathion , except for Kuala Lumpur where a probable resistance is observed with 91% mortality . Similarly , a full susceptibility was observed against dieldrin except in Johru Bharu where a moderate resistance is observed with 88% mortality in females [10] . These samples have been used for the transcriptional analysis . A series of microarray experiments were conducted to identify the genes potentially associated with the metabolic resistance in the Ae . aegypti populations across Malaysia . The genome-wide transcription profiling was carried out in comparison with a susceptible strain ( New Orleans ) to investigate the differential expression profiles of the populations . Microarray hybridizations were done using the 8 x 15k Agilent Ae . aegypti chip containing eight replicated arrays of 60-mers oligo-probes representing more than 14 , 320 different Ae . aegypti transcripts from AaegL1 . 2 Vectorbase annotation and several control ( ArrayExpress accession number A-MEXP-1966 ) . This 8 x 15k microarray enables a high coverage across the whole genome [27] and at the same time reduces cost . It also increases throughput compared to the previous Aedes Detox chip which includes only 204 detoxification genes probes [26] . Total RNA was extracted from 3 replicates of pools of 10 adult unexposed mosquitoes from all four strains . Only samples from Penang ( PG ) , Kuala Lumpur ( KL ) and Kota Bharu ( KB ) were used in the microarray . Samples from Johor Bharu ( JB ) were omitted from the microarray experiment due to the population having similar resistance profile to PG ( but at a lower level ) and therefore possibly sharing similar resistance mechanisms [10] . However , samples from JB were used in the validation of the candidate genes through qRT-PCR . RNA from the susceptible New Orleans ( NO ) strain was also used . Quality and quantity of RNA were assessed using the Nanodrop ND-1000 ( Thermo Scientific , Delaware , USA ) and the Agilent 2100 Bioanalyzer ( Agilent Technologies , California , USA ) . RNA ( 100ng ) sample from each sample were amplified and labelled using the two-colour low input Quick Amp Labelling kit ( Agilent Technologies ) . Labelled cRNAs were hybridized to the arrays for 17h at 65°C according to the manufacturer’s protocol . Three different comparisons were made: NO susceptible lab strain vs . PG , NO vs . KL and NO vs . KB ( non-exposed vs susceptible ) ( S1 Fig ) . A total of five replicates ( including three biological replicates and two dye swaps ) were performed for each comparison as previously successfully done in other species to detect the resistance genes [28 , 29] . Microarray data were analysed using the Genespring GX 12 . 0 software ( Agilent Technologies , US ) . Mean expression ratios were assessed using a t-test against zero with a multiple testing correction ( Benjamini-Hochberg false discovery rate ) . Genes showing both t-test p-values less than 0 . 05 and a fold change ( FC ) value greater than 2 were considered significantly differentially transcribed between the two samples compared . Genes or entities that were considered as significantly differentially expressed were used for Gene Ontology ( GO ) enrichment analysis Blast2GO software ( BioBam Bioinformatics S . L . , Valencia , Spain ) . Descriptions and GO-terms of transcript-IDs were generated from Blast2GO extracted from VectorBase . GO term enrichment analysis was performed on the significantly up-regulated genes ( 72% of transcripts present on microarray have GO-terms ) using Blast2GO software with Fisher’s exact test and false discovery rate ( FDR ) < 0 . 05 [30] . Some of the most significantly differentially expressed genes from the microarray analysis were selected for qRT-PCR validation . Firstly , control samples not exposed to insecticides from all four locations ( including the JB sample that was not used in the microarray to also confirm their potential role in this location ) were compared to the susceptible NO strain to assess the geographical variation of these key genes . Secondly , samples alive after exposure to various insecticides were also compared to the control non exposed mosquitoes and to the NO strain to detect possible induction of these genes or whether their over-expression was more associated to a specific insecticide than others . This was done only in PG as this population is moderately resistant to all insecticides allowing a comparison to the control sample . Total RNA from 3 replicates of each samples were used for the qRT-PCR . Primers used are listed in S1 Table . Standard curves for each gene were generated using a serial dilution of cDNA to assess PCR efficiency and quantitative differences between samples . qRT-PCR amplification was performed as described previously [28 , 29] . The relative expression level and fold change ( FC ) of each target gene in field samples relative to the susceptible NO ( S ) were calculated according to the 2-ΔΔCT method incorporating the PCR efficiency [31] after normalization with the housekeeping genes ribosomal protein S7 ( RSP7; AAEL009496-RA ) and Tubulin ( AAEL009496-RA ) . Patterns of polymorphism for CYP9J27 were explored across Malaysian Ae . aegypti populations to detect possible correlation with resistance profile using the permethrin susceptible ( NO ) and unexposed Ae . aegypti mosquitoes from the four sites in Malaysia: JB , KL , KB and PG . Full-length coding region of CYP9J27 was amplified from cDNA using the same cDNA synthesized for qRT-PCR with the Phusion High-Fidelity DNA Polymerase ( Thermo Scientific ) , cloned into the pJET1 . 2/blunt cloning vector ( Thermo Scientific ) , and sequenced as described previously [24] . Primers used are listed in S2 Table . Polymorphic positions were detected through manual analysis of sequence traces using BioEdit version 7 . 2 . 1 and as sequence differences in multiple alignments using ClustalW [32] . Basic sequence statistics , including the number of haplotypes ( h ) , the number of polymorphism sites ( S ) , haplotype diversity ( Hd ) and nucleotide diversity ( π ) , were computed with DnaSP 5 . 10 [33] . The statistical tests of Tajima [34] , Fu and Li [35] was used with DnaSP to test non-neutral evolution and deviation from mutation-drift equilibrium . Different haplotypes were compared by constructing a maximum likelihood phylogenetic tree and polymorphic positions of amino acid sequences were generated using MEGA 6 . 0 [36] . To predict the ability of the CYP9J27 to bind the various insecticides homology models of the P450 were created using query amino acid sequences from the various study locations ( KL , PG , JB , KB ) , as well as the sequence from susceptible strain , NO . The 3D models of the P450s were created using the standalone tool EasyModeller [37] . CYP3A4 ( PDB: 1TQN ) [38] was used as a template with sequence identity of 31% for all the five CYP9J27 amino acid sequences . Virtual datasets of ligand insecticides: 1R-cis permethrin ( ZINC01850374 ) , deltamethrin ( ZINC01997854 ) , DDT ( ZINC01530011 ) and bendiocarb ( ZINC02015426 ) were retrieved from the library in ZINC12 database ( https://zinc . docking . org/ ) [39] . Docking simulations were carried out using the Blind Docking Web Server ( http://bio-hpc . ucam . edu/webBD/index . php/entry ) . For each ligand , 30 binding poses were generated and sorted according to the binding energy and conformation in the protein’s active site . Figures were prepared using the PyMOL 1 . 7 [40] .
The microarrays were used to perform a genome-wide transcription analysis between the susceptible strain ( New Orleans , NO ) and non-exposed field populations from Penang ( PG ) , Kuala Lumpur ( KL ) and Kota Bharu ( KB ) . The number of differentially transcribed genes after analysis is presented in the Venn diagram ( Fig 1B ) . Penang had the most number of differentially transcribed probes with 2605 probes , followed by Kuala Lumpur with 1480 gene probes , and Kota Bharu with 425 gene probes . The number of commonly up-regulated probes in all population is 204 while 41 probes were down-regulated . From the list of the genes up-regulated in a single location only , PG had the most number of up-regulated detoxification genes exclusively expressed in this location with a predominance of cytochrome P450 among which is the P450 CYP6P12 ( FC value 6 . 50 ) . The ortholog of this gene in Ae . albopictus was recently shown to be the main pyrethroid resistance gene in this species in KL [42] . CYP6P12 is also the ortholog of CYP6P4 in the malaria vectors An . gambiae , An . arabiensis and An . funestus . Other cytochrome P450 genes are the CYP6BB2 ( a different probe than in the commonly up-regulated in PG and KB ) , CYP6N6 , CYP325X1 , CYP4D37 , CYP9J10 , CYP6Z8 and others . Other genes belong to GSTs ( GSTS1 ) , carboxylesterase ( transcript AAEL000905-RA and CCEAE1B ) and the ABC transporters ( Table 1 ) . The set of probes up-regulated only in KL is mainly made of cytochrome P450s among which the CYP9J28 gene exhibited the highest FC value of 10 . 80 . This gene has previously been shown to confer pyrethroid resistance in Ae . aegypti [46] . Other cytochrome P450s over-expressed only in KL are CYP18A1 and CYP9M6 ( Table 1 ) . The genes up-regulated in KB only are a cytochrome P450 ( CYP9M7 ) , an ABC transporter and a UDP glucuronosyl transferase ( Table 1 ) . Seven candidate genes commonly up-regulated in all three locations were chosen to validate the microarray expression pattern . These genes were trypsin ( AAEL013623-RA ) , multisynthetase complex , CYP6CB1 , CYP9J26 ( AAEL014609-RA ) , CYP9J27-607 ( AAEL014607-RA ) , CYP9M4 and CYP9J27-616 ( AAEL014616-RA ) . The most over-expressed P450 for microarray in all 3 locations , CYP6CB1 was not significantly over-expressed from the qRT-PCR results in all four locations including JB ( Fig 1C ) . However , the over-expression of other five P450 genes was confirmed whereas the over-expression of the multisynthetase gene was not confirmed . The JB sample consistently exhibited a lower expression level for all these genes in comparison to the other 3 locations . This difference could be associated with the lower resistance level to pyrethroids and DDT in JB . Overall , for most of the genes , the highest over-expression was observed for the KL population which correlates with the high resistance level to pyrethroids and DDT observed in KL from bioassays [10] . The expression profile of the five significantly over-expressed genes from qRT-PCR was further assessed for various samples from mosquitoes alive after exposure to different insecticides . From this analysis , CYP9J27-616 was more over-expressed , although not significantly , in the DDT resistant mosquitoes from PG than the other insecticides ( Fig 1D ) . CYP9M4 was significantly more over-expressed in the PG bendiocarb resistant sample than other mosquito samples . No significant difference was observed between samples for CYP9J26 , CYP9J27-607 and for the trypsin gene . GO enrichment analysis was used to identify particular Gene Ontology ( GO ) terms that were over represented in the data set of transcripts up-regulated in all three resistant populations and in single populations . When comparing the commonly up-regulated genes in all three populations at p = 0 . 01 , a few GO terms that relates to detoxification was observed . These included NADPH-hemoprotein reductase activity , ATP binding and others ( S2 Fig ) . When observing the GO terms of single locations either in PG ( S3 Fig ) , KB ( S4 Fig ) or KL ( S5 Fig ) , interesting terms such as ATP binding , heme-binding and monooxygenase activity were over-represented possibly associated with the multiple resistance to insecticides detected in these locations . In order to functionally investigate the potential involvement of the commonly over-expressed CYP9J27 gene in conferring resistance to insecticides in Ae . aegypti , further attention was paid on this gene since particularly as the role of other P450s has already been assessed [46] . Full-length cDNA sequences ( 1611 bp ) of CYP9J27-616 ( hereafter CYP9J27 ) , one of the genes commonly over-expressed all the three populations from microarray were successfully generated for 23 clones from four locations across Malaysia and the susceptible NO strain allowing to assess their polymorphism ( GenBank Accesion Number: KX394421-KX394443 ) . Overall , analysis of the genetic variability revealed 54 polymorphic sites ( s ) , 10 haplotypes ( h ) , and 15 amino acid changes in total , with the lowest polymorphism observed in JB ( Table 2 , Fig 2A ) . Phylogenetic tree constructed using maximum likelihood confirm the relatively high genetic variability of this gene in each location except in JB ( Fig 2B ) . Haplotype diversity is high ( Hd>0 . 73 ) in most samples except JB suggesting that little directional pressure is acting in this population . However , the lack of diversity in JB should be further investigated . The difference between JB and other populations is further highlighted on the neighbour-joining tree of the genetic distances between various populations , with JB exhibiting high level of genetic differentiation based on KST estimates whereas KL and KB cluster together . When all the sequences were analysed as a unique sample , the Tajima D , was positive ( D = 2 . 0675 ) and statistically significant ( P<0 . 05 ) ( Table 2 ) . In the other hand , when the sequences were analysed according to the origin of each sample this statistics was negative but not statistically significant ( Table 2 ) . Negative values for these indexes may indicate an excess of rare polymorphisms in a population and suggest either population expansion or background selection [35] . However , further analysis with more sequences is needed to test for a signature of selection on this gene . For each of the four locations as well the NO , CYP9J27 model with the highest ERRAT score ( which is based on the patterns of non-bonded interaction ) [47] was chosen for molecular docking . For the type I pyrethroid permethrin , the KL model exhibited the lowest binding energy ( 10 . 1 kcal/mol ) which was comparable to values obtained from JB and KB ( -9 . 9 kcal/mol respectively ) ; but lower than observed with PG ( -9 . 6 kcal/mol ) ( S6 Table ) . NO model exhibited the highest binding energy of -9 . 1 kcal/mol , however this binding energy predicts good binding of the permethrin to the model from the susceptible strain , even though lower than obtained from the other models . However , with the exception of KL model in which permethrin binds with the benzyl ring perpendicular above the heme plane ( 2 spot of benzyl ring located 4 . 1Å from the heme iron ( Fig 3A ) , predicting ring hydroxylation to generate 2-hydroxypermethrin ) , the insecticide docked in all the other models with the ester oxygen projecting toward the heme iron , susceptible to ester hydrolysis ( Fig 3B–3D; S9A Fig ) . As with permethrin , docking of type II pyrethroid deltamethrin to the active site of KL CYP9J27 model produced the lowest binding energy ( -10 . 3 kcal/mol ) , lower than observed with the other models , with NO model producing the highest binding free energy ( -9 . 3 kcal/mol ) ( S6 Table ) . Also , as with permethrin , deltamethrin docked in the active site of KL model with the phenoxy ring above the heme iron and the 4´ spot located within 4 . 2Å of the heme iron ( S6A Fig ) . In this posture ring hydroxylation to generate 4´-hydroxydeltamethrin is predicted . For JB and the NO models the α-cyano group docked above the heme iron within 2 . 4Å of the heme iron ( S6B and S6D Fig ) , while for PG it is the ester oxygen which projects toward the heme iron at a distance of 2 . 3Å ( S6C Fig ) . For KB , it’s the benzyl ring which docked within 2 . 9Å from the heme iron with possibility of ring hydroxylation ( S9B Fig ) . With the exception of NO model , DDT docked in the models of KL , JB , PG and KB with low binding energy ( S6 Table ) and productively with carbon atom of the trichloromethyl group positioned within a distance of 5 . 4Å , 4 . 3Å and 4 . 8Å respectively for KL , JB and PG ( S7A–S7C Fig ) . The same pose was obtained with KB model with the carbon atom of the trichloromethyl group at a distance of 5 . 6Å from the heme iron ( S9C Fig ) . In this posture reductive dechlorination to generate DDE is predicted . In contrast , DDT docked unproductively in the active site of NO with the chlorine atoms projecting toward the heme iron at a distance of 3 . 6Å ( S7D Fig ) . Generally , the free energy of DDT binding in the models from the resistant populations is higher than obtained with the pyrethroids but low enough to warrant good binding , with the exception of NO model to which the binding energy of DDT is very high ( -5 . 3 kcal/mol ) . In KL , JB , PG and KB models , bendiocarb bind productively with the carbamic acid ester group oriented toward heme at a reasonable distance of 3 . 4Å , 3 . 9Å , 3 . 7Å and 3 . 9Å , respectively ( S8A–S8C and S9D Figs ) . In this posture ester hydrolysis to generate benzodioxol-4-ol and carbamic acid is predicted . Free binding energy for all the models is lower than obtained from docking with pyrethroids and DDT; however , NO has the highest binding energy of all ( -5 . 1 kcal/mol ) indicative of low affinity compared with the other models . Not surprisingly , bendiocarb docked in the active site of NO model away ( 13 . 2Å ) from heme catalysis centre for any meaningful interaction to take place .
Elucidating the molecular basis of insecticide resistance in the dengue vector Ae . aegypti is an important step for the design of suitable resistance management strategies to control this vector . In this study genome-wide transcriptional analyses carried out using microarray showed that metabolic resistance plays an important role in conferring resistance to insecticides in Ae . aegypti across Malaysia . This further supports the synergist assay result with PBO showing a recovery of susceptibility notably for pyrethroids [10] . The role of metabolic resistance was supported by the over-expression of many genes belonging to detoxification families in PG , KL and KB when comparing them to the susceptible NO strain . The most prominent detoxification gene family established in this Aedes populations were the cytochrome P450 genes which were the only detoxification family discovered except for one unique carboxylesterase commonly over-expressed in the three locations . The most over-expressed cytochrome P450 is the CYP6CB1 which has also been reported in a strain from Isla Mujeres in Mexico [46] . Unfortunately , the microarray over-expression of CYP6CB1 was not supported by the qRT-PCR validation for all four populations tested . This discrepancy between the two methods could be caused by differences in the sequences of the microarray probes and the qRT-PCR primers . Recent functional analysis has shown that this CYP6CB1 could not metabolise pyrethroids [46] although it could metabolise other insecticides . CYP9J26 ( AAEL014609 ) gene was among the top up-regulated detoxification gene which has also been reported in Cuba , Thailand and Grand Cayman [30] and has also been functionally validated to confer pyrethroid resistance [46] . Overall , several P450 genes belonging to the CYP9 family were over-expressed in the control- versus susceptible across Malaysia including two transcripts of CYP9J27 , CYP9J26 , CYP9J28 , CYP9M6 while only few cytochrome P450s from the CYP6 family ( CYP6P12 , CYP6BB2 ) were over-expressed and usually at lower fold change . This is further supported by previous studies worldwide showing that contrary to Anopheles mosquitoes , genes from the CYP9 family play a more important role than those from the CYP6 family in insecticide resistance in Ae . aegypti [7 , 26 , 30 , 48 , 49] . This was different than the Aedes albopictus transcription analyses performed with samples collected from same locations at the same time as more P450 genes belonging to the CYP6 family were over-expressed in the C-S comparison of Ae . albopictus samples in Malaysia including CYP6N3 , CYP6P12 , CYP6Z6 , CYP6AG6 , while only few cytochrome P450s from the CYP9 family were over-expressed [42] . Interestingly , the top most commonly over-expressed gene was the anionic-trypsin which is found in the midgut of mosquitoes and shown to hydrolyse proteins after blood meals . This serine proteinase is found to be over-expressed in deltamethrin resistant strain of Culex pipiens pallens from China [50] . Few glutathione S-transferases were detected compared to cytochrome P450s despite the very high DDT resistance notably in KL . The PBO synergist assay previously indicated a recovery of susceptibility from 0 to 35% in KL for DDT [10] . The low expression of GSTs notably that of the known DDT metaboliser GSTe2 ( FC = 4 . 5 ) [42 , 45] suggests that knockdown resistance could be responsible for most of the remaining 65% loss of DDT susceptibility . Similar assessment of pyrethroids shows a recovery of susceptibility after PBO assay from 1% to 26% for permethrin and from 0% to 71% for deltamethrin . This suggests that metabolic resistance through P450 up-regulation is more important for deltamethrin than permethrin resistance and kdr playing a more important role for permethrin than deltamethrin . This will be in line with the higher correlation previously observed between permethrin and F1534C genotypes than with deltamethrin [10] . However , the molecular docking predicted CYP9J27 to bind and metabolise pyrethroids and DDT , especially KL model , compared to NO model to which DDT binds unproductively indicating lack of affinity and activity towards this organochlorine insecticide by NO strain . Because PBO assays with bendiocarb also revealed a nearly full recovery of the susceptibility to this insecticide [10] it is likely that some of the cytochrome P450 genes detected in this study are responsible for this resistance although future functional characterization will identify the specific genes . Of course , with the exception of NO model , docking analyses with CYP9J27-616 models predicted productive binding and good affinity towards bendiocarb suggesting the ability to metabolise this carbamate insecticide . The possible role of P450s in carbamate resistance will explain the low expression of carboxylesterase genes observed in this study and suggests an absence of any Ace-1 mutations as previously reported [49] . The higher polymorphism level of CYP9J27-616 gene across Malaysian samples ( apart from JB ) ( Table 2 ) suggests of little directional selection pressure favouring a specific SNP or amino acid change is acting on this gene in Malaysia despite it consistent over-expression . This suggests that CYP9J27 potential role in the resistance if confirmed would be through a mechanism involving genetic variation in the regulatory regions such as promoter beside potential variation in the coding sequence . This is similar to cases of some P450 genes in other mosquito species for which despite a high over-expression in resistant individuals , no significant association has been observed between polymorphism level and resistance phenotype , such as CYP6M7 in the malaria vector An . funestus [24] . Further analysis of these regulatory regions could help detect the specific genomic changes driving resistance through CYP9J27 . Nevertheless , the lowest binding affinity from docking exhibited by the CYP9J27 allele from the susceptible NO strain compared to those from Malaysian could suggest that allelic variation could also be playing a role in the resistance observed as recently demonstrated in the African malaria vector An . funestus [25] . Interestingly , negative values of neutrality tests Tajima’s , and Fu and Li’s observed for CYP9J27 suggest either recent demographic expansion or background selection [35] on this gene across Ae . aegypti Malaysian population .
This study revealed that the cytochrome P450 genes are associated in insecticide resistance in Ae . aegypti populations across Malaysia . However , further functional characterization work using either transgenic expression in Drosophila flies or recombinant enzymes expression in Escherichia coli coupled with metabolic assay has to be done to confirm the exact contribution of these candidate genes in the resistance profile . An alternative to pyrethroid is the organophosphate , malathion since all of the populations are susceptible to this insecticide [10] . This and other proper resistance management strategies should be adopted to reverse the spread and evolution of this resistance problem in Malaysia before it compromises control programmes . | Aedes aegypti is the major vector of emerging or re-emerging arboviruses such as dengue , chikungunya and Zika viruses worldwide . As there is still no vaccine or specific treatment for these diseases , their prevention relies mainly on insecticide-based interventions . Unfortunately , vector control programs are facing operational challenges with mosquitoes becoming resistant to commonly used insecticides in several regions across the world . To facilitate the design of suitable resistance management strategies for control of this vector , we investigated the molecular basis of insecticide resistance across four Malaysian Ae . aegypti populations . Transcriptional analysis revealed that 204 genes were commonly over-expressed with cytochrome P450 genes ( CYP9J27 , CYP6CB1 , CYP9J26 and CYP9M4 ) consistently found as the most up-regulated detoxification genes in all populations indicating that they likely play an important role in the observed resistance . Quantitative real-time polymerase chain reaction ( qRT-PCR ) validated the over-expression of all cytochrome P450s except for CYP6CB1 . Polymorphism analysis of CYP9J27 sequences revealed a high polymorphism ( except in Joho Bharu ) , suggesting a lack of directional selection on this gene . The predominant expression of cytochrome P450s suggests that synergist-based control tools could be used to improve control of this major dengue vector across Malaysia . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion",
"Conclusion"
] | [
"heme",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"animals",
"toxicology",
"ddt",
"genome",
"analysis",
"bioassays",
"and",
"physiological",
"analysis",
"insect",
"vectors",
"research",
"and",
"analysis",
"methods",
"agrochemicals",
"aedes",
"aegypti",
"proteins",
"epidemiology",
"gene",
"ontologies",
"disease",
"vectors",
"insects",
"agriculture",
"arthropoda",
"insecticides",
"microarrays",
"biochemistry",
"mosquitoes",
"detoxification",
"post-translational",
"modification",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"computational",
"biology",
"organisms"
] | 2017 | Pyrethroid Resistance in Malaysian Populations of Dengue Vector Aedes aegypti Is Mediated by CYP9 Family of Cytochrome P450 Genes |
The threat of the new pandemic influenza A ( H1N1 ) pdm09 imposed a heavy burden on the public health system in Finland in 2009-2010 . An extensive vaccination campaign was set up in the middle of the first pandemic season . However , the true number of infected individuals remains uncertain as the surveillance missed a large portion of mild infections . We constructed a transmission model to simulate the spread of influenza in the Finnish population . We used the model to analyse the two first years ( 2009-2011 ) of A ( H1N1 ) pdm09 in Finland . Using data from the national surveillance of influenza and data on close person-to-person ( social ) contacts in the population , we estimated that 6% ( 90% credible interval 5 . 1 – 6 . 7% ) of the population was infected with A ( H1N1 ) pdm09 in the first pandemic season ( 2009/2010 ) and an additional 3% ( 2 . 5 – 3 . 5% ) in the second season ( 2010/2011 ) . Vaccination had a substantial impact in mitigating the second season . The dynamic approach allowed us to discover how the proportion of detected cases changed over the course of the epidemic . The role of time-varying reproduction number , capturing the effects of weather and changes in behaviour , was important in shaping the epidemic .
The threat of the pandemic influenza A strain , A ( H1N1 ) pdm09 ( ‘swine flu’ ) , imposed a huge burden on the public health system in Finland in 2009 [1] . The first A ( H1N1 ) pdm09 season was part of the global pandemic and occurred from September 2009 through January 2010 with a major outbreak in November 2009 . To mitigate the epidemic , a national vaccination campaign was started in October 2009 , and by February 2010 approximately half of the Finnish population had been vaccinated against A ( H1N1 ) pdm09 . The second epidemic season occurred a year later from November 2010 through April 2011 . Only sporadic cases were observed before the first epidemic season and between the two seasons . It is well known that that laboratory-based surveillance of influenza misses the vast majority of infections that occur in the population . Underreporting follows from asymptomatic or non-diagnosed infection or incomplete reporting of influenza cases in primary and secondary health care . More severe cases are diagnosed and reported with a higher probability . This was true also for A ( H1N1 ) pdm09 , although special efforts were taken to record cases especially during the early phases of the first season . Bayesian methodology ( evidence synthesis ) has been used to analyse influenza outbreaks in the presence of underreporting and mixed data sources . In general , the underlying epidemiological models can be classified as static or dynamic . In static models , cases are typically aggregated by season and the unknown true incidence is estimated as an attack rate ( probability of becoming infected during the season ) [2–5] . In dynamic models , the process of spread of the infection via transmission is modelled explicitly [6] . The static approach is simpler and requires less computational resources while the dynamic model enables one to answer more complex questions . Based on a static model , we previously estimated that only 4% of the Finnish population were infected with A ( H1N1 ) pdm09 over the season 2009/2010 and an additional 1% during the 2010/2011 season [2] . The most affected age groups were children and teenagers with attack rates up to 10-12% . The attack rates were much lower in the second season , which was likely due to the relatively high immunity due to natural infection or vaccination in the most influential age groups . In particular , 74-81% of children aged less than 15 years had been vaccinated against A ( H1N1 ) pdm09 before the second season . However , a static model cannot address the impact of herd immunity induced by vaccination . To properly address the role of vaccination in mitigating the first-season epidemic and lowering the transmission potential before the second season , a more dynamic ( i . e . transmission ) model is needed . A dynamic model can also address questions about which age groups played the most important role in transmission or why there was a second season despite the fact the influenza strain did not evolve considerably between the seasons to escape population immunity [7] . The effect of time-varying conditions due to weather or public response to the outbreak can also be inferred using a dynamic model [8] . In this study , we built a dynamic probabilistic model of influenza transmission and disease . The model accounts for transmission of influenza in the population , the impact of vaccination , outcomes with varying severity and imperfect detection of infection . We calibrate the model to data on A ( H1N1 ) pdm09 cases and estimate the true incidence of A ( H1N1 ) pdm09 of the first two A ( H1N1 ) pdm09 seasons in Finland .
In all datasets used in this study , information about individuals was aggregated into 16 age groups: 0-4 , 5-9 , … , 70-74 , 75+ years of age . Fig 1 presents the data on registered A ( H1N1 ) pdm09 cases and the coverage of vaccination in Finland 2009-2011 . The population sizes were obtained from Statistic Finland ( www . stat . fi ) . We built a discrete-time dynamical model of influenza transmission and disease in the Finnish population . The time step was one week , corresponding to the resolution in the data . A period of 113 weeks was modelled from week 15/2009 ( one month before the first A ( H1N1 ) pdm09 cases of were registered in Finland ) through week 22/2011 ( after the end of the second season ) . Within the modelled period , two subperiods are referred to as the first epidemic season ( weeks 37/2009 through 1/2010 ) and the second epidemic season ( weeks 46/2010 through 17/2011 ) . The prior distributions of the model parameters are presented in Table 1 . All parameters except the age-dependent susceptibility p and transmission random effect wt were given informative priors . The severity parameters s ( sev/inf ) and s ( IC/sev ) were centred around 1% and 10% , respectively . The detection probabilities d t ( mild ) and d ( hosp ) were centred around 1% and 75% , respectively . These priors were consistent with the ones used in our earlier analysis of the same data [2] . The inflow of infection q was assumed to be extremely small . We estimated the joint posterior distribution of the model parameters and latent variables using Markov chain Monte Carlo computation ( MCMC ) with particle Gibbs sampler step [17] . In addition , we applied exact approximate MCMC [18] targeting a smoothed marginal posterior of the model parameters , p ( parameters|data ) 1/25 and p ( parameters|data ) 1/5 , to ensure that the peak area of the target posterior is unimodal and well-behaving . Details are provided in S3 Appendix . Posterior predictive checks were used to to explore how well the model describes the observed data . Sensitivity analysis was performed by comparing the posterior modes ( i . e . maximum a posteriori estimates ) under different prior settings . Details are provided in S5 Appendix .
The estimated true numbers of A ( H1N1 ) pdm09 infection are shown in Fig 5A and 5B . Fig 6A presents the attack rates , i . e . the numbers of infected per population size . We estimated that 440 000 – 550 000 individuals in total ( 8 . 2 – 10 . 4% of the population , posterior mean 500 000 , 9 . 3% ) were infected in Finland during the modelled period . Specifically , the numbers infected were 270 000 – 360 000 ( 5 . 1 – 6 . 7% , posterior mean 320 000 , 5 . 9% ) and 140 000 – 190 000 ( 2 . 5 – 3 . 5% , posterior mean 160 000 , 3 . 0% ) during the first and the second A ( H1N1 ) pdm09 epidemic seasons , respectively . Only a minor portion of infections ( 0 . 3 – 0 . 4% ) occurred outside the two epidemic seasons . In both seasons , the attack rate decreased with age . It was largest in the youngest age group ( 14 – 19% during the first and 5 . 5 – 7 . 6% during the second epidemic season ) and smallest in the oldest ( 5 . 0 – 6 . 6% and 4 . 6 – 6 . 4% ) . Fig 5C presents the cumulative age composition of the infected population per week . The mean age of infection increased with time . Before the peak of the first season , approximately half of all infections occurred among less then 15 years olds . During the second epidemic season only 25% of infections belonged to this age group . The oldest ( 65+ years ) never accounted for a significant portion of the infected population . Fig 7A shows the posterior distribution of susceptibility p ( probability of acquiring infection per contact with an infectious individual ) and inflow q ( probability of acquiring infection from outside the population ) . Susceptibility decreased with age: children aged less than 5 years had a 5-fold greater chance to acquire infection per contact than the oldest individuals . Individuals aged 20-29 years were most likely to acquire infection from outside the population . Fig 7B shows the posterior distributions of the severity parameters s ( sev/inf ) and s ( IC/sev ) . The hospitalization/infection ratio had a V shape , the infection being more severe among the youngest ( s ( sev/inf ) = 0 . 7 – 0 . 9% ) and the oldest ( 1 . 3 – 1 . 7% ) . Children aged 5-14 years had the smallest probability of severe disease per infection ( 0 . 3 – 0 . 4% ) . The IC/hospitalization ratio did not vary much across age groups , almost repeating the prior information . It was smallest among the youngest ( s ( IC/sev ) = 7 – 8% ) and largest ( 8 – 11% ) for those over 30 years . In our model , influenza transmission , including the outbreaks and periods between epidemic seasons , is modulated by a time-varying reproduction number R0 , t = R0 wt ( Fig 5E ) . Before June 2009 , R0 , t rose above 1 allowing for the minor pre-seasonal outbreak . A significant increase in R0 , t in the autumn of 2009 marked the onset of the first epidemic season . After the peak of the first season ( November 2009 ) R0 , t dropped below 1 leading to the end of the first outbreak . By the end of the first epidemic season about 22% of the population were vaccine-protected ( Fig 5D ) , especially in the youngest age groups ( 53% in <20 year olds , 14% in 20-64 olds , and 11% in >65 olds ) . This induced herd immunity in the population , so R0 , t could raise above 1 without causing an outbreak . By the second epidemic season , 41% of the population acquired immunity from vaccination ( posterior mean 52% , 34% , and 47% of individuals aged 0-19 , 20-64 and older than 64 years , respectively ) and 5 – 7% acquired immunity from infection . Around October 2010 R0 , t started gradually increasing , reaching its maximum in November 2010 and then slowly decreased . For the period November 2010—January 2011 , the reproduction number was above 3 . This marked the second epidemic season . The estimates of R0 , t outside the epidemic seasons are uncertain , as scarce data are available for these periods . Overall , the product R0 , t = R0wt was estimated with smaller uncertainty than wt and R0 individually ( see S4 Appendix ) . The largest number of potential infections was produced by individuals from age groups 5-14 years old ( 3 . 5 – 5 . 6 infections ) ( Fig 6B ) . The smallest number was produced by the oldest age group ( 0 . 4 – 0 . 5 infections ) . On average , only few infections per week were introduced from outside the population ( Fig 6C ) . The random effect wt and the detection probability d t ( mild ) increased simultaneously during the early phases of the epidemic seasons . However , for any time ( t ∈ 0 , … , T − 1 ) , the variables wt and d t ( mild ) did not have strong posterior correlation ( see S4 Appendix ) . Fig 6D shows the number of detected cases per the number of infected ( detection ratio; Table 2 ) . We estimated that 2 . 1 – 2 . 7% of all A ( H1N1 ) pdm09 infections were detected ( specifically 2 . 5 – 3 . 3% during the first epidemic season , 1 . 2 – 1 . 6% during the second and 1 . 5 – 2 . 0% outside seasons ) . The detection ratio varied by age with posterior means ranging from 3 . 7% to 1 . 9% . We estimated that the detection probability of the mild cases d t ( mild ) reached its maximum before the peak of the first season and decreased subsequently during the outbreak ( Fig 5F ) . During November 2009 , the observed numbers of mild infections decreased much faster than the observed numbers of hospitalized cases . According to the model , however , the true numbers of mild and severe infections decreased at the same speed and the observed difference was thus explained by the decline in d t ( mild ) . The posterior of the detection probability of hospitalized cases d ( hosp ) followed the prior closely . We measured the impact of the vaccination campaign as the number of cases prevented . To estimate this number , we simulated the incidence of infection , using parameter values sampled from the posterior and assuming that no one was vaccinated ( va , t = 0 ) . According to this analysis ( Fig 8 ) , the second season could have started earlier and caused a larger outbreak , leading to 4-8 times more infections overall ( total attack rate would have been 38 – 78% ) . By contrast , vaccination did not affect the first epidemic season . We also estimated the impact of the vaccination under a scenario where vaccines were distributed in the same amount but independent of age ( va , t = vb , t for all age groups a , b ) . In this situation our model predicts about twice as many infections overall ( total attack rate would have been 15 – 26% ) .
Using a dynamic transmission model , we estimated that 5 . 9% ( 90% credible interval 5 . 1 – 6 . 7% ) of the Finnish population was infected during the first year of the pandemic A ( H1N1 ) pdm09 strain of influenza in 2009/2010 . There was a second season a year later with an attack rate of 3 . 0% ( 2 . 5 – 3 . 5% ) of the population . The vaccination campaign launched in the middle of the first epidemic epidemic season was essential in mitigating the size of the second season , but occurred too late to have an impact on the first season . In both seasons , the proportion of the infected population decreased with age , with the youngest being at least an order of magnitude more likely to be infected than the oldest . The age distribution of the infected population evolved over time . Before the peak of the first season most infected individuals were children aged less than 15 years . According to the social mixing matrix , estimated from the available data , this age group forms the core group of transmission for infections that spread through droplets in close contact . After the end of the first season , as many as 72% of children aged less than 15 years either had had natural infection ( 18% ) or had been vaccinated ( 55% ) so that the importance of this age group in the chain of transmission decreased . During the second epidemic season , the mean age of infected individuals was higher . The posterior mean severity of influenza infection , as measured by the hospitalization/infection ratio ( parameter s ( sev/inf ) ) , was 0 . 7% when averaged over all age groups and had a clear V shape with the youngest and oldest requiring hospitalization more often . The IC/hospitalization ratio ( parameter s ( IC/sev ) ) was driven almost entirely by prior information ( around 8% across all age groups ) . We estimated that only 2 . 4% ( 90% credible interval 2 . 1 – 2 . 7% ) of infections were recorded by surveillance , i . e . there were 40 – 50 unobserved A ( H1N1 ) pdm09 infections for each detected case . The detection probability peaked early during the first epidemic season , with a clear decline towards the end of the season ( Fig 5F ) . This could reflect the public and governmental concerns increasing initially and then declining as the awareness of the relatively mild impact of the novel A ( H1N1 ) pdm09 virus was revealed . The detection ratio in the second epidemic season was smaller than in the first one . Similar patterns in the detection rates occurred in the UK during the first two years of the pandemic [19] . In our model , the spread of infection is modulated by four quantities: susceptibility to infection ( parameter p ) , the pattern of contacts ( contact matrix C ) , the time varying reproduction number ( R0 , t ) and the rate of inflow of infection ( q ) . Susceptibility to infection was estimated to decrease with age , which is likely to reflect higher levels of pre-existing immunity among older individuals [7] . The contact matrix was based on a survey of daily social contacts in Finland [10] . The standard deterministic SIR model assumes that outbreaks only stop by depletion of the pool of susceptibles . In particular , a second season would be impossible unless the virus evolves to escape the prevailing immunity in the population . Although this is known to happen for seasonal influenza [20] , the virus did not change much during the first two years of the pandemic [7] . Vaccination alone cannot explain the observations , as the first season ended in the population with 20% vaccine-induced protection while the second season started with 40% . Therefore , stochasticity in transmission and other mechanisms may be called for . We applied a time-varying reproduction number ( R0 , t ) of influenza transmission , capturing the impact of changing population behaviour or weather conditions as a stochastic process . In particular , cold and dry weather has been suggested as one of the drivers of influenza transmission [21] and the public behaviour may have changed as the epidemic appeared to be relatively mild . We estimated that R0 , t changed markedly with time ( Fig 5E ) . The model explains the appearance of the second epidemic season when almost half of the population was immune with extraordinary transmission circumstances: the reproduction number was very large ( R0 , t > 3 ) for a period October 2010 through January 2011 , possibly reflecting a seasonal ( weather ) effect . Dorigatti et al . [19] reported a similar finding regarding the 3rd wave of A ( H1N1 ) pdm09 in the UK , one year after the 1st and 2nd waves . They inferred that the basic reproduction number increased to 1 . 5 before the 3rd wave and concluded that this was likely due to the combination of favourable weather conditions and possible evolution of the virus . Of note , we did not factor the possibility of waning immunity in the analysis and , should it have occurred , our estimate of R0 , t would be too high . The rate of introduction of infection to the population ( parameter q ) mainly plays the role of a primer that initiates the outbreaks . Its influence during the outbreaks ( epidemic seasons ) was insignificant . Its role was to add stochasticity to the onsets of influenza seasons , thus removing the need to introduce index cases at any fixed time . The estimates of q were notable only for age group 15-29 years , reflecting the fact that the first detected cases in the country belonged to these age groups . According to our analysis , vaccination played an important role in mitigating A ( H1N1 ) pdm09 transmission in the second season . By the start of the second season , 41% of the population were vaccine-protected while less than 5 – 7% had acquired immunity from infection . We estimated that in the absence of vaccination the affected population would have been about 4 – 8 times larger . It should be noted , however , that these predictions rely heavily on the posterior estimate of the transmission random effect ( wt ) , which in turn may be strongly dependent on the conditions and data in the 2010/2011 season . In a previous analysis of the same data set [2] , we assumed that vaccination did not affect the first season at all , which agrees with the current estimates . Given the two weeks period needed to develop protective immunity after vaccination , it is likely that an effective proportion of immune individuals was only obtained after the end the first epidemic outbreak ( Fig 5D ) . In this study we modelled the vaccine as having a 80% chance to induce complete immunity against the infection . This efficacy was based on a cohort study conducted in a sub-sample of the same population during the same time ( i . e . the first season of A ( N1H1 ) pdm09 ) [11] . We used a discrete-time SIR model with a one-week time step to correspond to the available data . However , a shorter time step would have been more realistic for capturing the dynamics of influenza for which the infectious period is known to last less than a week [6] . In this case , each infection generation in our model likely reflects several actual generations , therefore the basic reproduction number R0 , t is an overestimate of what would have been obtained with a smaller time step or a continuous model . The estimability of model parameters was constrained by the amount of available data . We used informative priors on all of the model parameters except the susceptibility p and the transmission random effect wt and set a strong smoothness constraint on the time-varying processes of transmission ( wt ) and detection ( d t ( mild ) ) . We conducted several sensitivity analyses to study the impact of the choice of the prior distributions ( S5 Appendix ) . We found that increasing the variance of the prior distributions leads to smaller attack rates and vice versa . The prior of the detection probability for mild cases d t ( mild ) was the most influential one . Some estimated quantities were more robust to prior specification . The detection probability d t m i l d was always estimated to increase before the outbreak of the first season . The estimated trends , e . g . the decreasing susceptibility with age and the V shape in the age-specific severity , were also immune to the choice of the prior . The reproduction number R0 , t always increased before the seasonal outbreaks . In our previous study [2] we analysed the same period of the two first years of pandemic influenza in Finland using a static model . A dynamic approach allowed us to take into account the available time-series data to learn about trends in transmissibility and detection . The presented model estimated the total incidence to be 1 . 5 times higher ( see Table 3 ) . A dynamic model also produced larger estimates of the impact of vaccination as it is able to take into account herd immunity effects . Using a static model , we estimated previously that without vaccination the overall attack rate would increase only by 0 . 8 percentage points . The attack rates and severity of A ( H1N1 ) pdm09 varied considerably by geographical region ( see Table 3 ) . Such variation may be partly due to lack of precision , based on the differences in data availability and in the methods of analysis . Nevertheless , the estimated attack rate in Finland was still smaller than found in other studies . Because of the high per-population risk of hospitalization in Finland ( 0 . 06% ) , the severity of infection ( hospitalization/infection ratio ) was higher in Finland than elsewhere , probably reflecting differences in the health care system and surveillance . Such differences emphasise the need to calibrate transmission models in each particular setting to best address questions about the performance of surveillance and the impact of influenza seasons . | In 2009 , the threat of the new pandemic influenza A ( H1N1 ) pdm09 ( referenced in media as ‘swine flu’ ) created a heavy burden to the public health systems wordwide . In Finland , an extensive vaccination campaign was set up in the middle of the first pandemic season 2009/2010 . However , the true number of infected individuals remains uncertain as the surveillance missed a large portion of mild infections . We built a probabilistic model of influenza transmission that accounts for observation bias and the possible impact of the changing weather and population behaviour . We used the model to simulate the spread of influenza in Finland during the two first years ( 2009-2011 ) of A ( H1N1 ) pdm09 in Finland . Using data from the national surveillance of influenza and data on social contacts in the population , we estimated that 9% of the population was infected with A ( H1N1 ) pdm09 during the studied period . Vaccination had a substantial impact in mitigating the second season . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"finland",
"medicine",
"and",
"health",
"sciences",
"infectious",
"disease",
"epidemiology",
"influenza",
"immunology",
"geographical",
"locations",
"preventive",
"medicine",
"age",
"groups",
"seasons",
"infectious",
"disease",
"control",
"vaccination",
"and",
"immunization",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"epidemiology",
"people",
"and",
"places",
"infectious",
"disease",
"surveillance",
"immunity",
"earth",
"sciences",
"disease",
"surveillance",
"population",
"groupings",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"europe"
] | 2016 | Revealing the True Incidence of Pandemic A(H1N1)pdm09 Influenza in Finland during the First Two Seasons — An Analysis Based on a Dynamic Transmission Model |
Secondary plant compounds are strong deterrents of insect oviposition and feeding , but may also be attractants for specialist herbivores . These insect-plant interactions are mediated by insect gustatory receptors ( Grs ) and olfactory receptors ( Ors ) . An analysis of the reference genome of the butterfly Heliconius melpomene , which feeds on passion-flower vines ( Passiflora spp . ) , together with whole-genome sequencing within the species and across the Heliconius phylogeny has permitted an unprecedented opportunity to study the patterns of gene duplication and copy-number variation ( CNV ) among these key sensory genes . We report in silico gene predictions of 73 Gr genes in the H . melpomene reference genome , including putative CO2 , sugar , sugar alcohol , fructose , and bitter receptors . The majority of these Grs are the result of gene duplications since Heliconius shared a common ancestor with the monarch butterfly or the silkmoth . Among Grs but not Ors , CNVs are more common within species in those gene lineages that have also duplicated over this evolutionary time-scale , suggesting ongoing rapid gene family evolution . Deep sequencing ( ∼1 billion reads ) of transcriptomes from proboscis and labial palps , antennae , and legs of adult H . melpomene males and females indicates that 67 of the predicted 73 Gr genes and 67 of the 70 predicted Or genes are expressed in these three tissues . Intriguingly , we find that one-third of all Grs show female-biased gene expression ( n = 26 ) and nearly all of these ( n = 21 ) are Heliconius-specific Grs . In fact , a significant excess of Grs that are expressed in female legs but not male legs are the result of recent gene duplication . This difference in Gr gene expression diversity between the sexes is accompanied by a striking sexual dimorphism in the abundance of gustatory sensilla on the forelegs of H . melpomene , suggesting that female oviposition behaviour drives the evolution of new gustatory receptors in butterfly genomes .
Nearly 50 years ago Ehrlich and Raven proposed that butterflies and their host-plants co-evolve [1] . Based on field observations of egg-laying in adult female butterflies , feeding behavior of caterpillars , and studies of systematics and taxonomy of plants and butterflies themselves , they outlined a scenario in which plant lineages evolved novel defensive compounds which then permitted their radiation into novel ecological space . In turn , insect taxa evolved resistance to those chemical defences , permitting the adaptive radiation of insects to exploit the new plant niche . Ehrlich and Raven's theory of an evolutionary arms-race between insects and plants drew primarily from an examination of butterfly species richness and host-plant specialization . It did not specify the sensory mechanisms or genetic loci mediating these adaptive plant-insect interactions . Insects possess gustatory hairs or contact chemosensilla derived from mechanosensory bristles , scattered along a variety of appendages [2]–[4] . In adult butterflies and moths , gustatory sensilla are found on the labial palps and proboscis ( Figure 1 ) , the legs ( Figure 2A ) [5] , the antennae ( Figure 2B ) [6] , [7] , and the ovipositor [8] , [9] . In adult Heliconius charithonia legs , the 5 tarsomeres of the male foreleg foretarsus are fused and lack chemosensory sensilla , while female foretarsi bear groups of trichoid sensilla ( n = 70–90 sensilla/tarsus ) associated with pairs of cuticular spines [10] . Each trichoid sensilla contains five receptor neurons . These sensilla are sensitive to compounds that may be broadly classified as phagostimulants ( e . g . , sugars and amino acids ) , which promote feeding behavior , or phagodeterrents ( secondary plant compounds ) , which suppress it [11]; in adult females they may also modulate oviposition [12] . Genes for vision , taste and smell are likely to be crucial genomic loci underlying the spectacular diversity of butterfly-plant interactions . The availability of genomes for two butterfly species , the postman Heliconius melpomene ( Nymphalidae ) [13] and the monarch ( Danaus plexippus ) [14] , as well as the silkmoth ( Bombyx mori ) [15] , enables us to examine the evolutionary diversification of gustatory ( Gr ) and olfactory ( Or ) receptor genes that mediate insect-plant interactions . Each of these species feeds on hosts from different plant families . Silkmoth larvae feed on mulberry ( Morus spp . , Moraceae ) and monarch larvae feed on milkweed ( Asclepias spp . , Apocynaceae ) . The larvae of Heliconius feed exclusively on passion flower vines , primarily in the genus Passiflora ( Passifloraceae ) . In addition , adult Heliconius are notable for several derived traits such as augmented UV color vision [16] , pollen feeding ( Figure 1B ) [17] , [18] , and the ability to sequester substances from their host plants that are toxic to vertebrate predators such as birds [19] , [20] . In Drosophila melanogaster , the Gr gene family consists of 60 genes [21]–[24] , several of which are alternatively spliced , yielding 68 predicted Gr transcripts [24] . One or more of these Gr proteins including possibly obligatory co-receptors [25]–[27] may be expressed in each gustatory receptor neuron [11] . Originally considered members of the G-protein-coupled receptor ( GPCR ) family , insect Grs have an inverted orientation in the membrane compared to the GPCR family of vertebrate Grs [28] and are part of the same superfamily as the insect Ors [21] . Signalling pathways for insect Grs may be both G-protein dependent [29] , [30] , [31] and G-protein independent [32] . For the vast majority of Drosophila Grs the specific compounds to which they are sensitive remain unknown . Nonetheless , several receptors for sugars [33]–[35] , CO2 [26] , [36] , bitter substances [37]–[39] and plant-derived insecticides [25] have been identified in flies . Knowledge of the Gr gene family for insects outside Drosophila is sparse and has primarily relied on the analyses of individual reference genomes . Expression studies are challenging , due to the very low expression of Grs in gustatory tissues [21] , [23] . In addition , Grs and Ors typically have large introns , small exons and undergo fast sequence evolution , making their in silico identification using automated gene prediction algorithms from genomic sequences problematic . Thus , the large repertoire of Grs ( and Ors ) that have been examined in the reference genomes of the pea aphid [40] , the honey bee [41] , the red flour beetle Tribolium castaneum [42] , the mosquitoes Aedes aegypti [43] and Anopheles gambiae [44] , and several Drosophila spp . [45] , [46] have required extensive manual curation . In Lepidoptera , a large insect group which includes ∼175 , 000 species , completely described Gr ( and Or ) gene models from genomes are rare and limited to B . mori [47] , D . plexippus [14] and H . melpomene ( Grs , this study; Ors , [13] ) . In other lepidopteran species , only fragmentary Gr data are available: five sequences in Spodoptera littoralis [48] , three in Heliothis virescens [49] , two in Manduca sexta [50] , [51] and one in Papilio xuthus [52] . Adult females of each Heliconius species only lay eggs on a limited number of host plants [53] , and therefore need to recognize different species from among the large and diverse Passifloraceae family , which also show a remarkable diversity of chemical defences [54] . The evolutionary arms race between Heliconius butterflies and their hosts led us to hypothesize that Heliconius Grs ( and Ors ) might be subject to rapid gene duplication and gene loss as well as copy-number variation ( CNV ) . Recent work taking advantage of published Drosophila genomes has shown a relationship between host specialization and/or endemism and an increased rate of gene loss , as well as a positive relationship between genome size and gene duplication [46] , [55] . Moreover , Drosophila Grs appear to be evolving under weaker purifying selection than Ors [55] . We previously used the reference genome sequence for H . melpomene to annotate three chemosensory gene families , encoding the chemosensory proteins ( CSPs ) , the odorant-binding proteins ( OBPs ) , and the olfactory receptors ( Ors ) . This demonstrated a surprising diversity in these gene families . In particular there are more CSPs in the butterfly genomes than in any other insect genome sequenced to date [13] . We build on this work below by characterizing the Gr gene family in the reference H . melpomene melpomene genome and in two other lepidopteran species whose genomes have been sequenced , B . mori ( Bombycidae ) and D . plexippus ( Nymphalidae ) , by performing in silico gene predictions and phylogenetic analysis . We then analyzed whole-genome sequences of twenty-seven individual butterflies , representing eleven species sampled across all major lineages of the Heliconius phylogeny and including sixteen individuals from two species , H . melpomene and its sister-species H . cydno . We also generated RNA-sequencing expression profiles of the proboscis and labial palps , antennae and legs of individual adult male and female butterflies of the sub-species H . melpomene rosina from Costa Rica ( ∼1 billion 100 bp reads ) . We used these data to address four major questions: Are different chemosensory modalities less prone to duplication and loss than others ( e . g . , taste vs . olfaction ) ? Is there evidence of lineage-specific differentiation of Gr ( and Or ) repertoires between genera , species and populations ? What is the relationship between CNVs and the retention of paralogous genes over long-term evolutionary timescales ? Are the life history differences between males and females reflected in the expression of Grs and Ors as well as in the retention of novel sensory genes in the genome ? We find higher turnover of the Grs than the Ors over longer evolutionary timescales , and evidence for both gene duplication and loss among a clade of intronless Grs between lepidopteran species and within the genus Heliconius . We also find for H . melpomene and its sister species , H . cydno , evidence of copy-number variation ( CNVs ) within their Gr and Or repertoires . Lastly , our RNA-sequencing suggests both tissue-specific and sex-specific differences in the diversity of expressed Grs and Ors , with female legs expressing a more diverse suite of Grs than male legs . Our data set revealing the expression of 67 of 73 predicted Gr genes and 67 of 70 predicted Or genes in adult H . melpomene butterflies is the most comprehensive profiling of these chemosensory gene families in Lepidoptera to date , and suggests how female host plant-seeking behaviour shapes the evolution of gustatory receptors in butterflies .
In total , we manually annotated 86 , 870 bp of the H . melpomene melpomene reference genome ( Table S1 ) . Our 73 Gr gene models , consisted of 1–11 annotated exons , with the majority having three or four exons; six were intronless . We found genomic evidence ( but not RNA-seq evidence ) of possible alternative splicing of the last two exons of HmGr18 , bringing the total number of predicted Grs to 74 . Alternative splicing has not been previously described in the silkmoth B . mori [47] , but is known to occur in most other insects examined , including D . melanogaster , Anopheles gambiae , Aedes aegypti and T . castaneum [24] , [43] , [44] . We also identified eleven new putative Grs in the monarch butterfly genome , DpGr48-56 , DpGr66 and DpGr68 ( Table S1 ) [14] . All but five of our gene models contained more than 330 encoded amino acids ( AAs ) while individual gene models ranged from 258–477 AAs . Several Gr genes contained internal stop codons ( Table S1 ) . In at least one case , we found RNA-seq evidence of an expressed pseudogene–HmGr61–with two in-frame stop codons . In other cases , the 5′ end of our assembled transcripts was not long enough to verify the internal stop codons in the genome assembly . The Grs are located on 33 distinct scaffolds , with 58 forming clusters of 2–8 genes on 18 scaffolds , distributed across 14 chromosomes . To study the patterns of gene duplication and loss more broadly across the Lepidoptera , we next examined the phylogenetic relationships of Grs from the three lepidopteran reference genomes [13]–[15] . Across the gene family phylogeny a large number of duplications among the putative ‘bitter’ gustatory receptors of Heliconius or Danaus have occurred , while the putative CO2 and sugar receptors are evolving more conservatively , with only single copies in the H . melpomene reference genome ( see below ) ( black arcs , Figure 3 ) . A majority ( ∼64% ) of Gr genes found in the H . melpomene genome are the result of gene duplication since Heliconius shared a common ancestor with Danaus or Bombyx . This is in contrast to the more conserved pattern of evolution of the Ors ( Figure 4 ) [13] where a majority ( 37 of 70 or 53% ) of genes show a one-to-one orthologous relationship with either a gene in Danaus , in Bombyx or both . Within the genus Heliconius there is a great diversity of host plant preferences for different Passiflora species . To look at the relationship between gene duplication and loss over this shorter timescale , we focussed our efforts on a group of six intronless Grs , HmGr22-26 and Gr53 , because it is only feasible to identify single-exon genes with high confidence , given that the Illumina whole-genome sequencing approach leads to poorly assembled genomes ( Table S2 ) . These genes are also of interest as some members of this group are very highly expressed . Notably HmGr22 is one of the most widely expressed genes in our adult H . melpomene transcriptomes , which was verified by reverse-transcriptase ( RT ) -PCR and sequencing of the PCR products ( Figure 5A ) . In this regard HmGr22 resembles another intronless Gr , the silkmoth gene BmGr53 , which is expressed in adult male and female antennae and larval antennae , maxilla , labrum , mandible , labium , thoracic leg , proleg and gut [32] . The remaining five intronless Grs have much more limited domains of expression in adult H . melpomene ( see below ) . We searched for these genes in de novo assemblies of whole-genome Illumina sequences from eleven species across the Heliconius phylogeny . We investigate whether , as in Drosophila , a high turnover in putative bitter receptors is observed in species with host plant specializations or in species which are endemic and thus smaller in effective population size [46] . Although patterns of host plant use are complex within the genus , some notable host-plant shifts have occurred , leading to the prediction that gene loss may have occurred along more specialized lineages [46] . For example , H . doris unlike many Heliconius , tends to feed on large woody Passiflora that can support their highly gregarious larvae [53] . It also probably has a smaller effective population size than most other Heliconius species . From the 11 species studied , we identified a total of 44 intact or nearly intact intronless Grs , as well as three intronless pseudogenes ( Genbank Accession Nos . KC313949-KC313997 ) ( Table S2 and S3 ) . We also identified one intact intronless Gr each in monarch and silkmoth and one intronless Gr pseudogene in monarch . Phylogenetic analysis indicates that six intact intronless Gr genes were present at the base of the genus Heliconius while the intronless Gr pseudogene in monarch was the result of duplication since Heliconius and monarch shared a common ancestor ( Figure 5B , Figure 6 ) . Subsequent to the radiation of the genus Heliconius , there have been a number of gene losses . Whereas all members of the melpomene clade ( H . melpomene , H . cydno , H . timareta ) retained genomic copies of all six genes , members of the erato clade ( H . erato , H . clysonymus and H . telesiphe ) and sara-sapho clade ( H . sara and H . sapho ) have lost their copies of Gr22 and Gr25 . In addition , members of the so-called primitive clade ( H . wallacei , H . hecuba , and H . doris ) have lost Gr23 , while H . doris and H . wallacei have apparently lost Gr24 independently ( Figure 6 ) . The woody plant specialist , H . doris , has retained the fewest intronless Grs , apparently also having lost its copy of Gr53 , a pattern mirrored by Drosophila host plant specialists [46] . We have , however , no direct evidence that the intronless Grs are in fact involved in host plant discrimination so the observed patterns of loss may be better explained by other variables such as effective population size . We next tested whether the greater level of diversification of Grs as compared to Ors over long evolutionary timescales ( compare Figure 3 and Figure 4 ) , is similarly reflected in greater population level variation in Gr and Or duplicate genes . To test this hypothesis , we examined the incidence of CNVs among Grs and Ors that exist as single-copy genes in the reference H . melpomene genome with a one-to-one orthologous relationship with a gene in Danaus , Bombyx or both ( conserved ) ( red dots , Figure 3 and 4 ) , or as genes that are Heliconius-specific where no orthologue exists in either Danaus or Bombyx ( non-conserved ) . We used whole genome resequence data ( 12 genomes ) for three subspecies of H . melpomene ( H . melpomene amaryllis , n = 4; H . melpomene aglaope , n = 4; and H . melpomene rosina , n = 4 ) ( Figure 7 , inset ) and one sub-species of H . cydno ( H . cydno chioneus , n = 4 ) ( Table S4 ) . We first mapped genomic resequence reads to the H . melpomene melpomene reference genome , and then searched for regions of abnormal coverage using CNVnator [56] . More than half of Gr loci showed presence of CNVs ( 37 out of 68 loci ) . However , there were noticeably fewer CNVs in Gr loci that evolve conservatively over the long-term , such as among the putative CO2 receptors , while there was an excess of CNVs in loci that show patterns of Heliconius-specific duplication ( 11 . 1% vs . 54 . 9% , respectively ) ( Fisher's Exact Test , two-tailed , P = 0 . 0004 ) ( Table 1 ) ( Figure 7 ) . Intriguingly , many sugar receptor CNVs are sub-species specific; we observed fixed duplications relative to the reference genome in H . melpomene aglaope ( HmGr4 , Gr5 , Gr6 , Gr8 , Gr45 , Gr52 ) and H . melpomene amaryllis ( Gr4 , Gr5 , Gr6 , Gr7 , Gr8 , Gr45 , Gr52 ) , among genes that are found on different chromosomes ( Table S5 , Figure 7 ) . Although the majority of CNVs are likely to be evolving neutrally , this raises the possibility of local adaptation within the species range around the detection of sugars . As expected given their long-term stability , Ors also show a lower incidence of CNVs ( 12 out of 67 loci ) , with no association between gene duplication and CNV incidence at least in H . melpomene ( Table 1 , Table S6 ) . In H . cydno , a slight excess of Or CNVs was observed in loci that resulted in paralogous genes over longer evolutionary timescales ( Fisher's Exact Test , two-tailed , P = 0 . 0475 ) ( Table 1 ) ( Figure 8 ) . We have not experimentally verified the incidence of copy number variation in any of these genomes , and some of the regions identified as CNVs are likely to be false positives . To investigate the rate of false positives , we analysed resequence data from the reference genome itself and discovered 3 Gr and 3 Or CNVs , suggesting a false positive rate of around 4% . ( We therefore excluded these loci from our statistical tests . ) However , the fact that broad patterns of observed CNVs are consistent with the evolutionary patterns at deeper levels supports our conclusion that CNV , in the absence of strong purifying selection , is an important driver of gene family diversification . These results also provide a novel line of evidence that the butterfly Grs have a higher rate of evolutionary turnover as compared to Ors . The life histories of adult male and female butterflies are similar with respect to the need to find food and potential mates except that adult females are under strong selection to identify suitable host plants for oviposition . To ascertain host-plant identity , female butterflies drum with their legs on the surface of leaves before laying eggs [10] . This behaviour presumably allows the female to taste oviposition stimulants . Consistent with this behaviour , adult nymphalid butterfly legs are known to contain gustatory sensilla [57] , and it has been reported that while nymphalid butterfly females have clusters of gustatory sensilla on their foreleg foretarsi , males lack these entirely [10] , [58] . Here we confirm this mostly anecdotal evidence for sexual dimorphism using scanning electron microscopy ( SEM ) . The mid- and hindlegs of both male and female H . melpomene have similar numbers of individual gustatory sensilla along their entire lengths , but there is a striking difference in their abundance and distribution on the foretarsi of the female forelegs . Unlike males , females exhibit cuticular spines associated with gustatory ( trichoid ) sensillae ( n∼80 sensilla/foretarsus for females; n = 0/foretarsus for males ) ( Figure 2A ) [10] . We therefore hypothesized that the repertoire of expressed Gr and Or genes in H . melpomene legs might be more diverse in females as compared to males . Furthermore , if female-specific genes are used for assessment of potential host plants , then fast-evolving insect-host interactions might produce rapid duplication of these genes over evolutionary timescales . Accordingly , we examined the expression profiles of Grs and Ors in adult H . melpomene by RNA-sequencing of libraries prepared from mRNAs expressed in adult antennae , labial palps and proboscis , and legs from one deeply-sequenced male and female each of H . melpomene ( 6 libraries total ) ( Table S7 and S8 ) . The number of 100 bp reads per individual library ranged from 17 . 4 to 25 . 9 million for paired-end sequencing or 74 . 8–103 . 9 million for single-end sequencing ( Table S8 ) . To confirm these findings , we subsequently made 12 individual libraries from two more males and two more females ( Table S7 ) . As coverage was uneven across these libraries , we analysed them by merging biological replicates by sex and tissue type , and then downsampling so that an equal number of reads was analyzed for each treatment . The number of 100 bp reads analyzed for paired-end sequencing ranged from 19 . 4 to 49 . 6 million ( Table S8 ) . After downsampling , we examined the expression levels of the widely-expressed elongation factor-1 alpha gene in each of the libraries as a control , and found a comparable level of expression between sexes within each tissue type ( Table S8 ) . By careful visual examination of the uniquely-mapped reads to our 143 reference Gr and Or sequences , we found evidence of Gr and Or expression in all three adult tissue-types , with both tissue-specific and sex-specific differences as detailed below ( Figure 9 , Tables S9 , S10 , S11 , S12 , S13 , S14 ) . In total , we found evidence for expression of 67 of 73 Grs and 67 of 70 Ors identified in the H . melpomene reference genome . Strikingly , the sexual dimorphism of gustatory sensilla we observed among the foreleg foretarsi is reflected in Gr gene expression patterns . A total of thirty-two Grs are expressed in both male and female H . melpomene leg transcriptomes including three CO2 receptors , HmGr1-3 , four putative sugar receptors HmGr4 , Gr6 , Gr45 and Gr52 and a fructose receptor , HmGr9 ( Figure 9A , Table S9 , Supplementary Text ) . Many Grs showed sex-specific expression , however , with many more Grs in female ( n = 46 ) as compared to male leg transcriptomes ( n = 33 ) ( Figure 9B , C ) . In total 15 of these Grs expressed in female legs , HmGr10 , Gr24 , Gr26 , Gr29 , Gr40 , Gr41 , Gr48 , Gr50 , Gr51 , Gr16 , Gr55 , Gr57 , Gr58 , Gr60 and Gr67 , are the result of duplications since Heliconius and Danaus shared a common ancestor ( Figure 3 small arrows , Figure 9B , Table S9 ) . By contrast , only one of the three male-biased Grs , HmGr19 , evolved as a result of recent duplication . There is an excess of Heliconius-specific Grs but not Ors ( see below ) that are expressed in female legs ( Fisher's Exact Test , two-tailed , p = 0 . 019 ) ( Table 2 ) . Since male H . melpomene do not need to identify host-plants for oviposition , it seems likely that the 17 female-specific Grs in our leg transcriptomes are candidate receptors involved in mediating oviposition ( Figure S1 ) . Besides using their antennae for olfaction , female nymphalid butterflies also taste a host plant by antennal tapping before oviposition . This tapping behaviour presumably allows the host plant chemicals to come into physical contact with gustatory sensilla on the antennae . We therefore examined whether there was any difference in the abundance of gustatory sensilla on the antennae of male and female H . melpomene . Using scanning electron microscopy , we found individual gustatory sensilla scattered along each antennae of both male and female H . melpomene but no obvious sexual dimorphism in their abundance or distribution ( Figure 2B ) . We found 28 Grs expressed in both male and female H . melpomene antennae ( Figure 9A , Table S10 ) , including two sugar receptors , HmGr4 and HmGr52 , a putative fructose receptor HmGr9 and two CO2 receptors , HmGr1 and Gr3 . Besides the sugar and CO2 receptors noted , other conserved genes that are expressed in both male and female antennae include HmGr63 , a candidate Gr co-receptor ( see Text S1 ) , and HmGr66 , a candidate bitter receptor . We also found 11 Grs expressed in female H . melpomene antennae that did not appear to be expressed in male antennae . Two of these , HmGr47 and Gr68 , appeared in the top one-third of the most abundant female antennal Grs in terms of number of reads recovered from the individual butterfly transcriptome . In contrast , just four Grs were expressed in male antennae HmGr11 , Gr25 , Gr31 , and Gr69 but not female antennae ( Figure 9B , C , Table S10 ) . Six of the female-biased Grs and two of the male-biased Grs ( Gr31 , Gr69 ) expressed in antennae are the result of duplication events since Heliconius and Danaus shared a common ancestor . By contrast with the leg and antennal tissue , where more Grs are expressed in females compared to males , the labial palps and proboscis ( Figure 1 ) transcriptomes contained the largest number of Grs ( n = 35 ) expressed in both sexes ( Figure 9A , C , Table S11 ) . Five of the six candidate sugar receptors in the H . melpomene genome are expressed in both the male and the female transcriptomes along with two of the three conserved CO2 receptors , which may be used to assess floral quality [59] ( Figure 3 , Table S11 ) . A majority ( 21 of 35 ) of Heliconius Grs expressed in both male and female labial palps and proboscis libraries have no existing ortholog in the silkmoth genome , apparently the result of gene loss in B . mori or gene duplication along the lineage leading to Heliconius ( Figure 3 ) . This may in part reflect the fact that adult silkmoths have lost the ability to feed . Interestingly , four Grs expressed in both male and female labial palps and proboscis transcriptomes could not be detected in male and female antennae and legs ( HmGr12 , Gr20 , Gr35 , and Gr59 ) ( Figure 3 , red arrows , Figure 9B ) . Some of these Grs might play a role in the pollen-feeding behaviour that is specific to Heliconius , and which involves preferences for particular species of flowers in the plant families Rubiaceae , Cucurbitaceae and Verbenaceae ( see Discussion ) . In addition to the Gr gene expression described above , we examined Or expression in the three adult tissues . The expression of Ors in antennal tissue has been widely studied in a variety of insects including Drosophila and some Lepidoptera [50] , [60] . As expected , we observed that Or gene expression was high in the antennae . Unexpectedly , Or expression was about as prevalent as Gr expression in the proboscis and labial palps and leg transcriptomes ( Figure 9D , E , F ) . In total across all three tissues profiled , we found evidence for the expression of nearly all predicted Or genes ( 67 of 70 genes ) ( Table S12 , S13 , S14 ) in the H . melpomene reference genome [13] .
We have shown that like the opsin visual receptors [80] , the chemosensory superfamily composed of constituent Gr and Or families in Lepidoptera show rapid gene family evolution , with higher rates of copy-number variation and gene duplication among the Grs than the Ors , as well as gene losses in the Grs . In particular , there is a group of putative bitter receptors that show female-specific expression in the legs and that are especially prone to gene duplication , providing new material for sensory diversification in the insect-host plant arms race . We have also shown , for the first time , widespread expression of Ors in non-antennal tissues in a lepidopteran . With the most comprehensive data set on Gr and Or expression in butterflies to date we are one step closer to identifying the sensory and molecular genetic basis of the Heliconius-Passiflora co-evolutionary race that inspired Ehrlich and Raven in 1964 .
tBLASTn searches were conducted iteratively against the H . melpomene melpomene genome ( version v1 . 1 ) and haplotype scaffolds [13] using B . mori [28] , [47] and D . plexippus Grs [14] as input sequences . For these in silico gene predictions , intron-exon boundaries were identified by first translating the scaffold nucleotides in MEGA version 5 [81] , searching for exons identified in the tBLASTn searches , then back translating to identify splice junctions . Intron sequences were then excised to verify that the remaining exonic sequences formed an in-frame coding sequence . Insect Grs are defined by a conserved C-terminal motif TYhhhhhQF , where ‘h’ is any hydrophobic amino acid [21] . We inspected our predicted protein sequences for this motif or variants thereof , specifically ‘S’ , ‘M’ or ‘K’ instead of a ‘T’ or ‘L’ , ‘T’ or ‘I’ instead of ‘F’ . In the handful of cases where we were unable to find the last short exon that contains this motif , final assignment to the Gr gene family was based on using the predicted amino acid sequence as a search string for either tBLASTn or BLASTp against the nr/nt Genbank database . Gene annotations were submitted to the EnsemblMetazoa database http://metazoa . ensembl . org/Heliconius_melpomene/Info/Index as part of the H . melpomene v . 2 genome release ( for GeneIDs see Table S1 ) . Chromosomal assignments were based on published mapping of scaffolds in the H . melpomene melpomene reference genome [13] . Following amino acid alignment using ClustalW , preliminary phylogenetic trees were constructed in MEGA using neighbor-joining and pair-wise deletion to identify orthologous relationships with B . mori and D . plexippus Grs . Reciprocal tBLASTn searches against the B . mori and D . plexippus genomes as well as searches using the protein2genome module in EXONERATE [82] were then performed in order to search for ‘missing’ Grs in those genomes . Final phylogenetic analysis was performed using a maximum-likelihood ( ML ) algorithm and JTT model on an amino acid alignment that was inspected by eye and manually adjusted . These results were compared to a ML tree made from a Clustal-Omega alignment [83] and were found to be nearly identical . Once the initial H . melpomene Gr gene predictions were obtained , EXONERATE , Perl scripts and manual annotations in Apollo [84] were used to produce gff3 files for submission of the annotated H . melpomene genome scaffolds to EMBL-EBI . Butterfly pupae of H . melpomene rosina were obtained from Suministros Entomológicos Costarricenses , S . A . , Costa Rica . Adult males and females were sexed and frozen at −80°C . Total RNAs were extracted separately from antennae , proboscis together with labial palps , and all six legs of three males and three females of H . melpomene using Trizol ( Life Technologies , Grand Island , NY ) . A NucleoSpin RNA II kit ( Macherey-Nagel , Bethlehem , PA ) was used to purify total RNAs . Each total RNA sample was purified through one NucleoSpin RNA II column . Purified total RNA samples were quantified using a Qubit 2 . 0 Fluorometer ( Life Technologies , Grand Island , NY ) . The quality of the RNA samples was checked using an Agilent Bioanalyzer 2100 ( Agilent Technologies , Santa Clara , CA ) . 0 . 3–4 . 0 µg of purified total RNAs were used to make cDNA libraries . A TruSeq RNA sample prep kit ( Illumina , San Diego , CA ) was used to prepare 18 individual cDNA libraries . After being normalized according to their concentrations , the enriched individual libraries were pooled and then run on a 2% agarose gel . cDNA products ranging from 280 to 340 bp with an average of 310 bp were cut out and purified using a Geneclean III kit ( MP Biomedicals , Solon , OH ) to facilitate post-sequencing assembly . After being re-purified using Agencourt AMPure XP magnetic beads ( Beckman Coulter Genomics , Danvers , MA ) , the cDNA pool was quantified using the Qubit 2 . 0 Fluorometer , and quality control-checked using the Agilent Bioanalyzer 2100 . The cDNA pools were then normalized to 10 nM and run as either two paired-end or three single-end 100 bp runs on a HiSeq 2000 ( Illumina , San Diego , CA ) by the UCI Genomics High-Throughput Facility . mRNA sequences were demultiplexed , trimmed and sorted using Python and Perl scripts . A single de novo assembly of the combined libraries was performed using CLC Genomics Workbench 5 to check for missing exons in our gene models . The 73 corrected Gr gene models and 70 Or gene models were then used as an alignment reference to perform unique read mapping of each individual chemosensory transcriptome . To determine if an individual Gr or Or was expressed in a given tissue , each of the 1716 individual Gr and Or mapping alignments was inspected by eye for uniquely mapped reads , and any spuriously-mapped reads ( i . e . , reads <70 bp in length with indels or sequence mismatches at the ends ) were discarded . As a control for potential differences in RNA preparation between samples , we also quantified the number of uniquely mapped fragments to the widely-expressed elongation factor 1-alpha ( EF1α ) gene transcript and calculated the Fragments Per Kilobase of transcript per Million mapped reads ( FPKM ) [85] . Illumina reads for each of the libraries were deposited as fastq files in the ArrayExpress archive under the accession number: E-TAB-1500 ( Table S7 ) . One week old adult H . melpomene rosina butterflies were sexed , frozen at −80°C , then dissected and mounted for imaging on an FEI/Philips XL30 FEG scanning electron microscope at UCI's Materials Characterization User Facility . Forelegs , middle legs , hindlegs and antennae were examined for the presence of gustatory sensilla . We also examined resequenced genomes of twelve H . melpomene and four H . cydno individuals , including H . melpomene aglaope , H . melpomene amaryllis and H . melpomene rosina ( Table S4 ) , sequenced by The GenePool , University of Edinburgh , U . K . and the FAS Center , Harvard University , U . S . A . , for evidence of copy-number variation ( CNV ) in the Grs and Ors using CNVnator [56] . These sequences were deposited in the European Nucleotide Archive ( ENA ) under accession number: ERP002440 . The Illumina resequenced genomes were first mapped to the H . melpomene reference genome and the average read depth was calculated along a 100 bp sliding window . The output of CNVnator was parsed for candidate insertion and deletion variants , and those with estimated copy number of >2× were counted as potential duplications and <0 . 5× as potential deletions . The GenePool , University of Edinburgh , and the Oxford Genomics Centre , University of Oxford , U . K . , produced whole genome 100 bp sequences from H . cydno , H . timareta , H . wallacei , H . doris , H . clysonymus , H . telesiphe , H . erato petiverana , H . sara and H . sapho using the Illumina Pipeline v . 1 . 5–1 . 7 with insert sizes ranging from 300 to 400 bp . We deposited sequences for H . sapho and H . sara in the Sequence Read Archive ( SRA ) under accession number ERP002444 . We performed de novo assembly of the short reads using Abyss v . 1 . 2 [86] implemented in parallel at the School of Life Sciences , University of Cambridge , U . K . Based on previous results [87] , recommendations estimated by the software , and comparison of N50 values in preliminary experiments , we chose a k-mer size of 31 , a minimum number of pairs required n = 5 and the minimum mean k-mer coverage of a unitig c = 2 ( full command: abyss-pe n = 5 k = 31 c = 2 in = ‘for . fastq rev . fastq’ ) . In all assemblies , at least 96% of reads mapped back to the contigs . We created BLAST databases of these whole genome sequence assembly contigs ( Table S2 ) in Geneious Pro v . 5 . 5 . 6 . The lack of introns in the putative bitter receptor genes Gr22-26 and Gr53 permitted us to easily retrieve them from these BLAST databases . To confirm the identity and improve the quality of the sequences found , we mapped the reads to the assembled exon sequences in CLC Genomics Workbench v . 5 . 5 . 1 , using the following conservative settings to prevent mis-mapping of paralogous sequences: mismatch , insertion and deletion cost of 3; length fraction and similarity fraction of 0 . 9 . We then inspected all read-mappings by eye . Because the intronless Grs are closely related , we aligned the translated nucleotide sequences in MEGA using the ClustalW algorithm , and also inspected the alignment by eye . For all intronless Gr sequences except for the pseudogenes , sequence length was highly conserved ( i . e . , there were few indels ) . To illustrate the high substitution rate of the retrieved pseudogene sequences , we selected the neighbor-joining method for tree reconstruction and performed 500 bootstrap replicates . To infer the number of intronless Gr gene duplications and losses , we used the program Notung v . 2 . 6 [88] , [89] , which reconciles gene trees onto the species tree . The gene tree was made by a maximum likelihood analysis of 1074 nucleotide sites , aligned by Clustal-Omega , and 500 bootstrap replications . The species tree was derived from a phylogeny based on independent nuclear and mitochondrial DNA sequences [90] . We verified the presence of HmGr22 in several adult tissues using reverse-transcriptase PCR and primers for HmGr22 ( 5′-CCATAATTTTGTCATCCT-3′ and 5′-GATTTCGAAATAAGGTCTGT-3′ ) and EF1alpha ( 5′-CGTTTCGAGGAAATCAAGAAGG-3′ and 5′-GACATCTTGTAAGGGAAGACGCAG 3′ ) . RNA was extracted from fresh frozen specimens using Trizol and purified using the Nucleospin RNA II kit , which contains a DNAase-treatment step . RNA concentration was diluted to 12 . 5 µg/ml . Each 25 µl reaction had 2 . 5 µl 10× BD Advantage 2 PCR buffer , 2 . 5 µl dNTPs ( 2 mM ) , 0 . 5 µl ( 100 µM ) forward and 0 . 5 µl reverse primer , 0 . 5 µl ( 1∶20 diluted ) Stratagene Affinity Script Reverse Transcriptase , 0 . 5 µl 50× Advantage 2 Polymerase Mix , 17 µl H2O and 1 µl RNA . The PCR reaction consisted of 38 cycles of 95°C for 30 s , 55°C for 30 s , and 68°C for 55 s . The identity of the RT-PCR products was confirmed by Sanger sequencing . | Insects and their chemically-defended hostplants engage in a co-evolutionary arms race but the genetic basis by which suitable host plants are identified by insects is poorly understood . Host plant specializations require specialized sensors by the insects to exploit novel ecological niches . Adult male and female Heliconius butterflies feed on nectar and , unusually for butterflies , on pollen from flowers while their larvae feed on the leaves of passion-flower vines . We have discovered–between sub-species of butterflies-fixed differences in copy-number variation among several putative sugar receptor genes that are located on different chromosomes , raising the possibility of local adaptation around the detection of sugars . We also show that the legs of adult female butterflies , which are used by females when selecting a host plant on which to lay their eggs , express more gustatory ( taste ) receptor genes than those of male butterflies . These female-biased taste receptors show a significantly higher level of gene duplication than a set of taste receptors expressed in both sexes . Sex-limited behaviour may therefore influence the long-term evolution of physiologically important gene families resulting in a strong genomic signature of ecological adaptation . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genome",
"sequencing",
"genomics",
"gustatory",
"system",
"gene",
"expression",
"genetics",
"olfactory",
"system",
"molecular",
"genetics",
"comparative",
"genomics",
"biology",
"gene",
"duplication",
"evolutionary",
"biology",
"genomic",
"evolution",
"neuroscience",
"sensory",
"systems"
] | 2013 | Female Behaviour Drives Expression and Evolution of Gustatory Receptors in Butterflies |
Detailed observations of larval Drosophila chemotaxis have characterised the relationship between the odour gradient and the runs , head casts and turns made by the animal . We use a computational model to test whether hypothesised sensorimotor control mechanisms are sufficient to account for larval behaviour . The model combines three mechanisms based on simple transformations of the recent history of odour intensity at the head location . The first is an increased probability of terminating runs in response to gradually decreasing concentration , the second an increased probability of terminating head casts in response to rapidly increasing concentration , and the third a biasing of run directions up concentration gradients through modulation of small head casts . We show that this model can be tuned to produce behavioural statistics comparable to those reported for the larva , and that this tuning results in similar chemotaxis performance to the larva . We demonstrate that each mechanism can enable odour approach but the combination of mechanisms is most effective , and investigate how these low-level control mechanisms relate to behavioural measures such as the preference indices used to investigate larval learning behaviour in group assays .
It is well established that Drosophila larvae perform chemotaxis towards a wide range of odourants ( e . g . [1] ) . Our aim in this paper is to examine what sensorimotor mechanism ( s ) account for larval chemotaxis , looking for a minimal model that captures observed phenomena . This will allow us to examine the nature of sensory input and its processing , and identify possible key control outputs that are modulated by conditions or experience . In particular , we are interested in connecting models of odour discrimination and learning to the odour experience of the animal as it moves in a gradient . Many other organisms also exhibit chemotaxis , using a variety of different strategies [2] . The basic forms of orientation mechanism are reviewed in [3] . Bacteria alternate straight swimming and random tumbling , with the probability of switching modulated by the direction of change in chemical intensity [4] . In C . elegans , a similar modulation of the frequency of large re-orientations ( pirouettes ) by the odour gradient is accompanied by a more gradual directed bias of runs towards the odour [5] . Insects such as the silkworm moth that navigate in patchy odour plumes make upwind surges in response to odour encounters , interspersed with zig-zag and casting behaviours [6] . Flies approaching odour sources in relatively still air might do so by alteration of their visuomotor control , to increase straight flight and suppress turning if odour concentration is increasing [7 , 8] . Note that none of these strategies requires the use of spatially separated olfactory sensors to obtain instantaneous measurement of the direction of an odour gradient , but rather exploit temporal change due to movement of the animal , movement of the chemosensors , movement of the medium carrying the odour , or a combination of all three . However , in many cases a bilateral arrangement of sensors does make instantaneous assessment of the relative concentration across space possible , and this is sometimes exploited: for example , bees [9] and flies [10 , 11] exhibit turning towards the antenna experiencing higher concentration . Drosophila larvae’s olfactory sensors are located at the tip of the head [12 , 13] . As larvae have left and right olfactory sensory organs it would seem possible that they could compare between left and right odour concentrations to perform odour taxis . It has been reported that crude unilateral surgical ablation of sensory organs leads to increased turning towards the intact side [14] . However the separation between these sensors is very small and it seems unlikely that the minute instantaneous difference in concentration between left and right could be detected over environmental , sensory and neural noise [15] . Furthermore , using genetic rescue of single olfactory neurons , it has been demonstrated that while bilateral sensory input improves chemotaxis , it is not required [16] . The most salient features of the larva’s movement patterns also seem inconsistent with instantaneous lateral steering . Larval locomotion has two distinct modes [17] . During runs , consistent peristaltic waves cause the larva to move forwards in a relatively straight line ( but see below ) . During turns , unilateral contraction of one side of the body or the other causes the anterior section of the body to sweep from side to side , a behaviour referred to as ‘head casting’ . The effective direction of a turn is determined by the casting behaviour ending with the anterior section of the body at an angle to the rest of the body . In this case , when the larva resumes running , it moves off in a new direction with respect to the previous run . As it moves forward , the rear gradually realigns itself with the front . Larvae have been shown to produce run and turn behaviours without the brain [18] , suggesting they may have a ‘basic’ locomotion pattern embedded in the ventral nerve cord and motor system , which can be modulated by higher brain areas in response to sensory input . The most detailed behavioural description of larval Drosophila chemotaxis comes from [19] . By using an arena designed to produce a well-defined odour gradient ( described in [16] ) , and fine-grained tracking of individual larvae exploring this environment , the authors were able to decompose larval behaviours based on orientation with respect to the local odour gradient . This analysis revealed that larvae were 1 ) more likely to stop runs and start head casting when moving down gradient , and 2 ) more likely to turn ( i . e . finish head casting and return to running ) towards the direction of higher odour concentration . Similar results have been reported in a study using linear odour gradients [20] . But how do larvae determine when to turn and which direction to turn ? Turn initiation is typically preceded by a period of decreasing sensory experience ( defined as a normalised derivative of concentration ) corresponding to running down the gradient [19] . Turns to the direction of higher concentration are typically preceded by a large spike in sensory perception , corresponding to a head cast in the direction of higher concentration . Given that the direction of a turn ( the alteration in direction between two runs ) is determined by the direction of the head cast preceding the turn , a large spike in sensory perception could act as a signal to transition from head casting back to forward movement , resulting in turns generally being towards the direction of high concentration [19] . More recently a third factor contributing to odour-directed paths in larvae has been described [21 , 22] which has been termed ‘weathervaning’ . During runs , the larva’s path tends to be curved slightly but significantly towards the side of higher odour concentration . This behaviour can still be observed for larvae with single , unilateral olfactory receptors , and has been hypothesised to utilise active sensing of the lateral olfactory gradient through low amplitude head casts during runs [21] . In this paper we use an agent based model to determine if these three control mechanisms—initiating head casting when the odour intensity is decreasing , ending head casting when a sharp increase in odour is experienced , and ‘weathervaning’—can be derived from simple perceptual processing; whether they can replicate fine-grained statistics of larval behaviour; whether they are necessary and/or sufficient to produce chemotaxis; and whether they can be used to provide a low-level account for high-level behavioural descriptions such as preference indices .
We abstract the body of a larva as consisting of two sections of equal length , the head and body , with one articulation between them . The basic larva has two distinct behaviours , runs and head casts ( Fig 1a ) . During a run , the head section moves forward at constant speed vforward . Head orientation remains constant during a run , apart from slight modulation by the weathervane mechanism ( see below ) . The body section is ‘pulled’ behind the head section during runs; when the head body angle is not zero the body section gradually rotates to align with the head section as the larva moves forward . During head casting , the body section remains motionless while the head section rotates from side to side relative to the body section , at speed θ c a s t ′ . Upon reaching the limit of rotation , θmax_head_cast , in one direction , head rotation in the opposite direction begins immediately . Head casting may terminate with the head section oriented differently to the body section; this orientation will determine the direction of the following run , and thus the effective size and direction of turns . All our simulations consist of single larva trials , and we therefore do not consider collisions or interactions between larvae . The intensity of the odour at any location in the simulation is given by a single value; in this paper we consider only single-odour environments ( as used in many behavioural experiments ) . We use both artificial odour gradients ( e . g . a Gaussian distribution of concentration around an odour source ) , and data taken from measurements of real experimental odour landscapes in which larvae have been tested . For a simulated larva in a given landscape we use the odour concentration C at the tip of the head section as the input to the larva’s ‘perception’ . This perceptual value is generally the only information the simulated larva has about the environment , and all behavioural modulation is based on a limited history of this value . Following [19] , we approximate perception with a rule of the form: ϕ = 1 C · d C d t ( 1 ) This rule gives the larva access to a measure of the relative rate of change of the odour concentration . When moving up gradient the perceptual value will be positive , and when moving down gradient the value will be negative . Note that this perceptual processing is deliberately simple , intending to capture the hypothesis that a relative rate-of-change perceptual signal is sufficient to allow for larva-like chemotaxis . In reality there must be some level of odour which falls below the perceptual limits of the animal , and some level that entirely saturates the response , but these effects are not included in the current model . We address the issue of more realistic sensory processing in the discussion . Due to the normalisation in our perception rule , the scale of concentration values in our odour landscapes is arbitrary . Thus for simplicity , unless otherwise noted we normalise the values of all odour environments such that the peak concentration is 1 . Our starting point for behavioural control of the simulated larva is based directly on the hypothesis proposed by [19] , that directed behaviour emerges out of sensory-driven probabilistic transitions between running and head casting that are controlled solely by the recent history of perception ( Fig 1b ) . Specifically , it is assumed that larvae: For our purposes , it is simpler to first define rates of transitions , and convert these into probabilities of transitioning between behaviours on each time step by: p r u n _ t e r m i n a t e ( t ) = r r u n _ t e r m i n a t e ( t ) · d t ( 2 ) p c a s t _ t e r m i n a t e ( t ) = r c a s t _ t e r m i n a t e ( t ) · d t ( 3 ) Note that simulations proceed in discrete timesteps of length dt = 0 . 1s . We can now define our control problem as converting the larva’s perceptual history ( the only information it has available to it ) into these transition rates . We do this by defining a kernel for each transition , and obtaining a rate of transitions by convolving the perceptual history with the appropriate kernel ( i . e . element-wise multiplying and then summing ) . As larvae make transitions from runs to turns even in the absence of odour stimuli , we further include a base rate of making a transition regardless of the perceptual history . r r u n _ t e r m i n a t e ( t ) = r r u n _ t e r m i n a t e_b a s e + ∑ t ′ = 0 t r u n _ k e r n e l ϕ ( t - t ′ ) · k r u n _ t e r m i n a t e ( - t ′ ) ( 4 ) r c a s t _ t e r m i n a t e ( t ) = r c a s t _ t e r m i n a t e _ b a s e + ∑ t ′ = 0 t c a s t _ k e r n e l ϕ ( t - t ′ ) · k c a s t _ t e r m i n a t e ( - t ′ ) ( 5 ) To encode the desired behavioural controls into our model , we use simple linear kernels which resemble the average perceptual history preceding behavioural transitions in real larvae , as reported in [19] . Thus the kernel for run termination takes the form of a gradual negative slope , while the kernel for cast termination is a steeper positive slope with a similar duration to a single head cast ( see depictions in Fig 1c ) . Note that both transition rates are continuously calculated regardless of behavioural state , but only affect behaviour when the larva is in the corresponding state . The output of the perception rule ϕ and the control rules rrun_terminate ( t ) and rcast_terminate ( t ) for a section of simulated larva paths are shown in Fig 1c . Initial simulations using the control outlined above raised a number of issues leading to the following modifications to the control scheme . Twenty parameters need to be set for this model . Some can be taken from available data , but the appropriate values for others were less clear . We discuss here how we chose each of our parameters , with the final values used to generate the results in this paper summarised in table 1 . We take the forward movement speed vforward = 1mm/s from figure 2a in [19] . From analysis of paths of larvae in a no-odour environment we estimate the base rate of transitions from runs to turns at rrun_termination_base = 0 . 148s−1 , that is , turns occur on average every 7 seconds . To determine the corresponding base rate for turn to run transitions , we count the proportion of turns which have a single associated head cast . We assume that this gives the probability of transitioning from head casting to forward crawling behaviour during any given head cast ( making the implicit assumption that the distribution of number of head casts before a turn can be described by a geometric series ) . We then divide this probability by the duration of a head cast in our model to give rcast_termination_base = 2 . 0s−1 , that is , a probability of 0 . 7 of returning to running after a single head cast . Inspecting head casts from the no-odour larva data , we found that over 95% of casts did not exceed 120° , and so we set θmax_head_cast = 120° . The speed of head casts needs to be fast enough to allow up to 4 head casts within a 5 second window ( as seen in the larval data ) , so has been set to allow for a head cast ( out and in ) of maximal size in one second: θ c a s t ′ = 2 * θ m a x _ h e a d _ c a s t . To make turns effective , head casts should only terminate beyond some minimum angle from the centreline . We use the definition of head casts used to define turns in [19] , i . e . θmin_head_cast = 37° . The amplitude of weathervane casts was set to θmax_weathervane_cast = 20° , matching the size of small head casts shown in figure 8c in [21] . Weathervane cast speed was set to a moderate value , θ w e a t h e r v a n e _ c a s t ′ = 60 ° . The final parameters to be set are those defining the lengths and shapes of the kernels . As we only use linear kernels , they can be described with three parameters , the duration and the start and end values . We need a relatively long , negatively sloping kernel for the run termination kernel , and a short , positively sloping kernel for the cast termination kernel . On this basis we found approximate values for the kernels by adjusting until the simulated larva displayed navigation towards the odour source . We then further adjusted kernel parameters by hand until our model matched larval behaviour across a range of behavioural statistics ( see Results ) . This was achieved through gradual adjustment of parameters and visual inspection of resulting behavioural statistics . Automated optimisation of these parameters would have been possible in theory , however , defining a single optimisation criteria when the goal was to match across several distributions would in itself be a subjective process . The kernel parameters chosen are reported in Table 1; these values are used throughout unless otherwise noted . To establish a comparison between our model and experimental results from real larvae , we apply a number of metrics from [19] to paths of wild type larvae and simulated larvae . These metrics are constructed from the trajectories of larvae’s head , centroid and tail positions; note that in this analysis no use is made of the internal state of the model . The metrics used are as follows: We also extract times of turns and head casts from our simulations . These are defined as follows: Events are categorised as follows:
Our aim when picking kernel parameters was to produce a simulated larva which matches the behavioural statistics of real larvae ( n = 42 ) chemotaxing in an odour gradient of ethyl butyrate , as reported in [19] . We produce behavioural statistics for the simulated larva as follows using the same odour distribution as measured in [19] , in a virtual arena of size 65x100mm . Note that the arena size for the larval experiments was 80x120mm , however an estimate of the odour concentration could not be experimentally made at the outer edges . We run 500 simulated larvae in this arena , for 300s of simulated time each . Each simulated larva begins the run with random starting orientation , at a random position within a 12mm square centred on the odour source . The run of any simulated larva which touches the edge of the arena is truncated at that point , consistent with the acquisition of experimental data . From these simulated trajectories , we calculate various behavioural metrics ( as described in the previous section ) . These are used to produce the behavioural statistics shown in Fig 2 . Note that this process was repeated multiple times as we tuned kernel parameters; we show here only the behavioural statistics obtained with our final set of parameters as reported in Table 1 . Fig 2 shows two sample paths , and the match between real and simulated behavioural statistics . In the top panel of statistics it can be seen that the probability of initiating a turn ( an end-run transition ) relative to the odour bearing shows the same form as for the real larva . In the second panel , the probability of making a left turn is altered as expected relative to the bearing of the odour , turning left more often if the odour is on the left . In the third panel , the reorientation rate during runs shows a similar dependence on the bearing angle , and similar amplitude . These comparisons demonstrate that it is possible to choose kernel parameters which result in our model producing very similar behavioural statistics to the larva , on the three metrics which summarise the run termination , cast termination and weathervaning mechanisms . Given that we tuned kernel parameters to match these statistics it is perhaps not surprising that we achieve a close match; nonetheless , it is not trivially obvious that these statistics would be possible to obtain using only linear kernels convolved with the relative rate of change in odour concentration . The lower panels present comparisons of additional behavioural statistics , which show a number of emergent effects also match well between the model and the larva without additional tuning . The run termination mechanism produces distributions of turn initiation bearings and run lengths similar to the larva . Similarly , the head cast termination mechanism produces turn direction probabilities and numbers of pre-turn head casts comparable to the larva . Both simulated and real larvae tend to be headed away from the odour ( >90 degrees ) before a turn , and towards it ( <90 degrees ) after , but with a general undershoot , that is , only a partial correction in orientation . They also both show a similar distribution of bearings that result in correct ( to higher concentration ) rather than incorrect ( to lower concentration ) turns , with wrong turns more likely when heading near to 180 degrees away from the odour , a situation which produces the most ambiguous information during head casting . In the bottom panel , the relative frequency of patterns of head casts towards the direction of higher ( H ) or lower ( L ) concentration is shown . Overall the proportions are similar between the larva and the simulations . The larva and the model both show a bias in the direction of the first head cast , with a majority of head cast groups starting with a cast to high . The mechanism by which the larva creates this bias is not yet understood , however , our model produces a similar bias by simply using the angle of its head at the moment of run termination to determine its initial cast direction . Having set parameters for our model such that it matches the larva on these low level behavioural statistics , we go on to assess the model’s similarity to the larva by comparing its chemotaxis performance to the larva in three different environments . Having tuned our model parameters to qualitatively match low level behavioural statistics of the larva when chemotaxing around a point source of odour , we ask whether this leads to our model quantitatively matching the larva on a higher-level metric , namely the distribution of larvae around the the odour source . Using the data described in the previous section , we computed the distance to the odour source for 42 real and simulated larvae every second for 150s . For comparison , we repeated this process for 19 real and simulated larvae in a ‘no odour’ condition; for the simulated larva this means all behavioural transitions are made at their base rates , with no perceptual modulation . Fig 3 shows sample paths , the temporal evolution of the distance to the source over time , and a snapshot of distances to the source at 120s , for each of these groups . In the absence of an odour source real larvae gradually disperse; the model performs similarly to the larva in this condition . With the odour source present , real and simulated larvae both remain located around the source . We use the performance of simulated larvae at 120s as a quantitative measure to determine how closely our model , with parameters tuned to match larvae’s low level behavioural statistics , matches the larva’s chemotaxis performance . The larva and the model’s distances to the source are not significantly different ( Mann-Whitney U Test , p>0 . 05 ) , although this does not provide evidence for the null hypothesis that the medians of the groups are the same . However , using bootstrapping , we can state with a 95% confidence level that the difference in median distance to the source for real and simulated larvae is between -2 . 4 and 1 . 9mm , around a single larval body length . Having confirmed that our simulated larvae show chemotaxis performance at a similar level to real larvae when circling around the odour source , we consider how well they match the larva’s directness of approach to a distant odour source . For this experiment , we used a second odour gradient of ethyl butyrate ( also from [19] ) , with an odour source centred at one end of a rectangular arena . Simulations were carried out as above , with a different odour landscape and different starting positions; each simulated larva begins the run at a random position within a 20mm square in line with the odour source on the short axis of the arena and 68mm distant on the long axis . Starting orientations were chosen randomly from a distribution of plus or minus 30 degrees relative to the direction of the odour source . Runs were truncated at the point at which they came within 5mm of the odour peak; runs which did not reach this area were discarded . For this condition we compare 43 real and simulated larvae . We calculated distances to the odour source every second as above . Following [23] , we also use a path tortuosity metric to compare the efficiency of orientation in this landscape; a ‘straightness index’ is assigned to each individual by calculating the ratio of the length of its path to the length of the vector travelled . Paths leading directly to the odour peak will have a straightness index of 1 , while paths which follow a less direct route will have a lower straightness index . Fig 4 shows sample paths , the temporal evolution of the distance to the source over time , and boxplots of straightness indices for real and model larvae . We see a good match between the larva and the model’s approach to the odour peak . The larva and the model’s straightness indices are not significantly different ( Mann-Whitney U Test , p>0 . 05 ) , and using bootstrapping , we can state with 95% confidence level that the difference in median straightness index for real and simulated larvae is between -0 . 11 and 0 . 03 . We next consider the behaviour of the simulated larva in linear vs . exponential odour slopes . We compared the model’s paths to paths of wild type larvae in gradients of isoamyl acetate ( landscapes from [16] ) . Runs were truncated at the point at which they came within a 15mm zone at the peak end of the arena; runs which did not reach this area were discarded . Larvae started within a 20mm square in line with the odour source on the short axis of the arena and 80mm distant on the long axis , facing towards the odour peak . We used the same number of simulated larvae as real larvae in each condition: 20 for exponential , 14 for steep linear , and 11 for shallow linear . Initial results suggested that with parameters set as described above ( to match behaviour in a single source environment of ethyl butyrate ) , simulated larvae performed significantly worse than the real larvae in this condition . We therefore also tested whether we could improve the performance of the model by scaling ( i . e . changing the slope of ) the model’s kernels , thereby increasing the strength of the simulated larva’s behavioural biases . We show here results for the model with default parameters , and with a scaling factor of 6 on all kernels . Fig 5 shows sample paths , the temporal evolution of the distance to the source over time , and boxplots of straightness indices for real and model larvae in each environment . These demonstrate that the model ( both with normal and scaled kernels ) can successfully navigate up both exponential and linear gradients . The ‘straightness index’ shows that the larva more directly approaches the odour peak in an exponential gradient than in a shallow linear gradient; this is also true for the model with either normal or scaled kernels . However , both the sample paths and the straightness indices highlight the failure of the model with default parameters; simulated larvae in this case follow paths which are clearly more tortuous than the larva . However , by scaling the kernels by a factor of 6 , we achieve a much closer match between the model and the larva . Why do we need to alter our model’s kernel parameters to replicate larval behaviour in this situation ? One possibility comes from the difference in the odour used in this condition; our model’s parameters were set to match the behaviour of larvae in an ethyl butyrate gradient , while these gradients were produced using isoamyl acetate . It is possible that the difference in odourants leads to an difference in chemotactic performances for the larva . Alternatively , the differences between model and larval performance in these gradients may be a result of our simplified sensory processing ( see discussion ) . In any case , the fact that the model does chemotax successfully in this environment even with default parameters demonstrates its robustness in a novel odour landscape . A common measure of larval behaviour in odour experiments is the ‘preference index’ [12 , 24] . In typical odour-based experiments , a number of larvae are allowed to freely explore a Petri dish ( typically 9cm diameter ) . One or both sides of the dish contain odour sources . At the end of an allotted time period , the number of larvae on each side of the dish ( excluding a 1cm centre region ) are counted . A preference index is then calculated as: P I = # s i d e 1 - # s i d e 2 # t o t a l ( 6 ) Preference indices therefore range from 1 to -1 , with a positive preference index indicating a preference for side 1 , and a negative preference index indicating a preference for side 2 . Typical median preference indices for innate chemotaxis range from 0 to close to 1 , depending on the odour and concentration used ( Schleyer and Reid , pers . comm . ) . Unfortunately , the odour environments used in learning experiments are not as carefully controlled and measured as the data we have used for comparisons so far . Odours are presented in the form of a point source , e . g . in a small cup or on a filter paper . Furthermore , the enclosed nature of the Petri dish is likely to lead to non-uniform distribution of the odour , which may also be changing over time . As there are no detailed recordings of odour gradients in these conditions from which we can draw , we assume a very simple odour distribution; a circular Gaussian ( σ = 30mm ) distribution centred on the odour source . For these trials the simulation arena consists of a circular wall with a radius of 45mm , with an odour source 5mm from the left hand side of the dish . For each odour condition , we ran simulations for 400 individual larvae , each of which was allowed to explore the arena for 5 minutes . Each larva began at a random position on the vertical centre-line of the dish , at a random orientation . At the end of 5 minutes , the position of each larva was recorded . The larvae were split into 20 groups of 20 , and for each of these groups a preference index was calculated . Initial results , using the parameter settings described above , showed extreme PIs; all simulated larvae ended the 5 minute run on the odour side of the arena , i . e . PI = 1 . We therefore proceeded to investigate how scaling ( changing the slope of ) the model’s kernels , thereby reducing strength of behavioural biases and the efficiency of chemotaxis , changes the observed PIs . Fig 6 shows the distribution of simulated larvae and the corresponding PIs for different kernel scaling factors . Also shown is the the effect of the kernel scaling on the statistics for run termination bearing , turn direction probability , and run reorientation in the original point source environment ( as for Fig 2 ) . A scaling of 0 . 1 produces a strong PI score , and a scaling of 0 . 05 a score still comparable to larval experiments . With scaling 0 , the simulated larva has no behavioural biases and indeed ends up equally distributed across the dish , with PI around 0 . Scaling by a negative value produces a negative PIs , that is , apparent repulsion from the odour . We might expect that our model’s original parameter settings , which we have demonstrated produce behavioural biases of the same magnitude as the larva’s , should produce larva-like PIs . However , we had to considerably reduce kernel scaling , and therefore reduce behavioural biases , to produce moderate PIs . There are several possible explanations for this requirement . First , note that our model has access to a clear odour gradient across the whole arena , unperturbed by noise or sensory thresholds; real larvae may not encounter a consistent gradient far from the source . Alternatively , moderate preference indices could coexist with high behavioural biases if only a fraction of the larvae were displaying those biases , or if all larvae were displaying those biases only a fraction of the time . In preference tests of relatively long duration , this does not seem unlikely . In either case , averaging some combination of strongly biased and unbiased behaviour would result in seeing lower overall behavioural biases for larvae in PI-type experiments , as our model suggests . Finally , there may be effects of group assays such as random reorientations caused by collisions . Unfortunately , data to resolve larval behaviour at an individual level during group assays is not available , restricting our ability to differentiate between these explanations . Our model combines three sensorimotor mechanisms that appear to operate in the larva to produce chemotaxis: Finally , we explored the performance of the model in a selection of distinctly different odour landscapes ( linear , Gaussian , and step ) , with different amounts of noise , and with different combinations of the three behavioural biases . This provides a useful parallel to the analysis in [25] for a model of C . elegans . We analyse each larva’s performance with a Chemotaxis Index ( CI ) ; each larva is assigned a CI equal to the proportion of time spent in a region of interest ( ROI ) ; note that due to the differences in ROI areas , direct comparisons of performance between conditions cannot be made . The gradients used were all contained within a 9cm diameter circular arena , with parameters as follows . Linear: Concentration varying linearly from 0 at leftmost edge to 1 at rightmost edge . Region of interest is the rightmost 30mm of dish . Gaussian: Gaussian odour distribution , centred at the centre of the dish , with peak concentration 1 and variance 16mm . Region of interest is a 25mm diameter circle around the centre of the dish . Step: Odour concentration of 0 on left side of the dish and 1 on right , with 5mm wide linear transition of concentration between the halves . Region of interest is the right half of the dish . For each combination of mechanisms ( as in Fig 7 ) we ran 500 simulated larvae in each of the three environments , starting at a random initial position and orientation . This was repeated with multiplicative noise added to the environment by dividing the arena into an 0 . 08mm square grid and multiplying the concentration in each square by a value picked from a normal distribution with mean 1 and variance 0 . 04 ( low noise ) or 0 . 1 ( high noise ) . CIs for each condition are shown in Fig 8 . The results suggest that some mechanisms may make smaller or greater contributions depending on the conditions . For example , including all three mechanisms seems to make a difference for the Gaussian gradient but not the other conditions , where weathervaning makes little contribution . CIs in the linear gradient are more affected by noise than the other conditions . The run termination mechanism seems more effective than the cast termination in the step gradient , and in general the contribution of cast termination is more affected by noise .
Our aim in this paper was to implement a minimally complex model that captures the observed odour taxis behaviour of Drosophila larvae . We show that larva-like behaviour can be achieved using relative change of odour concentration at the tip of the head combined with simple linear kernels to trigger transitions between behavioural states . Tuning the parameters of this model to match the detailed observations of larvae reported in [19 , 21] produces behaviour that also matches the larva on other measures ( such as proportions of head casts , under-correction of the heading angle , behaviour on different odour gradients ) . However , to reproduce typical preference indices reported for en masse assays we needed to alter the parameters to substantial weaken the effects of odour concentration on state transitions . This is discussed further below . Our model combined three mechanisms , acting to modulate the probability of state transitions between running and head casting . In the absence of odour , the simulated larvae exhibit exploratory behaviour , making runs with regular small ‘weathervane’ head casts that are paused on random occasions leading to shallow curves , and also randomly stopping the run and making larger head casts , with random transition back to running , which can produce sharp turns . If this behaviour occurs in an odour gradient , the probability of stopping a run is enhanced for decreases and suppressed for increases in the change in odour concentration . The probability of restarting a run is enhanced by sharp increases in odour concentration during head casting . Small head casts during runs are also paused more often when the experienced odour concentration increase is greater than that already occurring in the run , resulting in a ‘weathervane’ curve towards the odour . From our simulations , it appears that either of the first two mechanisms would be sufficient to get the larva to an odour in a smooth gradient . Combining both with weathervaning produces the best performance; the improvement is most apparent in a Gaussian distribution , which is perhaps closest to the expected gradient for a point source of odour . Run termination is more robust to noise , which might be expected as ( at least in our implementation ) it averages input over a longer time scale than the other mechanisms . The first mechanism ( alteration in the rate of transitioning from running to head casting depending on the change in concentration ) is equivalent to bacterial klinokinesis or the modulation of pirouette frequency observed in C . elegans [26] and is well-known to be sufficient to ascend a gradient . The second mechanism ( ending head casting and resuming running when the head cast produces a sharp change in concentration ) is klinotaxis , and we have shown it is also sufficient for chemotaxis on its own , producing similar performance to pure klinokinesis ( when the latter is tuned to match larval behaviour ) . The combination of these two mechanisms substantially improves the efficiency of odour localisation over either alone . However , pure klinokinesis can potentially produce better chemotaxis if the parameters are tuned to optimise its performance . The final mechanism , weathervaning , is seen to make marginal improvements to chemotaxis performance , although it does not lead to robust chemotaxis without additional biases in the other mechanisms . It has been hypothesised that biasing of larvae’s run curvature is facilitated by low amplitude head casts made while running [21] . Our model concretely implements this hypothesis by including continuous low-amplitude head casting , which is temporarily paused by increases in the perceptual signal . We demonstrate that this mechanism can produce biases in run curvature comparable to that of the real larva , but other mechanisms , such alteration in the size of these casts , are also possible . [19] show that larvae’s first head casts after terminating runs tend to be in the direction of higher odour concentration . We suggest that this is possible due to the larva having information about the lateral gradient from weathervaning during its run , and show that a similar level of first headcast bias can be produced if the simulated larva simply casts in the direction of its current head angle when terminating a run . This interaction of weathervaning and cast direction accords with the observation that the direction of run curvatures and subsequent turns are correlated [21] . Although treated here as a distinct mechanisms , klinotaxis and weathervaning could be interpreted as the same underlying orientation algorithm , i . e . , exploiting the lateral sweep of the head through the gradient to obtain information about the odour direction , and altering the timing or extent of the sweep to orient the animal up the gradient . In the case of C . elegans , their oscillatory forward locomotion naturally produces a substantial sweep ( relative to body length ) and can be altered to produce relatively tight curves . In larvae , the peristaltic propulsion during runs appears to be inconsistent with large head casts , so while biasing the production of small head casts can steer the animal up the gradient , direct approach is only possible by stopping to make larger casts ( or indeed , it may be that making a large cast forces a stop ) . It remains to be discovered how independent are the neural mechanisms underlying these behaviours in the larva . The majority of behavioural experiments on larvae report only preference indices ( PI ) , i . e . , a binary classification of larvae as either within or without a designated region defined in relation to the stimulus of interest . It is important to understand how such global measures relate to the underlying behavioural control if the neural circuits involved in innate and learned sensorimotor control are to be explained . An issue revealed by our analysis is the difficulty of interpreting behavioural statistics that are derived by summing over many individuals , and over relatively long time durations . It was necessary to make each of the behavioural biases around 20 times weaker in a simulated larva to obtain ‘typical’ PIs . The discrepancy between the level of behavioural bias required to match PI data and the low-level behavioural statistics reported in [19 , 21] could have multiple sources: different larvae may have different innate capabilities or preferences for particular odour sources; the attraction of an individual larva to an odour source may change over time due to habituation or changing motivational state or competition from other cues; or the odour gradient itself may vary substantially in the reliability with which it corresponds to the actual odour direction , both over time and space , in a typical Petri-dish experiment . It is clear that this issue can only be resolved by studies that track individuals over time in well-controlled or measurable stimulus conditions . It is important to note that both previous biological experiments [16] and our simulations indicate that the larva can locate odours with a single point sensor on its head , and does not need spatially separated sensors , even for weathervaning . Rather , gradient information is gained through stereotypical movements over time . Nevertheless it is clear that the perceptual response to the odour gradient used in the simulation , which performs perfect differentiation and normalisation , is not realistic . In the majority of the behaviour analysed here the larva spends a large proportion of its time close to the odour peak , and thus in a relatively limited range of concentrations . As such , normalisation should not have a large impact on the model’s global behavioural statistics . However , it is likely to play a significant part in the model’s ability to localise the peak from a distance , by making the simulated larvae unrealistically sensitive to small differences in areas of low odour concentration . In future work , it will be interesting to incorporate more detailed olfactory receptor responses ( such as described in [27] ) into the model , and see how these interact with both different odour gradients and the dynamics of the motor actions to shape the overall behaviour , particularly in relation to approaching an odour from a distance . We have also used a highly simplified model of the larva’s motor system . Although ‘runs’ and ‘head casts’ are reasonable approximations to the main observable actions by the larva , further analysis may reveal important subtleties . For example , the peristaltic pattern that produces the run also imposes a pattern on the sensory input , as the head moves forward and pauses on each cycle , and also changes its orientation with respect to the substrate . Similarly , the rather arbitrary distinction between ‘small’ and ‘large’ head casts used in the simulation may need more detailed representation of the form , size and location of body bends of which the larva is capable . Finally it may be interesting to ask whether a simpler control scheme than the state transitions illustrated in Fig 1b might give rise to qualitatively similar behaviour . It is interesting to note that the mechanisms used to produce the different behavioural transitions in our model are all fundamentally the same , involving differentiation , integration and a non-linear switch , and differ only in their timescales and their weighting of the perceptual signal . Should we assume the current characterisation will map onto distinct ‘decision’ circuits in the animal for changing between runs and head casts ? Or is it possible that these are emergent properties of lower level control that integrate the muscle contractions producing both peristalsis and body bends and modulates them in response to sensory input ? | The larvae of the fruitfly are attracted to many odours . We use a computational model in which simulated larvae stop , start and redirect their crawling behaviour in response to their experience of changes in odour . We show that three simple rules for switching between behaviours are sufficient to produce larva-like results in a simulated agent . | [
"Abstract",
"Introduction",
"Models",
"Results",
"Discussion"
] | [] | 2015 | A Model of Drosophila Larva Chemotaxis |
Schistosomiasis japonica is a parasitic disease that remains endemic in seven provinces in the People’s Republic of China ( P . R . China ) . One of the most important measures in the process of schistosomiasis elimination in P . R . China is control of Oncomelania hupensis , the unique intermediate host snail of Schistosoma japonicum . Compared with plains/swamp and lake regions , the hilly/mountainous regions of schistosomiasis endemic areas are more complicated , which makes the snail survey difficult to conduct precisely and efficiently . There is a pressing call to identify the snail habitats of mountainous regions in an efficient and cost-effective manner . Twelve out of 56 administrative villages distributed with O . hupensis in Eryuan , Yunnan Province , were randomly selected to set up the ecological model . Thirty out of the rest of 78 villages ( villages selected for building model were excluded from the villages for validation ) in Eryuan and 30 out of 89 villages in Midu , Yunnan Province were selected via a chessboard method for model validation , respectively . Nine-year-average Normalized Difference Vegetation Index ( NDVI ) and Land Surface Temperature ( LST ) as well as Digital Elevation Model ( DEM ) covering Eryuan and Midu were extracted from MODIS and ASTER satellite images , respectively . Slope , elevation and the distance from every village to its nearest stream were derived from DEM . Suitable survival environment conditions for snails were defined by comparing historical snail presence data and remote sensing derived images . According to the suitable conditions for snails , environment factors , i . e . NDVI , LST , elevation , slope and the distance from every village to its nearest stream , were integrated into an ecological niche model to predict O . hupensis potential habitats in Eryuan and Midu . The evaluation of the model was assessed by comparing the model prediction and field investigation . Then , the consistency rate of model validation was calculated in Eryuan and Midu Counties , respectively . The final ecological niche model for potential O . hupensis habitats prediction comprised the following environmental factors , namely: NDVI ( ≥ 0 . 446 ) , LST ( ≥ 22 . 70°C ) , elevation ( ≤ 2 , 300 m ) , slope ( ≤ 11° ) and the distance to nearest stream ( ≤ 1 , 000 m ) . The potential O . hupensis habitats in Eryuan distributed in the Lancang River basin and O . hupensis in Midu shows a trend of clustering in the north and spotty distribution in the south . The consistency rates of the ecological niche model in Eryuan and Midu were 76 . 67% and 83 . 33% , respectively . The ecological niche model integrated with NDVI , LST , elevation , slope and distance from every village to its nearest stream adequately predicted the snail habitats in the mountainous regions .
Schistosomiasis japonica , caused by Schistosoma japonicum , is transmitted by the snail intermediate host Oncomelania hupensis [1 , 2] . A reference group from the World Health Organization ( WHO ) reported that the disease burden associated with schistosomiasis was estimated to be 3 . 309 million disability-adjusted life years ( DALYs ) in 2010 [3 , 4 , 5 , 6] . In the People’s Republic of China ( P . R . China ) , after more than 60 years’ effort , schistosomiasis was interrupted in five out of twelve endemic provinces and the infection rates of both human and livestock were reduced to no more than 5% countrywide [7 , 8] . However , the disease is still endemic in seven provinces in lake regions along the Yangtze River and in the mountainous region in western China [9] . In 2010 , the Chinese government launched a schistosomiasis elimination programme with the goal of eliminating schistosomiasis as public health problem at the national level by 2015 [10] . It is widely acknowledged that the frequency and transmission dynamics of S . japonicum is closely related to its unique intermediate host , O . hupensis [11 , 12 , 13 , 14] . To eliminate schistosomiasis , controlling snail populations through mollusciciding was considered as one of the integrated control measures in P . R . China [15 , 16] . However , due to the complicated environmental conditions , snails in mountainous regions are difficult to control completely [17 , 18] . For example , recurrence rates were 6 . 15% ( 16/260 ) in the counties where transmission of schistosomiasis had been interrupted and 32 . 81% ( 21/64 ) in the counties under control from 1999 to 2003 , at the country level of P . R . China [19] . In the mountainous regions of western China , 33 . 33% ( 7/21 ) of counties with transmission control and 4% ( 1/25 ) of counties with transmission interruption in Sichuan Province were confirmed to have local disease transmission again in 2004 [20] . In Chuxiong Autonomous Prefecture of Yunnan Province , the cumulative recurrence of snail infested areas were 1 . 882 km2 from 1994 to 2011 [21] . What’s more , the infection cases of snails are easily missed , so the endemic situation is inevitably underestimated in surveys of mountainous regions [22] . The question of “where snail infestations may be found in a certain mountainous region” is an urgent need in the schistosomiasis elimination stage in order to improve the surveillance and response system locally [23] . To address this question , it would be useful to understand the snail ecology in the mountainous region , leading to the development of a method to detect the habitats of O . hupensis promptly and precisely in a cost-effective manner . Previous studies found that the distribution of O . hupensis was strongly influenced by geographical and environmental characteristics [24 , 25] . Normalized difference vegetation index ( NDVI ) and land surface temperature ( LST ) were considered to be most successful environmental factors for snail habitat prediction [14 , 26] . Another important ecological feature of O . hupensis , as an amphibious snail , is the focal distribution along the water network , such as rivers , streams , etc . [27 , 28 , 29] . This is of concern because surface water serves as the most important indicator of the habitat of the amphibious snail [5] . In particular , water-flow in mountainous region is determined based on the elevation of the environmental settings . Therefore , the digital elevation model ( DEM ) was used to simulate the surface stream network and calculate slope data rapidly and precisely to enable use of important ecological metrics in the prediction of the snail habitats . The ecological niche model is frequently used to predict the geographic distribution of a species [30 , 31 , 32] . For a certain species , it connects the distribution information and related environmental factors to reveal the relationship between them , and then predicts the distribution or potential habitats of the species . It has been widely applied in the research of animal habitat’s predictions , which were proved simple and convenient [14 , 33 , 34 , 35 , 36] . In this study , we aimed to identify O . hupensis habitats in the mountainous regions by ecological niche modeling based on various remote sensing derived data , i . e . NDVI , LST , elevation , slope and distance from every village to its nearest stream , so as to contribute to the development of surveillance tools for the national elimination programme of schistosomiasis and other snail-borne infectious diseases [37] .
Eryuan County is located in Yunnan Province , southwest of P . R . China , extending 25 . 80°-26 . 43° N and 99 . 54°-100 . 34° E . S . japonicum has been endemic there for more than 80 years based on historical records [38 , 39] . Up until 2012 , the historically snail-infested areas covered 43 . 32 km2 [40] . According to the records of snail survey before year 2011 , 56 of 90 administrative villages are historically endemic areas of S . japonicum . In this study , 12 villages with snail presence data were randomly sampled from 56 endemic villages to develop an ecological niche model . The model validation was carried out in 30 villages out of the rest of 78 villages ( 12 villages selected for building model were excluded from the villages for validation ) in Eryuan and 30 out of 89 villages in Midu ( 100 . 32°-100 . 78° E , 24 . 78°-25 . 53° N ) . Midu County is located about 120 km southeast of Eryuan and has similar ecological conditions as Eryuan . Among a total of 89 administrative villages in Midu County , 41 were endemic with schistosomiasis historically . The transmission of schistosomiasis in Midu County was interrupted in 1994 , but the infested areas of O . hupensis rapidly relapsed after 1997 . Up until 2008 , the historically accumulative snail-infested areas covered 24 . 01 km2 in Midu County [41] . All study villages were selected by a chessboard method [42] . Remote sensing images covering Eryuan and Midu were downloaded from the National Aeronautics and Space Administration ( Available at: http://reverb . echo . nasa . gov/ ) . NDVI and LST were retrieved from the Moderate Resolution Imaging Spectroradiometer ( MODIS ) , ranging from July 2002 to July 2011 with a temporal interval of eight days and sixteen days , respectively , and a spatial resolution of one kilometer . The Digital Elevation Model ( DEM ) was retrieved from the Advanced Space borne Thermal Emission and Reflection Radiometer ( ASTER ) with a spatial resolution of 30 meters . Slope , elevation and the distance from each village to its nearest stream were derived from DEM by the hydrological feature-based model in ArcGIS 10 . 0 ( ESRI , Redlands , CA , USA ) . Coordinates and snail presence data of 12 villages for model building in Eryuan County as well as digital administrative boundary data of Eryuan and Midu Counties at a scale of 1: 50 , 000 were collected from the local Schistosomiasis Control Station . All digitized data related to field data were imported into ArcGIS to construct a GIS database . All remote sensing images ( NDVI , LST and DEM ) were pre-processed in ENVI 4 . 7 ( The Environment for Visualizing Images ) , i . e . setting projection , mosaicking and extracting region of interest . The data of 9-year-averaged NDVI and LST for each pixel or cell in remote sensing images of Eryuan and Midu were calculated by the “Band Math” function . Then the remote sensing images of LST and NDVI over nine years were compiled into one image with annual data . The hydrological feature-based model was developed mainly by “Hydrological analysis” in ArcGIS 10 . 0 with the process of flow direction , flow accumulation , and stream net ( Fig 1 ) . The stream net could be generated by the Raster Calculator tool , e . g . “Flow accumulation ≥ 2 , 500” . Cells that have a flow accumulation beyond the threshold value were identified as stream net . In 12 villages for model development in Eryuan County , the range ( difference between maximum and minimum values ) of environmental factors were extracted , namely NDVI , LST , slope , elevation and the distance from every village to its nearest stream and the O . hupensis suitable survival conditions of environment factors for habitats were defined according to ecological knowledge of snails . Based on the suitable ranges of environmental factors for snail habitats , remote sensing images in Eryuan and Midu were extracted using the tools of “reclassification” . After that , a snail potential habitats map for each environmental factor covering Eryuan and Midu Counties was produced . Finally , all these five prediction maps were overlaid together in ArcGIS 10 . 0 , and the prediction potential habitats was extracted by “Map Algebra”-“Raster calculator” [con ( ( NDVI ≥ 0 . 446 ) & ( LST ≥ 2 . 70 ) & ( elevation ≤ 2 , 300 ) & ( slope ≤ 11 ) & ( distance ≤ 1 , 000 ) , 1 ) ] , which meets the requirements for snail survival under five environmental conditions simultaneously . Model validation was carried out by snail survey at 30 administrative villages each in Eryuan and Midu , respectively , which were randomly selected via the chessboard method [43] . In each validation village , 30 sites were chosen along ditches and farmland [44] and snail surveys were performed at 10 m intervals in each snail collecting site , where a square frame measuring 0 . 11 m2 ( 33 . 3 cm × 33 . 3 cm ) was placed . All snails within the frame in 30 collecting sites were collected into envelopes , labeled with location ID , and then recorded the presence situations of snails in each validation village into a table . Comparing the prediction potential habitats map and the field investigation results , the sensitivity , specificity and the consistency rate of the ecological niche model were calculated . The consistency rate was calculated according to eq ( 1 ) : Consistency rate = An+DnAn+Bn+Cn+Dn×100% ( 1 ) Where A denotes the number of validation villages with snail presence in the predicted suitable area , B denotes the number of validation villages with snail presence in the predicted unsuitable area , C denotes the number of validation villages without snail presence in the predicted suitable area , and D denotes the number of validation villages without snail presence in the predicted unsuitable area . Besides , “n” denotes the code of counties ( n = 1 , Eryuan County; n = 2 , Midu County ) . The National Institute of Parasitic Diseases , China CDC ( IPD ) , Eryuan and Midu schistosomiasis control stations facilitated and validated field work methods and results .
The GIS database of the study area was established including remote sensing images , historical snail data , digital administrative boundary files , digitized locations of endemic villages . A total of 108 monthly images and a 9-year-averaged image of NDVI and LST were generated , respectively . As shown in Table 1 , the ranges of environmental factors ( difference between maximum and minimum values ) of the 12 villages were extracted and the snail suitable conditions for habitats were defined according to biological characteristics of snails [1]: ( i ) NDVI in O . hupensis endemic areas was higher than 0 . 446 , which was defined as the survival limit for O . hupensis ( see Fig 2A ) . ( ii ) The lowest limit of LST was defined as 22 . 7°C . Therefore , two potential O . hupensis habitats were detected distributing in the east and west drainage basins ( see Fig 2B ) . ( iii ) The water flow corresponded to two main streams: Miju River and Heihui River , both of which belong to the Lancang River basin . Elevation lower than 2 , 300 m was considered suitable for O . hupensis breeding ( see Fig 2C ) . ( iv ) Slope less than 11° was treated as survival limit of O . hupensis , which mainly distributed in the east part of Eryuan ( see Fig 2D ) . ( v ) Villages infested with O . hupensis were located less than 1 , 000 m away from the stream net ( see Fig 2E ) . Using the five aforementioned environmental indices as the parameters in the development of the ecological niche model simultaneously , the potential O . hupensis habitats in Eryuan and Midu were highlighted in red ( Fig 3 ) . We found that the potential O . hupensis habitats in Eryuan distributed in the Lancang River basin and O . hupensis in Midu showed a trend of clustering in the north and spotty distribution in the south . The validation for the ecological niche model was done in both Eryuan and Midu . As shown in Table 2 , the situation of snail presence in each validation village was recorded and sorted into 4 categories ( An , Bn , Cn and Dn ) . In Eryuan and Midu , according to the field investigation , 23 and 25 validation villages ( An + Dn ) had the same snail presence status with the predictions of the model , respectively . However , the results of field investigation in 7 and 5 validation villages were inconsistent with model prediction in Eryuan and Mindu , respectively ( Bn + Cn ) . In Eryuan , the sensitivity and specificity of the prediction were 76 . 5% ( 13/17 ) and 76 . 9% ( 10/13 ) respectively . By comparison , the two rates in the field survey of Midu were 88 . 9% ( 16/18 ) and 75 . 0% ( 9/12 ) , respectively . The model consistency rates were 76 . 67% and 83 . 33% in Eryuan and Midu , respectively .
This study , innovatively using NDVI , LST , together with DEM , derived hydrological features ( i . e . , slope , elevation and the distance from every village to its nearest stream ) to develop an environment niche model . Together these indices defined the potential habitats of O . hupensis in the mountainous endemic areas of schistosomiasis in China . It is the first time to use this model with the remarkable result that the predicted snail habitats had a good consistency rate of 76 . 67% and 83 . 33% in Eryuan and Midu , respectively . The data collection and preparation is relatively straightforward , can be updated in a timely way and is free of charge , which provides an advantage in the public health applications , particularly in the surveillance of and response to water-associated diseases . Previous studies have proved that the distribution of O . hupensis is related to a close relationship with environment factors such as elevation , bodies of water , vegetation and temperature [24 , 27 , 44 , 45] . A study carried out in lake and marshland regions had found that low elevation was more suitable for survival of the snail than high elevation , which was consistent with our finding that O . hupensis survives below a ceiling elevation of 2 , 300 m [46] . Similarly , Chen found that the distances from 90% of the S . japonicum endemic counties to their nearest rivers are less than 1014 meters , which agreed with the finding of the “1 , 000 m buffer zone” in the present study [47] . However , most of the previous studies weren’t concerned with the multiple environmental factors found in mountainous regions . In this study , we integrated five environmental factors by developing an ecological niche model to predict snail habitats in the mountainous regions and this proved to be a cost-effective approach . Recent georeferenced topographical maps could be used to digitize rivers and other water bodies , but changes in rivers is variable over time related to landscape changes , which prevents recognition of the potential watershed . Further , the climate in the mountainous regions of Yunnan Province is characterized by distinct rainy and dry seasons . The stream sectors with less flow rate may have no water flow during the dry season , but have flowing water in the rainy season . Such stream sectors are always neglected by the published hydrological maps , but they have significant influence on the distribution of O . hupensis . Recently , other researchers used hydrological models to predict mosquito abundance within watersheds or potential resurgence of Schistosoma haematobium [29 , 48 , 49] . Hydrological feature-based modeling is considered to be a powerful tool for determining the potential habitats of snail . In this study , ASTER DEM was not only applied to extract the elevation and the slope but also to calculate the stream nets , which could show the potential water system hub effectively . The real temporary and even the potential streams could be investigated simultaneously with the assistance of different threshold settings . The relationship between the slope of mountainous areas and O . hupensis has been very rarely reported . In this study , we found that slope played an important role in the distribution of O . hupensis . Our field survey found that rice field terraces distributed anywhere in mountainous areas of Eryuan and Midu , were where O . hupensis was always found . The dams of terraces have a high slope value and flow velocity , which does not hold the water , and the flow velocity could be too fast for O . hupensis to maintain their existence in that region . In addition , steep areas are often used to plant economic trees , which impede the breeding of O . hupensis . Due to all these factors , the possibility for O . hupensis breeding decreases as the slope increases . Compared with scenes from Landsat TM or SPOT , NDVI and LST from MODIS have a relatively low resolution , but they are available free and easily accessed . However , the model can be improved in several ways . First , remote sensing is developing rapidly , and use of data with high resolution would be better , especially in “hot spots” of prediction . Second , in consideration of the complicated conditions in mountainous regions , the contribution of different influencing factors to the model can be calculated . Third , one shortcoming of the DEM is that the stream order and the elevation are correlated . For example , low flow orders are more likely to occur in the high areas and vice versa . Further research could be done to explore the prediction of O . hupensis , such as O . hupensis habitats in key irrigation canals and ditches with remote sensing images of higher resolution . Finally , in addition to environmental factors , social factors could be also considered . We concluded that the model presented here can be used to predict potential O . hupensis habitats with a good consistency rate in mountainous regions . The model could become important tools for the prediction of O . hupensis in mountainous areas , particularly in areas where snail survey is a difficult task . We encourage other groups to adopt and further develop our prediction approach in different geographical areas in relation to other neglected tropical diseases to facilitate spatial targeting of controlled interventions in a timely and cost-effective manner . | Schistosomiasis japonica is a parasitic disease caused by the infection of Schistosoma japonicum . Oncomelania hupensis , serving as the unique intermediate host of S . japonicum , has a distribution highly correlated with schistosomiasis epidemic . At present , elimination of O . hupensis is still an important target for disease control in the People’s Republic of China . In mountainous regions , compared with two other endemic regions , snails are hard to detect due to the complicated environmental conditions and poor transportation systems . In this study , we developed an ecological niche model to predict the potential habitats of O . hupensis using remote sensing data including vegetation index , land surface temperature , elevation , slope and the distance from every village to its nearest stream . Validation of the approach was performed in two counties with similar ecological conditions in Yunnan Province , P . R . China . Results revealed a model with a good consistency rate of 76 . 67% and 83 . 33% for the two counties , respectively . The model holds promise for snail surveillance in mountainous regions . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Ecological Model to Predict Potential Habitats of Oncomelania hupensis, the Intermediate Host of Schistosoma japonicum in the Mountainous Regions, China |
Most weakly electric fish navigate and communicate by sensing electric signals generated by their muscle-derived electric organs . Adults of one lineage ( Apteronotidae ) , which discharge their electric organs in excess of 1 kHz , instead have an electric organ derived from the axons of specialized spinal neurons ( electromotorneurons [EMNs] ) . EMNs fire spontaneously and are the fastest-firing neurons known . This biophysically extreme phenotype depends upon a persistent sodium current , the molecular underpinnings of which remain unknown . We show that a skeletal muscle–specific sodium channel gene duplicated in this lineage and , within approximately 2 million years , began expressing in the spinal cord , a novel site of expression for this isoform . Concurrently , amino acid replacements that cause a persistent sodium current accumulated in the regions of the channel underlying inactivation . Therefore , a novel adaptation allowing extreme neuronal firing arose from the duplication , change in expression , and rapid sequence evolution of a muscle-expressing sodium channel gene .
To identify Nav channels expressed in the EMNs , we generated transcriptomes from the posterior spinal cords ( where EMNs are abundant ) and trunk muscle of adults of 3 species of Apteronotids ( A . albifrons , A . leptorhynchus , and Parapteronotus hasemani ) , one myogenic electric gymnotiform ( Eigenmannia virescens ) , and a catfish ( Ictalurus punctatus ) as a nonelectric outgroup . We also obtained expression data from another myogenic gymnotiform , the electric eel ( Electrophorus electricus ) , from a previous study [27] . Data reported here confirm previous results; both scn4a paralogs are expressed in catfish muscle , while only scn4ab is present in the muscle of the myogenic electric fish E . virescens and in E . electricus ( Figs 1 and S2 ) . Unlike myogenic gymnotiforms but like nonelectric teleosts , scn4aa is significantly expressed in the muscle of Apteronotids ( also confirmed here by quantitative PCR [qPCR] , S3 Fig ) . We detected virtually no expression of scn4aa in E . virescens muscle , which confirms previous research that showed that scn4aa is electric organ specific and is not expressed in the muscle of myogenic gymnotiforms [2–4 , 27] . We also discovered a novel gene duplication in A . albifrons: 2 scn4ab paralogs—scn4ab1 and scn4ab2—that are about 95% identical in amino acid sequence , suggesting a recent duplication event . We confirmed that these are not assembly artifacts by cloning them by PCR from muscle mRNA and amplifying them from genomic DNA across several exons ( equivalent to zebrafish scn4ab exons 22–24 ) , with the intervening introns showing even greater divergence ( S4 Fig ) . These 2 paralogs were also found in A . leptorhynchus and P . hasemani . Scn4ab2 showed expression in muscle , and scn4ab1 expressed in muscle and spinal cord . In contrast to other Gymnotiform electric fish species which only express 1 subtype in muscle ( scn4ab ) , these 3 Apteronotid species show significant expression of 3 sodium channels ( scn4aa , scn4ab1 , and scn4ab2 ) . In catfish , E . virescens and E . electricus , spinal cord expression is dominated by scn8ab and , to a lesser degree , scn8aa and scn1lab , as reported for zebrafish [21 , 28] and their orthologs in mammalian spinal motorneurons [29 , 30] . While those genes are also expressed in the Apteronotid spinal cord , scn4ab1 makes up 21%–45% of total sodium channel expression there . The expression of scn4ab1 in A . albifrons spinal cord was confirmed by qPCR ( S3 Fig ) . This is the first observation of a muscle-typical Nav channel gene expressing in spinal cord in any vertebrate . A gene tree inferred from a nucleotide alignment of gymnotiform and non-gymnotiform scn4ab sodium channels with additional sequences from 2 more basal and widely separated Apteronotids ( “A . ” bonapartii , Adontosternarchus devenanzi ) ( S5 Fig ) shows strong support for the scn4ab1 and scn4ab2 paralogs forming a monophyletic clade in a derived lineage within Apteronotidae called the Apteronotini [31] . A time-calibrated phylogeny of the Apteronotidae ( Fig 2A ) estimates the duplication of scn4ab at the divergence of Apteronotini from other Apteronotids at approximately 14 . 5 MYA ( min = 5 . 36 , max = 23 . 66 ) with most amino acid substitutions fixed by approximately 12 . 4 MYA ( min = 3 . 66 , max = 21 . 21 ) , preceding the divergence of species within the Apteronotini . Therefore , the duplication and divergence of this gene likely occurred within an approximate 2 million–year window . Branch-specific nonsynonymous replacements per nonsynonymous site divided by the synonymous changes per synonymous site ( dN/dS ) ratios estimated by maximum likelihood [32] support an episodic burst of positive selection on scn4ab1 immediately after duplication followed by an elevated dN/dS ratio within that clade ( Fig 2B and 2C ) . The other paralog , scn4ab2 , which shows muscle-specific expression like other vertebrate scn4a sodium channels , shows coding sequence patterns consistent with purifying selection , suggesting that this gene maintained its ancestral function along with ancestral expression in muscle . We also estimated dN/dS using phylogenetic analysis by maximum likelihood ( PAML ) [33] . We compared a model in which the root branch of the scn4ab1 clade following the duplication had a unique dN/dS ratio versus a model in which all branches in the tree had the same ratio . A likelihood ratio test supports the more complex model with 2 ratios , which estimates that scn4ab1 evolved by positive selection soon after duplication ( dN/dS = 1 . 58 ) , while the rest of the branches in the tree show more conservative evolution with dN/dS = 0 . 090 ( 2ΔL = 38 . 8 , df = 1 , p < 0 . 0001 ) . We were unable to find statistically significant evidence of evolution by positive selection in the root branch of the scn4ab2 clade using this approach . This analysis further supports scn4ab1 neofunctionalized soon after duplication . Nav channels comprise 4 repeating domains ( D1–D4 ) , each of which has 6 membrane-spanning helices ( S1–S6 ) ( Figs 3A–3C and S6 ) [34] . The S4 helices in each domain are displaced by membrane depolarization , resulting in a conformation change in the channel that allows an inward flow of Na+ ions [34] . Within milliseconds of activation , channel conduction is spontaneously terminated via fast inactivation [34] . The molecular mechanism of this process is unresolved; however , functional experiments have identified several critical molecular features . A hydrophobic triplet of the amino acids isoleucine , phenylalanine , and methionine ( IFM ) in the intracellular loop between D3 and D4 [35] ( the so-called “inactivation particle” ) is required for fast inactivation; this motif likely moves in response to membrane depolarization [36] , and in the inactivated state , it may interact with a “receptor” containing amino acid side chains in the intracellular S4–S5 linkers of D3 [37] and D4 [38] . Naturally occurring mutations in these regions produce an INAP that is implicated in neurological and muscular diseases [39–41] . Fig 3 shows the relative density of amino acid substitutions of the Apteronotid A . albifrons scn4ab paralogs and a distant relative , the scn4ab ortholog , in the zebrafish , D . rerio . Using a minimum mutation parsimony criterion , amino acid changes and indels were mapped onto each of the 3 branches connecting the genes ( Fig 3A ) . For example , if zebrafish scn4ab and A . albifrons scn4ab2 have the same amino acid residue at a coding position and scn4ab1 is different , then an amino acid substitution is assumed to have happened in scn4ab1 at that position after duplication . With this procedure , the density of substitutions was measured along the length of the 3 proteins ( Fig 3B ) . The relative density of amino acid replacements was measured as the ratio of substitution density in an ortholog to the combined density of replacements of the other 2 proteins at the same position ( Fig 3C ) . Simulations in which substitutions were randomly distributed along sequences were used to generate empirical null distributions of branch-specific substitution densities . A threshold that is reached in <99% of simulations was derived for each protein . Every site in the scn4ab2 sequence is well below threshold . However , scn4ab1 greatly exceeds this threshold in 3 locations , each around the S4 and S5 transmembrane regions within domains 1 , 2 , and 4 ( Fig 3C ) . The strongest signal shows a nearly 10-fold higher density of amino acid substitutions than threshold in the aforementioned D4 S4–S5 linker within the putative inactivation particle receptor . These substitutions accumulated within approximately 2 million years of duplication ( Fig 2 ) despite the strong conservation of these sites across the Nav channel gene family of vertebrates , spanning approximately 550 million years of evolution [43] ( S6 Fig ) . While this paper was in preparation , a cryo-electron microscopy ( CryoEM ) structure of the skeletal muscle Nav channel of E . electricus , the electric eel , was reported [42] . This is the first high-resolution structure of a canonical eukaryotic Nav channel . The state of the channel is not conclusively determined , but it may represent a pre-inactivated conformation wherein the pore is trapped open by an unresolved detergent molecule . This structure suggests extensive interactions between the D3–D4 loop and D4 S4–S5 linker ( Fig 3D , green and orange regions ) . Specifically , the structure predicts intimate association of Apteronotini substitutions in the hydrophobic inactivation particle with those within the D4 S4–S5 linker ( Fig 3D inset ) . Of the substitutions in the D4 S4–S5 linker , 3 are located within a leucine zipper-like motif that has been previously proposed to interact with some part of the D3–D4 loop during fast inactivation [38] . The other two , a highly conserved isoleucine/glycine pair ( I1660/G1661 , which are F1660/S1661 in the Apteronotini ) , directly appose the critical phenylalanine ( Phe ) in the IFM inactivation particle ( Fig 3D , inset ) . None of the other spinal cord–expressing Nav channels in the Apteronotini ( scna8aa , scn8ab , scn1Lab ) have amino acid substitutions in these parts of the channel . Because the mechanism of inactivation and its structural determinants are highly conserved among animal Nav channels , we made use of an available human Nav1 . 5 sodium channel expression construct , engineered the amino acid substitutions into this channel , and expressed these channels in Xenopus oocytes . We predicted that the Apteronotini substitutions would generate an INAP . Wild-type ( WT ) hNav1 . 5 channels had no apparent INAP ( Fig 4A ) . Compared to vertebrates in general , Apteronotid scn4ab ( protein name; Nav1 . 4b ) channels have substitutions in the inactivation particle ( IFM → LFL ) , with that of the Apteronotini scn4ab1 having 1 further substitution ( DIFM → HLFL ) . None of these substitutions generated an INAP on their own ( Fig 4C , blue and orange traces ) . The substitution of the complete set of the 5 amino acids observed in the Apteronotini scn4ab1 channel D4 S4–S5 linker into the WT hNav1 . 5 linker ( Fig 3 ) caused a large INAP ( Fig 4B ) . At none of these positions do single alanine mutations generate persistent currents in the neuronal sodium channel Nav1 . 2 [38] . Therefore , the specific chemical nature and/or combination of mutations in this domain of scn4ab1 are necessary for the physiological persistent current . The addition of the partial ( LFL , found in all Apteronotids ) or complete ( HLFL , found in Apteronotini ) inactivation particle substitutions to the Apteronotini D4 S4–S5 linker substitutions modified the amount of INAP such that the combination that most resembled the Apteronotid scn4ab1 channel generated about 6% maximum INAP ( at −35 mV ) . Importantly , INAP activated at more hyperpolarized voltages ( approximately −60 mV ) and was a larger relative percentage of the total current ( S7 Fig ) in the range from −60 to −40 mV . The channel also showed less steady-state inactivation than the WT hNav1 . 5 current in this voltage range ( S8 Fig ) . Electrophysiological recordings of spontaneously firing Apteronotid EMNs show that these neurons never completely return to resting potential between action potentials but , instead , briefly reach −40 to −50 mV before the next action potential [19] . This is the voltage range in which INAP produced by the inactivation apparatus of the Apteronotini scn4ab1 channels is maximal . Furthermore , the recovery from inactivation , which influences a neuron’s maximum firing frequency , is faster in hNav1 . 5 ( HLFL + D4 S4–S5 ) than the WT hNav1 . 5 , but only at the earliest ( submillisecond ) recovery interval ( S9 Fig ) .
Just as the diversification of voltage-gated ion channels contributed to diversification of animal nervous systems , the evolution of novelty in the Nav channel gene family led to diversification of electric organ signals [24 , 26 , 44] . In both major clades of electric fish ( Gymnotiformes , Mormyroidea ) , a muscle-specific Nav channel gene ( scn4a ) that duplicated in the whole-genome duplication at the origin of teleosts retained muscle expression for approximately 100 million years before it was convergently recruited into the electric organ at the origin of both groups of electric fish . A clade of electric fish within the South American Gymnotiformes exhibits another more recent duplication at one of the muscle-expressing Nav channel loci; this time , scn4ab duplicated to scn4ab1 and scn4ab2 . This lineage of electric fish is specialized for high-frequency discharges from a unique electric organ composed of the axons of EMNs . Shortly after duplication , scn4ab1 gained expression in the spinal cord , where the EMN electric organ is located , and it shows comparative sequence patterns consistent with having evolved under positive selection ( Fig 2 ) . The EMNs of the neurogenic electric organ of some Apteronotids , including those that contain scn4ab1 , fire spontaneously and at the fastest rate known in any animal neuron . Our molecular evolution ( Fig 3 ) and biophysical ( Fig 4 ) analyses suggest that several amino acid substitutions within the inactivation particle and particle receptor lead to the generation of INAP in scn4ab1 . The magnitude of INAP we measured in hNav1 . 5—that was specifically attributable to the Apteronotini substitutions—was similar in magnitude to the INAP observed in spontaneously firing mammalian Purkinje neurons , wherein INAP equal to only a small percentage of the peak current [45 , 46] is sufficient to drive spontaneous firing [47 , 48] . Because pharmacological and electrophysiological data from Apteronotus EMNs indicate the presence of a persistent sodium current , this strongly supports the origination of scn4ab1 as an important molecular contributor to the evolution of EMN spontaneous high-frequency electric signals in the Apteronotini [3] . It is , however , important to note that the INAP measured in spontaneously firing Purkinje neurons is elicited by the interaction of the pore-forming Nav subunit with an auxiliary subunit protein [49] . We propose that the Apteronotid EMNs use a distinct mechanism , wherein substitutions within scn4ab1 are sufficient to generate INAP . The inactivation particle and receptor evolved on different evolutionary timescales ( Fig 3E ) . The substitutions observed within the putative scn4ab1 inactivation receptor ( D4 S4–S5 linker ) happened soon after the duplication of scn4ab; when substituted into the human Nav1 . 5 channel , they cause a large INAP ( Fig 4C left ) . The inactivation particle , however , evolved over a longer timescale , with 2 of the 3 substitutions preceding the duplication . One of the amino acid substitutions ( IFM to LFM ) occurred before the evolution of Gymnotiformes . The basal Apteronotid then gained another substitution ( LFM to LFL ) . When substituted alone into human Nav1 . 5 , these substitutions have no discernable impact on inactivation . When coupled with the 5 mutations in the putative receptor , however , they have a moderate mitigation of the persistent current ( Fig 4C middle ) . This suggests that the initial substitutions in the gating particle may have been benign to start but later interacted with and possibly facilitated substitutions in the receptor when scn4ab duplicated . One more substitution found near the canonical inactivation particle , DLFL to HLFL , occurred around the same time as the mutations found in the receptor . The addition of the HLFL-to-DLFL substitution further mitigates the persistent current ( Fig 4C right ) . The central F in the IFM particle remains conserved , which is consistent with data indicating that it has a much larger impact on inactivation than its flanking amino acids [35] . The impact of mutations from different evolutionary time periods suggests that scn4ab1 arose and played a role in the electric organ rather than a spinal cord Nav channel because its inactivation particle had acquired mutations that dampened the impact of receptor mutations on the electric organ . Despite the extensive changes in the 2 interacting domains , our experiments indicate that the functional interaction between inactivation particle and receptor is maintained in scn4ab1 , suggesting great evolutionary lability post duplication in these usually highly conserved parts of the channel . The gain of expression of scn4ab1 in the spinal cord may have caused or was caused by the changes in the inactivation gate specific to the Apteronotini . Alternatively , the gain in expression might have occurred because of the duplication itself . Genomic and transcriptomic investigations into more basal Apteronotini species will illuminate the coevolution of expression and the inactivation machinery in this gene . One puzzling observation in our data is that , despite scn4ab1 displaying an INAP , this gene still maintains abundant expression in the skeletal muscle , where presumably a persistent current would be disruptive to swimming behavior . Mutations in the scn4a gene in humans that results in INAP in muscles cause muscle diseases [50] . A critical question is this: what prevents an INAP and spontaneous firing in muscle ? There are many posttranscriptional mechanisms for compensating for this expression . INAP may be inhibited by other muscle-expressing Nav channel–associated proteins , known as beta subunits [3 , 4] ( which differ in their expression in the Apteronotid muscle and spinal cord [S10 Fig] ) . Translation of scn4ab1 may be suppressed by muscle-specific micro-RNAs ( miRNAs ) , or there may exist compensatory increases in hyperpolarizing ionic currents that are specific to the Apteronotini . Finally , it is possible that there is an INAP in Apteronotini muscle and that the current is part of the unique swimming patterns exhibited by these fish . Most electric fish species , and especially those that maintain high-frequency signaling , remain rigid during swimming and rely on the undulation of a long , derived anal fin . Perhaps a persistent current in these fishes’ muscle contributes to their muscle rigidity . These explanations are highly speculative at this time but may inspire further study . While the duplication of an scn4ab gene and its evolution likely underlies the ability of the Apteronotini to spontaneously fire their electric organs at high frequencies , some other Apteronotid species generate high-frequency electric organ signals often exceeding 1 , 000 Hz , although whether they fire spontaneously is unknown . It is also possible that another spinal cord–expressing channel ( e . g . , scn8ab ) has evolved the ability to support high-frequency firing in other species . Exploration of the sodium channel gene families of other Apteronotids will likely be a rich vein for future research . A complete understanding of EMN firing rates in the Apteronotini must also consider the activities of the other Nav channels expressed in EMNs , such as scn8aa and scn8ab , as well as the voltage-gated potassium channels expressed in these neurons . It will be interesting to know whether any of these other channels have also evolved mechanisms for rapid spontaneous firing and/or whether there are other examples of “cryptic duplication” in these channels . Identifying such highly similar paralogs as scn4ab1 and scn4ab2 requires a level of granularity in transcriptomic analysis that has historically been difficult to achieve . In this study , we have presented evidence that the continued divergence and diversification of electric organ signals is driven in part by the repeated duplication and neofunctionalization of a Nav channel gene . In 3 separate fish lineages , a duplicate originating from a muscle-type sodium channel was co-opted by a novel electric organ , twice by a muscle-derived organ and once by a neurogenic organ . It is surprising to see scn4a-type sodium channels involved in neurogenic electric organs rather than one of the sodium channel types normally expressed in spinal neurons , namely scn8ab . The recurring neofunctionalization of scn4a genes suggests that this type of sodium channel may be relatively free of selective constraints to evolve novel expression and structural innovations [22] . Phylogenetic placement of the many mutations involved in scn4ab1 inactivation elicits an interesting hypothesis that perhaps some preadaptive mutations in the inactivation machinery primed this channel for evolution in Apteronotid EMNs . Perhaps the genetic substrate for new adaptive phenotypes will be found in gene types and families not necessarily expressed in the tissues from which novel phenotypes are derived but more from types of genes whose duplication and neofunctionalization are less detrimental to the organism . Further research may find that the scn4a-type sodium channel displays this remarkable tendency to contribute to electric fish evolution because , compared to other sodium channels , it is the least disruptive to the organism when it duplicates and neofunctionalizes .
Animals were acquired through the aquarium trade . Fish were euthanized according to ethical guidelines set by IACUC at UT Austin and Indiana University Bloomington . Skeletal muscle samples were taken from the midtrunk hypaxial location and the spinal cord from the mid to tail location of 3 adult A . albifrons , I . punctatus , and E . virescens , as well as 1 adult A . leptorhynchus and P . hasemanii . Tissue samples were flash frozen in liquid nitrogen , and total RNA was extracted and DNA removed , following previously described protocols [23] . Total RNA samples were submitted to the University of Texas at Austin core genomics facility , where ribosomal RNA was removed; paired-end 100 bp RNA-seq ( A . albifrons and I . punctatus ) or paired-end 150 bp RNA-seq ( E . virescens , A . leptrohynchus , and P . hasemanii ) was performed on an Illumina HiSeq 2000 to produce between 30 and 34 million reads per sample . Raw reads were quality filtered and adapter trimmed with Trimmomatic v 0 . 35 . Muscle and spinal cord reads were combined and transcripts assembled from a single biological replicate from each species using Trinity [51] v . 2 . 0 . 6 . Alpha and beta sodium channel transfrags were extracted and annotated using reciprocal blast . All 8 sodium channel alpha subunits and 5 beta subunits from the D . rerio were downloaded from GenBank and used to blast sodium channel sequence from each of the transcriptomes with e-value threshold e-6 . This yielded 38–71 ( alpha ) and 17–45 ( beta ) transfrags per species . CDS’s of transfrags were found using Transdecoder v . 3 . 0 . 0 with minimum CDS of 100 amino acids . CD-hit-est v . 4 . 6 . 4 was used to cluster all coding sequences with >99% sequence similarity and >90% of the sequence overlaps with the longest sequence in each cluster . This resulted in 12–25 transfrags for the alpha subunits and 7–9 for the beta subunits in each species . These sequences were then blasted against the D . rerio proteome ( assembly GRCz10 ) with blastx ( BLAST+ V 2 . 2 . 28 ) where the top hit was designated the transfrag’s ortholog . The quality of transfrags was visually inspected by mapping reads ( RSEM; see below ) and using Integrative Genome Browser v 1 . 3 . 1 to inspect patterns among mapped reads . The quality was generally good , except for the scn4ab transfrags in the Apteronotid species . Reads from muscle showed high-frequency base mismatches distributed uniformly along the length of the transfrags . Inspection of the reads showed that all high-frequency mismatched bases existed within the same reads . This same pattern occurred in all 3 Apteronotids investigated . However , in all 3 species , these “polymorphisms” did not appear in mapped reads from the spinal cord samples , suggesting the presence of a duplicate of scn4ab that did not assemble well and is only expressed in muscle . The facts that more than half of the gene sequence was assembled in each species and that there were very few polymorphisms of reads mapping in the spinal cord together indicated that , while one paralog did not assemble very well , the other did and contained little or no chimeric sequence with its paralog . There were small fragments ( <1 kb ) that corresponded to the second duplicate in some of the Apteronotid species . To better assemble this gene , we extracted all reads that mapped to scn4ab in muscle with at least 2 mismatches and used SOAPdenovo-Trans [52] with kmer size of 55 to assemble large transfrags of this gene . After adding this gene to the transcriptome and remapping with RSEM , the polymorphisms largely disappeared from scn4ab in skeletal muscle in each species . The presence of duplicate scn4ab genes was confirmed with PCR and Sanger sequencing of DNA samples from A . albifrons . All sodium channel transfrags were extracted from the transcriptome and analyzed with RSEM [53] v . 1 . 2 . 28 to estimate the relative expression in transcripts per million ( TPM ) for each transfrag . To estimate gene-level expression , alternative splice forms as estimated by Trinity ( _i notation in Trinity V . 1 . 2 . 28 notation ) were summed for each gene ( _g in Trinity notation ) . When multiple loci ( _g ) were annotated as orthologs of the same gene , they were assumed to be fragments of the same locus at different , mostly nonoverlapping locations in the gene . Expression levels were alternatively estimated as the averaged TPM or maximum TPM across all these transfrags for each gene in each replicate . These 2 approaches yielded very similar relative expression patterns to the averaging approach just described ( S2 Fig ) . E . electricus expression was in units of FPKM . To make relative expression comparable between species , TPM or FPKM values were scaled by total sodium channel expression in each tissue . Three muscle and spinal cord total RNA samples from A . albifrons were reverse transcribed to cDNA with random hexamers and oligo-dT20 primers using the superscript III kit and standard protocol ( Life Technologies , Grand Island , NY ) . Primers and hydrolysis probes ( IDT , Coralville , IA ) specific to scn4ab1 , scn4aa , and the housekeeping gene RPL13a ( which was sequenced using degenerate primers , PCR , and Sanger sequencing ) were used with the TaqMan Universal Master Mix NO UNG ( Applied Biosystems , Branchburg , NJ ) to perform qPCR reactions in the 2 tissues of 3 biological replicates . Specificity of primers and probes was confirmed through PCR and Sanger sequencing . qPCR amplicons were designed to span multiple exons , and negative controls were performed where the reverse transcriptase was not added to the reaction mix . qPCR reactions were run on a Viia7 Real-Time PCR machine ( Applied Biosystems ) . RPL13a normalized expression levels were estimated for scn4aa and scn4ab1 in muscle and spinal cord using previously published protocol [23] . In brief , the raw amplification data were baseline corrected , and linear regression on the log-linear phase of amplification for each individual reaction was used to verify close to 100% doubling efficiency and select a common threshold for each sodium channel gene and the housekeeping gene . Cq and doubling efficiency was estimated and confirmed using the LinRegPCR software package [54–56] . Normalized expression was estimated using 2-ΔΔCq method [57] , which assumes 100% doubling efficiency . Scn4ab sequences were trimmed to their longest coding sequence with stop codons removed for codon alignment . A codon alignment was created on the GUIDANCE2 server . A GUIDANCE alignment quality threshold of 0 . 93 was selected , which removed 20% of the codon positions . We used MRBayes version 3 . 2 . 5 [58] to estimate the gene tree topology . We estimated the gene tree by model averaging over the space of all possible GTR models with gamma-distributed rates and data partitioned by the 3 codon positions ( lset nst = mixed rates = gamma; mcmc ngen = 1 , 000 , 000 ) . This analysis produced a consensus tree that was identical in topology and nearly identical in branch lengths to a tree generated without partitioning the codon positions . The consensus tree was very well resolved . All branches had posterior probability >0 . 99 except one , which had posterior probability of 0 . 75 . Two independent runs , each with 4 chains , converged for 2 million iterations after 1 million burn-in on identical marginal posterior distributions for all model parameters . Posterior distributions were found to be of good quality when visualized with Tracer . Maximum likelihood estimates of branch dN/dS were estimated for the majority rule consensus tree generated from unpartitioned codon alignment using the BSR method [32] on the Datamonkey server . Sequences used in this analysis are given in S1 Table . Maximum likelihood dN/dS was also estimated using PAML [33] version 4 . 9e with F3X4 for codon frequencies . The following two different models were evaluated: ( 1 ) all branches share the same dN/dS ratio and ( 2 ) the root branch of the scn4ab1 clade having a unique ratio . Models were compared by the likelihood ratio test . To locate regions where the duplicate sodium channels may have diverged in function , the density of amino acid substitutions along the peptide sequence was analyzed . The amino acid sequences of scn4ab1 and scn4ab2 were most complete in A . albifrons . These sequences were aligned against scn4ab of D . rerio . Substitutions or indels were mapped to each of the 3 branches of the resultant tree by a minimum mutation parsimony criterion . At positions where all 3 sequences differed , a substitution was assumed to have occurred on each of the 3 branches . This slightly inflates the density of substitutions at these regions for the whole tree but does not impact investigation of the relative density of amino acid substitutions . For each of the branches connecting the 3 genes to the internal node , the density of substitutions at each residue was calculated using the R function density ( package stats ) with binwidth set to 15 and then weighted by that gene’s share of the total number of substitutions inferred by parsimony . The relative density at position x on sequence y was calculated as the density of position x on sequence y divided by the sum of the density of the other 2 sequences at the same position . For example , if scn4ab1 has a relative density of 2 at a position in the sequence , then the density of amino acid substitutions is twice that of the other 2 sequences combined . We generated 10 , 000 simulations in which the number of amino acid substitutions mapped to each branch were randomly distributed along each of the 3 sequences , and the same relative density statistic was measured to get a NULL distribution for relative density and to determine if the clustering pattern of amino acid substitutions observed is unlikely to have arisen by chance . The dataset included sequences from 37 taxa for 3 genes: cytochrome c oxidase subunit I ( COI ) , cytochrome B ( CytB ) , and recombination activating gene 2 ( RAG2 ) . Most sequences were derived from concatenated consensus data ( see S2 Table of the manuscript for a full list of individuals , source IDs , and accession numbers ) , with supplemental Apteronotid sequences coming from [31] and outgroup sequences taken from NCBI ( accession numbers for supplemental Apteronotid and outgroup sequences listed in S2–S3 Tables ) . Sequences were concatenated and aligned using the L-INS-I protocol in MAFFT [59] . Maximum likelihood trees were estimated using a GTR + gamma substitution model in RAxML v . 7 . 4 . 2 [60] with D . rerio as the outgroup . Dating estimates were performed using the RelTime method [61] and a Tamura-Nei model [62] in the MEGA7 software package [63] . The model included 5 discrete gamma estimates , allowed for invariant sites , included all codon positions , and discarded alignment gaps . The tree was calibrated using 2 dating ranges estimated from [64]: ( i ) the split between I . punctatus and Serrasalmus maculatus was assigned a potential range of 115–150 MYA , and ( ii ) the split between S . maculatus and Orthosternarus tamandua was assigned a potential range of 75–115 MYA . Two electrode voltage clamp recordings in Xenopus oocytes were performed at 20–22°C using a Turbotec 03X amplifier ( NPI ) . Intracellular recording electrodes had resistances of 0 . 2–0 . 4 MΩ when backfilled with 3 M KCl . Voltage protocols are described in detail in the figure legends . For all experiments , the holding potential was −120 mV . Mutations were made into human cardiac sodium channel Nav1 . 5 in pcDNA 3 . 1 using mini-gene synthesis ( Biobasic , Markham , Ontario , Canada ) in conjunction with Gibson Assembly . All constructs were verified by sequencing through the entire open reading frame . RNA was transcribed using the Mmessage Mmachine T7 Ultra Kit ( Thermofisher , Waltham , MA ) after linearization with Not1 . Approximately 12 . 5–50 ng of cRNA from the alpha subunit was co-injected with 6 . 25–25 ng of cRNA from the β1 auxillary subunit , which is expressed ubiquitously in excitable cells and enhances sodium channel surface expression [65 , 66] . Recordings were conducted 24 to 48 hours later in oocyte Ringer’s solution containing ( in mM ) : 116 NaCl , 2 KCl , 1 . 8 CaCl2 , 2 MgCl2 , 5 mM HEPES , pH 7 . 4 . Traces were acquired at 50 Khz and filtered at 10 Khz for display in the figures . For calculation of persistent current , pClamp 9 . 2 was used to average the last 0 . 5 ms of each 30-ms pulse , and then this value was divided by the peak of the transient component , after linear leak subtraction . All statistical comparisons were by unpaired Student t test with two-tailed distribution . | The electrical properties of excitable cells , such as those in muscle and nervous tissue , were enabled in large part by the evolution of voltage-gated ion channel genes . The regulated conduction of ions through these channels results in the propagation of electrical signals , facilitating communication between cells . Here , we investigated how voltage-gated sodium ( Nav ) channels contributed to the evolution of a novel electric organ system in the Apteronotids—a lineage of weakly electric fish . This organ is developmentally derived from motor neurons and used for communication between individual fish , as well as for probing their nocturnal environment . We used transcriptomic data to show that the gene encoding a broadly conserved muscle-specific sodium channel was duplicated in an ancestral fish . One duplicated gene copy subsequently gained expression in the spinal cord , where the electric organ is located . Through evolutionary analysis and biophysical experiments , we demonstrate that sequence changes in this new sodium channel transformed its function to cause novel electrical properties that can facilitate spontaneous high-frequency action potentials . This study shows that duplicate genes can gain highly novel expression patterns and quickly adapt to contribute to the phenotypic evolution of novel organ systems . | [
"Abstract",
"Results",
"Discussion",
"Methods"
] | [
"taxonomy",
"medicine",
"and",
"health",
"sciences",
"fish",
"chemical",
"compounds",
"nervous",
"system",
"vertebrates",
"electrophysiology",
"neuroscience",
"organic",
"compounds",
"animals",
"animal",
"models",
"osteichthyes",
"phylogenetics",
"ion",
"channels",
"data",
"management",
"model",
"organisms",
"amino",
"acid",
"substitution",
"phylogenetic",
"analysis",
"experimental",
"organism",
"systems",
"amino",
"acids",
"research",
"and",
"analysis",
"methods",
"sodium",
"channels",
"sequence",
"analysis",
"spinal",
"cord",
"computer",
"and",
"information",
"sciences",
"sequence",
"alignment",
"animal",
"cells",
"proteins",
"bioinformatics",
"chemistry",
"short",
"reports",
"biophysics",
"evolutionary",
"systematics",
"evolutionary",
"genetics",
"physics",
"biochemistry",
"zebrafish",
"cellular",
"neuroscience",
"eukaryota",
"neuroanatomy",
"organic",
"chemistry",
"anatomy",
"cell",
"biology",
"physiology",
"neurons",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"evolutionary",
"biology",
"neurophysiology",
"organisms"
] | 2018 | Rapid evolution of a voltage-gated sodium channel gene in a lineage of electric fish leads to a persistent sodium current |
The explosion of bioinformatics technologies in the form of next generation sequencing ( NGS ) has facilitated a massive influx of genomics data in the form of short reads . Short read mapping is therefore a fundamental component of next generation sequencing pipelines which routinely match these short reads against reference genomes for contig assembly . However , such techniques have seldom been applied to microbial marker gene sequencing studies , which have mostly relied on novel heuristic approaches . We propose NINJA Is Not Just Another OTU-Picking Solution ( NINJA-OPS , or NINJA for short ) , a fast and highly accurate novel method enabling reference-based marker gene matching ( picking Operational Taxonomic Units , or OTUs ) . NINJA takes advantage of the Burrows-Wheeler ( BW ) alignment using an artificial reference chromosome composed of concatenated reference sequences , the “concatesome , ” as the BW input . Other features include automatic support for paired-end reads with arbitrary insert sizes . NINJA is also free and open source and implements several pre-filtering methods that elicit substantial speedup when coupled with existing tools . We applied NINJA to several published microbiome studies , obtaining accuracy similar to or better than previous reference-based OTU-picking methods while achieving an order of magnitude or more speedup and using a fraction of the memory footprint . NINJA is a complete pipeline that takes a FASTA-formatted input file and outputs a QIIME-formatted taxonomy-annotated BIOM file for an entire MiSeq run of human gut microbiome 16S genes in under 10 minutes on a dual-core laptop .
The advent of next-generation sequencing technologies , combined with major advances in molecular and bioinformatics techniques , have enabled rapid growth in the culture-independent sequencing of amplified marker genes ( amplicons ) from environmental microbial communities . The major benefit of amplicon sequencing is that it allows reasonable resolution of taxonomic composition in these communities at a fraction of the cost of deep metagenomic sequencing . Once these sequences are generated , a common analysis approach is to bin them by sequence identity into operational taxonomic units ( OTUs ) [1–4] . For environments containing a large fraction of novel taxa , one must rely on unsupervised ( “de novo” ) clustering of amplicons to convert the raw reads to features representing organisms belonging to distinct evolutionary clades . On the other hand , in habitats with mostly well-characterized microbes , we have the option of matching the generated amplicon sequences to reference databases containing example marker genes from known taxa [5] . A hybrid approach may also be used , where sequences are first compared to a reference database , with subsequent de novo clustering of those that failed to match . As the number of published culture-independent amplicon-based surveys of microbial communities continues to grow , our ability to rely on reference sequences also increases . However , although the crucial analysis step of mapping generated amplicons to reference marker genes has received much attention from the microbial bioinformatics field , with a variety of solutions proposed [6–10] , there is much room for improvement in terms of speed , accuracy , memory footprint , and openness of code . NINJA-OPS , our portable , open-source OTU picking pipeline , realizes these goals . Originally conceived as a means to make data more compressible , the Burrows-Wheeler transform ( BWT ) [11] is a lossless , reversible transformation that effectively positions series of like characters close to each other in a way that can easily be undone to recover the original data . It involves creating a circular suffix array , sorting the final column lexicographically , and storing that column as the transformed data for later compression . This algorithm also has the interesting property of enabling rapid substring search , with O ( 1 ) order of growth in finding exact string matches . As long as there is an efficient indexing scheme that stores the indices of the transformed bases into the original string , the BWT can be used for fast database substring search amounting to binary searching ( or looking up via rank matrix ) the transformed reference string representation and mapping back to the original , and has hence been employed in a number of commonly used DNA alignment tools [11–14] . Although these tools are approximate methods due to the high additional computational cost of performing optimal local or global alignment search when mismatches occur , they are generally fast and widely used in the genome-enabled research community ( http://bowtie-bio . sourceforge . net/bowtie2/other_tools . shtml ) . Here we demonstrate that BWT-enabled DNA alignment can be effectively used for accurate and fast assignment of marker-gene sequences to a reference database . We present the NINJA-OPS pipeline utilizing several novel contributions to achieve an order of magnitude speedup and higher accuracy when compared to commonly used approaches ( or up to two orders of magnitude when combined with denoising ) . To test the accuracy and efficacy of our approach , we perform closed-reference OTU-picking on a wide range of biological data sets from varied environments . Accuracy was evaluated using an optimal aligner which produces a BLAST-style %ID for each query sequence against the reference sequence chosen by the OTU picking method . Speed was assessed as the elapsed time from parsing the correctly-formatted input FASTA file , which is accelerated by NINJA’s fast C parser and simple format requirements , until the alignment ( against a pre-generated database ) has terminated . However , it may be useful to note that NINJA also significantly speeds up the subsequent steps of tallying reads , incorporating taxonomic annotations , and producing an OTU table in sparse BIOM 1 . 0 format , as well as other steps prior to the alignment such as reverse complementing and trimming reads . Hence , the NINJA pipeline accelerates many stages of the OTU-picking pipeline in addition to the alignment step .
The pipeline follows three stages: filtering , aligning , and parsing . After forming the concatenated reference string , called the “concatesome , ” from the individual references , NINJA applies a powerful filtering step which uses a 3-way radix quicksort on string pointers to rapidly de-duplicate millions of reads , construct a sample dictionary , and output a reduced-size filtered FASTA file and sample dictionary ( Fig 1 ) . The program implements this lossless filtering approach as well as a lossy variant , making use of singleton filtering as well as statistical profiling over the entire set of reads to exclude reads with a user-defined number of duplicates or rare segments ( k-mers ) appearing below a user-defined threshold of prevalence . The lossy filtering , which is not enabled by default , is intended to identify reads with probable read error independently , and speeds up the resulting alignment by excluding such reads from the BWT aligner . This adds an additional speedup because BWT string matching spends a disproportionate amount of search effort to align erroneous or low-identity reads . Because choice of k ( from 8 to 14 ) and prevalence threshold are highly domain- and dataset-specific , it is difficult to issue a general recommendation for this setting . Although we have found k = 8 to k = 14 at a threshold of 0 . 05%-0 . 01% to be a safe minimum for 16S data , user experimentation is recommended . The NINJA filter step also performs reverse-complementing and sequence length trimming at the same time as the other filtering steps . Because of this simultaneous multi-step filtering , no intermediate files are created prior to the alignment stage , and all filtering steps are performed rapidly in optimized C code on data structures already in memory . This takes a fraction of the time used by other filtering pipelines which perform sequential operations often written in general-purpose scripting languages and generate numerous large intermediate files after each step . Using the base NINJA filter parameters , the entire filtering process itself takes approximately 10–20% of the time it takes to align the resulting filtered file when using all optional filtering steps . Next , the filtered reads are aligned against a reference database containing the ( the concatesome ) via any BWT-derived short read aligner such as BWA[11] , Bowtie[13]/Bowtie2[12] , hpg-aligner [15] , SOAP2 [16]—or , more broadly , any read aligner whatsoever capable of outputting to headerless SAM format[17] and suppressing unmatched input reads . Utilizing SAM is much faster than BAM ( binary compressed SAM ) after deduplication , as the alignment step is not I/O bound and the overhead of BAM’s additional compression/decompression step can be significant . We have chosen to standardize NINJA around Bowtie2 for our tests and publish the command line options for Bowtie2 as we have found it to be suitable for the purposes of BLAST-identity-based OTU picking . Following alignment , the resulting SAM file is fed to the NINJA parsing step , which takes in the sample dictionary metadata as well as an optional taxonomy map to rapidly re-assign each de-duplicated read to the biological sample ( s ) in which it originally occurred , add taxonomy annotation to each picked OTU , bin all reads by their matched OTU into a sample-by-OTU matrix ( OTU table ) , and output the result in sparse BIOM 1 . 0 format or a tab-delimited human-readable legacy QIIME format . This can also be incorporated into an open-reference OTU-picking pipeline . The BWT has received a lot of attention in the alignment of short reads to a reference genome , and now enjoys routine use in clinical and other settings as a well-vetted technique for mapping short DNA reads to a longer reference , where it is known as Burrows-Wheeler alignment . The BWT is based on the principle that a long string of text can be reversibly transformed to reduce the complexity of substring queries to effectively two binary searches into the transformed representation of the original string , which is then converted back to indices into the original reference string with a short walk-back ( the BW Last-First , or LF walk ) or lookup . The efficacy of this approach in matching short reads to a reference database of numerous short reference marker genes has remained largely unexplored [1] .
The runtime performance of the database generation is significantly longer than is practical to perform on the fly . This step only needs to be performed once for each reference database . Although ninja_prep performs the concatenation of references rapidly ( it is I/O bound on the Macbook’s SSD ) , the BWT program may spend a long time generating the BW index . For bowtie2 on our test machine , this takes over half an hour ( with a maximum of one thread ) on the Greengenes 97% OTU representative sequence database . For this reason , it is best to store and use pre-compiled databases for all subsequent alignments , and NINJA-OPS is distributed with a number of pre-compiled databases for commonly performed 16S bacterial marker-gene OTU matching . NINJA filtering takes approximately 10–15% of the alignment time . For our 1 . 6 million read 175bp test data , without additional processing , filtering runs in 3 . 5 seconds and outputs a de-duplicated FASTA file approximately 1/5 the size of the original . Bowtie2 with the settings mentioned in methods aligns the entire test dataset of 1 . 6 million 175bp reads in under 40 seconds on a single thread of the test laptop . Performance for the default and maximum-fidelity ( “max” ) NINJA presets were measured ( Fig 2 ) . The “max” preset not only demonstrates higher accuracy than either the default preset or USEARCH , but also retains significantly more reads . The “fast” preset displays similar accuracy characteristics to “default , ” but misses about 2% of the alignments detected by the latter , usually of the lowest identity . Total pipeline runtime on the same dataset decreases to less than 10 seconds when using the recommended singleton-based denoising option ( parameter “D 2” ) , in combination with default preset . In addition , the speedup versus USEARCH 8 was measured using the default preset without denoising ( “D 0” ) across different datasets ( S2 Fig ) . RAM usage during alignment was 205MB in all cases , while that of USEARCH 8 was 720MB . Using multiple threads during alignment decreases the running time further , but speedup is sublinear , having somewhat more advantage in datasets with longer reads or higher error rates ( and hence more difficult alignments ) . Parsing with ninja_parse takes roughly 0 . 2–3 seconds on datasets in the size range included here ( 0 . 5–2 million sequences ) . Outputting to legacy tab-delimited format instead of BIOM increases the runtime by a second or two . A Python-based convenience wrapper distributed with NINJA adds additional overhead if the user requests a fasta file containing the sequences that failed to match the database . To assess the accuracy of the alignments found by NINJA , and to compare them to existing tools , we calculated the optimal alignment , using a semi-global version of the Smith-Waterman algorithm , of each query sequence with the reference sequence assigned by a given tool . We found that NINJA ( default preset ) generally finds higher-accuracy matches than USEARCH 8 ( Mann-Whitney U test p < 2 . 2e-16 ) ( Figs 3 and 4 ) . In a published dataset containing healthy subjects and patients with Crohn’s disease the two methods produced the same list of differentiated genera across disease conditions with occasional disagreements about the direction of the association ( S3 Fig ) . NINJA produced a comparable percentage of matches with default preset to USEARCH ( S4 Fig ) , and generally comparable taxonomic assignments despite some interesting differences ( S5 Fig ) .
Optimizations within NINJA-OPS include tweaks to the parsing and filtering programs to increase the throughput of the processes leading up to the alignment . Deduplication is a viable strategy in marker-gene sequencing contexts because samples usually consist of fewer taxa than there are reads , and in fact are often dominated by a few highly abundant species . This results in a large number of identical reads which can be filtered out to reduce the alignment time . In human gut datasets which are quality-trimmed ( or where the marker gene reads are of approximately equal length ) , this may result in losslessly discarding 80% of the reads as duplicates , depending on the microbial community sampled , which can speed up the downstream alignment step substantially ( S6 Fig ) . A sequence-to-sample ( s ) dictionary keeps track of the abundance of each sequence in each sample to ensure that each original sequence is properly accounted for wherever it was originally found . By default , NINJA-filter also performs read compaction ( parameter “-d 1" ) , which normalizes for variation in read lengths within a dataset by treating reads which are subsets of longer reads as copies of the longer reads . This increases consistency of OTU calling as well as decreasing runtime . This behavior can easily be disabled ( parameter “-d 0” ) . An optional beneficial feature during the filter step is the ability to perform lossy denoising . NINJA performs this in two ways . The first and most straightforward for amplicon reads is to discard singleton reads ( parameter “-d 2” ) ; that is , reads that have no identical match in the entire list of queries , or which are not perfectly contained in a longer read . This can be extended as the user desires from singletons to doublets and so on ( parameter “-d 3” , “-d 4” , etc . ) . The second form of denoising is discarding reads judged to be erroneous by breaking each read into its component overlapping k-mers and comparing each of these k-mers to the counts of that k-mer in an empirical distribution of all k-mers in the body of input reads . Reads with k-mers that fail to meet user-defined criteria for support ( appearing under a certain % in the dataset ) are discarded completely from subsequent analysis . The resulting speedup for the downstream alignment is often much greater than the proportion of reads discarded , because Burrows-Wheeler alignment programs expend a disproportionately large amount of effort attempting to align erroneous reads that will not match the database compared to non-erroneous reads which will often find perfect ( or near-perfect ) matches in a well-populated database . The BW substring search is designed for perfect substring searches , so it performs most efficiently in aligning reads that have few to no mismatches with a subsequence of the database . This is also why NINJA and BWT tools perform most effectively when the alignment identity is high ( %ID in the mid-to upper-90’s , with taxonomic resolutions at the level of genus or finer ) . Performance of BWT-based tools is expected to increase as the diversity of available reference sequences increases , because the probability of finding a perfect match likewise increases . One early concern as we were considering how to most effectively construct the concatesome was that some reads would align by chance to the boundary between two concatenated marker genes , which would produce a meaningless mapping . However , in practice , such an occurrence is exceedingly unlikely to occur in end-to-end marker gene alignments at genus-level or greater resolution due to the high identity expected over the entire length of the input read . This is even more true of marker gene alignment , where reads are much more similar to each other than in shotgun data , and the possible sites of alignment seeding are likewise similar , with significantly less randomness than would produce alignments with the boundary region by chance . The prevalence of such reads in our 16S test data is accordingly less than 1 in 1 , 000 , 000 reads aligned . Furthermore , in the unlikely event that such an alignment does occur , it is trivial to discard it in the final parsing step by testing whether the index at the end of the alignment is equal to or greater than the starting index of the subsequent marker gene . NINJA-OPS automatically discards reads that map to junctions between concatenated marker genes . An interesting finding that corroborates past findings [21] is that the commonly used bowtie , bowtie2 , and BWA alignment tools do not scale linearly with increasing read length . However , due to the ability to substitute alternative BWT-based alignment programs for the alignment step , it is possible to use alternatively optimized variants such as HPG Aligner , which uses uncompressed suffix arrays instead of “traditional” BWT but shares many of the same characteristics with the added benefit of better scaling for longer reads . GPU-accelerated variants of the original algorithm are also available [22] . Additionally , NINJA-OPS is not restricted to the domain of 16S OTU picking , although it is distributed with a pre-built 16S database . Marker genes such as ITS for fungal identification [24] , bacterial rpoB [25] , and the recently proposed Cpn60 universal bacterial barcode [26] are easily incorporated into NINJA-OPS simply by compiling the included “ninja_prep . c” and running it on an appropriately-formatted FASTA file containing the desired marker sequences , followed by the BWT-based aligner’s database generation step . Further , NINJA-OPS can be incorporated as a preliminary step in another pipeline; for instance , NINJA-OPS can be used to group reads prior to de novo assembly [27] . This flexibility of the pipeline in allowing substitution of the aligner itself , as well as the marker gene database used , makes NINJA-OPS applicable for situations and optimizations beyond what were envisioned at the time of writing , and enable the pipeline to keep pace with emerging technologies in the sequencing and computing spheres alike . | The analysis of the microbial communities in and around us is a growing field of study , partly because of its major implications for human health , and partly because high-throughput DNA sequencing technology has only recently emerged to enable us to quantitatively study them . One of the most fundamental steps in analyzing these microbial communities is matching the microbial marker genes in environmental samples with existing databases to determine which microbes are present . The current techniques for doing this analysis are either slow or closed-source . We present an alternative technique that takes advantage of a high-speed Burrows-Wheeler alignment procedure combined with rapid filtering and parsing of the data to remove bottlenecks in the pipeline . We achieve an order-of-magnitude speedup over conventional techniques without sacrificing accuracy or memory use , and in some cases improving both significantly . Thus our method allows more biologists to process their own sequencing data without specialized computing resources , and it obtains more accurate and even optimal taxonomic annotation for their marker gene sequencing data . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"sequencing",
"techniques",
"split-decomposition",
"method",
"marker",
"genes",
"database",
"searching",
"next-generation",
"sequencing",
"multiple",
"alignment",
"calculation",
"gene",
"sequencing",
"genome",
"analysis",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"genomics",
"sequence",
"alignment",
"bioinformatics",
"biological",
"databases",
"heuristic",
"alignment",
"procedure",
"molecular",
"biology",
"software-aided",
"sequence",
"analysis",
"sequence",
"databases",
"computational",
"techniques",
"database",
"and",
"informatics",
"methods",
"transcriptome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"computational",
"biology",
"dna",
"sequencing"
] | 2016 | NINJA-OPS: Fast Accurate Marker Gene Alignment Using Concatenated Ribosomes |
Leishmaniasis is a zoonotic disease of worldwide relevance . Visceral leishmaniasis is endemic in Brazil , where it is caused by Leishmania infantum with Lutzomyia longipalpis being the most important invertebrate vector . Non-human primates are susceptible to L . infantum infection . However , little is known about the role of these species as reservoirs . The aim of this study was to evaluate the transmissibility potential of visceral leishmaniasis by non-human primates through xenodiagnosis using the phlebotomine Lu . longipalpis as well as to identify phlebotomine species prevalent in the area where the primates were kept in captivity , and assess infection by Leishmania in captured phlebotomine specimens . Fifty two non-human primates kept in captivity in an endemic area for leishmaniasis were subjected to xenodiagnosis . All primates were serologically tested for detection of anti-Leishmania antibodies . Additionally , an anti-Lu . longipalpis saliva ELISA was performed . Sand flies fed on all animals were tested by qPCR to identify and quantify L . infantum promastigotes . Eight of the 52 non-human primates were positive by xenodiagnosis , including three Pan troglodytes , three Leontopithecus rosalia , one Sapajus apella , and one Miopithecus talapoin , with estimated numbers of promastigotes ranging from 5 . 67 to 1 , 181 . 93 per μg of DNA . Positive animals had higher levels of IgG anti-Lu . longipalpis saliva when compared to negative animals , prior to xenodiagnosis . Captive non-human primates are capable of infecting Lu . longipalpis with L . infantum . Our findings also demonstrate the relevance of non-human primates as sentinels to zoonotic diseases . Several phlebotomine species , including Lu . longipalpis , have been identified in the area where the primates were maintained , but only one pool of Lutzomyia lenti was infected with L . infantum . This study has implications for public health strategies and conservation medicine .
Visceral leishmaniasis is a zoonotic disease caused by obligate intracellular protozoa of the genus Leishmania ( order Kinetoplasta: family Trypanosomatidae ) , which infect macrophages of several mammalian species , including man [1] . There are different species of Leishmania that can cause visceral leishmaniasis in the world . L . infantum ( synonym L . chagasi ) has the broadest distribution [2] . The main form of transmission is through the bite of female flies of the subfamily Phlebotominae [1] , and the most relevant biological vector in Brazil is Lutomyia longipalpis [3] . Although leishmaniasis has a high incidence , morbidity and lethality , it is one of the most neglected zoonotic diseases in the world , affecting mainly deprived human populations from developing countries in tropical areas of the Americas , Asia and Africa , extending to temperate regions of Latin America [1] . Despite its known geographic distribution , leishmaniasis , as others vector born diseases , is a dynamic disease , which transmission circumstances undergo continuous changes dependent on environmental , demographic , and human behavior factors [4] . According to the World Health Organization [1] , potential wild reservoirs of visceral leishmaniasis in the New World are wild canids , especially the crab-eating fox ( Cerdocyon thous ) , and opossums ( Didelphis marsupialis and D . albiventris ) , although the domestic dog is recognized as the most important reservoir in urban areas [5] . Many other species of wild and synanthropic mammals have also been identified as potential reservoirs , including some species of bats , felines , and neotropical primates [6 , 7] . To classify an animal as a reservoir , some criteria must be met , among which the capacity of the reservoir host to maintain the availability of parasites in the skin in sufficient numbers to be transmitted to the vectors [1 , 8] . Xenodiagnosis is an efficient tool to evaluate the capacity of a mammalian host to transmit this pathogen , which characterizes the host as a potential reservoir [9 , 10] . Although the risk of transmission of leishmaniasis by potential wild reservoirs , such as crab-eating fox [11] , wild rabbits [12] , maned wolves [9] , and bush dogs [9] , has already been evaluated by xenodiagnosis , such risk is completely unknown in the case of non-human primates . Non-human primates can be affected by leishmaniasis , with clinical and pathological manifestations that are similar to those reported in human patients , sometimes remaining asymptomatic [13–15] . However , there are only a few epidemiological studies on the occurrence of leishmaniasis in non-human primates [16] , and a complete absence of scientific data on the role they play in transmission of leishmaniasis . This study aimed to evaluate the transmissibility potential of visceral leishmaniasis by non-human primates through xenodiagnoses using the phlebotomine Lu . longipalpis . In addition , we performed identification of the phlebotomine species prevalent in the area where the primates were kept in captivity and assessed infection by Leishmania in captured phlebotomine specimens .
The experimental protocol employed in this study has been approved by the Ethics Committee on the Use of Animals of the Universidade Federal de Minas Gerais ( CEUA/UFMG ) , under protocol number 94/2013 . CEUA/UFMG adheres to the Brazilian legislation ( law 11794 –October 8 , 2008 ) under supervision of the Conselho Nacional de Controle de Experimentação Animal—CONCEA . Fifty two non-human primates kept in captivity at the zoological garden in Belo Horizonte ( Brazil ) were included in this study , totaling 13 species , 11 neotropical primates and two old world primate species ( Table 1 ) . Most of the primates ( 27/52 ) were born at the Belo Horizonte zoo , and all of them were housed at the zoo for at least one year prior to this study . A detailed description of the origin of each primate included in this study is provided in a Supplementary Table ( S1 Table ) . Phlebotomine capture was performed at 10 sites within the zoo area from February 2014 to February 2015 using traps as previously described [20] . Phlebotomines were identified by morphologic examination based on previously described criteria [21] . DNA samples were extracted from captured phlebotomines and used for nested PCR ( LnPCR ) amplification of the Leishmania SSUrRNA gene [22] . All reactions included positive ( 20 ng of L . infantum genomic DNA ) and negative controls ( target DNA replaced with water ) . PCR products were sequenced and blasted against the GenBank database for identification of the Leishmania species . Frequencies of positivity of non-human primates by xenodiagnosis were compared using the chi-square test with confidence interval of 95% ( p < 0 . 05 ) . Kruskall-Wallis and Mann-Whitney tests were performed to compare all other data . Agreement between the three serologic tests was calculated by Kappa analysis . All statistical analyses were performed using the Prism software version 7 . 0 ( GraphPad ) .
Seven of the 52 non-human primates tested ( 13 . 46% ) were serologically positive for Leishmania spp . using rKDDR as antigen ( Table 2 ) . The others two tests ( ELISA with rK39 and RAPID test ) were a little less sensitive , with 9 . 61% ( 5/52 ) of positivity . However , all serological tests had strong agreement , with kappa coefficient equal to 0 . 87 , using ELISA rKDDR as the main test . One L . rosalia was positive only at qPCR pool of sand flies , but serologically negative . In total , positive animals included: three P . troglodytes ( 100% 3/3 ) , three L . rosalia ( 20% 3/15 ) , one S . apella ( 16 . 67% 1/6 ) , and one M . talapoin ( 50% 1/2 ) . P . troglodytes were more predisposed to be serologically positive than the other species included in this study ( p = 0 . 015 ) . Both seropositive and seronegative animals were subjected to xenodiagnosis . All positive animals were capable to infect at least one sand fly , whereas one animal ( Leontopithecus rosalia ) that was serologically negative was also positive by xenodiagnosis ( Fig 1 ) . The number of promastigotes/μg of DNA varied from 5 . 67 to 1 , 181 . 93 in positive sand flies ( Fig 2 ) . We observed that L . rosalia was more efficient than S . apella to infect sand flies ( p = 0 . 0328 ) , with average of 194 and 10 promastigotes/μg of sand fly DNA , respectively . Although M . talapoin had one infected sand-fly with also 10 promastigotes/μg of sand fly DNA we could not perform a statistical test because the low number of infected sand fly in this case . P . troglodytes had an average of 59 promastigotes/μg of sand fly DNA . Fig 2 shows the quantity of promastigotes/μg of DNA of each positive sand fly from each species of the study . Although we did not observe significant differences between the O . D . of all families evaluated , animals positive to leishmaniasis had a significantly higher O . D . when compared to negative animals ( p = 0 . 0049 ) ( Fig 3 ) . A total of 1 , 392 phlebotomine specimens were captured , including the following species: Psathyromyia aragaoi ( Costa Lima , 1912 ) , Evandromyia bacula ( Martins , Falcão and Silva , 1965 ) , Evandromyia cortelezzii ( Brèthes , 1923 ) , Lutzomyia ischnacantha ( Martins , Souza and Falcão , 1962 ) , Lutzomyia lenti ( Mangabeira , 1938 ) , Pintomyia monticola ( Costa Lima , 1932 ) , Lutzomyia longipalpis ( Lutz and Neiva , 1912 ) , Pintomyia pessoai ( Coutinho and Barretto , 1940 ) , Microppygomyia quinquefer ( Dyar , 1929 ) , Sciopemyia sordellii ( Shannon and Del Ponte , 1927 ) as detailed in Table 3 . Seasonal distribution of captures is detailed in a Supplementary Table ( S3 Table ) . After taxonomic identification of all captured females , 226 species-specific pools with up to 10 phlebotomines each ( Table 3 ) were used for DNA extraction and PCR . These pools included all species with the exception of Psathyromyia aragaoi , which had only one male captured . Only one pool of Lutzomyia lenti was PCR positive for Leishmania spp . The amplified sequence had 97% identity with L . infantum .
Natural infections and Leishmania-associated disease in non-human primates have been occasionally reported , and there is also evidences of infection based on serology and PCR affecting captive and free-living animals [7 , 13–15 , 23–26] . Non-human primates have been also extensively used for experimental infections with Leishmania spp . , especially for vaccinology and clinical or immunopathological studies , with similar outcomes when compared to human patients [27–30] . Xenodiagnosis is the only tool capable of confirming the ability of a potential reservoir to infect the parasite vector . This study describes for the first time the competence of asymptomatic non-human primates to transmit L . infantum to Lu . longipalpis , the most important invertebrate vector of visceral leishmaniasis in the New World . Parasite loads in infected sand flies observed in this study were considered low , although they were similar to parasite loads previously found in phlebotomines fed on asymptomatic or symptomatic Leishmania-infected dogs that had averages of 10 and 84 parasites , respectively [31] . However , in that particular study one symptomatic dog infected sand flies with 29 , 774 parasites . In our study there was also one L . rosalia that infected a sand fly resulting in a high parasite load ( 1 , 181 . 93 parasites/μg of DNA ) . Infected sand flies may be classified as “super-spreader” when they carry a high infectious dose ( > 600 parasites ) , which is responsible for epidemic high-disease burden; and “endemic-spreader” with low parasite loads , and responsible for endemic low-disease burden [32] . Therefore , our results suggest that some non-human primate species may play a role in the endemic Leishmania cycle , transmitting lower infective doses of parasites to sand flies , favoring the circulation of “endemic-spreader” sand flies , which perpetuate a “mild/asymptomatic mode” of leishmaniasis . This situation is unlikely to favor emergence of clinical disease , but it may result in maintenance of Leishmania in a given population within an endemic area [32] . The infective dose of promastigotes in naturally infected sand flies is still unknown , with reports indicating hundreds to thousands promastigotes being required for establishment of infection in a mammalian host [31 , 33 , 34] . Some studies indicate that physical obstruction of the sand fly anterior midgut is required for actual transmission of the parasite to mammalian hosts [35 , 36] . Such obstruction is the result of an association of promastigotes and promastigote secretory gel ( PSG ) forming a sausage-like plug distending the sand fly anterior digestive tract [36 , 37] . Obstruction of the anterior gut leads to regurgitation of metacyclic promastigotes during blood feeding , resulting in infection of mammalian host [37] . A minimum number of promastigotes is needed to produce enough PSG to act as a blocking plug . However , the number of promastigotes increase about 16 times within the sand fly [36 , 37] , and sand flies tend to become increasingly parasitized after a second blood meal even if this second meal takes place on an uninfected host [38] . Therefore , it is reasonable to consider that even low parasite loads , as observed in sand flies that fed on non-human primates in this study , could eventually lead to an infective parasite load in an endemic environment . Importantly , we performed qPCR using DNA extracted from sand flies at five days after the blood meal . At five days , ingested blood has been eliminated by the sand fly through defecation [39] so ingested Leishmania DNA fragments do not interfere with PCR amplification at that time point . Although later time points may result in higher parasite loads , the PCR technique employed in this study is highly sensitive so sampling of sand flies at 5 days post blood meal was considered appropriate for the goals of this study . The competence of human individuals to infect sand flies is enhanced in symptomatic and immunocompromised patients [40 , 41] . Conversely , this tendency is still questionable in domestic dogs [31 , 42 , 43] . Previous reports demonstrated that two non-human primates housed at the same institution where this study was done have been diagnosed with symptomatic visceral leishmaniasis: one Callicebus nigrifrons in 2008 [13] and one Gorilla gorilla in 2016 [15] , supporting the notion that captive non-human primates in endemic areas are susceptible to leishmaniasis . Three serological diagnostic methods were performed in this study: two ELISAs with different antigens ( rKDDR and rK39 ) and one Immunochromatographic ( RAPID test ) with rKDDR as antigen . Even though ELISA with rKDDR was more sensitive , all three tests had similar results . These results are in agreement with a recent study that demonstrated higher sensitivity and specificity of rKDDR for human or canine sera when compared to traditional serologic protocols [44] . Although the ELISA protocols employed in this study have anti-human secondary antibodies , their results were similar to those obtained with the RAPID test , which have direct interaction of primary antibody from the non-human primate serum with the specific antigen . Importantly , serological tests cannot discriminate infectious animals from those that were previously exposed to Leishmania , but are not infectious [9 , 45] . Therefore , we performed qPCR with a pool of sand flies from each animal , to ensure that positive sand flies could be detected even from serologically negative animals . Interestingly , we observed significantly higher levels of IgG anti-Lu . longipalpis saliva in positive animals , which corroborate previous studies in dogs and humans , which serology to sand fly saliva has been directly associated with host exposition to sand flies , and increased risk of infection [46–49] . Interestingly , one serologically negative animal ( Leontopithecus rosalia ) was capable of infecting female sand flies . Although in traditional experimental models the peak of parasite load coincides with higher serologic titers [50] , non-human primates experimentally infected with L . infantum often have detectable parasites in the bone marrow or Leishmania-induced lesions prior to developing a humoral response [51 , 52] . Therefore , our hypothesis in this case is that the animal was infected and infectious , but had not yet seroconverted at that time . Although naturally occurring symptomatic visceral leishmaniasis have been previously reported in non-human primates , it is considered uncommon . There is only one reported case of the disease in a neotropical primate [13] , and three recently cases in old world primates , affecting two adults orangutans [14] and one infant gorilla [15] . All of these cases occurred in endemic regions for leishmaniasis . Carneiro and coworkers [53] suggest that neotropical primates have an innate immunological resistance to L . infantum infection , whereas our results suggest that chimpanzees were predisposed to the infection when compared to other non-human primate species under captivity in an endemic area . All chimpanzees included in this study were positive by xenodiagnosis , and their levels of -Lu . longipalpis saliva suggested high exposure to sand flies . Among 1 , 392 phlebotomine specimens captured within the zoo area , and grouped into 226 pools , only one pool of Lutzomyia lenti was PCR positive for L . infantum . These results may suggest a possible role of Lu . lenti in the transmission of leishmaniasis . However , these data are insufficient to support this hypothesis since according Killick-Kendrick [54] , in order to be considered a biological vector , a given invertebrate species must: ( i ) feed on humans and animal reservoir species; ( ii ) support the development of the parasite; ( iii ) carry parasites that are indistinguishable from the ones isolated from patients; and ( iv ) be capable of transmitting the parasite through bite . In conclusion , this study demonstrated for the first time that captive non-human primates might be susceptible to Leishmania infection and capable of transmitting the pathogen to sand flies . This study raises awareness regarding the need for improved public health strategies focusing on controlling the vector and the disease . This study also emphasized the importance of non-human primates , wild or captive , as sentinels for zoonotic diseases [55 , 56] , which is highly relevant under a conservation medicine point of view . | Visceral leishmaniasis is a zoonotic disease with worldwide distribution . The disease is endemic in several Brazilian regions , including the city of Belo Horizonte , where visceral leishmaniasis is caused by Leishmania infantum and transmitted by Lutzomyia longipalpis . This study evaluated the competence of non-human primates to infect Lutzomyia longipalpis with Leishmania infantum . Eight of 52 non-human primates were positive to leishmaniasis by xenodiagnosis , i . e . capable of infecting sand flies , with averages of 5 . 67 to 1 , 181 . 93 promastigotes/μg of DNA . Positive animals had higher levels of IgG anti-Lu . longipalpis saliva when compared to negative animals , prior to xenodiagnosis . This study highlights the importance of non-human primates in the leishmaniasis cycle , providing information that is relevant for development of better public health strategies , and to conservation medicine . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2019 | Competence of non-human primates to transmit Leishmania infantum to the invertebrate vector Lutzomyia longipalpis |
There is great interest in passive transfer of broadly neutralizing antibodies ( bnAbs ) and engineered bispecific antibodies ( Abs ) for prevention of HIV-1 infections due to their in vitro neutralization breadth and potency against global isolates and long in vivo half-lives . We compared the potential of eight bnAbs and two bispecific Abs currently under clinical development , and their 2 Ab combinations , to prevent infection by dominant HIV-1 subtypes in sub-Saharan Africa . Using in vitro neutralization data for Abs against 25 subtype A , 100 C , and 20 D pseudoviruses , we modeled neutralization by single Abs and 2 Ab combinations assuming realistic target concentrations of 10μg/ml total for bnAbs and combinations , and 5μg/ml for bispecifics . We used IC80 breadth-potency , completeness of neutralization , and simultaneous coverage by both Abs in the combination as metrics to characterize prevention potential . Additionally , we predicted in vivo protection by Abs and combinations by modeling protection as a function of in vitro neutralization based on data from a macaque simian-human immunodeficiency virus ( SHIV ) challenge study . Our model suggests that nearly complete neutralization of a given virus is needed for in vivo protection ( ~98% neutralization for 50% relative protection ) . Using the above metrics , we found that bnAb combinations should outperform single bnAbs , as expected; however , different combinations are optimal for different subtypes . Remarkably , a single bispecific 10E8-iMAb , which targets HIV Env and host-cell CD4 , outperformed all combinations of two conventional bnAbs , with 95–97% predicted relative protection across subtypes . Combinations that included 10E8-iMAb substantially improved protection over use of 10E8-iMAb alone . Our results highlight the promise of 10E8-iMAb and its combinations to prevent HIV-1 infections in sub-Saharan Africa .
The World Health Organization estimated that in 2015 , approximately two-thirds of the 2 million new HIV-1 infections globally , were in sub-Saharan Africa . Since HIV-1 infection cannot be cured , effective vaccines or other prevention measures are needed to mitigate the impact of HIV/AIDS on global health . Successful antibody ( Ab ) -based vaccines prevent infection , and T-cell-based vaccines enhance control of infection , but the development of such vaccines has proven challenging [1] . Pre-exposure prophylaxis ( PrEP ) with reverse transcriptase inhibitors is effective in prevention of HIV-1 infections , and is in current use [2] . PrEP efficacy , however , depends on adherence , which is challenging given that four or more doses a week are required , and associated costs and toxicity [2 , 3] . Thus , alternative approaches to PrEP using broadly neutralizing antibodies ( bnAbs ) or long acting antiretroviral formulations are being explored [4] . Many bnAbs isolated from chronically infected individuals can potently neutralize a substantial fraction of diverse global panels of HIV-1 pseudoviruses in vitro . Their characterization has provided insights for vaccine design [5–9] , enabling progress in strategies for eliciting bnAb responses [10 , 11] . The best bnAbs are also promising candidates for passive transfer to prevent HIV-1 infections , a more readily achievable goal [4 , 12] . Several preclinical studies have shown efficacy for prevention of HIV-1 infections following passive transfer of bnAbs [13–21] . In a recent repeated low-dose simian-human immunodeficiency virus ( SHIV ) challenge study in rhesus macaques using SHIVAD8-EO [21] , a single bnAb infusion delayed infection by weekly SHIV challenges to medians of 8–14 weeks , depending on the bnAb , compared to median of 3 weeks for infection of control animals . This underscores a main advantage of bnAbs over most small-molecule drugs–the long in vivo half-lives of bnAbs can result in prolonged protection by a single dose . Antibodies can be engineered to extend in vivo half-life even further [22 , 23] . Other advantages include Fc-mediated effector functions [24 , 25] , reduced side effects , and the availability of alternative approaches in situations of emerging drug resistance . Based on such encouraging data , several promising bnAbs are being clinically developed , and have either begun ( PGT121 in clinical trial NCT02960581 ( ClinicalTrials . gov identifier ) and VRC07-523LS in NCT03015181 ) , or completed preliminary human testing ( VRC01 , 3BNC117 and 10–1074 ) [26–28] . The first phase 2b efficacy trials using the bnAb VRC01 are underway in three continents ( NCT0271665 , NCT02568215 ) . The potency of particular bnAbs against different pseudoviruses tested in global panels is highly variable ( Fig 1 ) ; some Envs for any given bnAb will be completely resistant or have less potent IC80 titers [29 , 30] . Patterns of Env sensitivity are similar for bnAbs targeting similar epitopes , but differ across epitope classes [30] . Thus , a natural solution to the problem of limited breadth/potency is to combine bnAbs targeting different epitopes [29] . Neutralization for bnAb combinations in vitro can be very accurately modeled using individual bnAb data , suggesting that bnAbs targeting different epitopes act independently when used in combination [30] . Other solutions include engineering artificial bispecific antibodies with two Fab arms derived from different bnAbs [31–33] , or bispecific antibodies with one arm targeting the HIV receptor or co-receptor on host cells , and the other targeting HIV-1 Env [34–36] . A different approach involves arms derived from the CD4 receptor , with the Ab base including a CCR5 co-receptor mimetic peptide [37] . All these approaches can increase neutralization breadth and potency against diverse viruses , and several are under clinical development . As promising bnAb and bispecific candidates are developed , it will be important to assess their potential for in vivo prevention , and to compare in vivo performance to in vitro measures of neutralization . This will help inform choices regarding candidates Abs for subsequent advancement in the clinical testing pipeline . For successful clinical outcomes , Abs or Ab combinations will need to be effective against diverse circulating strains and the diversity in viral quasispecies that accumulates in each chronically infected donor . Furthermore , genetically identical virus samples can have Ab resistant subpopulations [30 , 38] , due to phenotypic heterogeneity in glycosylation profiles [39] and protein conformations [40 , 41] . Economic factors must also be considered , as Ab manufacturing costs may be higher than for small molecule drugs . Here , we analyzed the potential of several leading conventional and bispecific Ab candidates to prevent HIV-1 infections in sub-Saharan Africa . We obtained in vitro neutralization data for leading Ab candidates against virus panels of subtypes A , C & D , the dominant subtypes in this region , and used our modeling approach to predict neutralization by all Abs and 2 Ab combinations [30] . We compared the in vitro performance of Abs and Ab combinations using realistic in vivo target concentrations and previously developed metrics [30] that measure breadth-potency of neutralization , and efficacy against within-host viral diversity and viral phenotypic heterogeneity . Finally , we modeled Ab-mediated in vivo protection as a function of in vitro neutralization using data from a macaque SHIV challenge study , and used this to predict the relative protection afforded by Abs and Ab combinations in this study .
We collected in vitro neutralization data ( Methods ) for leading candidate bnAbs and bispecifics for PrEP: CD4 binding site ( CD4bs ) bnAbs 3BNC117 [42] , N6 [43] , VRC01 [44] and VRC07-523LS [23]; V2 glycan ( V2g ) apex bnAbs CAP256-VRC26 . 25 [45] and PGDM1400 [46]; V3 glycan ( V3g ) bnAbs 10–1074 [47] and PGT121 [48]; and bispecific antibodies 10E8v2 . 0-iMAb ( 10E8-iMAb for brevity ) [36] , which targets membrane proximal external region ( MPER ) on Env and host-cell CD4 , and 3BNC117-PGT135 [32] , which targets CD4 binding site and V3 glycan epitopes on Env . We studied the efficacy of the above antibodies against subtype A , C and D viruses , which make up 81 . 1% of the Los Alamos HIV database sequences from sub-Saharan Africa ( Fig 1A ) . The subtype C panel is a 100 pseudovirus subset of a previous panel of 200 early/acute viruses from southern Africa , designed to preserve the breadth-potency profiles of bnAbs as seen for the larger panel [49 , 50] . Subtype A panel includes 25 pseudoviruses , subtype D 20 , cloned from chronically infected individuals from five sub-Saharan Africa countries each spanning years 1992–2008 and 1993–2008 , respectively ( S1 Table ) . 22 out of 25 subtype A pseudoviruses ( including 5 transmitted-founder viruses ) and 11 out of 20 subtype D viruses ( with 4 transmitted-founder viruses ) were isolated from acute/early infections . Majority of subtype A and D pseudoviruses were cloned using single genome amplification or limiting dilution PCR . IC80 titer heatmaps of antibodies are shown in Fig 1B–1D , and IC50 and IC80 data are reported in S1 and S2 Data . These data recapitulate previously observed bnAb neutralization profiles , e . g . [29 , 30]: V2g and V3g bnAbs can be very potent , but have limited breadth ( Fig 1 ) , and CD4bs bnAbs are generally less potent but show higher breadth . Also , V2g and V3g bnAbs tend to have complementary reactivity patterns; Envs that are insensitive to V3g bnAbs are sensitive to V2g bnAbs , and vice versa . Several Abs showed subtype-specific differences in IC80 potency ( see Fig 1E for levels of statistical support ) . CD4bs bnAbs VRC01 , 3BNC117 and bispecific 3BNC117-PGT135 were significantly more potent against subtype A viruses; PGT121 less potent against subtype D; and 10E8-iMAb was less potent against subtype A . We next characterized the performance of individual Abs using in vitro IC80 breadth and potency and completeness of neutralization ( Fig 2 ) . Some bnAbs incompletely neutralize pseudoviruses even at very high concentrations [30 , 38 , 51] , due to phenotypic heterogeneity in a clonal pseudovirus sample arising from heterogeneity in glycan occupancy and/or processing [39] , dynamics of Env trimers [40] and alternate variable loop configurations [41] . We modeled the fraction of pseudovirus neutralized by bnAbs using the Hill curve parametrization of neutralization curves ( Methods ) , which accurately predicts observed neutralization profiles [30] . As exact thresholds for in vivo efficacy are not yet well characterized , we used >95% for complete neutralization as before [30] . This threshold is not unreasonably high , as our analysis of a low-dose SHIV challenge study below showed that a few macaques got infected in spite of serum bnAb levels corresponding to 95–99% in vitro neutralization of the challenge pseudovirus [21] . In the ongoing Phase 2b VRC01 clinical trials , the minimum VRC01 in vivo serum concentrations are predicted to be 5–16 μg/ml [52] . Based on this , we chose the target minimum concentration of 10 μg/ml total for our modeling of Abs , individually or in combinations . We used 5 μg/ml for bispecifics because of their greater potency ( Fig 1 ) . Since post-infusion concentrations will be higher than the minimum concentrations we assumed , our results yield conservative estimates of Ab efficacy . Even with using half the target concentration , 10E8-iMAb had the best neutralization metrics when compared against all other individual Abs for each subtype ( Fig 2 and S2 Table ) . It showed best IC80 potency ( median IC80 of 0 . 0045–0 . 015 μg/ml across all subtypes , 4–54 fold more potent than the next most potent Ab , p = 9 . 4x10-8–0 . 037 using Wilcoxon rank-sum test ) and completely neutralized ( >95% neutralization ) 96–100% of panel viruses . For subtypes A and C , the next best performing single Ab was N6 , with 95–96% IC80 coverage and complete neutralization of 87–92% panel viruses at 10 μg/ml . For subtype D , the bispecific 3BNC117-PGT135 showed the second most potent IC80 titers after 10E8-iMAb ( median IC80 of 0 . 2425 μg/ml ) and completely neutralized 85% of viruses at 5 μg/ml . While V2g bnAbs were very potent against sensitive viruses , they had relatively low IC80 breadth ( 55–59% for CAP256-VRC26 . 25 and 50–68% for PGDM1400 ) and lower proportion of completely neutralized viruses ( 32–47% for CAP256-VRC26 . 25 and 40–64% for PGDM1400 ) . Similarly , V3g bnAbs , which were slightly less potent than V2g , also show limited IC80 breadth and a low proportion of viruses completely neutralized . Among CD4bs bnAbs , N6 was best , VRC07-523LS was nearly comparable followed by 3BNC117 and VRC01 ( Fig 2 ) . To partially mitigate potential sampling biases of our pseudovirus panels ( particularly the smaller subtype A and D panels ) , we performed bootstrap resampling to understand the robustness of our results ( Methods ) . We generated 1 , 000 bootstrap realizations to match the size of each panel , and characterized the median and 95% confidence intervals ( CI ) for each of the above metric for each Ab; these results are presented in S3 Table . We found that 10E8-iMAb still showed the best metrics for all subtypes , with few bnAbs showing any metrics that were within the bootstrap 95% CI of the respective metric for 10E8-iMAb ( S3 Table ) . Next , we analyzed combinations of 2 conventional bnAbs , since combinations improve performance over single bnAbs [29 , 30] . We used the Bliss-Hill model on single bnAb IC50 and IC80 data to predict combination IC80 titers and pseudovirus fraction neutralized for combinations of 2 bnAbs targeting different epitopes ( Methods ) . This approach was shown to accurately predict experimental data [30] . We assumed equal concentrations for both bnAbs , and used a total target concentration of 10 μg/ml for the combination , i . e . 5μg/ml per bnAb . As before , we used IC80 breadth-potency and completeness of neutralization as metrics to evaluate performance . We also used coverage with both bnAbs active as a metric for prevention success , assuming a virus is actively neutralized by both bnAbs in a combination if IC80 < 5μg/ml for each bnAb individually . The rationale behind this metric is that strains from within-host quasispecies will have a lower chance of resistance to both bnAbs [30] . We analyzed three classes of 2 bnAb combinations: CD4bs + V2g , CD4bs +V3g and V2g + V3g . For each subtype the best overall performance across all metrics was observed for CD4bs + V2g combinations , however , the best specific combination was subtype dependent . For subtype A , the best combination was N6 + PGDM1400 , with the lowest median IC80 titer ( 0 . 027 μg/ml , combination titers are reported as total concentration of both bnAbs ) ( Fig 3A , S2 Table ) , best IC80 coverage ( 96% ) , second best coverage of complete neutralization ( 92% , best was 96% ) , and best coverage with both bnAbs active ( 68% ) . For subtype C , the best combination was N6 + CAP256-VRC26 . 25 , with the second lowest median IC80 titer ( 0 . 041 μg/ml , best was 0 . 025 μg/ml ) , best IC80 coverage ( 99% ) , best coverage of complete neutralization ( 93% ) , and third best coverage with both bnAbs active ( 52% , best was 56% ) . For subtype D , the best combination was 3BNC117 + CAP256-VRC26 . 25 , with the most potent median IC80 titer ( 0 . 114 μg/ml ) , second best IC80 coverage ( 95% , best was 100% ) , best coverage of complete neutralization ( 90% ) and second best coverage with both bnAbs active ( 45% , best was 50% ) . CD4bs + V3g combinations had somewhat lower performance than CD4bs + V2g combinations . The V2g + V3g combinations had some of the most potent median IC80 titers , however , they also had the lowest IC80 coverage , completeness of neutralization and especially coverage with both bnAbs active , due to the complementarity between V2g and V3g bnAb neutralization profiles ( Fig 1B–1D ) . Several of these results were robust to bootstrap variation ( S3 Table ) , however , a few differences were observed . For subtype D , the best combination was predicted to be VRC07-523LS + 10–1074 , based on each of its metrics being the best or within 95% bootstrap CI of the best metric . This combination showed the best IC80 breadth ( 100% ) and best coverage with complete neutralization ( 90% ) among combinations of two conventional bnAbs for subtype D . In general , we found that several combinations showed metrics that were within 95% bootstrap CI from the best metric for the subtype A and D panels , consistent with their smaller size ( n = 25 and 20 , respectively ) . This suggests that larger , representative panels for these subtypes might be needed to accurately inform ranking of 2 bnAb combinations . As expected , the best 2 bnAb combinations improved performance over individual conventional bnAbs ( Fig 2 , S2 and S3 Tables ) with the same total target concentration ( 10 μg/ml ) . Improvements were observed mainly in median IC80 titers ( 2 . 3–5 . 2 fold more potent across subtypes , p = 1 . 5 x 10−6–0 . 035 using Wilcoxon rank sum test on IC80 titers ) and complete neutralization ( 0–15% increase , not significant ) , while IC80 coverage was comparable ( 0–4% increase , not significant ) . These results reinforce the notion that it is better to combine bnAbs than use the same concentration of a single bnAb . Moreover , given the extent of complete neutralization , passive transfer of the best 2 bnAb combinations has the potential to prevent infection by most diverse strains across all subtypes . Remarkably , the 10E8-iMAb bispecific performed better than the best 2 bnAb combinations across all subtypes ( Fig 3 , S2 and S3 Tables ) , despite a target concentration of 5μg/ml , half of that for the 2 bnAb combinations . 10E8-iMAb was strikingly more potent ( 1 . 8–22 . 8 fold lower median IC80 than the most potent 2 bnAb combinations across subtypes; p = 0 . 0001–0 . 0256 using Wilcoxon rank sum test; median IC80 below the 95% bootstrap CI of the best 2 bnAb combinations for subtypes C and D ( S3 Table ) ) , and had higher complete neutralization coverage ( 3–10% , not significant ) . For IC80 coverage , 10E8-iMAb matched the coverage of the best 2 bnAb combination for subtype A , had 3% lower coverage for subtype C and had 5% higher coverage for subtype D . 10E8-iMAb also has the potential to match 2 bnAb combinations in terms of two independent targets [36] , although the strong synergy between the components makes it difficult to measure coverage of 10E8-iMAb with both specificities active . Building on the impressive performance of 10E8-iMAb , we next investigated the performance of 2 Ab combinations involving bispecifics . We assumed a target concentration of 5μg/ml for each Ab in the combination , and used the Bliss-Hill model to predict the neutralization for the combinations of both bispecifics , and of a bispecific with a conventional bnAb , such that epitope targets are not repeated ( combinations of 3BNC117-PGT135 with CD4bs or V3g bnAbs were not considered ) . As before , we used median IC80 titers , IC80 coverage , coverage of complete neutralization and coverage with both Abs active as metrics to evaluate performance ( Fig 4 and S2 and S3 Tables ) . For subtype A , the bispecific combinations with the best overall performance were 10E8-iMAb with VRC07-523LS , N6 or PGDM1400 ( S2 Table ) . The former two combinations showed median IC80 titers of 0 . 016–0 . 019 μg/ml , completely neutralized all viruses , and neutralized 88% viruses with both Abs active . 10E8-iMAb + PGDM1400 had the lowest median IC80 titer ( 0 . 007 μg/ml ) , completely neutralized all viruses , and neutralized 64% viruses with both Abs active . Bootstrap analysis favored 10E8-iMAb + VRC07-523LS as its median IC80 fell within the 95% bootstrap CI of 10E8-iMAb + PGDM1400 ( S3 Table ) . For subtype C , the best combination was 10E8-iMAb + N6 with median IC80 titer of 0 . 015μg/ml , complete neutralization of all viruses and neutralization of 90% viruses with both Abs active ( S2 and S3 Tables ) . For subtype D , the best combination was 10E8-iMAb + 3BNC117-PGT135 with the second lowest median IC80 titer of 0 . 007 μg/ml , complete neutralization of all viruses and neutralization of 95% viruses with both Abs active . Across all subtypes , the best combination was 10E8-iMAb + N6 , which had relatively less potent median IC80 titers than the best , but completely neutralized all viruses and neutralized 85–90% viruses with both Abs active . It was comparable to the combination of both bispecifics for IC80 potency and coverage of complete neutralization , however , it showed higher coverage with both Abs active for subtypes A and C ( 8 and 10% , respectively ) ( for subtype D it showed 10% lower coverage with both Abs active ) ( S2 Table ) . The differences in coverage with both bnAbs active for subtypes A and D were not robust to bootstrap ( S3 Table ) . The best combinations involving bispecifics performed better than either 10E8-iMAb alone or the best 2 conventional bnAb combinations . The improvements were complete neutralization of all viruses ( 10E8-iMAb and best 2 bnAb combinations incompletely neutralized 0–10% viruses across subtypes ) and substantial increase in coverage with both Abs active ( 20–50% increase over best 2 bnAb combinations; not significant for subtype A and p = 2 . 9 x 10−9 and 0 . 001 for subtypes C and D , respectively , using Fisher’s exact test; higher than 95% bootstrap CI for best 2 bnAb combinations for subtypes C and D ( S3 Table ) ) . The latter could be important as viral resistance can emerge in chronically infected mice treated with 10E8-iMAb [36] . In such cases , combinations involving bispecifics may be advantageous as bispecific + conventional Ab combinations effectively have three independent targets and the combination of bispecifics have four independent targets . In terms of potency , the overall best bispecific combinations sometimes had less potent median IC80 titers than 10E8-iMAb alone because of the conventions of equal concentration of Abs in the combination and combination IC80 titers reported as the total Ab concentrations . However , V2g bnAbs combined with 10E8-iMAb showed more potent IC80 titers than for 10E8-iMAb alone ( S2 Table ) . The protective effect of passively transferred Abs in preventing SHIV infections has been shown in macaques [13 , 15–18 , 21 , 22] . While these studies highlight the potential of bnAbs for protection , no strategy exists to predict in vivo protection using the in vitro neutralization profile of a given Ab against a given challenge virus . Here we begin to address this question by modeling Ab mediated protection using data from a repeated , low-dose SHIV challenge macaque study by Gautam et al . [21] . In this study , a single injection of 20 mg/kg of one monoclonal antibody , 10–1074 , 3BNC117 , VRC01 or the longer half-life variant of VRC01 ( VRC01-LS ) , was given to six macaques per Ab group , and nine macaques were used as controls . Each animal was challenged weekly with a low-dose SHIVAD8-EO inoculum by the intrarectal route until they got infected . The in vivo protective effect of each Ab was significantly higher than control , and modeling of protection as a function of Ab concentration showed that the more potent the Ab against SHIVAD8-EO , the higher the protective effect . To explore whether differences in the in vivo protective effect between Ab groups could be predicted using in vitro potency of Abs , we modeled in vivo protection as a function of in vitro neutralization corresponding to the Ab titers at the time of each challenge . We predicted the fraction neutralization of the SHIVAD8-EO pseudovirus using measured or interpolated in vivo concentrations of Abs , and the in vitro pseudovirus IC50 and IC80 Ab titers reported in Gautam et al ( Methods ) . We transformed the fraction neutralization to instantaneous inhibitory potential ( IIP ) [53] as our dependent variable , since it provided better fits . IIP is defined as -Log10 ( 1 . 0 –fraction neutralization ) , and measures the Log10 reduction in a single round of infection . We modeled the binary variable “protected” or “not protected” for each challenge as a function of IIP using modified logistic regression models , with model parameters determined using likelihood maximization ( Methods , S1 Text ) . These modified models account for the baseline probability of infection for the low-dose challenge by having a scale parameter for the maximum probability of infection ( p0 ) that is fit using experimental data . We compared the fits of the experimental data using two models: a ) with the same parameters across all Abs ( 3 parameters in total ) , and b ) with different parameters for all Abs ( 9 parameters in total ) ( S1 Text ) . Using model selection criteria the maximum-likelihood model with same parameters across Abs was better ( difference in AIC = 1 . 98 , and in BIC = 24 . 90 ) , and both models provided similar likelihoods ( p = 0 . 12 using likelihood ratio test ) . We tested the goodness of fit of the model with same parameters by using the Hosmer-Lemeshow test ( Methods ) , which estimates the statistical significance for rejecting the hypothesis that the fitted model is the true model [54] . For the above model , we obtained p = 0 . 9871 , which indicates a good fit of our model to the experimental data . Thus , these results together suggest that the 4 Abs from Gautam et al . provided similar in vivo protection as a function of IIP , and that the differences in protection in this study may be explained by differences in potency and pharmacokinetics of the Abs . The maximum likelihood model with same parameters for Abs is shown in Fig 5A . This model has a baseline probability of infection for the low-dose SHIV challenge of 22 . 42% , consistent with the 9 out of 33 challenges resulting in infection of control animals ( p = 0 . 53 using binomial test ) . The probability of infection was significantly negatively correlated with IIP ( p = 6 . 35x10-12 , using likelihood ratio test , S1 Text ) . However , the protective effect of Abs was seen at high fraction neutralization , with < 5% relative protection ( defined as 100 - % relative probability of infection ) for < 96 . 1% neutralization . Above this , the protection probability was predicted to have a sharp transition , with 50% relative protection for 97 . 9% neutralization and > 95% relative protection for > 98 . 8% neutralization . To estimate the robustness of the model parameters , we simulated 1 , 000 bootstrap randomizations ( Methods , S1 Fig ) ; the estimated parameters for baseline probability of infection ( median = 23 . 68% , interquartile range = 22 . 37–25 . 29% ) and neutralization for 50% relative protection ( median = 97 . 77% , interquartile range = 97 . 18–98 . 05% ) were robust and close to the best-fit model . However , the slope of the scaled logistic curve ( median = 11 . 97 , interquartile range = 7 . 64–37 . 93 ) showed relatively higher variation , with ~22% realizations showing slopes >100 , suggesting that the slope of the best-fit model might show some dependence on the exact data used . We used the above model to predict the in vivo protection offered by Abs and Ab combinations in this study . Since the baseline rates of human HIV-1 infections are much lower than those estimated for the low-dose SHIV challenge above , we used the relative probability of infection to model the performance of Abs and Ab combinations . We calculated IIP for Abs and combinations using the Bliss-Hill model ( Methods ) for a range of concentrations and predicted the relative probability of infection for each virus and the average relative probability of infection for a panel by averaging over all viruses in the panel ( Fig 5B ) . The average relative probability of infection as a function of concentration for Abs and Ab combinations are shown in Fig 6 and numerical values at the target concentrations of 5 or 10μg/ml are reported in S2 Table . The results from bootstrap simulations of each pseudovirus panel are shown in S2–S4 Figs and bootstrap medians and 95% CI are reported in S3 Table . Among single Abs , 10E8-iMAb had the lowest average relative probability of infection of 3–5% across subtypes at 5μg/ml . It was significantly better than the next best , N6 , with 11–39% across subtypes at 10μg/ml ( p = 7 . 2x10-12–0 . 0006 using Wilcoxon rank sum test on relative probability of infection values for each virus in a panel ) ( Fig 6A , S2 Table ) . The differences between 10E8-iMAb and other bnAbs were significant for subtypes C and D using 95% bootstrap CI ( S3 Table ) . VRC07-523LS was comparable to N6 , but other single Abs were predicted to have limited protection across subtypes . VRC01 showed 40% average relative probability of infection for subtype A , but 78% for subtype C and 85% for subtype D . Similarly , V2g and V3g bnAbs had minimum average relative probability of infection of 40–63% across subtypes . 3BNC117-PGT135 showed intermediate performance with 16–50% average relative probability of infection across subtypes . The best 2 conventional bnAb combinations showed 4–18% average relative probability of infection across subtypes and were different for each subtype: N6 + CAP256-VRC26 . 25 for subtype A , N6 + PGT121 for subtype C and 3BNC117 + CAP256-VRC26 . 25 for subtype D ( Fig 6B , S2 and S3 Tables ) . While N6 + CAP256-VRC26 . 25 showed slightly better protection than 10E8-iMAb for subtype A ( 0 . 27% lower average relative probability of infection , not significantly different using bootstrap analysis ( S3 Table ) ) , 10E8-iMAb was better than the best 2 bnAb combinations for subtypes C ( 4 . 3% lower average probability of infection , p = 0 . 06 , Wilcoxon rank sum test ) and D ( 14 . 7% lower average probability of infection , p = 0 . 005 Wilcoxon rank sum test and significantly different using 95% bootstrap CI ( S3 Table ) ) , in spite of having half the total target concentration . Combinations of 10E8-iMAb with other Abs reduced the average probability of infection across all subtypes ( Fig 6C , S2 and S3 Tables ) . 10E8-iMAb + N6 showed the best protection with average relative probability of infection of < 1 . 3 x 10−7% for subtypes A and D and of 0 . 54% for subtype C , a significant improvement over 10E8-iMAb ( p = 3 . 6 x 10−18–0 . 002 across subtypes using Wilcoxon rank sum test , and significantly different using 95% bootstrap CI ( S3 Table ) ) . 10E8-iMAb + 3BNC117-PGT135 was next with an average relative probability of infection of 7 . 2 x 10−8–1% across subtypes . These results raise the possibility that 10E8-iMAb combinations with N6 or 3BNC117-PGT135 , may be very effective at preventing almost all subtype A , C and D infections .
Given the absence of effective Ab-based vaccines against HIV-1 , and the difficulties in adherence to antiretroviral drug PrEP , passive transfer of Abs is a promising alternative prophylactic modality [4 , 12] . Here , we characterized the potential of the most clinically advanced conventional and engineered Abs , and their two antibody combinations , to prevent infections in sub-Saharan Africa , by analyzing in vitro neutralization metrics and modeling of in vivo protection . Modeling of data from a macaque challenge study highlighted the potential challenges for Ab-mediated in vivo protection . In particular , the protective effect of Abs was observed only beyond 96% neutralization , suggesting that near-complete neutralization , even beyond our assumed cutoff of 95% neutralization , might be important for consistent in vivo protection . Since the number of infectious challenges was relatively small in this dataset ( n = 33 out of 337 total ) , it is possible that our modeling would miss low levels of protection at low neutralization , rendering our estimates to be conservative . However , our result is consistent with previous findings that plasma ID50 titers of ~40–200 , using different bnAbs , can protect against SHIV challenges with different viruses , doses and routes of challenge [15 , 17 , 18 , 23 , 55] . These serum titers correspond to Ab concentrations of 40–200 times IC50 titers , which assuming an average neutralization curve with slope = 1 [38 , 56] , yield ~97 . 5–99% in vitro neutralization [38 , 56] . Nonetheless , we found that 10E8-iMAb , which targets MPER and host-cell CD4 , alone and in combination with other Abs , can still meet these stringent requirements of near complete neutralization . It is not clear how applicable for human infections is the above modeling of in vivo protection using macaque data due to several potential differences . First , even the low-dose SHIV challenge is much more infectious than typical human sexual transmissions ( ~30% baseline infection rate versus ~0 . 1–1% [57 , 58] ) . Second , our model was derived using data on the single subtype B SHIVAD8-EO challenge , and whether it will hold true for different viruses with different baseline infection rates is not clear . Third , while we found that the four bnAbs studied in Gautam et al . could be modeled using the same parameters , it is not clear whether this will apply for all the Abs/combinations in this study . Fourth , our model does not account for any potential contribution of effector functions or other non-neutralizing antibody functions to in vivo antibody mediated protection; although , the good fits of experimental data suggest that these effects could be minor in comparison to neutralization . Nonetheless , our modeling clarifies the relationship between in vitro neutralization and in vivo protection in macaques , and introduces a novel statistical framework to explore the question whether or not there are universal features of Ab protection , as future human and animal studies are undertaken . The use of in vitro neutralization metrics to inform the in vivo performance of Abs intrinsically has limitations . Two issues that may impact this study are that the in vitro assays used here were based on pseudoviruses grown in 293T cells , which can result in bnAb-specific neutralization differences relative to molecular clones grown in PBMCs [38 , 59] . As the latter are more relevant in vivo , such differences could impact the relative ranking of Abs and Ab combinations obtained using pseudovirus-based in vitro metrics . Another important factor missing from our analysis is in vivo Ab stability , which can vary between Abs and can impact the choice of optimal Abs and Ab combinations for clinical efficacy . The variable performance of single bnAbs highlights the difficulty of meeting the challenges in the prevention setting . While V2 and V3 glycan bnAbs were some of the most potent , they potently neutralized less than 50% of the pseudoviruses tested ( IC80 < 0 . 1 μg/ml ) ( Fig 2 ) , and had low coverage of complete neutralization and predicted relative protection in vivo across subtypes . The CD4bs bnAbs improved coverage of complete neutralization and relative protection; however , this performance was observed for some but not all subtypes . For example , VRC01 efficacy was predicted to be lower for subtypes C and D , and higher for subtype A , highlighting the importance of considering viral subtypes among breakthrough cases in the phase 2b VRC01 Antibody Mediated Prevention ( AMP ) clinical trials [60] . Given the limitations discussed above , our models may not be predictive of outcomes in a clinical setting , however , data from clinical trials will be invaluable in understanding how predictive in vitro neutralization can be of in vivo protection in human infections , and will help refine the models developed here . Our results indicated that combinations of 2 conventional bnAbs would substantially improve the performance over the single bnAbs , even with the same total concentration . This improvement was most notable for complete neutralization and relative probability of infection . The best combinations consisted of one CD4bs and one V2 glycan bnAb , however , the optimal combination differed for each subtype . Overall , N6 with CAP256-VRC26 . 25 or PGDM1400 showed best performance across subtypes , although their performance was limited for subtype D . Thus , at the total target concentration of 10 μg/ml , even combinations of 2 bnAbs might be insufficient for prevention of some infections , across subtypes prevalent in sub-Saharan Africa . Nonetheless , 2 bnAb combinations are predicted to be highly preferable over single bnAbs . Since combinations target two independent targets , they will also increase the coverage of the donor quasispecies diversity , as chronically infected donors can have viruses resistant to single bnAbs . Bispecific Abs offer a way to increase breadth and potency by combining different paratopes in a single molecule , thus overcoming some of the above challenges . The bispecific 3BNC117-PGT135 was comparable to the best conventional bnAbs , but was outperformed by some 2 bnAb combinations . However , we found that 10E8-iMAb showed superior performance over all single Abs and , remarkably , even over all combinations of two conventional bnAbs . This performance was found across subtypes at the lower assumed concentration of 5μg/ml , half that of bnAbs/bnAb combinations . 10E8-iMAb also has two independent components , both of which are individually quite broad , with significant synergy between them in the context of the bispecifics [36] . Still , 10E8 resistance mutations in Env allowed escape in most chronically infected mice treated with 10E8-iMAb [36] , suggesting that combining 10E8-iMAb with other Abs can improve coverage of within-host quasispecies and reduce the opportunity for emergence of resistance . Indeed , our modeling indicated that 10E8-iMAb when combined with N6 or 3BNC117-PGT135 showed 80–95% coverage with both Abs active and very low average relative probability of infection ( <0 . 6% ) across subtypes . Since 10E8-iMAb does not retain Fc effector functions such as antibody dependent cellular cytotoxicity ( ADCC ) , combining 10E8-iMAb with Abs with ADCC activity might provide an additional advantage . It is not clear how important Fc effector functions might be for sterilizing protection relative to neutralization , however previous studies suggest a beneficial role , as such functionalities may help clear infections as they are beginning to disseminate [15 , 20 , 61] . Thus , combining 10E8-iMAb with potent Abs like N6 or 3BNC117-PGT135 substantially improves the already impressive predicted potential of 10E8-iMAb to prevent HIV-1 infections in sub-Saharan Africa .
This study was designed to analyze the potential of passively transferred Abs and Ab combinations to prevent HIV-1 subtype A , C and D infections . We collected in vitro neutralization data for 10 conventional and bispecific Abs against a total of 145 pseudoviruses and used computational modeling on these data to predict neutralization data for Ab combinations . We analyzed these data to compare the performance of Abs and Ab combinations . We also modeled Ab-mediated in vivo protection using data from a published macaque challenge study and used this to predict the relative in vivo protection offered by Abs and Ab combinations in this study . Panels of HIV-1 Env pseudoviruses representative of clades A ( n = 25 ) , C ( n = 100 ) , and D ( n = 20 ) were utilized to assess the breadth and potency of bnAb neutralizing activity . The clade C virus panel is a subset of the larger 200 virus panel of early/acute isolates previously described [49 , 50] . The clade A and D pseudoviruses are derived from isolates from HIV-infected patients from sub-Saharan Africa obtained as part of the CAVD Comprehensive Antibody Vaccine Immune Monitoring Consortium’s ( CAVIMC ) Standard Virus Panel Project and exhibit a Tier 2 neutralization phenotype; information about these viruses is presented in S1 Table . Env pseudovirus stocks were generated by transfection of 293T/17 cells ( American Type Culture Collection ( ATCC ) , Manassas , VA ) as previously described [62] . While the pseudoviruses panels used here are a resource shared throughout the field , developed with the intention of being representative of circulating viruses , still they are subject to bias . Two documented issues suggest they may not be fully representative of the levels of resistance that would be encountered in a prevention trial . In particular , HIV-1 is diversifying over time , and becoming increasingly resistant at the population level , measurable on the time scale of decades [49 , 63] . As neutralization panels take years to develop , the original samples from which the pseudoviruses were derived were often sampled 10–20 years ago . In addition , transmitted-founder viruses tend to be more resistant to antibodies , and many of the viruses in our panels were sampled during chronic infection [49 , 64] . The panel of 10 broad and potent monoclonal antibodies tested here was selected based on their current use in passive infusion clinical trials , or considered advanced candidates for clinical development . Cloned human antibodies were generated in the laboratories of M . Nussenzweig ( 3BNC117 , 10–1074 ) , D . Burton ( PGT121 , PGDM1400 ) , M . Connors ( N6 ) , or at the NIH Vaccine Research Center ( VRC01 , VRC07-523 , CAP256-VRC26 . 25 ) . Engineered bispecific antibodies were generated in the laboratories of D . Ho ( 10E8-iMAb ) and J . Ravetch ( 3BNC117-PGT135 ) . Neutralizing antibody titers were determined using a luciferase-based reporter assay in TZM . bl cells as previously described [65 , 66] . Starting concentrations of individual Abs ranged from 10–50 ug/ml depending on available supply at the time of testing . All Abs were serially diluted seven times using a 5-fold titration series . All assays were performed in a laboratory meeting GCLP standards . Data for 3BNC117 , 10–1074 , VRC01 , VRC07-523LS and CAP256-VRC26 . 25 for some A and D pseudoviruses were used from previous studies [23 , 29 , 36 , 43 , 45 , 67] . We used the previously developed Bliss-Hill model [30] to predict the IC80 and fraction maximum inhibition values for 2 Ab combinations , using the web tool , CombiNAber ( https://www . hiv . lanl . gov/content/sequence/COMBINABER/combinaber . html ) . The individual Ab IC50 and IC80 experimental data was the input , and a target concentration of 5μg/ml of each Ab in the combination was used . We predicted incomplete neutralization values for single Abs against panel viruses by assuming a Hill curve ( f ( c ) = cm / ( IC50m + cm ) , where f is fraction neutralized , c is concentration of Ab and m = log ( 4 ) / [log ( IC80 ) –log ( IC50 ) ] ) , as implemented in CombiNAber . To understand variation of the metrics studied here with respect to finite sampling , we used a bootstrapping approach . 1 , 000 bootstrap replicates for each pseudovirus panel were generated by randomly sampling viruses with replacement , with each replicate matching the size of the pseudovirus panel . For each bnAb/combination and for each subtype , each metric was evaluated for these bootstrap replicates and the medians and 95% CI were calculated ( S3 Table ) . We performed statistical comparisons using packages implemented in the Stats module from SciPy [68] . Non-parametric tests were preferred and two-sided p-values are reported . We used the in vivo Ab concentrations reported in Gautam et al . [21] to obtain concentration at the time of each challenge for each macaque . If the concentration at the time of challenge was not reported , we interpolated the concentration assuming a log-linear decay of Ab concentration between the timepoints immediately before and after the challenge with reported concentrations; or extrapolated the concentration by using the previous two time points with reported concentrations . To match the experimental data , a minimum threshold of 0 . 1 μg/ml was used for predicted concentrations . We used Hill curves to predict the fraction neutralization at a given concentration of an Ab using the IC50 and IC80 titers against SHIVAD8-EO pseudovirus from Gautam et al . , and transformed this to IIP using IIP = -Log10 ( 1 . 0 –fraction neutralization ) . We used modified versions of logistic regression models to model the probability of infection in vivo as a function of IIP ( S1 Text ) . The best model with same parameters across Ab groups was: p ( x ) = p0 / [1 + exp ( a x + b ) ] , where x is IIP and parameters p0 , a and b were fixed using maximum likelihood on data from all challenges in all animals across control and Ab groups . We used the SciPy package for constrained optimization algorithm ‘L-BFGS-B’ [68 , 69] to estimate maximum likelihood parameters , which were p0 = 0 . 2242 , a = 11 . 2994 and b = -18 . 8970 . The goodness-of-fit for the above model was tested using the Hosmer-Lemeshow test [54] , as implemented in R . Given the size of our dataset ( 337 data points ) , we used 10 groups for the Hosmer-Lemeshow test as recommended by Paul et al . [70]; however , our result was robust when we used 5–15 groups . Likelihood ratio tests were performed to compare models and estimate significance of parameters as reported in S1 Text . For bootstrap simulations , we generated 1 , 000 realizations using random sampling of observed data with replacement to obtain the same number of infected and uninfected data points as in the observed data , and fit the above scaled logistic model to each bootstrap realization . For prediction of in vivo relative protection of Abs and Ab combinations from this study , the above model was used with IIP values at a given concentration predicted either using Hill curves for single Abs or using Bliss-Hill model for Ab combinations as implemented in CombiNAber [30] . Average relative protection for a virus panel was obtained by averaging over relative protection for all viruses in the panel at a given concentration of Ab or Ab combination . | In the absence of effective vaccines , the use of passive transfer of conventional and engineered antibodies to prevent HIV-1 infection is being considered . This approach is promising because of broad efficacy and long in vivo lifetimes of antibodies . We analyzed the potential of leading antibody candidates , and combinations of two antibodies , to prevent HIV-1 infections in sub-Saharan Africa , the hardest-hit region in the world . We used in vitro antibody neutralization data to predict neutralization metrics that might be relevant for in vivo success , and modeled antibody-based in vivo protection as a function of in vitro neutralization using data from a macaque study . By systematic comparison , we found , as expected , that combinations of two conventional antibodies significantly outperformed individual conventional antibodies , even with same total concentration . However , different antibody combinations were optimal for the different HIV-1 subtypes analyzed . The engineered bispecific 10E8-iMAb , which targets epitopes on HIV Env and host-cell CD4 , was predicted to reduce infection probability by 20–30 fold , and outperformed all individual antibodies and combinations of two conventional antibodies . This performance was further improved by combining 10E8-iMAb with other antibodies . Thus , our results suggest that passive transfer of current antibody candidates , especially 10E8-iMAb and its combinations , might be successful in prevention of HIV-1 infections in sub-Saharan Africa . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"biotechnology",
"antibody",
"engineering",
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"engineering",
"and",
"technology",
"pathogens",
"synthetic",
"biology",
"immunology",
"geographical",
"locations",
"microbiology",
"vertebrates",
"animals",
"mammals",
"synthetic",
"bioengineering",
"retroviruses",
"primates",
"immunodeficiency",
"viruses",
"viruses",
"rna",
"viruses",
"antibodies",
"africa",
"old",
"world",
"monkeys",
"bioengineering",
"immune",
"system",
"proteins",
"monkeys",
"proteins",
"medical",
"microbiology",
"hiv",
"microbial",
"pathogens",
"hiv-1",
"macaque",
"people",
"and",
"places",
"biochemistry",
"eukaryota",
"macromolecular",
"engineering",
"virology",
"physiology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"lentivirus",
"amniotes",
"protein",
"engineering",
"organisms"
] | 2018 | Potential of conventional & bispecific broadly neutralizing antibodies for prevention of HIV-1 subtype A, C & D infections |
Newcastle Disease Virus ( NDV ) is a pathogenic strain of avian paramyxovirus ( aPMV-1 ) that is among the most serious of disease threats to the poultry industry worldwide . Viral diversity is high in aPMV-1; eight genotypes are recognized based on phylogenetic reconstruction of gene sequences . Modified live vaccines have been developed to decrease the economic losses caused by this virus . Vaccines derived from avirulent genotype II strains were developed in the 1950s and are in use globally , whereas Australian strains belonging to genotype I were developed as vaccines in the 1970s and are used mainly in Asia . In this study , we evaluated the consequences of attenuated live virus vaccination on the evolution of aPMV-1 genotypes . There was phylogenetic incongruence among trees based on individual genes and complete coding region of 54 full length aPMV-1 genomes , suggesting that recombinant sequences were present in the data set . Subsequently , five recombinant genomes were identified , four of which contained sequences from either genotype I or II . The population history of vaccine-related genotype II strains was distinct from other aPMV-1 genotypes; genotype II emerged in the late 19th century and is evolving more slowly than other genotypes , which emerged in the 1960s . Despite vaccination efforts , genotype II viruses have experienced constant population growth to the present . In contrast , other contemporary genotypes showed population declines in the late 1990s . Additionally , genotype I and II viruses , which are circulating in the presence of homotypic vaccine pressure , have unique selection profiles compared to nonvaccine-related strains . Collectively , these data show that vaccination with live attenuated viruses has changed the evolution of aPMV-1 by maintaining a large effective population size of a vaccine-related genotype , allowing for coinfection and recombination of vaccine and wild type strains , and by applying unique selective pressures on viral glycoproteins .
Live attenuated virus vaccines have been successfully employed in veterinary medicine to prevent the economic impact of many diseases in poultry and livestock . However , the role of vaccination with attenuated viruses on the evolution of wild type strains is not often considered . Antigenic escape because of strong selection by vaccines , emergence of new strains through recombination , and increased virulence to expedite transmission of new genotypes in vaccinated populations are of potential concern . In this paper , we explored the consequences of vaccination on the evolution of class II aPMV-1 , which is the etiological agent of ND . NDV , a single-stranded , non-segmented , negative-sense RNA virus of the genus Avulavirus , family Paramyxoviridae , infects a wide range of domestic and wild bird species worldwide , and causes a significant economic burden to the poultry industry [1] . The first outbreaks of NDV were reported during the mid 1920s in Java , Indonesia and Newcastle-upon-Tyne , England [2] , and within a few years were occurring throughout the world [3] . The name ND is reserved exclusively for the disease that results from infection with strains of aPMV-1 that are pathogenic for domestic chickens [4] . aPMV-1 has been grouped by virulence phenotype , with lentogenic , mesogenic , and velogenic strains representing increasing levels of virulence , ranging from subclinical infections with moderate respiratory involvement to extensive hemorrhagic lesions and neurological signs [5] . Inactivated vaccines were first made commercially available to the poultry industry in 1946 , but because they provided incomplete protection against ND [6] , they were replaced with live lentogenic NDV vaccines . Although these vaccines reduce disease , they do not always prevent infection and birds can shed both vaccine and challenge strains of the virus [7] , [8] , [9] . aPMV-1 genome size is approximately 15 kb and encodes six genes , which produce nucleocapsid protein ( NP ) , phosphoprotein ( P ) , matrix protein ( M ) , fusion protein ( F ) , hemagglutinin-neuraminidase ( HN ) , and polymerase protein ( L ) [10] , [11] . RNA-editing of P gene creates two additional proteins , V and W [12] . There are nine serotypes of aPMV-1; viruses associated with ND are in serogroup 1 . Within serogroup 1 there are 2 major subdivisions , class I and II , based on phylogenetic grouping of the F gene [13] . Class I aPMV-1 are primarily recovered from waterfowl or samples from U . S . live bird markets , while the isolates from class II are commonly derived from poultry and other avian species [5] , [14] . Eight genotypes of class II aPMV-1 can be identified [13] . Viruses belonging to genotypes I-IV have circulated since the 1930's . Genotype I and II consist of both lentogenic and velogenic viruses and have been associated with ND outbreaks in Australia and North America , respectively [5] , [15] , [16] . These viruses have been attenuated in culture and are used as modified live vaccines [16] . Genotypes V-VIII were first recognized in the mid-1960s [13] and contained only virulent viruses [16] . Genotype V was responsible for the second panzootic of ND in Europe from 1970–1974 and has been detected sporadically thereafter [17] . Genotype VI was described mainly from the Middle East and Asia during the 1980's–1990's [18] and genotype VII and VIII were reported in the 1990's from several countries [19] , [20] , [21] , [22] , [23] , [24] . All genotypes , except IV [16] , are still in circulation . RNA viruses typically have a high mutation rate due to low fidelity and processivity of their polymerase [25] , which coupled with a high replication rate and short generation time [26] lead to high evolutionary rates . In addition , evidence is accumulating that recombination is an important process driving genotype diversity for many RNA viruses [27] , [28] , [29] . Although recombination was not thought to contribute to aPMV-1 evolution [30] , [31] , evidence of recombination in NDV has recently been reported [32] , [33] , [34] . This debate may be due , in part , to the reliance on a single gene for determining virus diversity and phylogeny . Because recombination can lead to the emergence of novel virus strains of unknown virulence [35] , [36] , [37] , [38] , a better understanding of the role of recombination in circulating aPMV-1 is warranted . In this study , we explored how vaccination strategies in poultry farming have shaped the evolution of this important avian virus using complete genome sequences available in GenBank . Specifically our objectives were to 1 ) determine if recombination was evident among full length class II aPMV-1; 2 ) estimate evolutionary rates of each genotype; 3 ) estimate the effective population size of each genotype; and 4 ) determine the selective forces on vaccine-related and nonvaccine-related wild type genotypes . Our results confirm that recombination is an important process in this negative sense RNA virus and that vaccine-related strains have an evolutionary history that is unique from nonvaccine-related strains , which includes distinct evolutionary rates , temporal changes in population size , and selection profiles .
The current phylogenetic classification of aPMV-1 strains is based on either full or partial nucleotide sequence of the M , F , or L genes . To determine if all genes in the viral genome provided consistent phylogenetic profiles , we obtained 54 full length class II aPMV-1 sequences from Genbank and generated nucleotide data sets for each of the six genes and a concatenated sequence of all protein coding regions . Maximum likelihood ( ML ) phylogenies were reconstructed for all sequence data sets under the appropriate nucleotide substitution model selected for each data set ( Figure 1 ) . Each gene tree and the concatenated tree revealed seven distinct genotypes within the class II aPMV-1 . However , genotype affiliations were not congruent among different genes ( Figure 1A ) . While genotype III , IV , V , VI , and VII were monophyletic , genotype I and II showed inconsistent phylogenetic relationships ( Figure 1A and B ) . Three distinct patterns of affiliations of genotype I and II were observed among different gene trees . The NP , HN , L , and concatenated gene trees consistently placed genotypes I and II in a sister clade to other genotypes . In the P gene tree , genotype II formed a basal clade while genotype I clustered with the remaining genotypes and in the M and F gene trees , genotype I was the basal clade and genotype II clustered with the remaining genotypes . Several taxa changed genotype affiliations in different gene trees and all discordant taxa were affiliated with vaccine-related genotype II in some gene trees ( Figure 1C ) . For example , isolate AY562985 ( Cockatoo/14698/Indonesia/1990 ) was affiliated with genotype II in the NP gene tree but with genotype VII in all other gene trees . This isolate occupied a long branch in the genotype VII sequences in the P gene tree . Isolate AY562989 ( Dove/2736/Italy/2000 ) was affiliated with genotype II in the M gene tree but with genotype VI in all remaining trees except the P gene , in which it was an outlier to all other genotypes . Isolate AY225110 ( HB92 isolate V4 vaccine/China ) affiliated with genotype I in M and L gene trees but with genotype II in other gene trees . Isolate EU167540 ( Layer/SRZ03/China/2003 ) affiliated with genotype VII in all gene trees but it occupied a long branch in genotype II in F gene tree . A Shimodaira-Hasegawa test ( SH-test ) provided statistical support of taxon incongruence ( p<0 . 005 ) among the gene trees ( data not shown ) . Phylogenetic incongruence among genes suggests that recombination might play a role in class II aPMV-1 diversity . To further investigate the possibility of recombination among the full length aPMV-1 sequences , we used seven different algorithms implemented in the RDP3 program [39] , [40] . Chimeric NDV vaccine strain EU140955 , which has the genotype II La Sota vaccine strain backbone and the F and HN genes from a contemporary genotype VII virus ( Figure 1C ) , was included as a control to evaluate the prediction capability of the program . The predicted recombination breakpoint ( detected by five methods with p-value<10−5 ) at position 7119 of the concatenated EU140955 matched correctly with the end of a SpeI restriction site of this chimeric strain where HN sequences were inserted from the KBNP-4152 strain . The MluI restriction site used to generate the chimera is within the intergenic region between M and F genes and is not present in our sequences . However , the RDP3 program reasonably identified the 5′ breakpoint at the 8th nucleotide of the F-gene . Two additional recombination breakpoints within this insert were also detected by the GENECONV and Bootscan methods ( p-value 2 . 29×10−3 and 3 . 35×10−2 ) . These corresponded to the positions in the F gene segment of KBNP-4152 strain that were mutagenized to attenuate recombinant strain EU140955 . Thus , we conclude that the RDP3 program accurately identifies recombination if five or more methods have statistical support of p≤10−5 for the breakpoints and we propose that any breakpoints statistically supported with only one or two methods should be carefully interpreted . Using the stringent criteria defined above , a total of five putative recombinant isolates were detected ( Figure 2 ) . Four of these isolates , AY562985 , EU167540 , AY562989 , AY225110 , were those that showed discordant phylogenies described above ( Figure 1C ) , and in each case , some recombinant regions were derived from genotype II sequences . Isolate AY562985 is predominantly genotype VII and had evidence of two recombination events based on RDP3 predictions . We confirmed that regions 508 to 926 ( in NP ) and 927 to 1511 ( NP and P ) are related to genotypes V and II , respectively by partitioning the data sets at the predicted break points and reconstructing a ML phylogeny ( Figure 3 ) . In the region 927 to 1511 , AY562985 had seven unique synonymous substitution sites compared to the other sixteen genotype II sequences in our dataset . Isolate AY562989 , which is predominantly of genotype VI origin , also contained a putative recombinant region spanning positions 2039 to 3225 , which was derived from genotype II and had three unique non-synonymous substitution sites compared to other genotype II sequences . Isolate AY225110 is a chimera of genotype I and II sequences . Compared to genotype II sequences , there were seven unique sites , four of which were non-synonymous substitutions , from the 5′ end of NP to position 2702; two synonymous and two non-synonymous substitution sites in the 3757–7149 fragment; and two synonymous and seven non-synonymous substitution sites within the region from 13758 to the 3′ end of L . For isolate EU167540 , a putative recombination region between position 3753 and 4345 was affiliated with genotype II and had one non-synonymous substitution site compared to other strains . Thus , genotype II , which is used in vaccines globally , has recombined with at least three other class II aPMV-1 strains and the recombinant viruses have been isolated from both domestic and wild birds . The fifth virus identified by all RDP3 methods , DQ485230 , was a genotype VII isolate that contained a small region within the HN gene contributed by genotype III . In addition , a region spanning 1479 to 3751 in P and M appeared to be derived from a different genotype VII virus . Inter-genotype recombination was also detected by fewer than five of the RDP3 methods in DQ486859 ( GM/China ) , DQ485231 ( Guangxi11/China/2003 ) and AF309418 ( Fowl/B1/USA/1947 ) . Genotype VII isolate DQ659677 ( NA-1/China ) contained a 640 bp region within L contributed from genotype VI . The origin of the putative recombinant fragment spanning the M and F genes of EF065682 ( rAnhinga/USA ) could not be determined . These data provide compelling evidence that genotypes II and VII are most commonly associated with recombinant viruses and that both intra- and intergenic recombination events can be detected using full genome sequence analysis . Phylogenies of individual and concatenated genes were reconstructed after the removal of the five putative recombinant isolates and chimeric vaccine strain EU140955 ( Figure 4 ) . Consistent with Figure 1A and 1B , genotypes III - VII clustered together as a monophyletic group in all trees . All taxa were consistently affiliated with a single genotype and there were no long branches associated with any genotype . Two of the three original patterns of phylogenetic affiliation were retained following removal of recombinant sequences . The placement of genotypes based on HN and L was the same with or without recombinants; genotype I and II were sister groups to genotype III , IV , V , VI , and VII ( Figure 1A; Figure 4 ) . All of the remaining trees presented a topology similar to that of the P gene-tree before recombinant removal , which placed genotype I with III-VII . It is noteworthy that in the absence of recombinant sequences , genotype II is never clustered with genotypes III-VII , as was seen with trees based on M and F in the presence of recombinant sequences ( Figure 1A ) . We inferred the evolutionary rates and past population dynamics of class II aPMV-1 using a Bayesian coalescent approach [41] . This analysis was based on all full length genome sequences in the data set which had a date of isolation and excluded the six recombinant sequences . Bayesian estimates of the evolutionary rates of each gene and concatenated coding genome of class II aPMV-1 were between 0 . 98×10−3–1 . 56×10−3 substitutions/site/year ( Table 1 ) . Evolutionary rate estimates under a relaxed clock with HKY+G4 ( Table S4 ) and GTR+G6 ( Table 1 ) substitution models were consistent . The time to the most recent common ancestor ( TMRCA ) of class II aPMV-1 was estimated to be between 114 and 137 years before 2005 , or between year 1868 and 1891 . Bayesian skyline plots ( BSP ) were used to infer how effective population size has changed with time [41] , [42] . All six protein-coding genes and the concatenated genome maintained constant effective population size until the late 1990's ( Figure 5A ) . In 1997-8 there was an abrupt decline in the population with recovery from this event in the early 2000 . To determine if all genotypes exhibited the same population history observed for the composite genotype data set , we repeated the analysis based on 97 dated full length F genes , which provided a larger data set for this analysis . Isolates from genotype III , IV , and V were not included because limited numbers of dated sequences were available . The TMRCA for genotype II was estimated to be 1899±20 years , and the estimated evolutionary rates were between 0 . 3–1 . 1×10−4 substitutions/site/year , making this the slowest evolving aPMV-1 genotype ( Table 2 ) . Genotype I and VI emerged in the early 1960's and had higher evolutionary rates than genotype II . The most recently emerged strain was genotype VII , which dates to the late 1970's and had the highest evolutionary rate . BSP analyses based on the F gene demonstrated that each genotype had a unique population history . Prior to the emergence of genotype VII in the 1970s ( phase i ) , genotype II showed an increase in population size ( Figure 5B ) . After the emergence of genotype VII ( phase ii ) the population size of genotype VI began to increase , while that of genotypes I and VII were relatively constant . Phase iii depicts the time of the population bottleneck observed in Figure 5A , which was based on all genes in the composite genotype data set . Only genotypes I and VI show a trend for decreasing population size during this time . The last decade has been the most dynamical for the four genotypes of aPMV-1 ( phase iv; Figure 5B ) . Genotypes I and VII showed a marginal increase in effective population size followed by a decline; genotype I has continued to decline , whereas genotype VII appears to have stabilized . Genotype VI showed continuous decline in population size in phase iv ( Figure 5B ) . Although estimates of population sizes for genotypes I , VI , and VII , have some degree of overlaps at the 95% posterior limit , genotype II shows no sign of reduction in effective population size since its origin ( Figure 5B ) . We compared the selection profiles on protein coding genes of genotypes I and II , which include strains that are circulating in the face of homotypic vaccination pressure ( designated as the vaccine-related group ) , and genotypes III-VII ( designated as the nonvaccine-related strains ) ( Table 3 ) . Overall , the global rate of non-synonymous to synonymous substitutions ( dN/dS ) for all protein coding genes were less than 1 , indicating purifying selection has been the major driving force in the evolution of class II aPMV-1 viruses . There were no codons identified to be under positive selection in M and L genes in either group ( Table 3 ) . However , there was a clear difference in the codon based selection profiles of N , P , F and HN genes between the vaccine- and nonvaccine-related groups . In the vaccine-related group , only the surface protein encoding genes , F and HN , had positively selected codons . Both groups had a single site identified in F; these were at codon 115 within the F0 cleavage site for the vaccine-related group and codon 28 in the signal peptide for the nonvaccine-related group . There were 3 positively selected sites identified in the HN gene in the vaccine-related group and one in the non vaccine-related group . In contrast , the P gene of nonvaccine-related genotypes III-VII had three sites predicted to be under positive selection but no sites were identified in the vaccine-related group . Thus , selection is focused on HN in vaccine-related groups and on P in nonvaccine-related genotypes and there are no shared sites under positive selection between the two groups .
Our study explored the forces shaping the evolutionary history of class II aPMV-1 using available full genome sequences . We demonstrated that genotype affiliations based on individual genes and concatenated full length genomes of class II aPMV-1 were not consistent . This may account for discrepancies reported for genotype groupings that are based on partial or complete sequences of a single gene [13] , [18] , [43] . Topological incongruence among the gene trees reflects different evolutionary histories of each gene [44]; recombination is the most plausible explanation for this . The role of recombination in the evolution of aPMV-1 , and negative sense RNA viruses in general , has been debated . For example , Sakaguchi et al . [30] and Toyoda et al , [31] reported consistent topological placement of different NDV strains in both F and HN gene trees . Seal et al . [45] also reported that there was no evidence of recombination among NDV M gene sequences . Although recombination is more common in positive-sense RNA viruses and can be explained by several genetic mechanisms [46] , there is increasing evidence of homologous recombination in several non-segmented negative-sense RNA viruses [32] , [47] , [48] , [49] , [50] . Our approach differs from those used in previous studies of aPMV-1 evolution because we evaluated full genome sequences and tested individual recombinant regions with phylogeny-based incongruence tests . Thus , we show that all class II aPMV-1 genes have evidence of recombination breakpoints , that multiple recombination events are discernable in some isolates , and that both intragenic and intergenic recombination events are evident . We considered the possibility that the recombinants detected in our analysis were the result of laboratory artifacts , as has been previously suggested [32] , [51] . Laboratory contamination is of concern because vaccine derived strains contributed the majority of the recombinant regions and these strains might have been present in laboratories sequencing field aPMV-1 isolates . The presence of unique nucleotide substitutions in the recombinant regions compared to the comparable region of the predicted parental genotypes suggests that these regions did not arise due to contamination with vaccine strains deposited in the sequence databases . Our identification of recombinants derived from vaccine strains indicates that birds can be simultaneously infected with the live virus vaccine and other circulating aPMV-1 genotypes . Indeed vaccination is reported to protect poultry from disease but not always from infection with other strains [7] , [8] , [9] . Suboptimal vaccination strategies could also lead to birds becoming infected with both a vaccine strain and circulating genotype , which can alter viral virulence [16] . This is an important issue for poultry management . Lack of vaccine efficacy has not frequently been reported in the United States but other countries such as Nigeria [52] , Korea [53] , Taiwan [54] , and China [34] , [55] have experienced vaccine failures . Recombination between wild type virus and vaccine strains is not unique to aPMV-1; vaccine recombinants of bovine viral diarrhea virus ( associated with fatal mucosal disease ) [35] , poliovirus ( associated with paralytic poliomyelitis ) [36] , [37]; and infectious bursal disease virus [38] have all been reported . This raises concerns that modified live virus vaccines , although efficacious , may facilitate emergences of new strains with unpredictable phenotypes through recombination with circulating viruses . The evolutionary rates presented here for class II aPMV-1 are compatible with the rates estimated for other RNA viruses ( e . g . [56] , [57] , [58] ) , suggesting that class II aPMV-1 is also a rapidly evolving RNA virus . The relatively lower evolutionary rate for genotype II is consistent with the rate that was previously reported for avirulent NDV [33] . The larger effective population size , which counters the impact of genetic drift , is a possible explanation for lower evolutionary rates of genotype II . Based on these rate estimates , the TMRCA of this virus is estimated to be between 1868–1891 , which is earlier than the first recorded outbreak of ND in Indonesia and England in the 1920's [2] . However , our data are in line with observations by Macpherson [59] , who suggested that an outbreak of disease in domestic birds that occurred in Northwest Scotland from 1897–1898 was due to NDV . The demographic history of class II aPMV-1 determined by Bayesian skyline plots indicated that there was an abrupt decline in population size during 1997–98 . Although the factors responsible for such an abrupt decline in class II aPMV-1 are not known , the impact of a severe El-Nino event during that time frame [60] or the slaughter of millions of domestic fowl during the first outbreak of H5N1 avian influenza virus in 1997 [61] , [62] could be possible explanations . In contrast to the other genotypes , genotype II was not impacted by factors causing population decline in the late 1990s . We expected that vaccination , which started worldwide in the 1950s , should have limited the number of susceptible avian hosts , thus causing a bottleneck for this genotype . However , the impact of NDV vaccination is not seen in the BSP . It is possible that the data available in GenBank is insufficient to capture the population history of this genotype . However , a plausible explanation for the absence of a population bottleneck could be that genotype II NDV is maintained as an asymptomatic infection because it is continually introduced to susceptible populations as a modified live vaccine . Vaccination effectively prevents birds from developing disease when exposed to a virulent strain , but does not prevent shedding [7] . Thus , the number of available susceptible hosts may not dictate genotype II population demographics . Although overall all genes of class II aPMV-1 are evolving under purifying selection consistent with other paramyxoviruses [33] , [56] , [63] , distinctive profiles of positively selected codons were shown in both vaccine- and nonvaccine-related groups . Notably , P gene had the highest dN/dS ratio of any gene and had three sites predicted to be under selection in the nonvaccine-related group; there were no sites under selection in P gene in the vaccine related group . In contrast , HN had three sites predicted to be under selection in the vaccine related group . Of interest , codon 115 in the F cleavage domain is positively selected only in vaccine-related groups . Previous studies have reported that a single amino acid substitution at codon-115 , which falls within the F0 cleavage site , resulted in a dramatic change from an avirulent infection to highly virulent NDV [64] , [65] , [66] . In contrast to the results reported in the present study , Miller et al [33] did not identify codon115 in F gene under positive selection . This is likely due to differences in the data sets because factors such as sequence length , sequence divergence , and the number of sequences can determine the ability to detect positively selected sites [67] . Thus , the vaccine-related genotypes I and II maintain a phenotypic mixture of strains with different infection and pathogenic potential and selection profiles .
A total of 54 complete genome sequences of class II aPMV-1 representing different avian hosts , geographic regions , year of isolation , and genotypes ( based on previous published phylogenetic grouping ) were retrieved from GenBank . The coding genome sequences were aligned using MEGA version 4 [68] . Six separate coding gene sequences datasets ( for NP , P , M , F , HN and L genes ) were generated ( see Table S1 ) and a concatenated genome sequence from these six coding gene sequences was generated using Mesquite version 1 . 12 ( http://mesquiteproject . org ) . Appropriate model of nucleotide substitution for each dataset was selected by the hierarchical likelihood ratio test implemented in Modeltest version 3 . 7 [69] . Maximum likelihood ( ML ) trees were reconstructed for all data sets using the heuristic search option , implementing stepwise addition with 100 random addition replicates and tree bisection-reconnection branch swapping in PAUP* version 4beta10 [70] and PHYML 3 . 412 [71] with 100 non-parametric bootstrapping replicates analyses . The inferred trees were visualized with FigTree version 1 . 12 ( http://tree . bio . ed . ac . uk/software/figtree/ ) and the congruency of topology placement of class II aPMV-1 genotypes based on each gene and concatenated genome was tested using the Shimadoira , Hasegawa ( SH ) test [72] implemented in PAUP . The concatenated tree was constrained and tested versus other gene trees . The recombination predictions of the concatenated genome sequences were conducted with a suite of programs within the RDP3 package [39] , [40] . The individual programs RDP [39] , GENECONV [73] , Bootscan [40] , Maximum Chi [74] , Chimaera [75] , SiScan [76] and 3Seq [77] , were implemented for the analysis . Since no single program provided optimal performance under all conditions , any event supported by five or more methods with p-values ≤10−5 was the criteria used for positive recombination breakpoints identification . The breakpoint position and the putative parental sequences were also determined . Twenty six full length genome sequences for which year of isolation was available were used to infer evolutionary rate and dates using BEAST version 1 . 4 . 8 [41] . Demographic history of a population/species using multi-locus data , even from a small number of individuals , can precisely recover past bottlenecks in population size that cannot be characterized by analysis of a single locus [78] . Given this fact , estimates based on the whole genome sequence data are expected to be more reliable . To determine the population history of individual genotypes , 149 complete F gene sequences , which had dates of collection , were retrieved from GenBank . Phylogenetic analyses were done as described previously ( Table S2 ) . From the ML tree , a total of 97 isolates from genotype I , II , VI , and VII were selected to infer evolutionary rates and population dynamics . These included all available dated full length sequences from Genotypes I ( 24 sequences ) , II ( 28 sequences ) , and VI ( 23 sequences ) . The majority ( 95% ) of genotype VII sequences were of Chinese origin . Thus , we included the seven non-Chinese sequences and picked an additional 15 sequences based on phylogenetic diversity to represent genotype VII in our analyses . The evolutionary rate ( nucleotide substitutions per site per year ) of each gene and concatenated genome was estimated using the Bayesian Markov chain Monte Carlo analyses ( independent assumption of codivergence ) . Substitution models of both HKY + G4 and GTR + G6 with estimated base frequencies , gamma and invariable site portion were used , with uncertainty in the data reflected in the 95% high-probability density ( HPD ) intervals . Strict clock and uncorrelated exponential ( UCED ) relaxed clock models were attempted independently , and the best-fit clock model was determined to be UCED based on the Bayes Factor calculated from their posterior distributions ( Table S3 ) . The Coalescent Bayesian skyline plot ( BSP ) was used to infer the past population dynamics . The BSP was constructed using the growth rate and demographic parameters from the selected best-fit models . Bayesian Markov Chain Monte Carlo ( BMCMC ) analyses were run for 5–10×108 generations depending on each dataset . Convergence of trees was checked using Tracer v1 . 4 . 1 ( http://beast . bio . ed . ac . uk/Tracer ) . Selection analyses were done based on datasets without putative recombinant sequences because recombination can result in falsely identifying positive selection [79] . The datasets were split into a vaccine-related group , which included strains from genotypes I and II , and nonvaccine-related group , which included strains from genotypes III , IV , V , VI , VII ( see Table S2 ) . Positively selected codons were detected using Fixed- Effect Likelihood ( FEL ) via the Datamonkey website ( http://www . datamonkey . org/ ) and ML approach implemented in CODEML ( PAML package version 3 . 15 ) [80] . For FEL analysis , p-values less than 0 . 05 were used to support positive selection . For PAML analysis , the likelihood ratio test was used to compare M1a , M7 and M8a models that assume no positive selection ( ω<1 ) with those M2a and M8 models that assume positive selection ( ω>1 ) [81] . | Modified live virus ( MLV ) vaccines have been effective in reducing disease burden and economic loss caused by Newcastle Disease ( ND ) in domestic poultry . Because the vaccine is a live virus , it is transmissible among birds . Thus , vaccination strategies have the potential to impact the evolutionary genetics of wild type strains of aPMV-1 including those that cause ND . In this report , we provided evidence that viruses isolated from wild and domestic birds have recombined with vaccine strains , because vaccinated birds are protected from disease but not infection with other strains of aPMV-1 . Despite the use of vaccines since the 1950s , the population size of the strain from which the most widely used vaccine was derived has steadily increased . In contrast , other contemporary genotypes , which emerged in the 1960s , experienced a decline in population size in 1998 , which may reflect a change in poultry farming practices or disease . Vaccination imposed a unique selection profile on the genotypes derived from the vaccine-related strains when compared with nonvaccine-related strains . Although modified live viruses are important for controlling Newcastle Disease , the potential of vaccination strategies to change viral diversity and population dynamics should be considered . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"evolutionary",
"biology/microbial",
"evolution",
"and",
"genomics",
"virology/virus",
"evolution",
"and",
"symbiosis"
] | 2010 | The Effect of Vaccination on the Evolution and Population Dynamics of Avian Paramyxovirus-1 |
DNA sequencing identifies common and rare genetic variants for association studies , but studies typically focus on variants in nuclear DNA and ignore the mitochondrial genome . In fact , analyzing variants in mitochondrial DNA ( mtDNA ) sequences presents special problems , which we resolve here with a general solution for the analysis of mtDNA in next-generation sequencing studies . The new program package comprises 1 ) an algorithm designed to identify mtDNA variants ( i . e . , homoplasmies and heteroplasmies ) , incorporating sequencing error rates at each base in a likelihood calculation and allowing allele fractions at a variant site to differ across individuals; and 2 ) an estimation of mtDNA copy number in a cell directly from whole-genome sequencing data . We also apply the methods to DNA sequence from lymphocytes of ~2 , 000 SardiNIA Project participants . As expected , mothers and offspring share all homoplasmies but a lesser proportion of heteroplasmies . Both homoplasmies and heteroplasmies show 5-fold higher transition/transversion ratios than variants in nuclear DNA . Also , heteroplasmy increases with age , though on average only ~1 heteroplasmy reaches the 4% level between ages 20 and 90 . In addition , we find that mtDNA copy number averages ~110 copies/lymphocyte and is ~54% heritable , implying substantial genetic regulation of the level of mtDNA . Copy numbers also decrease modestly but significantly with age , and females on average have significantly more copies than males . The mtDNA copy numbers are significantly associated with waist circumference ( p-value = 0 . 0031 ) and waist-hip ratio ( p-value = 2 . 4×10-5 ) , but not with body mass index , indicating an association with central fat distribution . To our knowledge , this is the largest population analysis to date of mtDNA dynamics , revealing the age-imposed increase in heteroplasmy , the relatively high heritability of copy number , and the association of copy number with metabolic traits .
As the “cellular power plant” , each mitochondrion encodes some of its constituent proteins in resident mitochondrial DNA ( mtDNA ) . Human mtDNA is a circular molecule of 16 , 569 bases , and mutations that have become fixed in the sequence of every mtDNA may cause several genetic diseases . Accumulation of variants during growth has been suggested to have an important role in aging and cancer[1–3] . However , although the degree to which mtDNA varies heritably and somatically has been much discussed , it has not been analyzed on a population basis . Modern high-throughput sequencing facilitates systematic identification of common and rare DNA variants , including many associated with complex diseases and quantitative traits[4 , 5] . To extend comparable sequence analysis to mtDNA , an important step in the analysis pipeline must be modified . Variant identification for nuclear DNA using sequencing data has been greatly refined , typically using a likelihood-based model to combine information from sequence reads and predict the genotype with the highest posterior probability at a site[6 , 7] . But mtDNA analysis is one of a number of instances ( see Discussion ) in which scoring allelic variation is more complicated , because there are more than the three discrete genotype states found in nuclear DNA . Instead of having two copies of each autosome ( chromosomes 1–22 ) , human cells have 100–10 , 000 separate copies of mtDNA , and different copies of mtDNA may differ in DNA sequence at any base . Thus , the conventional nuclear DNA variant caller must be adapted to identify mtDNA variants . We describe an algorithm specifically tailored to identify mtDNA variants from sequencing data , and apply it to 2 , 077 participants in the SardiNIA project[8] . We analyze both homoplasmies ( conventionally defined as variants affecting all of the mtDNA copies within a cell compared to a standard sequence ) and heteroplasmies ( defined as the presence of a mixture of more than one type of mtDNA within a cell ) . We examine transition/transversion ratios , coding vs . noncoding changes , and changes with age . Analyses are extended with a method to assess mtDNA copy number . Copy number is a critical determinant of mitochondrial function and has been proposed as a potential biomarker for disease . For example , studies have shown that elevated mtDNA copy number is associated with cancer risk[9 , 10] . Given that there are two copies of autosomal DNA in a cell , our method infers mtDNA copy numbers based on the observed ratios of sequence coverages between mtDNA and autosomal DNA . We estimate the heritability of copy number and show its correlations with gender , age , and waist circumference and waist-hip ratio .
All participants gave written informed consent , with protocols approved by institutional review board of the National Institute on Aging ( 04-AG-N317 ) . Recent analyses of mtDNA variants[11 , 12] have taken an approach that determines homoplasmic and heteroplasmic sites directly based on allele counts of sequence reads . This approach does not account for error rates in sequence reads , and hence potentially results in both false positive and false negative variant calls . We propose a likelihood-based model that takes into account the sequencing error rate at each base in each sequence read . The algorithm builds on the conventional autosomal DNA variant callers[6 , 7] , but is modified to allow for allele fractions ( i . e . , heteroplasmic levels ) at a variant site to vary across individuals . The mtDNA variant caller was applied to whole-genome sequence data of 2 , 077 individuals selected among 6 , 921 participants in the “SardiNIA” study of the genetics of quantitative traits in the Sardinian founder population [8] . The details on the selection of individuals to be sequenced are in S1 Text . Sequence data were generated at the University of Michigan Medical School Core Sequencing Lab . DNA was extracted by a standard salting-out method from whole blood samples after a red blood cell lysing step . Libraries were generated from 3–5 μg of genomic DNA using sample preparation kits from Illumina and New England Biolabs . Paired-end sequence reads ( typically , 100 to 120 bp in length ) were generated with Illumina Genome Analyzer IIx and Illumina HiSeq 2000 instruments . Samples were sequenced to an average depth of 4 . 2X . Reads were aligned to the human reference genome ( GRCh37 assembly with decoy sequences , as available in the 1000 Genomes Project ftp site , ftp://ftp . 1000genomes . ebi . ac . uk ) using BWA ( version 0 . 5 . 9 ) , allowing at most 5 differences ( mismatches or gaps ) , and trimming read tails with average base quality <15 . After alignment , base qualities were recalibrated and duplicate reads were flagged and excluded from analysis . We reviewed summary metrics generated using QPLOT and verifyBamId for each aligned sample , to remove samples with low sequencing depth , poor coverage of regions with high or low GC content , or evidence for sample contamination . We then extracted from all sequence reads those that were uniquely mapped to the mtDNA reference genome by bwa with a mapping quality score ≥20 ( i . e . , the theoretical probability of wrong alignment ≤1% ) . The mtDNA variant caller was applied only to these mtDNA reads . Fig 1 outlines the pipeline for aligning sequence reads to the whole genome reference , extracting mtDNA reads , and applying the mtDNA variant caller taking into account the circularity of the mtDNA genome . In implementing the likelihood-based model to identify mtDNA variants , we also applied quality control filters to help avoid the inclusion of false variants because of sequencing errors . At a position of interest in mtDNA , we considered only reads with base sequencing error rate ≤1% ( i . e . , recalibrated base quality score ≥20 ) . We also applied sequencing depth filters: we required an overall mtDNA median depth > 100 for an individual to be included in the analysis; and at each base of interest for variant calling , we required a raw depth ≥40 and a depth ≥10 after base quality score filtering at 20 . To call a heteroplasmy , we further required that 1 ) all alleles of the called genotype are observed at least once in both forward and reverse strand sequence reads , and 2 ) the minor allele fraction ( MAF ) for an individual is ≥ 4% . The MAF threshold was chosen based on simulations in which we mimicked the SardiNIA sequencing experiments and simulated similar coverage data with reads of comparable quality scores ( see S1 Text for details ) . An MAF threshold of 4% corresponded to an empirical false discovery rate ( i . e . , the proportion of false heteroplasmies among all identified heteroplasmies ) of 2% . We also used data from a deeply sequenced parent-child trio ( ~80-fold average coverage for nuclear DNA and ~6 , 000-fold average coverage for mtDNA ) to evaluate the accuracy of variant calling for the same three individuals from the low-pass sequencing data . Using the results from deep-sequencing data as gold standards , we confirmed all the heteroplasmies identified in the three individuals by our variant caller with the 4% MAF cut-off ( We found no false negatives in the child or the mother of the trio , but did see one false negative in the father ) . By contrast , if we lowered the MAF cut-off to a less stringent threshold of 1 . 6% , we observed an average false discovery rate of 30% . Looking at deep sequencing data from a few individuals could not provide a definite guideline about the MAF cut-off , but supports well the choice of a 4% MAF cut-off for this dataset . We have also considered the possibility that nuclear copies of parts of mtDNA sequence ( i . e . , nuclear mitochondrial DNA , or NUMTs ) might be the source of false positives for heteroplasmies . In our analysis , as mentioned above , we included only reads that were uniquely mapped to mtDNA . With further analyses on sampled individual cases , we found that any representation of NUMTs is minimal , and should therefore not restrict the utility of the method ( See S1 Text for a detailed discussion ) . We used publicly available online software HaploGrep ( http://haplogrep . uibk . ac . at/ ) to classify SardiNIA individuals into different haplogroups based on mtDNA , and ANNOVAR[14] to annotate the called variants on mtDNA and assess whether homoplasmies and heteroplasmies show different distributions in functional categories . Assuming autosomal and mtDNA are handled and sequenced with no significant differences , average sequencing coverage should be proportional to DNA copy number for autosomal and mtDNA: mtDNAaveragecoveragemtDNAcopynumberpercell=autosomalDNAaveragecoverageautosomalDNAcopynumberpercell As a proof of principle , we looked at the average depth of coverage across the 22 autosomal chromosomes for 100 randomly selected individuals and observed that as expected , sequencing depth was largely flat across 22 chromosomes for each individual ( S2 Fig ) . Because there are two copies of autosomal DNA in a cell , we could infer the mtDNA copy number by: mtDNAcopynumberpercell=mtDNAaveragecoverageautosomalDNAaveragecoverage×2 ( 3 ) We used SAMtools ( http://samtools . sourceforge . net/ ) to obtain the coverage of each base in the genome from the aligned bam[15] files . The average coverages for autosomal DNA and mtDNA were then calculated accordingly . We applied our computational method to the same whole-genome sequencing data from 2 , 077 Sardinians and estimated mtDNA copy number for each sample . We also used a NovaQUANT Human Mitochondrial to Nuclear DNA Ratio Kit ( EMD Chemicals Inc . ) to validate a random group of 18 samples experimentally by qPCR ( see S1 Text for more details ) . We tested for age and gender effects on the mtDNA copy number , and also assessed any association of mtDNA copy number with eleven quantitative traits collected for the cohort that include 5 anthropometric traits ( height , weight , BMI , waist circumference , and waist-hip ratio ) , 2 frailty traits ( walking speed and grip strength ) , and 4 lipid traits ( HDL-cholesterol , LDL-cholesterol , total cholesterol , and triglycerides ) . In addition , POLY software ( http://www . sph . umich . edu/csg/chen/public/software/poly/ ) was used to estimate the heritability of mtDNA copy number based on the known family structure in the SardiNIA cohort . We note that a similar framework has been suggested by Chu and colleagues[16] to estimate copy number . However , their method identified only reads mapped to mtDNA and counted all the remaining reads as mapped to nuclear DNA . As a result , it would include unmappable reads in calculations and would thus overestimate the nuclear DNA sequence coverage . Furthermore , in calculating the effective length of human genome , their method did not exclude the regions in the nuclear DNA that could not be covered by sequencing . Our computational method avoids those pitfalls and hence should provide more accurate copy number estimates . Here we also validate the method with Q-PCR and implement it to estimate mtDNA copy number in a large-scale population study . Software programs implementing our methods are freely available at http://lgsun . irp . nia . nih . gov/hsgu/software/mitoAnalyzer/index . html .
Applying our mtDNA variant calling program to the cohort of 2 , 077 Sardinians , we identified an overall average of 22 . 2 homoplasmies and 0 . 73 heteroplasmies per individual . S1 Table and S2 Table provide complete lists of homoplasmies and heteroplasmies , respectively . S3 Fig shows histograms of the numbers of homoplasmies and heteroplasmies per individual . The distribution of the number of homoplasmies per individual was bimodal . One group showed relatively fewer homoplasmies ( mode of 11 ) , whereas the other had a mode of 32 . Compared to the current reference phylogenetic tree , the two modes represented different European haplogroups . The former fell into the HV subgroup ( the mitochondrial reference genome sequence rCRS also belongs to this subgroup , accounting for the smaller number of homoplasmies ) and the latter was predominantly correlated with several other clades , including J , T and K subgroups . Looking at the sharing of the mtDNA variants among the 2 , 077 individuals , we observed significantly higher sharing of homoplasmies than heteroplasmies ( Fig 2 ) . For example , 10% of homoplasmies were shared by more than 100 individuals and 2% of homoplasmies were shared by more than 500 . By contrast , only ~1% of heteroplasmies were shared by more than 20 individuals . For both homoplasmies and heteroplasmies , we further investigated numbers of transition and transversion base changes ( Fig 3 ) . The transition/transversion ratio was greater than 10 for both homoplasmies and heteroplasmies , which is far higher than the ratio of 2 . 1 in human nuclear DNA ( 2 . 19 in the sequenced Sardinians; see Discussion ) . Homoplasmies and heteroplasmies showed very similar patterns of base changes ( Fig 3; see Discussion ) . Using ANNOVAR to annotate the identified variants , we grouped them into four functional categories: 1 ) intergenic; 2 ) structural RNA ( rRNA and tRNA ) -encoding; 3 ) synonymous protein-coding; and 4 ) non-synonymous protein-coding . As shown in Fig 4 , compared to heteroplasmies , homoplasmies are less likely to be RNA-encoding or non-synonymous ( chi-square test p-value = 3×10−13 ) , consistent with the notion that heteroplasmies represent new mutations and that natural selection makes it challenging for deleterious variants to become fixed as homoplasmies . To assess any age effect on the number of mtDNA variants , we used a Poisson loglinear model to test the association between the number of variants and age among unrelated individuals ( see S1 Text for the procedure to select unrelated individuals from the whole cohort ) . We observed no relationship between age and the number of homoplasmies ( S4 Fig ) . By contrast , with a minor allele fraction threshold of 4% , we observed a significant increasing trend of the number of heteroplasmies with age ( p-value = 6 . 2×10−5 ) . The increasing slope is small , yielding an average increase of ~1 heteroplasmy between ages 20 and 90 ( Fig 5 ) , but the slope increased and became more significant when we repeated the analyses with MAF thresholds of 1 . 6% and 3% ( Fig 5; p-values equal to 1 . 1×10−15 and 2 . 7×10−11 , respectively ) . When lowering the MAF threshold , one expects to include more true heteroplasmies together with more false positives . However , false heteroplasmies have no likely relationship with age , whereas the additional true heteroplasmies will strengthen the trend . On the other hand , when we raised the MAF threshold to 5% and 6% ( i . e . , applying more stringent thresholds ) , the trend remained but p-values were less significant ( 8 . 9×10−4 and 0 . 017 , respectively ) . Thus , given the minor allele fraction thresholds applied , there is appreciable accumulation of heteroplasmies with age ( see Discussion ) . We used 333 parent-child trios included in the cohort to investigate the sharing of mtDNA variants between parents and their children and to assess features of the inheritance pattern of mtDNA variants . Children and their mothers share essentially all homoplasmies . We observed 7 , 273 homoplasmic sites in 333 children , among which 7 , 238 ( 99 . 5% ) were also observed in their mothers ( Table 1 ) . At the same time , 2 , 940 ( 40 . 4% ) homoplasmies were also observed in their fathers . This observation is not incompatible with maternal inheritance , because many homoplasmies are shared across the Sardinian population , and thus children and their fathers could share many by chance . This is further supported by the observation that fathers also shared ~40% of mothers’ homoplasmies ( Table 1 ) . Concerning heteroplasmies , we observed 207 heteroplasmic sites in 333 children , among which 66 ( 31 . 9% ) were observed in their mothers and 1 ( 0 . 4% ) was observed in their fathers . These results indicate that children inherit a proportion of heteroplasmies from their mothers , whereas new heteroplasmies arise both in their own and in their mothers’ lymphocytes during life . The sharing of a single heteroplasmy between a child and its father could well have occurred by chance or may represent rare true patrilineal transmission . Further discussion on the inheritance of heteroplasmies can be found in S1 Text . Estimating mtDNA copy number for each sample , we observed a range from 50 to 350 , with most individuals between 75 and 150 ( with mean of 111 . 5 and standard deviation of 25 . 0; histogram in S5 Fig ) . To assess copy number by a standard biochemical assay for comparison , we carried out Q-PCR experimental validation for 18 randomly chosen samples . The Q-PCR measures have considerable intrinsic variability of their own; but taking an average of two experiments , the computational estimates and experimental measurements shared a similar range ( scatterplot in S6 Fig ) , with a correlation of 0 . 82 . We observed a significant gender effect on mtDNA copy number: on average , females have 6 . 7 ( 6 . 2% ) more copies of mtDNA than males ( p-value = 1 . 6×10−9 ) . After adjusting for gender and average sequencing coverage , the estimated mtDNA copy number decreases significantly with age ( p-value = 2 . 7×10−6 , with an expected 1 . 5 copy number decrease for every 10 years of age increase , S7 Fig ) . When testing the association of mtDNA copy number with a set of 11 quantitative traits , after adjusting for age , gender , and average sequencing coverage , we observed that mtDNA copy number is significantly associated with waist circumference ( p-value = 0 . 0031; after adjusting for multiple tests , Bonferroni-corrected p-value = 0 . 034 ) and waist-hip ratio ( p-value = 2 . 4×10−5; Bonferroni-corrected p-value = 2 . 6×10−4 ) , but not with BMI ( p-value = 0 . 42 ) or two frailty-related traits ( walking speed , p-value = 0 . 32; grip strength , p-value = 0 . 11 ) . We also estimated the heritability of mtDNA copy number as 54% , implying strong genetic regulation of mtDNA level .
The method presented here improves mtDNA variant calling in two ways . First , it directly models sequencing error rates ( i . e . , uncertainty ) in likelihood calculations , and therefore more accurately identifies mtDNA variants and estimates minor allele fractions of heteroplasmies in individuals . Second , the pipeline is adapted to the circular mtDNA genome , aligning sequencing reads to two linear mtDNA reference genomes by a “double-alignment” strategy . This strategy greatly increases the coverage in the hypervariable “junction” region—very important , for example , for phylogenetic studies—that otherwise would have very poor coverage , and can also be useful in aligning sequence reads of other circular genomes . Recently , several pioneering studies have also used next-generation sequencing technologies to assess mtDNA heteroplasmies . These include several studies applying deep sequencing to amplified mtDNA [11 , 12 , 17 , 18] and others performing whole-genome sequencing and then extracting mtDNA sequence reads [4 , 13] . Most of these studies used a set of filters to account for sequencing errors and technical artifacts in heteroplasmy identification , which generally included number of mismatches , mapping quality , base quality , minimum depth , double strand validation , and minor allele fraction ( MAF ) . S3 Table compares criteria for calling heteroplasmy among these studies and ours . Not surprisingly , depending on the sequencing coverage , different studies applied different MAF thresholds in calling heteroplasmies , which makes a direct comparison of results impossible . To do a fair comparison , we applied to five studies a common MAF threshold of 10%—the maximum threshold used in any of the studies . In our study , at least one heteroplasmy is observed in 21 . 8% of individuals . This is very comparable to the value of 24 . 4% found by Li and colleagues [12] in 133 individuals sequenced with a mean coverage of 85X , and to 23 . 1% observed by Rebolledo-Jaramillo and colleagues [18] after sequencing amplified mtDNA from 39 mother-child pairs to 20 , 000-fold coverage with additional steps to exclude PCR and sequencing errors . In addition , their high-quality data showed that on average one person possessed about 1 heteroplasmy with MAF > 1% , which is in good agreement with our findings of an average of 0 . 73 heteroplasmies per person with MAF > 4% based on direct sequencing of total DNA . By contrast , we note that the 1000 Genomes Pilot Project sequenced 163 individuals and found that 45% possessed heteroplasmies with an MAF cut-off at 10% [4] . Similarly , Ye et al . [13] found the prevalence of heteroplasmy to be 44 . 4% in the full 1 , 085 member cohort of 1000 Genomes Project . And a sixth study sequenced 40 HapMap individuals and found 65% possessing heteroplasmies with an MAF cut-off at 9% [19] . As a possible explanation of the difference in heteroplasmy prevalence between the two sets of studies , it is suggestive that the first three studies ( including ours ) used DNA samples extracted directly from cells or tissues , whereas the second set of three studies used DNA samples from transformed ( lymphoblastoid ) cell lines . Heteroplasmies could thus arise during the expansion of cell lines in culture; and if so , it may be prudent to assess mtDNA content level and heteroplasmies from untransformed cells . The extent of recovery of alleles in the Sardinian cohort at ~180-fold average coverage ( see below ) is robust enough to infer some other salient characteristics of mtDNA variation . The analyses indicate that the inheritance pattern of mtDNA variants can be explained by maternal inheritance of homoplasmies and a portion of heteroplasmies , with further heteroplasmic sites arising during life in both children and mothers up to the time of cross-sectional sampling of the cohort . This is consistent with observations from deep sequencing data for two CEPH families[11] . By looking at 333 trios , our study showed that 31 . 9% of heteroplasmies in children were inherited from their mothers , consistent with the study of Rebolledo-Jaramillo et al . [18] , who found an inherited proportion of heteroplasmies ( with MAF > 1% ) of 28 . 9% and 27 . 7% in blood and buccal cells , respectively ( results inferred from their Supplementary Materials ) . Considering the nature of the allelic variants , the transition/transversion ratios for homoplasmies and heteroplasmies are both about 10 , about 5-fold higher than the ratio observed in human nuclear DNA ( ~2 . 1 ) . This is consistent with the transition/transversion ratio estimated for mitochondrial mutations occurring at a frequency ≤ 1%[20] , and is also consistent with the hypothesis that misincorporation by DNA polymerase gamma and the deamination of nucleotides are major sources of the base changes[20] . Homoplasmies and heteroplasmies share similar distributions among the different types of base change ( Fig 3 ) , indicating that the mechanism creating base changes is likely to be the same for homoplasmies and heteroplasmies . In addition , compared to homoplasmies , heteroplasmies are enriched at RNA encoding and non-synonymous protein-coding sites , consistent with new mutations being more likely to be detrimental than those that survive selection during fixation . As for the much-discussed possibility that heteroplasmies may accumulate with age , we did observe a significant trend . To the best of our knowledge , this is the first direct analysis relating mtDNA variant number with age in a large-scale population study [We note that Rebolledo-Jaramillo et al . [18] compared 39 mother-child pairs and show a similar positive association in mothers but not in children , perhaps because the children were too young or the sample size was too small] . However , estimation of the true rate of increase requires further identification of lower-level heteroplasmies . As mentioned above , to recognize true heteroplasmies against a background of sequencing errors , we applied a conservative minor allele fraction threshold in calling heteroplasmic sites and required that the minor and major alleles are observed from both forward and reverse strands . These filters were necessitated by the level of average mtDNA coverage in each sample . The filters could be relaxed in several ways–for example , by increasing read coverage by deep sequencing of total DNA . As another example , Rebolledo-Jaramillo et al . [18] lowered the heteroplasmy threshold to 1% by analyzing mtDNA purified and amplified before sequencing . That approach , however , requires additional experimental procedures and eliminates the possibility of the simultaneous determination of mtDNA copy number . Additional approaches could use family structures to expand available reads or sequence DNA from single cells , which would assess only 100–1 , 000 mtDNA molecules , instead of that number augmented by millions of copies from all the cells in a lymphocyte sample . The algorithm could extend the power of analysis using other study designs as well . In particular , longitudinal analysis of a study cohort would be informative , but would require repeated sampling over time . Several approaches could also improve the variant calling algorithm itself . Calling variants by considering multiple individuals jointly or by considering linkage disequilibrium ( LD ) structure among variants would probably add little to variant calling accuracy , because we observed that heteroplasmies are rarely shared among individuals ( Fig 2 ) . But extensions that detect insertions and deletions ( i . e . , indels , etc . ) as well as single base variants could provide a more complete catalogue of variation . In assessing mtDNA copy number based on the sequencing coverage ratio between mtDNA and autosomal DNA , we assume that no significant differences are generated between reads of mtDNA and autosomal DNA during the sequencing and processing steps that include genomic DNA fragmentation , adapter ligation , PCR amplification , sequencing , and sequence alignment/mapping . The relatively high correlation with Q-PCR measurements supports this assumption , and the relatively high heritability of copy number ( 54% ) is also in accord with the reliability of the estimates . We were thus encouraged to use inferred mtDNA copy number in further analyses . We observed that females on average have slightly but significantly higher mtDNA copy numbers than males ( 6 . 7 more copies ) , and the average mtDNA copy number decreases with age , consistent with general decline of cellular energy metabolism during aging and with observations of other researchers in independent cohorts [21] . We note that lymphocytes , the cells studied here , have characteristically low levels of cytoplasm , and the average number of mtDNA molecules per cell is relatively low ( average of 111 . 5 ) compared to what may be expected in actively growing cells . In fact , further applications of our method to data from the 1000 Genomes Project , which sequenced DNA from transformed lymphocytes , observed an average number of mtDNA copies that is ~6-fold greater ( work in progress ) . This is consistent with the increased mtDNA copy number associated with augmented risk of several types of cancer[9 , 10] . The correlations of mtDNA copy number with waist circumference and waist-hip ratio , but not with BMI ( nor in further assessments , with height or lipid levels; Table 2 ) , are intriguing . The findings hint that mtDNA copy number may also be relatively associated with central obesity and body fat distribution , though further study is needed to find the basis for the correlation . The relatively high heritability of copy number ( 54% ) is also suggestive . Future analyses , including genome-wide association studies , should identify genetic factors underlying this substantial heritability . Although our variant calling algorithm was designed to identify mtDNA variants from sequencing data , its potential utility can easily extend to other instances of comparable sequence heterogeneity . For example , such allelic heterogeneity is a major characteristic of cancer , where mutations are expected to occur in only a subset of the sequenced cells because of tumor heterogeneity . Our method could be implemented to detect somatic mutations in cancer cells with allele fractions ranging between 0 and 1 ( a similar framework has recently been proposed[22] ) . As another example , although the extent of RNA editing is unclear[23–27] , the method has the potential to study RNA editing from sequencing data and infer the editing level—another parameter that ranges between 0 and 1 . The method could also be adapted to the analysis of cell-free fetal DNA in maternal plasma ( i . e . a non-invasive way to sequence human fetus[28] ) , where an important step is to estimate the proportion that is fetal in origin in DNA isolated from maternal plasma during pregnancy . Overall , our study with ~2 , 000 individuals is the largest population-scale study of mtDNA variation thus far , and the application of the two mtDNA analysis tools to the SardiNIA study indicates that when whole-genome sequencing data are available , a set of analyses on mtDNA variants and copy numbers can be performed with no additional experimental cost . Many population studies collecting large-scale sequencing data could thus extend the analyses of genetic factors affecting mtDNA levels , their inheritance , and their relation to aging and disease . | We present a new program that provides a general solution for the analysis of variation of mtDNA ( the small circular genome in mitochondria , separate from the DNA in the nucleus ) . This is needed because many large-scale genetic studies are using new DNA sequencing technologies to help assess genetic variation and its effects on disease , but the mitochondrial genome is often ignored because it exists in many copies in a cell , complicating analyses . Our approach both identifies variants on mitochondrial genome and estimates mtDNA copy number . Applying the programs to DNA sequence from ~2 , 000 SardiNIA project participants , we show that heteroplasmies ( mtDNA variants with more than one allele at a DNA site ) increase with age , and that copy number is relatively highly heritable and is correlated with metabolic traits , particularly central fat levels . The program package can facilitate comprehensive mtDNA analysis from any whole-genome sequencing data , with an increase in the understanding of mtDNA dynamics and its potential role in aging and metabolism . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2015 | Assessing Mitochondrial DNA Variation and Copy Number in Lymphocytes of ~2,000 Sardinians Using Tailored Sequencing Analysis Tools |
Anti-dengue T-cell responses have been implicated in both protection and immunopathology . However , most of the T-cell studies for dengue include few epitopes , with limited knowledge of their inter-serotype variation and the breadth of their human leukocyte antigen ( HLA ) affinity . In order to expand our knowledge of HLA-restricted dengue epitopes , we screened T-cell responses against 477 overlapping peptides derived from structural and non-structural proteins of the dengue virus serotype 3 ( DENV3 ) by use of HLA class I and II transgenic mice ( TgM ) : A2 , A24 , B7 , DR2 , DR3 and DR4 . TgM were inoculated with peptides pools and the T-cell immunogenic peptides were identified by ELISPOT . Nine HLA class I and 97 HLA class II novel DENV3 epitopes were identified based on immunogenicity in TgM and their HLA affinity was further confirmed by binding assays analysis . A subset of these epitopes activated memory T-cells from DENV3 immune volunteers and was also capable of priming naïve T-cells , ex vivo , from dengue IgG negative individuals . Analysis of inter- and intra-serotype variation of such an epitope ( A02-restricted ) allowed us to identify altered peptide ligands not only in DENV3 but also in other DENV serotypes . These studies also characterized the HLA promiscuity of 23 HLA class II epitopes bearing highly conserved sequences , six of which could bind to more than 10 different HLA molecules representing a large percentage of the global population . These epitope data are invaluable to investigate the role of T-cells in dengue immunity/pathogenesis and vaccine design .
Dengue is a member of the genus Flavivirus with a positive sense , single stranded RNA genome of ∼10 kb . The genome encodes for a polyprotein that is co- and post-translationally cleaved into 10 proteins: three structural ( capsid , precursor membrane and envelope ) , which constitute the virus particle; and seven non-structural proteins ( NS1 , 2a , 2b , 3 , 4a , 4b and 5 ) , which are proteases that cleave the viral polyprotein and contribute to the formation of the replication complex [1] , [2] , [3] . The virus exists in nature as a complex population of four dengue serotypes ( DENV1 , 2 , 3 and 4 ) , consisting of up to 86% homology of amino acid sequences between serotypes [4] . DENV infection , transmitted primarily by the Aedes aegypti mosquito , is a major global health problem in tropical and subtropical areas [5] . The typical spectrum of the dengue disease ranges from asymptomatic to a mild form of the disease , dengue fever ( DF ) . However , a small fraction of the patients develops a severe form of the disease , characterized by increased vascular permeability ( dengue hemorrhagic fever - DHF ) that can lead to hypovolemic shock ( dengue shock syndrome , DSS ) and even death . The leading theories underlying DHF immunopathology are based on the observation that sequential infection with different dengue serotypes leads to greater risk of developing a more severe form of the disease . The earliest postulated mechanistic theory proposes that cross-reactive and non-neutralizing antibodies would form immune complexes with the viruses , that can mediate enhanced infection of Fcγ receptor-expressing cells [6] , [7] . Several dengue vaccine candidates have been shown to induce memory T-cell response that can confer protection against dengue infection [8] , [9] . The importance of protective cytotoxic T lymphocyte ( CTL ) responses in primary dengue infection has been demonstrated in IFNα/βR knock out mice model [10] . Despite the lack of IFN type I responses in this animal model , immunization with dengue CTL and T-helper ( Th ) cell epitopes has been shown to contribute towards faster clearance of the virus [10] , [11] . However , a number of studies have suggested a possible involvement of cross-reactive HLA class I T-cells epitopes in dengue pathogenesis . Memory T-cell clones generated during a primary infection in response to epitopes from one dengue serotype , would cross-react with epitope variants presented during a subsequent infection with a different dengue serotype , to elicit abnormal responses ( cytokine storm ) associated with capillary leakage [12] . Despite the increased acknowledgement that T-cells play a role in both the pathology of and protection from dengue infection , a more comprehensive analysis of T-cell activation during dengue infection is hampered by the small repertoire of known dengue T-cell epitopes in humans . Most of the known epitopes are associated with DENV2 , and are restricted to a small number of human leucocyte antigens ( HLAs ) [9] , [13] , [14] , [15] , [16] , [17] , [18] , [19] . Reported T-cell epitopes from DENV3 , however , are limited and mostly in NS3 protein [15] , [16] , [18] , [20] , [21] , [22] , [23] , [24] , [25] , [26] , [27] , [28] , [29] , [30] . The search for DENV3 T-cell epitopes has been motivated by the association of DENV3 with major outbreaks in the Americas and Southeast Asia , infecting adults and children and causing a wide spectrum of disease severity [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] . Thus , the search for DENV3 T-cell epitopes is necessary . We previously showed that T-cell responses elicited either by attenuated yellow fever vaccine ( 17DD ) or peptide immunization were similar in terms of epitope repertoire and immune dominance [39] . We also demonstrated a correlation between the strength of binding to HLA class I and epitope immunogenicity [39] , [40] . Based on these studies , we devised an optimized strategy for identification and characterization of DENV3 T-cell epitopes by use of overlapping peptide libraries which were constructed based on protein sequences of DENV3 isolates of our human cohort [37] followed by in silico and biochemical characterization of the immunogenic peptides . This strategy was applied to HLA transgenic mice , an effective animal model to use for identifying potential epitopes recognized by human T-cells . The repertoire of epitopes identified in these animal models has been correlated with those identified in humans [41] , [42] , [43] , and thus , the strategy was proven to be an effective platform for epitope discovery [44] . Recently , IFN α/βR KO mice were backcrossed with different HLA transgenic mice [17] and used to discover HLA-A02-restricted T-cell epitopes upon infection with a mice-adapted DENV2 S221 strain . In the current study , however , we used a mouse model with functional IFN α/βR and identified T-cell epitopes of DENV3 Envelope , NS1 , NS3 and NS5 proteins by use of transgenic mice expressing HLA class I ( A2 , A24 and B7 ) and II ( DR2 , DR3 and DR4 ) molecules . The affinity of the epitopes to their specific HLAs was confirmed by binding assays . A subset of the epitopes identified the presence of activated memory T-cells from subjects naturally infected with DENV3 , and also effectively primed naïve T-cell clones from dengue IgG negative individuals ( dengue naïve ) . Additionally , analysis of intra- and inter-serotype variants was carried out for one HLA class I epitope for identification of possible altered epitope ligands in other DENV serotypes . This could facilitate the identification of epitope variants that could cause aberrant memory T-cell activation . An effective dengue vaccine needs to provide broad population coverage and induce immune response against majority variants of the four dengue serotypes . Therefore it would be desirable for a dengue vaccine to target the T-cell responses against conserved amino acid sequences that can be presented by multiple HLAs . In our previously reported preliminary study , we showed that epitopes containing highly conserved dengue sequences ( pan-dengue sequences; defined as being present in each of the four dengue serotypes with an incidence of 80% or more [45] ) are immunogenic in HLA TgM . Herein , we showed that such sequences can also be quite promiscuous , binding as many as 14 different HLA molecules . The results suggest that T-cell epitopes containing highly conserved in each of the four dengue serotypes could be immunogenic in a large percentage of the global population and , thus , potentially useful for further exploration as a vaccine target against DENV .
This study was performed in strict accordance to the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . The TgM protocol was approved by The Johns Hopkins University ( JHU ) Institutional Animal Care & Use Committee ( MOO7M78 ) . Adult subjects with history of dengue infection analyzed in this study were recruited from a cohort of suspected dengue cases established in Recife , Pernambuco , Brazil and described by Cordeiro et al . [37] . All patients provided their written informed consent to participate and this study was reviewed and approved by the ethics committee of the Brazilian Ministry of Health ( CONEP: 4909; Process n° 25000 . 119007/2002-03; CEP: 68/02 ) . In addition , the Johns Hopkins University Institutional Review Board reviewed and approved the study ( protocol JHM-IRB-3: 03-08-27-01 ) . Synthetic peptides covering the full-length Envelope ( Env; 95 peptides ) , NS1 ( 75 ) , NS3 ( 150 ) , and NS5 ( 156 ) protein sequences of DENV3 Philippines/H87/1956 isolate ( UniProtKB/Swiss-Prot Accession: P27915; GenPept: AAA99437 ) were obtained for the study ( Supplementary Table S1 ) . The Env , NS1 and NS3 peptides were 15-mers , overlapping by 10–11 amino acids ( aa ) , and synthesized by Schafer-N ( Denmark ) , while the NS5 peptides were 13- to 17-mers , overlapping by 11 to 13 aa , and obtained from BEI Resources , NIH ( Supplementary Table S1 ) . All the 9-mer peptides used for HLA class I minimal epitope discovery and analysis of the variants of DENV NS3399–407 were purchased from Genscript Corporation . All peptides were dissolved in 10% ( v/v ) dimethyl sulfoxide ( DMSO; Sigma ) at 2 mg/mL , aliquoted and stored at −20°C until use . Murine H-2 class II-deficient , HLA-A02 ( A*0201 ) [46] , HLA-B07 ( B*0702 ) [47] , HLA-A24 ( A*2402 ) ( Lemonnier et al . unpublished ) , HLA-DR2 ( DRA1*0101/I-Eα; DRB1*1501/I-Eβ ) [48] , HLA-DR3 ( DRA1*0101; -DRB1*0301 ) [49] , [50] , and HLA-DR4 ( HLA-DRA1*0101; -DRB1*0401 ) [51] , animals were bred and maintained in the Johns Hopkins University School of Medicine Animal Facility . Specific pathogen-free colonies were maintained in a helicobacter-negative mice facility . HLA expression of the experimental transgenic mice was evaluated by flow cytometry ( data not shown ) . Ten to 15 mice were injected twice , two weeks apart , with a pool of peptides ( 1 µg/peptide ) emulsed in Titermax gold ( Titermax ) or only Titermax gold ( negative control ) according to the manufacturer directions . The immunization was performed through subcutaneous route . Two weeks after the last immunization , the mice were sacrificed and their spleens were aseptically removed and placed into a disposable sterile petri dish containing 5 mL of RPMI media , followed by manual dissociation and grinding ( pressing tissue against a sterile 70 mm cell strainer ) . After washing the strainer with 15 mL of RPMI media , the splenocytes were transferred into a 50 mL conical tube and centrifuged 500×g for 5 minutes at room temperature ( RT ) . Red blood cells with the splenocytes were lysed by treating the pellets with ammonium-chloride-potassium ( ACK ) lysing buffer ( 3 mL per spleen ) for 3 minutes at RT . The splenocytes were then washed by addition of 20 mL of RPMI media , centrifuged at 500×g for 5 minutes at 4°C , resuspended , strained and counted using a Vi-Cell analyzer ( Beckman Dickson ) . The isolated splenocytes were depleted of murine CD4 ( HLA class I TgM ) or CD8 ( HLA class II TgM ) positive cells by the use of antibody-coated microbeads and LD columns manufactured by Miltenyi Biotec following the instructions of the manufacturer . The cells from immunized and control mice were screened by IFN-γ ELISPOT assays for memory T-cell response against either the pooled or individual peptides used in the immunization peptide pool . The strategy for epitope mapping , as depicted on Figure 1 , included a first round of screening of overlapping peptides arrayed in different matrix pools ( Envelope , 10×10; NS1 , 9×9; NS3 and NS5 , 13×12 ) , whereby each peptide is present in two different pools [52] . This was followed by a deconvolution of the positive pools . The peptide concentration in both of these steps was 10 µg/mL . Lastly , the functional avidity of the immunogenic peptides was assessed by measuring T-cell response against the peptides at different concentrations ( 10 µg/mL , 1 µg/mL and 0 . 1 µg/mL ) as previously reported [53] . Low resolution HLA typing for HLA-A , HLA-B , HLA-Cw , HLA-DR , and HLA-DQ loci was performed using Sequence Specific Primers amplification methods as described elsewhere [54] . The primary virus strains DENV-1 ( PE/97-42735 ) , DENV-2 ( PE/95-3808 ) , DENV-3 ( PE/02-95016 ) and DENV-4 ( IEC ) isolated in Brazil were expanded on African green monkey kidney cells and used in a standard plaque reduction neutralization test with the same cell line and heat-inactivated patient sera as described [55] . Serum samples were used at two-fold dilutions ranging from 1/20 to 1/2560 . The 50% end-point dilution of each serum , corresponding to the dilution at which 50% of the wells were completely protected from infection , was determined according to standard methods . Fifty percent plaque neutralization titers ( PRNT50 ) were calculated as the highest dilution of Ab reducing virus plaques by 50% . Peripheral blood samples were obtained from 5 subjects enrolled in this cohort who had history of dengue infection with different virus serotypes , including DENV3 , for at least a year , prior to the blood collection . Demographic , dengue serotype-specific immunity status and HLA typing information are shown in supplementary Tables S1 and S2 . Blood from the subjects was collected in heparinized tubes ( BD Biosciences ) 1 to 3 years year after onset of symptoms . PBMCs were isolated by gradient density using Ficoll-Paque Plus according to the instructions of the manufacturer ( GE Healthcare ) . The cells were then washed once with PBS ( Phosphate Buffer Saline ) pH 7 . 2 and the red cells were lysed with ACK Lysing buffer ( Biosource International , Inc ) for 3 minutes at room temperature ( RT ) . The cells were then centrifuged at 530×g for 5 minutes at 4°C and re-suspended , strained and counted by use of cell analyzer Vi-Cell XR ( Beckman Coulter ) . PBMC from HLA-A02 or HLA-B07 positive volunteers were used to assess the immunogenicity of the HLA-class I epitopes while HLA-DR2 positive donors were used to assess the immunogenicity of HLA-class II epitopes . The CD4+ T-cells and CD8+ T-cells were depleted from the HLA-A02/B07 positive donors and the HLA-DR2+ donors respectively . Cell depletion was carried out by magnetic bead cell isolation using Miltenyi CD4+ and CD8+ microbeads and LD columns ( Miltenyi Biotec ) according to the manufacturer manual . Following the cell depletion step , the remaining cells were suspended at 2 . 5×106 cell/mL and cultured at 37°C , 5% CO2 in RPMI media containing 8% ( v/v ) T-cell growth factor ( TCGF ) , 5% ( v/v ) autologous plasma . T-cells were activated with individual peptides at 10 µg/mL . On day 4 , the cell cultures were replenished with TCGF and autologous plasma . On day 8 , the cells were harvested and washed with serum-free media ( Gibco ) to prepare for ELISPOT analysis as described below . ELISPOT assays for IFN-γ detection were performed using the IFN-γ ELISPOT set from BD Biosciences , according to the manufacturer's instructions . Briefly , the ELISPOT plate ( 96-well ) was coated with anti-mouse IFN-γ at 5 µg/mL and incubated at 4°C overnight . The plate was blocked with RPMI-1640 containing 10% heat-inactivated fetal bovine serum ( HyClone ) , 1% L-Glutamine ( Gibco ) , 1% penicillin/streptomycin ( Gibco ) for 2 h at RT . Peptide pools or individual peptides were then plated in duplicate . The cells were then added at a range of 5 to 10×105 cells/well . Concanavalin A ( Sigma ) , at 2 . 5 µg/mL , was used as a positive control while media was used as a negative control ( background ) . After an 18 h incubation at 37°C and 5% CO2 , the plate was washed and incubated with biotinylated anti-mouse IFN-γ at 2 µg/mL for 2 h at RT . Streptavidin-HRP 100-fold diluted was then added and incubated for 1 h at RT . The plate was washed and reactions were developed with 3-amino-9-ethylcarbazole ( AEC ) substrate ( BD Biosciences Pharmingen ) . Spot development was stopped after 30 min incubation by washing the plate with distilled water . The plate was dried at RT and the spots were counted with the Immunospot Series 3B Analyser ELISPOT reader ( Cellular Technologies Ltd ) using the program Immunospot software version 3 . 0 ( Cellular Technologies Ltd ) . The spot average was normalized and expressed as the number of spot-forming cells ( SFC ) per 1 million cells . ELISPOT assays were also used to assess human T-cell responses using a protocol similar to the above using a human IFN-γ ELISPOT kit ( BD Biosciences ) ; phorbol 12-myristate 13-acetate ( PMA; Sigma ) at 250 ng/mL and ionomycin ( Sigma ) at 250 ng/mL were used as a positive assay control . Peptide pools and individual peptides were considered positive in the ELISPOT assay when all three criteria described bellow were met: These criteria were used consistently throughout this study . Monocytes were isolated from peripheral blood of healthy donors ( Central Blood Bank of Pittsburgh , USA ) and cultured for 5–7 days in 24-well plates at 5×105 cells per well in the presence of GM-CSF and IL-4 ( both 1000 IU/ml ) . The immature DCs were then exposed to the following combination of activation factors for 48 h: rhIL-1β ( 25 ng/ml ) , rhTNFα ( 50 ng/ml ) , rhIFNα ( 1000 IU ) , Poly IC ( 20 µg/ml ) and IFNγ ( 1000 IU/ml ) . As previously described [56] , high IL-12 p70 producing type-1 polarized DC ( αDC1 ) were induced CTL were generated using a protocol similar to what we previously described [56] . Briefly , CD8+ T-cells were isolated from the peripheral blood of dengue virus naïve ( dengue IgG negative ) HLA-A2+ donors by negative selection using the EasySep system ( Stem Cell Technologies ) . The T-cells were plated at 7 . 5×105cells/well in 48 well plates ( Falcon , BD Labware ) and sensitized with 9-mer peptide-pulsed autologous αDC1 ( 7 . 5e4 cells/well ) . To mimic their interaction with CD40L-expressing CD4+ T-helper cells , as a surrogate we added to the cultures γ-irradiated ( 3000 Rad ) J558-CD40L cells ( 5e4 cells/well ) ( provided as a gift from Dr . P Lane , University of Birmingham , Birmingham , U . K . ) . From day 4 onwards , rhIL-2 ( 50 units/ml ) , rhIL-7 ( 10 ng/ml ) , and rhIL-15 ( 100 IU/ml ) were added to the cultures every 4 days upon media replacement . The cultures were provided one round of in vitro sensitization at day 14 by adding relevant peptide-pulsed γ−irradiated ( 3000 Rad ) A2+ T2 cells ( provided by Dr . W Storkus , University of Pittsburgh ) at target cells to T-cell responder ratio of 1∶5 . On day 20 , the T-cells were screened for the presence of antigen specific CTL by IFN-γ ELISPOT assay described previously . Long term CTL lines from ELISPOT positive cultures were established and maintained under the described culture conditions with IL-2 concentration being increased to 500 IU/ml . The cultures were re-sensitized with peptide pulsed γ−irradiated T2 cells every 10 days and used periodically during the study . Prediction of HLA class I T-cell epitopes for HLA-A*0201 , HLA-A*2402 and HLA-B*0702 molecules was performed by use of the ANN ( artificial neural network ) algorithm based prediction tool at the Immune Epitope Database ( IEDB ) and Analysis Resource ( http://tools . immuneepitope . org/analyze/html/mhc_binding . html ) . The peptide sequences ( 13-mer to 17-mer ) were input to the prediction server and prediction was carried out for all possible lengths ( 8-mer to 11-mer ) . Peptide sequences predicted with IC50 below 500 nM were considered for further analysis . For determination of HLA class II core region interaction within the peptides bearing pan-dengue conserved sequences , stabilization matrix method ( SMM ) , artificial neural network ( NN ) and Sturniolo alignment algorithms [57] , [58] , [59] were used . The peptide sequences were input to the IEDB server for identification of HLA class II epitopes under the IEDB recommended prediction method . Then , the 9-mer sequence of align core ( core region interaction ) for the lowest predicted IC50 was annotated for each prediction method . Purification of HLA class I ( A*2402 and B*0702 ) and HLA class II molecules ( DRB1*0101 , DRB1*0301 , DRB1*0401 , DRB1*0404 , DRB1*0405 , DRB1*0701 , DRB1*0802 , DRB1*0901 , DRB1*1101 , DRB1*1302 , DRB1*1501 , DRB3*0101 , DRB4*0101 and DRB5*0101 ) by affinity chromatography and the quantification of peptide binding based on competitive inhibition assay against binding of a high affinity radiolabeled standard peptide were performed as detailed elsewhere [60] , [61] , [62] , [63] , [64] . Briefly , EBV transformed homozygous cell lines were used as sources of HLA molecules . A high affinity radiolabeled peptide ( 0 . 1–1 nM ) was co-incubated at room temperature or 37C with purified HLA in the presence of a cocktail of protease inhibitors . Following a two-day incubation , HLA bound radioactivity was determined by capturing HLA/peptide complexes on Ab coated Lumitrac 600 plates ( Greiner Bio-one , Frickenhausen , Germany ) , and measuring bound cpm using the TopCount ( Packard Instrument Co . , Meriden , CT ) microscintillation counter . The concentration of peptide yielding 50% inhibition of the binding of the radiolabeled peptide was calculated . Under the conditions utilized , where [label]<[HLA] and IC50≥[HLA] , the measured IC50 values are reasonable approximations of the true Kd values . Each competitor peptide was tested at six different concentrations covering a 100 , 000-fold range , and in three or more independent experiments . As a positive control , the unlabeled version of the radiolabeled probe was also tested in each experiment . Binding assay for HLA-A*0201 was performed using HLA-A2:Ig fusion protein . The labeled peptide ( ALMDKVLKV ) was commercially synthesized by GenScript Corporation as a reference peptide for the binding assay . A fluorescein moiety was introduced into position P8 of the designed peptide ( ALMDKVLKV ) . This chemical modification has been previously shown to not affect the peptide's affinity for the HLA*A2 binding pocket [65] . The lyophilized labeled peptide was diluted in 100% DMF ( Dimethylformamide ) to a final concentration of 10 mM . Subsequent dilutions were carried out using 0 . 5 mg/ml bovine-γ-globulin ( BGG ) in 1×PBS . Soluble HLA-A2:Ig dimer complexes were obtained from BD Biosciences . This fusion protein consists of two extracellular portions of the major histocompatibility complex ( MHC ) class I HLA-A2 domains fused to the variable regions of a mouse IgG1 antibody . Additional β2-microglobulin ( Fitzgerald Industries ) was added to the binding assays to retain the fusion protein in a functional state . Fluorescence polarization was measured by a SpectraMax M5 reader ( Molecular Devices , Sunnyvale CA ) using an excitation wavelength of 485 nM and an emission wavelength of 530 nM . The competitive binding assays were carried out at room temperature for an incubation period of 3 days . Both labeled and unlabeled peptides were added to the reaction mixture before the HLA-A2:Ig fusion protein , ensuring that both peptides were presented simultaneously to the HLA molecule . To avoid peptide photobleaching , the reaction mixture was incubated within aluminum foil covered microcentrifuge tubes . After the 3 days incubation period , 20 µl of the reaction mixture was loaded into individual wells of a black , non-binding surface ( NBS ) 384 well plate ( Corning ) . Controls included only buffer ( blank ) , protein and pFITC . Each experiment was performed in quadruplicate and reported as the mean with standard deviation . The reported IC50 values were obtained by fitting the mean of 13 data points to a four-parameter logistic equation:where , max indicates the upper plateau of the curve; min , the lower plateau and B , is the slope factor , which describes the steepness of the curve transition . NS3 protein sequences of each DENV serotype ( 696 for DENV1 , 681 DENV2 , 585 DENV3 , and 71 DENV4; as of Feb . 2009 ) were collected from the NCBI Entrez Protein Database and aligned , following the method in Khan et al . , ( 2008 ) [45] . The region of each serotype alignment corresponding to the DENV3 NS3399–407 epitope sequence ( KLNDWDFVV ) was extracted and analysed for all the peptides that were variant/different to the epitope sequence by at least one amino acid difference . The incidence ( % occurrence ) of the individual variants in each DENV serotype NS3 alignment was determined . Inter-serotype variant peptides ( DENV1 , 2 and 4 ) with an incidence of 5% or more and predicted to be a potential epitope of HLA-A*0201 were synthesized and their binding affinity and immunogenicity assayed .
The immunogenicity of each of the 477 peptides from DENV3 envelope ( Env ) , NS1 , NS3 and NS5 proteins ( Supplementary Table S3 ) were tested in six HLA transgenic mice ( TgM ) strains ( A02 , A24 , B07 , DR2 , DR3 and DR4 ) using IFN-γ ELISPOT as the initial screening readout ( Figure 1 ) . A total of 13 novel peptides were shown to be immunogenic in HLA class I transgenic mice ( TgM ) [A02 ( n = 5 ) , A24 ( n = 3 ) and B07 ( n = 5 ) ] , six of which were located in Env , three in NS3 and four in NS5 ( Table 1 ) . No HLA class I epitope was found in NS1 protein . A02 and B07 epitopes were presented in Env , NS3 and NS5 , whereas A24 epitopes were present only in Env . In contrast to HLA class I , a total of 173 peptides were shown to be immunogenic in HLA class II TgM [DR2 ( n = 64 ) , DR3 ( n = 50 ) and DR4 ( n = 59 ) ( Supplentary material S3 ) ] , 36 of which were located in Env , 31 in NS1 , 48 in NS3 , and 58 in NS5 ( Supplementary Table S4 ) . Affinity of the immunogenic peptides to the respective HLA of the TgM was confirmed by use of HLA-binding assay . HLA class I binding groove fits peptides between 8 to 12 amino acid ( aa ) long ( typically 9 aa ) . Hence , in silico analysis was first performed to identify 9-mer ( s ) within the immunogenic peptides ( 15-mers to 17-mers ) that were potential binders of the respective HLA of the TgM . Nine out of the 13 immunogenic peptides encompassed binding motifs for the HLA they were immunogenic in the animal studies and , thus , they were selected for peptide synthesis and binding assay analysis , which further confirmed the affinity of all tested peptides ( Table 1 ) . The open topology of the HLA class II groove , fitting longer peptides than class I , allowed direct use of the immunogenic peptides ( 15-mer to 17-mer ) for the binding assay , without the preliminary in silico binding prediction . Binding assay was performed against the HLA of three class II TgM [DR2 ( DRB1*1501 ) , DR3 ( DRB1*0301 ) and DR4 ( DRB1*0401 ) ] , as well as eleven additional HLA class II molecules ( see methods ) for the analysis of epitope promiscuity . HLA binding affinity was confirmed for 39 DR2- , 26 DR3- and 47 DR4-specific epitopes , representing 65% , 53% and 84% of the total positive peptides of DR2 , DR3 and DR4 TgM respectively ( Supplementary Table S5 ) . Furthermore , many of the peptides tested showed affinity to multiple HLA molecules ( supplementary material S1 ) . The number of peptides that elicited T-cell responses in two HLA class II TgMs ranged from 5 to 6 peptides , whereas only 2 peptides were immunogenic in all three HLAs tested in TgM ( Figure 2A ) . The majority of the peptides were positive in only one TgM strain analyzed . However , the binding assay analysis suggests that these peptides are more promiscuous . The majority of the class II peptides had binding affinity to more than one HLA . The number of peptides that had affinity to DR2/DR3 , DR2/DR4 , DR3/DR4 and DR2/DR3/DR4 were 9 , 18 , 6 and 17 , respectively ( Figure 2B ) . TgM and binding affinity assays are useful tools for determination of HLA affinity of an epitope . However , experimental validation of the T-cell epitopes in humans is necessary for accurate interpretation of results . Hence , a subset of the T-cell epitopes , identified herein by use of TgM with HLA affinity confirmed by binding assays , was selected for immunogenicity study in humans by use of PBMC collected from subjects immune to DENV3 ( Supplementary Tables S1 and S2 ) . In addition , DR2 epitopes were selected based on their functional avidity , defined as the ability to activate CD4 T-cells at concentrations below 0 . 1 µg/mL and high affinity to HLA-DR2 molecule ( IC50 below 20 nM ) . This additional criterion was applied in order to reduce the number of DR2 peptides to test and also to select only the peptides with increased likelihood to induce T-cell responses . PBMCs from A02 ( n = 2 ) , B07 ( n = 2 ) and DR2 ( n = 3 ) positive subjects were either CD4 depleted ( A02 and B07 ) or CD8 depleted ( DR2 ) and cultured for a week with a peptide pool containing the HLA-matched epitopes , followed by analysis of T-cell activation by use of ELISPOT to detect IFN-γ secretion . Among the A02 positive subjects tested , consistent response against peptide NS3399–407 was observed in all subjects , whereas only one of the two responded against the peptide NS5318–326 ( Figure 3B ) . No T-cell response was detected against the peptides Env106–114 and NS5325–333 ( Figure 3A & 3B ) . All subjects analyzed for B07 epitopes elicited T-cell response against the peptide NS5389–398 ( Figure 3C & 3D ) . However , T-cell response against Env226–234 and NS3593–601 was observed only in one of the two subjects analyzed ( Figure 3C ) . The peptides Env126–140 and NS185–99 reproducibly activated memory T-cells on all the subjects analyzed for DR2 epitopes ( Figure 4 ) . On the other hand , the peptides Env231–245 , NS169–83 and NS3357–371 elicited T-cell response in one of the volunteers while the remaining peptides did not elicit any memory T-cell response ( Figure 4 ) . Therefore , majority of the HLA class I and II T-cell epitopes analyzed activated memory T-cell response in at least one individual that had experienced DENV3 infection in the past . This suggests that the epitopes are naturally processed and presented to T-cells during DENV3 infection . In addition to showing that the epitopes identified herein are recognized by human memory T-cells , we tested if a subset of the epitopes identified herein could prime naïve T-cells for relevance in epitope-based vaccine design . CD14 positive monocytes were isolated from PBMC harvested from blood collected from healthy donors that were known to be IgG negative for dengue . Monocytes isolated from either HLA-A02 or HLA-B07 positive individuals were differentiated into mature dendritic cells and pulsed with peptide pool containing either A*0201 or B*0702 epitopes , respectively , and co-cultured with autologous lymphocytes responders . After 20 days , the in vitro-sensitized T-cells were harvested and reactivity was assessed by IFN-γ ELISPOT using target cells ( peptide-pulsed autologous monocytes ) expressing either HLA-A*0201 and HLA-B*0702 molecules pulsed with individual peptides . Among the epitopes analyzed , the peptide DENV3 NS5389–398 ( B*0702-restricted ) and NS3399–407 ( HLA-A*0201-restricted ) reproducibly primed naïve T-cells ( Figures 5 and 6 respectively ) . Thus , a selected repertoire of epitopes identified using HLA TgM did not only trigger memory T-cell response in individuals with history of dengue infection , but also primed T-cell from dengue naïve subjects . Altered peptide ligand is thought to be a mechanism driving memory T-cells toward an aberrant cytokine response associated with disease severity . Intra- and inter-serotype variant analysis of the A02-restricted epitope ( DENV3 NS3399–407 - KLNDWDFVV ) were performed to assess the potential of the peptides to act as altered peptide ligands . The DENV3 NS3399–407 epitope had an incidence of 63% among all analyzed DENV3 sequences collected from the NCBI Entrez Protein Database ( see Methods section ) . The remaining ∼37% of the sequences constituted four intra-serotype variants of the epitope ( KLNDWDFVV ) with only one peptide of incidence more than 1% ( RLNDWDFVV; ∼36% ) . In contrast , there were 11 inter-serotype variants , but only four ( DENV1: KNNDWDYVV; DENV2: RTNDWDFVV and RANDWDFVV; DENV4: KLTDWDFVV ) had an incidence of about 5% or more among all analyzed DENV sequences of each serotype ( Table 2 ) . Subsequently , HLA binding prediction analysis was performed on the inter-serotype variants to assess the effect of one or two amino acid substitutions to HLA binding . Variants predicted to retain the HLA affinity were synthesized and tested in binding assay experiments . Prediction analysis indicated that only the DENV2 and DENV4 variants retained the affinity to HLA-A*0201 , which was experimentally confirmed ( Table 2 ) . Further , the DENV2 and DENV4 variants were tested for their ability to prime naïve T-cells was assessed as described in the Methods . DENV2 variant failed to prime naïve T-cells , but DENV4 variant could reproducibly trigger naïve T-cell response in vitro ( Figure 6 ) and , thus , it could potentially be an altered peptide ligand to the DENV3 epitope KLNDWDFVV . A vaccine should provide broad HLA coverage for relevance at the population level . Promiscuous epitopes recognize multiple HLA molecules and are , thus , candidate epitopes for vaccine design consideration [66] . Additionally , a vaccine should also target conserved epitopes for broad coverage of viral variants . Therefore , theoretically an effective dengue vaccine would include promiscuous HLA epitopes that are highly conserved in each of the dengue serotypes . Previously , we reported several immunogenic , pan-dengue conserved sequences by use of HLA class II TgM [45] . We provide in Table 3 the list of HLA class II TgM epitopes that had HLA affinity confirmed by binding assay and contain at least nine consecutive amino acids that are pan-dengue conserved sequences . We then compared the ability of these epitopes to induce T-cell responses in TgM and to bind to multiple HLAs ( among the 14 HLA molecules analyzed ) . The results show that peptides containing highly conserved regions are equally immunogenic ( Figure 7A ) and can bind to as many HLAs ( Figure 7B ) as compared to peptides with non-conserved sequences . Additionally , we investigated if these immunogenic conserved dengue sequences were present in the 9-mer core regions encompassing the major pockets in the groove of HLA-DR2 , -DR3 and -DR4 molecules . In silico analysis was carried out using SMM , NN and Sturniolo prediction algorithms and the core regions were annotated and depicted in Supplementary Table S6 . For most of the peptides analyzed , the three prediction methods consistently showed that pan-dengue conserved sequences directly interact ( either partially or entirely ) with anchor motifs in the HLA molecules , suggesting that these sequences indeed are important for HLA specificity and affinity and , thus , are suitable for epitope-based vaccine development .
CD4 and CD8 T-cells have been shown to mediate protection against lethal dengue virus challenges in a mice model [10] , [11] . However , limited dengue-specific T-cell epitopes have been reported , specifically for DENV3 , a serotype responsible for several outbreaks worldwide [31] , [32] , [33] , [34] , [35] , [36] , [37] , [38] . Herein , we have performed an extensive T-cell epitope mapping and identified more than 90 novel potential T-cell epitopes for Env , NS1 , NS3 and NS5 proteins of DENV3 , by use of TgM expressing a set of HLA class I and II molecules highly frequent in the Caucasian population . Herein we identified 185 immunogenic peptides from the entire peptide library analyzed , 13 of which were HLA class I-restricted epitopes , whereas 172 were HLA class II-restricted epitopes . The number of epitopes found on HLA class II TgM outnumbered the HLA class I counterpart , which is consistent with other studies using the same TgM animal model and similar immunization strategies [53] , [67] . The combination of immunization strategy ( with peptide pool ) , the difference in topology between HLA class I and II molecules and the HLA class I epitope processing mechanism [68] might explain the reasons for the paucity of HLA class I epitopes identified in our study and , thus , revealing to be a caveat for identification of HLA class I restricted epitopes . Binding assay analysis was done in order to confirm the HLA affinity suggested by TgM immunogenicity data . HLA affinity correspondence between the two data was observed for 9 HLA class I and 97 HLA class II epitopes . A schema of DENV3 proteome with the location of each confirmed epitope as well as those reported in the literature are illustrated in Figure 8 . A number of these epitopes were shown to be quite promiscuous , binding to as many as 14 HLA class II molecules . Besides HLA promiscuity improving vaccine population coverage , multiple HLA binding motifs within a peptide antigen have been shown to enhance T-cell immunogenicity in humans vaccinated with attenuated yellow fever 17DD vaccine [69] . A comparison between DR2 , DR3 , and DR4 TgM immunogenicity and binding affinity data reported herein showed that not all the peptides with binding affinity to the HLA ( s ) induced an immune response in the transgenic mice . This is likely related to how T-cells are activated . In order to trigger T-cell responses there are two essential steps: ( 1 ) the epitope must bind to the HLA molecule; and ( 2 ) the T-cell receptor ( TCR ) on T-cells must recognize the HLA-epitope complex . Binding affinity only takes into consideration the step 1 , while immunogenicity is more complex , taking into account both steps . Anything that influences epitope presentation ( e . g . epitope dominance ) will affect the T-cell activation and , consequently , will compromise the step 2 . Thus , discrepancies observed between binding affinity and immunogenicity were most likely due to the fact that some peptides failed to activate T-cells possibly due to epitope competition and/or dominance during animal immunization . We selected a subset of epitopes that HLA affinity and TgM immunogenicity were corresponding for assessment of their ex vivo immunogenicity ( ability to activate memory T-cells ) in individuals with matched HLA and history of DENV3 infections . These individuals were enrolled in a dengue cohort established in Recife , Brazil , and described elsewhere [37] . The peptides NS3399–407 and NS5318–326 ( A02 epitopes ) ; NS5389–398 , Env226–234 and NS3593–601 ( B07 epitopes ) ; Env126–140 , Env231–245 , NS169–83 , NS185–99 and NS3357–371 ( DR2 epitopes ) activated memory T-cells of at least one of the subjects analyzed , suggesting that these epitopes are naturally processed and presented to T-cells during the course of dengue infection and , thus , are relevant to humans . However , we cannot rule out the possibility that the epitopes tested could have been presented by HLA molecules other than those studied herein or the memory T-cell clones activated were only cross-reactive to DENV3 . Further studies are needed to determine the HLA restriction and the specificity of these T-cell clones . We also analyzed the ability of the epitopes to prime naïve T-cells . Two selected DENV3 epitopes , NS3399–407 ( and its variants on DENV2 and DENV4 serotypes ) and NS5389–398 , were analyzed for their ability to activate naïve T-cells from subjects either HLA-A02 or HLA-B07 positives who had never been exposed to dengue virus infection by any serotype ( dengue IgG negative ) . Naïve T-cells were successfully primed by both DENV3 epitopes ( NS3399–407 and NS5389–398 ) and the NS3399–407 variant in DENV4 , but not by the DENV2 variant . The residues lysine ( K ) and leucine ( L ) at the amino termini of the peptide that are part of the HLA-A*0201 binding motif ( www . syfpeithi . de ) seemed to determine the immunogenicity of the peptides as both residues were present in the DENV3 NS3 epitope and its DENV4 variant , but not in the DENV2 variant ( Table 2 ) . Notably , DENV3 NS3399–407 -primed CD8 T-cells have been shown to recognize both DENV2 and DENV4 variants leading to a dysfunctional T-cell response [70] . Thus , more studies are needed to determine the role of these altered peptide ligands on disease outcome . Nonetheless , the data suggests that the epitopes are naturally processed and can potentially prime naïve T-cells , which is important for vaccine design . Our epitope screening strategy revealed HLA class I epitopes found only in non-conserved regions of the virus proteome . We analyzed the intra- and inter-serotype variants of the A02-restricted epitope NS3399–407 ( KLNDWDFVV ) among the sequences deposited in the NCBI Entrez Protein Database [45] to assess for the possibility of altered peptide ligands within DENV3 and between the serotypes . Based on the analysis , an intra-serotype variant ( RLNDWDFVV ) and four inter-serotype variants were identified ( Table 2 ) . In silico analysis showed that all , but DENV1 variant had HLA-A*0201 binding motif , and , thus , capable of potentially acting as an altered peptide ligand , leading to aberrant memory T-cell responses . A study by Hertz et al [71] using targeting efficiency analysis of different HLA class I molecules has shown that HLA molecules preferentially target conserved regions on the proteome of different viruses [71] . This was thought to be the case on DENV as well [12] . Nevertheless , virus belonging to Flaviviridae family , such as DENV , are exception to this rule , since there is an increased preference of HLA molecules to bind non-conserved regions among the different virus serotypes [71] . However , the authors found an association between preferential HLA targeting to conserved regions and protection against dengue severity caused by DENV2 [71] . Hence , it seems that targeting conserved sequences is desirable not only in terms of achieving a broad and efficient pan-dengue immune responses , but also to protect against disease severity . In contrast to the HLA class I , where we did not find conserved immunogenic regions , we identified 23 HLA class II immunogenic peptides that contained sequences highly conserved in each of the four dengue serotypes ( pan-dengue conserved sequences ) [45] . In TgM model , these conserved peptides were as immunogenic ( among DR2 , DR3 and DR4 ) and bound to as many HLA molecules as the peptides containing non-conserved sequences . Notably , these sequences are predicted to actively interact with the major pockets in the groove of HLA-DR2 , -DR3 and -DR4 molecules , thus , defining the importance of these pan-dengue conserved sequences for the affinity and specificity of the epitopes identified . Moreover , some of the epitopes bearing pan-dengue conserved sequences could bind as many as 14 different HLA molecules , suggesting that these immunogenic conserved sequences could trigger T-cell responses in a vast number of people . However , it is important to highlight that these immunogenic and HLA promiscuous peptides are not optimum epitopes , thus , additional analysis are needed in order to identify the exact sequences responsible for HLA affinity and ultimately T-cell activation . The importance of these epitopes in preventing dengue infection and disease severity as well as population coverage also needs to be further addressed . Currently there are as many as 15 reported human T-cell epitopes for DENV3 in the literature ( Table 4; Figure 8; as of December 2012 ) . Among these , the DR2-restricted epitope NS3348–362 ( GNEWITDFVGKTVWF ) reported by Mangada & Rothman [25] using a cohort of individuals immunized against DENV3 virus correlated with four overlapping DENV3 DR2 TgM epitopes identified herein , covering a larger region ( NS3345–370 ) . Two of the four overlapping epitopes ( NS3349–363 and NS3357–370 ) had the specificity for DR2 molecule confirmed by binding assay . Thus , there are possible multiple overlapping epitope in the region described by Mangada & Rothman . Notably , the peptide NS3357–370 ( GKTVWFVPSIKAGND ) , which overlapped 6 amino acids with the epitope reported by Mangada & Rothman [25] , was observed to bind to DR2 with even greater affinity and elicit memory T-cell response at a lower concentration . Additionally , we confirmed the HLA affinity of some of the previously reported epitopes . For instance , we identified a potential DR2 ( EEMFKKRNLTIMDLH ) and a DR4 ( KKRNLTIMDLHPGSG ) epitope overlapping the NS3187–201 ( RNLTIMDLHPGSGKT ) with no reported HLA affinity; similarly , a B07 ( RPRWLDA ) epitope within the NS3585–599 ( KEGEKKKLRPRWLDA ) [25] . Recently , Weiskopf et al [17] reported DENV2 T-cell epitopes using IFN α/βR KO HLA transgenic mice , an animal model that supports dengue replication and also expresses HLA-A*0201 , HLA-B*0702 , among other HLAs . In their study , the authors used DENV2 predicted epitopes for splenocyte challenge and a mouse-adapted DENV2 strain for immunization , and among the epitopes identified were five with identity greater than 89% in DENV3 ( identity not observed for other serotypes ) . Among these were three HLA-B*0702 epitopes ( NS3205–213: LPAIVREAI; NS3223–232: APTRVVAAEM; NS3276–283: VPNYNLIIM ) , however , none was positive in our study with DENV3 involving TgM . This could be because of the amino acid differences between the serotypes , however the increased virus replication within the animal model used might have contributed to the increased breadth and magnitude of T-cell response . Additionally , other factors might concomitantly facilitate the T-cell responses in the IFN α/βR KO HLA transgenic mice , such as increased antigen availability . The use of HLA TgM has been shown to be useful tool for the identification of potential T-cell epitopes , although a validation step is required [44] . Several studies have shown correlation between epitopes identified in HLA TgM and humans [42] , [43] , [53] . Herein , we have identified a vast repertoire of potential DENV3 T-cell epitopes by use of TgM , with HLA affinity validated for many by use of binding assay , and immunogenicity confirmed for a subset in HLA-matched subjects with history of dengue infection . The analysis of intra- and inter-serotype variants performed is important to identify and understand the role of altered peptide ligands and T-cell activation in the context of disease severity caused by DENV and on the other hand population coverage . Notably , a more extensive human validation study is currently underway employing the same peptide library used herein . The data to date reveal that twenty-seven ( 27 ) of the TgM immunogenic peptides ( considering only Envelope , NS1 and NS3; data not shown ) are also immune-prevalent ( recognized by at least 10% of the subjects tested ) in a human cohort of subjects naturally exposed to DENV3 ( Souza et al . , manuscript in preparation ) . The most advanced dengue vaccine candidates in clinical trials include the Walter Reed Army Institute of Research-GlaxoSmithKline ( WRAIR-GSK ) new live attenuated virus vaccine that are in phase II [72] and the live chimeric virus vaccines , from Sanofi Pasteur ( ChimeriVax ) , currently in phase III [73] . The WRAIR-GSK vaccine can induce immune responses against dengue structural and non-structural viral proteins , while ChimeriVax induces responses to dengue prM and Env proteins and yellow fever non-structural proteins . The ability of these vaccines to induce T-cell responses has been reported [74] , [75] , [76] . The complete T-cell repertoires of these vaccines are not known , but they can potentially induce T-cell responses to both conserved and non-conserved epitopes . Herein we report highly conserved , HLA promiscuous dengue T-cell epitopes ( mostly found in the non-structural proteins of the virus ) that are capable of inducing T-cell responses against the majority of dengue variants of each of the four serotypes and applicable to human population . We postulate that it would be possible to engineer a dengue vaccine that could target specifically the conserved T-cell epitopes . This may be addressed by priming the immune system with highly conserved , HLA promiscuous T-cell epitopes concomitantly with dengue prM/Env proteins . This may not only help to produce high affinity neutralizing antibodies , but also would expand T-cell clones specific for highly conserved dengue T-cell epitopes . These T-cells would be effectively activated after virus exposure in the field to produce cytokines ( e . g . IFN-γ ) associated with disease protection . Further studies are needed to identify the cytokine repertoire that such epitopes may induce and understand their role in dengue protection/pathogenesis , as well as establish proper epitope delivery for efficient antigen presentation and T-cell activation . The data reported herein are valuable resource for further studies investigating the role of T-cells in dengue protection/immunopathogenesis and design of tetravalent dengue vaccine . | Although there is an increased recognition of the role of T-cells in both dengue pathogenesis and protection , comprehensive analysis of T-cell activation during dengue infection is hampered by the small repertoire of known human dengue T-cell epitopes . Although dengue serotype 3 ( DENV3 ) is responsible for numerous outbreaks worldwide , most of the known epitopes are from studies of dengue 2 serotype ( DENV2 ) . In this study , we identified novel DENV3 T-cell epitopes in HLA transgenic mice that were confirmed by HLA binding assays . A subset of these epitopes activated memory T-cells from subjects who were dengue IgG positive and primed naïve T-cells from dengue IgG negative individuals . Notably , some of HLA class II epitopes bearing highly conserved regions common to all four dengue serotypes could bind to multiple HLAs . We postulate that these highly conserved and HLA promiscuous T-helper epitopes can be important components of a dengue tetravalent vaccine . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Identification of Conserved and HLA Promiscuous DENV3 T-Cell Epitopes |
Telomerase , particularly its main subunit , the reverse transcriptase , TERT , prevents DNA erosion during eukaryotic chromosomal replication , but also has poorly understood non-canonical functions . Here , in the model social amoeba Dictyostelium discoideum , we show that the protein encoded by tert has telomerase-like motifs , and regulates , non-canonically , important developmental processes . Expression levels of wild-type ( WT ) tert were biphasic , peaking at 8 and 12 h post-starvation , aligning with developmental events , such as the initiation of streaming ( ~7 h ) and mound formation ( ~10 h ) . In tert KO mutants , however , aggregation was delayed until 16 h . Large , irregular streams formed , then broke up , forming small mounds . The mound-size defect was not induced when a KO mutant of countin ( a master size-regulating gene ) was treated with TERT inhibitors , but anti-countin antibodies did rescue size in the tert KO . Although , conditioned medium ( CM ) from countin mutants failed to rescue size in the tert KO , tert KO CM rescued the countin KO phenotype . These and additional observations indicate that TERT acts upstream of smlA/countin: ( i ) the observed expression levels of smlA and countin , being respectively lower and higher ( than WT ) in the tert KO; ( ii ) the levels of known size-regulation intermediates , glucose ( low ) and adenosine ( high ) , in the tert mutant , and the size defect’s rescue by supplemented glucose or the adenosine-antagonist , caffeine; ( iii ) the induction of the size defect in the WT by tert KO CM and TERT inhibitors . The tert KO’s other defects ( delayed aggregation , irregular streaming ) were associated with changes to cAMP-regulated processes ( e . g . chemotaxis , cAMP pulsing ) and their regulatory factors ( e . g . cAMP; acaA , carA expression ) . Overexpression of WT tert in the tert KO rescued these defects ( and size ) , and restored a single cAMP signaling centre . Our results indicate that TERT acts in novel , non-canonical and upstream ways , regulating key developmental events in Dictyostelium .
Each time a chromosome replicates , it loses some DNA from each of its ends . This is not necessarily problematic , because the chromosome is initially capped at each end by a sacrificial strand of non-coding DNA , a telomere [1–3] . Further instances of replication , however , can expose the coding DNA , unless the cell can keep repairing the shortened telomeres , by the action of the enzyme complex , telomerase . Accordingly , telomerase , whose main subunits comprise a reverse transcriptase ( TERT ) , and the telomerase RNA component ( TERC ) [4] , has much significance in the biology and pathology of multicellular organisms . As somatic tissues age , for example , telomerase is downregulated , and the resulting telomeric dysfunction can lead to chromosomal instability and various pathologies , including disrupted pregnancies and cancer [5–7] . In other cases , the upregulation of telomerase is also associated with , and a biomarker of , some cancers , because it allows the unchecked proliferation of immortalised tumour cells [6 , 8] . Telomerase also has many non-canonical roles , in which telomere maintenance , or even telomerase activity , is not required [9 , 10] . For example , telomerase is known to have non-canonical roles in neuronal differentiation [11] , RNA silencing [12] , enhanced mitochondrial function [13] , cell adhesion and migration [14 , 15] and various cancers [9 , 16] . Our understanding of telomeres and telomerase began , and has continued to develop , through the study of model organisms such as Drosophila , Zea mays , Tetrahymena , yeast and mice [2 , 3 , 17–21] . One model system in which the possible roles of telomerase have not yet been addressed is Dictyostelium discoideum . This system has proved its usefulness in many contexts , including the study of human diseases [22–26] . One of its advantages is that the processes of cell division ( i . e . growth ) and development are uncoupled [27] , making the organism a highly tractable system for the study , in particular , of differentiation and tissue size regulation [28–35] . In culture , when its bacterial food source is abundant , D . discoideum multiplies as single-celled amoebae . This leads to denser colonies , and exhaustion of the food supply . The rising concentration of a secreted glycoprotein , CMF , triggers the organism to switch to a multicellular mode of development [34 , 36] . With no resources for further cell proliferation , the amoebae move , in a radial pattern of streams , towards centres of aggregation . Rising levels of secreted proteins , of the counting factor ( CF ) complex [37 , 38] , trigger a series of changes that lead to breaking up of the streams , which therefore no longer contribute cells to the original aggregate . Each aggregate , which will typically contain 20 , 000 to 100 , 000 cells [39] , now rounds up into a mound , which then proceeds through several life-cycle stages , finally forming a spore-dispersing fruiting body about 1-2mm high [34 , 40] . Mounds can also develop from the breaking-up of a large stream ( or aggregate ) , a process similarly regulated by CF [29 , 41] . The generic term , ‘group’ , can be used to address the fact that mounds develop from clusters that arise in these slightly different ways , but in this paper we will refer to ‘mounds’ . Some of the processes and regulators involved in our very abbreviated account of the life-cycle are shown in Fig 1 , which focuses on those elements examined in this study . In addition to being uncoupled from growth , development in D . discoideum has other features that make it potentially useful as a model system for the understanding of telomerase-based pathologies , in particular cancers that arise from disruption of non-canonical functions . First , as indicated in Fig 1 , development in D . discoideum depends on properly regulated cell motility and cell adhesion , two processes fundamental to metastasis . Second , the switch to multicellular development , and the control of aggregate , mound and hence fruiting body size are influenced by various secreted factors that , respectively , promote aggregation and regulate tissue size , in ways analogous to the regulation of tumour size by chalones [42 , 72] . Third , a putative TERT has been annotated in the D . discoideum genome . It is not known if the RNA component of telomerase ( TERC ) is present [73] and , in any case , extrachromosomal rDNA elements at the ends of each chromosome in D . discoideum suggest a novel telomere structure [74] . Thus , telomerase in this organism may have a separate mechanism for telomere addition or might have non-canonical roles . As yet , however , there have been no functional studies of TERT reported for D . discoideum . In this study , we characterize the gene tert in D . discoideum , showing that it has both RT and RNA binding domains . We describe the pattern of tert’s expression levels during all stages of development , assay for any canonical telomerase function , and examine the effects of knocking out the gene’s function on development . The tert mutant exhibits a wide range of developmental defects , suggesting that wild-type TERT targets multiple elements of the regulatory network depicted in Fig 1 . Most interestingly , these defects , and the results of experiments by which we attempt to rescue , or phenocopy , the tert KO phenotype , suggest that this telomerase influences the activity of smlA , and processes downstream of it . Tert thus emerges as one of the upstream genes of the cell-counting pathway , and its overall influence indicates that , despite having no obvious canonical activity , a telomerase can nevertheless play major regulatory roles by virtue of its non-canonical targets .
Extending previous predictions of tert encoding a protein with telomerase motifs [75] , our use of the Simple Modular Architecture Research Tool ( http://SMART . embl-heidelberg . de ) and UniProt ( Q54B44 ) revealed the presence of a highly conserved reverse transcriptase domain and a telomerase RNA binding domain ( S1 Fig ) . These are characteristic of a telomerase reverse transcriptase [76] , supporting the idea that the gene we characterized indeed encodes for TERT . The Dictyostelium TERT protein shares 23% and 18 . 7% identity with human and yeast TERT protein respectively ( Pairwise sequence Alignment-Emboss Needle ) . The protein sequence identities between the TERT of D . discoideum and five other species are tabulated in S1 Table . In the case of the identity with the TERT of humans , the strongest homologies are seen in the reverse transcriptase domain . We did a phylogenetic analysis to examine the relatedness of DdTERT with that of other organisms . For this , TERT amino acid sequences from different organisms were obtained from the NCBI database or Dictybase ( http://www . dictybase . org/ ) or SACGB database ( http://sacgb . leibniz-fli . de ) and compared with TERT of D . discoideum . Multiple sequence alignment of the TERT amino acid sequences of various organisms including other social amoebae were used to create the phylogenetic tree , employing the MUSCLE alignment feature of MEGAX software [77] . The phylogenetic analysis suggests that D . discoideum TERT falls in a separate clade and is likely to be a distant relative of vertebrate homologs ( S2 Fig ) . The evolutionary history was inferred using the Neighbor-Joining method [78] . The evolutionary distances were computed using the p-distance method and the units shown are the number of amino acid differences per site . Further , using the fold recognition technique on the I-TASSER server , the structure of D . discoideum TERT was predicted using Tribolium castaneum ( telomerase in complex with the highly specific inhibitor BIBR1532; PDB-5cqgA ) as a template ( S3 Fig ) . The modeled structure of Dictyostelium TERT also suggests that D . discoideum has a structurally conserved TERT ( S3 Fig ) . Telomerase activity , if any , can be ascertained by performing a Telomeric Repeat Amplification Protocol ( TRAP ) assay , and activity has been successfully detected in organisms such as humans , C . elegans , yeast , Daphnia , and plants [79–84] . However , while human cell lines ( HeLa , HEK ) did show telomerase activity , we did not detect any telomerase activity in D . discoideum cell extracts ( S4 Fig ) . This concurs with previous findings , namely that the telomeres of D . discoideum have a novel structure [85] , and that , in other organisms , TERT has several non-canonical roles [11–13] . In humans , telomerase expression is reported to be low in somatic cells compared to germline and tumour cells [86] . To ascertain if tert expression is differentially regulated during growth and/or development , we performed qRT-PCR using RNA from different developmental stages ( 0 , 4 , 8 , 10 , 12 , 16 and 24 h after starvation ) . Tert expression is higher in development than during growth , ( 8h and 12 h ) ( Fig 2 ) , implying that tert plays a prominent role beyond the point at which D . discoideum is responding to starvation . Expression also shows a marked biphasic pattern , with the first peak at 8h ( when streams are forming ) , a big dip during stream breaking ( 10h ) and then rising gradually again to peak at about the time of mound formation ( 12h ) . To understand the possible non-canonical roles of tert in development of D . discoideum , tert KO cells generated by homologous recombination were seeded at a density of 5x105 cells/cm2 on non-nutrient buffered agar plates and monitored throughout development . While aggregates appeared by 8 h in the wild-type , and streams began to break at 10 h , in the mutants there was a further 8 h delay before aggregates were seen , and stream breaking began at about 18 h . Because of these delays , ‘during aggregation’ , in this study , refers to 8 h in WT and 16 h in the tert KO , and ‘during stream breakup’ refers to 10 h in WT and 18 h in the tert KO . Wild-type cells formed long streams of polarized , elongated cells leading to aggregation , but tert KO cells did not form well-defined streams , failing to aggregate even at 5x104 cells/cm2 ( wild-type cells aggregated even at a density of 2x104 cells/cm2 ) , suggesting an inability to respond to aggregation-triggering conditions ( S5 Fig ) . The mutant’s streams were also larger ( Fig 3A ) . In contrast to streams moving continuously towards the aggregation centre in WT , tert KO streams break while they aggregate ( S1 and S2 Videos ) . They did eventually form aggregates , largely by clumping . During the early stages of aggregate formation , the number of aggregation centres formed by the tert KO was only 10% of that formed by WT ( Fig 3B , p<0 . 0001 ) . Due to uneven fragmentation , the late aggregates were also of mixed sizes . The tert KO cells did eventually form all of the typical developmental structures , but by the mound stage , continued fragmentation had resulted in the mounds being more numerous , and smaller , on average , than in the WT . This was also the case for fruiting bodies . Thus , with reference to Fig 1 , tert appears to play roles in multiple aspects of Dictyostelium development: the timing of aggregation; streaming; and the regulation of the size of the mound and fruiting body ( Table 1A and 1B ) . Given the wide-ranging phenotypic defects seen in the tert KO , it seemed likely that tert is one of the key regulators of development in D . discoideum , affecting many of the processes and regulators depicted in Fig 1 . We thus monitored the activity or levels of a number of those elements , comparing the wild-type and tert KO ( summarised in Table 1A and 1B ) . As that summary shows , the tert KO showed significant changes from the wild-type in three broad areas: components of the mound-size regulation pathway; cAMP-related processes/regulators; and adhesion-related processes/regulators . As is clear from Fig 1 , the factors that influence these features overlap considerably , both in terms of interacting with each other , and in regulating more than one of the various developmental stages disrupted in the tert KO . Nevertheless , we think it is useful to consider each of them in turn . As we do so below , we describe a series of experiments that largely fall into two broad categories , as shown in summary form in Tables 2 and 3: Those that attempt to rescue the normal phenotype in tert KO cells ( Table 2 ) ; and those that attempt to phenocopy , or induce , the tert KO phenotype in wild-type cells ( Table 3 ) . First , however , we describe some experiments that support the direct involvement of tert in the effects already noted . To support the idea that the changes observed in the tert KO are , in the first instance , due to changes involving tert itself , and not some other factor , we took two approaches: Overexpression of tert , and the use of TERT inhibitors . Most importantly , overexpression of wild-type TERT ( act15/gfp::tert ) in tert KO cells rescued all three of the phenotypic defects ( Fig 4A , S3 Video; Table 2 ) , suggesting that the tert KO phenotype is not due to any other mutation . Next , we treated wild-type cells with structurally unrelated TERT specific inhibitors , BIBR 1532 ( 100nM ) and MST 312 ( 250nM ) . BIBR 1532 is a mixed type non-competitive inhibitor , whereas MST 312 is a reversible inhibitor of telomerase activity ( see Methods ) . Both inhibitors strikingly phenocopied two features of the tert mutant , in that we observed large early aggregate streams that broke and eventually resulted in mounds ( Fig 4B; Table 3 ) and fruiting bodies that were small . The developmental delay , however , was not induced . Since the two inhibitors phenocopied the tert KO to a remarkable degree , it is likely that the inhibitor binding sites of Dictyostelium TERT are conserved . Human TERT [87] , which shares a 23% homology with Dictyostelium TERT , failed to rescue the tert KO phenotype ( S6 Fig ) . Surprisingly , the morphologies of TERT-overexpressing lines in the wild-type did not show any significant difference to those of the untreated wild-type ( Fig 4A ) . Overall , these results strongly support the idea that the relevant changes in the tert KO involve tert itself . The fact that the TERT inhibitors induced only two of the three tert KO defects is interesting . Given the lack of any apparent interconnection between the pathway that regulates the switch to aggregation , and that regulating mound size , it seems likely that TERT acts on more than one molecular target . It could be that the inhibitors do not perturb that part of TERT that interacts with the target that regulates the switch to development . Given the perturbations seen in the tert KO , one would predict some abnormalities associated with cAMP dynamics [44–46 , 88–90] . The role of cAMP in streaming , in particular , has been much studied . Below we examine how various cAMP processes or factors , related to streaming and developmental delay , were affected in the tert KO . Cell-substratum adhesion is also important for migration and proper streaming . By shaking cells at different speeds ( 0 , 25 , 50 and 75 rpm ) , it is possible to vary substratum dependent sheer force . Thus , by counting the fraction of floating cells at different speeds , it is possible to check substratum dependent adhesion . Although both wild-type and tert KO cells exhibited a sheer force-dependent decrease in cell-substratum adhesion , tert KO cells exhibited a significantly weaker cell-substratum adhesion ( S12 Fig , p<0 . 0001 ) , affecting cell motility in a way that might also contribute to stream breaking . Cell-cell adhesion is also an important determinant of streaming and mound size in Dictyostelium [41] . To examine if adhesion is impaired in the mutant , we checked the expression of two major cell adhesion proteins , cadA , expressed post-starvation ( 2 h ) and csaA expressed during early aggregation ( 6 h ) . cadA-mediated cell-cell adhesion is Ca2+-dependent and thus EDTA-sensitive , while csaA is Ca2+ independent and EDTA-resistant [67] . Both csaA and cadA expression were significantly down-regulated ( Fig 14A and 14B ) . Further , cell adhesion was monitored indirectly by counting the fraction of single cells not joining the aggregate . Aggregation results in the gradual disappearance of single cells and thus it is possible to measure aggregation by determining the ratio of single cells remaining . To examine Ca2+-dependent cell-cell adhesion , the assay was performed in the presence of 10 mM EDTA . Both EDTA-sensitive and resistant cell-cell adhesion were significantly defective in tert KO cells ( Fig 14C , p = 0 . 0033 and 14D , p = 0 . 0015 ) . The levels of csaA and cadA were also lower in the tert KO during aggregation when compared to the WT ( Fig 14E , p = 0 . 0037 and 14F , p = 0 . 0508 ) . Thus , the delay in tert KO development might be the basis for differences in gene expression . These results imply that defective cell-substratum and cell-cell adhesion might play roles in the abnormal streaming and mound-size regulation of the tert KO . One interesting observation was that the only treatment that fully rescued the tert KO cells was the overexpression of wild-type tert . Also , the only other treatment that rescued the developmental delay itself was mixing wild-type cells with the tert KO cells at a 1:1 ratio ( Fig 15; Table 2 ) . Even though caffeine and glucose rescued streaming and mound size , and apparently this was at least partly mediated via their impact on cAMP-regulated processes , neither of the compounds rescued the delay , even though abnormalities of cAMP-regulated processes are commonly reported causes of delay in other Dictyostelium studies [44–46] . Thus , we examined polyphosphate levels in the tert KO because of their known importance to developmental timing in Dictyostelium [43] . We stained the CM with DAPI for 5 minutes and checked the polyphosphate specific fluorescence using a spectrofluorometer . The CM of tert KO cells has reduced polyphosphate levels ( 49 . 55±2 . 02 μM ) compared to wild-type ( 60 . 62±1 . 95 μM ) , implying that low polyphosphate levels might also contribute to the delay in initiating development in this system ( Fig 16 , p = 0 . 0009 ) . Our results reveal that TERT plays an important role in many aspects of Dictyostelium development . The tert KO exhibited a wide range of developmental defects . Despite suitable environmental conditions for multicellular development to begin , the start of the streaming phase is delayed by 8 h . Having once begun , development proceeds and ends abnormally , with large streams , uneven fragmentation , and , eventually , small mounds and fruiting bodies . The wide-ranging developmental defects are associated with changes to the levels , or expression , of genes and compounds that are known to be highly upstream regulators of the various stages of development , such as streaming and mound/fruiting body formation . Based on the perturbations in the tert KO , and our other experiments , Fig 17 depicts the possible extent of processes , and potential mediating factors , that might depend upon normal tert expression/TERT activity in the wild-type . Note that the arrows that connect tert/TERT to any element in the diagram are not meant to suggest that TERT directly regulates that element , only that TERT is important , perhaps in some indirect way , for the normal levels , or activity , of that element . One of the most striking findings was that TERT appears to regulate , or is at least necessary for , the normal activity of what was previously known as the most upstream regulator of mound size , smlA [28 , 30 , 32] . Expression levels of smlA were reduced in the tert KO , and we also observed a wide variety of the expected downstream effects of lowered smlA levels . All of these , and a wide variety of treatments that rescued the size-defect of the mutant phenotype , support the idea that the reduction of mound size in the tert KO was indeed mediated via the abnormal functioning of the previously-identified elements of the mound-size regulation pathway . In addition to the rescue approach , treatments that attempted to phenocopy the tert KO phenotype in the wild-type , also suggest TERT is one of the upstream regulators of mound size . In particular , given that size regulation in D . discoideum depends upon secreted factors of the CF complex , one would have predicted the effects we observed when tert KO CM was added to wild-type cells . Another strong indication that tert acts upstream , at least of CF , was that the inhibition of tert activity in countin mutants failed to phenocopy the tert KO phenotype . A similarly rich range of results ( involving the tert KO phenotype , and its rescue , and phenocopying ) support the idea that TERT also plays a high-level role in the regulation of streaming . During the streaming phase , two genes associated with cAMP related-processes in D . discoideum ( acaA , carA ) were significantly downregulated ( compared to the wild-type ) , and the levels of several other genes trended lower . This was also accompanied by lower cAMP levels . This might explain the defective chemotaxis and cell motility of the tert KO . Of course , the regulation of streaming is not entirely isolated from that of size . Glucose , one of the central elements of the CF pathway , influences several cAMP-related processes [64] . Thus , it was not surprising that adding 1 mM glucose to the tert KO cells rescued both the size and streaming defects . This study , however , provided a new insight into how the rescue of streaming occurs , because added glucose also reduced adenosine levels . Thus , in the tert KO , the low glucose levels might lead to higher adenosine levels , allowing it to inhibit cAMP related processes ( via pathway i , Fig 17 ) . In normal development , given the known sequence of the telomere repeats of D . discoideum ( A-G ( 1–8 ) ; [74] ) , and the fact that telomerase activity would therefore recruit cellular stores of adenosine , it is possible that normal TERT activity keeps adenosine levels low . As yet , however , whether TERT actually acts as a functional telomerase in D . discoideum is not known . The tert gene we characterized includes the conserved domains and structure of a telomerase reverse transcriptase . Also , supplementing structurally unrelated but specific inhibitors of TERT to wild type cells phenocopies the mutant phenotype . The widely used method to test telomerase activity is the TRAP assay . However , this method failed to detect telomerase activity in D . discoideum and there may be both technical and innate limitations . For example , possible reasons for the lack of any observed activity are that: ( i ) the presence of rDNA palindrome elements in the chromosomal ends , suggesting a novel telomere structure and the possible role of TERT in maintaining both rDNA and chromosomal termini [74] . This could be an alternate pathway of telomere maintenance in D . discoideum; and ( ii ) polyasparagine repeats , present in the TERT protein of Dictyostelium , splitting the functional domain into two halves . For telomerase activity , a functional TERT is important in humans [100–102] . In yeast as well as humans , truncation of one of the TERT protein domains is known to abolish its function [103 , 104] . While it is not yet clear whether the apparent absence of canonical TERT function in D . discoideum is due to the absence of normal eukaryote telomeres [105] , other studies suggests that TERT is not always associated with telomerase activity . The silkworm genome contains a telomerase gene , but the telomerase itself displays little or no enzymatic activity [106 , 107] . The telomeres of silkworm consist of the telomeric repeats typical of insects , but also harbor many types of non-LTR retrotransposons [106 , 108 , 109] . Also of interest is that species of Calcarea ( sponges ) , Cnidaria ( sea anemones and jellyfish ) and Placozoa , all have metazoan telomeric sequences , but display little or no telomerase activity [110] . D . discoideum might employ an alternative mode of telomere addition , such as the recombination seen in yeast [111] or the retrotransposition of Drosophila [112 , 113] . The discussion so far , while it establishes that TERT is needed for several developmental processes to take place , does not help to distinguish whether or not it acts more than once , or if it has more than one target . Could TERT for example act more like the much studied homeodomain proteins , master regulators of animal development , but which only act during very early embryological life [114 , 115] ? Likewise , in D . discoideum , CMF appears to act only once [34] . Two lines of argument suggest that TERT is different . First , the biphasic nature of tert's expression pattern suggest that it could possibly act during two stages of development . In the wild-type , tert expression builds up to its first peak at 8 h , thus being a potential candidate for enabling streaming to begin , and to proceed correctly , around this time . It then dips markedly to a low point at 10 h , whereby it might help to enable stream break-up by its relative absence . Then , it begins its climb to its second peak at 12 h , when mound size is being finalised . However , it is also possible that the later-occurring defects seen in the tert KO correspond to pleiotropic effects of TERT being absent at a much earlier time-point . Second , while it is well known that cAMP-related processes play important roles in allowing streaming to begin and to proceed properly , and while we have shown that TERT influences multiple cAMP related processes , the pathway by which TERT influences the initiation of streaming seems distinct from that used for maintaining it . Both glucose and caffeine , for example , rescued the streaming and size defects of the tert KO , but the delay was unaffected . Complementarily , when wild-type cells were mixed at 50% with tert KO cells , they rescued the delay defect only . In fact , the only treatment that fully rescued the tert KO was the overexpression of wild-type tert . Interestingly , MAP kinase kinase ( MEK1 ) disruption results in a stream-breaking phenotype similar to the tert KO [56] , suggesting that MEK1 could be involved in either CF secretion or signal transduction . Also , signals transmitted through p38 mitogen-activated protein kinase ( MAPK ) regulate hTERT transcription in human sarcoma [116] . We speculate that MEK1 might regulate countin levels through TERT , thus helping to regulate tissue size in D . discoideum . Also , it is known that MST 312 ( a TERT inhibitor ) treatment reduces tumour size by 70% in a mouse xenograft model and this inhibition preferentially targets aldehyde dehydrogenase-positive cancer stem cell-like cells in lung cancer [117] . In Dictyostelium , disruption of aldehyde reductase increases group size [118] and , since aldehyde dehydrogenase and aldehyde reductase have opposing activities ( oxidation and reduction of aldehydes respectively ) , they might have opposite functions in group size regulation as well . TERT might possibly be regulating aldehyde reductase activity in determining mound size in D . discoideum . Other genes are also known to play a significant role in aggregate size determination in Dictyostelium , such as dio3 [119] and pkc [120] . However , it is not known if they interact with TERT in determining mound size . This study indicates for , the first time , that TERT acts in several non-canonical ways in D . discoideum , influencing when aggregation begins , the processes involved in streaming , and the eventual size of the fruiting body . TERT's influences appear to occur upstream of many other regulators of streaming and fruiting body size . Curiously , as yet we have no evidence that TERT acts as a canonical telomerase , nor is it known whether any other enzyme protects the unusually sequenced telomeres of this species . Given that telomere research is still in progress , we cannot even rule out that TERT’s apparently non-canonical roles in D . discoideum development are in fact mediated via some as-yet unidentified action on its unusual telomeres . In the most heavily studied stages of the organism’s life-cycle , that is , those that occur in response to starvation , replication has ceased , so further study of this particular point should focus on the amoeboid stage . More generally , this study has revealed a previously unreported non-canonical process influenced by a telomerase , tissue size regulation . This role of TERT , together with its influence on cell motility and adhesion , and the levels of chalone-like secreted factors , bear consideration by those engaged in cancer research .
Wild-type D . discoideum ( AX2 ) cells were grown with Klebsiella aerogenes on SM5 plates , or axenically , in modified maltose-HL5 medium ( 28 . 4 g bacteriological peptone , 15 g yeast extract , 18 g maltose monohydrate , 0 . 641 g Na2HPO4 and 0 . 49 g KH2PO4 per litre , pH 6 . 4 ) containing 100 units penicillin and 100 mg/ml streptomycin-sulphate . Cells were also grown in Petri dishes as monolayers . Other dictyostelid species ( D . minutum and D . purpureum ) were grown with Klebsiella aerogenes on SM5 plates and cells were harvested when there was visible clearing of bacterial lawns . To trigger development , cells were washed with KK2 buffer ( 2 . 25 g KH2PO4 and 0 . 67 g K2HPO4 per liter , pH 6 . 4 ) and plated on 1% non-nutrient KK2 agar plates at a density of 5x105 cells/cm2 in a dark , moist chamber [121] . To study streaming , cells were seeded in submerged condition ( KK2 buffer ) at a density of 5x105 cells/cm2 . BIBR 1532 is a specific non-competitive inhibitor of TERT with IC50 value of 93 nM for human telomerase [122] . To find the optimal dose response of BIBR 1532 in Dictyostelium , starved cells were plated in phosphate buffered agar with different concentrations of BIBR 1532 ( 10 nM , 25 nM , 50 nM , 100nM and 200 nM ) and 100nM was found to be the minimal effective dose in inducing complete stream breaking . MST 312 , which is structurally unrelated to BIBR 1532 , is a reversible inhibitor of TERT with IC50 value of 0 . 67 μM for human telomerase [123] . The minimal effective dose in Dictyostelium was found to be 250 nM . Inhibitor treatments were carried out with freshly starved cells resuspended in KK2 buffer and plated on KK2 agar plates . The TRAP assay takes advantage of the low substrate specificity of telomerase , and involves replacing the telomere sequence with a synthetic template . The telomerase first extends the synthetic substrate primer by adding telomere repeats and these primary products are further amplified by PCR . The primer must have certain modifications , such as an anchor sequence at the 5’ end and two mismatches within the telomerase repeats [124 , 125] . For the TRAP assay in Dictyostelium , we have used different primer sets ( S2 Table ) according to the basic design principles [124] . The KO vector for tert disruption was designed following standard cloning procedures . A 5' fragment of 678 bp and a 3' fragment of 322 bp spanning the tert gene ( DDB_G0293918 ) and intergenic regions were PCR amplified and cloned on either side of a bsR cassette in pLPBLP vector ( S13 Fig ) . Restriction endonuclease digestion and DNA sequencing were carried out to confirm the integrity of the KO vector . The tert KO vector was transfected to D . discoideum cells by electroporation . Axenically grown AX2 cells were washed twice with ice-cold electroporation buffer and 1x107 cells were resuspended in 100 μl EP++ buffer containing 10 μg of linearized tert KO vector . The cell suspension mixed with linearized KO vector was transferred to pre-chilled cuvettes ( 2 mm gap , Bio-Rad ) and electroporated ( 300 V , 2 ms , 5 square wave pulses with 5 s interval ) using a BTX ECM830 electroporator ( Harvard Apparatus ) . The cell suspension was then transferred to a Petri dish containing 10 ml of HL5 medium and incubated at 22°C . After 24 h , the cultures were replaced with fresh HL5 supplemented with 10 μg/ml blasticidin ( MP Biomedicals ) . Blasticidin-resistant clones were screened after three days . Genomic DNA isolated from tert KO clones were subjected to PCR analysis to confirm tert disruption using different primer combinations ( S3 Table ) . Using genomic DNA as template , a 3 . 8kb tert sequence was PCR amplified using ExTaq polymerase ( Takara ) and ligated in pDXA-GFP2 vector by exploiting the HindIII and KpnI restriction sites . This vector was electroporated to tert KO and AX2 cells and G418 resistant ( 10 μg/ml ) clones were selected and overexpression was confirmed by semi-quantitative PCR . Primer sequences used for generating the vectors are mentioned in S4 Table . Conditioned medium was prepared as described previously with slight modifications [126] . Briefly , log phase cells of AX2 and tert KO were resuspended at a density of 1x107 cells/ml and kept under shaking conditions for 20 h . Cells were pelleted and the supernatant was further clarified by centrifugation . The clarified supernatant ( CM ) was used immediately . To check the effect of CM on aggregate size , cells were developed in the presence of CM on non-nutrient agar plates and development was monitored . KK2 buffer was used as control . To deplete extracellular CF with anti-countin antibodies , cells were starved in KK2 buffer . After 1 h , the cells were developed with anti-countin antisera ( 1:300 dilution ) in KK2 buffer [65] . To examine countin protein expression levels during aggregation , a Western blot was performed with anti-countin antibody . Cells were resuspended in SDS Laemmli buffer , and boiled for 3 min . Subsequently , the samples were run in a 12% SDS-polyacrylamide gel and Western blots were developed using an ECL Western blotting kit ( Bio-Rad ) . Rabbit anti-countin antibodies were used at 1: 3000 dilution . Log phase cells were starved at a density of 1x107 cells/ml in KK2 buffer in shaking conditions at 22°C for 4 h . At the beginning of starvation , 4x107 cells were removed and resuspended in 2 ml Sorensen phosphate buffer , vortexed vigorously and 0 . 4 ml of cell suspension was pipetted immediately in vials containing 0 . 4 ml ice-cold Sorensen phosphate buffer or 0 . 4 ml of 20 mM EDTA solution . The cell suspension was then transferred to a shaker and incubated for 30 min and 0 . 2 ml of 10% glutaraldehyde was added to each sample at the end of incubation and stored for 10 min . Then , 7 ml Sorensen phosphate buffer was added to each vial . Cell adhesion was indirectly measured by counting the number of single cells left behind using a hemocytometer [127] . To measure cell-substratum adhesion , 5x105 cells were seeded in 60mm Petri dishes and incubated at 22°C for 12 h . The Petri dishes with the cell suspension was placed on an orbital shaker at different speeds ( 0 , 25 , 50 , 75 rpm ) . After 1 h , adherent and non-adherent cells were harvested , counted using a hemocytometer and the fraction of adherent cells was plotted against the rotation speed [58] . To visualize cAMP wave propagation , 5x105 cells/cm2 were plated on 1% non-nutrient agar plates and developed in dark moist conditions at 22°C . On a real-time basis , the aggregates were filmed at an interval of 30 s/frame , using a Nikon CCD camera and documented with NIS-Elements D software ( Nikon , Japan ) . For visualizing cAMP optical density waves , image pairs were subtracted [92] using Image J ( NIH , Bethesda , MD ) . The under agarose cAMP chemotaxis assay was performed as described previously [128] . Briefly , 100 μl of cell suspension starved at a density of 1x107 cells/ml in KK2 buffer was added to outer troughs and 10 μM cAMP was added in the middle trough of a 1% agarose plate . Cells migrating towards cAMP was recorded every 30 s for 15 min with an inverted Nikon Eclipse TE2000 microscope using NIS-Elements D software ( Nikon , Japan ) . For calculating the average velocity , directionality and chemotactic index , each time 36 cells were analyzed . The cells were tracked using ImageJ . Velocity was calculated by dividing the total displacement of cells by time . Directionality was calculated as the ratio of absolute distance traveled to the total path length , where a maximum value of 1 represents a straight path without deviations . Chemotactic index was calculated as the ratio of the average velocity of a cell moving against a cAMP gradient to the average cell speed . It is a global measure of direction of cell motion . Total RNA was isolated from AX2 and tert KO cells at the indicated time points ( 0–24 h ) using TRIzol reagent ( Life Technologies , USA ) [129] . RNA samples were quantified with a spectrophotometer ( Eppendorf ) and were also analyzed on 1% TAE agarose gels . cDNA was synthesized from total RNA using cDNA synthesis kit ( Verso , Thermo-scientific ) . 1 μg of total RNA was used as a template to synthesize cDNA using random primers provided by the manufacturer . 1 μl of cDNA was used for qRT-PCR , using SYBR Green Master Mix ( Thermo-scientific ) . qRT-PCR was carried out to analyze the expression levels of tert , acaA , carA , pdsA , regA , pde4 , 5’NT , countin and smlA using the QuantStudio Flex 7 ( Thermo-Fischer ) . rnlA was used as mRNA amplification control . All the qRT-PCR data were analyzed as described [130] . The primer sequences are mentioned in S5 Table . cAMP levels were quantitated using cAMP-XP assay kit as per the manufacturer’s protocol ( Cell Signalling , USA ) . AX2 and tert KO cells developed on 1% KK2 agar , were lysed with 100 μl of 1X lysis buffer and incubated on ice for 10 min . 50 μl of the lysate and 50 μl HRP-linked cAMP solution were added to the assay plates , incubated at room temperature ( RT ) on a horizontal orbital shaker . The wells were emptied after 3 h , washed thrice with 200 μl of 1X wash buffer . 100 μl of tetramethylbenzidine ( TMB ) substrate was added and incubated at RT for 10 min . The reaction was terminated by adding 100 μl of stop solution and the absorbance was measured at an optical density of 450 nm . The cAMP standard curve was used to calculate absolute cAMP levels . Glucose levels were quantified as per the manufacturer’s protocol ( GAHK20; Sigma-Aldrich ) . Mid-log phase cells were harvested and resuspended at a density of 8x106 cells/ml in KK2 buffer and kept in shaking conditions at 22°C . Cells were collected again and lysed by freeze-thaw method . 35 μl of the supernatant was mixed with 200 μl of glucose assay reagent and incubated for 15 min . The absorbance was measured at an optical density of 540 nm . The glucose standard curve was used to calculate absolute glucose levels . Adenosine quantification was performed as per the manufacturer’s protocol ( MET5090; Cellbio Labs ) . Cells grown in HL5 media were washed and seeded at a density of 5x105 cells/cm2 on KK2 agar plates . The aggregates were harvested using the lysis buffer ( 62 . 5 mM Tris-HCl , pH 6 . 8 , 2% SDS , 10% glycerol ) . 50 μl sample was mixed with control mix ( without adenosine deaminase ) or reaction mix ( with adenosine deaminase ) in separate wells and incubated for 15 min . The fluorescence was measured using a spectrofluorometer ( Ex- 550 nm , Em- 595 nm ) . The adenosine fluorescence in the sample was calculated by subtracting fluorescence of control mixed sample from reaction mixed sample . The adenosine standard curve was used to calculate absolute adenosine levels . The conditioned media was incubated with 25 μg/ml DAPI for 5 min and polyphosphate specific fluorescence was measured using a spectrofluorometer ( Ex- 415 nm , Em- 550 nm ) as previously described [131] . Conditioned medium samples were prepared in FM minimal media to reduce the amount of background fluorescence . Polyphosphate concentration , in terms of phosphate monomers were determined using polyphosphate standards . ICP-OES was performed as described previously [99] . Cells were developed on KK2 agar , washed five times in Sorensen phosphate buffer and pelleted . Then , 1 ml of concentrated HNO3 ( 70% ) was added to each sample , and these were further digested by microwave heating . After digestion , the volume of each sample was brought to 9 ml with ultrapure water , filtered with 0 . 45 mm filter and analysed by ICP-OES ( Perkin Elmer Optima 5300 DV ICP-OES ) . Sample digestion and metal quantification were carried out at the SAIF facility ( Sophisticated Analytical Instrument Facility , IIT Madras ) . A Nikon SMZ-1000 stereo zoom microscope with epifluorescence optics , Nikon 80i Eclipse upright microscope or a Nikon Eclipse TE2000 inverted microscope equipped with a digital sight DS-5MC camera ( Nikon ) were used for microscopy . Images were processed with NIS-Elements D ( Nikon ) or Image J . Microsoft Excel ( 2016 ) was used for data analyses . Unpaired Student's t-test and two-way ANOVA ( GraphPad Prism , version 6 ) were used to determine the statistical significance . | When cells divide , their chromosomes are prone to shrinkage . This risk is reduced by an enzyme that repairs protective caps on each chromosome after cell division . This enzyme , telomerase , also has several other important but unrelated roles in human health . Most importantly , via one or other of its functions , both high and low levels of the enzyme can contribute to cancer . We have studied , for the first time , the roles played by telomerase in the life-cycle of the cellular slime mould , Dictyostelium discoideum , a model system with a rich history of helping us understand human biology . While we did not find any evidence of telomerase having the features typically needed to repair a chromosome , telomerase was necessary for many aspects of development . The Dictyostelium telomerase mutant we generated shows delayed aggregation and forms irregular fruiting bodies . The tert mutant miscalculates , in effect , how big those fruiting bodies should be , and they end up being too small . These results are significant because they show , for the first time , that a telomerase can influence tissue size regulation , a process central to a wide range of cancers . | [
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Methods"
] | [
"cell",
"motility",
"glycosylamines",
"chemical",
"compounds",
"chromosome",
"structure",
"and",
"function",
"dictyosteliomycota",
"carbohydrates",
"organic",
"compounds",
"glucose",
"telomeres",
"model",
"organisms",
"experimental",
"organism",
"systems",
"molecular",
"biology",
"techniques",
"adenosine",
"research",
"and",
"analysis",
"methods",
"dictyostelium",
"discoideum",
"slime",
"molds",
"artificial",
"gene",
"amplification",
"and",
"extension",
"animal",
"studies",
"chromosome",
"biology",
"hyperexpression",
"techniques",
"protozoan",
"models",
"chemistry",
"molecular",
"biology",
"chemotaxis",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"gene",
"expression",
"and",
"vector",
"techniques",
"biochemistry",
"eukaryota",
"organic",
"chemistry",
"cell",
"biology",
"phenotypes",
"polymerase",
"chain",
"reaction",
"protists",
"nucleosides",
"monosaccharides",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"dictyostelium",
"glycobiology",
"organisms",
"chromosomes"
] | 2019 | A telomerase with novel non-canonical roles: TERT controls cellular aggregation and tissue size in Dictyostelium |
Podoconiosis ( endemic non-filarial elephantiasis ) is a chronic , non-infectious disease resulting from exposure of bare feet to red-clay soil in tropical highlands . This study examined lay beliefs about three under-researched aspects of podoconiosis patients’ care: explanatory models , health-seeking behaviours and self-care . In-depth interviews and focus group discussions were undertaken with 34 participants ( 19 male , 15 female ) between April-May 2015 at podoconiosis treatment centres across East and West Gojjam regions in north-west Ethiopia . Explanatory models for podoconiosis included contamination from blood , magic , soil or affected individuals . Belief in heredity or divine punishment often delayed clinic attendance . All participants had tried holy water treatment and some , holy soil . Herbal treatments were considered ineffectual , costly and appeared to promote fluid escape . Motivators for clinic attendance were failure of traditional treatments and severe or disabling symptoms . Patients did not report self-treatment with antibiotics . Self-care was hindered by water being unavailable or expensive and patient fatigue . A pluralistic approach to podoconiosis self-treatment was discovered . Holy water is widely valued , though some patients prefer holy soil . Priests and traditional healers could help promote self-care and “signpost” patients to clinics . Change in behaviour and improving water access is key to self-care .
Podoconiosis ( endemic non-filiarial elephantiasis ) is a chronic non-infectious disease affecting individuals whose feet have been exposed to red clay soil found in the tropical highland areas . [1] A very recent study from Molla et al . demonstrates the importance of phyllosilicate clay minerals , particularly smectite , mica groups and quartz ( crystalline silica ) in the pathogenesis of podoconiosis . [2] A strong genetic element has been identified , following an autosomal co-dominant major gene pattern , [3] with identified variation in the human leukocyte antigen ( HLA ) class II locus increasing risk by 2 to 3 times . [4] Those most commonly affected are lower socioeconomic groups , who often struggle to afford shoes , socks or water to wash with . Worldwide , an estimated 4million people live with podoconiosis . [1] A particularly high prevalence has been documented in highland areas of East and Central Africa . [5–7] The prevalence in Ethiopia is 4% , with increasing prevalence at older ages . [8] Approximately 1million people are affected by the disease in Ethiopia . [1] Podoconiosis is characterised by a prodromal phase consisting of itching and burning sensations in the forefoot and lower leg . Early changes observed include plantar oedema , splaying of the toes , hyperkeratosis with the formation of moss-like papillomata and rigid toes . [1] Long-standing disease is associated with fusion of interdigital spaces and ankylosis of interphalangeal or ankle joints . [1] On average , five times per year , patients suffer from episodes of acute adenolymphangitis that are characterised by pyrexia and intense pain , often necessitating days off work . [1] It is estimated that podoconiosis patients lose 45% of total working days per year . [9] In one Ethiopian zone of 1 . 5million people , podoconiosis is estimated to cost over US$16million per year . [9] Stigmatisation against podoconiosis patients is common , [10] with patients being excluded from school , churches , mosques , and barred from marriage with unaffected individuals . [11] Given this background , the discovery of high levels of mental distress and overall lower quality of life amongst podoconiosis patients is unsurprising . [12 , 13] Primary prevention consists of the use of footwear . Secondary prevention involves foot hygiene ( washing daily with water and soap , and using antiseptics and emollients ) , with compression bandaging to reduce soft swelling . A small uncontrolled clinical evaluation from Sikorski et al . demonstrated that clinical improvements can be achieved if these simple measures are strictly adhered to . [14] In 2010 , International Orthodox Christian Charities ( IOCC , an international non-governmental organisation ) started a programme aiming to prevent and treat podoconiosis in East and West Gojjam zones in the Amhara region of Ethiopia . Several studies have identified significant barriers to podoconiosis patients’ attendance at IOCC clinics , with major challenges being: geographical isolation , cost , domestic duties and stigma . [15 , 16] It was hypothesised that additional factors may influence peoples’ health-seeking behaviour , especially early in their illness . Early health-seeking behaviour models were based on pathways to care , in which a stepwise journey starts at identification of symptoms and ends with the use of care . [17] More recent research in this field has moved from pre-defined trajectories and has largely sought to capture wider determinants of health-seeking behaviour . For example , Andersen’s model which groups factors that influence utilisation into three main categories: environmental , predisposing/enabling and health system . [18] This was due to the criticism that health-seeking behaviour research places excessive emphasis on individual choice and an assumption that individuals are autonomous in decision making . [19] We therefore adopted a dynamic notion of health-seeking behaviour where perceived eligibility to receive care represents “a continually negotiated property of individuals , subject to multiple influences arising both from people and their social contexts and from macro-level influences on allocation of resources and configuration of services” . [20] Health-seeking behaviour , therefore , is ultimately determined by the interaction between the individual , their socioeconomic situation and the agents of available treatment services , i . e . not a pre-defined trajectory . An additional dimension of understanding health-seeking behaviour in early stage podoconiosis was to explore explanatory models and their influence on treatment choice and compliance with such treatment . Previous studies highlight that podoconiosis patients fail to comply with daily washing treatment regimes . [21] Poor adherence to treatment regimes in relation to chronic conditions is associated with increased rates of complications and higher treatment costs . [22] Effective communication around conflicting explanatory models is shown to be a “major determinant of patient compliance , satisfaction and appropriate use of medical facilities” . [23] The present study , thus , also aimed to explore explanatory models and beliefs about and barriers to self-care of amongst podoconiosis patients .
Amhara is the second largest region in Ethiopia with a 2007 population of 17 , 214 , 056; of whom only 12 . 27% resided in urban areas . [24] The majority of inhabitants in the Gojjam area are agricultural workers , only 19 . 5% in West Gojjam and 11 . 4% in East Gojjam reported as being in non-farm related occupations . [24] The soil and climatic conditions that induce podoconiosis , in particular high altitude ( >1000m ) , seasonal rainfall ( over 1000mm annually ) and the presence of red-clay soil , [25] are all present in Amhara region , especially East and West Gojjam zones . Participants for the present study were enrolled from treatment sites in these zones at Dembecha , Chertekle , Fenote Selam , Debre Elias , Amanuel and Bure ( see Fig 1 for a map of the study area ) . A cross-sectional exploratory study was conducted between April-May 2015 at IOCC treatment centres across East and West Gojjam zones in North-West Ethiopia ( Fig 1 ) . Eleven IDIs , three FGDs and two key-informant interviews ( KIIs ) were conducted . Convenience sampling was used in order to recruit participants . Interviews were conducted in Amharic by a bi-lingual researcher experienced in conducting qualitative research . A semi-structured interview guide was developed to encourage podoconiosis patients to discuss their understanding[23] and recognition of their illness , [20] when they sought help and who this was from , any traditional treatments tried , their motivations for attending IOCC services were and beliefs about adherence to self-care regimes . Questions were designed to be open and the interviewer given freedom to explore respondents’ responses , to elucidate deeper understanding and to identify themes not anticipated by the research team . Questions were translated into Amharic , piloted among local staff and then modified to verify cultural meanings before use with participants . As particular themes emerged , interviews were modified to include questions to explore these themes more fully . FGD topics were created following themes that emerged from the IDIs . Current IOCC patients were interviewed on the day of their attendance at clinic . For past IOCC users and non-IOCC patient’s , a mutually convenient time was agreed on when to conduct the IDI . IDIs were conducted in a private setting for up to 58 minutes . FGDs also took place at treatment centres in a private setting . These took up to 2 hour 10 minutes and all included both male and female participants who were all affected by podoconiosis . Interviews were recorded , transcribed , translated and entered into Microsoft Word to facilitate data coding , text-searching and analysis . Data collection continued until no new themes emerged , i . e . until data saturation was reached . Following transcription , data were organised using manual coding techniques to categorise and generate themes . The method for generating codes involved using an integrated approach to developing code structure . This process involves both inductive development of codes as well as a deductive organising framework for code types . [26] Validity was promoted through having two independent coders organise the data . The coders resolved issues and incorporated new themes through regular meetings and discussion . Study participants were only recruited once they were fully aware of the purpose of the research and the methods of data collection . Since many participants were illiterate , individual consent was requested orally and signed by an Amharic-speaking witness , once the participant had read and/or discussed the participant information sheet . Participants were informed that they had a right to stop or skip questions during the interview at any time , and that if they changed their mind about being part of the study they could contact the IOCC and their data would be removed . Participants were given unique identification numbers to maintain confidentiality and audio recordings were destroyed following anonymized transcription . In the results below , quotations are identified as follows: [F2 , 2] indicates that the participant was labelled as participant 2 from FGD 2; [8] is participant 8 from the IDIs , [K1] was the first key-informant interview .
A total of 34 participants were interviewed , comprising the following: past and present IOCC patients ( 13 male , 12 female ) , non-IOCC patients ( 2 female ) , patient association leaders ( 4 male , 1 female ) and health workers ( 2 male ) . The age range of participants was 26–90 years . Most participants were subsistence farmers or daily labourers , rural dwellers , married and unable to read and/or write . See Table 1 for study population characteristics and assigned participant codes . Most patients had sought non-biomedical treatment for podoconiosis and these came in four forms: herbal , magical , folk remedies and faith-based healing . Herbal treatments were often prepared by diviner-wizards who guarded their secrets closely . Treatment most commonly involved tying a poultice to the leg , which “created a wound that weeps” [F3 , 5] . The herbs applied such treatment were: “Zigba” ( Podocorpus falcatus ) [27] , “Demakese” ( Ocimum lamiifolum ) [27] and an unidentified plant “Gishila” . Other herbal treatment involved boiling the leaves and inhaling the vapours to provoke diaphoresis , including: “Harage Resa” ( Zehneria scabra ) [27] and an unidentified herb named “Shingung” in Amharic . Some herbs were unavailable during the dry season , and participants reported concerns about addiction and withdrawal symptoms , especially with Shingung . In addition to advising on which herbs to use , certain diviner-wizards were also able to perform magical spells and rites in order to relieve some symptoms of podoconiosis . In some cases , the diviner-wizards would cast the herbs involved in a patient’s treatment into the road as part of a spell . Folk treatments involved tying fertilizer ( a mixture of human urine , soil and water ) to a patient’s leg and leaving them to heal . This treatment was very expensive ( 500-600birr/£15–18 ) and in some cases aggravated the patient’s symptoms . Other practices included cleansing the legs with lemon and salt water , tying the skin of a dead snake to the affected person’s legs or through lay surgical techniques such that “the swollen leg is penetrated by blade then the blood is collected by a cup . ” [F2 , 2] Faith-based healing involved washing or drinking holy water or applying holy soil into the legs . Although there are many holy water sites in Northern Ethiopia known for their curative benefits , water could be rendered holy if prayed over by a priest in church , or if the source was the “result of a miracle” [K1] . Every patient participant had visited holy water treatment sites and believed them to provide valuable treatment and psychological support with their illness: “There is holy water , and it relieves . I have visited many holy waters” [6] . Participants were very amenable to discussing traditional treatments . Perhaps unsurprisingly ( given that interviews were conducted in biomedical treatment centres ) the majority expressed negative views towards traditional treatments , with the exception of holy water . This is succinctly summarised by one participant: Podoconiosis patients tended to wait until their symptoms were severe before attending services . This appeared to be due to a belief that their symptoms would remit or that visiting a clinic would waste valuable working time for key activities , such as childcare: “Its only when I fall and am unable to feed my children that I come to the health center . Unless I am troubled by the disease , I don’t come to health centers . ” [F1 , 5] . Only a minority of participants suggested it was better to visit a clinic early in the stages of the disease . Decisions to attend were predominantly made by men , however , it was stated that women could lead on healthcare access “if the female is educated” [F2 , 5] . Traditional forms of treatment were highly valued in the community . Farmers and other members of the community “inform where the people who give the traditional treatment are found and encourage to go there” [8] . It was apparent that there is an inherent distrust of modern treatment services , with some participants expressing that such services were “too poisonous” [F3 , 5] , ineffective or not modern enough . A fundamental issue was that community members did not believe that the disease “can be treated simply by washing” [8] . In one case , a participant stated her fear of clinical errors arising from IOCC treatment because the “government [could] put me to death by putting the wrong ointment on my leg” [6] . A few participants expressed that their decision to come to the clinic was due to advice from IOCC workers . Generally , it was after the failure of one particular treatment ‘regime’ that participants sought different options . These tended to be other forms of traditional treatment , though when traditional options were exhausted , Western medical clinics were a last resort . Participants with no previous contact with IOCC services mentioned that they were not even aware of modern services and when advised about these services responded with: “Wow ! What a pleasant news it would be ! If there I will go” [10] . Self-care practices were attributed to valuing wellbeing and “keeping oneself tidy” [4] . Participants stated that wellness and general hygiene concepts were taught by the church , whereas specific podoconiosis-related self-care practices were only delivered by the IOCC clinics . Participants expressed that they only really learnt about the importance of washing their legs through observing “the difference that I feel when I wash and when I don’t wash” [4]; in some instances , this was ‘discovered’ before attendance at any health-care services . The most commonly cited reason for failing to adhere to self-care practices was “laziness” . This appeared to amount to exhaustion following long-working days: “When we arrive home from work , we feel tired and we prefer to directly go to bed [rather than wash]” [F3 , 7] . In addition , collecting water was extremely difficult for those suffering from podoconiosis due to extreme pain when walking . During the dry seasons in Ethiopia , streams and wells could dry up and many patients understandably “prefer to save the water for drinking than for washing” [F3 , 2] . Some were able to purchase water in local shops , though “the time we are living in is difficult” [11] due to this expense . For the lucky few with access to hand-pumps in their village , they tended to avoid public places due to fear of and past experience of stigma and abuse directed at them .
The results of this qualitative study indicate the importance of explanatory models , pluralistic health-seeking behaviour and self-care practice in treatment access and care for podoconiosis . It also highlights the interlinked nature of these factors and the need for culturally relevant strategies to improve awareness of , and engagement with , treatment services . This calls for greater understanding and co-operation with priests and traditional healers , including exploration of potential roles in signposting to treatment centres . As reported elsewhere , a further way to engage with lay health beliefs , tackle community stigma and promote access to IOCC services could be through “expert patients” . [15] This study underlines significant challenges to self-care , especially in collecting adequate water for washing . Expanding access to clean water is of utmost importance for the effective treatment of affected persons . Collaboration with OWNP would help water to be available in convenient locations , free of charge and to be collected in such a way as to not promote stigma . | Podoconiosis is a disease that is caused by long-term exposure to red clay soil found in tropical highland areas . This causes swelling of the legs , episodes of intense pain and severe disability . Sufferers of the condition experience stigma and exclusion from the community . Previous work has demonstrated a low rate of re-attendance to clinics and that many patients fail to adhere to their treatment regime of improving foot hygiene . The present study explored areas of health-seeking behaviour and self-care practices to discover the reason behind these failures . It was found that explanatory models of disease causation had a significant impact on decisions to seek healthcare and that participants only turned to Western medical clinics after failure of traditional treatments and with particularly severe symptoms . We identified several form of traditional treatment for the disease , and these tended to be based on cleansing , fluid extraction or faith/symbolism . The most important barrier to self-care and adhering to treatment regimens was an inability to collect adequate water . We call for greater integration with traditional healers , improved access to water through collaboration with other NGOs and the government and the use of expert patients to disseminate information and signpost patients to clinics . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"legs",
"health",
"services",
"research",
"behavioral",
"and",
"social",
"aspects",
"of",
"health",
"tropical",
"diseases",
"limbs",
"(anatomy)",
"parasitic",
"diseases",
"herbs",
"health",
"care",
"farms",
"neglected",
"tropical",
"diseases",
"patients",
"plants",
"public",
"and",
"occupational",
"health",
"musculoskeletal",
"system",
"elephantiasis",
"podoconiosis",
"behavior",
"agriculture",
"anatomy",
"biology",
"and",
"life",
"sciences",
"organisms"
] | 2016 | Using Qualitative Methods to Explore Lay Explanatory Models, Health-Seeking Behaviours and Self-Care Practices of Podoconiosis Patients in North-West Ethiopia |
Erythema Nodosum Leprosum ( ENL ) is a humoral immunological response in leprosy that leads to inflammatory skin nodules which may result in nerve and organ damage , and may occur years after antibiotic treatment . Multiple episodes are frequent and suppression requires high doses of immunosuppressive drugs . Global occurrence is unknown . Systematic review of evidence on ENL incidence resulted in 65 papers , predominantly from India ( 24 ) and Brazil ( 9 ) , and inclusive of four reviews . Average incidences are based on cumulative incidence and size of study populations ( n>100 ) . In field-based studies 653/54 , 737 ( 1 . 2% ) of all leprosy cases , 194/4 , 279 ( 4 . 5% ) of MB cases , and 86/560 ( 15 . 4% ) of LL cases develop ENL . Some studies found a range of 1–8 per 100 person-years-at-risk ( PYAR ) amongst MB cases . Hospital samples indicate that 2 , 393/17 , 513 ( 13 . 7% ) of MB cases develop ENL . Regional differences could not be confirmed . Multiple ENL episodes occurred in 39 to 77% of ENL patients , with an average of 2 . 6 . Some studies find a peak in ENL incidence in the first year of treatment , others during the second and third year after starting MDT . The main risk factor for ENL is a high bacteriological index . Few studies reported on ENL as a primary outcome , and definitions of ENL differed between studies . Although , in this review averages are presented , accurate data on global and regional ENL incidence is lacking . Large prospective studies or accurate surveillance data would be required to clarify this . Health staff needs to be aware of late reactions , as new ENL may develop as late as five years after MDT completion , and reoccurrences up to 8 years afterwards .
Erythema Nodosum Leprosum ( ENL ) , the main symptom of a type-2 reaction in leprosy , is caused by a humoral immune response to Mycobacterium Leprae [1] . Patients develop fever and tender/painful subcutaneous nodules , often in the face or extensor surfaces of the limbs [2]–[4] . ENL may also damage nerves , skin , eyes , and testes , and involves systemic illness including fever , weight loss and pain [5] , all of which result in extreme discomfort . The majority of patients develop multiple episodes of ENL . Severe cases require the use of potent immunosuppressants , and the steroid-induced side effects may increase mortality and morbidity [3] , [6] . Furthermore , the limited use of teratogenic thalidomide presents another challenge [5] . The economic impact of ENL is unknown , but likely to be considerable . ENL is confined to leprosy patients classified as BL or LL ( Ridley-Jopling ) , comprising the multi-bacillary ( MB ) patient group , as defined by WHO . In 1981 this concerned patients with a bacteriological index ( BI ) of 2 or more , changing to any positive skin smear in 1988 . In 1995 this was widened further; MB comprising any patients with more than five skin lesions [7] . The proportion of MB cases among new leprosy patients varies between countries and is increasing [8] , [9] . Global incidence of MB leprosy was 139 , 125 in 2009 , and is decreasing [8] . ENL may occur before , during or after antibiotic treatment , and several years later [10] . It can occur as a single acute episode , but frequently develops into a chronic condition with recurrent episodes [3] , [5] . Immune responses causing ENL are triggered by high loads of fragmented bacilli in skin tissue [11] . Although adequate surveillance systems are used to estimate global leprosy prevalence and inform drug supply , this is not available for estimating incidence , frequency and severity of ENL [12] . Geographic variation in ENL prevalence complicates accurate estimations [13] , and hampers logistics in drug supplies . For this reason , a systematic literature review was conducted to determine global incidence of ENL , inclusive of incidences of recurrent and severe ENL and contributing factors .
A systematic literature search was conducted in January 2011 in five databases ( Pubmed ( MEDLINE ) , EMBASE , LILACS , SCOPUS , Scielo , and Ajol ) . Keywords used were: <lepro* OR lepra* OR hansen* , Erythema Nodosum OR ENL OR ( type 2 ) > , AND <incidence OR prevalence OR cohort> . Reference lists of included studies were checked and national leprosy control managers and leading leprologists were asked for additional ( un- ) published articles . Studies , published after 1980 , presenting data on incidence or prevalence of ENL were selected . Focus was on papers in English , whereas Portuguese , Spanish or French studies were included after Google-translation . No separate search was conducted on adverse events and risk factors , but information was retrieved from the included studies . A distinction was made between acute and chronic ENL as well as severe and mild forms [2] . We included all studies reporting on the onset of ENL . The following forms of ENL were included: single acute episodes , multiple acute episodes , and chronic ENL ( ENL lasting for more than 6 months , in either single or multiple episodes ) [2] . Data extraction regarding onset , risk factors , severity and reoccurrence of ENL was completed by the first author and co-reviewed by the second author . A structured form was designed to retrieve data on the setting ( country , region , place studied , other characteristics ) , methods ( study period , design , sampling , data sources , representativeness ) , study design and characteristics ( sample size , population , leprosy classification ( Ridley-Jopling ) , inclusion criteria , ethnicity , gender , age group , other ( health ) characteristics and study variables ( follow-up time , loss to follow up , and MDT- , ENL- , or other treatment , serious adverse events ) . Evidence was graded according to the Oxford Centre for Evidence Based Medicine guidelines [14] . Depending on availability , incidence rates of ENL are presented in person years at risk ( PYAR ) . Where proportions or actual numbers of patients developing ENL were reported , ENL incidence is based on the proportion of persons at risk ( i . e . total number of leprosy cases , MB cases or specific Ridley-Jopling classifications ) . We considered cases MB as reported in the articles . Occurrence is only presented when sample sizes exceeded 100 at risk ( MB ) population , for field and hospital studies separately . The average incidence of ENL was calculated taking all different sample sizes together .
The search resulted in 914 records ( Figure 1 ) . Scanning the references and consultation with experts resulted in an additional 10 papers . 65 papers met the inclusion criteria . Four literature reviews were analysed separately [2] , [12] , [15] , [16] . One relevant workshop report was included [17] . The majority of studies were from India ( 24 ) and Brazil ( 9 ) , the two countries with the highest incidence of new leprosy cases [8] . Table 1 summarises the characteristics of included studies . Approximately one third of the studies included a minimum of 300 persons at risk for ENL and another third between 100 and 300 persons . 23 studies had sample sizes below 100 persons at risk [10] , [18]–[39] . Studies were either cross-sectional or retrospective cohort analyses . Less than half of them aimed specifically at ENL occurrence . The majority reported ENL frequency while their main focus was on clinical or epidemiological aspects of leprosy . Only five studies reported ENL incidence rates in person years at risk ( PYAR ) . Follow up varied between 2 and 7 years . Incidence rates ranged from 1 to 8 per 100 PYAR [40] , [41] among MB leprosy patients ( figure 2 ) . Six prospective [30] , [41]–[45] and five retrospective studies [17] , [40] , [46]–[48] gathered data from a control programme and most accurately reflected ENL occurrence . Table 2 demonstrates that cumulative ENL incidence varied from 0 . 2% among all leprosy patients in an Indian study [49] and up to 4 . 6% in a Chinese study [48] , with an average of 1 . 2% . ENL incidence among MB cases varied from 1 . 0% in a one year cross-sectional Indonesian study [46] to 8 . 9% in an Indian cohort [47] , with an average of 4 . 5% . From the latter study , it was not clear if referral cases were included , which may explain the relatively high percentage . Three prospective studies were from the ALERT leprosy control services [41] , [42] , [45] . Interestingly , cumulative ENL incidence was 2 . 5% among MB cases after an average follow-up of 2 . 5 years [45] , whereas after 10 years this was doubled [41] . Table 3 indicates the cumulative ENL incidence in 28 studies ( >300 patients ) , ranging from 2–28 . 9% of MB cases . Calculation from studies with at least 100 patients reveals that on average 13 . 7% of MB cases developed ENL . In four studies this was more than 30% [50]–[53] . Studies with largest population sizes indicated lower cumulative incidence rates . Sixteen studies reported ENL occurrence for the Ridley-Jopling classifications ( Figure 3 ) . Findings differed widely between countries . Among the four field studies [41] , [42] , [44] , [47] ENL for LL leprosy ranged from 11 . 1% [42] to 26% [44] with an average of 86/560 ( 15 , 4% ) . For BL cases this varied from 2 . 7% [42] to 5 . 1% [47] , on average 51/1231 ( 4 , 1% ) . In hospital based studies higher proportions were found , in Brazil up to 56 . 4% [52] and in India a range of 24 . 2 [54] to 50 . 9% [55] . ENL reoccurrence was disproportionately higher in hospital-based studies . Multiple episodes were found in 39% [56] to 77 . 3% [50] of ENL patients , with an average of 2 . 6 episodes . Various studies reported 24% of all ENL cases having more than four episodes: the longer the follow-up the more episodes were recorded . Three larger studies ( >100 ENL cases , see Table 4 ) found a range from 49% [57] to 64 . 3% [58] . Similar ranges were found in field based studies: 44 to 63% of all ENL cases have multiple ENL episodes [41] , [44] , [45] . There was discrepancy in the average number of ENL episodes , as is evident in the following findings . In a cohort from Zaire [59] there was an average of 1 . 8 episodes , compared to 3 . 2 episodes ( CI 2 . 7–3 . 5 ) , in a study from India [60] . A Thai cohort revealed that ENL episodes often occurred more than 4 times [50] . A large hospital study in India reported that 23 . 5% of reoccurring cases ( 15 . 1% of all ENL cases ) had four or more episodes [58] . Similar proportions were found in a Brazilian cohort [52] , whereas other studies in India [47] and Nepal [57] found four or more episodes among 5 and 7% of ENL patients respectively . In Ethiopia , almost one third of ENL patients developed a chronic condition lasting more than 2 years [41] . Episodes lasted from 14 days [19] to 26 . 1 weeks [61] . Total ENL episodes and ENL-free intervals in India found an average of 18 . 5 months ( CI 15 . 4–21 . 5 ) [60] . Six studies distinguished between mild and severe ENL , finding that 30–50% of ENL cases are ( moderate to ) severe . They represented 0 . 7–2 . 0% of all MB leprosy patients and 0 . 7% of all newly detected cases [46] , [62] . However , descriptions of severity differed between the studies . Shortened MDT duration ( 12 months ) almost doubled the incidence of moderate to severe ENL [61] , [63] . Poor referral practices leave some severe reactions under-diagnosed [40] , while hospital figures misrepresent the field situation [47] . Findings on the onset of ENL differ . Most studies indicated that the incidence of ENL during MDT was at least twice as high than at the time of the initial diagnosis [37] , [42] , [44] , [50] , [64] , [65] . ENL incidence was highest in the first year of MDT [17] , [37] , [42] , [44] , [57] , [58] , [64] . There were a few exceptions , a from the Philippines ( 10 year follow-up ) [43] , [61] and India ( 13 years follow-up ) [58] where most ENL was diagnosed during the second and third year after starting MDT , as was the case in Ethiopia [41] . A study conducted in an Indian hospital found 3% of MB patients developed ENL two years after completing MDT ( follow-up 74 months ) [58] . Longer term follow up showed ENL three [37] , five [66] , seven [41] , or even eight years after MDT [58] . Similar findings ( ENL occurring 5–7 years later ) were reported in India [17] . Multiple studies [22] , [23] , [52] , [57] , [58] , [60] , [62] reported a correlation between the bacteriological index ( BI ) and ENL up to a 8 . 6 ( CI 2 . 3–32 ) times higher risk when having a BI of six [41] . Discrepancies are evident Nepali patients with a BI>4+ had a 39% higher risk of ENL ( OR; 1 . 39 ( CI 1 . 11–1 . 76 ) adjusted for age ) [57] and in India a BI≥4 was associated with an Odds Ratio of 5 . 2 ( 2 . 1–12 . 9 ) [60] . Inherent to BI , lepromatous leprosy is a significant risk factor [58] , [67] . An Ethiopian study found a 9 . 6 times higher ENL incidence among LL patients compared to BL or BB ( X2 = 18 . 7 , p<0 . 005 ) [42] . Odds ratios for the prevalence of ENL in LL as compared to BL varied from 2 . 8 ( 1 . 59–5 . 2; adjusted for age and BI ) [57] to 8 . 4 ( CI 4 . 6–15 . 4 ) [60] . LL cases have higher chances to suffer multiple rather than single ENL episodes ( OR 2 . 94 , p = 0 . 052 ) [57] . This finding was disputed , however , by a controlled clinical trial conducted in India , which reported no such differences [55] . It has been claimed that the risk of developing ENL has decreased since introducing MDT [42] , [51] , [54] , [57] , due to the ENL suppressant effect of clofazimine [22] , [51] , [68] . A recent multi-country cohort study indicated more severe and longer-lasting episodes of ENL among patients who received 12 as compared to 24 months of MDT , although ENL frequency as such was similar [61] , [63] . The Bombay Leprosy Project had similar findings: 55 . 9% and 35 . 8% of cases receiving 12 and 24 months MDT respectively had a type1 or 2 reaction [17] . Gender is generally not a risk factor for ENL [41] , [52] , [55] , [57] , [60] , [62] , [63] . Some studies appear to challenge this , as a large hospital study in India found a male predominance [69] , and a large Indian cohort reported a higher risk for women [58] . These differences , however , may be due to differences in health seeking behaviour [69] . Seemingly , age is not a risk factor for ENL [41] , [50] , [58] , [60] , [63] , although a Nepali cohort indicated decreased risk for those older than 40 ( adjusted OR 0 . 69 , CI 0 . 5–0 . 94 ) [57] , and a higher ENL incidence was seen in patients diagnosed with leprosy in their adolescence , but these findings are not supported elsewhere [50] . Pregnancy and lactation appears to be a significant precipitating factor for severe and recurrent ENL [54] . Additionally , hormonal changes are implicated in a study from India: 62% of 32 ENL in women were associated with pregnancy or lactation and 21% with menopause [69] . A major Ethiopian study among pregnant leprosy patients found an increased ENL incidence ( 22% among BL and 59% among LL patients ) . Some episodes continued until 15 months after delivery [24] . Minimal evidence has been published regarding co-morbidities as risk factors for ENL , with the exception of HIV that suggested a 5 . 3 times higher risk for developing ENL ( RR 5 . 3 , CI 1 . 0–2 . 8 ) . However , numbers ( n = 10 ) were too low to be conclusive [41] . A recent review concluded there is no reliable data on the effect of HIV [13] . In other studies , malaria and tuberculosis were reported to trigger ENL [24] , [54] .
Most of the literature regarding ENL occurrence was descriptive data , and only a few studies had an adequate sample of patients . Characteristics of cases and populations , definitions , outcomes and procedures were not always systematically described , making a statistical meta-analysis impossible . To what extent study samples reflected the leprosy population at large was often difficult to assess , as distinction between field and hospital based studies was not clear in each publication . Higher ENL rates were found in hospital based studies , although it is not known how many severe ENL cases actually arrive in referral clinics . In the hospital based studies the population size of which these cases are drawn is not known . Field based studies often only report patients with ENL who actually seek help . Only few appropriate prospective studies could be found that are representative for the most peripheral level . The majority of publications lacked both a clear case definition of ENL and a clear description of the diagnostic procedure . Both may vary between settings and studies . Only a few studies make a distinction between mild and severe ENL [60]–[63] , and mild ENL may have been overlooked and thus incidence rates underestimated . Considering the limited evidence and the significant differences in ENL rates , country specific data should be interpreted with great caution . The wide range in cumulative incidence and variation of ENL found in this review is most likely explained in terms of duration of treatment and follow-up of the subjects . Furthermore , the widening definition of MB leprosy since 1981 [4] , [72] would have decreased rates of ENL . LL patients would be the most appropriate risk group for ENL to report on , especially in research papers . In this study , however , MB was the most common denominator in the articles that were identified . Ideally , future studies on ENL should report incidence in person years at risk , both for MB and Ridley-Jopling classification . None of the studies included in this review looked at explicitly at the social and medical costs related to ENL . This review provides a systematic overview of available evidence regarding ENL occurrence . Wide ranges were found between and within different countries . Despite these limitations , a global average incidence was calculated . This review has established that reliable data on ENL occurrence is lacking , and could only be obtained through large comprehensive prospective studies or data obtained from accurate ENL surveillance . Furthermore , studies investigating risk and precipitating factors for ENL would be useful in diagnosis and prevention . | This systematic review addresses an underpublicized and yet highly significant leprosy topic . Erythema Nodosum Leprosum ( ENL ) is a serious complication in multi-bacillary ( MB ) leprosy that may lead to severe disability . Inflammatory skin nodules may result in nerve and organ damage and require high doses of immunosuppressive drugs . ENL can occur long after patients are released from antibiotic treatment . Frequency and severity of ENL is unknown; this review confirms the lack of accurate data at global , regional , and national levels . Available data indicates that ENL incidence ranges between 0 . 7–4 . 6% of all MB cases and late reoccurrence up to 8 years after release from treatment . ENL episodes often reoccur , with an average of 2 . 6 times . The main risk factor for ENL is a high bacteriological index . Additionally , data indicate a wide variation of ENL occurrence between and within countries . The conclusions demonstrate the need for increased awareness about ENL , in research , patient surveillance , and in programme management . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | A Systematic Review on the Epidemiological Data of Erythema Nodosum Leprosum, a Type 2 Leprosy Reaction |
Viruses have exerted a constant and potent selective pressure on human genes throughout evolution . We utilized the marks left by selection on allele frequency to identify viral infection-associated allelic variants . Virus diversity ( the number of different viruses in a geographic region ) was used to measure virus-driven selective pressure . Results showed an excess of variants correlated with virus diversity in genes involved in immune response and in the biosynthesis of glycan structures functioning as viral receptors; a significantly higher than expected number of variants was also seen in genes encoding proteins that directly interact with viral components . Genome-wide analyses identified 441 variants significantly associated with virus-diversity; these are more frequently located within gene regions than expected , and they map to 139 human genes . Analysis of functional relationships among genes subjected to virus-driven selective pressure identified a complex network enriched in viral products-interacting proteins . The novel approach to the study of infectious disease epidemiology presented herein may represent an alternative to classic genome-wide association studies and provides a large set of candidate susceptibility variants for viral infections .
Infectious diseases represent one of the major threats to human populations , are still the first cause of death in developing countries [1] , and are therefore a powerful selective force . In particular , viruses have affected humans before they emerged as a species , as testified by the fact that roughly 8% of the human genome is represented by recognizable endogenous retroviruses [2] which represent the fossil remnants of past infections . Also , viruses have probably acted as a formidable challenge to our immune system due to their fast evolutionary rates [3] . Indeed , higher eukaryotes have evolved mechanisms to sense and oppose viral infections; the recent identification of the antiviral activity of particular proteins such as APOBEC , tetherin , and TRIM5 has shed light on some of these mechanisms . Genes involved in anti-viral response have therefore been presumably subjected to an enormous , continuous selective pressure . Despite the relevance of viral infection for human health , only few genome-wide association studies ( GWAS ) have been performed in the attempt to identify variants associated with increased susceptibility to infection or faster disease progression [4]–[5] . These studies have shown the presence of a small number of variants , mostly located in the HLA region . This possibly reflects the low power of GWAS to identify variants with a small effect . An alternative approach to discover variants that modulate susceptibility to viral infection is based on the identification of SNPs subjected to virus-driven selective pressure . Indeed , even a small fitness advantage can , on an evolutionary timescale , leave a signature on the allele frequency spectrum and allow identification of candidate polymorphisms . To this aim we exploited the availability of more than 660 , 000 SNPs genotyped in 52 human populations distributed world-wide ( HGDP-CEPH panel ) [6] and of epidemiological data stored in the Gideon database .
Previous studies [7]–[9] have suggested that the number of the different pathogen species transmitted in a given geographic location is a good estimate of pathogen-driven selection for populations living in that area . Indeed , pathogen diversity is largely dependent on climatic factors [10] and might more closely reflect historical pressures than other estimates such as the prevalence of specific infections . We therefore reasoned that virus diversity can be used as a measure of the selective pressure exerted by virus-borne diseases on human populations and , as a consequence , that SNPs showing an unusually strong correlation with virus diversity can be considered genetic modulators of infection susceptibility or progression . To explore this possibility we used a large set of SNPs that have been genotyped in the HGDP-CEPH panel , a collection of DNAs from almost 950 individuals sampled throughout the world ( Table 1 ) . Virus diversity estimates were derived from the Global Infectious Disease and Epidemiology Network database: for each country where HGDP-CEPH populations are located we counted the number of different virus species ( or genera/family as described in materials and methods ) that are naturally transmitted ( Table 1 ) . One simple prediction of our hypothesis whereby virus diversity is a reliable estimator of virus-driven selective pressure is that genes known to be involved in immune response are enriched in SNPs significantly associated with virus richness . In order to verify whether this is the case we analysed the InnateDB gene list which contains 2 , 915 genes involved in immune response and showing the presence of at least one SNP in the HGDP-CEPH panel . Correlations with virus richness were calculated using Kendall's partial rank correlation; since allele frequency spectra in human populations are known to be affected by demographic factors in addition to selective forces [11]–[12] , each SNP was assigned a percentile rank in the distribution of τ values calculated for all SNPs having a minor allele frequency ( MAF ) similar ( in the 1% range ) to that of the SNP being analysed . A SNP was considered to be significantly associated with virus diversity if it displayed a significant correlation ( after Bonferroni correction with α = 0 . 01 ) and a rank higher than 0 . 99 . As shown in Table 2 , 104 SNPs in InnateDB genes showed a significant association with virus diversity . All SNPs in InnateDB genes that correlated with virus diversity are listed in Table S1 . By performing 10 , 000 re-samplings of 2 , 915 randomly selected human genes ( see materials and methods for details ) we verified that the empirical probability of obtaining 104 significantly associated SNPs amounts to 0 . 010 , indicating that genes in the InnateDB list display more virus-associated SNPs than expected . It is worth mentioning that amongst these genes , UNG ( MIM 191525 ) , encoding uracil DNA glycosylase , functions downstream of APOBEC3G ( MIM 607113 ) to mediate the degradation of nascent HIV-1 DNA [13] . SERPING1 ( MIM 606860 ) , a regulator of the complement cascade , is also involved in HIV-1 infection ( MIM 609423 ) as its expression is dysregulated in immature dendritic cells by Tat [14]; moreover , the protein product of SERPING1 is cleaved by HCV and HIV-1 proteases [15]–[16] . Genes involved in the biosynthesis of glycan structures have also been considered as possible modulators of infection susceptibility . Indeed , since Haldane's prediction in 1949 [17] that antigens constituted of protein-carbohydrates molecules modulate the resistance/susceptibility to pathogen infection , protein glycolsylation has been shown to play a pivotal role in viral recognition of host targets [18] , as well as in antigen uptake and processing and in immune modulation [19]–[20] . We therefore computed a list of genes involved in glycan biosynthesis from KEGG pathways and Gene Ontology annotations . Again these genes displayed significantly more virus-associated SNPs than expected if randomness alone were responsible ( empirical p = 0 . 0138 ) ( Table 2 and Table S2 ) . Several virus-associated SNPs were located in genes coding for sialyltransferases ( ST6GAL1 ( MIM 109675 ) , ST3GAL3 ( MIM 606494 ) , ST6GALNAC3 ( MIM 610133 ) , ST8SIA1 ( MIM 601123 ) , ST3GAL1 ( MIM 607187 ) and ST8SIA6 ( MIM 610139 ) ) . Notably , sialic acids represent the most prevalent terminal monosaccharides on the surface of human cells and determine the host range of different viruses including influenza A [21]–[22] , polyomaviruses ( i . e JCV and BKV in humans ) [23] , and rotaviruses ( the leading cause of childhood diarrhea ) [24] . Sialyltransferases also play central roles in B and T cell communication and function . In particular , the generation of influenza-specific humoral responses is impaired in mice lacking ST6GAL1 [25] , while ST3GAL1 regulates apoptosis of CD8+ T cells [20] . Interestingly , ST8SIA6 is expressed in NK cells , possibly playing a role in the regulation of Siglec-7 lectin inhibitory function in these cells [26] . Four other genes ( XYLT1 ( MIM 608124 ) , HS3ST3A1 ( MIM 604057 ) , UST ( MIM 610752 ) and CHSY3 ( MIM 609963 ) ) carrying SNPs associated with virus diversity are involved in the biosynthesis of either heparan sulphate or chondroitin sulphate . The former is an ubiquitously expressed glycosaminoglycan serving as the cell entry route for herpesviruses [27] , HTLV-1 [28] and papillomaviruses [29] . Chondroitin sulphate is similarly expressed on a wide array of cell types and functions as an auxiliary receptor for binding of herpes simplex virus [30] as well as a facilitator of HIV-1 entry into brain microvascular endothelial cells [31] . Finally , we identified LARGE ( MIM 603590 ) among the genes subjected to virus-driven selective pressure ( Table 2 ) . Recent studies have demonstrated that the post-translational modification of α-dystroglycan by LARGE is critical for the binding of arenaviruses of different phylogenetic origin including Lassa fever virus and lymphocytic-choriomeningitis virus [32]–[33] . Therefore our data support the previously proposed hypothesis whereby viruses represent the selective pressure underlying the strong signal of positive selection at the LARGE locus [34] . Since genes involved in immune response and in the biosynthesis of glycan structures are likely to be subjected to selective pressures exerted by pathogens other than viruses , we verified whether a set of genes directly involved in interaction with viral proteins also displays more SNPs significantly correlated with virus diversity . To this aim we retrieved a list of 1 , 916 genes known to interact with at least one viral product and displaying at least one genotyped SNP in the HGDP-CEPH panel ( see materials and methods ) . In order to perform a non-redundant analysis , genes included in the InnateDB list and involved in glycan biosynthesis were removed; the remaining 987 genes displayed 80 SNPs correlated with virus diversity , corresponding to an empirical p value of 0 . 017 ( Table 2 and Table S3 ) . Notably , when this same analysis was performed using the diversity of pathogens other than viruses ( bacteria , protozoa and helminths ) , no significant excess of correlated SNPs was found ( all empirical p values>0 . 05 ) . Given these results , we wished to identify SNPs significantly associated with virus richness on a genome-wide base . We therefore calculated Kendall's rank correlations between allele frequency and virus diversity for all the SNPs ( n = 660 , 832 ) typed in the HGDP-CEPH panel . We next searched for instances which withstood Bonferroni correction ( with α = 0 . 05 ) and displayed a τ percentile rank higher than the 99th among MAF-matched SNPs . A total of 441 SNPs mapping to 139 distinct genes satisfied both requirements . Table 3 shows the 30 top SNPs ( or SNP clusters ) located within genic regions and associated with virus diversity , while the full list of SNPs subjected to virus-driven selective pressure is available on Table S4 . It is worth noting that the SNP dataset we used contains less than 200 variants mapping to HLA genes ( both class I and II ) , therefore covering a minor fraction of genetic variability at these loci; as a consequence HLA genes cannot be expected to be identified as targets of virus-driven selective pressure using the approach we describe herein . We next verified whether the correlations detected between the SNPs we identified and virus diversity could be secondary to climatic variables . Hence , for all countries where HGDP-CEPH populations are located we obtained ( see materials and methods ) the following parameters: average annual minimum and maximum temperature , and short wave ( UV ) radiation flux . Results showed that none of the SNPs associated with virus diversity significantly correlated with any of these variables ( Table S5 ) . Previous works have reported an enrichment of selection signatures within or in close proximity to human genes [12] , [35] . In line with these data we verified that virus-associated SNPs are more frequently located within gene regions compared to a control set of MAF-matched variants ( χ2 test , p = 0 . 026 ) . We investigated the role and functional relationship among genes subjected to virus-driven selective pressure using the Ingenuity Pathway Analysis ( IPA , Ingenuity Systems ) and the PANTHER classification system [36]–[37] . Unsupervised IPA analysis retrieved two networks with significant scores ( p = 10−17 and p = 10−12 ) which were merged into a single interaction network ( Figure 1 ) . The network contains 23 genes showing a significant correlation with virus diversity and , among these , 10 encode proteins interacting with viral products ( Figure 1 ) . Based on the number of observed human-virus interactions , this finding is unlikely to occur by chance ( χ2 test , p = 0 . 0013 ) as 2 . 88 human-virus interactions would be expected for 23 genes . Analysis of the whole network indicated that a 31 of 66 genes encode proteins interacting with viral products ( Figure 1 ) : again this number is higher than expected ( expected interactions = 8 . 27; χ2 test , p = 2 . 8×10−10 ) . Thus , the interaction network we have identified is enriched in genes subjected to virus-driven selective pressure and in genes coding for proteins interacting with viral products . It is worth mentioning that , in agreement with previous findings [38] , many viral-interacting proteins represent hubs in the network . Conversely , most of the genes we found to be subjected to virus-driven selective pressure , irrespective of their ability to interact with viral proteins , tend to display very low connectivity ( low-degree nodes ) . This observation might be consistent with previous indications [39]–[41] that in eukaryotes hub genes are more selectively constrained compared to low-degree nodes , these latter being more likely to evolve in response to environmental pressures . In addition to proteins directly interacting with viral products , several network genes showing correlation with virus diversity might play central roles during viral infection . DNMT1 ( MIM 126375 ) and MGMT ( MIM 156569 ) are involved in DNA methylation and repair , respectively , two processes that are often dysregulated during viral infection . In particular , altered expression of DNMT1 is induced by diverse viruses including HIV-1 [42] , EBV [43] , BKV and adenovirsuses [44]; also , DNMT1 plays a pivotal role in the expansion of effector CD8+ T cell following viral infection [45] . A relevant role in HIV-1 infection is also played by HSPG2 ( MIM 142461 ) , the gene coding for perlecan , a cell surface heparan sulfate proteoglycan which mediates the internalization of Tat protein [46] . We next investigated the over-representation of PANTHER classification categories among genes subjected to virus-driven selective pressure . Table 4 shows the significantly over-represented PANTHER molecular functions and biological processes with the contributing genes . In line with the results we reported above , genes involved in immune response , as well as genes coding for proteins involved in cell adhesion and extracellular matrix components , resulted to be over-represented; these latter genes might mediate viral-cellular interaction and facilitate viral entry .
The identification of non-neutrally evolving loci with a role in immunity can be regarded as a strategy complementary to classic clinical and epidemiological studies in providing insight into the mechanisms of host defense [47] . Here we propose that susceptibility genes for viral infections can be identified by searching for SNPs that display a strong correlation with the diversity of virus species/genera transmitted in different geographic areas . Similar approaches have previously been applied to study the adaptation to climate for genes involved in metabolism and sodium handling [48]–[50] . These analyses , including the one we describe herein , rely on similar assumptions and imply some caveats . First , we implicitly considered virus diversity , as we measure it nowadays , a good proxy for long-term selective pressure . This clearly represents an oversimplification , as new viral pathogens have recently emerged and the virulence of different viral species or genera might have changed over time . Still , previous studies have indicated that the geographic distribution of virus diversity is strongly influenced by climatic variables such as temperature and precipitation rates [10] , suggesting that , despite significant changes in prevalence and virulence , virus diversity might have remained relatively constant across different geographic areas , possibly representing the best possible estimate of long-standing pressure . In line with these considerations , we calculated virus diversity as the number of all viral species ( or genera/families ) that can cause a disease in humans , irrespective of virulence or pathogenicity ( Table S6 ) . The second issue relevant to the data we present herein is that environmental variables tend to co-vary across geographic regions: the distribution of different pathogens ( e . g . parasitic worms and viruses/bacteria/protozoa ) is correlated across HGDP-CEPH populations [9] and , as reported above , virus diversity is influenced by climatic factors . Therefore , our genome-wide search was preceded by analyses aimed at verifying whether virus diversity is a reliable and specific estimator of virus-driven selective pressure . In particular , we verified that genes involved in immune response and in the biosynthesis of glycans display significantly more variants associated with virus diversity than randomly selected human genes; this finding supports the idea that pathogens rather than climate or demography has driven the genetic variability at these loci . Notably , we also analysed genes that encode proteins interacting with viral components: since loci involved in immune response and in glycan biosynthesis were removed from this list , the remaining genes are expected to be specific targets of viral-driven selective pressure; consistently , we verified that a significant excess of SNPs correlating with virus diversity map to these loci . Conversely , a SNP excess was not noticed when the diversity of other human pathogens was used for the analysis , suggesting that , despite the correlation among different pathogen species across geographic locations [9] , the selective pressure imposed by viruses can be distinguished from that exerted by other organisms . As a further control for the possible confounding effects of other environmental factors , we verified that the variants we identified at the genome-wide level do not correlate with climate ( temperature ) and UV radiation . This analysis was motivated by the known association of virus diversity and biodiversity in general , with temperature [10] , [51] and by the fact that both climate and UV exposure have long been considered among the strongest selective pressures in humans [52] . Since none of the SNPs we identified correlated with either short wave radiation flux or temperature , we consider that their geographic distribution is likely to have been shaped by virus-driven selective pressure . In this respect it is worth mentioning that UV irradiation has been shown to be immunosuppressive in mice ( reviewed in [53]–[54] ) , but the effect of sun exposure on immune functions in humans is still poorly understood . Yet , herpes viruses ( both simplex and zoster ) and some papillomavirus types have been shown to be reactivated by UV exposure , suggesting that the link between short wave radiation flux and virus-driven selective pressure might be more complex than simply predicted on the basis of geographic variation . Our genome wide search for genes subjected to virus-driven selection allowed the identification of a gene interaction network that is enriched in both genes associated with virus diversity and in genes encoding proteins that interact with viral products . Many of the genes included in the identified network are of great interest as they are known to be involved in the activation of mechanisms that have direct or indirect protective effects against viruses . Thus , beside the well known activities of IL1A ( MIM 147760 ) and B ( MIM 147720 ) , IL4 ( MIM 147780 ) , TGFB1 ( MIM 190180 ) , IL16 ( MIM 603035 ) , IFNG ( MIM 147570 ) and TNF ( MIM 191160 ) , OAS2 ( MIM 603350 ) encodes a protein that activates latent RNases , resulting in the degradation of viral RNA and in the inhibition of viral replication [55] . CCL17 ( MIM 601520 ) induces T lymphocytes chemotaxis , thus potentiating the immune responses , and PPP3CA ( MIM 114105 ) , also known as calcineurin , activates NFATc [56] , a key factor in the up-regulation of IL2 ( MIM 147680 ) [57] , the main cytokine responsible for T lymphocytes growth and differentiation . Finally , ULBP2 ( MIM 605698 ) encodes an MHC1-related protein that binds to NKG2D ( MIM 602893 ) [58] , an activating receptor expressed on CD8 T cells as well as on NK cells , NKT cells and γδ T cells . In the light of the viral pathogenesis of a growing number of neoplasia , it is very interesting that other members of the network play a well described role in the inhibition of tumoral growth . In particular , E2F1 ( MIM 189971 ) is known to have a pivotal role in the control of cell cycle and in the activation of tumour suppressor proteins and , together with TP53I3 , TADA3L , and TP53BP2 mediates p53-dependent and independent apoptosis [59]–[60] . CCND3 ( MIM 123834 ) is involved in cell cycle progression through the G2 phase , whereas RAD23A ( MIM 600061 ) up-regulates the nucleotide excision activity of 3-methyladenine-DNA glycosylase [61] , therefore playing a role in DNA damage recognition in base excision repair . Finally , NR4A2 ( MIM 601828 ) encodes a nuclear orphan receptor expressed in T cells and involved in apoptosis [62] . NR4A2 is also known to play a central role in eliciting the production of inflammatory cytokines in multiple sclerosis ( MS ( MIM 126200 ) ) [63] . Notably , variants in PPP3CA ( Figure 1 ) have recently been reported to correlate with MS severity as well [64] . We therefore investigated whether other genes carrying SNPs which correlate with virus diversity have been identified in GWAS for MS susceptibility or severity . Three additional genes , JMJD2C ( MIM 605469 ) , C20orf133 ( also known as MACROD2 , ( MIM 611567 ) ) and CSMD1 ( MIM 608397 ) have been associated with MS [64] and display SNPs significantly correlated with virus diversity ( Table S1 ) . While the function of C20orf133 is unknown , JMJD2C encodes a histone demethylase expressed at very high levels in B cells and cytotoxic lymphocytes ( see materials and methods ) , a pattern consistent with its being subjected to virus-driven selective pressure . Finally , CSMD1 , in analogy to the aforementioned SERPING1 , acts as a regulator of the complement system [65]; notably , complement activation plays a central role in both response to viruses and inflammatory reactions , particularly in the central nervous system [66] . Analysis of the 30 stronger associations ( Table 3 ) indicated that several genes are part of the network described above or have been involved in immune response ( see InnateDB gene list , Table 2 ) . Conversely , others encode relatively unknown products ( e . g . KIAA1529 ( MIM 611258 ) , LHFPL3 ( MIM 609719 ) , LOC51149 , RNF217 , TMEM132B , LEPREL1 ( MIM 610341 ) , ANKFN1 , MYO5C ( MIM 610022 ) , ANXA4 ( MIM 106491 ) and SCRN3 ) . Among these genes , MYO5C , ANXA4 and SCRN3 are involved in membrane trafficking events along exocytotic and endocytotic pathways , suggesting that they might play a role in either viral cell entry [67] or lytic granule exocytosis; this might be the case for ANXA4 which is expressed at high levels in NK cells ( see materials and methods ) . Most interestingly , EYA4 ( MIM 603550 ) ( Table 3 ) has recently been described as a phosphatase involved in triggering innate immune responses against viruses [68] . Finally , both PDE2A ( MIM 602658 ) and SCNN1A ( MIM 600228 ) might play a role in maintaining lung epithelial barrier homoeostasis during viral infection . Indeed , both genes can be induced by TNF-alpha in lung epithelial cells [69]–[70] and can influence lung fluid reabsorption and , therefore , edema formation . In line with these observations , expression of the amiloride-sensitive epithelial Na+ channel ( SCNN1A codes for the α subunit ) is affected by infection with influenza virus , severe acute respiratory syndrome coronavirus and respiratory syncitial virus . In humans , resistance to infectious diseases is thought to be under complex , multigenic control with single loci playing a small protective role [47] . This concept also holds for viral infection as demonstrated by the role of genetic variants in modulating the susceptibility to HIV infection or disease progression ( reviewed in [71] ) . Classic GWAS offer a powerful resource to identify susceptibility loci for infectious diseases; yet GWAS typically have limited power to detect variants with a low frequency or a small effect . Indeed , recent GWAS for SNPs determining the host control of HIV-1 [4]–[5] failed to identify most known loci with a role in AIDS progression . The alternative approach we have proposed here is based on the identification of variants subjected to virus-driven selective pressure . Similarly to the GWAS results mentioned above we did not identify well known antiviral-response genes . Still , we noticed that variants in TRIM5 ( MIM 608487 ) ( rs2291845 , τ = 0 . 44 , p = 1 . 86×10−5 , rank = 0 . 97 ) and IFIH1 ( MIM 606951 ) ( also known as MDA5 , rs10439256 , τ = 0 . 51 , p = 5 . 4×10−7 , rank = 0 . 99 ) showed significant associations with virus-diversity , although they did not withstood genome-wide analysis . Also , it is worth mentioning that variants with a well established role in resistance to viral infections may be neutrally evolving; this is the case for the Δ32 allele of CCR5 ( MIM 601373 ) for example , which confers protection against HIV-1 infection and possibly against other pathogens , but displays no selection signature [72] . This is possibly due to how long and how strong the selective pressure has been exerted . Conversely , variants subjected to selective pressure must have ( or have had along human history ) some selective advantage , indicating that the SNPs we have identified can be regarded as candidate modulators of infection susceptibility or disease progression .
Virus absence/presence matrices for the 21 countries where HGDP-CEPH populations are located were derived from the Global Infectious Disease and Epidemiology Network database ( Gideon , http://www . gideononline . com ) , a global infectious disease knowledge tool . Information in Gideon is weekly updated and derives from World Health Organization reports , National Health Ministries , PubMed searches and epidemiology meetings . The Gideon Epidemiology module follows the status of known infectious diseases globally , as well as in individual countries , with specific notes indicating the disease's history , incidence and distribution per country . We manually curated virus absence/presence matrices by extracting information from single Gideon entries . These may refer to either species , genera or families ( in case data are not available for different species of a same genus/family ) . Following previous suggestions [7]–[9] , we recorded only viruses that are transmitted in the 21 countries , meaning that cases of transmission due to tourism and immigration were not taken into account; also , species that have recently been eradicated as a result , for example , of vaccination campaigns , were recorded as present in the matrix . A total of 81 virus species/genera/families were retrieved ( Table S6 ) . The same approach was applied to calculate the diversity of other pathogens , namely bacteria , protozoa and helminths [9] . The annual minimum and maximum temperature were retrieved from the NCEP/NCAR database ( http://www . ngdc . noaa . gov/ecosys/cdroms/ged_iia/datasets/a04/ , Legates and Willmott Average , re-gridded dataset ) using the geographic coordinates reported by HGDP-CEPH website for each population ( http://www . cephb . fr/en/hgdp/table . php ) . Similarly , net short wave radiation flux data were obtained from NCEP/NCAR ( http://www . esrl . noaa . gov/psd/data/gridded/data . ncep . reanalysis . surfaceflux . html , Reanalysis 1: Surface Flux ) ; these data were read using Grid Analysis and Display System ( GrADS , http://www . iges . org/grads/ ) . Daily values for four years ( 1948–1951 ) were averaged to obtain an annual mean . Since virus diversity , due to data organization in Gideon , can only be calculated per country ( rather than per population ) , the same procedure was applied to climatic variables . Therefore the values of annual temperature and radiation flux were averaged for populations located in the same country . This assures that a similar number of ties is maintained in all correlation analyses . Data concerning the HGDP-CEPH panel derive from a previous work [6] . Atypical or duplicated samples and pairs of close relatives were removed [73] . A SNP was ascribed to a specific gene if it was located within the transcribed region or no farther than 500 bp upstream the transcription start site . MAF for any single SNP was calculated as the average over all populations . The list of immune response genes was derived from the InnateDB website ( http://www . innatedb . com/ ) and it contains a non-redundant list of 5 , 070 immune genes derived from ImmPort , IRIS , Septic Shock Group , MAPK/NFKB Network and Immunome Database; it only includes genes derived from curated immune gene lists . Genes involved in glycan biosynthesis were obtained by merging genes from two KEGG pathways ( “Glycan structures - biosynthesis 1” and “Glycan structures - biosynthesis 2” ) . Additional genes were identified by searching Gene Ontology categories for genes that act as glycosyltransferases ( GO:0016757 ) and are located in either the Golgi or the endoplasmic reticulum ( GO:0005783 , GO:0005793 and GO:0005794 ) . The list of human genes coding for proteins interacting with viral products was derived from three sources: a previously published study [38] , the VirHostNet website [74] ( http://pbildb1 . univ-lyon1 . fr/virhostnet/ ) and the HIV-1 Human Protein Interaction Database [75] ( http://www . ncbi . nlm . nih . gov/RefSeq/HIVInteractions/ ) . Expression data were obtained from SymAtlas ( http://symatlas . gnf . org/ ) . The location of genomic elements that are highly conserved among vertebrates was derived from UCSC annotation tables ( http://genome . ucsc . edu/; “PhastCons Conserved Elements , 44-way Vertebrate Multiz Alignment” track ) . All correlations were calculated by Kendall's rank correlation coefficient ( τ ) , a non-parametric statistic used to measure the degree of correspondence between two rankings . The reason for using this test is that even in the presence of ties , the sampling distribution of τ satisfactorily converges to a normal distribution for values of n larger than 10 [76] . In order to estimate the probability of obtaining n SNPs located within m genes and significantly associated with virus diversity , we applied a re-sampling approach: samples of m genes were randomly extracted from a list of all genes covered by at least one SNP in the HGDP-CEPH panel ( number of genes = 15 , 280 ) and for each sample the number of SNPs significantly associated with virus diversity was counted . The empirical probability of obtaining n SNPs was then calculated from the distribution of counts deriving from 10 , 000 random samples . A SNP was ascribed to a gene if it was located within the transcribed region or in the 500 upstream nucleotides . Analysis of PANTHER over-represented functional categories and pathways was performed using the “Compare Classifications of Lists” tool available at the PANTHER classification system website [77] ( http://www . pantherdb . org/ ) . Briefly , gene lists are compared to the reference list using the binomial test for each molecular function , biological process , or pathway term in PANTHER . All calculation were performed in the R environment [78] ( http://www . r-project . org/ ) . Biological network analysis was performed with Ingenuity Pathways Analysis ( IPA ) software using an unsupervised analysis ( www . ingenuity . com ) . IPA builds networks by querying the Ingenuity Pathways Knowledge Base for interactions between the identified genes and all other gene objects stored in the knowledge base; it then generates networks with a maximum network size of 35 genes/proteins . We used all genes showing at least one significantly associated SNP as the input set; in this case a SNP was ascribed to a gene if it was located within the transcribed region or in the 25 kb upstream . All network edges are supported by at least one published reference or from canonical information stored in the Ingenuity Pathways Knowledge Base . To determine the probability of the analysed genes to be found together in a network from Ingenuity Pathways Knowledge Base due to random chance alone , IPA applies a Fisher's exact test . The network score represents the -log ( p value ) . | Viruses have represented a constant threat to human communities throughout their history , therefore , human genes involved in anti-viral response can be thought of as targets of virus-driven selective pressure . Here we utilized the marks left by selection to identify viral infection-associated allelic variants . We analyzed more than 660 , 000 single nucleotide polymorphisms ( SNPs ) genotyped in 52 human populations , and we used virus diversity ( the number of different viruses in a geographic region ) to measure virus-driven selective pressure . Results showed that genes involved in immune response and in the biosynthesis of glycan structures functioning as viral receptors display more variants associated with virus diversity than expected by chance . The same holds true for genes encoding proteins that directly interact with viral components . Genome-wide analysis identified 441 variants , mapping to 139 human genes , significantly associated with virus-diversity . We analyzed the functional relationships among genes subjected to virus-driven selective pressure and identified a complex interaction network enriched in viral products-interacting proteins . Therefore , we describe a novel approach for the identification of gene variants that may be involved in the susceptibility to viral infections . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"genetics",
"and",
"genomics/genetics",
"of",
"the",
"immune",
"system",
"evolutionary",
"biology/human",
"evolution",
"immunology/immune",
"response",
"virology/mechanisms",
"of",
"resistance",
"and",
"susceptibility,",
"including",
"host",
"genetics",
"infectious",
"diseases/viral",
"infections",
"virology/host",
"antiviral",
"responses",
"immunology/immunity",
"to",
"infections",
"genetics",
"and",
"genomics/population",
"genetics"
] | 2010 | Genome-Wide Identification of Susceptibility Alleles for Viral Infections through a Population Genetics Approach |
Arenaviruses such as Lassa virus ( LASV ) can cause severe hemorrhagic fever in humans . As a major impediment to vaccine development , delayed and weak neutralizing antibody ( nAb ) responses represent a unifying characteristic of both natural infection and all vaccine candidates tested to date . To investigate the mechanisms underlying arenavirus nAb evasion we engineered several arenavirus envelope-chimeric viruses and glycan-deficient variants thereof . We performed neutralization tests with sera from experimentally infected mice and from LASV-convalescent human patients . NAb response kinetics in mice correlated inversely with the N-linked glycan density in the arenavirus envelope protein’s globular head . Additionally and most intriguingly , infection with fully glycosylated viruses elicited antibodies , which neutralized predominantly their glycan-deficient variants , both in mice and humans . Binding studies with monoclonal antibodies indicated that envelope glycans reduced nAb on-rate , occupancy and thereby counteracted virus neutralization . In infected mice , the envelope glycan shield promoted protracted viral infection by preventing its timely elimination by the ensuing antibody response . Thus , arenavirus envelope glycosylation impairs the protective efficacy rather than the induction of nAbs , and thereby prevents efficient antibody-mediated virus control . This immune evasion mechanism imposes limitations on antibody-based vaccination and convalescent serum therapy .
For most viral vaccines in clinical use today , neutralizing antibodies ( nAbs ) represent the main correlate of protection [1 , 2] . However , viral immune evasion strategies such as antigenic variation and so-called “glycan shields” on viral envelope proteins [3–8] can undermine the protective , neutralizing capacity of antibody immunity . An understanding of the mechanisms underlying viral interference with the host’s antibody defense is , therefore , of pivotal importance to refine vaccination strategies . Members of the Arenaviridae are found worldwide , reflecting the geographic distribution of each virus’ natural rodent host [9] . Several arenaviruses , categorized as high-risk pathogens , can cause lethal hemorrhagic fever in humans and require biosafety level 4 containment . Most prominently , Lassa virus ( LASV ) is endemic in West Africa and accounts for estimated 300’000 human infections with several thousand deaths each year [10] . Similarly , the South American clade B viruses Junin ( JUNV ) , Guanarito , Machupo and Sabia virus cause Argentine , Venezuelan , Bolivian and Brazilian hemorrhagic fever , respectively . Despite these viruses’ socio-economic impact , the live-attenuated JUNV strain Candid #1 [11] remains the only arenavirus vaccine in clinical use [12] . Besides life-supporting intensive care , ribavirin is the only therapeutic option in Lassa fever but shows limited efficacy [13] . Hence the development of a LASV vaccine remains a priority . The human B cell response to LASV infection allows for a timely diagnosis by immunofluorescence and complement fixation [14] . But the kinetics of such non-protective , binding antibody responses contrast with those of nAbs . Already shortly after the identification of Lassa virus in the early 1970ies , Casals and colleagues noted a “lack of synchrony in the development of antibodies detected by the different tests” [14] . Indeed , nAbs are undetectable in the first two to three months after the onset of clinical symptoms , and in most patients remain at or below the 1:100 titer range throughout several months of follow-up [15] . With most convalescent serum donors never reaching an effective titer range [15 , 16] , passive serum therapy in human LASV infection evidenced only limited efficacy [17] . Intriguingly , the discrepancy between binding and neutralizing antibodies was also observed in monkeys immunized with gamma-irradiated Lassa virions [18] . This argued against infection-associated lymphoid depletion and immunosuppression as sole reasons for poor LASV nAb induction [19 , 20] . In contrast to LASV , passive serum therapy represents an efficient treatment against Argentine hemorrhagic fever [21] and formalin-inactivated JUNV , unlike LASV , can induce potent nAb responses [22] . The reasons underlying differential behavior of JUNV and LASV have remained unclear though . A serological response pattern analogous to the one of humans to LASV is observed when mice are infected with lymphocytic choriomeningitis virus ( LCMV ) , a close relative of LASV . Antibodies binding to the nucleoprotein ( NP ) and the glycoprotein-2 subunit ( GP-2 ) are elicited early after infection and reach high titers , whereas neutralizing antibodies target exclusively GP-1 [23] and remain undetectable for the first 40–60 days after infection [24–26] . Furthermore , nAbs only arise in animals with protracted viremia , which is thought to drive continuous somatic hypermutation and antibody evolution [25] . Using reverse genetic techniques to swap glycoproteins between LCMV and vesicular stomatitis virus , we have previously demonstrated that delayed and weak LCMV-neutralizing antibody induction represents a GP-intrinsic feature [27] . Irrespective of the isolation of rare clones of neutralizing monoclonal antibodies ( mAbs ) against LCMV [28 , 29] and LASV [30] which can exhibit therapeutic efficacy in vivo [29] , these observations supported the notion of a neutralization evasion mechanism in these Old World arenavirus glycoproteins . The arenavirus envelope carries a single glycoprotein ( GP ) complex . It is synthesized as GP-C precursor , which is post-translationally cleaved into a stable signal peptide , an outer globular domain ( GP-1 ) and the membrane-anchored GP-2 stalk . These resulting GP complexes consisting of GP-1 , GP-2 and the stable signal peptide remain non-covalently associated on the virion surface and are responsible for receptor binding and membrane fusion . The GPs of LCMV and LASV contain six and seven N-linked glycosylation motifs in GP-1 , respectively , all of which are used during protein biosynthesis [31 , 32] . This was established in earlier mutagenesis studies using rLCMV [32] and plasmid-based expression of LASV-GP [31] , respectively , demonstrating that mutation of each individual N-linked glycosylation site resulted in the predicted reduction in the GP’s molecular mass . Table 1 provides a comparative overview on N-linked glycosylation motifs in thirty arenavirus GP-1 sequences of all clades , which we aligned based on amino acid sequence homology ( see also S1 Fig ) . We numbered the glycans from 1 to 15 ( Glc1 –Glc15 ) , to allow for a comparison of homologous glycans in diverse arenaviruses . N-linked glycosylation impacts protein expression and function [33] , and thus influences LCMV-GP processing , transport and cell fusion [32] . As an additional potential role , early monoclonal antibody ( mAb ) work suggested that Glc12 in GP-1 masked a neutralizing epitope [34] . In support of this hypothesis , a recent mutagenesis study with LCMV found that most GP-1 glycans but not Glc9 and Glc12 affected viral fitness [35] . Here we performed infection experiments with recombinant LCM viruses expressing a range of arenavirus GPs and glycosylation variants thereof . We assessed nAb induction and measured viral sensitivity to neutralization by human and mouse antisera as well as by mAbs . Our findings establish specific viral GP-1 glycans as key mediators of arenavirus nAb evasion in mice and humans . GP-specific antibody responses were readily elicited but reacted predominantly if not exclusively with glycan-deficient viral variants . These observations delineate a viral immune evasion strategy , which prolongs viremia in primary infection and remains to be overcome in antibody-based vaccination against human-pathogenic arenaviruses such as LASV .
We sought support for our hypothesis that N-linked glycosylation represented an arenaviral strategy for nAb evasion . A review of historical data documented that nAb induction differed considerably between individual arenaviruses [37] . Interestingly , we noticed that the Pichinde and Parana viruses with 11 and 10 GP-1 glycosylation motifs , respectively , [38] were reported to elicit lower nAb titers than their relatives Tacaribe , Junin , Amapari , Machupo and Tamiami with only 4–6 such motifs ( Fig 1A ) . Not only the GP-1 as molecular target of nAbs , but also the viral backbone could have influenced nAb induction in the infected host . We therefore engineered recombinant LCM viruses ( rLCMV ) , which expressed the Tacaribe , Junin , Amapari , Machupo , Guanarito or Tamiami GPs instead of LCMV-GP . These GPs were chosen because they all were of clade B phylogeny but spanned a range of between four to six predicted N-linked GP-1 glycans . Upon infection of mice with rLCMV carrying either the Tacaribe or Junin GP ( rLCMV/TAC , rLCMV/JUN; 4 GP-1 glycans ) nAbs were detectable within 8 to 14 days after infection and reached appreciable titers ( Fig 1B ) . rLCMV expressing either the Amapari , Machupo or Guanarito virus GPs ( rLCMV/AMA , rLCMV/MACV , rLCMV/GTO; 5 GP-1 glycans ) induced detectable nAb responses within 14 to 25 days , with lower titers than elicited against the former two recombinant viruses carrying only 4 GP-1 glycans . Finally , nAbs to rLCMV/TAM ( Tamiami virus GP; 6 GP-1 glycans ) remained only marginally above technical backgrounds throughout the 35 days observation period . This suggested an inverse correlation between neutralizing antibody responses and the number of GP-1 glycans . Conversely , all Clade B GP-recombinant LCM viruses elicited comparable LCMV-NP-specific antibody titers ( Fig 1C ) , supporting the concept that differential nAb induction was an intrinsic feature of the individual Clade B GPs ( Fig 1C ) . For further comparison to the Clade B GP-recombinant viruses spanning a range of between 4–6 GP-1 glycans , Fig 1B shows also that rLCMV expressing the Old World LASV GP ( rLCMV/LAS ) with seven GP-1 glycans [38] did not induce any detectable nAbs response within the time frame of our experiment . Altogether , these findings supported the hypothesis that GP-1 glycans represent an impediment to rapid and potent nAb formation by the arenavirus-infected host . Glc9- and Glc12-deficient LCMV-GPs reportedly exhibit normal cell surface expression , and the corresponding viruses ( rLCMVΔGlc9 , rLCMVΔGlc12 ) grow normally in cell culture [35] . Here we infected mice with rLCMVΔGlc9 or rLCMVΔGlc12 to analyze nAb responses . Both , rLCMVΔGlc9 and rLCMVΔGlc12 induced a more rapid nAb response of higher titer than a cDNA-derived control virus with wt GP ( rLCMV , Fig 2A ) . The effect of Glc9 was , however , more pronounced than the one of Glc12 and we therefore centered the remainder study around Glc9 . rLCMVΔGlc9 elicited lower NP-specific antibody responses than rLCMV wt ( S2A Fig ) . This was apparently due to accelerated elimination of rLCMVΔGlc9 and reduced antigen loads over time ( see below ) . Thus , Glc9 deficiency exerted a distinct effect on nAb titers without augmenting antibody responses to the viral backbone . Extending the mutagenesis study to LASV-GP we considered that in addition to the glycans in LCMV-GP , LASV-GP contained Glc5 , which thus might have served antibody evasion purposes . Indeed , rLCMV/LASΔGlc5 induced a rapid nAb response that clearly exceeded the one to the corresponding WT virus ( Fig 2B ) . In light of the above LCMV-GP data we hypothesized that in the context of Glc5 deficiency , Glc9 might also play a role in delaying and weakening nAb induction to LASV-GP . Indeed , when removing the Glc9 motif in addition to Glc5 on rLCMV/LAS ( rLCMV/LASΔGlc5 , 9 ) a stepwise increase and acceleration of the nAb response resulted ( Fig 2B ) . In contrast to these clear differences in nAb responses , all rLCMV/LAS variants induced similar NP-specific antibody titers ( S2B Fig ) . Junin vaccine strains lack Glc11 , which is present in clinical isolates [39] . Hence we compared nAb induction by rLCMV expressing either the Junin vaccine strain XJ Clone 3 GP ( rLCMV/JUN-vacc ) or by the analogous virus , in which the consensus motif for Glc11 had been restored ( rLCMV/JUN ) . rLCMV/JUN induced a less potent nAb response than rLCMV/JUN-vacc , again correlating inversely with GP-1 glycan density ( Fig 2C ) . Importantly , the above results were obtained when assessing serum nAb titers against the very virus used for immunization . Conversely , rLCMVΔGlc9-induced serum antibodies failed to neutralize rLCMV ( Fig 2D ) . Similarly , antibodies elicited by rLCMV/LASΔGlc5 , 9 neutralized the immunizing virus but failed to detectably neutralize rLCMV/LAS or rLCMV/LASΔGlc5 ( Fig 2E ) . Also rLCMV/JUN-vacc immune sera neutralized preferentially the homologous virus ( Fig 2F ) , analogously to earlier observations in vaccinated monkeys [40] . In line with the clinical efficacy of live-attenuated Junin vaccines [11 , 41] , rLCMV/JUN-specific neutralizing activity was also detected in rLCMV/JUN-vacc-immune sera but was comparably lower . These findings suggested that partially deglycosylated GP-1 variants elicited an accelerated and more potent nAb response that was , however , largely specific to the glycan-deficient immunogen . In an inverse approach we assessed whether the antibody response to fully glycosylated wt GPs neutralized the respective partially glycan-deficient variants . Infection with rLCMV induced a late albeit detectable nAb response against itself ( Fig 3A ) . Conversely , rLCMVΔGlc9-neutralizing activity in the same sera was detected earlier and reached higher titers . Even more pronounced , rLCMV/LAS infection elicited a rapid and potent nAb response against the rLCMV/LASΔGlc5 , 9 variant , but no detectable neutralizing serum activity against rLCMV/LASΔGlc5 or rLCMV/LAS used for infection ( Fig 3B ) . We further corroborated the key contribution of Glc9 in reducing nAb sensitivity of LASV by assessing LASV-GP glycan variants lacking individually either glycosylation motifs 3 , 5 , 6 , 9 , 12 or 15 ( S3A Fig ) . Unlike the other mutants tested , rLCMV/LASΔGlc9 was potently neutralized by rLCMV/LAS-immune serum . Finally , the rLCMV/JUN-induced antibody response neutralized rLCMV/JUN-vacc more potently than rLCMV/JUN ( Fig 3C ) . To assess the relevance of these findings for the human immune response to a pathogenic arenavirus , we extended our analysis to LASV-convalescent human sera with known seroreactivity as determined by indirect immunofluorescence [42] . Four out of nine patient sera exhibited detectable neutralizing activity against rLCMV/LAS ( Fig 3D ) . The potency of these “WT neutralizers” sera increased stepwise when tested against the glycan-deficient rLCMV/LASΔGlc5 and rLCMV/LASΔGlc5 , 9 variants , respectively . In further three out of nine patients ( “mutant-only neutralizers” ) , neutralizing activity was only detectable against rLCMV/LASΔGlc5 and/or rLCMV/LASΔGlc5 , 9 . Two patient sera ( “non-neutralizers” ) failed to detectably inhibit the infectivity of either virus . These data showed that preferential neutralization of glycan-deficient LASV-GP variants , as observed in rLCMV/LAS-infected mice ( Fig 3B ) , extended to humans infected with wt LASV in the field . These observations suggested that , both in mice and humans , arenavirus infections elicited serum antibodies that neutralized predominantly glycan-deficient viral variants . In support thereof , a panel of rLCMV-immune mouse sera exhibited a statistically significant correlation between their neutralizing potency against rLCMVΔGlc9 and WT rLCMV ( Fig 4A ) . On average the former activity exceeded the latter one by about four-fold . Analogous observations were made when mice were infected with recombinant LCMV expressing the Armstrong strain GP ( rLCMV/ARM ) or a variant thereof lacking Glc12 ( rLCMV/ARMΔGlc12 , corresponding to the formerly described Armstrong 4 isolate [34] , S3B Fig ) . These findings raised the possibility that a proportion of serum antibodies reacted against WT virus and additionally , with higher potency , also neutralized glycan-deficient variant viruses . In support of this hypothesis we found that KL25 , a widely used WT LCMV-induced mAb [28] , neutralized rLCMVΔGlc9 at roughly 20-fold lower concentration than rLCMV carrying the wt GP ( Fig 4B and 4D ) . Conversely , the IC50 of the WEN3 mAb on rLCMV and rLCMVΔGlc9 was not significantly different ( Fig 4C and 4D ) . In concert with this observation , point mutations in KL25 escape variants cluster around Glc9 [43 , 44] and don’t affect WEN3 binding or neutralization , suggesting the two mAbs recognize distinct epitopes . An electron microscopic assessment of virion labeling with saturating concentrations of KL25 and WEN3 , respectively , indicated that glycoprotein densities on rLCMV and rLCMVΔGlc9 differed by less than 1 . 5-fold ( S4 Fig ) . This was in line with earlier observations on unimpaired cell surface expression of LCMV-GPΔGlc9 [35] and suggested differential recognition rather than differential availability of the KL25 epitope on rLCMV and rLCMVΔGlc9 . Similarly to the behavior of KL25 against LCMV-GP , a panel of WT JUNV-induced mAbs exhibited consistently higher potency against rLCMV/JUN-vacc ( 3 GP-1 glycans ) than against rLCMV/JUN ( 4 GP-1 glycans; Fig 4E ) . The relative differences in potency against the two viruses varied , however , between mAbs . Taken together , these observations suggested that neutralizing antibodies , which were induced in response to fully glycosylated wt GPs , exhibited higher potency when specific glycans were removed from the target antigen . In an inverse approach , we introduced Glc5 in LCMV-GP ( rLCMV+Glc5 ) thus mimicking LASV glycosylation . rLCMV+Glc5 was viable [35] but it was ≥100-fold less sensitive to KL25 or WEN3 neutralization than WT virus ( Fig 4F ) , further attesting to the capacity of Glc5 to shield arenaviruses against neutralizing antibodies . We hypothesized that facilitated binding was accountable for glycan-dependent differences in neutralization potency of the KL25 mAb . We transfected 293T cells with LCMV-GPwt or LCMV-GPΔGlc9 expression plasmids , respectively , and used flow cytometry to establish saturation curves and resulting EC50 values for the Glc9-sensitive KL25 and the Glc9-insensitive WEN3 mAbs . The EC50 of KL25 on LCMV-GPΔGlc9 was approximately five-fold lower than on LCMV-GPwt , whereas comparable WEN3 concentrations were required to bind the two LCMV-GP versions ( Fig 5A ) . Thus , higher KL25 occupancy of LCMV-GPΔGlc9 as compared to its wt counterpart contrasted with the indiscriminate behavior of WEN3 , matching the neutralization behavior of these mAbs ( compare Fig 4B–4D ) . To further dissect these interactions we performed surface plasmon resonance measurement of KL25 and WEN3 Fab binding to the soluble ectodomains of LCMV-GPΔGlc9 and LCMV-GPwt . Counter to expectations based on neutralization sensitivity ( Fig 4B ) , the overall affinity of KL25 Fab binding to LCMV-GPΔGlc9 was modestly lower than its binding to LCMV-GPwt ( i . e . higher KD = kd/ka; Figs 5B and S5 ) . This was due to a higher off-rate ( kd ) that partially counterbalanced an elevated on-rate ( ka ) . Thus , like for mAbs against members from other viral families [46 , 47] , the higher on-rate of the KL25 Fab on LCMV-GPΔGlc9 than on LCMV-GPwt represented the best correlate of increased neutralizing potency of the dimeric full length mAb . Conversely , WEN3 affinity was slightly lower on LCMV-GPΔGlc9 than on LCMV-GPwt , and on-rate differed only modestly , which was in agreement with virtually identical neutralizing potency and binding in flow cytometry . Iso-affinity plots illustrated the non-discriminative binding behavior of WEN3 , which contrasted with the differential on-rate but comparable overall affinity of KL25 for the wt and Glc9-deficient GP variants , respectively ( Fig 5C ) . Taken together , the reduction in antibody on-rate offered a mechanistic explanation how Glc9 shielded LCMV-GP against antibody neutralization . The only arenavirus pre-fusion GP-1 structure that has been resolved is that from Machupo virus [48 , 49] . Despite the low sequence homology of MACV GP-1 with LASV GP-1 , secondary structure predictions ( Fig 6A ) indicated that their core folds were conserved , and a web-based algorithm [50] calculated 100% confidence for structural homology . This prompted us to map the location of LASV GP-1 N-linked glycans onto the MACV GP-1 surface ( Fig 6B ) . Glc5 , Glc9 and Glc12 , which apparently reduce neutralization sensitivity of Old World arenaviruses , all projected onto solvent-exposed loops outside the receptor-binding footprint on MACV ( Fig 6C ) . A limitation of this model consists in the fact that clade B viruses , such as MACV , utilize transferrin receptor 1 for entry [51] whereas alpha-dystroglycan serves as receptor for the Old World arenaviruses LASV and LCMV [52] . Receptor binding sites for the latter two viruses have not yet been mapped . Nevertheless , the model supports the mechanistic postulate that Glc5 , Glc9 and Glc12 serve to shield arenaviruses against antibodies by reducing their access to highly immunogenic protein loops on the GP-1 surface . The clustering of KL25 mAb escape mutations around Glc9 [43 , 44] is also in line with this concept but additional structural information on arenavirus envelope GPs will be required to formally test these assumptions . nAb responses not only protect against viral reinfection but can also help resolving primary infection [26 , 53] . Hence we tested the possibility that the arenavirus glycan shield impeded efficient virus control by promoting nAb evasion . For this we exploited T11μMT mice , which mount normal CD4+ and CD8+ T cell responses to LCMV [26] , but have a quasi-monoclonal B cell repertoire recognizing virtually exclusively the LCMV-unrelated glycoprotein of vesicular stomatitis virus . Accordingly , T11μMT mice failed to mount nAb responses when infected with either rLCMVΔGlc9 or WT rLCMV ( Fig 7A ) . In contrast , wild type mice mounted a rapid and potent nAb response against rLCMVΔGlc9 but not against fully glycosylated rLCMV , as expected ( Fig 7B ) . Therefore the comparison of viral loads in these two congenic strains of mice allowed us to directly assess the impact of the rapid nAb response on rLCMVΔGlc9 control . In concert with identical growth of rLCMVΔGlc9 and rLCMV in cell culture ( S6 Fig ) , the two viruses persisted at indistinguishable levels in the blood of T11μMT mice throughout the observation period of 19 days ( Fig 7C ) . In contrast , rLCMVΔGlc9 was cleared from the blood of C57BL/6 wt mice by day 19 , whereas the glycan-shielded rLCMV virus persisted ( Fig 7D ) . This protracted course of infection was expected for the LCMV strain Clone 13-based viruses used in our experiments [54 , 55] . Viral loads in blood of rLCMVΔGlc9- and rLCMV-infected C57BL/6 mice were significantly different from day 12 onwards , which was in line with the early onset of the nAb response . Altogether , this demonstrated that glycan-mediated nAb evasion promotes protracted LCMV infection .
Glycan shielding of arenavirus GPs provides an explanation for the consistent failure to induce potent LASV-specific antibody immunity by either vaccination or natural infection [14 , 15 , 17 , 18 , 56–59] . In both circumstances , specific ELISA titers were high while neutralizing activity remained modest at best [56–58] . In line with these observations , our data suggest that glycosylation does not primarily prevent GP-1-specific antibody induction , but it impairs the capacity of these antibodies to neutralize . This shielding mechanism we propose differs from previous concepts such as the supposed “hole” in the arenavirus GP-1-specific B cell repertoire [25] . Neither would the arenavirus GP-1 represent the equivalent of an “immunologically silent face” in HIV-1 [60] , i . e . GP-1 does not seem to lack immunogenicity owing to glycan resemblance to “self” . Our observations are more reminiscent of the “glycan shield” concept for HIV-1 [3] , proposing that glycans impair antibody access to neutralizing epitopes on gp120 . In stark contrast to HIV-1 , however , the available sequence data suggest that arenavirus GP-1 glycans are invariable between isolates . Conversely , the difference in glycan density between the GP-1 of JUNV and LASV represents a likely reason why the excellent therapeutic success of convalescent serum in Argentine hemorrhagic fever does not find a parallel in Lassa fever [17 , 21] . The presence of comparably fewer glycans in JUNV-GP-1 also is likely to facilitate the induction of antibody-mediated protection by the live-attenuated JUNV vaccine Candid#1 [11 , 41] . In contrast , our data indicate that glycan-deficient LASV-GPs as immunogens will not overcome these structural hurdles . Removal of Glc9 from LCMV-GP ( LCMV-GPΔGlc9 ) increased the association rate of the neutralizing mAb KL25 , thus augmenting the antibody’s ability to neutralize in spite of its slightly lower affinity for this deglycosylated target antigen . This interpretation is in line with the result from large-scale mutagenesis studies on the therapeutic RSV antibody Palivizumab , demonstrating that a fairly modest increase in association rate can translate into a considerably heightened neutralizing potency . Conversely , dissociation rates were found of comparably minor impact owing to the dimeric nature of IgG binding [46] . Analogously mathematical models predict Ab on-rate as a major determinant of HIV-neutralizing Ab potency [47] . We hypothesize therefore that key glycans such as Glc9 and Glc5 in LASV-GP-1 shield the virus against nAbs by reducing their access to neutralizing epitopes . Glycan-mediated stabilization of a distinct poorly accessible pre-fusion conformation represents an alternative and not mutually exclusive mechanism [61] . Effective prevention of Lassa fever remains a priority in West Africa where LASV is endemic . Additionally , a vaccine would allow for the timely containment of potential future outbreaks and for the protection of healthcare workers . As illustrated by the recent Ebola epidemic , which has ravaged the same geographic area [62] , viral hemorrhagic fevers can rapidly emerge to a global health concern . Hence , international efforts at developing a LASV vaccine should be intensified up to the level of human clinical trials [12] . Amongst a larger number of candidates ( reviewed in [12 , 63] ) , the LASV-related apathogenic arenavirus Mopeia ( MOPV ) , a chimeric LASV/MOPV reassortant virus ( ML29 ) , recombinant vaccinia viruses expressing LASV structural proteins and a replicating vesicular stomatitis virus expressing LAS-GP have shown safety and efficacy in non-human primate models [56–59] . Intriguingly , the protective efficacy of all of these vaccines has been accredited to cell-mediated immunity . Also for MOPV , a high level of sequence similarity to LASV ( 76% , 74% , 57% and 58% for GP , NP , L and Z , respectively ) , and the induction of LASV-specific T cell responses in MOPV-infected mice [64] support this interpretation . As a second example of heterologous arenavirus immunity , the attenuated Junin vaccine strain XJ clone 3 induced negligible MACV-specific nAb titers but protected against disease upon MACV challenge [65] . High sequence similarity ( 69% , 88% , 73% and 76% for GP , NP , L and Z , respectively ) as a basis for cross-protective T cell immunity seems a likely mechanism , but an accelerated nAb response upon challenge [65] could also have contributed to MACV control . In light of the present findings , modest glycan density on MACV-GP-1 as compared to LASV-GP-1 may have facilitated this response ( compare Fig 1B ) . The use of life-attenuated or replicating vectored vaccines can be associated with significant reactogenicity as well as with anticipated ( infancy , pregnancy ) or unexpected safety issues [66 , 67] . These constraints are of lesser concern in an outbreak control setting , and may also be acceptable if protection requires potent T cell responses , notably of the CD8+ subset . Conversely , if nAbs could be exploited as effectors of protection , inactivated or subunit vaccines might be preferable for population-wide vaccination campaigns as will be needed to control endemic Lassa fever in West Africa . In order for a LASV vaccine to reproduce the success of the numerous antibody-based vaccines in clinical use today [1 , 2] , a profound understanding of the hurdles on this path is of paramount importance . As recently exemplified for respiratory syncytial virus ( RSV ) , epitope-focused scaffold-based vaccine design can generate artificial vaccine antigens for challenging antibody targets [68] . A detailed mechanistic understanding of LASV nAb evasion will represent an essential basis to generate analogous scaffold-based approaches for this virus . Additional structural information on arenavirus GPs [48 , 49] including analyses of their interactions with nAbs will also be required . Still , it remains uncertain whether scaffold-embedded epitopes as immunogens can induce potent nAb responses against glycan-shielded viral epitopes [69] . Hence , our results can also be taken as a rationale to argue that for densely glycosylated arenaviruses such as LASV , nAb-based vaccination may not be feasible . By providing this mechanistic explanation , our data will help justifying the clinical use of more reactogenic vaccine delivery strategies such as life-attenuated and replicating vectored vaccines , which can induce potent T cell-based protection [56–59] . For Lassa fever , high viral loads are a predictor of lethal outcome [13] . Augmented and prolonged viremia due to glycan-mediated nAb evasion ( Fig 7 ) may thus suggest that the envelope glycan shield represents an arenavirus virulence factor . For Junin virus , reverse genetic mapping studies have been conducted both in suckling mice and guinea pigs , and have unanimously identified an attenuating mutation in the GP-2 transmembrane domain of the Candid #1 vaccine strain , which reduces virion infectivity [39 , 70] . Unlike in suckling mice , however , the guinea pig model has provided evidence for at least one additional attenuating mutation in GP , with impact on viral dissemination and disease [70] . It seems tempting to speculate that Glc11 deficiency of Candid#1 , which we show can facilitate antibody neutralization , may have contributed to the vaccine’s attenuated phenotype in guinea pigs . The failure to detect a putative Glc11 effect in suckling mice [39] could have been due to these young animals’ immunological immaturity , which entails reduced antibody responsiveness [71] . In addition , the more rapid disease course in mice ( ~10 days ) as compared to guinea pigs ( ~18 days ) may have outpaced nAb effects [53 , 70 , 72] . Alternatively , cell-mediated immunity may control primary Candid#1 infection largely independently of nAb responses [53] . In summary , our study shows that specific GP-1 glycans shield the arenavirus envelope against efficient antibody neutralization , thus limiting the protective capacity of humoral immune defense and promoting protracted infection . This lends a novel perspective on these viruses’ immune evasion strategies and provides strategic guidance for LASV vaccine development .
C57BL/6 mice were bred at the Institute for Laboratory Animal Sciences of the University of Zurich or were purchased from Charles River Laboratories . Animal experiments were performed at the Universities of Zurich , Geneva and Basel . Experimental groups were sex and age-matched . Anonymised human sera were obtained from a serum bank at the University of Marburg , Germany . They originate from a highly LASV-endemic area of Guinea and were identified as seropositive for LASV by immunofluorescence . They correspond to the previously characterized sera from 1999 [42] . LCMV clone 13 expressing either the LCMV WE strain glycoprotein ( referred to as rLCMV WT herein ) , heterologous arenavirus GPs or glycosylation variants thereof were generated from cDNA by reverse genetic techniques [73] . J . C . de la Torre generously provided a cDNA of the Lassa virus strain Josiah GP . Reverse transcribed virion RNAs of Machupo and Guanarito virus were generously provided by R . Charrel . The GPs of the Junin vaccine strain XJ clone 3 , of Amapari and Tamiami virus were RT-PCR cloned from virion RNA . The viruses were kindly provided by R . Zinkernagel . To substitute the GP ORF in the LCMV S segment cDNA for heterologous GP cDNAs , a PCR cloning strategy was utilized as previously outlined in detail [74] . N-linked glycosylation sites were deleted by either a two-way PCR or a circular PCR strategy , mutating the N-X-S/T motif to Q-X-S/T based on a double-nucleotide change in the respective codon . Additional glycosylation sites were introduced by analogous procedures . The sequences of all cDNAs used for virus rescue have been validated by DNA sequencing . Oligonucleotide primer sequences are available from the authors upon request . Viruses to be used for neutralization assays were generally grown on BHK-21 . Viruses for immunization of mice were grown on BHK-21 or 293T-GP cells [75] . Vero cells were used for work with rLCMV/TAC . BHK-21 , 293T and Vero cells were obtained from the American Type Culture Collection ( ATCC ) . All viruses were titrated as previously described [75 , 76] . The neutralizing capacity of mAbs and immune serum was tested in immunofocus reduction assays [27 , 76] . Sera were typically pre-diluted 1/8 or 1/10 , followed by serial two-fold dilution steps , and were tested against a constant amount of virus . Neutralizing titers of mouse serum are expressed as the serum dilution yielding 50% immunofocus reduction . To reflect this assay setup in the figures , neutralizing titers are reported as negative log2 values , which must be multiplied by the pre-dilution factor given in each figure legend . For example , a neutralizing titer of 3 determined in 8-fold pre-diluted serum indicates a 50% neutralizing titer at a serum dilution of 1:32 . For a more precise assessment of the potency of human LASV-convalescent sera and mAbs , the number of foci at any given serum dilution or antibody concentration was expressed in percent of the average number of foci obtained in the absence of serum or mAb . To obtain a precise IC50 value for mAbs , the latter type of measurement was automated for high throughput using an immunospot reader ( Cellular Technology Ltd . ) , and neutralization curves were drawn using Graphpad Prism software . The LCMV monoclonal antibodies have been described [27 , 28] . JUNV-specific antibodies [45] were generously provided by the Biodefense and Emerging Infections Research Resources Repository ( BEI Resources; catalog numbers: GB03 ( NR-2564 ) , GD01 ( NR-2565 ) , LD05 ( NR-2569 ) , OD01 ( NR-2567 ) , QC03 ( NR-2566 ) ) . Contributors to the BEI catalog were NIH Biodefense and Emerging Infections Research Resources Repository and BEI Resources . To measure LCMV-specific IgG titers , 96-well plates were coated with 100 μl of recombinant , bacterially expressed LCMV-NP at 3 μg/ml in sodium carbonate buffer ( pH 9 . 6 ) . Plates were blocked for 2h with 5% milk in PBS-Tween 0 . 05% ( PBS-T-milk ) . In a parallel 96-well plate serum samples were prediluted 1:100 in PBS-T-milk , and threefold dilution series were performed . 100 μl of the diluted serum samples was then transferred to the NP-coated plate for 1h . Finally , the plates were incubated for 1h with HRP-coupled goat anti-mouse IgG Ab ( Jackson 115-035-062 ) diluted 1:1’000 in PBS-T-milk . HRP was detected by addition of ABTS color reaction . All steps were carried out at room temperature . Plates were washed three times with PBS-T between each step . NP-specific IgG titers were defined as the log3 dilution resulting in an optical density at 450 nm that was twofold above background . To assess KL25 and WEN3 mAb binding to the native membrane-bound form of LCMV-GP , 293T cells were transfected with LCMV-GPwt or ΔGlc9 using saturating plasmid amounts . 48 hours later , we harvested the cells and stained them with titrated concentrations of KL25 or WEN-3 mAb for 5 min at RT followed by detection with PE-conjugated goat anti-mouse IgG . The fluorescence signal was measured on an LSR Fortessa flow cytometer ( BD ) and was analyzed using FlowJo software . For use in SPR assays , the ectodomain of the LCMV-GPwt ( WE strain , aa 1–430 ) and the respective Glc9 mutant version were C-terminally fused to streptag II ( SA-WSHPQFEK ( GGGS ) 2GGSAWSHPQFEK; Twin-Strep-tag , IBA GmbH , Germany ) and were expressed in transiently transfected 293T cells . The protein was purified for SPR assays using Strep-tactin purification columns according to the manufacturer’s instructions ( IBA GmbH , Germany ) . KL25 and WEN3 Fabs were obtained by recombinant expression and enzymatic cleavage , respectively . Affinity and kinetics of Fab binding were determined on a Biacore 2000 ( GE Healthcare , Uppsala , Sweden ) . A CM5 sensor chip ( GE Healthcare , Uppsala , Sweden ) was covalently coupled with the StrepMAB-Immo antibody ( IBA BioTAGnology , St . Louis , MO ) by amine coupling . The surface was activated for 7 min at 10 μL/min with a mix 1:1 containing EDC and NHS solutions to final concentrations of 200 and 100 nM respectively . The StrepMAB-Immo was diluted in 10 mM acetate buffer at pH 5 . 0 and injected at 10 μL/min for 7 min . Unused activated chip surface was blocked by injecting 1 M ethanolamine for 7 min . This process resulted in the immobilization of StrepMAB-Immo antibody at densities ranging from 5000 to 10000 RU . Then , soluble LCMV-GP ( wt or Glc9 mutant ) was injected for 5 min at a concentration of 50 μg/mL and a flow rate of 5 μL/min , leading to capture levels between 800 and 1500 RU . Kinetics were performed at 25°C , in HBS EP buffer ( GE Healthcare , Uppsala , Sweden ) , at a flow rate of 30 μL/min . KL25 and WEN3 Fabs were injected for 5 min in duplicate and randomly at five and six decreasing concentrations , starting from 500nM and 1000nM , respectively . The dissociation phase was monitored for 30 min . Regeneration was assessed using a 10 mM glycine pH 1 . 5 solution injected for 3 min . Curves were fitted according to the 1:1 Langmuir binding model and using the BIAevaluation 4 . 1 . 1 software ( GE Healthcare , Uppsala , Sweden ) . A double referencing was applied for analysis to subtract buffer signal drift on coated surface and unspecific background signal on a reference channel . All experiments were performed in triplicates . BHK-21 cells were infected with either rLCMV or rLCMVΔGlc9 at MOI 0 . 1 . 40 h later , the cells were fixed for 60 min at RT in 0 . 1 M phosphate buffer ( pH 7 . 4 ) containing 2% paraformaldehyde and 0 . 02% glutaraldehyde . After washing , the cells were scraped off the culture dishes , embedded in 12% gelatin , infused with 2 . 3 M sucrose , frozen in liquid nitrogen , and sectioned with a EMFCS ultracryomicrotome ( Leica ) . Ultrathin sections were immunostained for 15 h with either KL25 or WEN-3 mAb at a saturating concentration of 100 μg/ml , followed by a 20 min RT incubation with Protein A-coated 10 nm gold particles [77] . Cryosections were screened and photographed using a CM10 electron microscope ( Philips , Eindhoven , The Netherlands ) . For the evaluation of GP labeling density on virions , 160–180 cells were photographed at 21 , 000-fold magnification and the number of gold particles per virion was manually counted . Amino acid sequence alignments and automatic N-X-S/T motif searches were performed using the Jalview software [78] . The secondary structure of LASV GP-1 was predicted with NPS@ [79] . Fold prediction was performed using the Phyre2 fold prediction tool [50] . The MACV GP-1 structure was previously published ( PDB accession number 2WFO [48] ) and residues forming contacts with human TfR1 were determined with the PISA EBI server [80] using PDB accession number 3KAS [49] . Protein sequence similarities of LASV/MOPV and JUNV/MACV described in the discussion section were calculated online with BLASTp , using comparison of two protein sequences [81] . For statistical analysis , the GraphPad Prism software ( version 5 . 04 , GraphPad Software , San Diego , California ) was used throughout . Titers values were log-converted to obtain a near-normal distribution . To assess significant differences between single measurements of 2 groups we used two-tailed Student’s t tests . Differences between multiple measurements of 2 or more groups were assessed by two-way ANOVA followed by multiple t tests with Bonferroni adjustment for multiple comparisons if the F test of ANOVA indicated statistically significant differences . To analyze correlations , linear regression was performed and the Pearson’s correlation coefficient as well as a two-tailed p-value were calculated . P-values <0 . 05 were considered statistically significant ( * ) , p<0 . 01 was considered highly significant ( ** ) and p>0 . 05 was considered not statistically significant ( ns ) . Animal experiments were approved by the Cantonal Veterinary Office of the Canton of Zurich ( permission 176/2005 ) , the Direction Générale de la Santé ( permissions 1005/3312/2 and 1005/3312/2-R ) of the Canton of Geneva , and the Cantonal Veterinary Office of the Canton of Basel ( permission 24257/2666 ) , respectively . All animal experiments were performed in accordance with the Swiss law for animal protection . The measurements of LASV-nAbs in anonymised human sera were performed with ethical approval by the Ethik-Kommission des Kantons Zürich ( KEK , Ref . Nr . : StV 49–2006 ) . The Genbank accession numbers for genes and proteins mentioned in this study are shown below in parentheses . LUJV ( FJ952384 ) , DANV ( EU136038 ) , LCMV-WE ( AJ297484 ) , LCMV-ARM ( AY847350 ) , LASV ( J04324 ) , MOBV ( AY342390 ) , MOPV ( AY772170 ) , MORV ( EU914103 ) , IPPYV ( DQ328877 ) , FLEV ( AF512831 ) , ALLV ( AY012687 ) , PARV ( AF485261 ) , PIRV ( AF277659 ) , PICV ( K02734 ) , OLVV ( U34248 ) , LATV ( AF485259 ) , JUNV-vacc ( HQ126699 ) , JUNV ( AY358023 ) , TCRV ( KP159416 ) , MACV ( AY619643 ) , AMAV ( AF512834 ) , GTOV ( AF485258 ) , CPXV ( AF512832 ) , SABV ( U41071 ) , CHP ( EU260463 ) , BCNV ( AY924391 ) , CATV ( DQ865244 ) , NAAV ( EU123329 ) , SKTV ( EU123328 ) , TAMV ( AF512828 ) , WWAV ( AF228063 ) . The arenaviruses’ full names corresponding to the above acronyms can be found in Table 1 . | Neutralizing antibodies ( nAbs ) represent a key principle of antiviral immunity . Protective vaccines aim at inducing nAbs to prevent viral infection , and infusion of nAbs in convalescent patient serum can offer a potent antiviral therapy . Certain viruses , however , have found ways to evade nAb control . Amongst them are high-risk pathogens of the arenavirus family such as Lassa virus ( LASV ) , which is a frequent cause of hemorrhagic fever in West Africa . Here we unveil the molecular strategy by which arenaviruses escape antibody neutralization and avoid efficient immune control . We show that their surface is decorated with sugar moieties , serving to shield the virus against the neutralizing effect of the host’s antibodies . This immune evasion strategy differs from those described for other viruses , in which sugars impair primarily the induction of antibodies or allow for viral mutational escape . The arenavirus sugar coat renders the host nAb response inefficient and as a consequence thereof , the host fails to promptly control the infection . Our results offer a compelling explanation for the long history of failures in trying to make a nAb-based vaccine against LASV or in using convalescent serum for therapy . These mechanistic insights will support vaccine development efforts against arenaviruses such as LASV . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Arenavirus Glycan Shield Promotes Neutralizing Antibody Evasion and Protracted Infection |
Accurately describing synaptic interactions between neurons and how interactions change over time are key challenges for systems neuroscience . Although intracellular electrophysiology is a powerful tool for studying synaptic integration and plasticity , it is limited by the small number of neurons that can be recorded simultaneously in vitro and by the technical difficulty of intracellular recording in vivo . One way around these difficulties may be to use large-scale extracellular recording of spike trains and apply statistical methods to model and infer functional connections between neurons . These techniques have the potential to reveal large-scale connectivity structure based on the spike timing alone . However , the interpretation of functional connectivity is often approximate , since only a small fraction of presynaptic inputs are typically observed . Here we use in vitro current injection in layer 2/3 pyramidal neurons to validate methods for inferring functional connectivity in a setting where input to the neuron is controlled . In experiments with partially-defined input , we inject a single simulated input with known amplitude on a background of fluctuating noise . In a fully-defined input paradigm , we then control the synaptic weights and timing of many simulated presynaptic neurons . By analyzing the firing of neurons in response to these artificial inputs , we ask 1 ) How does functional connectivity inferred from spikes relate to simulated synaptic input ? and 2 ) What are the limitations of connectivity inference ? We find that individual current-based synaptic inputs are detectable over a broad range of amplitudes and conditions . Detectability depends on input amplitude and output firing rate , and excitatory inputs are detected more readily than inhibitory . Moreover , as we model increasing numbers of presynaptic inputs , we are able to estimate connection strengths more accurately and detect the presence of connections more quickly . These results illustrate the possibilities and outline the limits of inferring synaptic input from spikes .
Neural computation requires fast , structured transformations from presynaptic input to postsynaptic spiking [1–3] . Changes in these transformations underlie learning , memory , and recovery from injury [4 , 5] . Tools for identifying synaptic weights and tracking their changes , thus , play a key role in understanding neural information processing . Traditionally , synaptic integration and plasticity are studied using intracellular recordings in vitro , where synaptic weights can be directly measured as the amplitude of postsynaptic potentials or currents . Although there are singular studies employing simultaneous intracellular recordings from several neurons in vivo [6–8] , recording intracellularly from connected neurons in vivo is technically prohibitive . On the other hand , methods for recording extracellular spike trains are advancing at a rapid pace [9 , 10] and allowing the simultaneous recording of hundreds of neurons . Estimation of synaptic interactions from extracellularly recorded spike trains requires development of sensitive data analysis tools . Although strong synapses are usually readily detectable using cross-correlation analysis [11–17] , where they appear as asymmetric , short latency peaks on cross-correlograms [18 , 19] , in general , it is difficult to link the statistical relationships between spike trains to specific synaptic processes [20 , 21] . Here we provide empirical tests of statistical tools for such analysis using in vitro current injection where the true synaptic input is known . As techniques for large-scale electrical [22] and optical [23] neural recordings continue to improve , methods for inferring interactions between the recorded neurons are needed to provide insight into the connectivity and information processing of neural circuits . Although correlational methods have long been used to study interactions between pairs of neurons [18 , 19] , recent work has shown that statistical inference methods may be able to substantially improve our ability to detect neuronal connectivity and predict neural activity [24–26] . These model-based methods [22 , 27 , 28] are important in removing the confounds that occur with simultaneous recordings [20 , 29] and have revealed highly structured functional interactions , that accurately reflect the known circuit architecture , in the retina [30] and invertebrate systems [31] . However , it has proven difficult to relate functional connectivity reconstructed from spikes to the known anatomy and physiology of cortical connectivity [26 , 32–34] . Sparse sampling of neurons and large electrode spacing may contribute somewhat to the difficulty in interpreting the results of functional connectivity analyses of cortical circuits , but it is also unclear what information these inference methods can provide about actual synaptic inputs and what limitations there are to the use of these methods in general . Here we examine to what extent the functional connections estimated from spike trains correspond to simulated synaptic processes in a highly controlled setting . We use in vitro intracellular recordings from layer 2/3 pyramidal cells in slices from rat neocortex as they respond to simulated , current-based presynaptic input . The fully-defined input is composed of excitatory and inhibitory postsynaptic currents produced by firing of large number of simulated presynaptic neurons . Since we know both the spike timing of the input presynaptic neurons and spike timing of the postsynaptic neuron , we can examine the limits of functional connectivity inference . We ask how well synaptic inputs of different amplitudes can be detected , how much data is necessary to reconstruct the amplitudes of excitatory and inhibitory synaptic inputs , and how precisely synaptic weights can be estimated from spikes alone . We briefly examine the feasibility of tracking changes in synaptic weight over time . Finally , we examine how accurately the firing of the postsynaptic neuron can be predicted from presynaptic spiking , and to what extent knowledge of multiple presynaptic inputs improves the accuracy of spike prediction .
In a first set of experiments the input was partially-defined . We injected into layer 2/3 pyramidal cells current consisting of three components: simulated , artificial excitatory postsynaptic current ( aEPSC ) from a single presynaptic neuron , fluctuating noise with standard deviation σ , and a DC offset ( Fig . 1A ) . We adjusted the gain of the injected fluctuating current to produce membrane potential fluctuations with ∼15–20 mV peak to peak amplitude and DC current to achieve postsynaptic spiking ∼5Hz . After this adjustment , the overall standard deviation of the scaled fluctuating current σ was between 70 and 110pA . On top of this fluctuating noise , we injected aEPSCs produced by the single presynaptic input and varied the amplitude of aEPSCs in different realizations . We then analyzed the postsynaptic responses aiming 1 ) to examine whether input can be detected based on spikes alone , 2 ) to quantify how much data is necessary to detect a synaptic input of a given strength , 3 ) to quantify how much data is necessary to detect changes in input strength , and 4 ) to determine how accurately such pairwise models describe and predict spiking of the postsynaptic neuron . Traditionally , the effects of synaptic input on postsynaptic spiking are assessed using descriptive , cross-correlation methods ( Fig . 1C ) . Here we use a common model for estimating functional connectivity from point-process observations: a generalized linear model ( GLM ) . We assume that postsynaptic spiking is generated by a Poisson process with a rate determined by a baseline firing rate , the recent history of the neuron’s firing , as well as input produced by presynaptic spikes ( see Methods for details ) . To determine whether an input of a certain amplitude can be “detected” given a specific set of spike trains we use the log likelihood ratio ( LLR ) . Specifically we compare a model that predicts postsynaptic spikes based only on the recorded neuron’s spike history ( Model 1 ) with a model that uses both spike history and presynaptic input ( Model 2 ) . These models accurately capture two different aspects of postsynaptic spiking: the spike history term captures the fact that immediately after an action potential , the probability of spike generation decreases , while the coupling terms capture the variable ( excitatory , in these experiments ) effect of the presynaptic input ( Fig . 2A ) . In Model 2 the estimated presynaptic inputs correlate well with the actual amplitude of EPSCs ( Fig . 2A and 2E ) . Note that these effects do not correspond to exact biophysical processes , but merely capture the statistics of postsynaptic firing . The model does not aim to distinguish between the underlying processes that determine spike timing . For instance , in vivo results show that , although absolute refractoriness lasts only few milliseconds , generation of an action potential can influence the spike threshold for up to 1s [35 , 36] . Similarly , although postsynaptic potentials ( PSPs ) persist only over 10’s of ms , the effect of an EPSP on postsynaptic spike timing may last substantially longer [36 , 37] . Here , the estimated post-spike history and coupling effect in the models persist over 50–200ms . Both the history-only Model 1 and Model 2 with constant coupling capture statistics of the spike trains and predict spikes with reasonable accuracy . Model 2 provides an increasingly accurate prediction as the input amplitude increases ( Fig . 2B ) . Using ROC analysis ( 1ms timescale , 10-fold cross-validated ) , we find that spikes are predicted with an area-under-the-curve ( AUC ) of 0 . 71±0 . 01 for Model 1 , while Model 2 yields 0 . 72±0 . 01 , 0 . 74±0 . 01 , and 0 . 77±0 . 01 as the input amplitude increases from 0 . 5 σ , to 1 . 00 σ , to 1 . 5 σ . The relatively high accuracy of history-only Model 1 predictions reflects the fact that these neurons show substantial regularity in their ISIs ( average CV = 0 . 54±0 . 02 ) . However , Model 2 provides better spike prediction ( AUC ) in all cases ( paired t-tests , p = 0 . 01 , p<10−5 , p<10−3 ) . Model 1 accounts for auto-correlations fairly well , but the spike history alone cannot account for the cross-correlations between pre- and postsynaptic spikes ( Fig . 2C , black lines ) . Model 2 , using the spike history and coupling parameters together , captures both the auto- and cross-correlations present in the observed spiking ( Fig . 2C , red lines ) . Consistent with results of cross-correlation and area-under-the-curve analysis , the model with coupling provides a better fit to the data than the spike-history alone model when the amplitude of added synaptic input is 0 . 5 σ or larger . For a fixed amount of data ( recording length 200s ) , modeling a synaptic input with amplitude < 0 . 5σ does not improve prediction of the postsynaptic spikes over a spike history-only model ( Fig . 2D ) . Nevertheless , even for these weaker inputs ( 0 . 1 σ , 0 . 2 σ , 0 . 25 σ , and 0 . 3 σ ) the model parameters are able to accurately reproduce the relative amplitude of the presynaptic input ( Fig . 2E ) . Here we use the kernel mean over the first 25ms to summarize the coupling strength in Model 2 ( see Methods ) . Detectability of synaptic connections from spike trains depends strongly on how much data is available . With only a short recording of pre- and postsynaptic spikes it is difficult to determine if two neurons are “connected” . Both Model 1 with spike history-only and Model 2 with coupling tend to over-fit data from recordings shorter than ∼5–10s ( LLR<0 when compared to a homogeneous Poisson model which describes only the baseline firing; Fig . 2F ) . As the amount of available data increases over-fitting is reduced , and , with sufficient data , Model 2 is more accurate than Model 1 if there is truly a synaptic connection between the neurons . We define detection time as the length of data at which Model 2 with coupling provides a better fit than Model 1 with history only ( Fig . 2F , red point ) . For the data in Fig . 2F , where the pre- and postsynaptic neurons each have 5Hz firing rates and the amplitude of the synaptic connection is 1σ , this crossing point occurs around 20s . By varying the input amplitude we find that the recording time needed to detect an input of amplitude x falls off as approximately c / x2 ( Fig . 2G ) , with c = 16 . 0±0 . 1s . In addition to detecting whether two neurons are connected or not , for studying plasticity , it is important to determine whether the strength of a connection changes , for instance , after applying an experimental manipulation , such as high frequency stimulation or a conditioning session . To study the detectability of a change in connection strength , we constructed artificial data sets from the recorded data , in which the strength of the synaptic connection between two neurons changes at time point t , from amplitude a1 , during time period from 0 to t , to amplitude a2 , for time t to 2t ( Fig . 3A ) . As in previous analysis , we use the likelihood ratio to determine whether the synaptic weight has changed . We compare Model 2 , used previously , with a single synaptic input of a constant amplitude throughout the recording time ( 0 to 2t ) to a new Model 3 which allows the synaptic input to differ before and after time t ( Fig . 3B ) . As expected , both Models 2 and 3 accurately account for the auto-correlation statistics and the cross-correlations between pre- and postsynaptic spiking on the full data ( Fig . 3C , left ) . However , only Model 3 can accurately track changes of firing statistics produced by changes of synaptic strength and accurately describe the cross-correlation before and after the change-point t ( Fig . 3C , right ) . As in the case of detecting a connection , the parameters of Model 3 accurately reproduce the amplitude changes in the presynaptic input ( R2 = 0 . 96±0 . 01 , data not shown ) . Detectability of synaptic weight changes in long recordings is comparable to the detectability of connections of constant strength . In 200s long recordings , a likelihood ratio test between Models 2 with constant coupling and Model 3 with variable coupling reveals the change only for Δa > 0 . 25σ ( Fig . 3D , compare to Fig . 2D ) . Again , the ability to detect a change strongly depends on the recording length . With less data available , detecting changes in synaptic strength becomes more difficult than simply detecting a connection . The recording length necessary to detect a change x = Δa of connection strength after a known change-point falls off as approximately c / x2 with c = 21 . 5±0 . 1s ( Fig . 3E ) . In a second experimental protocol , rather than injecting aEPSCs from a single synaptic input immersed in noise , we inject current that is produced by the activity of a large number ( N = 1024 ) of spiking presynaptic neurons ( Fig . 1B ) . We used a presynaptic population consisting of the equal number of excitatory and inhibitory neurons , with log-normal distribution of synaptic amplitudes ( same distribution , positive weights for excitatory , negative weights for inhibitory ) , and PSC kernels consisted of the same difference of two exponentials with time constants of 0 . 5ms and 5ms . The average input had an amplitude of 0 . 15 σ ( corresponding to ∼15pA , depending on σ ) , and membrane potential responses to injection of this current again mimicked the statistics of membrane potential fluctuations in vivo with amplitudes of 15–20mV [8 , 38 , 39] . This paradigm for injection of fully-defined current allows us to examine the detectability of excitatory and inhibitory inputs of multiple amplitudes using the same recording [40] . Using this paradigm we examined 5 additional cells , 3 of which were driven to fire at ∼5Hz , and the other 2 were each driven at several different rates: ∼1Hz , ∼5Hz , and ∼10Hz ( produced by varying the DC offset ) . We then analyze the postsynaptic responses using the same pair-wise model comparison techniques as we used for analyzing single-input experiments . Similar to results from the single input experiments , analysis of the fully-defined input experiments shows that the larger an input is the larger an effect it has on postsynaptic spike prediction , and the more readily it is detected . High-throughput experiments with injection of the fully-defined input allowed us to reveal further features of input detection . The detectability is influenced not only by amplitude , but also by the sign of the input ( excitatory vs inhibitory ) and the postsynaptic spike rates ( Fig . 4A ) . Inhibitory inputs are detected less readily than excitatory . They have less impact on the postsynaptic firing , and thus are less accurate in predicting output spikes compared to excitatory inputs of the same magnitude ( the log likelihood ratios comparing Model 2 with coupling to Model 1 with spike-history alone are 58±2% smaller for inhibitory inputs ) . As in single-input experiments , detection time decreases with amplitude approximately as c / x2 . With these data we find c = 111 , 32 , and 18s for detection of excitatory inputs at 1Hz , 5Hz , and 10Hz output rates and c = 182 , 46 , and 29s for detection of inhibitory inputs at these rates . Across postsynaptic firing rates r , these times are well approximated by c / rx2 ( Fig . 4B ) . On average , inhibitory inputs required 32±3% more data for detection than excitatory inputs . Although detection time likely depends also on presynaptic firing rates as well as time course of PSCs ( not just their amplitude ) , here , for making the comparison clear , presynaptic rates for all inputs were held at 5Hz and PSC kernels had the same time course , differing only in amplitude . In the fully-defined input setting , we can examine , in a single cell , how accurately model estimates of synaptic weights ( of different amplitude and sign ) capture the actual values . The coupling coefficients accurately reconstruct both excitatory and inhibitory input amplitudes over a broad range , and this reconstruction becomes more accurate with higher postsynaptic firing rates ( Fig . 4C ) . The pair-wise cross-correlations ( 0–25ms following presynaptic spikes ) also reflect the input amplitudes fairly well but are much more nonlinear ( Fig . 4D ) . Uncorrected cross-correlations tend to underestimate the relative amplitude of strong inhibitory inputs , but overestimate the relative amplitude of strong excitatory inputs . Next , we ask how modeling multiple inputs simultaneously might affect the detection of individual inputs . Model 2 , with post-spike history and coupling to a single input , can be extended to capture multiple presynaptic inputs simply by adding extra coupling parameters for each additional input . However , modelling all 1024 inputs in detail is computationally prohibitive . Here we consider two alternatives: 1 ) a model with 64 coupling terms where inputs of the same amplitude are grouped together , and 2 ) a bilinear model where the 1024 inputs are restricted to have the same shape . Since the presynaptic inputs in our experiments were specifically grouped together and all have the same PSC shape , these alternative models both provide accurate descriptions of the data ( see Methods ) . It is important to note , though , that these models are simplifications . In typical multi-electrode spike recordings grouping information would not be available , and PSCs at different synaptic connections would have different shapes . As with the pair-wise analysis above , a model that includes multiple presynaptic inputs allows accurate reconstruction of their true amplitudes . The accuracy of the reconstruction for both the grouped and bilinear models increases as more inputs are included in the model ( Fig . 5 ) . With 5% of the inputs observed , the coupling coefficients estimated by the GLM ( average kernel from 0–25ms ) are already well correlated with the underlying input amplitudes ( R2 = 0 . 87 for the grouped model and R2 = 0 . 69 for the bilinear model ) . With all inputs observed this correlation increases to R2 = 0 . 94 and R2 = 0 . 82 for the grouped and bilinear models , respectively . Since the bilinear model has many more parameters , it is not unsurprising that there is more uncertainty in the parameter estimates given the same amount of data ( 200s in this case ) . However , in both the grouped and bilinear models , including multiple inputs provides more accurate reconstruction than pairwise model fit to the same data ( Fig . 5B ) . The increased accuracy in estimating synaptic weights using models of all inputs suggests that postsynaptic spikes might be more readily associated with or disassociated from the spiking of individual presynaptic inputs . Previous methods for detecting synaptic inputs , by using pairwise spike statistics alone , do not take advantage of this additional predictability . For instance , in the traditional , non-parametric tests the cross-correlation between spiking of two neurons is compared to the cross-correlation obtained when spike timings of the presynaptic neuron are shuffled ( Fig . 6A ) . Using boot-strapping we can create a distribution of the total spike count in a cross-correlation over a window following presynaptic spikes ( Fig . 6B ) . Comparison of the distributions obtained using the observed vs shuffled spike trains allows us to test whether an input has a statistically significant effect on the firing of the postsynaptic neuron ( see Methods ) . As with the model-based methods , we can vary the amount of data ( recording length ) used for computing this statistic , and determine a minimal recording time necessary for the detection that , as expected , depends strongly on the amplitude of an input ( Fig . 6C ) . Using data from experiments with fully-defined input we find that , even after optimizing the detection window for the nonparametric test ( Fig . 6D ) , the pair-wise , model-based method yields slight but significant reductions in detection time ( to 94±2% on average compared to the nonparametric test with 15ms time window , paired two-sided sign test p = 0 . 005 ) . A grouped model that includes all inputs yields additional reductions in detection time ( to 86±4% ) compared to the nonparametric approach ( paired two-sided sign test p<10−5 ) . Using all inputs in the grouped model also reduces detection times relative to the pair-wise , model-based tests , where spiking of only one presynaptic neuron is included in the model , while all other inputs are treated as noise ( to 94±2% , paired two sided sign test p = 0 . 04 ) . One reason for the faster detection of connections between neuron pairs by model-based inference than by cross-correlation based non-parametric tests might be that the non-parametric test requires us to specify a fixed interval for counting spikes on the cross-correlogram ( 0–15ms is optimal here ) . Making this interval too long or too short will reduce the sensitivity of the test ( Fig . 6D ) . However , the model-based methods are more robust to these differences , since the coupling kernel can flexibly account for changes in postsynaptic spike statistics produced by inputs with different time-courses . This advantage may be especially useful for reconstructing connectivity from in vivo data , where the shape of PSCs varies broadly across connections . Indeed , previous work has shown how model-based methods can produce a more accurate picture of connectivity compared to descriptive ( cross-correlation ) methods [22] . These results obtained in a fully-defined experimental setting illustrate how model-based methods can be applied to improve input detection over descriptive methods , as well . In addition to examining the detectability of individual inputs , the fully-defined input protocol allows us to extend the basic GLM framework to predict postsynaptic spiking based on recent activity of multiple presynaptic inputs . The pairwise analyses above indicate that the spike times of individual presynaptic inputs are useful in predicting postsynaptic spikes . Since inputs are independent , observing more presynaptic neurons should improve the overall model accuracy . Because spike times of all N = 1024 excitatory and inhibitory presynaptic neurons composing the total input to the postsynaptic neuron ( Fig . 7A ) are defined , we can analyze how increasing the fraction of observed inputs improves the prediction of postsynaptic spikes . The post-spike history term always plays a large role in predicting postsynaptic spikes ( e . g . Fig . 7B , top ) . This term describes the strong influence that generation of an action potential has on the probability of future spikes . These dynamics are intrinsic to the spike generation mechanism and are present no matter how many inputs are included in the model . With a small number of observed inputs the history term strongly dominates the prediction of spikes . Fig . 7B shows the time course of changes of gain that are due to the post-spike history term and due to the influence of N = 4 inputs included in the model for a typical neuron fit using the grouped model ( middle plot ) . A linear combination of these few sparse , excitatory ( spikes indicated by small bars above the trace ) and inhibitory ( spikes below the trace ) inputs has only weak influence on the spiking , and can account for only a small portion of the variations in spike timing . However , as the number of inputs included in the model grows , their total contribution becomes comparable to that of the post-spike history term ( Fig . 7B , bottom ) . Although the model used here does not explicitly describe the underlying fluctuations in membrane potential that result from the current injection , the contribution of the cumulative coupling terms of all inputs ( N = 1024 , red trace in Fig . 7B , bottom ) tends to be correlated with the total injected current ( Fig . 7B , bottom , gray trace , R = 0 . 54 for this example ) . As more inputs are observed , overall spike prediction becomes more accurate . Fig . 7C shows a combined influence of the post-spike history term and an increasing number of inputs ( N = 4; 256; 512; 1024 ) on the probability of postsynaptic spiking . With few inputs , the postsynaptic firing is dominated by post-spike effects and timing of postsynaptic spikes is predicted relatively inaccurately ( AUC = 0 . 78 for this example with 4 inputs ) . With an increasing fraction of inputs included in the model , their contribution to spike prediction increases , and timing of individual spikes is predicted with progressively increasing accuracy ( AUC = 0 . 82 , 0 . 86 , and 0 . 98 for 256 , 512 , and 1024 inputs , respectively ) . When all inputs are observed , the model spike probability tends to be sharply peaked around actual spike times ( vertical lines in Fig . 7C ) , and the coupling effects contribute substantially to the rate variation ( 81% of the variance in the log rate is due to coupling ) . These qualitative results are corroborated by quantitative analysis of the contribution of the different model components to spike prediction . When only few synaptic inputs are included in the grouped model the post-spike history accounts for nearly all of the variability in the firing rate ( Fig . 8A ) . As more inputs are modeled the contribution of coupling becomes progressively stronger , and , in the grouped model , with inclusion of ∼70–90% of the inputs it becomes as important as the intrinsic dynamics of spike generation described by the post-spike history ( Fig . 8A ) . Note that the effect of individual inputs remains weak ( Fig . 8A , bottom ) , but the total contribution from all inputs becomes substantial . Similarly , as more inputs are included in the grouped model , the contribution of the coupling terms becomes increasingly correlated with the injected current ( Fig . 8B , black curve ) . Since the total current is simply the sum of all PSCs produced by all presynaptic inputs , we can also extract the “partial” current that is the current contributed only by a fraction of the observed neurons . This partial current is fairly consistently correlated with the coupling term ( Fig . 8B , red curve ) . The reduced correlation of both the total and the partial current with the coupling term when the fraction of observed inputs approaches 100% appears to be due to the differences in the timescales of the coupling terms under the grouped model . The coupling term tends to follow the injected current with a slightly longer timescale as more inputs are grouped together ( Fig . 8B , inset ) . Including larger number of inputs in the model improves spike prediction quantitatively , as well . Log likelihood ratios ( relative to a homogeneous Poisson model ) increase monotonically with the increasing fraction of observed inputs ( Fig . 8C ) . Note that , unlike in vivo data where the LLR saturates as more neurons are observed [26] , here the model accuracy shows no signs of saturation . This is likely due to the fact that the synaptic inputs simulated here are independent , while firing of presynaptic neurons in vivo can be much more correlated . Using ROC analysis we find that cross-validated area-under-the-curve increases from 0 . 72±0 . 01 when 5% of the inputs are observed to 0 . 98±0 . 01 in the model that includes all ( 100% ) inputs ( Fig . 8D ) . Assessing spike prediction accuracy with single trial data is not always intuitive . Here , an area-under-the-curve of 0 . 98 signifies that a randomly chosen time bin containing a spike in the recorded response will have a higher p ( spike ) than a randomly chosen non-spike bin 98% of the time . To unpack the relationship between the post-spike history and presynaptic input as more inputs are included in the model we compare observed responses of a neuron to a repeated fully-defined input stimulus and responses of a grouped model to the same stimulus ( Fig . 9 ) . Here we simulate a neuron receiving the same presynaptic inputs on each trial and use the simulated post-spike history to generate model responses . Simulating multiple trials , we find that there is substantial trial-to-trial variability when only a small fraction ( 5% ) of the inputs is used to generate the model response ( Fig . 9 , top ) . With this small number of inputs , the average response only weakly matches the peri-stimulus time histogram ( PSTH ) of observed spike responses ( R = 0 . 09±0 . 02 , PSTHs estimated by a maximum likelihood , fixed bandwidth kernel density estimator ) . When all inputs are modeled , trial to trial variability is substantially reduced ( Fig . 9 , middle ) , and the simulated PSTH more accurately matches the observed responses to repeated trials ( R = 0 . 73±0 . 06 ) . Similarly , the bilinear model reproduces the PSTH with R = 0 . 10±0 . 02 and R = 0 . 55±0 . 04 when 5% and 100% of the inputs are observed , respectively . Even when all inputs are modeled , the GLMs do not capture the full precision of the data ( Fig . 9 , right ) . These results suggest that , although single trial spike prediction with the GLM is quite accurate , more precise models , with additional nonlinearities [41 , 42] or explicit estimates of the underlying membrane potential [43] , might be necessary to provide a full account of the transformations that occur as fluctuating current input leads to spiking output . One potential challenge in generalizing the results of in vitro current-injection experiments to data collected in vivo is that these experiments omit several important features of real synaptic integration . In vivo , nonlinear dendritic integration , probabilistic release , and plasticity will all introduce certain variability in transmission at a synapse , making it more difficult to detect the connection and to estimate its strength . It is also important to note that here we use current-based synaptic input , while at real synapses currents are generated by changes of conductance . This causes the PSC amplitude to vary as a function of the membrane potential . To test whether the nature of current-based and conductance-based inputs could have a large effect on their detectability and estimation of their strength , we used post-synaptic spikes generated by two model neurons with either current-based or conductance-based synapses , and fit pairwise models to these new data . To simulate postsynaptic neurons that match the Layer 2/3 cells recorded in vitro , we fit two adaptive exponential integrate-and-fire models: one with current-based synapses and one with conductance-based synapses [44] . Using the same presynaptic spike times and weights delivered to the observed neurons , we then optimize the parameters of these models to match both the observed membrane potential and spike timing ( Fig . 10A , see Methods ) . After optimization , the sub-threshold fluctuations of the models reproduce the observed fluctuations to 4 . 1mV RMSE – with similar accuracy between current-based and conductance-based models . Overall spike statistics ( post-synaptic inter-spike interval distributions and cross-correlations ) were well-matched between the models and data ( S1 Fig . ) , and the spike coincidence factor [45] with Δ = 4ms averaged Γ = 0 . 35 for the current-based models and with the conductance-based models performing slightly worse at Γ = 0 . 31 ( S2 Fig . ) . Using the simulated spike trains from the two adaptive exponential integrate-and-fire model neurons , we performed the same pair-wise analyses for estimating synaptic amplitude and detection time that we used for the real data ( Fig . 10B-D ) . As with the experimental data , we find that , in both the current-based and conductance-based models , higher amplitude synapses are easier to detect and there is an asymmetry between excitatory and inhibitory inputs with inhibitory inputs being more difficult to detect than excitatory inputs of the same magnitude . As before , the postsynaptic firing rate has a large effect on detectability and estimation accuracy with higher rates resulting in faster detection of inputs . Estimation accuracy for the current-based model , in this case , was R2 = 0 . 85 for 1Hz output , 0 . 93 for 5Hz output , and 0 . 96 for 10Hz output ( for clarity only 5Hz condition is shown in Fig . 10C ) . The conductance-based model yields similar results with R2 = 0 . 84 , 0 . 93 , and 0 . 96 for the three output rates . Finally , we find that detection times decrease as ∼c/x2 with increasing input amplitudes , and are shorter for higher post-synaptic firing rates . Together , these results show that the form of synaptic input – conductances vs currents – does not substantially alter our results once spike statistics are matched .
Here we use in vitro current injection experiments to assess how well functional connectivity inferred from spikes corresponds to underlying synaptic inputs and to examine the limits of postsynaptic spike prediction from presynaptic spike times . Our results show that 1 ) Simulated synaptic inputs with amplitudes >0 . 25σnoise can be reliably inferred from several minutes of spiking activity of pairs of neurons . Stronger connections can be detected even faster , but the amount of data required for detecting weaker connections increases rapidly . Thus , in typical experiments only a subset of connections can be detected , with a low amplitude limit depending on recording time and firing rate . 2 ) The inferred coupling strength is strongly correlated with the true amplitudes of synaptic inputs . 3 ) Excitatory inputs are more readily detected than inhibitory . 4 ) Detectability of changes of synaptic weights follows same rules and has same limitations as detection of individual synaptic connections . 5 ) Observing and modeling larger fractions of the presynaptic input improves model accuracy and reduces the necessary recording time for detecting inputs . Thus , model-based inference of functional connectivity allows accurate detection of individual inputs , reconstruction of synaptic amplitudes , and prediction of postsynaptic spiking . Experiments with injection of currents composed of mixtures of input signals provide a flexible experimental framework for exploring how synaptic input is translated into output postsynaptic spiking . These recordings provide a link between properties of synaptic connections typically measured in intracellular experiments in vitro , such as PSC amplitude , and estimates of connectivity made from in vivo extracellular recordings , such as inferred functional connectivity and coupling strength . We have used a real spike generation mechanism of an actual neuron to study spike responses to artificial PSCs with realistic time courses , distribution of weights and firing statistics of presynaptic neurons [46 , 47] , consistent with experimental observations . Injected currents induced membrane potential fluctuations typical for in vivo activity [8 , 48 , 49] . However , we have also made some simplifying assumptions and neglected to model many known effects , such as nonlinear dendritic integration [50] , short-term plasticity [51] , possible input correlations [52] , probabilistic release , channel noise [53] , and the fact that actual synapses are conductance-based [54 , 55] . Since these effects can have a substantial impact on spike statistics , future experimental work and statistical modeling may yield improvements over the current methods . These additional experimental parameters will dramatically increase the complexity of future experimental results and their interpretation . Here we have neglected these effects in order to simplify our analyses and determine the basic constraints of what can be inferred about simulated synaptic connectivity from spiking of neuronal ensembles . Although these experimental simplifications may limit the extent to which our results can be directly generalized to extracellular recordings , the trends described here provide concrete links between synaptic currents and the detectability of synaptic inputs . Applying model-based methods to spike trains generated by model neurons fit to our observed data , we found that our results are robust to both the form of input ( conductance-based vs current-based synapses ) and the spike generation mechanism ( integrate-and-fire vs biophysical in real neurons ) . In addition to these physiological caveats , it is important to note that there is also room for improvement in modeling . The model-based methods used here are based on a tractable class of rate models and fail to describe some fundamental aspects of spike statistics . Here we used a common GLM approach where neurons are assumed to emit spikes according to a Poisson random variable with a rate that is determined by recent pre- and postsynaptic spiking . While these models can describe the auto- and cross-correlations present in the data , neurons often have more complex nonlinear behavior [56 , 57] and generate spikes much more reliably than Poisson processes [58] . In general , rate models may fail to capture some of the important nonlinear dynamics of real neural systems [59] . However , several statistical models have been developed that explicitly aim to describe the underlying membrane potential dynamics [60 , 61] and tend to yield more accurate spike prediction . Although inference from spike trains is more difficult in these models , they are able to model the high reliability of spikes in many experimental settings [43] and provide detailed insight into the membrane dynamics and changing conductances underlying the observed spiking [62] . Including coupling between neurons in these voltage-based models , as latent synaptic conductances , may lead to improved detection , estimation , and tracking of inputs from spikes . Despite these caveats , GLM-based estimation of functional connectivity offers some important advantages over previous methods . Descriptive statistics of functional connectivity [18 , 19 , 63 , 64] and recent extensions [65] are widely applied tools for understanding interactions between neurons . However , as many previous studies have noted , correlations that are not due to direct monosynaptic connections can confound estimates of connectivity and compromise the interpretation of cross-correlograms [20 , 29] . Common-input , the dynamic effects of post-spike history on future spikes , poly-synaptic effects , and changing synaptic weights will all affect basic pair-wise spike statistics . One key advantage of statistical inference model-based methods , over descriptive techniques , is that the many factors contributing to postsynaptic spiking can be modeled simultaneously and , to some extent , disentangled . Our initial results show that estimates of synaptic coupling by statistical-inference model are sensitive to changes in synaptic strength , and thus can provide a tool for assessing plasticity of synaptic connections from neuronal spiking . We demonstrated that detecting a synaptic weight change that might occur after a specific experimental intervention , such as tetanization , requires only moderately more data than detecting the presence of a connection . However , statistical methods that can infer synaptic changes from extracellularly recorded spike trains could also be applied to situations where change-points are unknown or not defined , for instance , when plasticity is a result of ongoing natural activity . New statistical methods are beginning to address these issues by explicitly modeling ongoing plasticity [66 , 67] , considering state-changes with unknown change points [68] , and providing more flexible descriptions of time-varying neural dynamics [69–72] . In our analysis of stationary coupling , the use of a broad range of synaptic amplitudes composing the injected currents allowed us to test the limits of detectability and determine how much data is necessary to detect the effect of presynaptic spiking on postsynaptic spike statistics . We find that detection time for an input of amplitude x is approximately proportional to 1 / x2 and also depends on firing rates . Importantly , inhibitory inputs are more difficult to detect than excitatory inputs of the same amplitude , and inhibition required ∼30% more data for detection with pair-wise tests . We further show that when an increasing fraction of presynaptic neurons can be included in the model , spike prediction accuracy increases and less data is necessary for detection of individual connections . Here we found that with ∼15min of data and the pre- and postsynaptic cells firing at ∼5Hz , synapses could only be reliably detected when they had amplitudes > 0 . 25σ . This corresponds to PSCs amplitudes of 18–27 pA , which is in the range of average EPSCs amplitudes in L2/3 cells ( 15–25 pA , [73] ) . These results suggest that , to fully map functional connectivity estimated from spikes to actual physiological connectivity , longer recording lengths may be necessary to detect weak connections . At the same time , pairs of neurons with strong synapses or high firing rates may be evident even in short in vivo recordings . Previous work has shown that modeling partial , putative presynaptic input can improve both spike prediction and decoding [26 , 30 , 74] . Here we consider a limiting case where the input is fully controlled during the experiment . We find that a basic rate model ( GLM ) can provide surprisingly accurate spike predictions and approximately capture the true input current . These findings suggest that models of functional connectivity may begin to provide insight into actual synaptic effects as modern techniques allow recording of simultaneous spiking from increasing numbers of neurons in vivo . In vitro current injection has been used to study temporal integration of presynaptic spikes [3 , 37] , the effects of input synchrony on spiking [75] , and signal propagation through networks [76] . Here we focused on the detectability and identification of individual inputs from simultaneously recorded spike trains . Our results highlight some of the important scientific possibilities offered by the use of statistical methods for understanding large-scale interactions between neurons . Modeling the interplay between post-spike history effects and presynaptic input will be essential to describe biophysical diversity within [77 , 78] and across [79 , 80] cell types . Extending both the statistical and experimental tools used here will provide a deeper understanding of the transformations of synaptic inputs into a neuron’s spiking output .
All animal use procedures were approved by the institutional animal care and use committee at the University of Connecticut , and conform to the principles outlined in the Guide for the Care and Use of Laboratory Animals ( National Institutes of Health publication no . 86-23 , revised 1985 ) . In vitro intracellular recordings were made from layer 2/3 pyramidal neurons in slices of rat visual cortex ( Wistar rats , P17-P30 ) , as described in previous work [3 , 39] . The solution used during the preparation of the slices and during the recordings contained ( in mM ) 125 NaCl , 2 . 5 KCl , 2 CaCl2 , 1 MgCl2 , 1 . 25 NaH2PO4 , 25 NaHCO3 , 25 D-glucose and was bubbled with 95% O2 and 5% CO2 . Whole-cell recordings were made with patch electrodes ( 4−6 MΩ ) filled with K-gluconate based solution ( in mM: 130 K-Gluconate , 20 KCl , 4 Mg-ATP , 0 . 3 Na2-GTP , 10 Na-Phosphocreatine , 10 HEPES ) using the bridge mode of a Dagan BVC-700A amplifier ( Dagan Corporation , USA ) . Data were digitized at 20 kHz ( Digidata 1440A , Molecular Devices , USA ) and stored for further processing . To systematically study the detection and tracking of synaptic weights we injected neurons with partially-defined or fully-defined synthetized current . 1 ) Partially-defined current was produced by a single simulated , excitatory presynaptic input immersed in fluctuating noise . Several amplitudes of artificial EPSCs were used in these experiments . 2 ) Fully-defined current was produced by the firing of a large population of simulated presynaptic excitatory and inhibitory neurons whose postsynaptic currents ( PSCs ) sum to mimic fluctuating , naturalistic input . Current for injection in this first set of experiments was composed of 1 ) a fluctuating component ση ( t ) , where η ( t ) is a standardized ( zero mean , unit variance ) Ornstein-Uhlenbeck ( OU ) process with a correlation time of τ = 5ms rescaled to have standard deviation σ , 2 ) artificial EPSCs of several different amplitudes: 0 . 1 , 0 . 2 , 0 . 25 , 0 . 3 , 0 . 5 , 1 . 0 and 1 . 5 of the noise standard deviation σ , and 3 ) a DC component tuned to maintain a desired firing rate , around ∼5 Hz . Seven cells were injected with ( 0 . 5 , 1 . 0 , 1 . 5 σ inputs ) , one cell with ( 0 . 2 , 0 . 3 , 0 . 5 , and 1 σ inputs ) , and two cells with ( 0 . 1 , 0 . 25 , 0 . 5 , 1 . 0 σ inputs ) . This approach was similar to that used in previous work [2 , 3] . Note that this approach based on injection of current through the intracellular electrode ignores the transformation of synaptic conductances into postsynaptic currents , as well as effects of dendritic integration of synaptic inputs . However , the injected fluctuating current reproduces well the membrane potential fluctuations recorded in the soma of neocortical neurons in vivo [8 , 48 , 49] . Presynaptic spike timing for the artificial EPSCs was generated by a gamma renewal process ( shape k = 2 , scale θ = 2 . 5 ) . This corresponds to a neuron firing at a rate of 5Hz with more regularity ( CV ≈ 0 . 7 ) than a Poisson process . Spikes were then convolved with a synaptic integration kernel generated by a difference of exponentials – rise time of 1ms and decay time of 10ms – to generate an EPSC trace . This kernel was previously used in [3] . In vivo , there is substantial variation in spike statistics [80 , 81] and dendritic integration [82 , 83] . Although a complete exploration of the parameter space is beyond the scope of these experiments , the values used here were chosen to be consistent with in vivo and in vitro observations . Current was injected in episodes of 46s , with intervals of 40–90s between the episodes . We recorded from 10 cells with 10–60 episodes recorded from each cell . The analysis ( described below ) assumes that the responses are stationary , such that the files can be reordered and concatenated without loss of generality . That is , if episodes are recorded with aEPSC amplitudes 0 . 1 , 0 . 2 , 0 . 5 , 0 . 2 σ , we assume that we can examine an amplitude change of 0 . 4 σ by concatenating files 1 and 3 and that concatenating files 2 and 4 is equivalent to a single longer recording with 0 . 2 σ input . In all cases , the intervals between episodes , where there was no postsynaptic spiking , as well as the first and last 3s of each record were excluded from the analysis . In this first set of experiments we do not consider the effects of IPSCs . In a second set of experiments , rather than injecting current composed of a single input plus noise , we aim to model complete presynaptic input by simulating the spiking of large number of presynaptic neurons . Several factors determine the statistics of the injected current: 1 ) the statistics of presynaptic spiking , 2 ) the amplitudes of synaptic “weights” , and 3 ) the time-course of the PSCs . Here we make several simplifying assumptions . We assume that population of presynaptic neurons consists of an identical pool of gamma renewal processes , 50% of which are excitatory and 50% of which are inhibitory , and we assume a log-normal distribution of synaptic weights based on observations from paired in vitro cortical recordings [46] . Finally , we generated the PSCs as a difference of exponentials with a rise time of 0 . 5ms and decay time of 5ms . Thus , the model parameters are N ( the number of presynaptic neurons ) , k and θ ( the shape and scale parameters for the homogeneous Gamma renewal processes ) , μ and σ ( the shape and log-scale parameters of the log-normal amplitude distribution ) , and τ1 and τ2 ( the time constants of the artificial PSCs ) . Here we use N = 1024 with amplitudes drawn pseudo-randomly from a discrete approximation to the log-normal distribution described in [46] ( μ = 0 . 702 , σ = 0 . 9355 ) . Absolute PSC amplitudes are constrained to 32 discrete , log-spaced values to simplify the analysis , and the proportion of amplitudes in each bin is exact to maintain excitatory-inhibitory balance . Excitatory and inhibitory inputs are assumed to have the same distribution , differing only by the sign . We use k = 2 , θ = 2 . 5 for the gamma renewal processes and resample short ( <10ms ) ISIs to avoid strongly overlapping PSCs . It is important to note that these artificial PSCs only partially capture the dynamics of real synaptic currents , which exhibit synaptic noise , as well as , short-term depression and facilitation . Moreover , this basic population spiking model does not capture all of the details of cortical circuitry . There are known differences in the spike statistics and time course of PSCs for different cortical cell types [80] . However , this model maintains balanced , approximately OU statistics and allows us to examine the detectability of a broad range of synaptic weights . The simulation parameters here were chosen with the goal of making inputs physiologically realistic , but structured enough to analyze statistically . Using these parameter settings we recorded an additional 5 cells where the DC component was tuned to maintain a desired firing rate . Three cells were recorded around ∼5 Hz , while two cells were driven at multiple firing rates: ∼1 Hz , ∼5 Hz and ∼10 Hz . For detection of synaptic inputs from spiking , we use a basic statistical model that describes the probability of postsynaptic spikes as a function of the presynaptic spike timing – a generalized linear model ( GLM ) with Poisson observations . For a single presynaptic input we assume We aim to predict postsynaptic spiking npost ( t ) , assuming that spikes are generated as Poisson random variables with a rate λ ( t ) . This rate is determined by a linear combination of a baseline firing rate b0 , the influence of spiking history , that is the previous spikes of the postsynaptic neuron , parameterized by bpost , and a term that describes the effect of the input from presynaptic spikes , parameterized by bpre . A set of basis functions fk provides a smooth expansion of each of these effects . The parameters of this model can then be readily estimated by maximizing the log-likelihood [84] b^MLE=arg maxb∑t ( n ( t ) logλ ( t ) −λ ( t ) ) . ( 2 ) In the results presented here we often summarize the effect of coupling with a single “coupling weight” , defined as the average kernel over the first 25ms: 〈∑k=1Kbkfk ( τ ) 〉0<τ<25 . . This basic GLM can then be extended to include multiple inputs by using In this case , with O ( kN ) parameters , rather than straight-forward maximum likelihood estimation , regularization is important to prevent over-fitting . Here we use L1-regularization and find the maximum a posteriori ( MAP ) parameters estimates The regularization hyper-parameter is optimized by maximizing cross-validated log-likelihood , and no regularization is performed on the baseline or post-spike history terms . Here we use 10-fold cross-validation both to optimize the hyper-parameter and evaluate model accuracy ( see below ) . With L1-regularization we assume that input weights are exponentially distributed a priori , and this assumption is useful in that it prevents overfitting and leads to sparse solutions where many input weights bi , k are set to 0 . This assumption is useful for analysis of in vivo data , since only a small number of potential presynaptic inputs are likely to actually be connected to the postsynaptic neuron , so such sparse solutions are desirable . Other regularizers , such as spike-slab , group-L1 , or even the log-normal prior used to generate the input weights here , may produce more accurate estimates . However , the fact that the log-posterior is concave with L1-regularization provides the additional advantage of a unique , global optimum . For both MLE and MAP ( L1-regularized ) optimization we use scaled sub-gradient maximization using the L1General toolbox [85] . Even with regularization , fitting a model with 1024 inputs is computationally prohibitive . Here we use two alternative simplifications: 1 ) a grouped GLM where inputs of the same weight are grouped into a single covariate , and 2 ) a bilinear extension of the GLM where the inputs are constrained to take the same shape . The grouped model takes advantage of the fact that the simulated amplitudes are highly structured . Namely , groups of presynaptic neurons have identical amplitudes distributed according to a discrete approximation of the log-normal distribution [46] . We can thus collect inputs into groups to simplify the analysis from Eq . ( 3 ) – instead of N = 1024 simulated presynaptic input neurons , we can study an equivalent system with only N = G inputs , where G is the number of amplitude “groups” ( G = 64 used in our simulations ) by combining spike trains Fitting a GLM for this reduced set of inputs as λ ( t ) =exp ( b0+∑k=1Kbpost , kfk ( npost ( t ) ) +∑g=1G∑k=1Kbg , kfk ( ng ( t ) ) ) is much more computationally tractable than modeling all individual inputs when N ≫ G . Here we make use of this group simplification to examine how detectability and overall model accuracy are affected by the fraction of observed inputs . In this case we randomly sample a fraction p of the total inputs N and include these “observed” presynaptic spikes in a grouped-GLM as described above . Note that , in most experimental settings , such group membership would be unknown , and a large number of unconnected pairs would be likely present . The results from the grouped model likely underestimate parameter uncertainty and overestimate spike prediction accuracy compared to the full estimation problem . However , accurately estimating connectivity using models with O ( N ) parameters is not a trivial problem for N>1000; structured regularization techniques [86] and detailed model comparison [34 , 87] will be essential as the number of connections in these types of models increases . Here as a point of comparison for the grouped model we also use a bilinear model that models individual inputs with separate parameters , but restricts all inputs to have the same shape . Here , rather than K parameters per input , there is only a single “weight” parameter wi per input . In this case we fit the model by coordinate ascent – alternating between maximizing the posterior with respect to the input weights while holding the shape constant and maximizing the likelihood ( no regularization ) with respect to the shape while holding the weights constant . This approach tends to work well in practice [66 , 88] , and effectively approximates the full model , which would have O ( Nk ) parameters , with a model with O ( N+k ) parameters . For all models we assume the basis {f1…fK} to be a set of gamma distributions tn exp ( −t / m ) / Z where n = {0…K −1}s and Z denotes normalization by the maximum . For the post-spike history term and coupling terms during pair-wise analyses we use m = 25ms , s = 1 . 2 , and K = 8 . In the full model , this same K = 8 basis is used for the post-spike history , but for simplicity coupling terms use a reduced basis with m = 45ms , s = 1 . 5 , and K = 4 . These basis sets allow us to fit a variety of band-limited , smooth functions on a range ∼0–500ms with faster variation near 0ms . Similar bases have been used in previous work [30 , 86] . Together with the baseline parameter , these settings give 9 total parameters for Model 1 , 17 for Model 2 , 25 for Model 3 , 265 for the full grouped model with all inputs , and 1037 parameters for the full bilinear model with all inputs . Given this model framework , we can now examine whether the effect of presynaptic spiking on postsynaptic spiking is detectable . In general , if we have two models H1 and H2 with Poisson observations the log likelihood ratio is given by LLR ( H1 , H2 ) =[∑tn ( t ) logλ1 ( t ) −λ1 ( t ) ]−[∑tn ( t ) logλ2 ( t ) −λ2 ( t ) ] ( 5 ) where the two models have conditional intensities defined by λ1 and λ2 ( log base 2 is used LLR ( H1 , H2 ) * log2 when reporting bits ) . Importantly , the log likelihood ratio quantifies the relative accuracy of the two models . For instance , when H2 is a homogeneous Poisson model that only describes the mean firing rate , the log likelihood ratio quantifies how much more accurately spikes are predicted by the model H1 over just predicting the mean . An important concept in functional connectivity analysis is whether or not an input is “detectable . ” One approach is to define an input as detectable if the effect bpre , from the model in Eq . 1 , results in a cross-validated log likelihood ratio >0 when compared to the nested model with bpre = 0 . A second approach is to use an explicit likelihood-ratio test ( without cross-validation ) . This test makes use of the fact that , for nested models , log likelihood ratios are approximately χ2 distributed with df2 – df1 degrees of freedom when df1 and df2 denote the number of free parameters in H1 and H2 , respectively . We then consider the effect of the presynaptic input to be statistically significant when the χ2-test gives p < 0 . 05 . Here we base “detection time” on the minimum amount of recorded data ( averaged over blocks length T ) needed to satisfy the cross-validated LLR criterion . In deciding whether a connection is present or not , it may be useful to compare the assumptions of the cross-validated log likelihood ratio and the un-cross-validated , explicit likelihood-ratio test . Especially for models with many parameters ( i . e . the bilinear model with many presynaptic inputs ) , cross-validation is essential to avoid over-fitting . Even though estimating confidence intervals for test-set log-likelihood is problematic [89] , defining “detectability” qualitatively , based on generalization performance , is more appealing in larger models . Here we use a convention of reporting and plotting the cross-validated LLR even for simple , pairwise models . In evaluating statistical significance , we limit our analysis to the pairwise models and use the un-cross-validated likelihood-ratio test . The likelihood ratio provides one , perhaps unintuitive , notion of accuracy . Although the models used here assume Poisson observations , potentially allowing >1 spike per bin , we can also perform ROC ( receiver operating characteristic ) analysis by limiting our predictions to no-spike p ( n ( t ) = 0|λ ( t ) ) vs spike ( 1 – p ( n ( t ) = 0|λ ( t ) ) ) classification . A bin size of 1ms is used for all analysis , such that , in practice , more than one postsynaptic spike per bin never occurred . The model , tests for detectability , and accuracy analyses described above are generic tools for model-based estimation of spike train statistics and have been applied to a variety of in vivo extracellular data [22] , in which the underlying synaptic weights are unknown . The data collected here , with the controlled synaptic input , offer several unique opportunities to study how functional connectivity estimation is related to synaptic input . In addition to comparing the estimated model parameters to the known PSC amplitudes and comparing components of the model to the injected current , we can also examine the detection of changes of synaptic inputs . We assume that responses to injected current containing PSCs of different amplitudes can be concatenated in order to study putative changes in synaptic strength . As with the detection of connections , here we define a weight change as detectable if modeling the effects bpre , t<t′ and bpre , t>t′ around a known change-point t′ results in a cross-validated log likelihood ratio >0 when compared to a model with a single coupling effect bpre for all observations . The model-based tools described so far provide a flexible framework for studying detectability of functional connections . However , one traditional method for determining whether a functional connection is present is to examine the cross-correlations between two spike trains . Using boot-strapping we can generate a distribution that reflects the expected rate change in the postsynaptic firing following a presynaptic spike , as well as the uncertainty in those rate changes . We then construct a null distribution by measuring the cross-correlations in data where the presynaptic ISIs have been shuffled . By taking differences between samples of the observed and shuffled distributions we can then estimate a p-value: the probability of observing , by chance , a rate change at least as extreme as the one actually observed . To compare this nonparametric test with the model-based framework , we define a test statistic based on the average cross-correlation in a range of lags 0 ≤ τ ≤ τ′ ms following the presynaptic spikes After computing this statistic for bootstrap samples of the observed data robs and data with shuffled ISIs rshuff , we can test against the null-hypothesis that there was no difference between observed and shuffled statistics . The null hypothesis is rejected at a confidence level α , when P ( robs – rshuff > 0 ) > 1 – α / 2 or 2 P ( robs – rshuff > 0 ) < α / 2 . Here we use τ′ = 15ms with 1ms bins and α = 0 . 05 . To compute detection times we use 4096 total samples for each distribution with recording length T , aggregated over 16 randomly selected time blocks . Our in vitro data were collected by injecting current into L2/3 pyramidal cells , where the current is generated by a sum of fixed post-synaptic currents from simulated pre-synaptic neurons . Since real synaptic currents are generated by changing conductances , it is possible that our detection results overestimated the detectability and estimation accuracy of synaptic weights from spikes . To examine this potential confound we simulate an adaptive exponential integrate-and-fire model neuron in two situations: receiving either current-based or conductance-based synaptic inputs . The adaptive integrate-and-fire model takes the general form Where the dynamics of the membrane potential V depend on the capacitance C , leak conductance gL , resting potential EL , an adaptation variable w , DC current input I0 , and fluctuating synaptic currents Isyn ( t ) . The behavior of the membrane near the spike threshold VT is determined by the exponential nonlinearity gLΔTexp ( V−VTΔT ) , and the adaptation variable has its own dynamics determined by τw and a . As with other integrate-and-fire models , spikes occur when the membrane potential crosses a threshold VT , after which the potential is immediately reset to Vreset . After spiking the adaptation term is also updated instantaneously with w ← w + b . Here the exponential nonlinearity and adaptation term allow us to better describe the post-spike and near-threshold behavior of the observed neurons . To examine the effects of current-based and conductance-based inputs we use the same set of simulated pre-synaptic spikes as were delivered to the observed neurons . For current-based inputs we know Isyn ( t ) exactly . Given the simulated presynaptic spike times ni ( t ) , PSC kernel p , and weights for each synapse vi these currents were generated as To modify the synaptic inputs for changing conductances we split the inputs into excitatory and inhibitory contributions with two separate reversal potentials The conductances are then given by gE ( t ) =aE∑i∈Evi∑τ ( ni ( t ) *p ( t−τ ) ) and gI ( t ) =aI∑i∈Ivi∑τ ( ni ( t ) *p ( t−τ ) ) where two additional parameters aE and aI weight the excitatory and inhibitory contributions . Here we choose standard values for EL = −70mV , EE = 0mV , and EI = −90mV and optimize all other parameters to match the membrane potential fluctuations and spike timing of the observed neurons . For the current-based models this formulation leaves the parameters {C , gL , VT , Vreset , ΔT , τw , a , b} to be optimized . For the conductance-based models , since we want the average PSC for each presynaptic input to match the original inputs , we introduce an additional constraint that the excitatory and inhibitory input be balanced: aI < V ( t ) – VI > = – < V ( t ) – VE > aE . Since Isyn ( t ) has 0 mean , we approximate the expectations using the equilibrium voltage far from threshold: Veq = VL + I0 / gL . Given this constraint we have the parameters {C , gL , VT , Vreset , ΔT , τw , a , b , aE} for the conductance-based model . For both the current-based and conductance-based inputs we then optimize model parameters to match the observed voltage fluctuations and spike timing using derivative-free search ( Nelder-Mead ) with random restarts . Specifically we aim to minimize a cost function that mixes voltage and spike timing terms Here the first term allows us to minimize the discrepancy between the observed and simulated membrane potential ( neglecting voltage changes within 5ms of observed post-synaptic spikes N ) . The second term , the van Rossum distance [90] , accounts for discrepancy between observed and simulated spike times , and here we use τR = 2 / f where f denotes the post-synaptic firing rate of the observed neuron . Together these terms allow us to closely approximate the current-injection experiments and simulate model neurons receiving similar current-based or conductance-based input . | Synapses play a central role in neural information processing – weighting individual inputs in different ways allows neurons to perform a range of computations , and the changing of synaptic weights over time allows learning and recovery from injury . Intracellular recordings provide the most detailed view of the properties and dynamics of individual synapses , but studying many synapses simultaneously during natural behavior is not feasible with current methods . In contrast , extracellular recordings allow many neurons to be observed simultaneously , but the details of their synaptic interactions have to be inferred from spiking alone . By modeling how spikes from one neuron , statistically , affect the spiking of another neuron , statistical inference methods can reveal “functional” connections between neurons . Here we examine these methods using neuronal spiking evoked by intracellular injection of a defined artificial current that simulates input from a single presynaptic neuron or a large population of presynaptic neurons . We study how well functional connectivity methods are able to reconstruct the simulated inputs , and assess the validity and limitations of functional connectivity inference . We find that , with a sufficient amount of data , accurate inference is often possible , and can become more accurate as more of the presynaptic inputs are observed . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2015 | Identifying and Tracking Simulated Synaptic Inputs from Neuronal Firing: Insights from In Vitro Experiments |
In Africa , many areas are co-endemic for the two major Schistosoma species , S . mansoni and S . haematobium . Epidemiological studies have suggested that host immunological factors may play an important role in co-endemic areas . As yet , little is known about differences in host immune responses and possible immunological interactions between S . mansoni and S . haematobium in humans . The aim of this study was to analyze host cytokine responses to antigens from either species in a population from a co-endemic focus , and relate these to S . mansoni and S . haematobium infection . Whole blood cytokine responses were investigated in a population in the north of Senegal ( n = 200 ) . Blood was stimulated for 72 h with schistosomal egg and adult worm antigens of either Schistosoma species . IL-10 , IL-5 , IFN-γ , TNF-α , and IL-2 production was determined in culture supernatants . A multivariate ( i . e . multi-response ) approach was used to allow a joint analysis of all cytokines in relation to Schistosoma infection . Schistosoma haematobium egg and worm antigens induced higher cytokine production , suggesting that S . haematobium may be more immunogenic than S . mansoni . However , both infections were strongly associated with similar , modified Th2 cytokine profiles . This study is the first to compare S . mansoni and S . haematobium cytokine responses in one population residing in a co-endemic area . These findings are in line with previous epidemiological studies that also suggested S . haematobium egg and worm stages to be more immunogenic than those of S . mansoni .
Schistosomiasis is a parasitic disease of major public health importance . Schistosoma mansoni and S . haematobium are the main human species . Both species are endemic in Africa , where their distributions show a great overlap [1] . Schistosomes are known to down-regulate host immune responses and to induce so-called modified Th2 responses . The exact phenotype of the induced response depends on a complex immunological ‘dialogue’ that involves cytokines and immune cells of Th2 , but also Th1 , Th17 and regulatory components of the immune system [2] . So far , little is known about differences in host immune responses to schistosomes and possible immunological interactions between S . mansoni and S . haematobium in humans . Yet , epidemiological studies have suggested that host immunological factors may play an important role in co-endemic areas . Interspecies differences in immunogenicity for example , may explain why infection-age curves and morbidity patterns differ between S . mansoni and S . haematobium . Also , immunological interspecies differences and/or immunological interactions between S . mansoni and S . haematobium may explain differences in morbidity levels between single and mixed Schistosoma infections . Cheever et al . reported a more pronounced reduction of S . haematobium than S . mansoni worm loads with age [3] . Similarly , in a mixed focus in northern Senegal , we found the age-infection curve of S . haematobium to decline more steeply after adolescence than that of S . mansoni [4] , indicating that protective immunity against S . haematobium may develop more rapidly . In addition , we found that mixed S . mansoni and S . haematobium infection as compared with single S . haematobium infection tended to decrease the risk of S . haematobium-specific urinary tract pathology [5] . This appeared mainly due to ectopically excreted , possible hybrid eggs [6] . Others also found S . mansoni to affect S . haematobium-specific morbidity and vice versa [7] , [8] , indicating that the two infections may have different effects on the egg-induced immune responses that provoke morbidity . The present study set out to compare Schistosoma-specific cytokine responses induced by S . mansoni and S . haematobium antigens , and to relate these to Schistosoma infection in a S . mansoni and S . haematobium co-endemic area . Schistosoma infection status ( single and mixed ) and infection intensities as well as Schistosoma-specific cytokine responses were determined in residents from a co-endemic focus in northern Senegal . A multivariate ( i . e . multi-response ) approach was used to allow a joint analysis of multiple cytokine responses ( interleukin ( IL ) -10 , IL-5 , interferon ( IFN ) -γ , tumor necrosis factor ( TNF ) -α , and IL-2 ) [9] .
This study was part of a larger investigation on the epidemiology of schistosomiasis and innate immune responses ( SCHISTOINIR ) for which approval was obtained from the review board of the Institute of Tropical Medicine , the ethical committee of the Antwerp University Hospital and ‘Le Comité National d'Ethique de la Recherche en Santé’ in Dakar . Informed and written consent was obtained from all participants prior to inclusion into the study . For minors , informed and written consent was obtained from their legal guardians . All community members were offered praziquantel ( 40 mg/kg ) and mebendazole ( 500 mg ) treatment after the study according to WHO guidelines [10] . This study was conducted in Ndieumeul and Diokhor Tack , two neighboring communities on the Nouk Pomo peninsula in Lake Guiers . Details on the study area have been described elsewhere [4] , [5] . Between July 2009 and March 2010 , parasitological data were collected from 857 individuals [4] . A random subsample of 200 subjects was followed up immunologically . These subjects were between 5 and 53 years of age . Individuals who had lived in an urban area in the 5 years preceding the study ( n = 7 ) , had taken praziquantel within the last year ( n = 2 ) , or had clinical signs of malaria ( recruited upon recovery ) , and pregnant women ( n = 18 ) were excluded from the immunological study . Two feces and two urine samples were collected from each participant on consecutive days . Infection with Schistosoma spp . was determined quantitatively ( by Kato-Katz and urine filtration ) , and infection with soil-transmitted helminths ( STHs ) Ascaris lumbricoides , Trichuris trichiura and hookworm , was assessed qualitatively ( by Kato-Katz ) , as described elsewhere [4] . Aliquots of the first fecal samples were preserved in ethanol to confirm microscopy results by multiplex PCR ( A . lumbricoides , hookworm and Strongyloides stercoralis ) ( n = 198 ) [11] . Infection with Plasmodium was determined by Giemsa-stained thick blood smears . Five hours after venipuncture , heparinized blood was diluted 1∶4 in RPMI 1640 ( Invitrogen ) supplemented with 100 U/ml penicillin , 100 µg/ml streptomycin , 1 mM pyruvate and 2 mM glutamate ( all from Sigma ) . This mixture ( 200 µl sample volume ) was incubated in 96-well round bottom plates ( Nunc ) at 37°C under 5% CO2 atmosphere for 72 h , together with one of four schistosomal water-soluble antigen preparations at a final concentration of 10 µg protein/ml: Medium ( see above ) without stimulus was used as a negative control . After harvesting , supernatants were stored at −80°C . Schistosoma eggs and adult worms were isolated from either S . mansoni- or S . haematobium-infected golden hamsters . SEAm , SEAh , AWAm and AWAh were prepared from this material using identical procedures . In brief , eggs or worms were freeze-dried and then homogenized in phosphate-buffered saline ( PBS ) with 10% n-octyl-β-D-glucopyranoside . Subsequently , this mixture was sonicated , frozen , thawed and washed with PBS . The resulting pellet was dialyzed and filter-sterilized . While AWAm and AWAh batches were lipopolysaccharide ( LPS ) -free , SEAm and SEAh antigens contained equivalent amounts of LPS ( final concentrations of 1–5 ng/ml ) . IL-10 , IL-5 , IFN-γ , TNF-α , and IL-2 in culture supernatants were analyzed simultaneously using custom Luminex cytokine kits ( Invitrogen ) according to the manufacturer's instructions . Samples with concentrations below the detection limit were assigned values corresponding to half of the lowest value detected . Lowest values detected were 0 . 063 pg/ml for IL-10 , 0 . 044 pg/ml for IL-5 , 0 . 090 pg/ml for IFN-γ , 0 . 051 pg/ml for TNF-α , and 0 . 063 pg/ml for IL-2 . Results were considered significant when the p-value was <0 . 05 . The Pearson Chi-square test was used to determine the association between infection status on the one hand , and age and gender on the other . Nonparametric techniques were chosen because cytokine concentrations were not normally distributed . Univariate statistics were used to compare single antigen-induced responses within individuals ( IBM SPSS 21 . 0 ) . McNemar's tests were used to compare cytokine response frequencies between S . mansoni and S . haematobium antigen-induced responses within individuals ( e . g . SEAm- versus SEAh-induced responses ) . Similarly , Wilcoxon Signed Rank tests were used to compare cytokine response levels between S . mansoni and S . haematobium antigen-induced responses within individuals . Multivariate ( i . e . multi-response ) statistics were used to collectively analyze multiple cytokine responses – i . e . cytokine profiles - in the study population , and to investigate interrelationships between these responses [9] . We chose the nonparametric technique nonmetric multidimensional scaling ( nMDS; in R with the ‘Vegan’ package [12] , [13] ) . This is a variant of the parametric principal component analysis ( PCA ) , but with fewer assumptions about the nature of the data and the interrelationship of the variables [14] . This is important because cytokine response levels were not normally distributed , even after log-transformation . Also , levels of different cytokines typically correlate with one another . Upon computation of the cytokine profiles , associations between these cytokine profiles and Schistosoma infection were assessed . The approach is illustrated in Supporting Information S1 . Before nMDS , cytokine concentrations in the negative control were subtracted from those in antigen-stimulated samples to obtain net cytokine responses . Negative values were set to zero . Net cytokine responses were normalized by log ( base 10 ) -transformation after adding 1 pg/ml to allow for zeroes . Schistosoma infection intensities were normalized after adding half of the detection limit ( i . e . 5 eggs per gram of feces and 0 . 5 eggs per 10 ml of urine for S . mansoni and S . haematobium , respectively ) . One nMDS was performed for each of the four Schistosoma-specific whole blood stimulations ( either SEAm , SEAh , AWAm or AWAh ) using the ‘metaMDS’ function [13] . Each nMDS was repeated several times to assess the robustness of the resulting pattern [14] . The Euclidean dissimilarity index was used [13] , and cytokine profiles - i . e . the matrix of IL-10 , IL-5 , IFN-γ , TNF-α , and IL-2 - were plotted in three dimensions ( 3D ) to adequately represent the variation in the data [14] . Afterwards , gradients of the separate cytokine responses , on which the nMDS was based , were fitted using the ‘envfit’ function [13] . The same function was used to fit infection intensities onto each 3D nMDS , and to statistically test associations of antigen-induced cytokine profiles with Schistosoma infection intensity or infection status , i . e . uninfected , single S . mansoni , single S . haematobium , versus mixed S . mansoni and S . haematobium infection . The ‘ordiellipse’ function was used to fit average group scores - with their 95% confidence intervals ( CIs ) - for different infection statuses [13] . In contrast to individual S . mansoni- and S . haematobium-induced cytokine responses which can be compared quantitatively within individuals as described above ( univariate statistics ) , qualitative differences between S . mansoni- and S . haematobium-induced cytokine profiles could only be assessed visually by nMDS , not by formal statistical testing .
The study population consisted of 88 males and 112 females with a median age of 16 ( range 5–53 ) years . Malaria and STHs T . trichiura and hookworm were absent in this population , and A . lumbricoides and S . stercoralis rare ( n = 3 and 2 , respectively , with 100% concordance between microscopy and PCR ) . In contrast , 137 ( 69% ) subjects were infected with S . mansoni , and 116 ( 58% ) with S . haematobium . Sixty percent ( 95/158 ) of all Schistosoma infections were mixed S . mansoni and S . haematobium infections ( Table 1 ) . The distributions of S . mansoni and S . haematobium infections in the study population according to age and gender are shown in Table 2 . Both Schistosoma infections peaked in adolescents ( 10 to 19 year-olds ) , but gender differences were not statistically significant . Epidemiological patterns of infection have been described in more detail elsewhere [4] . Insight into the different antigen-induced cytokine responses relative to one another was obtained by nMDS . Figure 1 and 2 show the variation in multivariate cytokine responses in the study population , with dots representing individuals . Distances between dots approximate inter-individual dissimilarities in cytokine responses with stress values ( i . e . discrepancies ) of 0 . 051 for SEAm , 0 . 041 for SEAh , 0 . 058 for AWAm , and 0 . 061 for AWAh . Red arrows indicate increasing gradients of IL-10 , IL-5 , IFN-γ , TNF-α and IL-2 responses , respectively . The level of a cytokine response increases in the direction of the corresponding arrow ( see also Supporting Information S1 ) . The length of a cytokine arrow indicates the goodness of fit of that arrow ( or cytokine gradient ) . The nMDS outcomes for the first axis ( nMDS1 ) show that for each of the four antigen stimulations , all cytokine responses point to the left . Individuals plotted on the left produced consistently higher levels of all cytokines measured than those on the right . In other words , nMDS1 indicates a gradient of high ( left ) to low ( right ) cytokine responses . In analogy , the second axis ( nMDS2 ) , indicates a gradient of Th1-like ( IFN-γ and TNF-α , top ) to Th2-like ( IL-5 , bottom ) phenotypes for each of the antigen stimulations . In contrast to SEA-induced IL-5 , AWA-induced IL-5 was not accompanied by production of IL-10 . IL-2 levels increased with Th1 cytokines , except for SEAm . The third axis ( nMDS3 ) indicates a gradient of TNF-α and IL-2 ( left ) to IFN-γ and IL-10 ( right ) . In contrast to antigen-induced cytokines , spontaneously induced levels of cytokines in the control ( medium only ) , did not show significant gradients , except for IL-5 on the third nMDS axis ( stress = 0 . 11 , data not shown ) . Figure 1 and 2 indicate that S . mansoni and S . haematobium antigens induced very similar cytokine profiles; cytokine profiles differed more between adult ( AWA ) and egg ( SEA ) life stages of the parasite than between the two Schistosoma species . Within individuals , S . haematobium-induced cytokine response levels were higher than those induced by S . mansoni ( Table 3 ) . This was statistically significant for all SEA- and AWA-induced cytokine responses that were measured , except for SEA-induced IFN-γ and IL-10 . Subsequently , we related the above-described variation in cytokine responses in the study population ( i . e . plotted cytokine profiles ) to infection intensity . Table 4 shows that all associations between Schistosoma antigen-induced cytokine profiles and infection intensity were statistically significant . In Figure 1 , the direction of the black arrows represents the increasing gradients of S . mansoni and S . haematobium infection intensity , respectively ( see also Supporting Information S1 ) . On the first axis , which indicates cytokine response levels ( see above ) , these arrows generally point into the opposite direction of cytokine responses . This indicates that people with elevated Schistosoma infection intensities are more likely to have lower cytokine responses , and vice versa . On the second axis , which indicates the Th1 versus Th2 response phenotype ( see above ) , infection intensity generally increases with IL-5 and decreases with Th1 cytokines TNF-α , IFN-γ , and IL-2 ( except for SEAm-induced IL-5 which decreases with increasing infection intensity ) . Briefly , as infection intensity increased , cytokine response levels decreased and the Th2 phenotype became more pronounced . The association between infection intensity and reduced cytokine responsiveness was more pronounced for SEA than for AWA stimulation . Schistosoma infection intensity increased with AWA-induced IL-5 , but decreased with SEA-induced IL-5 levels , indicating that people with higher infection intensities produced more of a Th2-like response against AWA and more of a suppressive response ( i . e . with low cytokine response levels ) against SEA than people with lower infection intensities , and vice versa . We did not observe differences in induced cytokine profiles between the two Schistosoma infections . Associations between cytokine profiles and infection intensity were comparable for S . mansoni and S . haematobium infections ( Figure 1 ) . Table 4 shows significant correlations between cytokine profiles and Schistosoma infection intensity for homologous combinations ( i . e . infection intensity and antigen stimulation of the same species ) as well as for heterologous combinations ( i . e . infection intensity of one and antigen stimulation of the other species ) . Schistosoma antigen-induced cytokine profiles were significantly associated with Schistosoma infection status , except upon stimulation with AWAm ( Table 4 ) . Figure 2 shows how antigen-induced cytokine profiles differed according to infection status ( except for AWAm , which was not significantly associated with infection status ) , with 95% CI ellipsoids indicating the average nMDS scores per infection group: uninfected ( ‘N’ ) , single S . mansoni ( ‘M’ ) , single S . haematobium ( ‘H’ ) , versus mixed ( ‘MH’ ) Schistosoma infection group . In analogy with Figure 1 , uninfected individuals had higher cytokine responses than Schistosoma-infected subjects , and their cytokine profiles were skewed more towards the Th1 phenotype . On the whole , there was a gradient in cytokine profiles from uninfected individuals , to people with single and then mixed Schistosoma infections ( Figure 2 ) and these profiles were in the same direction as the gradient of infection intensity ( Figure 1 ) . In other words , people with low cytokine responses of the Th2 phenotype tended to have both mixed and heavier infections , people with strong Th1 responses tended to be uninfected , and those with an intermediate cytokine profile tended to have both single and lighter Schistosoma infections . For the SEAm-induced cytokine profile , there was a clear difference ( i . e . separation between ellipsoids ) between S . mansoni-infected individuals ( with either single or mixed S . mansoni ) , and those without S . mansoni ( no Schistosoma infection , or single S . haematobium infection; Figure 2A ) . There were no significant differences in this cytokine profile between single and mixed S . mansoni infections , or between uninfected individuals and those with single S . haematobium infections . This indicates that , in contrast to S . mansoni , S . haematobium infection status was not associated with SEAm-induced cytokine profiles . Schistosoma haematobium-induced cytokine profiles on the other hand , showed similar relationships with S . mansoni as well as with S . haematobium infection status . Cytokine profiles of people with single and mixed infections differed significantly from those of uninfected people , and cytokine profiles did not appear to differ between single S . mansoni and single S . haematobium infections .
In conclusion , this is the first study to comprehensively investigate S . mansoni- and S . haematobium-induced cytokine responses in a S . mansoni and S . haematobium co-endemic area , and to relate these cytokine responses to Schistosoma infection . The present study demonstrates that nMDS can be used successfully as a tool for the joint analysis of multiple cytokine responses in relation to Schistosoma infection . We showed strong associations between Schistosoma infection and Schistosoma-induced cytokine profiles , and provided a first insight into potential differences and interactions between human S . mansoni and S . haematobium infections . This knowledge will contribute to an improved understanding of the mechanisms underlying Schistosoma infection and morbidity in co-endemic populations . | In the developing world , over 207 million people are infected with blood-dwelling parasitic Schistosoma worms . Schistosoma haematobium and S . mansoni are the most widespread species . In Africa , they often occur together in the same area , with many people carrying both species . Yet , little is known about the differences in immune response that the human host develops against these two species . It is also unknown whether the presence of one species may affect the immune response to the other . We here investigated 200 people from an area in the north of Senegal where both species occur . They were examined for Schistosoma infections , as well as for immune responses to the two species . We observed that both infections were characterized by very similar cytokine responses . However , S . haematobium antigens induced higher levels of cytokines than S . mansoni . This suggests that S . haematobium may give rise to stronger immune responses , and may help to explain differences between the two most important Schistosoma species regarding the occurrence of infection and morbidity . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"and",
"health",
"sciences",
"cytokines",
"pathology",
"and",
"laboratory",
"medicine",
"immunity",
"to",
"infections",
"immunology",
"tropical",
"diseases",
"parasitic",
"diseases",
"developmental",
"biology",
"molecular",
"development",
"neglected",
"tropical",
"diseases",
"infectious",
"diseases",
"pathogenesis",
"immune",
"response",
"immune",
"system",
"helminth",
"infections",
"schistosomiasis",
"clinical",
"immunology",
"immunity",
"host-pathogen",
"interactions",
"biology",
"and",
"life",
"sciences",
"acquired",
"immune",
"system"
] | 2014 | Cytokine Responses to Schistosoma mansoni and Schistosoma haematobium in Relation to Infection in a Co-endemic Focus in Northern Senegal |
The chemotherapy of schistosomiasis currently depends on the use of a single drug , praziquantel . In order to develop novel chemotherapeutic agents we are investigating enzymes involved in the epigenetic modification of chromatin . Sirtuins are NAD+ dependent lysine deacetylases that are involved in a wide variety of cellular processes including histone deacetylation , and have been demonstrated to be therapeutic targets in various pathologies , including cancer . In order to determine whether Schistosoma mansoni sirtuins are potential therapeutic targets we first identified and characterized their protein sequences . Five sirtuins ( SmSirt ) are encoded in the S . mansoni genome and phylogenetic analysis showed that they are orthologues of mammalian Sirt1 , Sirt2 , Sirt5 , Sirt6 and Sirt7 . Both SmSirt1 and SmSirt7 have large insertion in the catalytic domain compared to their mammalian orthologues . SmSirt5 is the only mitochondrial sirtuin encoded in the parasite genome ( orthologues of Sirt3 and Sirt4 are absent ) and transcripts corresponding to at least five splicing isoforms were identified . All five sirtuins are expressed throughout the parasite life-cycle , but with distinct patterns of expression . Sirtuin inhibitors were used to treat both schistosomula and adult worms maintained in culture . Three inhibitors in particular , Sirtinol , Salermide and MS3 induced apoptosis and death of schistosomula , the separation of adult worm pairs , and a reduction in egg laying . Moreover , Salermide treatment led to a marked disruption of the morphology of ovaries and testes . Transcriptional knockdown of SmSirt1 by RNA interference in adult worms led to morphological changes in the ovaries characterized by a marked increase in mature oocytes , reiterating the effects of sirtuin inhibitors and suggesting that SmSirt1 is their principal target . Our data demonstrate the potential of schistosome sirtuins as therapeutic targets and validate screening for selective sirtuin inhibitors as a strategy for developing new drugs against schistosomiasis .
The current strategy for the treatment and control of schistosomiasis is the mass-treatment of populations in endemic areas using the only available drug , Praziquantel . Notably , the Schistosomiasis Control Initiative [1] in sub-Saharan Africa had dispensed more than 40 million doses of Praziquantel by 2008 . Although this ongoing programme will undoubtedly have a major impact on morbidity and mortality in the region ( estimated at 280000 deaths annually prior to the initiative [2] ) , this approach renders probable the eventual selection of resistant strains of schistosomes [3] , which have already been characterized in endemic areas [4] and can be selected in the laboratory [5] . The development of new drugs is therefore indispensable in order to ensure our capacity to treat schistosomiasis in the long term . In the search for new drug leads one of the possible approaches is to exploit strategies that have been successful for other pathologies . We have chosen to target a group of enzymes that is under active study for the development of anti-cancer drugs , the enzymes that effect posttranslational modifications of histones including the ( de ) acetylation and ( de ) methylation of lysine or arginine residues . Inhibitors of these enzymes have been shown to be generally more toxic for cancer cells than for normal cells [6] . Two such drugs ( Vorinostat and Romidepsin ) , both histone deacetylase ( HDAC ) inhibitors , have been approved for use in humans and a further 15 HDAC inhibitors are in clinical trials [7] . Our working paradigm is that schistosomes , like other parasites , have some of the characteristics of malignant tumours [8] . Their cell division ( for egg production ) is intense and outside the control of the host , they are practically invisible to the host immune response . They also have a high level of metabolic activity , which like tumours , is dependent on the use of large amounts of glucose that is metabolized by aerobic glycolysis ( culminating with the conversion of pyruvate to lactate rather than its use in oxidative phosphorylation ) within the mammalian host [9] , [10] . This type of metabolism was first shown to be a characteristic of cancer cells by Warburg [11] . Moreover , the reverse paradigm , that tumour cells behave like parasitic organisms to favour their survival and growth , has also been proposed [12] . This is based on the observation that metabolites ( including fatty acids , ketones , glutamine and glucose ) from “host” tissues promote tumour growth . The metabolic switch to lactate production in cancer cells has been linked to changes in their epigenetic state [13] . In consequence our expectation is that inhibitors of histone modifying enzymes will be significantly more toxic to the parasite than to the host , and , moreover , that analogues , or novel inhibitors , can be identified that will be selective for the schistosome target . The availability of the annotated genome sequence for Schistosoma mansoni [14] has allowed us to identify the schistosome histone modifying enzymes [8] . Here we have studied the S . mansoni sirtuins and attempted to evaluate their potential as therapeutic targets . Sirtuins are NAD+-dependent deacetylases that are also referred to as class III HDACs , although they are phylogenetically unrelated to the Zn2+-dependent class I and II HDACs [15] . Sirtuins can also act as mono-ADP-ribosyltransferases . For example , human sirtuin 4 ( Sirt4 ) is a mitochondrial enzyme that down-regulates glutamine dehydrogenase by ADP-ribosylation [16] . Moreover , Sirt5 has been recently demonstrated to preferentially hydrolyze succinyl and malonyl lysine [17] , whilst Sirt6 has low deacetylase activity , but efficiently removes long-chain fatty acyl groups , such as myristoyl , from lysine residues [18] . Sirtuins can be divided into five classes , one of which ( class U ) is only represented in bacteria and archaea [19] . The seven human sirtuins are grouped into the four other classes and have distinct subcellular localizations . Sirtuins 3 , 4 and 5 are localized in the mitochondria , Sirts 6 and 7 are exclusively nuclear , Sirt1 has a dual nuclear/cytosolic localisation and Sirt2 is cytosolic [20] . Many different target proteins have been described for these enzymes and even the nuclear sirtuins act on proteins other than histones . For instance , Sirt1 deacetylates transcription factors such as p53 [21] and FoxO [22] . In keeping with this variety of substrates , in metazoans sirtuins have been associated with a wide variety of processes including transcriptional silencing , ageing , metabolic regulation and apoptosis [23 for review] . Despite the identification of inhibitors of Sirt1 , Sirt2 , and Sirt3 with a wide range of core structures [24 for review] , the need for potent and selective inhibitors , particularly of Sirt1 , remains to be fulfilled [7] . One compound ( Selisistat or SEN196 ) is in Phase II clinical trials for Huntington's disease [7] . However , the demonstration of the activity of sirtuin inhibitors such as Sirtinol , which induces apoptosis and autophagic cell death in MCF-7 human breast cancer cells [25] , or Salermide , which targets both Sirt1 and Sirt2 and , in so doing , induces cell death and p53 acetylation , again in MCF-7 cells [26] , shows the potential of sirtuin inhibitors in cancer therapy . The inhibition of sirtuins has been less well studied for therapeutic potential against parasites than has the inhibition of class I and II HDACs . In the case of Plasmodium falciparum this is partly due to the fact that both of the sirtuins present , PfSir2A and PfSir2B , can be genetically disrupted without an impact on parasite viability in vitro [27] . However , both Nicotinamide and the synthetic inhibitor Surfactin inhibit PfSir2 activity and are potent inhibitors of intra-erythrocytic growth of the parasite [28] , [29] . Disruption of the gene encoding a cytosolic Sir2 homologue in Leishmania infantum showed its necessity for parasite survival and the sirtuin inhibitor Sirtinol inhibited the in vitro growth of the parasite via the induction of apoptosis [30] . In the present study we have identified and characterized the five sirtuins encoded in the S . mansoni genome and established their homology relationships through phylogenetic analysis . We further investigated the relative expression of their transcripts at different life-cycle stages . We next showed that inhibitors of human sirtuins , including Sirtinol and Salermide , induced the death of schistosomula in culture via the induction of apoptosis . Further , these inhibitors provoked the separation of adult worm pairs in culture , a reduction in egg laying and Salermide treatment induced massive modifications to the ovaries and testes . Finally , the knockdown of SmSirt1 transcripts by RNAi in adult worms led to similar modifications in the ovaries as seen after Salermide treatment .
All animal experimentation was conducted in accordance with the European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes ( ETS No 123 , revised Appendix A ) and was approved by the committee for ethics in animal experimentation of the Nord-Pas de Calais region ( Authorization No . AF/2009 ) and the Pasteur Institute of Lille ( Agreement No . A59-35009 ) . A Puerto Rican strain of S . mansoni is maintained in the laboratory using the intermediate snail host Biomphalaria glabrata and the golden hamster Mesocricetus auratus as definitive host . Adult worms were obtained by whole-body perfusion of 6-week infected hamsters [31] . Eggs were obtained from the livers of infected hamsters and hatched out under light to obtain miracidia [32] . Newly transformed miracidia were maintained in complete Chernin's balanced salt solution [32] ( CBSS ) supplemented with 1 mg/mL glucose and 1 mg/mL trehalose , for 48 h to achieve in vitro transformation into primary sporocysts . Cercariae were released from infected snails , harvested on ice as described previously [33] and schistosomula were obtained in vitro by mechanical transformation [34] . Total RNA was isolated from the different stages of S . mansoni with TRIzol® reagent ( Invitrogen ) according to the manufacturer's instructions , followed by treatment with RNase-free DNase ( Turbo DNA-free kit , Ambion ) . Sirtuins encoded in the S . mansoni genome were identified by screening using Hidden Markov Model profiles derived from the Pfam database [35] . Predicted protein sequences were manually annotated by integrating data from InterProScan [36] and reverse PSI-BLAST analysis . In order to verify and complete the predicted sequences , we carried out 5′ and 3′ RACE ( GeneRacer Kit , Invitrogen ) using oligonucleotides ( Table S1 ) based on these sequences and generated full length cDNA sequences . The integrity of the sequences was verified by performing PCR using the Advantage 2 Polymerase mix according to the manufacturer's procedure ( Clontech ) and oligonucleotides encompassing the coding region . The PCR product was purified from agarose gels using the extraction kit Wizard SVGel and PCR clean up system ( Promega ) and inserted into pCR2 . 1-TOPO before transformation of chemically competent Escherichia coli cells ( One Shot T0P10 , Invitrogen ) . Selected clones were sequenced by GATC Biotech . Sequence analyses and alignments were performed using the LASERGENE package ( DNAStar ) and the BioEdit v7 . 1 . 11 package ( http://www . mbio . ncsu . edu/BioEdit/bioedit . html ) . Sirtuin conserved catalytic domain ( Pfam: PF02146 ) sequences , delimited based on alignments in [15] , were aligned using BioEdit v7 . 1 . 11 and the Clustal W program implemented therein . In order to determine the relationship of S . mansoni sirtuins to defined classes and families , we included sirtuin sequences from vertebrates , ecdysozoan invertebrates and one other lophotrochozoan ( Clonorchis sinensis ) in the analysis . An unrooted phylogram was generated using the MEGA4 neighbour-joining method [37] based on a Poisson correction substitution model and the figure was generated using FigTree ( http://tree . bio . ed . ac . uk/software/figtree/ ) . The confidence levels for the phylogenetic tree were estimated by bootstrapping using 100000 replicates . Secondary structures of sirtuins were obtained with a secondary structure prediction server , Jpred 3 [38] with human sirtuins as templates . Complementary DNAs were obtained by reverse transcription of total RNA using the Thermoscript RT-PCR System ( Invitrogen ) and used as templates in triplicate assays for PCR amplification using the SYBR Green PCR Master Mix and ABI PRISM 7000 sequence detection system ( Applied Biosystems ) . Primers specific for S . mansoni sirtuins ( Table S1 ) were designed by the Primer Express Program ( Applied Biosystems ) and used or amplication in triplicate assays . Measurements of real time PCR efficiency for each primer pair allowed the ratios of expression to be calculated using the 2−ΔΔCt ratio [39] with S . mansoni α-tubulin as the reference transcript [40] . Sirtinol [41] , MS3 ( compound 26 in [42] ) , MS13 ( compound 19 in [43] ) , HR103 ( compound 5a in [44] ) , and CS13 ( bis ( 4-nitrobenzylidene ) pyrrolidine-2 , 4-dione; F . B . , unpublished ) were synthesized in our labs according to reported procedures . Purity and identity were assured using MS , NMR and HPLC . Salermide was purchased from Santa Cruz Biotechnology Inc . The assay to determine the effects of sirtuin inhibitors on the viability of S . mansoni schistosomula was carried out as previously described [40] . Briefly , schistosomula ( 2000 ) were incubated at 37°C in a humid atmosphere containing 5% CO2 during 5 days in 6-well plates containing 2 mL of M199 medium ( Invitrogen ) kept at pH 7 . 4 with HEPES 10 mM and supplemented with penicillin ( 50 U/mL ) , streptomycin ( 50 µg/mL ) , gentamycin ( 15 µg/mL ) and rifampicin ( 60 µg/mL ) and 10% fetal calf serum ( Gibco ) ( hereafter referred to as M199 complete medium ) with two different concentrations ( 10 and 20 µM ) of sirtuin inhibitors dissolved in DMSO . Culture medium was refreshed daily . Parasite mortality was assessed by eye each day using three criteria: absence of motility , tegument defects and granular appearance . A minimum of 300 larvae was observed for each condition , and the ratio of dead larvae to total larvae calculated . Two different assays were performed for each condition and three independent biological replicates ( different batches of schistosomula ) were carried out . To measure the effect of sirtuin inhibitors on adult worm pairing in culture [45] S . mansoni adult worms obtained from hamsters were washed in M199 medium and ten pairs of adult worms were transferred to each well of a 6-well culture plate containing 2 mL of M199 completed medium . The worms were cultured during 6 days at 37°C in a humid atmosphere containing 5% CO2 , and then two different concentrations of sirtuin inhibitors ( 10 and 20 µM ) were tested . Culture medium and the inhibitors were refreshed daily . The number of paired couples was estimated every day by microscopy . In each well , medium containing the eggs was harvested every day and centrifuged . The total number of eggs was determined by microscopy and two different assays were performed for each condition and repeated with three independent biological replicates . Detection of DNA strand breaks in Salermide , Sirtinol and MS3 treated schistosomula was done using the Terminal deoxynucleotidyl transferase dUTP Nick End Labelling ( TUNEL ) method using the In Situ Cell Death Detection Kit , TMR red ( Roche ) . The method designed for cell suspensions was followed as described in the manufacturer's instructions with modifications , Briefly , 2000 schistosomula were treated or not for 48 h with 10 and 20 µM Salermide , Sirtinol and MS3 , in 6-well plates containing 2 mL of complete medium . Culture medium was removed and the schistosomula were centrifuged ( 1000 rpm , 2 min ) washed three times in PBS , then fixed in formaldehyde 2% for 60 min . Schistosomula were washed once more in PBS and permeabilization solution ( Triton X-100 0 . 1% , sodium citrate 0 . 1% ) was added for 10 min on ice . Labeling of schistosomula with DAPI and TMR red dUTP was performed according to the manufacturer's instructions and TUNEL-positive parasites were observed by fluorescence using an AxioImager Z1-Apotome microscope ( Zeiss ) . After 6 days in culture , worms were fixed for at least 24 h in AFA ( ethanol 95% , formalin 3% and glacial acetic acid 2% ) , stained for 30 min with 2 . 5% hydrochloric carmine ( Certistain , Merck ) , and destained in acidic 70% ethanol . Following dehydration in 70% , 90% and 100% ethanol , 5 min each , worms were preserved as whole-mounts in Canada balsam ( Merck ) on glass slides . To study the morphology of the reproductive organs of parasites , CLSM images were taken using a Leica TCS SP2 microscope with a 488 nm He/Ne laser and a 470 nm long-pass-filter under reflection mode . Two fragments of 500 bp SmSirt1-dsRNA templates were generated by PCR using gene targeted primers containing T7 promoter sequences ( Table S1 ) . A luciferase dsRNA template of equivalent size was generated similarly using the pGL3-basic plasmid ( Promega ) as template . dsRNA was prepared and purified using the Megascript RNAi kit ( Ambion ) according to the manufacturer's instructions , and concentrations were determined spectrophotometrically ( NanoVue Plus™ , GE Healthcare ) . To deliver the dsRNA , 8 adult worms/group in 100 µL M199 medium containing 25 µg dsRNA , were electroporated in a 4 mm cuvette by applying a square wave with a single 20 ms impulse , at 125 V and at room temperature , as described [46] . Parasites were then transferred to 4 mL complete M199 . After two days in culture , 2 mL of medium was removed and 2 mL of fresh complete M199 culture medium was added . Gene knockdown was monitored by qRT-PCR 5 days after dsRNA treatment as described above . Microscopic examination of RNAi-treated worms was carried out exactly as described below . Four independent experiments were carried out . The statistical significance of the level of SmSirt1 transcript knockdown was evaluated using Student's t-test in the GraphPad Prism programme ( GraphPad Software Inc . ) .
Five protein sequences corresponding to sirtuins were identified in the S . mansoni predicted proteome [14] and initial sequence similarity searches using Blastp [47] provisionally identified them as potential homologs of mammalian sirtuins . The S . mansoni proteins include Sirt1 ( Smp_138640 ) , Sirt2 ( Smp_084140 ) , Sirt5 ( Smp_055090 ) , Sirt6 ( Smp_134630 ) and Sirt7 ( Smp_024670 ) . Each of the corresponding coding sequences was present on a separate genome scaffold and each corresponded to a single copy gene . In order to verify the predicted sequences and to detect eventual splicing isoforms we carried out 5′ and 3′ RACE PCR using oligonucleotides based on the predicted coding sequences . This allowed us to confirm the existence of each of the predicted proteins as transcripts and to correct a number of assembly errors in the predicted sequences . The corrected sequences ( including splice variants ) have been submitted to the NCBI with accession numbers ABG78545 and KC993850 to KC993857 . The alignment of the catalytic domain ( Pfam: PF02146 ) of S . mansoni sirtuins ( SmSirt1 , 2 , 5 , 6 and 7 ) with homologues from other species is shown in Fig . 1A , B and C . This alignment shows that although the schistosome sirtuin sequences diverge from those of vertebrates or ecdysozoan invertebrates , crucial residues involved in NAD+ binding , acetyl-lysine peptide binding or zinc binding are generally conserved , suggesting functional conservation of these sirtuins in the parasite . The SmSirt1 transcript ( 1832 nt ) encodes a protein of 568 aa . The alignment of the catalytic domain of SmSirt1 ( aa183–466 ) with orthologues from C . sinensis , Caenorhabditis elegans , Drosophila melanogaster and Homo sapiens , is shown in Fig . 1A . Although this domain is well conserved , showing an overall sequence identity of about 70% compared to all three sequences , it also has a large insertion of 63 aa , not shared with Sirt1 orthologues ( Fig . S1 ) . This insertion contains a putative PEST motif ( KVDPSSLLPDEMNDSESTNH ) at its N-terminal , characterized by an enrichment for proline , glutamate ( or aspartate ) , serine and threonine , flanked by lysine , asparagine or histidine residues , that targets proteins for rapid destruction [48] . The short insertion present in D . melanogaster Sirt1 may also represent a PEST motif . The insertion in SmSirt1 further contains a C-terminal phosphorylation motif for PKB/Akt ( Fig . S1 ) [49] that is also absent from all orthologues examined , including Sirt1 from the related trematode C . sinensis ( oriental liver fluke ) . However , we were concerned that this insertion might represent either a cloning artefact , a splice variant or be specific for the Puerto-Rican strain of S . mansoni maintained in our laboratory . We therefore performed PCR , using primers flanking the insertion , on cDNAs from our strain and two isolates from Guadeloupe and Brazil ( kind gifts from G . Mitta , UMR 5244 , CNRS EPHE , University of Perpignan ) . Results ( not shown ) indicate that the same fragment is obtained whatever the S . mansoni strain tested , indicating that the only transcript produced contains the insertion . SmSirt2 is encoded by a unique transcript of 1460 nt that encodes a protein of 337 aa the sequence of which is identical to the prediction ( Smp_084140 ) . In contrast , we identified five different SmSirt5 transcripts , only one of which ( isoform 4 ) seems to contain a complete catalytic domain . This isoform is encoded by a 1346 nt transcript ( 299 aa ) and is the sequence shown in Fig . 1 . However , even this sequence is not identical to the proteome prediction ( Smp_055090 ) , the latter starting at a Met residue 6 aa upstream of that of isoform 4 , which was not confirmed by our RACE experiments . The discrepancy is due to the presence of an intron just upstream of the ATG start codon and a non-coding 5′ exon ( not shown ) . The other isoforms detected are due to alternative splicing sites within exon 3 , leading to the splicing out of parts of this exon , or to an alternative 3′ exon . These forms encode peptides with the catalytic domain truncated at the N- or C-terminal ends respectively ( summarized in Fig . S2 ) the function of which remains to be determined . Although the assembly and annotation of the S . mansoni genome has been improved since its initial publication [14] , [50] some problems with gene assembly and annotation remain . The predicted protein sequence of SmSirt6 ( Smp_134630 ) is a chimera between Sirt6 and a mannosyltransferase . Our RACE experiments allowed us to show not only that SmSirt6 is not chimeric , but also that about 80 aa were missing from the predicted sequence . Full length SmSirt 6 is a protein of 386 aa encoded by a single transcript of 1287 nt . The coding sequence of SmSirt7 was found to be identical to the proteome prediction , but close examination of the full-length transcript revealed that the first 33 nt at the 5′ end corresponded to the spliced leader sequence described by Davis et al [51] that is present on a subset of S . mansoni transcripts and that the translation initiating ATG codon was also provided by the spliced leader as described for other transcripts [52] . The full-length transcript ( 1782 nt ) encodes a 517 aa protein , which , like SmSirt1 , contains a large insertion ( about 190 aa ) within the catalytic domain ( Fig . 1 ) . This insertion is also conserved in the sequence of C . sinensis Sirt7 although in this species the insertion is slightly smaller . In order to verify the assignment of orthologies of the S . mansoni sirtuins , we carried out phylogenetic analysis using neighbor-joining methodology . Other models were tested , but gave similar results . We present a phylogram ( Fig . 2 ) that shows that all the S . mansoni sirtuins group with their orthologues within the four classes of eukaryotic sirtuins defined by Frye [15] . The figure omits the U class sirtuins present in archaea and bacteria . From this analysis it is clear that schistosomes have no orthologues of mammalian Sirt3 ( classe Ib ) or Sirt4 ( class II ) . Along with Sirt5 the two latter sirtuins both localize to mitochondria in humans [20] . Consequently SmSirt5 is probably the only mitochondrial sirtuin present in S . mansoni and its predicted localization ( PSORTII , [53] ) is in agreement with this . Similarly , the predicted localizations of SmSirt1 ( predominantly nuclear ) , SmSirt2 ( cytoplasmic ) , SmSirt6 and SmSirt7 ( nuclear ) are the same as the effective localizations of their human orthologues . Quantitative real-time RT-PCR was carried out at all parasite stages to determine the levels of expression of each of the S . mansoni sirtuin transcripts . In all cases the lowest level of transcript expression was detected in male adult worms and consequently transcript levels at other stages were expressed relative to this stage . Three distinct patterns of expression were evidenced , corresponding to the different sirtuin classes [15] ( Fig . 3 ) . The class I sirtuins , SmSirt1 and SmSirt2 , showed similar profiles of expression ( Fig . 3A ) with the highest transcript levels being present in miracidia and in sporocysts , whilst lower levels were present in male and female adult worms and in cercariae . However , maximal differences in expression levels throughout the life cycle were only about tenfold . SmSirt5 mRNA ( Fig . 3B ) , the only class III sirtuin identified , was most highly expressed in cercariae and schistosomula and much less ( about 50-fold ) abundantly in male adult worms . The expression profiles of SmSirt6 and SmSirt7 mRNAs ( Fig . 3C ) , the class IV sirtuins , were very similar and showed the greatest amplitude of expression during the life-cycle . Indeed , both SmSirt6 and SmSirt7 transcripts were more than 50-fold more expressed in the larval stages than in male and female adult worms . In order to determine whether S . mansoni sirtuins are potential chemotherapeutic targets , we chose a panel of six sirtuin inhibitors for testing on both schistosomula and adult worms in culture . Sirtinol is a potent Sirt1 inhibitor and treatment of MCF-7 human breast cancer cells with this inhibitor leads to hyperacetylation of the Sirt1/2 target p53 and induces apoptosis [25] . Salermide is an inhibitor of Sirt1 and Sirt2 [26] that causes apoptotic tumor-specific cell death in a variety of human cancer cell lines . MS3 and MS13 are thiobarbiturate inhibitors of Sirt1 , Sirt2 [24] and the splitomicin derivative HR103 is a potent Sirt2 inhibitor . CS13 is a tetramic acid derivative . With the aim of determining the capacity of these sirtuin inhibitors to affect the viability of schistosome larvae maintained in culture , 3 h-old schistosomula were cultured for 5 days with a daily renewal of the medium containing the inhibitors . Parasite death was assessed by optical examination each day using three criteria: absence of motility , tegument defects and granular appearance . All sirtuin inhibitors tested induced the mortality of schistosomula in a time and dose-dependent manner ( Fig . 4 ) . However , Salermide ( Fig . 4A ) , Sirtinol Fig . 4B ) and MS3 ( Fig . 4C ) significantly reduced the viability of the schistosomula at 10 µM ( respectively by 78% , 68% and 84% ) and killed all the larvae at 20 µM after 5 days . These inhibitors were clearly more potent than HR103 ( Fig . 4D ) , MS13 ( Fig . 4E ) and CS13 ( Fig . 4F ) . For example , treatment at 10 and 20 µM with HR103 , MS13 and CS13 affected parasite viability , but at a much lower level , with about 50% mortality after 5 days compared to the untreated controls incubated with DMSO alone . One of the principal effects of sirtuin inhibitors on cultured cancer cells is the induction of apoptotic cell death [25] , [26] . Moreover , in our laboratory we have already shown that inhibitors of class I and II HDAC ( s ) induce apoptosis in larvae maintained in culture [40] . We therefore tested the capacity of sirtuin inhibitors to induce apoptosis in the cells of schistosomula using a TUNEL assay . In this experiment , schistosomula were treated with 10 or 20 µM of Salermide ( Fig . 5B , 5F ) , Sirtinol ( Fig . 5C , 5G ) and MS3 ( Fig . 5D , 5H ) for 48 h , then fixed and stained with DAPI and TUNEL . The results indicate that these three inhibitors induce fragmentation of DNA , which may be due to the induction of apoptosis , within 48 h at the same concentrations that induced the mortality of the schistosomula within 3 to 5 days . To determine the effect of sirtuin inhibitors on adult worms , we decided to study the stability of pairing and the production of eggs of parasites maintained in culture . In these experiments , adult worm couples were treated with different concentrations of sirtuin inhibitors over a period of 6 days . The results indicate that during this period , sirtuin inhibitors affect pairing stability in a time and dose-dependent manner ( Fig . 6 ) . We showed that Salermide ( Fig . 6A ) , Sirtinol ( Figure 6B ) and MS3 ( Fig . 6C ) had a greater effect on worm pairing than HR103 ( Figure 6D ) , MS13 ( Fig . 6E ) and CS13 ( Fig . 6F ) compared to the controls incubated with the DMSO solvent alone . After 6 days of treatment with 10 µM of sirtuin inhibitors , worm couples were less affected than at 20 µM , with a reduction of 30% of the pairing for the most potent compounds ( Salermide and Sirtinol ) and a reduction of between 5% and 20% for the less potent compounds ( MS3 , HR103 , CS13 and MS13 ) . All worm couples had separated after 6 days of treatment with 20 µM of Salermide and MS3 . Sirtinol at the same concentration induced a reduction of 70% in pairing . We next determined the total number of eggs laid by the female worms during the 6 days of treatment for each sirtuin inhibitor tested . The results show that all sirtuin inhibitors decreased the number of eggs laid during the treatment compared to the control , in a dose-dependent manner ( Fig . 7 ) . At 10 µM , Salermide ( Fig . 7A ) , Sirtinol ( Fig . 7B ) and MS3 ( Fig . 7C ) induced a reduction of about 30% in egg production and HR103 ( Fig . 7D ) , MS13 ( Fig . 7E ) and CS13 ( Fig . 7F ) a reduction of around 20% . Treatment at 20 µM with Salermide had a drastic effect on egg production , inducing a decrease of 95% compare to the control . With Sirtinol and MS3 , at the same concentration , we observed a lesser reduction around 60% and around 40% for HR103 , MS13 and CS13 . To complement these results , we analyzed by confocal laser scanning microscopy the phenotypic effect on the ovary and the testis of adult worms treated with Salermide at 10 and 20 µM during 6 days . A remarkable effect on the morphology of the gonads of adult worms treated with this inhibitor was evidenced ( Fig . 8 ) . In untreated male worms , the testes are composed of several testicular lobes containing numerous spermatogonia and spermatocytes in different stages of maturation ( Fig . 8A ) . However , after treatment with Salermide at 10 µM ( Fig . 8B ) we observed a drastic reduction in numbers of germinal cells in the testes and this effect was enhanced at 20 µM ( Fig . 8C ) . In untreated female worms , the ovaries have an oval form and are composed of small oogonia with immature oocytes in the anterior part , and larger primary oocytes in the posterior part ( Fig . 8D ) . In ovaries of worms treated with Salermide at 10 µM ( Fig . 8E ) we observed a dramatic disorganization , where immature oocytes were less abundant and the mature cells seemed to invade the whole ovary . As in the case of the male worm this effect was enhanced when the concentration of inhibitor was increased ( Fig . 8F ) . In order to determine whether Sirt1 inhibition was sufficient to explain the effects of the sirtuin inhibitors on adult worm morphology , we carried out RNA interference studies in adult worms to knockdown SmSirt1 transcripts , followed by laser scanning microscopy of the male and female reproductive organs . The results ( Fig . 9 ) show that SmSirt1 knockdown leads to disorganization of the ovary , similar to that caused by Salermide treatment , with a marked increase in mature oocytes ( Fig . 9B , D ) and the appearance of mature oocytes in the anterior part of the ovary . However , no effects were seen on the testes ( Fig . 9B , C ) , suggesting that the inhibition of other sirtuins by Salermide contributed to the phenotype observed after treatment with this drug . Four independent experiments were carried out with similar results .
The corrected sequences ( including splice variants ) of the S . mansoni sirtuins have been submitted to the NCBI with accession numbers ABG78545 ( SmSirt1 ) KC993850 ( SmSirt2 ) , KC993851 ( SmSirt5 isoform 1 ) , KC993852 ( SmSirt5 isoform 2 ) , KC993853 ( SmSirt5 isoform 3 ) , KC993854 ( SmSirt5 isoform 4 ) , KC993855 ( SmSirt5 isoform 5 ) , KC993856 ( SmSirt6 ) and KC993857 ( SmSirt7 ) . | Schistosomiasis is a disease affecting more than 200 million people in tropical and sub-tropical countries caused by parasitic flatworms of the genus Schistosoma . The current reliance on a single drug , Praziquantel , for the treatment and control of the disease renders urgent the development of new therapeutic agents . The strategy that we have chosen is to target the enzymes that carry out epigenetic modifications of the chromatin in the parasite and in particular the histone deacetylases ( HDACs ) . Inhibitors of HDACs have been developed as drugs against cancer and our aim is to exploit structural differences in the catalytic domains of the schistosome enzymes in order to develop selective inhibitors that will be drug precursors . Sirtuins are histone deacetylases that have an NAD+-dependent catalytic mechanism . In this study we have characterized all the Schistosoma mansoni sirtuins and show that they are expressed throughout the parasite life-cycle . Sirtuin inhibitors cause the death of schistosome larvae , the separation of adult worm pairs and tissue damage to the worm reproductive organs . These results demonstrate the validity of S . mansoni sirtuins as therapeutic targets . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"biochemistry",
"gene",
"expression",
"drug",
"discovery",
"biology",
"microbiology",
"molecular",
"cell",
"biology",
"parasitology",
"histone",
"modification"
] | 2013 | Schistosoma mansoni Sirtuins: Characterization and Potential as Chemotherapeutic Targets |
Implantable retinal stimulators activate surviving neurons to restore a sense of vision in people who have lost their photoreceptors through degenerative diseases . Complex spatial and temporal interactions occur in the retina during multi-electrode stimulation . Due to these complexities , most existing implants activate only a few electrodes at a time , limiting the repertoire of available stimulation patterns . Measuring the spatiotemporal interactions between electrodes and retinal cells , and incorporating them into a model may lead to improved stimulation algorithms that exploit the interactions . Here , we present a computational model that accurately predicts both the spatial and temporal nonlinear interactions of multi-electrode stimulation of rat retinal ganglion cells ( RGCs ) . The model was verified using in vitro recordings of ON , OFF , and ON-OFF RGCs in response to subretinal multi-electrode stimulation with biphasic pulses at three stimulation frequencies ( 10 , 20 , 30 Hz ) . The model gives an estimate of each cell’s spatiotemporal electrical receptive fields ( ERFs ) ; i . e . , the pattern of stimulation leading to excitation or suppression in the neuron . All cells had excitatory ERFs and many also had suppressive sub-regions of their ERFs . We show that the nonlinearities in observed responses arise largely from activation of presynaptic interneurons . When synaptic transmission was blocked , the number of sub-regions of the ERF was reduced , usually to a single excitatory ERF . This suggests that direct cell activation can be modeled accurately by a one-dimensional model with linear interactions between electrodes , whereas indirect stimulation due to summated presynaptic responses is nonlinear .
Implantable neural stimulation devices have demonstrated clinical efficacy , from the facilitation of hearing for deaf people using cochlear implants [1] to the treatment of neurological disorders such as epilepsy , Parkinson's disease , and depression using deep brain stimulation [2] . Additionally , neural stimulators are being used clinically for the restoration of sight [3–5] . Most stimulating neuroprostheses operate in an open-loop fashion; they do not adjust the stimulation by sensing how the stimulation affects the system . Devices that can both sense and stimulate will enable the development of new implants that may offer tighter control of neural activation and lead to improved patient outcomes [6] . The success of future retinal prostheses may benefit greatly from the ability to control spatiotemporal interactions between stimulating electrodes . For example , this may allow the design of stimulation strategies that better approximate the spiking patterns of normal vision . To this end , mathematical models that can predict responses to electrical stimuli are critical . A successful approach for extracting visual receptive fields uses models estimated from optical white noise stimulation patterns , which predict retinal responses [7–9] and responses in visual cortex [10 , 11] . These models use high-dimensional random stimuli and rely on the identification of a low-dimensional stimulus subspace to which the neurons are sensitive . The features , or receptive fields , describe the spatial , temporal , or chromatic ( for light stimuli ) components of the stimuli to which the neurons are most sensitive . The low-dimensional subspace is commonly identified using spike-triggered average ( STA ) and spike-triggered covariance ( STC ) analyses [7 , 12 , 13] but other methods , such as spike information maximization , can be used [14–17] . In all of the aforementioned models , a stimulus is projected onto a feature subspace and then transformed nonlinearly to estimate the neuron’s firing rate . Generally , the accuracy of the model depends on the accurate identification of the low-order subspace . Our previous work [12] demonstrated that short-latency RGC responses to electrical stimulation could be accurately described using a single linear ERF , and similarly for cortical responses [18] . In Maturana et al . [12] , short-latency intracellular recordings were analyzed ( i . e . , responses within 5 ms of stimulus onset for which synaptically mediated network effects were not apparent ) . In the present study , we used extracellular recording because this is currently the only clinically viable method to measure retinal signals . Due to the presence of stimulation artefacts , we analyzed long-latency activity ( >5 ms from stimulation onset ) , which arises largely from the activation of retinal interneurons [19] . For such indirect activation , we find that ERFs often have multiple sub-filters that can be estimated using a Generalized Quadratic Model ( GQM ) [16] , with maximum likelihood methods , to accurately identify the low-dimensional subspace . Such maximum likelihood approaches have been shown to outperform regular STC analysis , revealing additional feature dimensions and more accurately predicting responses [15–17] . We present an approach using the GQM to recover spatiotemporal ERFs during electrical stimulation of the retina . This study explores RGC responses to biphasic charge-balanced pulse stimulation applied at three stimulation frequencies: 10 , 20 and 30 Hz . Our work used large diameter stimulating electrodes because of their relevance to clinical application [5] . Moreover , we explore suprathreshold electrical pulses of a similar duration to those used clinically ( 0 . 5 ms per phase ) . Nonlinear responses in RGCs during light stimulation have been shown to arise from activation of presynaptic neurons [20–26] . Therefore , it is conceivable that electrical stimulation of the retina may also lead to nonlinear responses in RGCs . Indeed , our results demonstrate that electrical stimulation of the retina leads to a range of nonlinear excitatory and suppressive effects . The GQM accurately predicts nonlinear neural responses and outperforms models based on a linear subspace . When synaptic transmission is blocked , we show that the model becomes linear and can be described by a simpler one-dimensional description , as in our previous study [12] .
The responses from 77 RGCs were recorded . These included 16 ON , 35 OFF , and 12 ON-OFF cells; the remaining 14 recordings had no identifiable light response or the cell morphology could not be recovered ( for intracellular recordings ) . However , the responses to electrical stimulation from the non-classified cells were analyzed to verify the ability of the model to deal with responses of arbitrary types . Initially , cells were stimulated with a spot of light centered at the recording electrode location . Responses to light stimuli lasting 50 s ( consisting of repeated periods of 10 s light on , 10 s light off ) were recorded and repeated . The average spike rate and instantaneous change in spike rate were analyzed . A spike cluster analysis of the light response recordings confirmed whether the recordings were from single isolated cells . Fig 1A depicts a typical ON cell response during a transition from light off to light on . The white triangle shows the time when the light was switched on . This cell responded with more spikes during periods of light on compared to light off and produced more spikes during a transition from light off to light on . In this recording , smaller action potentials from another cell are also visible , however , these were not used . Fig 1B shows the average response during periods of light on and light off ( top ) and the instantaneous firing rate one second before and after transitions to light on and light off ( bottom ) . While we could not be certain that extracellular recordings were from the cell soma , in general , the recorded spikes had a distinctive biphasic shape with a large positive deflection towards the end of the spike ( Fig 1A , inset ) . Studies have suggested that the origin of these spikes are somatic or proximal to the soma rather than distal axonal spikes [27 , 28] . Stimulation artefacts were removed online using a 5 ms sample-and-hold circuit . High quality spikes between the stimulation pulses were observed , enabling online spike detection ( Fig 1C ) . Note that the filter used during electrical stimulation had a much narrower band-pass range than that used during light stimulation; hence , the spike shape changed considerably ( see Methods ) . The filter did not influence the frequency or timing of the spikes , only the appearance of the spikes . Stimulation artefacts during intracellular recordings were also blanked ( i . e . removed from the analysis ) in a similar manner to extracellular recordings; however , this was only for the duration of the stimulation pulse ( ~1 ms ) . Using this technique , high quality action potentials with very low latency were recorded ( Fig 1D ) . The major aim of this study was to develop mathematical models that could accurately relate RGC responses to arbitrary patterns of electrical stimulation . Specifically , we are interested in the influence of pre-synaptic neurons , which are known to produce nonlinear effects on RGCs [25] . To model RGC responses , electrical Gaussian white noise stimulation was applied to the retina using a 20-electrode array . The array was positioned subretinally; i . e . on the photoreceptor side of the retina . Every RGC response was related to the applied stimulus over a 300 ms window preceding the response . An initial estimate of the spatiotemporal linear filters that affected the neuron’s response to electrical stimulation were obtained using spike-triggered average ( STA ) and spike-triggered covariance ( STC ) analysis ( Materials and Methods ) . A second estimate of the spatiotemporal filters was obtained from the log-likelihood maximization of a General Quadratic Model ( GQM , Eq ( 4 ) , Materials and Methods ) initialized with the significant filters obtained from STA/STC . Electrical pulses were applied at three stimulation frequencies: 10 , 20 and 30 Hz . From the total of 77 cells , 30 cells were stimulated at 10 Hz , 22 cells at 20 Hz , and 44 at 30 Hz . Note that a subset of cells ( 11 of the 77 ) were stimulated at all three frequencies . The steps of the analysis for a sample cell stimulated at 20 Hz are shown in Fig 2 . This ON cell produced four significant eigenvalues from the STC analysis , two of which were excitatory and two suppressive ( Fig 2A ) . The four corresponding GQM component filters after log-likelihood maximization are shown in Fig 2C , where v→1 and v→2 represent the excitatory components and v→3 and v→4 represent the suppressive components . These filters can be interpreted as components of the electrical receptive fields ( ERFs ) of the cell . While a total period of T = 300 ms was explored , only stimulus frames in the period up to 50 ms significantly affected the neuron’s response . A linear component ( v→0 ) was also computed but it was small , with no significant electrodes determined by a bootstrap test ( see Materials and Methods ) and hence not shown . The observed responses were binned and plotted against the generator signal , which integrates the output of the four filters via a quadratic nonlinearity ( Eq ( 2 ) Materials and Methods; each bin contained 200 stimuli , red dots in Fig 2B ) . From this , a second nonlinearity could be estimated by fitting a sigmoid ( Eq ( 5 ) ) to the mean of each bin . A log-exponential fit is also shown for comparison , which shows a poor fit for high values of the generator signal . Above each plot in Fig 2C are histograms showing the distribution of input stimuli ( black ) and the number of spikes per stimulus ( grey ) as a function of the stimulus projection onto each ERF component . For the excitatory components , the responses were bimodal , with v→1 showing a slight preference towards negative amplitudes . All cells had one or more significant excitatory components from STC analysis . Most also had significant suppressive components . Fig 3A and 3B summarizes the number of excitatory and suppressive ERF components for all cells . Stimulation at 20 Hz and 30 Hz tended to produce a larger number of significant components . The duration over which electrical stimulation affected a neuron’s response was unique to individual cells ( Fig 3C ) . Electrodes that significantly contributed to a cell’s excitatory response tended to be constrained within a short latency from the response . For example , v→1 and v→2 in Fig 2B had significant electrodes only at a latency of 0 ms . Suppressive ERF components could extend up to 200 ms , but this varied across cells . For example , v→3 and v→4 in Fig 2B had significant electrodes at a latency of -50 ms . For the cell in Fig 2B , no components had significant electrodes that extended past -50 ms , hence this period constituted the integration time of this neuron . The average integration time for excitatory components was 22 ( ±40 SD ) ms and for suppressive components was 58 ( ±54 SD ) ms . A major goal of electrical stimulation of the retina is to achieve selective stimulation of the ON and OFF retinal pathways . We compared responses between ON , OFF , and ON-OFF cells to see if any differences could be exploited . Excitatory ERFs , where all electrode amplitudes had the same polarity ( i . e . , v→1 from Fig 2C ) , were compared to see if there was a cell-type specific preference to cathodic-first or anodic-first stimulation . To compare the preference to stimulus polarity , the normalized maximum spike-rate achieved for positive and negative amplitudes along the ERF was used ( e . g . arrows in histogram from v→1 from Fig 2C ) . Overall , a higher saturation rate was observed with anodic-first stimulation in ON cells ( mean anodic-first 0 . 90 c . f . mean cathodic-first 0 . 73 , F ( 1 , 27 ) = 5 . 02 , p = 0 . 03 , N = 14 ERFs ) , suggesting a higher dynamic range could be achieved with anodic-first pulses ( Fig 3D ) . The opposite was observed in OFF cells; OFF cells preferred cathodic-first pulses ( mean anodic-first 0 . 79 c . f . mean cathodic-first 0 . 92 , F ( 1 , 77 ) = 10 , p = 0 . 002 , N = 39 ) . No statistical difference was found among ON-OFF cells ( F ( 1 , 27 ) = 0 . 14 , p = 0 . 71 , N = 12 ) . No other differences ( number of components , temporal or spatial characteristics ) were found between ON , OFF or ON-OFF cells . To test the accuracy of the model , we compared the model prediction ( trained on 80% of data ) against data from a validation set ( 20% of data ) not used during estimation of the model parameters . Fig 4 shows the model validation for the example cell depicted in Fig 2 . Fig 4A compares the prediction to the mean observed response; each point represents the mean response to 200 stimuli . The prediction and observed responses agreed very well ( R2 = 0 . 95 ) , producing an average error of approximately 0 . 12 spikes per bin ( 4 . 3% error relative to max response ) . By comparison , a best-case R2 value of 0 . 98 ( computed by assuming a Poisson process , see Materials and Methods ) was obtained . To examine the assumption of Poisson spike train statistics , the variance of the responses was computed and normalized by the maximum predicted response at each bin location ( Fig 4B ) . If a cell has Poisson spiking statistics , the variance should be equal to the mean ( grey dashed line ) . For this example cell , the variance was approximately equal to the mean , which is consistent with Poisson-like statistics ( R2 = 0 . 84 , F ( 1 , 145 ) = 784 , p = 3 . 9x10-60 ) . The contours of fixed response expectation ( Eq ( 6 ) , Materials and Methods ) were computed using the predicted model parameters and compared to the measured response contours when the data was projected onto two of the linear filter components ( Fig 4C ) . The experimental data conformed well to the curves predicted by the model , although this was noisy in the cases of the projections onto v→2 and v→3 . This is most likely because these components were significantly smaller ( see components in Fig 2A ) than v→1 and v→4 . This example also highlights a constraint of the GQM in that it assumes a symmetric quadratic nonlinearity , which may not be the case in general . However , the analysis provided a qualitative confirmation that this model is an appropriate description of neural responses to electrical stimulation . Moreover , it demonstrated that the data was highly nonlinear , and a model with a linear generator function is not an appropriate description of the spike probability . The model could accurately predict responses for most cells , demonstrated by a high coefficient of determination ( Fig 5A ) . The average error for all cells was 0 . 14 spikes/bin ( average of 6 . 45% error relative to max response ) . No significant differences were found between the model performance at 10 , 20 , and 30 Hz ( Fig 5C ) . For each cell , the calculated coefficient of determination was compared to the best-case value , measured by simulating a Poisson process with a mean given by the model prediction . Overall , most cells achieved a high coefficient of determination ( Fig 5D , mean R2 = 0 . 75 ) . In all cells where R2 was below 0 . 5 , relatively few spikes were collected ( <10 , 000 spikes , c . f . average ~30 , 000 spikes for all other cells ) . For many cells , the responses were Poisson-like; the variance of the response was approximately equal to the mean ( Fig 5B , grey ) . However , a trend of lower variance at high amplitudes was observed ( Fig 5B , black ) , suggesting that stimulating in a binary manner ( zero or maximum current ) rather than using intermediate amplitudes could lead to responses with lower relative variance . RGCs are known to spatially integrate light stimuli in a nonlinear fashion , which is mediated by presynaptic neurons [25] . Electrical stimulation of the retina activates retinal interneurons [19 , 29] , which are likely to produce nonlinear responses . For this reason , our model included nonlinear subunits . This is in contrast with previously published models [12 , 30 , 31] , that found that electrodes interact in a largely linear fashion . An important difference between the present study and the investigations by Jepson et al . [30] and Maturana et al . [12] is that the latter only considered short latency spikes that occurred generally <5 ms from stimulus onset . In contrast , the extracellular recordings reported above were all long-latency spikes ( >5 ms ) . If the origin of the nonlinear subunits is indeed a result of presynaptic neurons , then abolishing presynaptic input should recover a linear model similar to previous investigations if the short-latency spikes are recorded . To test this , we recorded from nine cells using an intracellular patch clamp ( which allowed the recovery of short-latency spikes ) , with normal Ames extracellular solution mixed with 250 μM CdCl2 , which interferes with the release of neurotransmitters from nerve terminals . Fig 6A and Fig 6C compare raster plots of the spike times of a sample cell before and after the application of CdCl2 . Prior to CdCl2 , the histogram of spike times shows a distinct long latency peak , which is abolished by the addition of CdCl2 , leaving only the short latency peak . This confirms that the long latency spikes are of presynaptic origin ( >5 ms ) while short latency spikes are the result of direct activation ( <5 ms ) , and is consistent with previous studies [19 , 32] . Fig 6B shows the eigenvalues from STC analysis of the responses , which shows three significant components . All but the first component disappeared after application of CdCl2 ( Fig 6D ) , suggesting that two components , including the suppressive component , were of presynaptic origin . Importantly , the reduction from two or more to a single component simplifies the model from a nonlinear to a linear generator function . To see this , we note that a model of the form E ( St→ ) =N ( v1→ . St→ , v2→ . St→ , … , vN→ . St→ ) is composed of a nonlinear function of a generator potential , which is in turn a function of the stimulus subspace: E ( St→ ) =N ( g ( v1→ . St→ , v2→ . St→ , … , vN→ . St→ ) ) . The generator potential , g ( v1→ . St→ , v2→ . St→ , … , vN→ . St→ ) , is akin to the cell membrane potential , while the nonlinearity captures the conversion of this potential into an expected spike rate . In the simplest case , if the generator potential is linear in each argument , vN→ . St→ , then it can be rewritten as a one-dimensional linear subspace: g=∑ ( winvN→ . St→ ) = ( ∑wnvN→ ) . St→=vnew . St→ , where vnew=∑wnvN→ defines the new one-dimensional subspace . If , however , the generator potential is a nonlinear function of two of more augments , then the subspace remains of two or more dimensions . In other words , a generator function of two or more dimensions is nonlinear except in the case where it can be written as a linear sum , thereby becoming one-dimensional . Application of CdCl2 reduced the number of significant components observed in the nine cells , simplifying the model . Fig 7 shows the numbers of significant components before and after application of CdCl2 . Seven of nine cells were reduced to a single significant excitatory component , and eight of nine cells were reduced to zero suppressive components . This result is consistent with our previous observation [12] , which showed that short-latency responses always produced a significant excitatory component , and sometimes some other small significant components . However , omission of these smaller components did not alter the accuracy of a model with a single linear filter . Application of CdCl2 also altered the duration over which the neuron integrated electrical stimulation . After application of CdCl2 , the average integration time was reduced from an average of 89 ms to 44 ms . To test if the GQM presented here was comparable or better than other models , we tested the performance of the GQM against other commonly used models . First , we compared the subspace spanned by the linear component v→0 , to the subspace spanned by the higher order components . It is possible that v→0 , which represents a linear approximation to the quadratic model , may span part of the stimulus space that is not included in higher order components . To test this , we compared the linear filter , v→0 , to higher order quadratic components from the GQM by estimating how much of v→0 lies in the subspace spanned by the higher order components . Fig 8A show v→0 for two sample cells along with the weighted sum of higher order components ( u→0 , Eq ( 7 ) ) . In both cases , u→0 and v→0 appear very similar , demonstrating that v→0 spans the same subspace as the higher order components . For all cells , a high cosine similarity index was produced between u→0 and v→0 . Next , we compared the coefficient of determination between the predicted and measured responses for three other models , on a data set not used to fit the model . The GQM presented here was compared to a linear model using a projection onto the first STC component [STC1 , 12] , a two-dimensional STC analysis [STC2 , 33] , and the NIM [15] ( Fig 8B ) . A linear model using only the STA is not shown , since for most cells it could not predict the responses . The main reason for this was that most cells were sensitive to both polarities of stimulation , leading to an almost zero STA . Similar to the GQM model , both STC and NIM models incorporate a sum of non-linear receptive field terms , but do not make the assumption that the non-linearity is quadratic . While this makes them more general , it also means they can require more data to extract the same effective number of receptive fields . On average , the GQM produced a higher coefficient of determination than the other three models ( Fig 8B ) . The GQM performed significantly better than both STC1 ( F ( 1 , 191 ) = 10 . 03 , p = 0 . 002 ) and the STC2 ( F ( 1 , 204 ) = 5 . 03 , p = 0 . 026 , 1-way ANOVA ) . No significant differences were found between the GQM and NIM models . Patterns of electrical stimulation in blind patients is likely to contain high order correlations that are not present in Gaussian stimuli . Our model should therefore be robust to other stimuli that contain correlations . To gauge the capacity of the model to predict responses to non-white stimuli , eleven cells were also stimulated with a set of structured stimuli generated from a series of images ( see Materials and Methods ) . The images included static wave gratings , spots and lines , and gratings moving in random directions . Importantly , these stimuli contain second and higher order correlations between electrodes . For each cell , the model parameters were estimated from white noise stimuli and used to predict responses to the stimuli from the images . Fig 9A shows the distribution of the electrode amplitudes for Gaussian stimuli ( black ) and the images ( grey ) for a sample cell , demonstrating very different stimulus distributions between the two sets of stimuli , even without considering correlations between electrodes . With both types of stimuli , the model could accurately predict responses ( Fig 9B ) , suggesting that the model generalized well for non-white and correlated stimuli . The average error for Gaussian and patterned stimuli was 0 . 12 spikes/bin ( ~3 . 6% error relative to maximum response ) in both cases . The accuracy of model predictions for Gaussian stimuli tended to be higher than for the images ( Fig 9C ) . The GQM and NIM performed significantly better than an STC1 ( p < 5x10-5 in both cases , 1-way ANOVA ) . Performance for the GQM and NIM were similar ( F ( 1 , 21 ) = 0 . 02 , p = 0 . 88 , Fig 9D ) . The model extracted diverse varieties of spatiotemporal ERFs . For most cells , the significant electrodes in the excitatory ERFs were limited to the stimulus frame immediately prior to the response . Cells that had more than one excitatory component tended to have electrodes that occurred in the same stimulus frame but were spatially different ( e . g . , v→1 and v→2 , Fig 2B ) . However , some cells did have excitatory components with significant electrodes that extended over more than one stimulus frame . Fig 10A shows an example of a cell stimulated at 30 Hz that produced three excitatory components . The first two components ( v→1 and v→2 ) have significant electrodes in the same stimulus frame but are spatially different . The third component ( v→3 ) is spatially aligned with v→1 but has significant electrodes in a stimulus frame prior to that of v→1 . Cells that produced significant suppressive components tended to be spatially aligned to the excitatory ERFs but occurred in preceding stimulus frames ( e . g . , v→3 and v→4 , Fig 2B ) . However , suppressive components could also have significant electrodes in the same stimulus frames as the excitatory ones but with different spatial characteristics . Many cells produced suppressive components that extended over several stimulus frames . Fig 10B shows the suppressive components for the same cell in Fig 10A . Both v→4 and v→5 were spatially aligned , and aligned also to v→1 , but were not similar in their temporal characteristics to any excitatory component . Moreover , both suppressive components have significant electrodes at a lag of -33 ms , which is when the excitatory component , v→3 , occurred . To test the effect of stimulation frequency , responses from a subset of 11 of the 77 cells were compared at the three stimulation frequencies: 10 , 20 and 30 Hz . In all cases , the ERFs were similar across frequency , suggesting that stimulation frequency had little influence on the shape or temporal properties of the ERFs . Fig 11 demonstrates two example responses that are representative of all cells in the subset . The cell in Fig 11A produced two excitatory components ( v→1 and v→2 ) and a single suppressive component ( v→3 ) . All components occurred in the same stimulus frame ( lag zero ) but were spatially different . Across the three stimulating frequencies , the components were very similar , with the greatest difference occurring for v→2 at 10 Hz . The cell in Fig 11B produced one excitatory component ( v→1 ) and two suppressive components when stimulating at 20 and 30 Hz ( v→2 and v→3 ) . Due to the different stimulation rates , pulses occurred at different times preceding the response between 0 and 100 ms but coincided at these two times . Both v→2 and v→3 show a continuity of the significant electrode over this time period across frequencies , suggesting that the effect of stimulation frequencies is to simply sample from a single consistent ERF . Note also that , while v→2 and v→3 were different at intermediate lags between 0 and -100 ms , the two components were the same at -100 ms . Hence only one suppressive component ( v→2 ) was observed when stimulating at 10 Hz . Fig 12 shows a sample of the various ERFs observed at the three stimulation frequencies . The green dots in the frames denote the approximate recording electrode locations and red frames denote suppressive ERF components . The average ERF size ( 0 . 42 ± 0 . 46 SD μm ) , was calculated by computing a weighted average distance between significant electrodes and the recording location [12] .
This work compared the performance of a General Quadratic Model ( GQM ) to three other models: a linear model using a one-dimensional projection , a two-dimensional STC analysis , and the Nonlinear Input Model ( NIM ) . The performances of GQM and NIM were very similar; both models could accurately predict responses and the correlation between prediction and response was similar in both cases . In general , the number of quadratic components in the GQM and components in the NIM were the same . However , the time taken to fit the NIM parameters was considerably longer than for GQM and increased with the number of components . The simplest description of the spike-triggered subspace is found by analyzing the spike-triggered average . Sekhar et al . [31] analyzed the STA produced by electrical stimulation of one electrode . The amplitudes of currents produced from the STA were found to change over time , often changing polarity from negative to positive or vice versa . Similarly , our results showed that for many cells stimulated at 20 and 30 Hz , the linear component , v→0 , produced significant electrodes that spanned a number of stimulus frames and changed polarity over time ( i . e . Fig 8A ) . In contrast , the majority of excitatory or suppressive ERFs had significant electrodes that were confined to one stimulus frame . For all cells with a significant linear component , v→0 could be accurately described as a linear combination of the quadratic components from the GQM . Also , the STA could be accurately described as a linear combination of significant STC components , suggesting that the higher order components spanned the same subspace as the STA . However , the advantage of using a model with second-order components is that the components separately characterize excitation and suppression and it can provide more predictive power by capturing the important nonlinearities in the responses , which are neglected by the STA .
Methods conformed to the policies of the National Health and Medical Research Council of Australia and were approved by the Animal Experimentation Ethics Committee of the University of Melbourne ( Approval Number: 1413306 ) . Data were acquired from retinae of Long-Evans rats ranging from 1 to 6 months of age . The animals were initially anesthetized with a mixture of ketamine and xylazine prior to eye enucleation . After enucleation , the rats were sacrificed with an overdose of sodium pentobarbital ( 350 mg intracardial ) . Dissections were carried out in dim light conditions to avoid bleaching the photoreceptors . After hemisecting the eyes behind the ora serrata , the vitreous body was removed and each retina was cut into two pieces . The retinae were left in a perfusion dish with carbogenated Ames medium ( Sigma ) at room temperature until used . Pieces of retina were mounted on a multi-electrode array ( MEA ) with ganglion cell layer up and were held in place with a perfusion chamber and stainless steel harp fitted with Lycra threads ( Warner Instruments ) . Once mounted in the chamber , the retina was perfused ( 4–6 mL/min ) with carbogenated Ames medium ( Sigma-Aldrich , St . Louis , MO ) at 32–35°C . Recordings were obtained using two extracellular electrodes as described below in more detail , or using whole cell intracellular recordings . Extracellular potentials were recorded with custom-made carbon fiber electrodes ( diameter ~7 μm ) pulled in a glass pipette ( Fig 13 ) . The carbon fiber electrodes were used because they could reliably detect high-quality single cell spikes . Two electrodes were manipulated above the retinal surface ( Sutter Instrument , MPC-200 ) until action potentials from both electrodes were obtained . Voltage recordings from two or more cells on each electrode were simultaneously collected , amplified , digitized with 18-bit precision at 50 kHz ( Tucker Davis Technologies: RZ2 base station and PZ2 multichannel acquisition ) , and stored for offline analysis . Recordings lasting approximately 4 hours were obtained during each experiment . Extracellular recordings were filtered and processed , and spikes were detected online using a custom-built circuit that was programmed into the digital signal processor on the Tucker Davis Technologies RZ2 base station . A first-order Butterworth band-pass filter with frequency range 1–5 kHz was used to make the stimulation artefact easier to remove , allowing for online detection of spikes . Stimulation artefacts were removed using a sample-and-hold circuit that captured the voltage prior to stimulation and held its value for 5 ms , such that a constant signal was recorded during stimulation . Spikes in the remaining waveform were detected and counted using threshold detection ( see following sections for details ) . For each extracellular recording , a signal-to-noise ratio ( SNR ) was calculated to assess the quality of the extracellular recording . The SNR was calculated by collecting all detected spikes and calculating the ratio of the spike amplitude to the standard deviation of the waveform noise [49] , SNR=max ( W¯ ) −min ( W¯ ) 2ϑn , ( 1 ) where W¯ is the average time-course of the spike waveform and ϑn is the standard deviation of the noise . The SNR was calculated for every stimulus train . Recordings that produced SNR values lower than 4 were discarded . To confirm the origin of the observed responses , intracellular recordings were conducted in some experiments . This allowed high-quality recording of low latency activity ( <5 ms from stimulus onset ) , which is difficult to obtain with extracellular recordings . In these experiments , synaptic transmission was blocked after approximately 1 hour of recording responses in normal Ames medium . To block synaptic transmission , extracellular solution containing 250 μM Cadmium Chloride ( CdCl2 , Sigma-Aldrich ) was applied . This agent was applied directly to the Ames medium and perfused over the retina . Intracellular action potentials were detected using threshold crossings after the stimulation artefacts were blanked ( 1 ms blanking ) . This allowed the detection of spikes with very low latency ( <5 ms ) as demonstrated previously [12] . Whole cell intracellular signals were recorded using standard procedures [50] . To obtain responses from RGCs , a small hole was made in the inner limiting membrane to expose some RGC somas . A pipette was then filled with internal solution containing ( in mM ) 115 K-gluconate , 5 KCl , 5 EGTA , 10 HEPES , 2 Na-ATP , 0 . 25 Na-GTP ( mosM = 273 , pH = 7 . 3 ) , Alexa Hydrazide 488 ( 250 mM ) , and biocytin ( 0 . 5% ) . The initial pipette resistance in the bath was in the range 6–8 MΩ . Prior to recording , the pipette resistance was nulled using a bridge balancing circuit and the capacitance was compensated on the amplifier head stage . Voltage recordings were obtained in current-clamp mode , amplified ( SEC-05X , NPI electronic ) , and digitized with 16-bit precision at 25 kHz ( National Instruments BNC-2090 ) . Cells were classified as ON , OFF , or ON-OFF by their responses to a spot of light ( ~100 μm diameter ) centered at the recording electrode location . Cells were illuminated with stimuli lasting a total of 50 sec , consisting of alternating periods of light and dark , with each period lasting 10 sec . The stimuli were repeated 5–10 times . The average spike rate during dark periods was compared to the average spike rate during light periods . Additionally , the instantaneous change in spike rate from one second before to one second after a change in illumination was compared . Cells were classified as ON if they produced a larger number of spikes during light on periods compared to light off periods , or if they produced a greater number of spikes during the transition to a light on period . The opposite was true for OFF cells . Cells were classified as ON-OFF if they produced spikes during a transition from light on to light off and during a transition from light off to light on . Extracellular action potentials in response to the light stimuli were recorded and filtered online using a first-order Butterworth band-pass filter with bandwidth 500–15 , 000 Hz . A threshold of 5 times the standard deviation of the recording was set for spike detection . Spikes crossing this threshold in the positive or negative direction were detected . To determine if detected spikes were from the same cell , we performed an offline spike cluster analysis ( WAVECLUS , Quiroga et al . [51] ) on the detected spikes from the light response recordings . WAVECLUS compares and clusters recorded spikes to determine which spikes are similar . Only recordings where single cells were identified were used . Electrical stimulation was applied using an electrode array described previously [12] . The array was fabricated on a glass coverslip using lithographic methods to produce 20 platinum electrodes . Electrodes were spaced in a hexagonal arrangement with center-to-center spacing of 1 mm ( Fig 13 ) . Each electrode was circular with a diameter of 400 μm . The total stimulating area spanned approximately 3 . 5 mm x 3 . 5 mm . Stimulation consisted of biphasic pulses applied simultaneously to all electrodes at a frequency of 10 , 20 or 30 Hz . Pulses of equal phase duration ( 500 μs per phase ) with an interphase gap ( 50 μs ) and random amplitude were chosen independently for each electrode . The random amplitudes were sampled from a Gaussian distribution with standard deviation σ = 150 μA . A stimulus vector , s→ , of length 20 ( equal to the number of active electrodes ) was generated by sampling each element from a Gaussian distribution . If the amplitude of stimulation of an electrode exceeded the stimulator limits ( ±300 μA ) , then the amplitude value was discarded and a new value was generated from the distribution . Positive stimulus amplitudes refer to anodic first stimuli , while negative amplitude refer to cathodic first . A stimulus train lasting 1 minute was generated and applied to the retina . The experiment continued for up to 4 hours with approximately 100 , 000 total stimuli applied at 10 Hz ( proportionally more at higher frequencies ) . Stimulation waveform signals were generated by a custom-made MATLAB ( MathWorks , version 2014a ) interface commanding a multichannel stimulator ( Tucker Davis Technologies: RZ2 base station and IZ2 multichannel stimulator ) . Similar to Park and Pillow [16] , we consider a General Quadratic Model ( GQM ) where a quadratic generator signal is used to map a stimulus onto the real line: g ( St→ ) =v→0∙St→+∑i=1Nwi ( v→i∙St→ ) 2 , ( 2 ) where St→ represents the set of stimuli presented over the time period [t−T , t] , v→i ( i = 0 , … , N ) represent linear filters that span the relevant stimulus subspace , and N represents the dimension of the stimulus subspace . wi is either +1 or –1 depending on whether v→i is excitatory ( positive ) or suppressive ( negative ) . A 300 ms window was chosen for T since this has been shown to be a conservative estimate of the duration over which the retina integrates visual information and , hence , constitutes the “memory” of RGCs [7] . Furthermore , preliminary experiments showed that this captured the period over which stimulus history appeared to influence RGC responses to electrical stimulation . The spike rate was estimated by a nonlinearity operating on the generator signal: E ( St→ ) =N ( g ( St→ ) ) , ( 3 ) where E ( St→ ) represents the expected number of spikes in response to stimulus St→ . The parameters of Eqs ( 2 ) and ( 3 ) were estimated by assuming that each neuron’s spikes were described by an inhomogeneous Poisson process with the firing rate function given by E ( St→ ) , and maximizing the log-likelihood of the model parameters given an observed set of responses: L=∑tR ( t ) log ( E ( St→ ) ) −E ( St→ ) , ( 4 ) where R ( t ) is the observed response at time t . The optimization was carried out using the NIMtoolbox [15] in MATLAB . When the linear filters v→i ( Eq ( 2 ) ) were found , the components were orthogonalized and scaled according to their eigenvalues ( such that the length of v→i was the square-root of its eigenvalue ) . The optimization procedure used in the NIMtoolbox assumes a log-exponential type nonlinearity for Eq ( 3 ) . However , for our observed responses , the log-exponential nonlinearity was not an accurate fit for large values of the generator function , so a sigmoid nonlinearity was used instead: N ( g ( x ) ) =a1+exp ( −b ( g ( x ) −c ) ) , ( 5 ) where a represents a scaling factor that determines the saturation amplitude of the sigmoid , b represents the gain of the sigmoidal curve , and c represents the threshold ( 50% of the saturation level ) . To determine which electrodes in each receptive field significantly affected the cell’s response , the optimization procedure was repeated 1000 times using randomly time-shifted versions of the spike-triggered stimuli; i . e . , randomly time shifting R ( t ) by t′ . This gave a distribution for the linear and second-order components to which the true components could be compared . Electrode amplitudes that were greater in absolute value than two standard deviations were considered significant . Fig 14A depicts an example where a cell’s filter is compared to a distribution over three stimulus epochs ( t−2 , t−1 , and t0 ) . Each epoch represents the current amplitude applied to the 20 stimulating electrodes at the corresponding times preceding the response . A diagrammatic description of the whole model is shown in Fig 14B . A fixed value of the generator function ( Eq ( 2 ) ) describes a quadratic surface of constant mean response . By projecting the stimulus-response data onto any two major components , the surface can be estimated in two dimensions by an ellipsoid ( for excitatory-excitatory dimensions ) or hyperbola ( for excitatory-suppressive dimensions ) with major and minor axes weighted by the strength ( or norm ) of each component . The equation describing this surface is: g′ ( x , y ) =kz+kx+ky+wix2+wjy2 ( 6 ) where x=v→i∙St→ , y=v→j∙St→ . kx and ky describe an offset from zero ( due to a non-zero projection of v→0 onto either v→i or v→j in Eq ( 2 ) ) , kz describes a DC offset , and w are the same weights as in Eq ( 2 ) . The contours of the surface described by g′ ( x , y ) were used as a qualitative measure to test the suitability of the GQM by comparing it to the contours in the response data . The simplest low-dimensional subspace that characterizes the neuron’s response can be described by the spike-triggered average ( STA ) , v→STA [7] . This is found by taking the average stimulus that generated a response . v→STA can be used to capture the deviation between the spike-triggered and raw set of stimuli if the neuron has a nonlinearity such that the mean of the spike-triggered distribution is shifted away from the raw stimulus set . Second-order components that best capture the difference between the spike-triggered stimuli and the raw ensemble can be found by performing a spike-triggered covariance ( STC ) analysis [12 , 13 , 33] . Significant eigenvectors of the STC matrix span a stimulus subspace that generates an excitatory or suppressive influence on the neuron . Prior to maximization of Eq ( 4 ) , components of Eq ( 2 ) were initialized by v→STA for v→0 and STC components for v→i ( i > 0 ) . The number of significant excitatory and suppressive components in Eq ( 2 ) ( i . e . , the dimensionality N ) was found by incrementally increasing the number of excitatory and suppressive components until the best model performance was obtained . Model performance was measured by comparing the model prediction to the observed response from a validation set of data ( see next section for details ) . The accuracy of the model was evaluated by comparing model predictions to observed responses . The data from each cell was divided into five equal sets , four of which were combined to estimate the model parameters and one that was used to test the model’s prediction . To ensure that each result was not an effect of a particular choice of data , the results were cross-validated by repeating the analysis with a different set of the originally partitioned data . The mean values of the five cross-validated sets are reported . All reported p values were calculated using the Tukey-Kramer test unless otherwise stated . Fig 15A demonstrates the validation for an example cell . The measured spike counts from the validation set ( grey dots ) are plotted against the predicted spike count ( black line ) , binned into groups of 200 stimuli . The average spike rate in each bin ( black dots with error bars ) was used to compute an average prediction error . The average error was calculated as the mean difference between the predicted and measured spike count for each bin . For the example in Fig 15A , the average error is 0 . 15 spikes/bin ( 4 . 5% error relative to maximum response of 3 . 3 spikes ) . Additionally , a coefficient of determination between the mean response and the model fit was computed ( R2 = 0 . 93 for the example in Fig 15A , p < 0 . 01 ) . To obtain an upper bound of this value , a best-case coefficient of determination was estimated by assuming the neuron could be described by a Poisson distribution ( this assumption is later tested ) . At each bin location , a Poisson process was simulated ( black dots in Fig 15B ) , with a mean given by the model prediction ( black line in Fig 15B ) . Qualitatively , both the raw data ( grey dots ) and Poisson simulated data were similar . From the simulated data , a coefficient of determination was computed and compared to the observed value . For the example in Fig 15B , the best-case coefficient of determination is 0 . 98 . The spike triggered average ( STA ) is often used to describe retinal responses to light [7] and to electrical stimulation [31 , 52] . In these models , a linear kernel described by v→STA is used to model the neural responses . However , v→STA can often describe a linear combination of higher order components [17] and , therefore , may describe a combination of excitatory and suppressive stimuli . To test this , we compared the subspace spanned by the quadratic components in our model to the linear component , v→0 . We computed a weighted sum of higher order components , where the weights of each component were fitted to v→0 . The weighted vector was given by u→0 , such that u→0+ϵ→=Vb→+ϵ→=v→0 , ( 7 ) where b→ is a vector of weights that scale a linear combination of second order components ( v→i ) contained in the columns of the matrix V , and ϵ→ is the orthogonal remainder . b→ can be found by the relation b→=VTv→0 , ( 8 ) given orthonormal components in V . To test for the similarity between v→0 and u→0 , a cosine similarity measure was used , which is the cosine of the angle between two vectors . A similarity of 0° produces a value of 1; any non-zero angle produces a value less than 1 . Only cells that produced a v→0 with electrodes that contributed significantly to a cell’s ERF were used ( approximately half the cells ) . Significance was determined by the nested bootstrap method described in the previous section . Our model was compared to three other models . First , we compared our model to a linear model , described by projecting the spike-triggered data onto a single linear component . The linear component was found by taking the first excitatory component of an STC rather than the v→STA since , for many cells , the v→STA was close to zero . Furthermore , the first component from the STC produced a far more accurate prediction . The data was projected onto the component and a double sided sigmoid was fit to the projected data [12] . Second , we compared the GQM to an STC analysis using a two-dimensional nonlinearity . Only a two-dimensional model was used since including more components generally led to over-fitting and required more data for a robust prediction . The two-dimensional nonlinearity was estimated by projecting the spike-triggered data onto the two most significant components ( whose eigenvalues departed most from the null distribution , usually the first and last ) . From this , a two-dimensional smoothed surface was fit in MATLAB . A lowess fit with span of 0 . 05 was chosen for the surface . Third , we compared the GQM to the nonlinear-input model [NIM , 15] . The NIM models the synaptic inputs from excitatory and suppressive subunits using half-wave rectifying-type nonlinearities . This model is a more generalized version of the GQM that makes fewer assumptions about the higher order nonlinearities , but results in more parameters that need fitting . Like the GQM , the optimal number of components in the NIM were chosen by incrementally increasing the number of excitatory and suppressive components until the highest coefficient of determination between prediction and measured response , in a cross-validated set , was recorded . To test the model’s capability to predict responses to structured , non-Gaussian stimuli , a subset of cells were activated with stimuli generated from a set of images . The images were randomly generated from one of the four following modes defined on a 500 x 500 grid . Each pixel in an image was assumed to span 10 μm , such that the entire 500 x 500 image spanned 5 x 5 mm . The following examples are shown in Fig 16: All random features in modes 1–4 ( i . e . , spot locations , line orientations , etc . ) were chosen using a uniform distribution . The orientations of lines and gratings were randomly chosen in the interval of 0–180° in steps of 1° . The speed of the moving wave grating was randomly chosen in steps of 0 . 1 Hz , up to a maximum of 2 Hz . Each image from modes 1–3 was scaled such that the highest amplitude pixel had a value of +250 μA or –250 μA . Each image stayed constant for 180 ms . For the moving wave gratings , the set of images were scaled such that the maximum pixel in the set of images was +250 μA or –250 μA . The stimulating electrodes were superimposed on each image in their correct locations and the pixel amplitude at the center of each electrode was applied to the corresponding electrode . Stimulation was applied at 30 Hz in all cases , with each stimulus train lasting 30 sec . | Implantable neural stimulation devices are being widely used and clinically tested for the restoration of lost function ( e . g . cochlear implants ) and the treatment of neurological disorders . Smart devices that can combine sensing and stimulation will dramatically improve future patient outcomes . To this end , mathematical models that can accurately predict neural responses to electrical stimulation will be critical for the development of smart stimulation devices . Here , we demonstrate a model that predicts neural responses to simultaneous stimulation across multiple electrodes in the retina . We show that the activation of presynaptic neurons leads to nonlinearities in the responses of postsynaptic retinal ganglion cells . The model is accurate and is applicable to a wide range of neural stimulation devices . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] | [
"medicine",
"and",
"health",
"sciences",
"action",
"potentials",
"membrane",
"potential",
"ocular",
"anatomy",
"electrophysiology",
"light",
"neuroscience",
"electromagnetic",
"radiation",
"surgical",
"and",
"invasive",
"medical",
"procedures",
"eigenvalues",
"mathematics",
"functional",
"electrical",
"stimulation",
"ganglion",
"cells",
"algebra",
"membrane",
"electrophysiology",
"visible",
"light",
"bioassays",
"and",
"physiological",
"analysis",
"research",
"and",
"analysis",
"methods",
"animal",
"cells",
"electrophysiological",
"techniques",
"physics",
"cellular",
"neuroscience",
"retina",
"electrode",
"recording",
"cell",
"biology",
"anatomy",
"linear",
"algebra",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"afferent",
"neurons",
"ocular",
"system",
"physical",
"sciences",
"retinal",
"ganglion",
"cells",
"neurophysiology"
] | 2018 | Electrical receptive fields of retinal ganglion cells: Influence of presynaptic neurons |
The pathogen interference phenotype greatly restricts infection with dengue virus ( DENV ) and other pathogens in Wolbachia-infected Aedes aegypti , and is a vital component of Wolbachia-based mosquito control . Critically , the phenotype’s causal mechanism is complex and poorly understood , with recent evidence suggesting that the cause may be species specific . To better understand this important phenotype , we investigated the role of diet-induced nutritional stress on interference against DENV and the avian malarial parasite Plasmodium gallinaceum in Wolbachia-infected Ae . aegypti , and on physiological processes linked to the phenotype . Wolbachia-infected mosquitoes were fed one of four different concentrations of sucrose , and then challenged with either P . gallinaceum or DENV . Interference against P . gallinaceum was significantly weakened by the change in diet however there was no effect on DENV interference . Immune gene expression and H2O2 levels have previously been linked to pathogen interference . These traits were assayed for mosquitoes on each diet using RT-qPCR and the Amplex Red Hydrogen Peroxide/Peroxidase Assay Kit , and it was observed that the change in diet did not significantly affect immune expression , but low carbohydrate levels led to a loss of ROS induction in Wolbachia-infected mosquitoes . Our data suggest that host nutrition may not influence DENV interference for Wolbachia-infected mosquitoes , but Plasmodium interference may be linked to both nutrition and oxidative stress . This pathogen-specific response to nutritional change highlights the complex nature of interactions between Wolbachia and pathogens in mosquitoes .
Wolbachia pipientis ( Rickettsiaceae ) is an obligate bacterial endosymbiont that shows great potential as a natural control agent for a range of clinically important mosquito-transmitted pathogens , including those responsible for diseases such as malaria and dengue in humans [1 , 2] . Wolbachia naturally infect an estimated 40% of terrestrial insect species [3] . Infection often results in manipulation of host biology , with the nature and extent of these manipulations varying depending on the host and infecting Wolbachia strain [4 , 5] . Wolbachia are maternally transmitted , and heavily infect host ovaries . The bacterium is often associated with extreme manipulation of the host reproductive process , furthering bacterial propagation at the expense of host fitness [6] . These manipulations allow the bacterium to naturally spread into uninfected insect populations , and to move across large distances [7] . The most common reproductive manipulation is cytoplasmic incompatibility ( CI ) . CI-causing Wolbachia strains prevent or restrict viable egg production when uninfected females mate with Wolbachia-infected male insects , while Wolbachia-infected females can successfully breed with either infected or uninfected males . Infection can affect other host physiological processes including oogenesis [8] , chemosensory perception [9] and parasitism [10] . Some strains form mutualistic relationships with their hosts , contributing resources [11] , or enhancing fitness [12] , while others are metabolically dependent on their hosts , and the resources they consume [13 , 14] . Wolbachia naturally infect many mosquito species including Aedes albopictus and Culex pipiens , but not the primary dengue vector Aedes aegypti or most anopheline vectors of human malaria . Infections in Ae . aegypti have been generated in the laboratory via transinfection [15] , through the injection of cytoplasm from the eggs of a Wolbachia-infected donor species into Ae . aegypti embryos [16–18] . These infections cause pathogen interference [19 , 20] , a Wolbachia-induced decrease in susceptibility to infection with pathogens including the dengue ( DENV ) , chikungunya , yellow fever and West Nile viruses , filarial nematodes and some bacteria [17 , 21–24] . Pathogen interference can result in decreased pathogen load , and largely prevent disseminated viral infection and salivary transmission [17 , 25 , 26] . Interference against DENV has been thoroughly studied in Ae . aegypti infected with the wMel and wMelPop Wolbachia strains , with the strength and prevalence of the interference phenotype dependent on the viral isolate and serotype [25] . Wolbachia can also interfere with Plasmodium infection in mosquitoes , however interaction between the bacterium and these parasites appears to be more variable . The only stable Wolbachia transinfection in an anopheline mosquito , wAlbB in Anopheles stephensi , reduced Plasmodium falciparum oocyst and sporozoite numbers [27] . Infection with wMelPop in Ae . aegypti produced stronger interference against P . gallinaceum [23] . However , this effect may not be representative of how Wolbachia interacts with Plasmodium species that infect humans , given that the Plasmodium species that infect different animals are phylogenetically distinct [28] , and that there are genetic , metabolic and immunological differences between anopheline and culicine mosquitoes [29 , 30] . Prior to transinfection , Wolbachia infection in anophelines was studied using transient infection via somatic injection of Wolbachia . Some of these associations produced pathogen interference , however for wAlbB infections of Anopheles gambiae , Plasmodium berghei infection was enhanced [31] . This enhancement may be temperature dependent [32] , and has also been observed for some native Wolbachia infections , including in Culex pipiens where Wolbachia protects the host against Plasmodium-induced mortality , but also increases susceptibility to infection [33–35] . Interestingly , such enhancement has never been observed for Plasmodium species that infect humans , or in a mosquito with a stable Wolbachia transinfection . The process underlying pathogen interference remains poorly understood , while potential causes of enhancement are only hypothetical [31 , 36] . Strong pathogen interference is typically associated with high Wolbachia density [17 , 37] . The effect has been linked to activation of immune effector genes [22 , 23 , 38 , 39] , increased induction of reactive oxygen species ( ROS ) and related genes , which serve as part of the host defence against pathogens [38 , 40] , and competition for host cholesterol in Drosophila melanogaster [41] . Critically , none of these effects occur universally amongst the species and Wolbachia strains where pathogen interference has been observed , which suggests that the underlying mechanism may be complex , and that it could potentially be dependent on the length of the host-symbiont relationship [39 , 42 , 43] . Pathogen interference and CI serve as the basis for a form of Wolbachia-dependent mosquito control through mosquito population replacement [2 , 44] , which is currently being utilised for Ae . aegypti and dengue ( www . eliminatedengue . com ) . This involves the release of Wolbachia-infected mosquitoes that mate with the wild population , where CI increases the Wolbachia infection frequency over successive generations [45] . High prevalence of pathogen interference in these mosquito populations would then potentially reduce disease transmission amongst humans [25] . Successful Wolbachia invasion is dependent on local environmental conditions , and a high proportion of infected individuals [44 , 46] . Another critical factor is the competitiveness of released mosquitoes [47] , with high fitness costs , as seen with the wMelPop strain [16 , 48 , 49] , leading to rapid loss of infection in field populations [50] . In contrast , the wMel strain has minimal fitness costs [17] , and a stably infected population has persisted for several years in the field [51] , with no loss of pathogen interference observed since the initial release [52] . Nutritional status and diet are key factors in an insect’s ability to resist infection with a pathogen [53–55] . Likewise , many pathogens are dependent on host nutritional resources , and can manipulate host metabolic process in order to facilitate infection [56–59] . Recent evidence has demonstrated that Wolbachia has a similar metabolic dependency [41 , 60 , 61] , and this suggests that there is great potential for tripartite interactions between Wolbachia , pathogens , and host metabolism and nutrition to play a role in pathogen interference . To that end , we have used dietary carbohydrate concentration as a platform to study the influence of host nutrition on the complexity and plasticity of pathogen interference and associated processes in female wMel-infected Ae . aegypti . We investigated the role of diet-induced nutritional stress on interference against DENV and P . gallinaceum , and levels of immune gene expression and H2O2 , which have previously been linked to the phenotype . Through these experiments we sought to further understanding of how Wolbachia can influence pathogen infection .
In all experiments described below , adult mosquitoes were fed one of four carbohydrate regimes ( 1% , 5% , 10% or 20% raw sugar solution ) . Two experimental infections with a recently circulating Brazilian DENV-3 isolate were performed to determine if altered carbohydrate diets affected pathogen interference against DENV . Mosquitoes were fed on the carbohydrate regimes for 7 days post-eclosion , and then orally challenged with DENV . In both replicates , no DENV RNA was amplified from any wMel sample at either 7 ( Fig 1A & 1C ) or 14 days post-infection ( Fig 1B & 1D ) , for any diet . In contrast , the Tet infection rate varied between 35% and 82% , depending on diet , and the duration of infection . Prevalence ( proportion infected with DENV-3 ) was consequently significantly higher for Tet mosquitoes than for wMel , for each diet ( Fisher’s exact test; P = 0 . 0033 - <0 . 0001 ) . As no wMel mosquitoes became infected , only the viral load data for Tet mosquitoes were compared statistically . There was a significant difference in viral load due to host nutrition at 7 dpi for both replicates , characterized by higher DENV levels on the 1% diet ( Kruskal Wallis; R1—P = 0 . 0489; R2—P = 0 . 0084 ) . At 14 dpi there were higher DENV levels on the 1% diet in the first replicate ( Kruskal Wallis; P = 0 . 0015 ) , but no effect in the second replicate . Three replicate P . gallinaceum infection experiments were performed to assess the impact of dietary carbohydrate levels on the ability of the wMel Wolbachia strain to interfere with Plasmodium infection ( Fig 2 ) . Prevalence ( proportion of mosquitoes infected ) and intensity ( number of oocysts in infected mosquito midguts ) of Plasmodium infection were measured at 7–8 days post-infection , and data were compared independently for each experiment using binomial regression to determine the effects of Wolbachia infection and diet ( S1 Table ) . In both experiment 1 ( Fig 2A ) and 2 ( Fig 2B ) , Wolbachia was a significant factor affecting prevalence ( Binomial models; E1 - , P = 0 . 0014; E2—P < 0 . 0001 ) . There was a strong inhibitory effect of wMel on the 10% diet in both experiments , however this attenuated as dietary carbohydrate levels changed , as evidenced by a significant effect of diet ( Binomial models comparing each diet against the 10% diet; E1−1% diet and 20% diet—P < 0 . 0001; E2−1% , 5% and 20% diets—P < 0 . 0001 ) . Interestingly , pairwise comparisons of prevalence for wMel and Tet mosquitoes on each diet revealed that some level of interference was maintained for all diets except the 1% ( Fisher’s exact test; E1—P < 0 . 05 , E2—P < 0 . 001 ) . In experiment 3 ( Fig 2C ) , the prevalence of Plasmodium infection observed for the 10% diet was greater than in the other two experiments , although the parasitemia level was lower . In this experiment there was no overall effect of Wolbachia on prevalence ( Binomial regression; P = 0 . 646 ) , however there was still a significant effect of diet ( Binomial models; 1% diet—P < 0 . 05: 5% and 20% diets—P < 0 . 001 ) . Plasmodium intensity data were compared independently for each experiment using binomial negative regression , and we observed a significant effect of Wolbachia infection only in experiment 2 ( Binomial negative regression against 10% diet; P < 0 . 0001 ) . As for prevalence , a change in diet led to increased intensity of infection for all three experiments when compared to the 10% diet ( Binomial negative regression: E1 , E2—All diets: P < 0 . 0001 , E3−1% and 5%: P < 0 . 0001 , 20%: P > 0 . 05 , all comparisons in reference to the 10% diet ) . However , pairwise comparisons for each diet indicated that a significant interference effect due to Wolbachia was induced only for low carbohydrate diets , as wMel infection reduced the intensity of infection on the 1% diet in all 3 experiments ( Mann Whitney U test; E1—P = 0 . 0151; E2—P < 0 . 0001; E3—P = 0 . 0231 ) , and for the 5% diet in experiments 2 and 3 ( Mann Whitney U test; E2—P < 0 . 0001; E3—P = 0 . 0376 ) . To obtain a broad indicator of fitness changes due to host nutrition , we compared the effects of altered dietary carbohydrates on the longevity of Wolbachia-infected wMel mosquitoes using Cox Regression . Average mosquito survival was greater with higher dietary carbohydrate levels ( S1 Fig ) . Average ( ± s . e . m . ) survival time on the control diet was 29 . 48 ± 1 . 15 days , which was 12 . 28 days and 4 . 26 days longer than the average survival times for mosquitoes reared on the 1% and 5% diets , but 5 . 57 days shorter than the average for the 20% diet . Diet was a significant factor affecting mosquito longevity ( Cox Regression; P < 0 . 0001 ) . The 1% ( B = 4 . 41 , 95% CI = 3 . 32–5 . 88 , P < 0 . 0001 ) and 5% ( B = 1 . 38 , 95% CI = 1 . 06–1 . 79 , P = 0 . 016 ) regimes were associated with significantly higher hazard ratios than the control diet ( 10% ) , however the 20% diet led to a lower hazard ratio than the control ( B = 0 . 60 , 95% CI = 0 . 46–0 . 78 , P < 0 . 0001 ) . Expression levels of the Wolbachia gene wsp were quantified relative to the host rps17 in mosquitoes after 7 days of feeding on the different dietary regimes , in order to determine if the different regimes altered Wolbachia levels in wMel mosquitoes ( Fig 3 ) . These data were compared statistically using univariate general linear models , which indicated that there was no statistically significant effect of diet ( GLM; P = 0 . 468 ) . We looked at whether changing dietary carbohydrate levels affected the expression of four genes associated with immune activation by Wolbachia . These genes , Cecropin E ( cece ) and Defensin C ( defc ) , both antimicrobial peptides stimulated by the Toll and IMD immune pathways , C-type lectin galactose binding 5 ( ctlga5 ) , a carbohydrate-binding protein involved in bacterial recognition , and Transferrin ( tsf ) , an iron transport protein , were all strongly upregulated by wMel infection in Australian mosquitoes [39] . Expression data for each gene were analysed independently using general linear models to determine if Wolbachia infection status or nutrition had a major effect ( Fig 4 ) . Data for all immune assays were obtained from mosquitoes fed on carbohydrates for 7 days . These mosquitoes were not blood fed or infected with a pathogen . Levels of cece ( Fig 4A ) and defc ( Fig 4B ) were significantly affected by both Wolbachia infection and diet ( GLM; P < 0 . 0001 ) , however there was no effect of interaction between diet and Wolbachia infection . Analysis of individual treatments revealed that cece and defc levels were higher in wMel mosquitoes than in Tet for all diets ( Student’s t tests; P < 0 . 001 ) . Levels of cece were increased on the 1% diet for both wMel and Tet mosquitoes , while defc expression in Tet mosquitoes was increased on the 1% and 5% diets , although average expression levels were still lower than for wMel . Wolbachia did not have a significant effect on ctlga5 expression in the overall GLM model , however both diet and the Wolbachia x diet interaction ( GLM; P < 0 . 0001 ) were significant factors . Expression levels of ctlga5 were higher in wMel mosquitoes than in Tet for all diets except the 1% ( Fig 4C; student’s t tests—5% & 10%; P < 0 . 01 , 20%; P < 0 . 05 ) , where levels in wMel mosquitoes remained high , but Tet levels were slightly higher ( Student’s t test; P = 0 . 0324 ) . There was a decrease in expression for wMel mosquitoes on the 20% diet , where levels were on average 39 . 06% lower than for the 10% diet . Expression levels of tsf ( Fig 4D ) were not significantly affected by diet in the overall model , but were affected by both Wolbachia ( GLM; P < 0 . 0001 ) and Wolbachia x diet ( GLM; P = 0 . 015 ) . tsf levels were significantly higher for wMel than Tet for all 4 diets ( Student’s t tests; P < 0 . 0001 ) . However , tsf expression in wMel mosquitoes on the 1% diet was significantly lower than for the 10% and 20% diets ( Student’s t test; P < 0 . 05 ) . The expression of 8 genes with putative regulatory roles in the mosquito IMD ( caspar and rel2 ) , JAK-STAT ( domeless and pias ) , JNK ( ap-1 and jnk ) and Toll ( cactus and rel 1A ) immune pathways was examined in order to determine whether diet x Wolbachia interactions had a broader effect on host immunity ( S2 Fig ) . We observed no effect of Wolbachia , diet , or Wolbachia x diet interaction in the expression of these genes . The one exception to this was for pias , a putative negative regulator of the JAK-STAT immune pathway , where Wolbachia but not diet or the Wolbachia x diet interaction was a significant predictor in the overall model ( GLM; P = 0 . 044 ) . In biological terms , this translated to higher pias expression in wMel mosquitoes compared to Tet , but only for the 10% diet . Expression levels of duox-2 ( Fig 5A ) , an important gene in mosquito reactive oxygen species production , were unaffected by wMel infection , diet , or diet x Wolbachia interaction ( GLM; P < 0 . 05 ) . Likewise levels of nos ( Fig 5B ) , which is involved in nitric oxide production , were unaffected by the presence of Wolbachia ( GLM; P < 0 . 05 ) . However , we observed a significant increase in nos expression associated with lower carbohydrate diets in the overall model ( GLM; P < 0 . 0001 ) , and independently for both Tet ( GLM; P < 0 . 0001 ) and wMel mosquitoes ( GLM; P = 0 . 001 ) . Average nos levels were 45 . 92% higher for wMel mosquitoes on the 1% diet than those on the 10% diet ( student’s t test; P = 0 . 0013 ) . For Tet mosquitoes , the 1% diet had on average 60 . 74% higher nos levels than the 10% diet ( student’s t test; P = 0 . 0005 ) , and those from the 5% diet had on average 38 . 56% higher nos levels ( student’s t test; P = 0 . 0099 ) . H2O2 levels were quantified in pairs of female mosquito after spending 7 days feeding on the different carbohydrate diets ( Fig 6 ) . Wolbachia infection ( GLM; P < 0 . 0001 ) , diet ( GLM; P < 0 . 0001 ) and Wolbachia x diet ( GLM; P = 0 . 0006 ) were all significant factors affecting H2O2 levels in mosquitoes . H2O2 levels in Tet mosquitoes did not change due to diet however mean H2O2 levels in wMel mosquitoes were positively correlated with dietary carbohydrate concentration . ROS induction due to Wolbachia infection was observed for each of the three highest concentration diets , where significantly higher levels were observed in wMel mosquitoes ( Student’s t tests; 5% diet—P = 0 . 0046; 10% diet—P = 0 . 0040; 20% diet—P = 0 . 0082 ) , however on the 1% diet there was no effect of Wolbachia ( Student’s t test; P = 0 . 2034 ) .
Pathogen interference in Wolbachia-infected Ae . aegypti restricts or prevents infection and transmission of DENV and other pathogens [22 , 23 , 25] . Interference is fundamental to transmission-blocking strategies that use Wolbachia to combat mosquito-transmitted disease [44] , yet the underlying biological processes remain poorly understood . Competition for nutrients is important to interference in Drosophila [41] , but no link with host nutrition had previously been made in Ae . aegypti . To that end , we fed wMel ( +Wolb ) and Tet ( -Wolb ) Ae . aegypti mosquitoes with 1 of 4 carbohydrate diets ( 1% , 5% , 10% or 20% sucrose solution ) , and challenged them with either DENV-3 or P . gallinaceum . We observed strong interference to both pathogens on the 10% ( control ) diet . For P . gallinaceum , wMel infection reduced the prevalence of infection but did not affect intensity , while no wMel mosquitoes became infected with DENV . Pathogen interference against P . gallinaceum had not previously been described for wMel-infected Ae . aegypti . This effect was not as strong as for wMelPop-infected Ae . aegypti where there was greatly reduced prevalence and intensity of infection [23] , although that strain has a higher bacterial density , which likely promotes stronger pathogen interference [17 , 25] . Our DENV interference results were similar to results from other DENV isolates , where mosquitoes were reared on 10% sucrose [17 , 25] . Altering host nutritional status by feeding 1% , 5% or 20% sucrose led to increased prevalence of P . gallinaceum infection in wMel mosquitoes , which could be interpreted as less effective pathogen interference . The effect was most striking on the 1% diet , where P . gallinaceum prevalence for both Tet and wMel mosquitoes was near 100% . This increased prevalence could have been driven by starvation , similar to what is seen with Plasmodium infection in mosquitoes that experience larval nutritional stress [55 , 62] . These data suggest that there are certain nutritional states or biological conditions that favour Plasmodium infection to the point where an inhibitory effect by Wolbachia is not possible . The fact that we also observed less effective interference on the 20% diet indicated that our results could not be solely explained by a starvation effect , and could have been due to a broader modulatory effect of host nutrition . Changing nutritional status also increased the median oocyst count for both Tet and wMel mosquitoes , particularly on the 1% and 20% diets , however there was still a statistically significant effect of wMel infection for the latter . Interestingly , wMel limited the increase in the intensity of infection on the 1% and 5% diets , suggesting that the interference effects of wMel at the intensity level occurred with the change in host nutritional status . We observed greater overall P . gallinaceum intensity , and a different effect of Wolbachia on P . gallinaceum prevalence in one experiment . Infection with P . gallinaceum is typically subject to high variability , with great differences in prevalence and pathogen levels resulting from mosquito , parasite and avian genetic factors , and environmental factors [33 , 63 , 64] . Each experiment involved different chickens , with different genetic and immune responses that could have influenced the course of infection [65] . Across the three experiments , a stronger pathogen interference effect was associated with higher parasitemia , with no effect of Wolbachia observed in the experiment with the lowest parasitemia . While we did observe some variation between experiments , our results did suggest that host nutritional status can alter the response of wMel to P . gallinaceum under some conditions , but also that this interference does not occur under all experimental conditions , and may only be induced during more severe infection . In contrast , we saw no effect of host nutrition on DENV interference as no wMel mosquitoes became infected on any diet across two experiments . This indicated that DENV interference is not affected by the change in host nutritional status , starvation or dietary excess . Furthermore , the different response to host nutritional status between the two pathogens suggests that there are potentially host biological factors that differentially affect interference against P . gallinaceum and DENV . We sought to determine if nutritional stress affected Wolbachia density , the expression of key immune genes and ROS levels , all of which have previously been linked to pathogen interference in either mosquitoes or Drosophila . These processes were characterized after mosquitoes fed on the different carbohydrate regimes for 7 days , the same time at which mosquitoes were infected with a pathogen in our experimental infection assays . These mosquitoes were not blood fed or infected with a pathogen in order to characterize basal changes due to diet and Wolbachia , and to avoid metabolic and transcriptional changes induced by blood feeding [56 , 66] . High Wolbachia levels appear to be a key driver of pathogen interference [17 , 67] , and reduction of bacterial density can lead to weaker interference [68] . Critically , we saw no effect of diet on Wolbachia expression . This could potential indicate that the loss of interference against P . gallinaceum was not associated with a change in Wolbachia density . Although it is possible that such a change could occur in response to feeding on Plasmodium-infected blood , or that changes in Wolbachia levels at the tissue level led to a loss of bacterial density . An alternative explanation is that there was amelioration of another biological process linked to the phenotype . Pathogen interference in mosquitoes has been strongly associated with the increased expression of key immune effector genes [22 , 23 , 38 , 39] . We observed that high expression levels of four of these genes , cece , defc , ctlga5 and tsf were consistently associated with Wolbachia infection for all diets . This could imply that a loss of immune gene activation did not underlie the less effective interference for P . gallinaceum that we observed on some diets , however it should be noted that we only measured basal immune gene levels , not in the context of Plasmodium infection , and this could potentially have led to different results . We did observe slight decreases in the expression of tsf on the 1% diet , and defc and ctlga5 on the 20% diet in wMel mosquitoes , and it is possible that our results could be explained by a similar effect across a large number of immune genes . Similarly , we saw no effect of diet on the expression of regulatory genes in the IMD , JAK-STAT , JNK , and Toll mosquito immune pathways that might explain our results . Given that Plasmodium and DENV infections stimulate different immune pathways [69–73] , it was possible that a diet-induced change in regulatory gene expression could stimulate higher infection levels . However , we only saw an effect of Wolbachia on the expression of pias , a negative regulator of the JAK-STAT pathway , and this change—higher expression in wMel mosquitoes than Tet only for the 10% diet—did not adequately explain our results , as wMel mosquitoes had similar pias levels across all diets . These results do not preclude an immune basis for the Plasmodium-specific response if it were to occur through genes or pathways other than those we measured . Diet can influence levels of ROS and oxidative stress in insects [74 , 75] , and we observed a clear effect of mosquito diet on ROS induction , with equivalent H2O2 levels in wMel and Tet mosquitoes from the 1% diet , and higher dietary carbohydrate levels associated with higher mean H2O2 levels in wMel mosquitoes . In contrast , H2O2 levels in Tet mosquitoes were unaffected by diet , suggesting that there was a Wolbachia-specific interaction between nutritional and oxidative stress . The ROS induction phenotype is strongly correlated with pathogen interference in both mosquitoes and Drosophila [38 , 40] . However , it is not universal amongst all host-strain associations where pathogen interference occurs , as is the case for wMel-infected Ae . albopictus , where there is interference against DENV and Chikungunya virus infection [43 , 76 , 77] . The fact that ROS induction occurs for wMel-infected Ae . aegypti suggests that its absence in Ae . albopictus is more likely due to the host mosquito than the wMel strain , potentially because of the residual effects of co-adaptation with its native Wolbachia strains wAlbA and wAlbB . The fact that loss of ROS induction occurred for the 1% diet , where wMel and Tet mosquitoes has a similar susceptibility to P . gallinaceum infection is particularly interesting . ROS induction is part of the natural response to Plasmodium infection , with higher oxidative stress levels promoting parasite melanisation [78 , 79] . Interestingly , levels of tsf in wMel mosquitoes were also decreased for that diet . This gene is involved in iron transport and changes in its expression could have contributed to decreased ROS production and may be indicative of broader alterations to host oxidative stress response under conditions of starvation in Wolbachia-infected mosquitoes . Critically , as less effective Plasmodium interference , and high H2O2 levels were observed for the 20% diet , changes to ROS induction are unlikely to be the sole factor causing the differential effect of host nutrition that we observed on Plasmodium and DENV infection . The stimulation of mitochondrial and oxidative stress gene expression by Wolbachia has been implicated in ROS induction , activation of the Toll immune pathway , and pathogen interference [22 , 38 , 39] . In wAlbB-infected Ae . aegypti , this effect was linked to a 23-fold increase in the expression of duox-2 , which is thought to be an important enzyme for ROS production [38] . However in our experiments , and in wMel-infected Ae . albopictus , Wolbachia did not affect duox-2 levels , potentially because the gene lacks peroxidase activity , and therefore cannot directly stimulate ROS [43] . As we observed ROS induction without an effect of Wolbachia on duox-2 , this implies that ROS induction occurs via a different process , potentially via the duox-1 gene . Likewise , duox-2 expression could not explain our ROS induction results , given the lack of an effect of host nutrition . The enzyme nos is involved in the production of nitric oxide and reactive nitrogen species , and high nos levels have been linked to the inhibition of both Plasmodium and DENV in mosquitoes [80–82] . We observed no change in nos expression due to Wolbachia , indicating that this gene was unlikely to contribute to pathogen interference . Interestingly , we observed higher nos levels on the 1% and 5% diets for both wMel and Tet mosquitoes , where the prevalence of P . gallinaceum infection was greater . This suggests that there is a link between nutritional stress and nos expression , and that nos levels can be induced under conditions of starvation without a strong effect on P . gallinaceum infection . It is possible that levels of H2O2 , nos or the immune genes that we examined could have been changed in response to blood feeding or severity of Plasmodium infection , as both factors are linked to oxidative stress response [56 , 66 , 83 , 84] . Additionally , there could have been systemic change in the mosquito oxidative stress and immune responses as a result of these processes , and this may have contributed to the response of Wolbachia to pathogens , even under conditions of starvation . We observed that changing host nutrition affected response to Plasmodium interference , ROS induction and nos expression . Furthermore there was differential fitness due to diet in the form of a longevity cost for low carbohydrate diets , which is not unexpected as dietary composition and insulin signalling affect lifespan in Wolbachia-infected insects [13 , 85] , and because Wolbachia increases the rate of resource depletion during starvation in larvae [86] . Starvation can stimulate immune response [53 , 87 , 88] , as we observed with immune gene and nos expression on the 1% diet . It also reduces the availability of arginine and therefore affects levels of nitric oxide , and consequently affects the prevalence and intensity of Plasmodium infection [55 , 89] . As such , it is possible that starvation-induced perturbations of the oxidative stress or nitric oxide response were the primary determining factor explaining our Plasmodium results from the 1% diet . Dietary excess is another form of nutritional stress , and in insects it causes obesity , alters the metabolism and biosynthesis of fats and carbohydrates , and alters oxidative stress response [90–93] , which could explain some of the results for the 20% diet . Metabolic interaction and competition for resources between the host and Wolbachia affects host gene expression , metabolic homeostasis , and physiological processes linked to metabolism [22 , 39 , 49 , 60 , 85 , 94] . Resource competition leads to less effective pathogen interference in D . melanogaster [41 , 60] , and could underlie diet-based differences in Plasmodium interference , particularly on the 20% diet . Both Plasmodium and DENV exploit host carbohydrate metabolism [54 , 95–97] , and infection alters host carbohydrate homeostasis [57 , 59 , 98] . However , Plasmodium are probably more heavily reliant on host sugars , which they use for glycolysis , carbohydrate metabolism , and fatty acid II synthesis [99] , and for development [95] , and thus could be more highly affected by competition with Wolbachia . As the type of carbohydrate intake can influence susceptibility to Plasmodium infection in mosquitoes , there could be similar effects on the ability of Wolbachia to interfere with infection [95] . The composition of the host microbial community can be affected by host diet [100 , 101] , can alter host metabolic profile [102–104] , and can affect response to pathogen infection [105–108] . Interestingly , the microbiota induce production of ROS , which can influence susceptibility to Plasmodium infection , and offers a potential explanation for the diet-induced changes we observed in oxidative stress response[109] . There is evidence of interaction between Wolbachia and the microbiota , in the form of mutual exclusion between Wolbachia and Asaia in anophelines [110] , and a microbial influence on the vertical transmission of Wolbachia in transiently infected An . stephensi [111] . But the full extent of the interactions between Wolbachia and host microbiota are unclear , and there is certainly scope for a nutrition-driven interaction , that could affect a range of physiological processes including pathogen interference . wMel-infected mosquitoes have been present in the field for several years , where they maintain high levels of interference against different DENV isolates [52] . The issue of nutritional stress and pathogen interference is particularly important in the field where mosquitoes are released in locations with complex environmental and nutritional factors , and high levels of endemic dengue transmission [50 , 112] . Adult Ae . aegypti nutritional needs are fulfilled by blood feeding when human hosts are available [113] , and plant sugars when they are not . Recent work suggests that repeated blood feeding does not affect interference against DENV in Ae . aegypti [114] . While a sucrose-based diet is unlikely to be perfectly reflective of mosquito carbohydrate intake in the field , our diets did induce varying levels of nutritional stress , which could be similar to what mosquitoes in a heterogeneous environment might experience . What our results suggest is that DENV interference appears to be quite robust in the face of variable host nutritional status , and such an effect would be greatly beneficial if it were to occur in the field Wolbachia-infected mosquitoes . These data should be further clarified using different DENV serotypes , genetic isolates , and viral titres , as well as for other types of host diet , as these factors can all influence pathogen interference [25] . Our results did show that changes in host diet led to significantly weaker pathogen interference against P . gallinaceum under some host nutritional conditions , and that this may correspond to altered oxidative stress response . Yet because Wolbachia-infected Ae . aegypti are unlikely to become infected with Plasmodium in the field this does not represent a large issue for current control efforts . Potential problems could arise if a similar nutrition-based interaction were to occur in Wolbachia-infected anophelines . Critically , P . gallinaceum does not infect humans , and the effect we observed here may not occur for the mosquitoes and parasites responsible for human malaria , given their different immune and metabolic interactions [29 , 30] . Pathogen interference has been observed against P . falciparum in wAlbB-infected An . stephensi [27] , and future studies should determine the extent to which this phenotype is subject to environmental factors including nutrition , as this will have implications for future malaria control programs involving Wolbachia . Perhaps the most interesting idea resulting from our data is the reinforcement and extension of the hypothesis of a complicated pathogen interference phenotype . Previous data indicates that the associated processes are not universal , with ROS induction being strain specific , and immune activation apparently specific to mosquitoes [37 , 39 , 43] . We have demonstrated that interference can also be pathogen specific , with diet-induced nutritional stress , and potentially starvation , affecting interference against P . gallinaceum but not DENV . It is also clear that host nutritional status can affect the ROS induction effect that has been linked to interference , and this should be further examined in the context of blood feeding , and experimental Plasmodium and DENV infection in order to characterize the effects of Wolbachia in a more natural nutritional state . These findings highlight the complicated nature of the phenotype , with the implication being that there is unlikely to be a ‘magic bullet’ explaining all occurrences of the phenotype . Rather , pathogen interference may arise through combinations of contributory factors with additive effects , and different pathways to interference occurring for different host-strain-pathogen combinations . The identity of these factors is currently unclear . However , given the breadth of Wolbachia’s effects on mosquito molecular biology , there are many potential candidates that have not yet been studied in great detail .
Two Ae . aegypti lines were used in these experiments . The first was infected with the Wolbachia strain wMel ( wMel ) . This line was derived from the wMel-transinfected line , originally generated in Ae . aegypti with an Australian genetic background [17] . The wMel infection was introgressed into a Brazilian genetic background by breeding infected females with uninfected , field-collected males over nine generations , as previously described [112] . A subset of these mosquitoes were treated with tetracycline to clear the Wolbachia infection and then had their gut microbiota recolonised by introducing larval water from untreated mosquitoes into rearing trays , as previously described , with this line serving as a Wolbachia-uninfected control line ( Tet ) [112] . 50 wildtype , Wolbachia-uninfected F1-F2 males were introduced into colony cages for both wMel and Tet lines each generation , in order to limit the occurrence of inbreeding and genetic divergence between the lines . These mosquitoes were collected near Rio de Janeiro , Brazil , and reared under laboratory conditions until eclosion , as described below . No wildtype males were introduced into experimental cages . wMel mosquitoes used in these experiments were from G14—G29 post introgression into the Brazilian genetic background . Tet mosquitoes were from G10—G25 post microbial recolonization . All mosquitoes in these experiments were reared under standard laboratory conditions in a climate-controlled insectary ( temperature—27 ± 1°C , RH -70 ± 10% , photoperiod—12 hours light: dark ) . Mosquito larvae were hatched in 3L RO water containing ½ of a tetramin tropical tablet ( Tetramin ) ground into powder . Larval density was reduced to 50 per litre 24 hours after hatching . Larvae were then fed ½ a tetramin tropical tablet as required , with food levels equating to 1mg of food per larva per day . Pupae were sexed , collected and moved to small cylindrical cages ( diameter– 16cm , height– 18cm ) for experiments , with a maximum adult density of 150 per cage . Adult mosquitoes were maintained on one of four different carbohydrate diets for the entirety of each experiment . The control diet was 10% sucrose , which was the same concentration provided to colony mosquitoes . The other three diets consisted of 1% , 5% and 20% sugar solution , with each inducing dietary stress either through starvation or excess . All diets were prepared by dissolving raw , granular cane sugar into RO water . Sucrose cups in experimental cages were changed every two days to prevent microbial contamination , with the solutions prepared fresh each time . The virus used in these assays , DENV-3 MG20 ( 375 ) was originally isolated from infected patient blood in Brazil in 2012 . The virus was cultured in C6-36 cells , titred using both the TCID-50 and plaque forming assay methods according to previously described methods [23] . Titre estimates were 1010–1013 infectious units/mL and 1 . 9x106 infectious units/mL , respectively . Viral aliquots were stored at -80°C until the day of feeding . Cages of approximately 60 female mosquitoes were reared on carbohydrate diets as described above , and were starved overnight prior to feeding . Virus was mixed with freshly drawn blood from a willing volunteer at a 1:1 ratio . Blood used for feeding was screened for dengue virus using the Dengue NS1 Ag Strip Test ( BioRad Laboratories ) . Mosquitoes were fed through glass feeders with pig intestine , using a heated waterbath system at a temperature of 37°C for 1 hour . Afterwards , non-blood fed , and semi-fed mosquitoes were removed and carbohydrate diets were re-introduced to cages . Half of the cage was collected at 7 days post-infection , and the other half collected at 14 days post-infection . Two independent feeding experiments were performed , using different aliquots from the same batch of virus . Whole mosquito samples were stored at -80°C , and total RNA was extracted using the TRIzol protocol ( ThermoFisher Scientific cat 15596–026 ) according to manufacturer’s instructions . Mosquitoes were homogenized in 200μL TRIzol using a mini beadbeater ( BioSpec products ) . Samples were quantified using a NanoDrop 2000 UV-Vis spectrophotometer ( ThermoFisher Scientific ) , and 1μg of total RNA was used for first strand cDNA synthesis using the M-MLV reverse transcriptase assay according to manufacturer’s instructions ( Promega cat: C118A ) . cDNAs were then diluted 1:10 in nuclease-free water and stored at -30°C . Absolute DENV levels were quantified in duplicate for each cDNA , using a TaqMan-based assay with primers and a probe generalized to all four DENV serotypes ( S2 Table ) . Each reaction contained the following: 2 . 5μL of cDNA , 2 . 50μL of TaqMan Universal Master Mix ( ThermoFisher Scientific cat: 4304437 ) , 0 . 50μL each of forward and reverse primers ( 10μM ) , 0 . 25μL of DENV probe ( 10μM ) , and 3 . 75μL of nuclease-free water . For a standard curve , we utilised a cloned DENV fragment , as previously described [23] . Serial dilutions of this fragment were run in triplicate between the concentrations of 107 and 103 copies for each plate . The run profile was 10 mins to denature at 95°C , followed by 40 amplification/cycles of 15 sec at 95°C followed by 1 min at 60°C using a Viia 7 Real-Time PCR System ( ThermoFisher Scientific ) . DENV copies per sample were normalised per 1μg of total RNA . 12–20 samples were quantified per treatment . The Plasmodium gallinaceum stock used in these experiments was a long-term laboratory line ( Brumpt , 1937 , strain 8A ) . Cultures were maintained in the laboratory stored in chicken blood at -80°C , and through regular passage in 1–2 week old Gallus gallus chicks . Chicks were obtained at 1–2 day olds from Rivelli Poultry Farms , Mateus Leme , MG , Brazil , and were maintained in the FIOCRUZ Animal Facility during the course of experiments . Chicks were infected with P . gallinaceum infected blood drawn from previously infected chickens by trained personnel . Blood parasitemia levels were monitored during the course of infection by counting infected cells in a Giemsa-stained blood smear , with the blood obtained through toenail clipping . In each experiment , approximately 70 female mosquitoes from each of the 8 treatments ( 4 diets x 2 Wolbachia infection statuses ) were fed on the different carbohydrate diets for 7 days . Mosquitoes were starved overnight and then allowed to feed on the same chick for 15 minutes per cage , with cages fed in random order . Blood parasitemia levels in the chicks on the day of feeding varied between experiments ( E1: 11 . 16% , E2: 35 . 38% , E3: 4 . 10% ) . Plasmodium stocks used in these experiments had been passaged a maximum of three times . Post-feeding , the appropriate diets were re-introduced to cages , and non-blood fed , and semi-fed mosquitoes were removed . There were no noticeable effects of Plasmodium feeding on mosquito survival . At 7–8 days post-blood feeding , midguts were dissected in sterile 1x PBS before staining in 2% mercurochrome for 10 mins . Oocysts were visualised and counted via light microscopy . Mosquito numbers ranged between 33–53 per treatment across the three experiments . 12 pairs of 7–8 day-old , female , wMel and Tet mosquitoes were collected after 7 days on their respective diets . Paired samples were used to reduce within treatment variation . This corresponded to the time when the mosquitoes in the pathogen infection assays were infected with either P . gallinaceum or DENV , however samples in these experiments were not infected with a pathogen . RNA extractions and first strand cDNA synthesis were performed as described above . The levels of 14 immune-related genes were quantified for all samples , while Wolbachia expression levels were quantified for only the wMel samples using the wsp gene ( S3 Table ) . Primer sequences used in these assays were either designed using Primer 3 V0 . 4 . 0 ( http://bioinfo . ut . ee/primer3-0 . 4 . 0/ ) , or as previously described [22 , 23 , 38] , ( S2 Table ) . Prior to use in experiments , each primer pair was assayed for specificity by melt curve analysis , with all pairs displaying only one peak . Additionally , we assayed primer efficiency by examining amplification performance with dilutions of cDNA samples . All primer pairs had an efficiency of between 90–100% at the dilution used in the experiments described below . The immune transcription assays comprised of three parts . The first was an examination of genes previously shown to be highly upregulated by wMel infection in Ae . aegypti with an Australian genetic background [39] . Four genes were examined: cecropin e ( cece ) , defensin c ( defc ) , transferrin ( tsf ) and c-type lectin galactose binding 5 ( ctlga5 ) . The second looked at regulatory genes in 4 different mosquito immune pathways . Eight genes were examined: rel2 and caspar from the IMD pathway , domeless and pias from the JAK/STAT pathway , ap-1 and jnk from the JNK pathway , and rel1a and cactus from the Toll pathway . The third part looked at two genes linked to stress response in mosquitoes . These were duox-2 , which is linked to oxidative stress , and nitric oxide synthase ( nos ) , which is linked to stress and Plasmodium infection . All genes were quantified in duplicate relative to the host ribosomal protein S17 ( rps17 ) . Each reaction contained the following: 2 . 50μL of cDNA , 7 . 50μL of SYBR Green PCR Master Mix ( ThermoFisher Scientific cat 4309155 ) , 0 . 75μL each of forward and reverse primers ( 10μM ) , and 4 . 50μL of nuclease-free water . The run profile was the same as described above . Mean normalised expression values were calculated for each gene using Q-Gene [115] . 17–22 pairs of 7–8 day-old , female , wMel and Tet mosquitoes were collected after 7 days on their respective diets . H2O2 levels in these samples were quantified using the Amplex Red Hydrogen Peroxide/Peroxidase Assay Kit ( ThermoFisher Scientific cat A22188 ) . Samples were collected on ice and then immediately homogenized in 200μL of 1x reaction buffer , using a mini beadbeater ( BioSpec products ) , and then centrifuged for 2 mins at 14 , 200 x g , at 4°C . 50μL the supernatant was used to run the H2O2 assay , according to manufacturers instructions . Assays were run in black Nunc MicroWell 96-well Optical Bottom plates ( ThermoFisher Scientific ) , and quantified using a Synergy 2 Multi-Mode Reader ( BioTek ) , with an excitation wavelength of 545nm and an emission wavelength of 590nm . A longevity assay was conducted to provide a basic measurement of the effects of host nutritional status on the fitness of wMel-infected Ae . aegypti mosquitoes . wMel larvae were reared as described above , and then female pupae were sexed and transferred to experimental cages , separated by carbohydrate diet . Pupal cups were removed from cages after 48 hours , so that all mosquitoes shared a similar age and development time . There were 3 replicate cages per diet , each containing approximately 45 females . Survival was monitored daily for the duration of the experiment , with dead mosquitoes removed from cages . Cage positions were rotated daily in order to normalize environmental variance . Maintenance of chickens , infections with P . gallinaceum and feeding of mosquitoes were conducted according to protocols that were reviewed and approved by The Commission of Ethical Animal Use ( CEUA ) / FIOCRUZ ( License—LW 38/12 ) . This complied with Brazilian law 11794/08 which governs the use of animals for scientific purposes and principles as dictated by the Brazilian Society of Science on Laboratory Animals ( SBCAL ) , and The National Council of Animal Experimentation Control ( CONCEA ) . The human blood used in these experiments was drawn from one willing , adult volunteer by trained medical personnel , after obtaining written consent . This process was conducted according to established guidelines , and approved by The Committee for Ethics in Research ( CEP ) / FIOCRUZ ( License—CEP 732 . 621 ) . Our use of human blood was in accordance with Brazilian laws 196/1996 and 01/1988 , which govern human ethics issues in scientific research in compliance with the National Council of Ethics in Research ( CONEP ) . P . gallinaceum infection data were analysed in two components; prevalence and intensity of infection . Uninfected mosquitoes were not considered in intensity analyses . Prevalence data were compared using binomial regression , and oocyst data by binomial negative regression , as there was overdispersion within the data set [116 , 117] . Due to the fact that different P . gallinaceum-infected chickens were used in each experimental infection , the three experiments were analysed independently . The test variable in these analyses was either infection status or oocyst number , while explanatory variables in the models were Wolbachia infection status , and dummy variables considering the effect of each diet , compared to the control 10% diet . A general effect of diet was not considered in the model , as we believed that the effect would differ between diets . Wolbachia x diet interaction terms were included , however these were generally not significant , and the models fit the data better after they were excluded ( S1 Table ) . Pairwise comparisons of differences in prevalence of Plasmodium infection due to Wolbachia for individual diets were calculated using Fisher’s exact test . Pairwise comparisons of oocyst levels for each diet were made using Mann-Whitney U tests . DENV prevalence data were compared by treatment using Fisher’s exact test . Viral intensity data for Tet mosquitoes were compared using one-way ANOVA . Longevity data were compared statistically using Cox Regression . Expression data for immune activation genes , immune pathway regulators , stress response genes , wsp levels , and H2O2 levels were compared independently using univariate general linear models . When significant effects were observed , interactions between treatments were compared post-hoc using student’s t-tests and then using a false discovery rate of 0 . 05 as a multiple test correction . Statistical tests were applied only if the data fit the underlying assumptions . Statistical analyses were performed using R , SPSS V17 ( IBM ) and Prism 6 . 0g ( Graphpad ) . Figures were prepared using Prism V 6 . 0g , Microsoft PowerPoint for Mac 2011 , and GIMP v 2 . 8 . 14 . From VectorBase ( https://www . vectorbase . org ) unless noted . ap-1 ( AAEL011650-RA ) , c-type lectin galactose binding 5 ( AAEL005641-RA ) , cactus ( AAEL000709-RB ) , caspar ( AAEL003579 ) , cecropin e ( AAEL000611-RA ) , defensin c ( AAEL003832-RA ) , domeless ( AAEL012471-RA ) , duox-2 ( AAEL007562-RA ) , jnk ( AAEL008634-RA ) , nitric oxide synthase ( AAEL009745-RA ) , pias ( AAEL015099-RA ) , rel1a ( AAEL007696-RA ) , rel2 ( AAEL007624-RA ) , ribosomal protein S17 ( AAEL004157 ) , transferrin ( AAEL015458-RA ) , wolbachia surface protein ( GenBank accession: EU395833 . 1 ) . | Mosquito-transmitted disease severely impacts human health around the world . One novel form of control involves infecting medically important mosquito species with the naturally occurring bacterium Wolbachia , which restricts dengue and malaria transmission through a process called pathogen interference . The interference phenotype is still poorly understood , and potentially involves multiple physiological changes to the mosquito . We examined the role of nutritional stress on pathogen interference in the dengue vector Aedes aegypti , in order to better understand factors that might lead to variable interference . We demonstrated that interference against malaria-causing Plasmodium gallinaceum was dependent on mosquito nutritional status , however interference against dengue was not , implying that pathogen interference operates differently for different pathogens . We then examined mosquito immune processes that had been previously correlated with pathogen interference , and demonstrated that mosquito nutrition did not affect the expression of key mosquito immune genes , but did affect levels of reactive oxygen species . Our results highlight the complexity of the phenotype , and importantly suggest that adult nutrition may not be a key determinant of interference against DENV . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"parasite",
"groups",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"plasmodium",
"pathogens",
"diet",
"carbohydrates",
"animals",
"wolbachia",
"parasitology",
"organic",
"compounds",
"apicomplexa",
"nutrition",
"insect",
"vectors",
"bacteria",
"genetic",
"interference",
"epidemiology",
"chemistry",
"pathogenesis",
"disease",
"vectors",
"insects",
"arthropoda",
"mosquitoes",
"organic",
"chemistry",
"host-pathogen",
"interactions",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms"
] | 2016 | Diet-Induced Nutritional Stress and Pathogen Interference in Wolbachia-Infected Aedes aegypti |
Rep68 is a multifunctional protein of the adeno-associated virus ( AAV ) , a parvovirus that is mostly known for its promise as a gene therapy vector . In addition to its role as initiator in viral DNA replication , Rep68 is essential for site-specific integration of the AAV genome into human chromosome 19 . Rep68 is a member of the superfamily 3 ( SF3 ) helicases , along with the well-studied initiator proteins simian virus 40 large T antigen ( SV40-LTag ) and bovine papillomavirus ( BPV ) E1 . Structurally , SF3 helicases share two domains , a DNA origin interaction domain ( OID ) and an AAA+ motor domain . The AAA+ motor domain is also a structural feature of cellular initiators and it functions as a platform for initiator oligomerization . Here , we studied Rep68 oligomerization in vitro in the presence of different DNA substrates using a variety of biophysical techniques and cryo-EM . We found that a dsDNA region of the AAV origin promotes the formation of a complex containing five Rep68 subunits . Interestingly , non-specific ssDNA promotes the formation of a double-ring Rep68 , a known structure formed by the LTag and E1 initiator proteins . The Rep68 ring symmetry is 8-fold , thus differing from the hexameric rings formed by the other SF3 helicases . However , similiar to LTag and E1 , Rep68 rings are oriented head-to-head , suggesting that DNA unwinding by the complex proceeds bidirectionally . This novel Rep68 quaternary structure requires both the DNA binding and AAA+ domains , indicating cooperativity between these regions during oligomerization in vitro . Our study clearly demonstrates that Rep68 can oligomerize through two distinct oligomerization pathways , which depend on both the DNA structure and cooperativity of Rep68 domains . These findings provide insight into the dynamics and oligomeric adaptability of Rep68 and serve as a step towards understanding the role of this multifunctional protein during AAV DNA replication and site-specific integration .
Adeno-associated virus ( AAV ) is a non-pathogenic human parvovirus that has evolved a unique mechanism of persistence in human cells by integrating its genome site-specifically into a defined locus of human chromosome 19 [1] . The single-stranded AAV DNA genome contains two open reading frames ( ORFs ) , REP and CAP that are flanked by inverted terminals repeats ( ITRs ) . The non-structural proteins of the REP ORF mediate AAV DNA replication , integration , transcriptional regulation and packaging of the AAV genome into preformed empty capsids . The REP ORF encodes four Rep isoforms , Rep40 , Rep52 , Rep68 , and Rep78 [2] . All Rep isoforms share a central AAA+ domain , which has ATPase and DNA helicase activities . Rep68 and Rep78 also contain the OID , which binds and nicks the ITR structure . Furthermore , Rep52 and Rep78 share a putative zinc-finger domain , which has been implicated in interacting with diverse cellular factors . Despite the apparent redundancy of functional domains , the biological functions of the small and the large Reps differ . Rep40 and Rep52 support efficient packaging of AAV DNA into AAV capsids [3] , [4] . Rep68 and Rep78 , on the other hand , are essential for AAV DNA replication [5]–[7] as well as site-specific integration of AAV DNA into human chromosome 19 at the AAVS1 locus [8] . The functional versatility shown by the AAV Rep proteins is in large part due to the presence of the AAA+ motor domain that structurally defines members of helicase superfamily 3 ( SF3 ) [9] . SF3 helicases are multifunctional proteins only found in small DNA and RNA viruses such as simian virus 40 ( SV40 ) , bovine papillomavirus ( BPV ) , and AAV . In addition to their DNA unwinding activity , LTag , E1 , and Rep68/78 helicases act as initiators of DNA replication on their respective viral origins [10] . This function is facilitated by the presence of the OID , which is positioned at the amino-terminus of the AAA+ motor domain . Once bound to the origin of replication , DNA melting of Ori sequences promotes the formation of an active helicase oligomer , which in the case of SV40 LTag and BPV E1 is double-hexameric ring . To date , the oligomeric nature of the AAV Rep initiation complex remains inconclusive . The oligomeric character of the large Rep68/Rep78 is still under debate due to their tendency to aggregate at low ionic strength conditions [11] . Several studies have suggested that Rep68/Rep78 form hexameric rings upon binding AAV origin , however the supporting evidence is not entirely conclusive [11]–[13] . Moreover , in contrast to the corresponding AAA+ motor domains from SV40 LTag and BPV E1 , the minimal AAA+ domain represented by Rep40 is monomeric [14] , [15] . Thus AAV Rep proteins stand apart from the other SF3 family members , as they appear to have evolved an additional requirement for cooperative involvement of both the OID and the AAA+ domain for oligomerization . It is tempting to speculate that this step is further regulated by ATP binding as well as by the nature of the various DNA targets these multifunctional proteins encounter during the replication and integration process . Accordingly , we hypothesized that Rep68/Rep78 assembles into different complexes depending on the nature of the DNA substrates . In order to address these questions we carried out a series of biochemical , biophysical , and structural studies using Cryo-EM and protein modeling in order to characterize the oligomeric nature of Rep68 in the absence and presence of different DNA substrates . Our analyses show that Rep68 assembles into a ring-shaped double octamer in the presence of ssDNA or ssDNA-dsDNA heteroduplex substrates . In contrast and consistent with previous suggestions , Rep68 assembles into a smaller complex in presence of RBS containing dsDNA . However , our analyses suggest this complex to be pentameric rather than hexameric as was proposed previously [11]–[13] . These results indicate a dynamic process during which Rep68 adopts different quaternary structures at distinct steps throughout the AAV DNA replication reaction .
Rep68 has two functional domains with independent DNA binding properties that are used at different stages of the viral life cycle: the OID binds the RBS double-stranded DNA specifically , while the AAA+ domain binds ssDNA or ss-dsDNA junctions non-specifically to perform the unwinding of DNA . We hypothesized that different oligomeric Rep68-DNA complexes are formed to carry out these diverse reactions . We first used size exclusion chromatography in order to investigate the in vitro oligomerization properties of Rep68 after binding either a 26-mer RBS dsDNA sequence or a 25-mer poly-dT oligonucleotide . Both complexes were analyzed on a Superose-6 column that was calibrated with proteins of known Stokes radii . As expected , the two complexes elute at different times: the Rep68-RBS complex elutes with an apparent molecular weight of ∼578 kDa ( Figure 1C ) , while the Rep68-ssDNA complex elutes earlier , with an apparent molecular weight of ∼2 . 3 MDa ( Figure 1B ) . The calculated Stokes radius indicates that the Rep68-ssDNA complex is roughly twice as large as the Rep68-RBS complex ( 106 Å and 73 . 9 Å respectively; Figure S1A and B ) . In the presence of non-specific dsDNA substrates ( Figure 1D ) , Rep68 did not efficiently oligomerize , although a slight difference in the elution profile can be observed when this complex is compared to apo Rep68 ( Figure 1A ) . Purified Rep68-RBS and Rep68-ssDNA complexes were further analyzed by small-angle X-ray scattering ( SAXS ) , and the radii of gyration were determined to correspond to 81 . 65+/−2 . 34 Å and 154 . 895+/−1 . 327 Å , respectively ( Figure S1C ) . These data are in agreement with the gel filtration results . Taken together , these results show that Rep68 can form different oligomers depending on the DNA substrate . In order to further examine the molecular weights of both Rep68 complexes , sedimentation velocity experiments were performed . Figure 1E shows that Rep68-RBS complex sediments with a coefficient S20 , w of 11 . 5S ( Figure 1E ) , while Rep68-ssDNA complex sediments faster , with a sedimentation coefficient S20 , w of 21 . 9S ( Figure 1F ) . A MW of ∼318 kDa was calculated for the Rep68-RBS complex . In contrast , for the Rep68-ssDNA complex a MW of ∼1 MDa was determined ( Table 1 ) . We further analyzed the Rep68-RBS complex using sedimentation equilibrium ( SE ) ultracentrifugation using two different concentrations at three increasing speeds . The complex was first purified by gel filtration and concentrated before SE . Global fitting yielded a molecular weight of ∼311 kDa for the complex at low complex concentrations , and ∼324 kDa when the concentration was 3-fold higher . Both values are in agreement with the value calculated from sedimentation velocity experiments ( ∼318 kDa ) . Taking into account that the theoretical MWs for pentameric and hexameric Rep68-RBS complexes with one DNA molecule are 321 . 3 kDa and 382 . 2 kDa , respectively , our data indicate that Rep68 assembles on the RBS DNA rather as a pentamer than the previously proposed hexamer [11]–[13] . The observed discrepancy with the molecular weights determined by gel filtration are likely due to the non-spherical nature of both complexes as suggested by their high frictional coefficient ratios f/f0 ( 1 . 79 and 1 . 83 for Rep68-RBS and Rep68-ssDNA complexes , respectively ) . Previous studies have indicated that Rep68/Rep78 has two regions that are required for oligomerization in vitro: A putative coiled-coiled region located in the OID and the AAA+ C-terminal domain [13] . We used size exclusion chromatography in order to determine if the individual domains are able to form higher molecular weight complexes in the presence of ssDNA . In the absence of ligands , Rep40 elutes as monomer with an estimated MW of 45 kDa ( Figure 2A ) . Unlike Rep68 , this profile does not change in the presence of ssDNA ( Figure 2B and C ) . The OID alone also elutes as monomer ( estimated MW ∼38 kDa; Figure 2D ) and does not oligomerize in the presence of ssDNA ( Figure 2E and F ) . Control experiments show that the OID is capable of forming a higher MW complex in the presence of RBS DNA ( Figure 2G and H ) . By means of sedimentation velocity analyses on the interaction of OID with RBS , we determined that the 5∶1 OID∶RBS stoichiometry is formed at a salt concentration of 50 mM ( data not shown ) . The conditions of the experiment shown in Figure 2G and H contain 200 mM NaCl , and support the formation of a complex with only 2 molecules of N208 bound to the RBS site ( data not shown ) . Therefore , these results together with our observation that Rep68 oligomerizes in the presence of ssDNA , demonstrate a requirement of both domains to form the Rep68-ssDNA complex . These studies further demonstrate that ssDNA elutes as a free form , suggesting that neither Rep40 nor the OID is interacting with the ssDNA under these conditions . This prompted us to examine the ssDNA binding affinities of the three protein constructs . Binding affinities were determined using fluorescent polarization on a fluorescein labeled poly- ( dT ) 38 oligonucleotide . Figure S2 shows the binding isotherms for all three proteins , with a Rep40 binding constant of ∼3500 nM while OID binds ssDNA with higher affinity and a binding constant of ∼130 nM . As expected , Rep68 exhibits the highest affinity to ssDNA , with a binding constant of 23 nM . The large difference in affinities shown by the individual domains suggests a significant level of cooperativity involved during Rep68 binding to ssDNA . This finding invited the question of whether residues involved in the respective DNA interactions by the individual domains influence the formation of the complex and thus contribute to the cooperativity . We have previously shown that B′ motif residues K404 and K406 located on β-hairpin-1 of the AAA+ domain of Rep40 , are essential for ssDNA binding and helicase activity [16] . On the other hand , R107 located on the OID was shown to be essential for origin binding and nicking , as well as plasmid integration into the AAVS1 site [17] . It was later shown that this residue directly interacts with origin DNA [18] . His-tagged variants of all mutants were used and shown to elute as a single peaks in the absence of ssDNA ( Figure 3C , E , G , and I ) . Albeit at somewhat lower efficiency , WT His-Rep68 oligomerizes in the presence of ssDNA ( Figure 3D ) , and shows a similar elution profile as non-tagged Rep68 with ssDNA ( Figure 3B ) . Mutation of either K404 or K406 did not affect His-Rep68 oligomerization in vitro ( Figure 3F and H ) , indicating that these ssDNA binding residues of the helicase domain are not required for ssDNA-dependent oligomerization . In contrast , mutation of R107 residue completely eliminated His-Rep68 oligomerization , which was accompanied by the appearance of a new peak ( ssDNA ) at later elution volumes ( Figure 3J ) . As a quality control and to rule out the possibility of an unfolded R107A mutant , we recorded the Circular Dichroism spectrum of both the HisRep68wt and R107 mutant proteins , which show similar profiles ( Figure S3 ) . Interestingly , R107A mutation is shifting Rep68 elution to species of lower molecular weight ( Figure 3I ) , suggesting that this residue is involved in the Rep68 oligomeric interface directly or indirectly , in addition to its role in DNA binding . Altogether , these findings suggest that R107 , and by extension the OID , is critical for ssDNA-dependent Rep68 oligomerization in vitro . Our sedimentation velocity experiments suggest that Rep68 assembles into a ∼1 MDa complex in the presence of ssDNA . In order to gain structural information of the Rep68-ssDNA/Rep68 complex , cryo-electron microscopy ( CEM ) combined with single-particle analysis was used . For this , we purified the complex by size-exclusion chromatography and analyzed frozen samples by EM . We readily observed ring-shaped molecules ( Figure 4A ) , a characteristic feature of AAA+ proteins , and SF3 helicases in particular [19] , along with other views of the complex . Reference-free 2D alignment and classification of 852 rings was performed without imposing symmetry . Surprisingly , all classes showed a ring with eight-fold symmetry ( a representative class is shown in Figure 4C ) . The octameric ring has an external diameter of 145 Å , and an internal diameter of 70 Å . In addition , we also observed elongated particles ( Figure 4B ) . Using the same approach , 363 elongated particles were aligned without references and classified . A representative two-dimensional average view is shown in Figure 4D . The averaged view shows a clear two-fold symmetry , indicating that Rep68 assembles into double octameric rings in the presence of ssDNA . The dimensions of this double octamer are 145×220 Å . Interestingly , this analysis indicates that the two rings are assembled in opposite orientation . The overall shape of this side view strongly resembles the double-hexameric LTag [20] , suggesting that Rep68 rings are interacting through their N-terminal domains . As a result , any additional domain attached to the N-terminal is likely to affect the formation of the complex . This possibility is in accordance with the observation that the His-tagged version of Rep68 does not form the complex as efficiently as the non-tagged protein ( Figure 3C ) . In order to put the experimental projections into a structural context , a double octameric atomic model was generated and 2D projections were deduced for comparison . The model was built from the coordinates of the available AAV5 OID1–197 and the AAV2 Rep40224–490 structures [14] , [21] . In this Rep68 model , the RBS interacting residue R107 is facing the internal channel and it is in proximity to the helicase domain , as it is found for the origin interacting residues in the double hexameric SV40 LTag structure [20] . Using the programs pdb2mrc and project3D from the EMAN package [22] , a 3D map and 2D projections were obtained , respectively , without enforced symmetry . As shown in Figures 5E and F , the calculated 2D projections of the double-octameric Rep68 model resembles those observed in the CEM analyses: first , the ring is octameric , and second , the two rings indeed interact through the origin interaction domains . Projections generated with a double-hexameric ring model did not resemble the properties of the experimental projections ( data not shown ) . In addition , inversion of the rings so that the helicase domains interact with each other did not yield projections that resemble the experimental CEM sideviews ( data not shown ) . Therefore , the CEM data supports the conclusion that ssDNA promotes the in vitro oligomerization of Rep68 into head-to-head double octameric rings . In order to examine the functionality of the observed Rep structure , heteroduplex DNA substrates were used to evaluate Rep68 oligomerization in vitro as compared to ssDNA containing complexes . Since a non-specific blunt-ended dsDNA does not promote Rep68 oligomerization ( Figure 1 ) , a heteroduplex substrate containing a 25-nucleotide ssDNA 3′- ( poly-dT ) tail juxtaposed to non-specific dsDNA was tested . As expected this ssDNA-dsDNA heteroduplex supported the formation of a Rep68 oligomer with an elution profile similar to that obtained with ssDNA ( Figure 5A and B ) . This Rep68-heteroduplex complex ( Figure 5D ) was further purified and analyzed by CEM , revealing ring-shaped and elongated particles with dimensions that are similar to those obtained with ssDNA ( Figure 5E and F ) . Reference-free 2D alignment of ring-shaped particles shows that Rep68 assembles into rings with dimensions of 148×68 Å , which are close to the dimensions of the Rep68-ssDNA ring ( Figure 5G ) , suggesting that the rings in the Rep68-heteroduplex complex are also octameric . The same heteroduplex DNA ( in this case labeled with Cy5 ) was then used to perform helicase assays . As expected , the Cy5-labeled heteroduplex supported Rep68 oligomerization , albeit with less efficiency , into a complex with similar elution time ( Figure 5C ) . Decreasing the heteroduplex concentration allowed efficient formation of the Rep68 oligomer and its subsequent purification ( Figure 5H ) . Notably , in the presence , but not in the absence of ATP and magnesium , we observed that the purified Rep68-heteroduplex complex was indeed capable of unwinding DNA ( Figure 5I ) , suggesting that Rep68 interacts with the ssDNA tail of the heteroduplex substrate , leading to the formation of an active helicase complex . Although we cannot discard the possibility that cofactors like ATP and Mg2+ could influence the ssDNA-dependent Rep68 oligomerization , we have observed that ATP , ATP/Mg2+ , and ATP/Ca2+ -at concentrations used in our helicase assay- supported the formation of the same Rep68-ssDNA complex with the same efficiency when compared to ssDNA alone ( Figure S4 ) .
Superfamily 3 helicases include the viral initiator proteins BPV E1 , SV40 LTag and AAV Rep68/78 , among others . The function of these initiator proteins during viral DNA replication relies on their ability to oligomerize upon binding and subsequently melt their respective origin DNA . For E1 and LTag , it has been shown that they assemble into double hexameric rings on viral origin DNA , the oligomeric structure that is required for viral DNA unwinding during replication . Rep68/78 has also been shown to oligomerize in the presence of its origin DNA . Although it as been suggested that Rep68/78 forms hexamers , its oligomeric structure remains to be determined . However , based on the structural similarity of its AAA+ domain with E1 and Tag , it has been hypothesized that Rep68/78 would assemble into hexameric rings . Thus far no definitive experimental evidence has been presented that proves this hypothesis . As these viral initiator proteins are necessarily multifunctional , we set out to investigate the ability of the DNA structure to modulate the oligomeric state of Rep68 . We found that Rep68 forms a complex with RBS dsDNA containing five subunits of Rep68 . This Rep685-RBS complex is in accord with the crystal structure of the OID-RBS complex , which shows five OIDs bound to the RBS DNA [18] . Thus we have demonstrated that the AAA+ motor domain does not influence the number of Rep68 subunits that bind the RBS . However , the motor domain might influence the overall structure of both the OIDs and RBS DNA in the complex; therefore , additional structural investigation of the Rep68-RBS complex is necessary to elucidate this question . Our results appear to be in contradiction with previous attempts to determine the stoichiometry of the Rep68/Rep78-origin complex but a closer look at the literature shows that this is not the case . For instance a report by Smith et al . [13] found that Rep78 forms a hexamer on an AAV ORI DNA molecule using gel filtration analysis . The DNA used is their study is 63 bp long , while we used a 26 bp DNA containing only the minimal RBS sequence . The chance of more Rep78 molecules binding to the longer DNA site is very likely . Moreover , the estimated stokes radius of the complex ( 64 Å ) appears to be too small , particularly when the length of the DNA used is almost 215 Å . In the same report cross-linked Rep78 to AAV ORI DNA was analyzed on SDS-PAGE . The gel shows six clear bands , however , the presence of higher molecular weight complexes that did not enter the gel was not taken into account [13] . In another report , Muzyczka and colleagues introduced the concept that Rep68 can adopt different oligomeric states on ITR DNA , depending on Rep68 concentration as well as on the presence of ATP [12] . These investigators used native polyacrylamide gels to determine the molecular weight of the different Rep68:ITR complexes . However no precise determination of the Rep:DNA stoichiometry could be obtained . Nevertheless , as the authors point out , the AAV ITR DNA used contained additional contact points that are recognized by Rep68/78 [18] , [23] , [24] , which could contribute to the binding of additional Rep68 molecules . Interestingly , at high Rep68 concentrations and in the presence of ATP , Rep68 binds the ITR mainly as a complex described as PDC5 , which appears to contain 5 molecules of Rep68 [12] . Dignam et al . calculated a S20 , w value for the Rep68-RBS complex of 13 . 15 [11] . We obtained an S20 , w value of 11 . 5S . However , the difference of almost 2S indicates a real distinction between the two complexes . This disparity can be attributed to either a difference in the DNA substrate and/or buffer conditions . The RBS DNA site used by Digman ( A stem ) contains compatible overhangs of 4 and 6 nucleotides that could hybridize to produce longer DNA substrates where more Rep68 molecules could bind . In contrast , our RBS substrate has blunt ends . The claim by the authors that the sedimentation coefficient of 13 . 15 is “consisted with a tight complex comprised of two A-stem per six Rep68 subunits” supports the possibility of two concatenated A-stem DNA sites . In fact , a calculated sedimentation coefficient from the atomic model of the RBS site using the program HYDROPRO [25] , predicts a sedimentation coefficient of ∼2 . 2S which is consistent with the experimental sedimentation coefficient of 2 . 4S that we obtained for the RBS site ( data not shown ) . In contrast , Dignam et al . obtained a sedimentation coefficient for the RBS site of 3 . 28S . Moreover , their experimental conditions at low salt ( 50 and 100 mM ) increases the likelihood of more Rep68 molecules binding . We propose that the pentameric assembly of Rep68 on the minimal RBS site could represent an intermediate complex that would require further assembly of Rep68 molecules in the presence of the full-length ITR origin molecule . Indeed , studies by Hickman et al . describe that they could detect a sixth OBD molecule using a longer RBS site than the 26-mer used in the crystallographic studies . However , using the same biochemical assay and full length Rep68 , the number of molecules bound is now 7 . 5 [18] . Clearly , further biophysical and structural analysis of the Rep68-ITR complex in a purified form will provide a better understanding of the oligomeric nature of Rep68 when bound to the AAV ITR . We further show that Rep68 self-assembles into double octamers upon binding ssDNA as well as heteroduplex helicase substrates , demonstrating a novel oligomeric structure of an SF3 helicase . This is in contrast to the hexameric-ring complexes formed by the equivalent AAA+ domains of both E1 and LTag upon binding ssDNA [10] , [26] . Although our current structural models do not provide conclusive data indicating a molecular basis for the formation of octameric rings , it is likely that subunit-subunit interactions within the AAA+ ring are more stable in the octamer as compared to a possible hexamer conformation of Rep68 under the conditions used in our experiments . In addition , the ssDNA substrate might direct Rep68 into a conformation that matches the dimensions required to efficiently support both DNA replication and integration through a complex that is assembled from cellular replication factors . We demonstrate that both the OID and the motor domain function cooperatively to assemble a double octamer and confer higher affinity binding to DNA . Therefore , the OID plays an important role in determining the symmetry of the ring by establishing subunit-subunit interactions in the OID ring that influence the interactions in the AAA+ ring . Although we cannot exclude the possibility that essential cofactors such ATP and magnesium could potentially influence the oligomeric state of Rep68 , our results show that in presence of ssDNA this is not the case . Moreover , we have observed that ATP alone supports the formation of ring-shaped Rep68 , whose dimensions are very similar to the rings obtained with ssDNA ( data not shown ) . Preliminary gel filtration analyses suggest that this Rep68-ATP complex corresponds to a single Rep68 ring , and it requires both the OID and the AAA+ domains ( data not shown ) . Altogether , our results suggest that Rep68 is poised to form octameric rather than hexameric rings . Nevertheless , it is noteworthy that the double octameric structure may only assemble under our experimental conditions . A detailed study of Rep68 in complex with ITR during the different steps of the terminal resolution reaction will be needed , in order to determine the biological relevance of this complex . To date , the molecular mechanisms of Rep68/78 assembly during ITR resolution and the function ( s ) associated to each Rep oligomer are yet to be determined . Based on our observations , we hypothesized that DNA structure plays an important role in controlling the oligomeric nature of Rep68/78 . In our in vitro conditions , we observe a stable Rep68:RBS complex that contains 5 molecules of Rep , and it may represents an initial complex that would require further assembly to initiate the ITR resolution reaction . It has been proposed that RBS melting is needed for the formation of hairpinned terminal resolution site ( TRS ) , which is followed by nicking of the TRS by Rep68 [27] . Although there is no structural information of the Rep68/78 complex after RBS melting , it is likely that a ring-shaped Rep68/78 will form because of the ssDNA RBS that appears during melting as is the case with SV40 Tag and BPV E1 . Our results further show that Rep68 is functional as an octameric helicase , and we propose that both helicase rings may be active in this bidirectional complex . Although the proposed structure might have implications for our current replication model , the exact role of a double-octameric Rep68 in AAV DNA replication and/or site-specific integration remains to be determined . However , several scenarios are plausible . The current model for AAV DNA replication does not envision bidirectional replication [28] , as it has been proposed for the SV40 and papilloma viruses . These viruses have a double-stranded DNA origin that contains two inverted repeats that are both recognized by the respective initiator protein . In contrast , AAV contains a single repeat ( the RBS ) in each ITR . Using LTag and E1 provide as precedence , Rep68 would be expected to require two inverted repeats in order to assemble a double octamer . In view of biochemical evidence , which suggests that Rep68 can form ternary complexes with 2 AAV ITRs [12] , the Rep68 double octamer may coordinate the resolution of two ITR molecules ( as may be the case of intermolecular unwinding ) . Another interesting scenario is the requirement of a double octamer during the refolding of ITR structures after completion of the ITR resolution and its subsequent duplication . Interestingly , two inverted RBS sequences are obtained after these steps , and , in theory , Rep68/78 proteins have the potential to recognize them and initiate their melting , followed by the formation of a double octamer , which would not only allow the refolding of the ITR structures but also the unwinding of the AAV dsDNA required for the following rounds of replication . Identifying the exact role of the Rep68 double octamer during AAV life cycle as well as its structural characterization will help to understand how Rep68 functions during the unwinding reaction . In addition to the complexes presented here , it is plausible that Rep68 will assemble into additional different structure with other DNA substrates . Among the SF3 helicases , AAV Rep68/78 initiator protein is unique because of its ability to nick its origin DNA . During ITR resolution , the terminal resolution site ( TRS ) hairpin DNA is formed after RBS melting . This TRS hairpin is recognized and nicked by Rep68/78 in a sequence-specific manner . Therefore , we suggest that there exists a coordinated Rep68/78 oligomerization during origin DNA binding , melting , and nicking . Finally , the initiator protein Rep68/78 is also required for the site-specific integration of the viral DNA into the AAVS1 locus [1] . This locus contains RBS- and TRS-like sequences , which represent the minimal cis elements required for AAV integration [29] . Besides the recognition and nicking of these sequences [17] , [30] , Rep68/78 has been shown to form ternary complexes with AAV ITR and AAVS1 RBS DNAs , implying the interaction of two origins complexed through oligomeric Reps in this process [31] . Our findings demonstrate the versatility of Rep68 regarding its ability to assemble into different quaternary structures depending on the DNA substrate provided . Moreover , the data supports the idea that Rep68 can oligomerize through distinct pathways , with a pathway that relies on the cooperativity between the OID and the motor domain – as in the case of ssDNA - and a pathway that only requires the OID – in the case when RBS DNA is recognized . We propose that this flexibility in oligomerization provides Rep68 with the possibility to accommodate the different DNA structures it encounters during its involvement in all aspects of the AAV life cycle . Furthermore , our findings show a striking difference in oligomerization potential between Rep40 and Rep68 , despite the fact that these two share the identical helicase domain . We hypothesize that AAV might have evolved to utilize a helicase domain that could support two different modes of DNA unwinding . This difference in the oligomerization-based mechanism may support the differential roles of Rep40 versus Rep68 in AAV DNA packaging and DNA replication/integration , respectively . In conclusion , our study demonstrates that DNA structure modulates Rep68 oligomerization , requiring specific domain contribution of Rep68 depending on the DNA ligand . AAV ITR resolution and genome integration into the AAVS1 locus are complex reactions , where distinct Rep68-DNA complexes are expected to arise . Structural studies of these complexes are central for the elucidation of the molecular mechanisms of AAV DNA replication and site-specific integration into the human genome .
Rep40 and OID-N208 Reps were expressed and purified as described [16] , [32] , except that the final buffer corresponded to Buffer A ( 25 mM Tris-HCl [pH 8 . 0 at 4°C] , 200 mM NaCl , 5% glycerol , and 1 mM TCEP ) . His6-PreScission Protease ( PP ) cleavage site-Rep68 fusion protein was expressed in BL21 ( DE3 ) pLysS bacteria at 37°C for 3 h , in LB medium containing 1 mM IPTG . Cell pellets were lysed in 1∶1 Ni-Buffer A ( 20 mM Tris-HCl [pH 7 . 9 at 4°C] , 500 mM NaCl , 10% glycerol , 0 . 05% NP-40 , and 5 mM imidazole ) : B-PER solution ( Pierce ) containing protease inhibitors ( 2 µg/ml aprotinin , 2 µg/ml leupeptin , 1 µg/ml pepstatin A , and 600 µM PMSF ) . After five 10-s cycles of sonication , the fusion protein was purified using a Ni-column –equilibrated in Ni-buffer A . Protein eluted with 300 mM imidazole was desalted using PP buffer ( 50 mM Tris-HCl [pH 7 . 0 at 4°C] , 200 mM NaCl , and 1 mM EDTA ) and a HiPrep™ 26/10 desalting column ( GE Healthcare ) . DTT was added to a final concentration of 1 mM , and His-PP tag was removed by PreScission protease treatment using 20 µg PP /mg His-PP-Rep68 . After overnight incubation at 4°C , buffer was exchanged using the same desalting column and Ni-Buffer A . Subsequent Ni-column chromatography was performed to remove the uncleaved fusion protein , and untagged Rep68 was eluted with 50 mM imidazole . Rep68 ( GE Healthcare ) was finally purified by gel filtration chromatography using a HiLoad Superdex 200 16/60 column and Buffer A . Purified Rep68 was concentrated up to 20 µM ( 1 . 2 mg/ml ) , flash-frozen in liquid N2 , and kept at −80°C until use . N-terminus His6-tagged WT and mutant Rep68 proteins were expressed and purified as above , except that proteins were directly concentrated after affinity purification , and loaded on the HiLoad Superdex 200 column . Rep68 ( 16 . 6 µM ) was incubated in the absence or presence of 16 . 6 µM ssDNA ( polydT25 ) , 16 . 6 µM RBS dsDNA ( generated with oligos JM-2: 5′ GCCTCAGTGAGCGAGCGAGCGCGCAGAG , and JM-20 CTCTGCGCGCTCGCTCGCTCACTGAGGC ) or 1 . 4 µM non-specific IRF3 dsDNA [33] for 30 min on ice . Samples ( 50 µL ) were chromatographed on a Superose 6 10/300 GL column ( GE Healthcare ) with a flow rate of 0 . 5 ml/min . For fractionation of Rep40 and OID proteins , Superdex 200 10/300 GL column ( GE Healthcare ) was used . Buffer A was used for all chromatographic analyses . Protein elution was detected by UV at 280 nm . For experiments using heteroduplexes , oligos JM-38 ( 5′-GGGAGAAGTGAAAGTGGGAA ( T ) 25 ) and JM-40 ( 5′-TTCCCACTTTCACTTCTCCC ) were used to generate the non-labeled heteroduplex , and oligos JM-37 ( same sequence as JM-38 with the Cy5 molecule at 3′ end ) and JM-40 were used to make the Cy5-heteroduplex . Formation of heteroduplexes was checked by gel filtration chromatography using the Superdex 75 column; in both cases a single peak was observed . MW standards of known Stokes radii ( GE Healthcare ) were used to estimate the MW and Stokes radius of Rep complexes . Sedimentation velocity experiments were carried out using a Beckman Optima XL-I analytical ultracentrifuge ( Beckman Coulter Inc . ) equipped with a four-position AN-60Ti rotor . Rep68 ( 1 mg/ml ) was incubated with 2 . 8 µM ssDNA ( polydT25 ) or RBS dsDNA in buffer A . Samples in aluminum double sector cells were centrifuged at 45 , 000 rpm at 20°C . Concentration profiles were recorded using UV absorption ( 280 nm & 260 nm ) and interference scanning optics , and analyzed using the program Sedfit [34] . We used a continuous distribution c ( s ) Lamm equation model with other prior knowledge that in this case is the number of species with different diffusion coefficients . We calculated the partial specific volume of the complex using the following equation: The vbar value used in the final calculation had a stoichiometry of 5∶1 ( Rep68∶RBS ) , but other stoichiometries were also considered during the analysis . The addition of an extra molecule of Rep to the Vbar only increases its value by 0 . 0016 thus having a small effect on the final molecular weight but without affecting the final conclusions . The sedimentation coefficients were corrected to standard conditions ( S20 , w ) using density and viscosity values calculated with SEDNTERP ( http://www . rasmb . bbri . org/ ) , a program developed by Hayes , Laue , and Philo . For sedimentation equilibrium experiments , the Rep68-RBS complex was purified by gel filtration , and concentrated to an OD260 of 0 . 25 or 0 . 75 . Each sample was analyzed at 4 , 000 , 5 , 000 , and 7 , 000 rpm at 20°C . Radial scans of the absorbance at 260 and 280 nm were taken every 4 h , and equilibrium was determined by comparing successive scans using WinMatch , a program developed by Yphantis and colleagues ( http://www . biotech . uconn . edu/auf/ ? i=aufftp ) . To obtain the background level at all three speeds , an over-speeding step at 42 , 000 rpm at 20°C for 6 h was performed , after which the speed was reduced to 4 , 000 rpm and radial scans were obtained . This procedure was repeated immediately for the other two speeds . After subtraction of the background level , the equilibrium concentration distributions were globally analyzed using HeteroAnalysis [35] . Rep68-ssDNA samples were prepared by purification of the complex by size-exclusion chromatography ( Superose 6 column ) . The central part of the peak was concentrated to about 0 . 4 mg/ml . For side view analysis , n-octyl β-D-glucopyranoside was added to a final concentration of 0 . 05% . This detergent did not affect ssDNA-dependent Rep68 oligomerization as determined by size-exclusion chromatography . Drops ( 3–4 µL ) of sample were applied to glow-discharged Quantifoil EM 300-mesh grids with 2-µm holes , which were then blotted and plunged into a bath of liquid ethane ( ∼−180°C ) . Grids were analyzed in a Tecnai F20 transmission electron microscope , using the Tecnai low-dose package . Images of particles suspended in ice were collected at a microscope magnification of 50 , 000 and a defocus of 3 µm on a Tietz F415 CCD camera . Particles were selected using the boxer program from the EMAN software package [22] . Reference-free 2D alignment and classification were done with both the EMAN and SPIDER [36] software packages with similar results . For the Rep68-ssDNA/dsDNA complex , an identical approach was taken , except that a limited number of endviews was used for 2D classification . The crystal structures of AAV5 OID1–197 [21] and AAV2 Rep40225–490 [14] were used to make a Rep68 atomic model , which lacks the linker region ( residues 198–224 ) as well as the last 46 aminoacids . The orientation of the domains in the oligomeric rings was based in the known crystal structures of the E1 and LTag hexamers [37] , [38] . The orientation of the rings in the double octamer was based on the CEM structure of the LTag double hexamer , in which both rings are interacting via their the N-terminal domains [20] . Dimensions of the double octamer were according to the cryo-EM data . 3D density maps at 30-Å resolution were obtained by using the EMAN program pdb2mrc . A series of 2D projections were obtained for each model by using the EMAN program project3d without symmetry imposed . Control reaction ( 10 µl ) contained 200 fmoles of Cy5-heteroduplex , 1 . 6 pmoles Rep68 , 25 mM Tris pH 7 . 4 ( 25°C ) , 20 mM NaCl , 5% glycerol , 1 mM MgCl2 , and 1 mM ATP . A negative control reaction contained the same components except Rep68 . To test activity of the Cy5-heteroduplex/Rep68 complex , Rep68 ( 1 mg/ml; 16 . 6 µM ) was incubated with 2 . 1 µM Cy5-heteroduplex in the presence of buffer A . After a 30-min incubation on ice , 100 µl of mix were loaded on the buffer A-equilibrated Superose 6 column . Fractions of 300 µl were collected , and the fraction corresponding to the central part of the complex peak was concentrated 4 times using Microcon concentrators ( 10 kDa cut-off; Millipore ) . The complex was incubated in the absence or presence of 1 mM MgCl2/1 mM ATP , with a final NaCl concentration of 20 mM . All reactions were carried out at 37°C for 30 min , and stopped by adding 7 µl of loading buffer ( 1× TBE , 0 . 5% SDS , 20% Glycerol ) , and immediately loaded on a native 16%-polyacrylamide gel . After electrophoresis , the Cy5-substrate and Cy5-ssDNA were detected using the STORM 860 phosphorimager set for red fluorescence detection . Oligo JM-37 was used as a marker for the ssDNA position . The DNA sequences of the proteins used in this manuscript are according to the AAV2 genome sequence . AAV2 genome: GenBank accession number AF043303 . Rep68 protein: GenBank accession number AAC03774 . Rep40 protein: GenBank accession number AAC03776 . See Text S1 for supporting materials and methods . | Adeno-associated virus ( AAV ) is a parvovirus with a linear single-stranded DNA genome . Thus far , it is the only eukaryotic virus known to integrate its genome in human cells in a specific region of chromosome 19 . Because no pathologies have been associated with AAV , there is great interest in using AAV as a vector for gene therapy . The genetic information of AAV encodes for both the structural Capsid proteins and the Rep proteins . We have studied a protein called Rep68 , which is essential for both AAV genome replication and site-specific integration in chromosome 19 , and found that it forms distinct structures in the presence of different DNA structures . Of particular interest is the formation of a Rep68 structure composed of two opposite rings , which resemble the structures formed by the large T antigen and E1 viral proteins of the tumor-inducing Simian virus 40 ( SV40 ) and papilloma viruses , respectively . The double-ring structure of these viral proteins is essential for viral DNA replication , which suggests that AAV has evolved a similar mechanism of DNA replication that relies on a double-ring Rep68 . Moreover , Rep68 encounters different DNA structures during viral genome replication , and our results show how Rep68 can adapt to these changes . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biophysics/replication",
"and",
"repair",
"biochemistry/replication",
"and",
"repair",
"biophysics/macromolecular",
"assemblies",
"and",
"machines",
"biochemistry/macromolecular",
"assemblies",
"and",
"machines"
] | 2009 | DNA Structure Modulates the Oligomerization Properties of the AAV Initiator Protein Rep68 |
P pili are important adhesive fibres involved in kidney infection by uropathogenic Escherichia coli strains . P pili are assembled by the conserved chaperone–usher pathway , which involves the PapD chaperone and the PapC usher . During pilus assembly , subunits are incorporated into the growing fiber via the donor–strand exchange ( DSE ) mechanism , whereby the chaperone's G1 β-strand that complements the incomplete immunoglobulin-fold of each subunit is displaced by the N-terminal extension ( Nte ) of an incoming subunit . P pili comprise a helical rod , a tip fibrillum , and an adhesin at the distal end . PapA is the rod subunit and is assembled into a superhelical right-handed structure . Here , we have solved the structure of a ternary complex of PapD bound to PapA through donor–strand complementation , itself bound to another PapA subunit through DSE . This structure provides insight into the structural basis of the DSE reaction involving this important pilus subunit . Using gel filtration chromatography and electron microscopy on a number of PapA Nte mutants , we establish that PapA differs in its mode of assembly compared with other Pap subunits , involving a much larger Nte that encompasses not only the DSE region of the Nte but also the region N-terminal to it .
Urinary tract infections , which include infections of the bladder ( cystitis ) and kidney ( pyelonephritis ) , are some of the most common bacterial infections . These infections are caused mainly by uropathogenic Escherichia coli [1] . Once uropathogenic E . coli is introduced , survival and persistence of these bacteria in the urinary tract require a specific set of virulence factors , including the expression of type P pili . P pili are specifically required for the ability of uropathogenic E . coli to bind Gal-α ( 1–4 ) -Gal moieties in human kidney cells and cause pyelonephritis [2 , 3] . P pili are encoded by the pap gene cluster and are assembled via the highly conserved chaperone–usher pathway , involving the periplasmic immunoglobulin ( Ig ) –like chaperone PapD and an outer membrane usher PapC [4 , 5] . P pili consist of six subunits making up a composite fiber with a short tip fibrillum composed of the PapE subunit joined to a more rigid helical rod composed of the PapA subunit [6 , 7] . PapG is the adhesin at the end of a tip fibrillum; PapK and PapF are adaptor subunits between the PapA rod and the PapE fibrillum and between the PapE fibrillum and the PapG adhesin , respectively; finally , PapH terminates P pilus formation [8 , 9] . The PapA rod is formed by more than 1 , 000 PapA molecules assembled in a right-handed helical manner , with 3 . 3 molecules per turn [6 , 10] . All pilin subunits adopt an Ig-like fold but lack the seventh , C-terminal G β-strand , thus producing a large hydrophobic groove on the side of the protein ( Figure 1 and [11 , 12] ) . In a process called donor–strand complementation ( DSC ) , the G1 β-strand of PapD inserts a conserved motif of three alternating hydrophobic residues ( called the P1 to P3 residues ) plus N101 ( P4 residue ) into four binding pockets in the hydrophobic groove of the pilus subunits ( P1 to P4 binding pockets ) . The G1 strand provides the structural information lacking in the pilus subunit by completing its Ig fold [11–13] . Pilus subunit assembly proceeds through a noncovalent polymerization process called donor–strand exchange ( DSE; Figure 1 ) . All subunits , except for the adhesin , possess an N-terminal extension ( Nte ) peptide of 11 ( PapK ) , 12 ( PapE and PapF ) , 19 ( PapA ) , or 33 ( PapH ) residues ( Figures 1 and S1 ) that is disordered and not part of their Ig-like structure . The Nte comprises a highly conserved array of alternating hydrophobic residues , called the P2 to P5 residues [14 , 15] . This array constitutes the DSE region of the Nte ( see Figure S1 for location of this DSE region ) . As chaperone–subunit complexes are differentially targeted to the usher [16 , 17] , each subunit donates its Nte to complete the Ig-fold of the subunit previously assembled by inserting its P2–P5 residues into the corresponding P2–P5 binding pockets , thus first displacing and then replacing the chaperone G1 strand in the groove of the previously assembled pilus subunit [18–21] . This process occurs through a zip-in–zip-out process whereby the DSE reaction is initiated by the insertion of the P5 residue of the Nte of one subunit into the P5 pocket of the groove of the other [21] . This binding event leads to the formation of a transient ternary complex , the formation of which is essential for the DSE reaction to proceed . Insertion of the Nte of PapH into the groove of a PapA subunit terminates pilus biogenesis because PapH lacks a P5 pocket and thus cannot provide the initiator-binding event required for the exchange reaction with another subunit [9] . Recently , in a departure from the more conventional model described above , Mu et al . [22] suggested , based on electron microscopy and image reconstruction of the PapA rod , that the DSE region of the Nte of PapA is not involved in DSE , but instead the region of the Nte N-terminal to the DSE region is involved in the process . Interestingly , the Nte of PapA is longer than the Nte of most Pap subunits except PapH ( Figure S1 ) , and thus the model proposed by Mu et al . [22] could potentially explain why residues N-terminal to the DSE region would be required in the process of P pilus biogenesis . This is explored further in this report , in which we describe the structures of a binary complex of PapD bound to PapA and of a ternary complex containing the chaperone PapD and two PapA subunits . In this ternary complex , PapD is bound to PapA through DSC , and this subunit is itself bound to another PapA subunit through DSE . These structures are used as a basis for a detailed mutational study dissecting the requirements for PapA polymer formation .
In order to investigate the PapD/PapA and PapA/PapA interactions within the P pilus rod , the first structure which we solved was that of PapD/PapA , where two mutations were introduced in the papA gene . This was necessary because PapA in purified wild-type PapD/PapA complexes tends to spontaneously polymerize , and self-polymerization of PapA has prevented structural studies . As PapA , like any other pilus subunits , polymerizes through DSE , mutating residues in the Nte was an obvious starting point to obtain a PapA mutant unable to polymerize . We first deleted the entire Nte ( residues 2 to 19; PapANtd2; Figure S1 ) , but coexpression of such mutants with PapD did not result in a material amenable to purification ( unpublished data ) . Thus , a second mutant was designed that conserved the entire Nte but introduced an Asn at position 15 . Position 15 , a Gly residue in the wild-type Nte of PapA , locates in the DSE region of the Nte and is strictly conserved among all Pap Ntes ( Figure S1 ) . G15 is indeed required because , after DSE , it lies in the P4 pocket of the subunit's grooves , which , in this region and in all Pap subunits , contains a bulky phenylalanine or a tyrosine ( F152 in PapA , Y146 in PapK , F138 in PapE , F137 in PapF , and Y162 in PapH ) [11 , 18] . We thus mutated G15 to N; however , PapAG15N in complex with PapD also undergoes spontaneous polymerization ( Figure S2A , top panel ) . Thus , in addition to the G15N mutation , a mutant where residues 2 to 8 were deleted was next constructed . The deleted region is just N-terminal to the DSE region of the Nte of PapA . This mutant ( PapANtd1_G15N ) did not undergo spontaneous polymerization , and formed a stable complex with PapD ( Figure S2A , middle panel ) . This experiment indicates that , as suggested by the electron microscopy study of Mu et al . [22] , the region of the Nte N-terminal to the DSE region is involved in PapA/PapA interaction . Crystals of PapD/PapANtd1_G15N diffracted to a resolution of 2 . 6 Å , and the structure was solved by molecular replacement using the PapD/PapK structure as a search model . This structure is very similar to the already known PapD/PapK , PapD/PapENtd , or PapD/PapHNtd1 structures [9 , 11 , 18] . Like PapK , PapE , or PapH , PapA lacks strand G of its Ig-fold; PapD complements this by donating its G1 strand . By itself , the PapD/PapANtd1_G15N structure is not very informative . However , by investigating a different PapA mutant with the additional mutation T101L , we obtained crystals of a complex containing one PapD molecule and two PapANtd1_G15N_T101L molecules . This T101L mutant was initially designed to fill in the PapA P5 pocket . Indeed , as explained in the Introduction , all Pap subunits , except PapH , have a clear P5 pocket , which serves as an initiator point , the occupation of which triggers the DSE reaction [9 , 21] . Thus , the PapANtd1_G15N_T101L was made to test the possibility that by filling its P5 pocket , PapA would become more like PapH in being unable to undergo DSE . As shown in Figure S2B , indeed , PapANtd1_G15N_T101L has reduced DSE activity compared with that of PapANtd1_G15N . However , this T101L mutation had the additional unexpected consequence of stabilizing a complex containing a 1:2 molar ratio of PapD and PapA ( Figure S2A , lower panel ) . PapD/ ( PapANtd1_G15N_T101L ) 2 was thus purified . Crystals were produced diffracting to 2 . 5 Å resolution . The structure of this ternary complex is shown in Figure 2 . It clearly shows one PapA molecule bound to PapD through DSC in a complex very similar to PapD/PapANtd1_G15N; however , this time , the Nte of the donor–strand-complemented PapA molecule is bound to the groove of another PapA molecule , and thus this ternary complex crystal structure provides a snapshot of PapA before and after DSE . In that respect , the PapD/ ( PapANtd1_G15N_T101L ) 2 complex is similar to the one obtained by Zavialov et al . for the Caf system [19] . Figure 3A shows a superimposition of the two PapA molecules in the PapD/ ( PapANtd1_G15N_T101L ) 2 structure , with the donor–strand-complemented PapA subunit ( dscPapA ) in purple and the donor–strand-exchanged PapA subunit ( dsePapA ) in orange . The core sheet structure of dsePapA is in a closer conformation than that of dscPapA , as the β-strands on each side of the groove of dsePapA are nearer to each other . Also , the “63–74” loop is ordered in dsePapA and not in dscPapA , as this molecule is missing residues 70 to 73 in this region . The truncated Nte of dscPapA ( as indicated above , residues 2 to 8 were removed to create PapANtd1 ) is clearly visible in the groove of dsePapA from residue 10 ( the two first residues of PapANtd1_G15N_T101L , which are residues 1 and 9 of full-length PapA , were not defined in the electron density ) . DsePapA is only visible in the electron density from residue 20 , as its Nte is not interacting in the groove of another PapA molecule and is thus disordered . Figure 3B shows the surface of dsePapA bound to the truncated Nte of dscPapA ( left panel ) and that of dscPapA bound to the G1 strand of the chaperone ( right panel ) : the PapD G1 strand interacts as expected at the P1 to P4 positions in the groove of dscPapA ( right panel ) , the Nte of which interacts in the P2 to P5 pockets in the groove of dsePapA ( left panel ) . It is noticeable in the left panel of Figure 3B that the dsePapA groove is extending beyond the region occupied by the P2 residue of dscPapA Nte . However , the groove of dscPapA is not extending beyond the P1 pocket , due to the disordered “63–74” loop . Modeling of the first nine residues of PapA Nte in the extended groove of dsePapA shows that this extended groove has the right length and shape to accommodate the nine missing residues in the Nte of PapANtd1 ( unpublished data ) . Thus , the groove of PapA is long enough to accommodate the extended Nte of another PapA molecule . Figure 4A shows details of the grooves of dscPapANtd1_G15N ( left panel ) and dsePapANtd1_G15N_T101L ( right panel ) . Figure 4A on the left panel shows that in the PapD/PapANtd1_G15N structure , the P4 position of the groove is formed by F152 . This configuration of F152 is similar to that seen in the equivalent position of the PapD/PapK , PapD/PapE , PapD/PapH , and PapE/KNte complexes [9 , 11 , 18] . As mentioned above , the bulk created in the P4 pocket by F152 and equivalents , by being able to accommodate only a conserved Gly in the Ntes , acts as a registering device that calibrates the positioning of Ntes in the subunits' grooves . In contrast to what is observed in the PapE/KNte or any other chaperone–subunit complex structures , the side chain of F152 has moved out of the P4 pocket of the DsePapA structure ( Figure 4A , right panel ) and the P4 pocket can now accommodate the N15 mutation ( substituted for wild-type G15 in the PapANtd1_G15N_T101L mutant to prevent higher-order polymerization ) . F152 in the dsePapA structure is allowed to move away from the groove position because of the T101L mutation . Indeed , as shown in Figure 4B , the T101L mutation and the insertion of N15 induce a rearrangement of the side-chains in the P4–P5 region , leading to T99 moving out of the P5 region , thereby leaving room for F152 to substitute in its place , the new F152 position being stabilized by L101 . Like dsePapA in the PapD/ ( PapANtd1_G15N_T101L ) 2 complex structure , PapE in the PapE/KNte structure is also donor–strand exchanged ( dsePapE ) , but with the Nte of PapK ( KNte ) [18] . Comparing dsePapE ( Figure 5A , right ) with dsePapA ( Figure 5A , left ) shows that , while the groove of dsePapE stops at the P2 pocket , the groove of dsePapA is extending beyond this pocket . This is due to the presence of the “63–74” loop in dsePapA , a loop which is much shorter in dsePapE , and to the closure of dsePapE groove by PapE N- and C-termini compared with the open groove in dsePapA ( Figure 5B ) . A number of PapA Nte mutants were next produced in order to evaluate the effect of these mutations on polymerization and pilus formation . In addition to PapA wild-type , six constructs were studied . In the DSE region , a G15N mutation and a Δ11–17 deletion ( where the entire DSE region is removed ) were made . In the region N-terminal to the DSE region of the Nte , an I4G single-site mutant and the Ntd1 deletion described above were studied . I4 is a bulky residue in that region of the Nte and thus would contribute significantly to the interface , were it to be involved in groove/Nte interaction ( see [21] for consideration regarding surface area contribution of residues in the Nte ) . We also combined the Ntd1 ( Δ2–8 ) deletion with the G15N mutation and the I4G and G15N mutations . Polymerization was assessed by gel filtration immediately after purification of the corresponding PapD/PapA complexes , and pilus formation was assessed after freeze–thaw of PapD/PapA complex preparations using electron microscopy ( EM ) . Thus , gel filtration provides information on the limited polymerization events taking place early on during polymerization , while freeze–thaw of PapD/PapA complexes followed by analysis by EM provides information on the ability of the various PapA molecules to form pili . Results obtained for each of the PapA constructs are presented in Figure 6 , where each panel provides the elution profile and EM micrograph for each wild-type and mutated PapA . For those complexes that exhibited a gel filtration peak corresponding to PapD/ ( PapA ) 2 , further polymerization was checked by pooling the fractions corresponding to PapD/ ( PapA ) 2 and running another gel filtration the next day . For those complexes that did not exhibit polymerization , the PapD/ ( PapA ) 1 peak was rerun on gel filtration to make sure that indeed no polymers were formed . Gel filtration of PapD/PapAwt complexes reveals a major 1:2 PapD/PapA ( PapD/ ( PapA ) 2 ) complex . Yet , this complex polymerizes , as ( 1 ) rerunning a gel filtration on the 1:2 complex 24 hours later results in higher order polymers being formed ( unpublished data ) , and ( 2 ) pili are formed readily ( Figure 6A ) . These results confirm the existence of a rate-limiting step in the DSE reaction whereby PapD/ ( PapA ) 2 complex formation appears to be required before DSE can proceed to higher-order polymer forms [23] . The reason for such a rate-limiting step is unclear; it may be that this provides a checkpoint mechanism before committing to full biogenesis of the rod . The PapD/PapAI4G behaves very much like the PapD/PapAwt in the gel filtration , but fewer pili appear to be made ( Figure 6B ) . The single G15N mutation in the DSE region appears to affect polymerization with the detection of 1:3 and 1:4 PapD/PapA complexes ( PapD/ ( PapA ) 3 and PapD/ ( PapA ) 4 ) , and while pilus formation does occur , the diameters of the pili and their central channels are increased ( Figure 6C ) . Thus , these mutations appear to slow down the reaction in such a way that polymer intermediates are observed after 24 h , but overall , these mutations do not seem to affect the process so stringently that it is unable to proceed to completion . More important is the effect of deleting the region preceding the DSE region ( PapD/PapANtd1; Figure 6D ) or deleting the DSE region ( PapD/PapAΔ11–17; Figure 6E ) . Neither of these mutants produces pili . Also , while PapD/PapANtd1 appears to be able to form 1:2 PapD/PapA complexes and aggregates of protein are visible by EM , the PapD/PapAΔ11–17 mutant is totally impaired . Thus , both regions ( the DSE region and the region N-terminal to it ) are important for PapA pilus formation , with the deletion of each of the regions blocking the process at two different stages of PapA polymerization . Combining the DSE ( G15N ) and non-DSE ( I4G ) mutations ( Figure 6F ) , or combining the non-DSE deletion ( Ntd1 ) and the single-site DSE ( G15N ) mutation ( Figure 6G ) lead to results that confirm our conclusion that both regions of the Nte are important for Pap polymerization and pilus formation . For the combined PapAI4G_G15N , large , linear aggregates are found , which are not pili . Higher-order limited polymers ( PapD/ ( PapA ) 2 , PapD/ ( PapA ) 3 , and PapD/ ( PapA ) 4 ) are observed , showing that this double mutant is not totally impaired , while PapANtd1_G15N appears to be severely impaired . In this report , we solved the structures of the PapA subunit , the major subunit of the P pilus , before and after DSE , and , based on these structures , we have examined the roles that the various regions in the Nte play in polymerization and pilus formation . We show that polymerization of wild-type PapA transitions through a 1:2 PapD/PapA complex ( PapD/ ( PapA ) 2 ) , and that a triple alteration combining a deletion of the residues preceding the DSE region of the Nte ( Ntd1 ) , a mutation of a conserved Gly residue in the DSE region ( G15N ) , and a mutation in the P5 pocket ( T101L ) stabilizes the PapD/ ( PapA ) 2 intermediate . The Ntd1 and G15N mutations , individually , do not appear to block the formation of the PapD/ ( PapA ) 2 intermediate complex , nor do they block formation of higher-order complexes ( Figure 6C and 6D ) . However , the combined Ntd1 and G15N mutations severely impair formation of these complexes ( Figure 6G ) . Thus , the T101L mutation appears to attenuate the severity of the combined Ntd1 and G15N mutations and stabilizes the PapD/ ( PapA ) 2 intermediate . This may be because partial filling of the P5 pocket by Leu alters the DSE reaction , resulting in increased PapD/ ( PapA ) 2 formation but abrogating further polymerization events . The crystal structure of the PapD/ ( PapANtd1_G15N_T101L ) 2 suggests that the groove of PapA is longer than the groove of any other Pap subunits of known structure , and that this is why it can accommodate a longer Nte . This led us to suggest that both the DSE region and the region N-terminal to it are important for pilus formation . Site-directed and deletion mutagenesis confirm this view and thus validate earlier published observations by Mu et al . [22] , which emphasized the role of the non-DSE region of the Nte . Thus , PapA uses an extended Nte and in that respect appears to be very similar to other major pilus subunits such as SafA of Salmonella or Caf1 of Yersinia [19 , 21] . The Nte of SafA , for example , is 17 residues long , and one residue outside its DSE region , F3 , was shown to be important in capping the process of DSE and driving the reaction to completion ( the DSE region of SafA consists of residues 11 to 17 ) . Indeed , a mutation of F3 to Ala in SafA results in an equilibrium between reaction species because in this mutant , DSE is allowed to proceed in reverse . A possible equivalent of F3 in PapA is I4 . Indeed , DSE is somewhat affected by the I4G mutation . Thus , some features common to the assembly of all major pilus subunits are emerging , which include the involvement of an extended protein–protein interface and a potential capping mechanism driving polymer formation to completion . The PapA polymer is however different from the SafA or Caf1 polymer in that it adopts a distinct tertiary superhelical structure . The structures presented here do not provide any clues as to how such a ternary structure could form . Indeed , packing interfaces observed in both the PapD/PapANtd1_G15N and PapD/ ( PapANtd1_G15N_T101L ) 2 crystals appear irrelevant ( unpublished data ) . However , the structure of PapA elucidated here provides the basis for complementing the work by Mu et al . [24] and characterizing further the PapA/PapA interactions that preside over superhelix formation .
See Text S1 . The eight PapDHis/PapA constructs ( PapDHis/PapAwt , PapDHis/PapANtd1 , PapDHis/PapAΔ11–17 , PapDHis/PapAG15N , PapDHis/PapAI4G , PapDHis/PapANtd1_G15N , PapDHis/PapAI4G_G15N , and PapDHis/PapANtd1_G15N_T101L ) were transformed one at a time into E . coli C600 cells and grown in a 5-l fermentor vessel containing Terrific Broth ( TB; Sigma , http://www . sigmaaldrich . com ) kept at 37 °C and shaken at 600 rpm . The cells were induced with 1 mM IPTG , once the OD600 reached a value of 0 . 9 , and kept growing for another 3 . 5 h . The complexes were purified after periplasmic extraction using Cobalt-affinity chromatography ( Talon; Clontech , http://www . clontech . com ) , followed by hydrophobic interaction chromatography ( phenyl source; GE Healthcare , http://www . gehealthcare . com ) , ending with a gel filtration step in 20 mM TrisHCl ( pH 7 . 5 ) and 20 mM NaCl , using a Superdex75 120 ml column . This last step was crucial for separating the different polymer forms of PapA in complex with PapD ( PapD/ ( PapA ) 1 , PapD/ ( PapA ) 2 , PapD/ ( PapA ) 3… ) . The PapDHis/ ( PapANtd1_G15N_T101L ) 2 complex was purified as explained above . However after the gel filtration step , there was a second major peak that eluted at a volume of around 54 ml ( the 1:1 complex eluted at around 60 ml ) and was interpreted as a 1:2 complex ( PapD/ ( PapA ) 2 ) . The 1:2 complex was concentrated to 13 mg/ml for crystallization trials . The PapD/PapANtd1_G15N complex was purified from periplasmic extracts using cation-exchange chromatography ( SP HiTrap HP column; GE Healthcare ) followed by hydrophobic interaction chromatography ( phenyl source ) . The purification was completed by a gel filtration step in 20 mM MES ( pH 6 . 0 ) and 20 mM NaCl on a Superdex75 120 ml column . The 1:1 complex eluted at a volume of around 60 ml and was concentrated to 8 mg/ml for crystallization trials . See Text S1 . PapD/PapANtd1_G15N: two crystal forms of the complex were obtained; in both cases PapD/PapANtd1_G15N was crystallized at room temperature in a hanging drop . In the first crystal form ( plates ) , the drop was equilibrated against a reservoir solution containing 25% PEG8K , 10% isopropanol , and 0 . 1 M MES ( pH 6 . 6 ) . In the second crystal form ( rods ) , the drop was equilibrated against a reservoir solution containing 2 M ammonium sulfate and 0 . 1 M Na acetate ( pH 5 . 6 ) . The plates belonged to space group C2 and diffracted to 3 . 2 Å , whereas the rods belonged to space group P3221 with cell dimensions a = 167 Å , b = 167 Å , c = 178 Å , and diffracted to 2 . 6 Å The solvent content is 71% , with 4 PapD/PapANtd1_G15N complexes per asymmetric unit ( Table S2 ) . The data from one single rod-shaped crystal was processed to 2 . 6 Å , and the structure solved by molecular replacement using PapD/PapK as a search model with the program AMoRe [25] . The PapK molecule was first modeled to a poly-alanine chain prior to refinement . The first refinements were performed using simulated annealing and noncrystallographic symmetry restraints for the four complexes in the asymmetric unit , using CNS [26] . Then successive cycles of manual rebuilding with O [27] and conjugate gradient minimization with CNS were performed . B factors were refined individually . Toward the end of the refinement , the noncrystallographic symmetry restraints were only applied to some parts of the β-sheet core of the complexes . The refinement converged to the final R values of R = 23 . 0% and Rfree = 26 . 6% with good stereochemistry . PapDHis/ ( PapANtd1_G15N_T101L ) 2: This complex was crystallized at room temperature in a hanging drop equilibrated against a reservoir solution containing 12% PEG8K , 5% isopropanol , and 0 . 1 M TrisHCl ( pH 7 . 5 ) . These plates belonged to space group C2 , with one complex per asymmetric unit ( 62% solvent ) and were improved by microseeding . The cell dimensions are a = 133 Å , b = 74 Å , c = 80 Å , and β= 109° . The data from one plate was processed to a resolution of 2 . 5 Å . The structure was solved by molecular replacement ( AMoRe ) using the previously solved PapD/PapANtd1_G15N structure as a search model . In the electron density map , there was extra density near the N-terminal extension of the PapA molecule . This density was good enough to enable manual fitting of another PapA molecule next to the first one . Then successive cycles of conjugate gradient minimization with CNS and manual rebuilding with O ( http://xray . bmc . uu . se/alwyn ) enabled rebuilding of some of the second PapA molecule loops that differed from the first one . B factors were refined individually , and no noncrystallographic symmetry restraint was applied between the two PapA molecules . The refinement converged to the final values of R = 22 . 3% and Rfree = 26 . 2% with good stereochemistry ( Table S2 ) . PapD/PapA ( chaperone/pilin ) samples of wild-type and mutant PapA proteins were frozen and thawed five times in liquid nitrogen at 150 μg/ml in 10–20 mM TrisHCl ( pH 7 . 6 ) and 20–50 mM NaCl . The frozen–thawed sample was used either on the same day , or allowed to sit at 4 °C for as long as 40 d to enhance oligomerization of PapA subunits . The sample ( 5 μl ) was placed on carbon-coated , glow-discharged grids , washed with 10 mM TrisHCl ( pH 7 . 6 ) , negatively stained with 1% uranyl acetate , and imaged on a Philips CM12 electron microscope ( no longer available ) .
The Protein Data Bank ( http://www . rcsb . org/pdb ) accession numbers for the coordinates for the structures of the complexes mentioned in this article are PapD/PapANtd1_G15N ( 2uy7 ) , PapDHis/ ( PapANtd1_G15N_T101L ) 2 ( 2uy6 ) , PapE/KNte ( 1N12 ) , and PapD/PapK ( 1PDK ) . | Bacterial adhesion to a host is a crucial step that determines the onset of bacterial infection . It is mediated through recognition of a receptor on the host cell surface by a protein called an adhesin displayed on the surface of the bacterium . Many adhesins are displayed at the tip of specialized organelles called pili , some of which are assembled by the ubiquitous chaperone–usher pathway . In this pathway , each pilus subunit is assisted in folding by a chaperone . The resulting chaperone–subunit complex is targeted to a pore located in the outer membrane , called the usher , that serves as assembly platform . There , pilus subunits dissociate from the chaperone and polymerize , resulting in a surface organelle , the pilus , that protrudes out of the usher . Here , we have determined the structure of the major subunit of the P pilus , PapA . The P pilus , produced in uropathogenic Escherichia coli , displays the adhesin PapG responsible for targeting the bacterium to the kidney epithelium . We have determined the structure of PapA either bound to its cognate chaperone , PapD , or bound to another PapA subunit . These structures provide a view of PapA before and after its assembly in the pilus and shed light on the mechanism of PapA assembly . | [
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] | [
"biophysics",
"infectious",
"diseases",
"eubacteria"
] | 2007 | Crystal Structure of the P Pilus Rod Subunit PapA |
Current knowledge about the dynamics of antigen presentation to T cells during viral infection is very poor despite being of fundamental importance to our understanding of anti-viral immunity . Here we use an advanced mass spectrometry method to simultaneously quantify the presentation of eight vaccinia virus peptide-MHC complexes ( epitopes ) on infected cells and the amounts of their source antigens at multiple times after infection . The results show a startling 1000-fold range in abundance as well as strikingly different kinetics across the epitopes monitored . The tight correlation between onset of protein expression and epitope display for most antigens provides the strongest support to date that antigen presentation is largely linked to translation and not later degradation of antigens . Finally , we show a complete disconnect between the epitope abundance and immunodominance hierarchy of these eight epitopes . This study highlights the complexity of viral antigen presentation by the host and demonstrates the weakness of simple models that assume total protein levels are directly linked to epitope presentation and immunogenicity .
The presentation of virus peptides ( epitopes ) to CD8+ T cells plays a pivotal role in anti-viral immunity . Recognition of these epitopes presented on MHC class I drives CD8+ T cell priming following interactions with professional antigen presenting cells ( APC ) and subsequently allows control of infection through killing of infected cells and secretion of cytokines . The process of MHC class I antigen presentation is complex and multi-staged . It starts with degradation of polypeptides , typically by the proteasome , followed by transport to the ER , loading onto MHC class I and finally egress to the cell surface [1] . Along the way other proteases and chaperones refine the peptides and perform quality control functions on peptide-MHC complexes ( pMHC ) [2] . Surprisingly , despite the large coding capacity and therefore antigenic potential of many viruses , CD8+ T cell responses are often skewed towards a small number of peptides in a phenomenon known as immunodominance [3] . This is exemplified by studies of humans and animals infected with large , complex dsDNA viruses , such as herpes- and poxviruses , where reproducible CD8+ immunodominance hierarchies emerge . For example , up to 20% of the CD8+ T cell response following infection of C57BL/6 mice with vaccinia virus ( VACV ) is directed towards a single immunodominant epitope and a handful of subdominant specificities account for much of the remainder [4] , [5] . Further , while MHC class I antigen presentation is well understood in principle [6] and bioinformatic predictions of MHC class I binding are often highly refined [7] , prediction of antigenicity and immunogenicity have remained elusive . In part this gap remains because kinetic studies to date have focused on single peptides [8] and broader scale studies of antigenicity have been limited to single time points [9]–[11] . This has reflected limitations of technology in that the best reagents for quantifying antigen presentation have been the few monoclonal antibodies generated to date that recognise specific pMHC complexes [8] , [12]–[15] . Proteome-wide biochemical approaches have typically required prohibitively large numbers of cells ( 1×109 and greater ) restricting experiments to single time points [16] , [17] . Although we have good examples showing the diversity of native virus epitopes presented and we know the consequences of manipulating expression levels and even translation rates for presentation of model antigens [8] , [18] , this information remains disconnected . As a consequence , while it is clear that increasing expression of a given antigen leads to higher presentation of epitopes , it is not known whether antigen expression level per se is a useful predictor of likely antigenicity across different viral proteins . Further , whether bulk protein abundance or expression levels correlate best with production of epitopes as a general rule is not known . Indeed , several recent studies have highlighted the diversity of source for MHC class I bound peptides and have implicated both products of translational infidelity ( defective ribosome initiation products ( DRiPs ) ) [10] , [19]–[22] as well as mature proteins [23] . For instance , some biochemical surveys of epitope versus transcript or steady-state antigen abundance suggest these are closely related at single time points [16] , [24] . However , most epitopes studied in detail are shown to be the products of recent translation and therefore need not be related to final antigen abundance [25]–[28] . Only studies that can link the kinetics of antigen synthesis and accumulation with epitope presentation for multiple native virus proteins will allow general conclusions to be drawn . Finally , antigen expression levels can be linked to immunogenicity for model antigens , but again whether this is useful for evaluating whole viral proteomes has not been approached . Here we present the first study that links the kinetics of virus protein build up and CD8+ T cell epitope presentation for multiple pMHC complexes . We used vaccinia virus , best known as the vaccine used to eradicate smallpox , taking advantage of its robust in vitro infections and a well characterised CD8+ T cell epitope hierarchy [4] , [5] . In addition there is good evidence that anti-VACV CD8+ T cells are directly primed by infected APC making this an ideal choice to study antigen presentation in vitro [29]–[31] . The abundance of 8 VACV epitopes was quantified simultaneously at multiple times after infection using the multiple reaction monitoring approach to tandem mass spectrometry [32] . The same method was applied in parallel to determine relative abundance of the relevant virus proteins using filter assisted sample preparation and whole cell tryptic digestion [33] . Together , these data provide an unparalleled insight into the dynamic nature of antigen presentation on class I during a virus replication cycle . Further they provide the most compelling evidence to date of the direct correlation between the timing of virus antigen expression and the appearance of epitopes derived from the same protein . Finally , while we can now add kinetics to our description of epitope presentation for multiple epitopes , these biochemical data still fail to predict the hierarchy of immunodominance in responding CD8+ T cell responses .
Previous studies aimed at understanding antigen presentation kinetics have focussed on single epitopes , most commonly the model peptide SIINFEKL ( presented by H-2Kb ) expressed from recombinant viruses , including VACV . Whilst these experiments have yielded much useful mechanistic insight , it is not clear whether kinetic data generated are representative of virus epitopes in general . To examine this issue , we first recapitulated published data showing the rapid rise of H-2Kb-SIINFEKL complexes on cells infected with a recombinant VACV strain WR-NP-S-GFP [8] , [13] . This virus expresses a chimera in which SIINFEKL is sandwiched between influenza virus nucleoprotein and enhanced green fluorescent protein [8] , [34] . DC2 . 4 cells , a dendritic cell-like line derived from C57BL/6 mice , were infected at a multiplicity of 10 pfu per cell and presentation of Kb-SIINFEKL complexes measured using the mAb 25D1 . 16 and flow cytometry at various times ( Figure 1A ) . Consistent with previous work that typically used L-Kb cells , in DC2 . 4 Kb-SIINFEKL complexes rose rapidly after infection and began to plateau by 6 hours post infection ( hpi ) . To test if the kinetics observed for Kb-SIINFEKL complexes is representative of all VACV epitopes we used polyclonal T cells isolated from infected mice since monoclonal antibodies to VACV epitope-MHC complexes are not available . If all VACV antigen presentation is like Kb-SIINFEKL , the fraction of polyclonal anti-VACV CD8+ T cells that can be stimulated by infected cells should rise over time with a simple , rapid kinetic . If on the other hand , new pMHC complexes first appear on the cell surface at different times after infection , then one might expect a more complicated curve as new populations of T cells are able to be activated once their epitope appears at the cell surface . Thus using DC2 . 4 and the same infection protocol , global VACV epitope presentation was probed up to 12 hpi using splenocytes taken from mice seven days after VACV infection and the percent of CD8+ T cells making IFNγ determined by intracellular cytokine staining ( ICS ) ( Figure 1B ) . In contrast to the simple rise of Kb-SIINFEKL presentation , the increase in number of CD8+ T cells recognising the infected cells was more complex . There were two phases of rising CD8+ T cell activation , one from 2 to 5 hours ( a similar time frame to Kb-SIINFEKL presentation ) followed by second , steeper rise from 5–7 hpi that continued until 12 hpi . While this reveals nothing about the kinetics of individual epitopes , it suggests that the onset of presentation differs across the native VACV epitopes . It is also consistent with published work using mono-specific T cell lines that shows presentation of some VACV epitopes is delayed for some hours after infection [35] . Together these data suggest that monitoring a single epitope does not reveal the true complexity of viral antigen presentation to T cells . We therefore sought to dissect in greater detail the presentation of individual VACV derived epitopes using mass spectrometry ( MS ) . Liquid chromatography coupled to multiple reaction monitoring mass spectrometry ( LC-MRM ) is the method of choice for detection of multiple known peptides [32] , [36] , [37] . LC-MRM MS affords high sensitivity and selectivity and has been recently applied to multiplexed qualitative and quantitative analyses of peptide epitopes eluted from MHC molecules [32] , [37] . For this study , eight VACV epitopes restricted by murine H-2 Kb were chosen based on their well characterised immunogenicity and their expression from a variety of different VACV proteins spanning different temporal phases of the infection ( Table 1 ) [4] , [5] . In addition , SIINFEKL was included in some experiments to allow a direct comparison of this model antigen with the native VACV epitopes . Optimal MRM transition conditions ( precursor ion charge , fragmentation energy and fragment ion selection ) for each VACV epitope listed in Table 1 were determined using synthetic peptides ( Table S1 and Figure S1 in Supporting Information ) . The resulting MRM method allowed for the simultaneous detection of all 8 VACV epitopes ( Figure 2A ) and also included transitions to measure SIINFEKL and isotopically-labelled ( AQUA ) SIIN*FEKL; inclusion of the SIIN*FEKL AQUA peptide was used to control for losses during processing of the MHC-bound peptides as described [32] . The unequivocal detection of peptide epitopes was achieved by several rigorous confirmatory steps in this LC-MRM workflow: firstly , RP-HPLC retention across multiple dimensions of purification ( correct eluting fraction during off-line RP-HPLC and correct on-line retention time during LC-MRM MS ) must be consistent with that measured for the synthetic version of each of the VACV peptides ( Figure S2 ) ; secondly , they must trigger all MRM transitions concurrently and in the correct transition hierarchy; and , as a final step , each peptide sequence must be further confirmed by an MRM-triggered MS/MS sequencing scan – a modality unique to the quadrupole linear ion trap mass spectrometer used in this study [38] . In order to verify the sample workflow ( Figure 2B ) , DC2 . 4 cells were incubated with a pooled mixture of the full set of 8 synthetic peptides representing VACV epitopes ( Table 1 ) . Following extensive washing to remove unbound peptides , cells were pelleted and snap-frozen and subjected to immunoaffinity purification of H-2Kb complexes , peptide elution and chromatographic separation as previously described [32] , [37] . The presence of each VACV epitope in the MHC eluate was confirmed by LC-MRM ( Figure 2C ) . The differing detection intensities across the peptide set reflects a combination of the varying ionisation efficiencies of the peptides and competition for binding to the Kb molecules during incubation . Next , MHC elution and LC-MRM were used for the detection of SIINFEKL and native VACV epitopes generated through VACV infection with the recombinant WR-NP-S-GFP . DC2 . 4 cells ( 1×108 ) were infected for 6 hours with WR-NP-S-GFP to compare the levels of SIINFEKL presentation with that of the 8 native VACV epitopes ( Figure 3 ) . Capture of Kb-peptide complexes was achieved as above , including the addition of 50 fmol of isotopically-labelled AQUA SIIN*FEKL in order to control for sample preparation losses post affinity purification of the MHC-peptide complexes [32] . The quantification of each VACV epitope was achieved by comparing the area under the MRM curve to that of 100 fmol of the corresponding synthetic epitope analysed separately ( Figure 2A ) . LC-MRM confirmed the detection of SIINFEKL and all 8 VACV peptides ( Figure 3A shows representative data for SIINFEKL , B820–27 and J3289–296 ) . Further it provides the first definitive evidence that the amino acid length and constitution of the VACV epitopes is exactly as described in the original mapping studies [4] , [5] . SIINFEKL presentation on WR-NP-S-GFP-infected cells at 6 hpi was calculated to be 2 . 3×104 and 3 . 1×104 copies per cell for two independent experiments ( Figure 3B ) . All 8 Kb-restricted VACV epitopes were detected at considerably lower estimated abundances to that of SIINFEKL . Further , abundance of the 8 VACV peptides varied over a wide range with 3 epitopes ( B820–27; A47138–146 and J3289–296 ) being presented at levels up to 1000-fold higher than the remaining 5 VACV epitopes . When compared to CD8+ T cell response elicited in mice infected for 7 days by the same virus , there is a striking dissociation between the epitope abundance and T cell immunodominance hierarchies ( Figure 3B ) . Next we sought to assess the presentation kinetics of the 8 VACV epitopes during the course of infection . This was done using non-recombinant VACV , to avoid any potential competing effects from the very high levels of presentation of SIINFEKL following infection with the recombinant WR-NP-S-GFP VACV strain . DC2 . 4 cells were infected for 0 . 5 , 3 . 5 , 6 . 5 , 9 . 5 and 12 . 5 hours , or mock infected as a negative control and epitope abundance at each time determined by LC-MRM analysis . All 8 VACV epitopes were detected and the kinetics of their presentation measured ( Figure 4A ) . Six of 8 peptides were detected by 0 . 5 hpi , with the remaining 2 epitopes ( A3270–277 and A1947–55 ) undetectable until 6 hours later . Peak expression occurred at 3 . 5 hpi for 5 epitopes , 6 . 5 hpi for two epitopes and at the final time point of 12 . 5 hours for a single epitope . We noted that the presentation of the immunodominant B820–27 epitope was unusual in that its onset was at 30 minutes , but instead of peaking at 3 . 5 hpi , like most of this group of epitopes , its peak was later at 6 . 5 hpi . The abundance profile spanned 3 logs , ranging from as low as an estimated 11 copies per cell for C4125–132 to as high as 32 , 400 copies of A47138 . These basic features of presentation with some epitopes showing peak presentation around 3 . 5 hours after infection , while others only appear at 6 . 5 hours have also been observed for cells infected with the MVA strain of VACV ( our unpublished observations ) . Thus abundance and kinetics of presentation are highly variable across different epitopes and robust presentation early after infection is not always maintained . In order to assess how the kinetics of epitope presentation correlates with source antigen expression , a sample of the cell lysate from each infection time point was subjected to reduction , alkylation and subsequent digestion with the enzyme trypsin prior to proteomic analysis . Proteotypic tryptic fragments from each of the 8 VACV protein antigens were chosen using Skyline [39] ( Table S2 and Figures S3 and S4 ) . Following initial screening of samples , 6 of the 8 VACV proteins were detected ( for A3 and J3 , multiple tryptic fragments were found to be amenable to MRM analysis and so all were included ) . Despite rigorous testing of multiple peptides , no positive signal could be detected for proteins L2 and C4 so these were not included further . In order to achieve normalisation of protein loading across the timecourse , 12 murine tryptic peptides ( corresponding to eight host proteins; Table S3 and Figure S3 ) were simultaneously analysed in the same LC-MRM method ( Figure S3 ) . These murine proteins were chosen as suitable candidates for normalisation based on the high copy number and long half life of their human homologues [40] , with the notion that such proteins will not be grossly affected by the VACV-mediated shutdown of host protein synthesis . In addition , a good correlation between the abundance of these representative proteins and cell number recovered post-infection was found suggesting that they were appropriate for normalisation ( Figure 3C ) . The uncorrected data is also shown in Figure S4 for comparison . MRM peaks at each time point for the 6 VACV proteins were used to determine relative protein expression over the course of infection and these were plotted alongside the relative levels of each epitope derived from the same protein ( Figure 4B ) . This approach allows relative expression of individual antigens to be determined at different time points but does not provide absolute quantitation of the antigen and therefore direct comparison between antigens is more qualitative . Expression profiles of the 6 proteins were consistent with their temporal expression cluster as reported by analyses of transcription and more recently defined promoters [41]–[43] , which gives further confidence of the method . Translation , as determined by tryptic peptide detection , was detected at 0 . 5 hpi for A47 , A8 and B8 , corresponding with the appearance of epitopes derived from those proteins . Whilst levels of A47 peaked at 6 . 5 hpi , all other proteins peaked ( at least within the limits of this time course ) at 12 . 5 hours . Proteins A3 and A19 , both of which are classified as late , were detected by 3 . 5 hours , but did not reach substantial levels until 6 . 5 hours and onwards; presentation of epitopes A3270–277 and A1947–55 tracked closely with the increase in protein levels . For epitopes A47138–145 , A8189–196 and J3125–132 , rapid and peak presentation following protein expression was followed by a sharp decline in epitope levels to almost zero by 12 . 5 hpi . However , epitopes B820–27 and A3270–277 , although decreasing following peak levels mid-infection , maintained a more constant level around 20–40% of the maximum; for A1947–55 , epitopes levels did not peak until the end of the time course , following an almost identical profile to A19 protein expression . Of note the B820–27 epitope appeared to display a lag between peak of protein expression and peak of epitope presentation . Next , in vitro protein and epitope presentation kinetics were correlated with CD8+ T cell immunodominance in vivo . C57BL/6 mice were infected with VACV WR by the intraperitoneal route ( i . p . ) and 7 days after infection , the percentage of CD8+ T cells responding to ex vivo stimulation with each peptide was determined by intracellular staining for IFNγ ( Figure 4C ) . This method of epitope detection has recently been shown to have a linear range that covers responses to all the epitopes investigated here [44] . As previously reported [4] , [5] , B820–27 dominated the response , A1947–55 was the weakest and the remaining 6 epitopes formed an intermediate hierarchy . Here , where the onset , peak level and longevity of epitope display were revealed ( as opposed to the single time point for the WR-NP-S-GFP in Figure 3 ) , there was still no obvious correlation between presentation and the CD8+ T cell dominance hierarchy . Although the immunodominant B820–27 was one of the most robust epitopes in peak and persistence of presentation , it is similar in this respect to the subdominant A47138–146 and J3289–296 . Further , A3270–227 and A8189–196 , which are the next 2 peptides in the dominance hierarchy after B820–27 , have very different presentation profiles with the former only appearing later ( 6 . 5 hpi ) and having better persistence but a substantially lower ( approximately 10-fold ) peak than the latter . | A major mechanism for the detection of virus infection is the recognition by T cells of short peptide fragments ( epitopes ) derived from the degradation of intracellular proteins presented at the cell surface in a complex with class I MHC . Whilst the mechanics of antigen degradation and the loading of peptides onto MHC are now well understood , the kinetics of epitope presentation have only been studied for individual model antigens . We addressed this issue by studying vaccinia virus , best known as the smallpox vaccine , using advanced mass spectrometry . Precise and simultaneous quantification of multiple peptide-MHC complexes showed that the surface of infected cells provides a surprisingly dynamic landscape from the point of view of anti-viral T cells . Further , concurrent measurement of virus protein levels demonstrated that in most cases , peak presentation of epitopes occurs at the same time or precedes the time of maximum protein build up . Finally , we found a complete disconnect between the abundance of epitopes on infected cells and the size of the responding T cell populations . These data provide new insights into how virus infected cells are seen by T cells , which is crucial to our understanding of anti-viral immunity and development of vaccines . | [
"Abstract",
"Introduction",
"Results"
] | [
"sequencing",
"biochemistry",
"antigen",
"processing",
"and",
"recognition",
"immunity",
"peptide",
"mapping",
"immunity",
"to",
"infections",
"immunology",
"biology",
"proteomics",
"immune",
"response"
] | 2013 | Kinetics of Antigen Expression and Epitope Presentation during Virus Infection |
Cell regulatory circuits integrate diverse , and sometimes conflicting , environmental cues to generate appropriate , condition-dependent responses . Here , we elucidate the components and mechanisms driving a protein-directed RNA switch in the 3′UTR of vascular endothelial growth factor ( VEGF ) -A . We describe a novel HILDA ( hypoxia-inducible hnRNP L–DRBP76–hnRNP A2/B1 ) complex that coordinates a three-element RNA switch , enabling VEGFA mRNA translation during combined hypoxia and inflammation . In addition to binding the CA-rich element ( CARE ) , heterogeneous nuclear ribonucleoprotein ( hnRNP ) L regulates switch assembly and function . hnRNP L undergoes two previously unrecognized , condition-dependent posttranslational modifications: IFN-γ induces prolyl hydroxylation and von Hippel-Lindau ( VHL ) -mediated proteasomal degradation , whereas hypoxia stimulates hnRNP L phosphorylation at Tyr359 , inducing binding to hnRNP A2/B1 , which stabilizes the protein . Also , phospho-hnRNP L recruits DRBP76 ( double-stranded RNA binding protein 76 ) to the 3′UTR , where it binds an adjacent AU-rich stem-loop ( AUSL ) element , “flipping” the RNA switch by disrupting the GAIT ( interferon-gamma-activated inhibitor of translation ) element , preventing GAIT complex binding , and driving robust VEGFA mRNA translation . The signal-dependent , HILDA complex coordinates the function of a trio of neighboring RNA elements , thereby regulating translation of VEGFA and potentially other mRNA targets . The VEGFA RNA switch might function to ensure appropriate angiogenesis and tissue oxygenation during conflicting signals from combined inflammation and hypoxia . We propose the VEGFA RNA switch as an archetype for signal-activated , protein-directed , multi-element RNA switches that regulate posttranscriptional gene expression in complex environments .
Mammalian cells integrate diverse , and sometimes conflicting , environmental signals to generate appropriate , condition-dependent responses . Tissue myeloid cells are exposed to a plethora of stimulatory and inhibitory signals , and thus its integrated response is particularly complex . This task is made more problematical , and possibly more critical , in dynamic , pathological environments . Myeloid cell vascular endothelial growth factor ( VEGF ) -A is critical for blood vessel formation during development , wound-healing , and tumorigenesis [1] . Hypoxia is possibly the most potent agonist of VEGF-A expression , working at the levels of transcription , mRNA stabilization , and translation [2] , [3] . VEGF-A synthesis is induced in monocyte/macrophages activated by pro-inflammatory agonists , including interferon ( IFN ) -γ and bacterial lipopolysaccharide . Overproduction of VEGF-A can cause excessive neovascularization , blood vessel permeability , and enhanced leukocyte recruitment , all hallmarks of chronic inflammatory conditions , including cancer and atherosclerosis [4]–[6] . Agents that inhibit VEGF-A or its receptor have been applied clinically to successfully limit colorectal and renal cell carcinoma [7] . Positive and negative regulation of VEGF-A expression has been reported in human macrophages in multiple stressed conditions . We have shown that VEGF-A expression in myeloid cells is translationally repressed by the IFN-γ-triggered GAIT ( interferon-gamma-activated inhibitor of translation ) system [8] , [9] . Importantly , under certain pathological conditions , for example within the avascular cores of tumors and in the thickened intima of atherosclerotic lesions , macrophages are simultaneously exposed to both inflammatory cytokines and hypoxia that act concurrently in multiple pathophysiological scenarios to regulate gene expression . Treatment of human monocytic cells with IFN-γ induces the synthesis of VEGFA mRNA and protein for up to about 12 to 16 h . However , VEGF-A synthesis and secretion are suppressed about 16 h after IFN-γ treatment despite the presence of abundant VEGFA mRNA [10] . Translational silencing of VEGFA and other GAIT targets requires binding of the GAIT complex to its cognate GAIT element in the target mRNA 3′UTR [10] . The GAIT element is a defined 29-nt stem-loop with an internal bulge and unique sequence and structural features . The human GAIT complex is heterotetrameric containing glutamyl-prolyl-tRNA synthetase ( EPRS ) , ribosomal protein L13a , NS1-associated protein–1 , and glyceraldehyde 3-phosphate dehydrogenase ( GAPDH ) [11] , [12] . A C-terminus truncated form of EPRS , termed EPRSN1 , functions as a dominant-negative regulator of GAIT complex activity and maintains basal expression of VEGF-A [13] . RNA-binding proteins ( RBPs ) that regulate mRNA stability or translation generally recognize their target mRNAs through structural or sequence-specific elements in the 5′ or 3′UTRs of mature mRNAs . The activity of trans-acting RBPs can be modulated by dosage ( in turn regulated by synthesis rate and stability ) , cellular localization , posttranslational modification , noncoding RNAs , and interacting protein partners . Heteronuclear ribonucleoprotein ( hnRNP ) L is a key posttranscriptional regulator of VEGF-A expression . Human hnRNP L has three consensus RNA recognition motifs ( RRM ) [14] and binds CA-rich elements ( CARE ) in coding and noncoding regions of multiple transcripts [15] . hnRNP L contributes to pre-mRNA splicing [16] , mRNA nucleocytoplasmic transport [14] , internal ribosomal entry site-mediated translation [17] , translational repression [18] , and mRNA stabilization [19] . The molecular mechanisms by which signal transduction systems integrate multiple environmental cues into a binary response that determines gene expression remain largely unexplored . We have reported that hnRNP L operates a hypoxia-stimulated , binary conformational RNA switch that overrides IFN-γ-induced GAIT-mediated translational silencing of VEGFA mRNA in human monocytic U937 cells and in primary human peripheral blood monocytes ( PBMs ) [20] . The proposed switch permits high-level VEGF-A expression under combined inflammatory and hypoxic stress . Here we elucidate the molecular mechanism underlying the IFN-γ- and hypoxia-dependent regulatory RNA switch . The switching mechanism involves condition-dependent posttranslational modification and relocalization of hnRNP L , and subsequent formation of an hnRNP L-containing heterotrimeric complex that stabilizes the VEGFA HSR in a translation-competent conformation .
HnRNP L is an essential component of the RNA switch that blocks GAIT-mediated translational silencing of VEGF-A mRNA , and permits high-level expression of VEGF-A in myeloid cells in the presence of IFN-γ and hypoxia ( Figure S1 ) [20] . To determine whether hnRNP L is sufficient for RNA switch function , the activity of recombinant protein was determined by in vitro translation of luciferase reporter bearing the VEGFA HSR in a wheat germ extract system in the presence of active GAIT complex from IFN-γ-treated U937 cells ( Figure 1A ) . hnRNP L failed to overcome the translational repression suggesting that posttranslational modification of hnRNP L or additional protein factors may be required . Identical results were seen using a rabbit reticulocyte lysate system ( not shown ) . Hypoxia-dependent hnRNP L binding partners were determined by RNA affinity purification using a 30-nt , 5′-biotinylated , extended CARE ( CARE-E ) from the VEGFA HSR ( Figure 1B ) . To reduce nonspecific binding , lysates from U937 cells incubated under normoxic or hypoxic conditions were pre-cleared with an excess of 5′-biotinylated antisense CARE-E RNA in which CA pairs were mutated to GU . Cleared lysates were incubated with biotinylated , wild-type CARE-E RNA and μMAC magnetic streptavidin microbeads , and applied to a magnetic column . Bound proteins were eluted with salt solution , concentrated , and subjected to SDS-PAGE and Coomassie stain ( Figure 1C ) . Bands enriched in lysates from hypoxia-treated cells were subjected to mass spectrometric analysis , and peptides corresponding to hnRNP L , hnRNP A2/B1 , and DRBP76 ( nuclear factor 90 or interleukin enhancer binding factor 3 ) were identified ( Table S1 ) . Binding of the proteins to CARE RNA was confirmed by RNA affinity isolation and immunoblot analysis of lysates from hypoxia-treated U937 cells . A hypoxia-inducible complex of hnRNP L , DRBP76 , and hnRNP A2/B1 ( HILDA ) was shown to bind wild-type but not mutant antisense CARE RNA; substantially less binding of the three proteins to CARE RNA was observed in normoxic lysates ( Figure 1D ) . The formation of an RNA-binding heterotrimeric complex was investigated by co-immunoprecipitation ( IP ) . Lysates from U937 cells and primary human PBM treated with IFN-γ under normoxic or hypoxic conditions were subjected to IP with anti-hnRNP L antibody , and probed with hnRNP A2/B1- and DRBP76-specific antibodies ( Figure 1E , left panel ) . A hypoxia-dependent interaction of hnRNP L with hnRNP A2/B1 and DRBP76 was observed . The interaction between hnRNP L and hnRNP A2/B1 was RNA-independent as shown by the lack of an effect of RNase A treatment . However , the RNase diminished the interaction between hnRNP L and DRBP76 , suggesting that the hnRNP L-DRBP76 complex is stabilized by RNA . The expression levels of the three HILDA complex constituents were not altered by hypoxia exposure ( Figure 1E , right panel ) . In vitro GST-pulldown experiments showed that recombinant GST-hnRNP L directly interacted with recombinant hnRNP A2/B1 and DRBP76 ( Figure 1F , left panel ) . In a parallel experiment , GST-hnRNP A2/B1 was found to directly bind hnRNP L but not DRBP76 ( Figure 1F , right panel ) . hnRNP L contains an N-terminal glycine-rich domain , three RNA-binding motifs ( RRM1–3 ) , and a proline-rich linker domain connecting RRM2 and RRM3 ( Figure 1G , top ) . Domain mapping experiments revealed that hnRNP A2/B1 binds the proline-rich linker in hnRNP L ( Figure 1G , left ) . In contrast , the RRM3-containing , C-terminal domain of hnRNP L was the binding site for DRBP76 ( Figure 1G , right ) . EPRS and hnRNP L from IFN-γ-treated U937 cells , in either normoxia or hypoxia , bind in vitro synthesized VEGF-A HSR in a mutually exclusive manner [20] . To provide in vivo evidence of the VEGF-A switch , RNA from cells treated with IFN-γ in the presence of normoxia or hypoxia for 24 h were immunoprecipitated with anti-EPRS and -hnRNP L antibodies and subjected to qRT-PCR using transcript-specific primers . GAIT complex EPRS and HILDA complex hnRNP L recognized and bound VEGFA mRNA following stimulation by IFN-γ under normoxic and hypoxic conditions , respectively , consistent with previous results ( Figure 2A ) [20] . To determine whether hnRNP A2/B1 or DRBP76 are required for hnRNP L binding to VEGFA mRNA , lysates from cells treated with IFN-γ and hypoxia were subjected to ribonucleoprotein IP ( RIP ) using anti-hnRNP L antibody , coupled with RT-PCR . hnRNP L interacted with VEGFA mRNA in control transfected cells; however , the interaction was substantially reduced following siRNA-mediated depletion of either hnRNP A2/B1 or DRBP76 ( Figure 2B ) . Similarly , the interaction of hnRNP A2/B1 or DRBP76 with VEGFA mRNA required the presence of the other ( Figure S2 , left and center panels ) . Moreover , the interaction of hnRNP A2/B1 and DRBP76 with VEGFA mRNA was abolished following hnRNP L depletion by siRNA-mediated gene silencing ( Figure S2 , right panels ) , suggesting that HILDA binding to VEGFA mRNA requires integrity of the entire complex . To begin to understand the roles of the individual protein components in RNA switch activity , their binding sites within the HSR region were mapped by UV-crosslinking . Of the three proteins , only hnRNP L and DRBP76 directly bind the VEGFA HSR . Interestingly , the two interacting proteins bind different regions of the HSR , hnRNP L binds the CARE , whereas DRBP76 binds the AU-rich stem loop ( AUSL ) ( Figure 2C ) . The less robust binding to the individual ascending ( AUSL-A ) and descending ( AUSL-D ) regions of the AUSL suggests that DRBP76 stabilizes the double-stranded AUSL in a conformation that prevents formation of the GAIT element , which overlaps AUSL-A ( Figure 1B ) . We determined the specific DRBP76-binding region by constructing a series of mutations in either AUSL strand . Mutation of M2 ( U404UAUAU409 to AAUAUA ) , but not M1 ( A416AUAUA421 to UUAUAU ) , inactivated the RNA switch of the HSR-bearing reporter RNA , suggesting the upper stem-loop region of the AUSL is critical ( Figure 2D and Figure S3A , B ) . Differences in luciferase activities of the mutant forms were due largely to altered translation as shown by comparable firefly luciferase mRNA levels determined by semi-quantitative RT-PCR ( Figure 2D , insert ) ; renilla luciferase mRNA levels were essentially the same for all transfections ( not shown ) . Complementary covariant mutations ( M2–M3 , A381UAUAA386 to UAUAUU ) on the M3 strand opposing M2 were introduced in an attempt to restore function . However , the M2–M3 double mutant failed to recover RNA switch activity , possibly due to disruption of the GAIT element structure by M2 mutation . Thus , we further created complementary mutations of U358UAUAU363 to AAUAUA ( M4 ) to restore the GAIT element structure at the distal 6-bp stem region . RNA switch activity was partially restored in the M2–M3–M4 triple mutant , indicating the stem structure , not the sequence , is critical for DRBP76 activity in the RNA switch . As controls , individual M3 and M4 mutants lacked GAIT-mediated translational silencing activity and RNA switch function . In the VEGFA HSR , the CARE adjoins the GAIT element with not even a single nt separating them ( Figure S3A ) [20] . To determine the maximum distance between the elements that permits RNA switch activity , we inserted 5- to 25-nt poly ( C ) spacers between them in an HSR-bearing reporter . Spacers up to 15 nt permitted RNA switch activity , but 20- and 25-nt spacers were inhibitory ( Figure 2E and Figure S3A , C ) , consistent with a distance limit for an effective interaction between the binding proteins hnRNP L and DRBP76 . The insertions did not affect mRNA expression of FLuc ( Figure 2E , insert ) and RLuc ( not shown ) significantly , indicating that altered translation was responsible for differential Luc activity . Together these results suggest that whereas hnRNP L is responsible for target selectivity , DRBP76 , through binding a nearby stem-loop region , has primary responsibility for stabilizing the RNA form lacking the GAIT structural element , thereby suppressing GAIT complex-directed translational silencing ( Figure 2F ) . Knockdown of DRBP76 did not significantly alter VEGFA mRNA half-life , providing additional evidence that DRBP76 influences VEGF-A expression primarily at the level of translation ( Figure S3D ) . By knockdown and overexpression experiments , we previously reported that hnRNP L is essential for hypoxia-induced switch activity in U937 cells [20] . To test the requirement for the other HILDA components , DRBP76 and hnRNP A2/B1 , both were subjected to siRNA-mediated knock-down ( hnRNP L knock-down served as positive control ) ( Figure 3A , top ) . Cells were treated with IFN-γ and hypoxia for up to 24 h , and lysates tested for their effect on in vitro translation of an HSR-bearing reporter . As seen before , 24-h lysates from IFN-γ-treated normoxic cells inhibited translation of the reporter , but 24-h lysates from hypoxic cells were inactive ( Figure 3A , bottom ) . However , deletion of either DRBP76 or hnRNP A2/B1 dramatically impaired the hypoxia-driven RNA switch to an extent comparable to that of hnRNP L knockdown , and permitted GAIT complex-mediated translation inhibition by 24-h lysates ( Figure 3A , bottom ) . We investigated the effect of these lysates on endogenous gene expression . As before , hypoxia prevented IFN-γ-mediated inhibition of expression of VEGF-A observed at 24 h ( Figure 3B ) . However , siRNA-mediated knock-down of either DRBP76 or hnRNP A2/B1 restored translational inhibition of VEGF-A without significantly altering the steady-state level of VEGFA mRNA ( Figure 3B ) . Polysome profiling was done to verify that the effects on VEGF-A expression were due to altered translation . IFN-γ activation of the GAIT pathway inhibited VEGF-A mRNA translation-initiation [21] , and this inhibition was reversed by hypoxia [20] . Indeed , following IFN-γ treatment under hypoxia , knock-down of either hnRNP A2/B1 or DRBP76 induced a dramatic shift of endogenous VEGFA mRNA from translationally active polysome pools to translationally inactive free mRNP pools ( Figure 3C and Figure S4 ) . hnRNP L expression is markedly reduced in normoxic , IFN-γ-treated cells , thereby permitting GAIT complex binding to the VEGFA mRNA and transcript-specific translational silencing [20] . Semiquantitative RT-PCR ( Figure 4A ) and Northern blot analysis ( Figure S5 ) showed that hnRNP L mRNA expression is unaltered by either hypoxia or IFN-γ treatment for up to 24 h , and that altered hnRNP L expression must be posttranscriptional . hnRNP L half-life was measured in the presence of cycloheximide to inhibit protein synthesis . In nonstressed monocytic cells ( normoxia , no IFN-γ ) the half-life of hnRNP L is about 12 h ( Figure 4B and Figure S6A ) . The half-life of hnRNP L was shortened to about 4 h by IFN-γ treatment in normoxia; however , hypoxia suppressed the effect of IFN-γ , restoring the half-life to about 12 h ( Figure 4C and Figure S6B ) . As shown previously , the proteasome inhibitor MG132 blocked IFN-γ-mediated hnRNP L degradation , indicating an important role of the ubiquitin/proteasome pathway in regulating hnRNP L expression [20] . To investigate the mechanism underlying IFN-γ-induced hnRNP L degradation , hnRNP L ubiquitination was determined . IFN-γ treatment in the presence of MG132 induced accumulation of a high molecular weight form of hnRNP L consistent with ubiquitination ( Figure 4D ) . Expression of HA-ubiquitin and detection with anti-HA-tag antibody confirmed formation of high molecular weight , ubiquitinated hnRNP L , and exposure to hypoxia dramatically diminished hnRNP L ubiquitination ( Figure 4E ) . We considered the von Hippel-Lindau ( VHL ) -containing ubiquitin ligase complex as a candidate E3 ubiquitin-protein ligase because of its normoxia-dependent role in regulation . VHL specifically targets proteins , e . g . , hypoxia inducible factor ( HIF ) -1α tagged by O2-dependent prolyl hydroxylation [22] . VHL was shown to interact robustly with hnRNP L , but not with hnRNP A2/B1 or DRBP76 , in an IFN-γ-dependent manner ( Figure 4F ) . Also , siRNA-mediated knockdown of VHL markedly reduced hnRNP L polyubiquitination ( Figure 4G , left panel ) with MG132 treatment , and increased hnRNP L stability following IFN-γ treatment in absence of MG132 ( Figure 4G , right panel ) . However , overexpression of VHL did not affect the stability of hnRNP L or the assembly of the HILDA complex in hypoxia , suggesting that HILDA complex formation might contribute to protection of hnRNP L from VHL-mediated degradation ( Figure S7 ) . In an in vitro ubiquitination system reconstituted with exogenous E1 and E2 enzymes and E3 ubiquitin ligase pVHL derived from lysate of 8 h , IFN-γ-treated U937 cells in normoxia further confirmed robust polyubiquitination of hnRNP L ( Figure S8 ) . In contrast , cell lysate from hypoxia-treated U937 cells failed to modify hnRNP L . Similar results were obtained with primary human PBM ( not shown ) . These results suggest that proteasomal degradation of hnRNP L in U937 cells and in human PBM is mediated by IFN-γ-triggered ubiquitination by a VHL-containing E3 ubiquitin ligase . hnRNP L is primarily localized in the nucleus in human monocytic cells but substantially redistributes to the cytoplasm during hypoxia [23] . Fluorescence visualization verified hypoxia-driven cytoplasmic relocalization of hnRNP L , even in the presence of IFN-γ ( Figure 5A ) . Similar hypoxia-stimulated cytoplasmic relocalization of hnRNP L was observed in primary human PBM-derived macrophages induced by macrophage colony stimulating factor ( M-CSF ) ( Figure S9 ) . Immunoblot analysis of cytosolic and nuclear fractions from IFN-γ- and hypoxia-treated cells further confirmed hnRNP L translocation ( Figure 5B ) . Cellular localization of RBPs can be regulated by their phosphorylation state [24]–[26] . Metabolic labeling with 32P-orthophosphate showed that hypoxia induced robust phosphorylation of hnRNP L at 8 h , and the modification was stable for at least 24 h ( Figure 5C ) . Immunoblot analysis of hnRNP L immunoprecipitated from hypoxia-treated cells with phospho-specific antibodies revealed strong phosphorylation at Tyr , but not at Ser or Thr ( Figure 5D ) . A time course experiment showed modest hnRNP L Tyr-phosphorylation after 0 . 5 h of hypoxia and maximal phosphorylation after 4 h in U937 cells ( Figure 5E ) and in primary human PBM ( not shown ) . Immunoblot analysis with anti-pTyr antibody showed Tyr-phosphorylated hnRNP L was almost completely restricted to the cytoplasm in hypoxia-treated cells ( Figure 5F ) . To identify the hypoxia-induced phosphorylation site , hnRNP L was immunoprecipitated from lysates of hypoxia-treated cells , and phospho-sites detected by mass spectrometry . Total coverage with three protease treatments was 84% , but phosphorylation events were not detected ( Figure S10 ) . Endogenous hnRNP L in U937 cells was knocked down with siRNA targeting the 3′UTR , and cells transfected with cDNA constructs containing specific , site-directed Tyr-to-Ala mutations at residues in regions not covered by the mass spectrometry analysis . Among the five hnRNP L mutants tested , only Y359A was not phosphorylated in U937 monocytic cells ( Figure 5G ) and in human PBM ( not shown ) . Tyr359 , and the surrounding sequence , is evolutionarily conserved from frogs to humans ( Figure 5H ) , and has been identified as a phospho-site by high-throughput proteomic survey ( www . phosphosite . org ) in both mouse and human ( in addition to Tyr phosphorylation at positions 47 , 48 , 92 , 267 , 285 , 333 , 340 , 363 , 375 , 565 , 574 , and 576 ) . To determine the role of Tyr359 phosphorylation in hnRNP L localization , cells were transfected with c-Myc-tagged wild-type cDNA or , phospho-dead ( Y359A ) or phospho-mimetic ( Y359D ) mutants . Under normoxic conditions , wild-type hnRNP L is primarily localized in the nucleus , but also present in the cytoplasm , as observed previously [23] . In contrast , the Y359A mutant was exclusively in the nucleus , and the Y359D mutant was exclusively cytoplasmic ( Figure 5I ) . Similarly , following IFN-γ stimulation under hypoxia , the Y359A and Y359F hnRNP L mutants were exclusively localized in the nucleus ( Figure S11 ) . As a control for specificity , Tyr130 mutants did not partition with the Tyr359 mutants . Cells were transfected with c-Myc-tagged wild-type or mutant hnRNP L , immunoprecipitated with anti-c-Myc antibody , and probed with hnRNP A2/B1 antibody . Y359D exhibited much greater binding to hnRNP A2/B1 compared to wild-type or Y359A mutant hnRNP L ( Figure 5J ) . Remarkably , the Y359D mutant , but not the Y359A mutant or wild-type protein , was completely resistant to IFN-γ-stimulated degradation ( Figure 5K ) . Consistent with the cellular translocation of hnRNP L ( Figure 5A ) , Tyr phosphorylation was induced by IFN-γ treatment in hypoxia ( Figure S12 ) . In summary , hypoxia-inducible Tyr359 phosphorylation of hnRNP L facilitates its cytoplasmic relocalization and prevents its degradation . Because hnRNP A2/B1 does not bind the HSR directly , it is more likely involved in regulation of its binding partner hnRNP L , than in operating the RNA switch itself . We tested the possibility that hnRNP A2/B1 contributes to hypoxia-induced stabilization of hnRNP L . siRNA-mediated knockdown of hnRNP A2/B1 resulted in hnRNP L destabilization following IFN-γ treatment in hypoxia ( Figure 6A ) . In contrast , hnRNP A2/B1 knockdown did not induce DRBP76 degradation ( Figure 6B ) . Also , siRNA-mediated knockdown of DRBP76 did not affect hnRNP L stability ( Figure S13 ) . Interestingly , hnRNP L was subject to IFN-γ-dependent Pro hydroxylation as shown by IP followed by probing with anti-hydroxyproline antibody ( Figure 6C ) . Hypoxia prevented the IFN-γ-inducible prolyl hydroxylation of hnRNP L ( Figure 6D ) . Knockdown of hnRNP A2/B1 under hypoxic condition and in the presence of IFN-γ and MG132 restored marked Pro hydroxylation of hnRNP L after 24 h ( Figure 6E ) . Finally , co-IP with anti-hnRNP L antibody revealed that hypoxia induced hnRNP A2/B1 binding to hnRNP L , and completely blocked hnRNP L recognition by VHL ( Figure 6F ) . These results indicate that the major function of hnRNP A2/B1 in the heterotrimeric switch is to protect hnRNP L from IFN-γ-triggered prolyl hydroxylation , ubiquitination , and subsequent degradation . Treatment of U937 cells with prolyl hydroxylase ( PH ) inhibitors L-mimosine and dimethyloxalylglycine ( DMOG ) blocked prolyl hydroxylation of hnRNP L and caused marked stabilization of the protein in the presence of IFN-γ under normoxia ( Figure S14 ) . Co-IP and RNA-binding studies suggest a model in which the interaction between DRBP76 and hnRNP A2/B1 is indirect and facilitated by hnRNP L and VEGFA HSR RNA ( Figure 2F ) . We investigated the interactions in detail by in vitro reconstitution using recombinant proteins and in vitro–transcribed RNA . DRBP76 and hnRNP A2/B1 by themselves did not bind , nor did the addition of either hnRNP L or HSR RNA restore their interaction significantly ( Figure S15 ) . However , when both hnRNP L and HSR RNA were added , then a modest interaction between hnRNP A2/B1 and DRBP76 was detected . A much stronger interaction was observed when phospho-mimetic hnRNP L ( Y359D ) was added together with HSR RNA , but not nonspecific RNA , thereby reconstituting the entire HILDA complex in vitro . To investigate the sufficiency of hnRNP L , hnRNP A2/B1 , and DRBP76 in operating the RNA switch , we determined the regulatory activity of the purified proteins in vitro . Phospho-mimetic hnRNP L ( Y359D ) was used to facilitate interaction with hnRNP A2/B1 . The three proteins were pre-incubated in several combinations , and their effect on in vitro translation of an FLuc reporter bearing the VEGFA HSR element ( and RLuc control RNA ) was determined in a wheat germ extract in the presence of 35S-Met and cytosolic extracts from IFN-γ-treated U937 cells . hnRNP L ( Y359D ) by itself or with either hnRNP A2/B1 or DRBP76 , did not restore translation in the presence of lysates from cells treated with IFN-γ for 24 h ( Figure 6G ) . However , the three proteins together substantially overcame the translational inhibition . Substitution of wild-type hnRNP L for the phospho-mimetic was ineffective , suggesting the posttranslational modification is not only required for maintaining a high level of cytoplasmic hnRNP L , but also is required for HILDA complex assembly . As positive controls , lysates from cells treated for 24 h with or without IFN-γ in hypoxia could rescue translation of HSR-bearing FLuc . These results support the role of the heterotrimeric HILDA complex in operating the RNA conformational switch .
The combinatorial activity of pairs of nearby elements has become an area of increasing interest , particularly with the recent recognition that microRNA binding to targets can influence protein binding to nearby target RNA elements [27] . There are few cases in which pairs of protein-binding RNA elements dictate the response . In one well-studied example , a combinatorial code in which the number and position of two elements—namely , the cytoplasmic polyadenylation element and Pumilio-binding element—determine translational activation or repression in Xenopus oocytes [28] . However , there is a dearth of studies on the mechanisms by which nearby RNA elements , and their cognate binding factors , integrate disparate environmental signals to generate a binary response and regulate gene expression . In one known case , the leader sequence of the Mg2+ transporter gene mgtA of Salmonella enterica contains a Mg2+-sensing riboswitch and an 18-codon , proline- or hyperosmotic stress-sensing ORF that integrate distinct signals to generate the cell response; however , an interaction between the disparate elements was not observed [29] . In the case of the GAIT system , we have reported that hypoxia prevents GAIT complex binding to the VEGFA 3′UTR by a switch in the conformation of RNA that masks the GAIT structural element [20] by converting the element into the ascending half of a long , double-stranded stem-loop . The switch is initiated by hypoxia-stimulated binding of hnRNP L to a 3′UTR CARE directly adjacent to the GAIT element . In this report we define the components of a heterotrimeric complex that constitutes the RNA switch , their regulation by IFN-γ and hypoxia , and their specific functions in directing the VEGFA mRNA switch in human monocytic cells . The requirement for each of the components of the HILDA complex to drive the RNA switch was shown by knockdown experiments in cells , and their sufficiency shown by in vitro reconstitution . The HILDA complex has not been previously described , but its individual components are known to regulate distinct mRNA-related functions . DRBP76 was initially identified through its binding to double-stranded RNA and to protein kinase R ( PKR ) [30] . DRBP76 exhibits multiple RNA-related functions including regulation of transcription , mRNA stability [31] , and translation [32] . DRBP76 also binds the VEGFA HSR in hypoxic breast cancer cells , increasing mRNA stability and translation , but the binding region within the VEGFA HSR in these experiments was not determined [33] . The double-stranded RNA-binding property of DRBP76 is most likely the critical function it performs in the context of the HILDA complex , stabilizing the conformation featuring a long , double-stranded stem loop , and disrupting the structure of the GAIT element . hnRNP A2/B1 , like hnRNP L , participates in splicing of pre-mRNAs and in translational regulation [34] . hnRNP A2/B1 also serves as a molecular motor-powered transporter of select mRNAs bearing specific hnRNP A2/B1 response elements ( A2RE ) , for example , neurogranin , Arc , and calmodulin-dependent kinase II [35]–[37] . Cytosolic complexes containing heterodimeric hnRNPs have been shown to interact with specific target mRNAs . For example , hnRNP L and I form a complex that binds murine inducible nitric oxide synthase mRNA , and regulates its translation [38] . Interestingly , the same pair of hnRNPs found in the HILDA complex , hnRNP L and A2/B1 , interacts with the glucose transporter 1 ( Glut1 ) 3′UTR , inducing translational repression and mRNA instability [18] . However , an interaction between DRBP76 and A2/B1 has not been described . hnRNP L is a critical component of the HILDA complex because it is uniquely responsible for stimulus sensing as well as target recognition . Our results show that the steady-state level and cellular localization of hnRNP L in myeloid cells are regulated both by IFN-γ and by hypoxia . Under normoxic conditions hnRNP L is distributed between the cytoplasm and nucleus , the latter for execution of mRNA processing functions . IFN-γ induces prolyl hydroxylation of cytoplasmic hnRNP L and consequent rapid , VHL-mediated ubiquitination and proteasomal degradation ( Figure 7 ) . Near-complete cytoplasmic depletion of hnRNP L permits GAIT complex binding to the VEGFA GAIT element in the translationally silent conformer , resulting in low-level translation of VEGFA mRNA . Hypoxia induces phosphorylation of hnRNP L on Tyr359 , which increases cytoplasmic localization by restricting transport into the nucleus . Hypoxia-inducible phosphorylation suggests the activity of a nonreceptor Tyr kinase such as a member of the Src , Abl , Jak , Syk , or Fak families . The sequence surrounding the Tyr359 phosphorylation site ( pRRGPSR359YGPQYGHPPPPPPPP ) exhibits 100% conservation in humans , rodents , rabbits , and frogs , and provides insight into the identity of the proximal kinase . “YG” is a specific Src kinase substrate motif ( PhosphoMotif Finder ) , and the downstream polyproline motif is a binding site for SH3-containing proteins , including Src family kinases . hnRNP A2/B1 binds Tyr359-phosphorylated hnRNP L and blocks recognition by VHL-containing E3 ubiquitin ligase complex , thus permitting cytoplasmic accumulation . The precise kinetics and binding order have not been determined , but our results suggest that the phospho-hnRNP L and hnRNP A2/B1 recruit DRBP76 to form the heterotrimeric HILDA complex that binds the VEGFA CARE . The interaction is weakened by nuclease treatment , indicating that the binding of DRBP76 to other complex members is enhanced by its interaction with the long , AU-rich stem-loop within the VEGFA HSR . The HILDA complex stabilizes the translationally permissive conformer that masks the GAIT element , thus resulting in uninhibited translation of VEGFA mRNA , even in the presence of IFN-γ-induced GAIT complex . The tumor suppressor protein VHL is an essential , target-specific component of a multifunctional E3 ubiquitin ligase complex involved in protein degradation [39] . The best-known target of VHL is hypoxia inducible factor ( HIF ) -1α and -2α , transcription factors that stimulate expression of multiple hypoxia-inducible transcripts , including VEGFA mRNA . In normoxia , O2-dependent prolyl hydroxylation of HIF-1α triggers recognition by VHL and consequent degradation , thereby inhibiting expression of HIF-1α targets [40] . However , prolyl hydroxylation of HIF-1α is inhibited in hypoxia , thereby stabilizing HIF-1α and increasing target mRNA transcription . Other VHL targets have been identified in renal cell carcinoma cell lines; interestingly , several are downregulated by VHL [41]–[43] . hnRNP A2/B1 has been reported to be targeted by VHL [44] . However , we find hnRNP A2/B1 binding to hnRNP L prevents targeting by VHL in human monocytic cells . Possibly , cell-type specificity of targets and directionality of regulation—i . e . , up or down—are promoted by additional factors within the VHL-bearing E3-ubiquitin ligase complex . Proline hydroxylase inhibitors DMOG and L-mimosine both block hnRNP L prolyl hydroxylation and consequent degradation . Collagen prolyl-4-hydroxylase ( C-P4H ) is a candidate because it is induced by hypoxia [45] , [46] and hydroxylates and destabilizes another RBP , Argonaute 2 ( Ago2 ) [47] . Likewise , HIF prolyl hydroxylase ( HIF-PH ) is a candidate because it modifies HIF-1α for poly-ubiquitination by pVHL and proteasomal degradation [48] . Long , noncoding regions of mRNAs , because of their manifold protein- and RNA-binding elements , are potentially ideal for integration of multiple inputs into a single output—i . e . , gene expression . Because of their unusually long length , the 3′UTR , which averages almost 600 nt in human mRNAs versus about 150 nt for 5′UTRs , is a particularly attractive target for signal integration [49] . A plethora of examples of posttranscriptional regulation have been described in which RBPs are activated by environmental signals that alter their binding behavior , generally by posttranslational modification and complex formation [50] . In most known cases , RBPs or complexes interact one-to-one with preformed sequence or structural elements [50] , [51] . More recently , regulatory processes have been described in which signals alter the conformation of the RNA to modulate gene expression [52] . The VEGFA 3′UTR RNA switch features alternative interaction of distinct protein complexes in response to environmental signals , culminating in regulated gene expression . The CARE element is analagous to a riboswitch aptamer domain , and hnRNP L acts as a “responder/selector , ” responding to environmental cues and determining HILDA complex mRNA target specificity . The AUSL element determines the expression outcome: VEGF-A expression is high when the double-stranded conformation is bound by the HILDA complex , and expression is depressed when the GAIT complex binds the GAIT element in the alternate conformation ( Figure 7 ) . To our knowledge there are not any previous reports of 3-RNA element switches . Likewise , the integration of two different signals—i . e . , hypoxia and inflammatory cytokine—by the VEGFA RNA switch lacks precedent . The principles , protein constituents , and mechanisms utilized by the VEGFA switch might be applicable to distinct mRNA switches . One possibility is that the HILDA complex recognizes other transcripts with sequence and structural elements analogous to the VEGFA switch region—i . e . , CARE and GAIT elements nearby DRBP76-binding double-stranded RNA stretches . Cytoplasmic hnRNP L binds VEGF-A mRNA and other transcripts in multiple cell lines [18] , [19] , [38] , suggesting that the HILDA complex might direct additional RNA switches . More generally , distinct RBPs may replace hnRNP L as the “specificity factor , ” but likewise recruit DRBP76 to stabilize nearby stem-loop structures and drive formation of alternate regulatory conformers . High-throughput screening has identified at least two RBPs hnRNP A1 and FUS ( fused in sarcoma ) that bind DRBP76 and might direct alternate RNA switches [53] , [54] . Alternatively , other inhibitory factors ( microRNA or proteins ) might replace the GAIT complex to drive the hnRNP L-directed GAIT-independent RNA switches in more general sense . We speculate that the VEGFA switch is a founding member of signal-activated , protein-directed , RNA switches that regulate posttranscriptional gene expression in vertebrates , and similar switches might be widespread RNA sensors in multicellular animals .
Phospho-safe extraction buffer was from Novagen ( Madison , WI ) . Rabbit reticulocyte lysate , wheat germ extract , large-scale RNA production system-T7 , and dual luciferase reporter assay system were from Promega ( Madison , WI ) . Human IFN-γ was obtained from R&D Systems ( Minneapolis , MN ) . Human monocyte nucleofactor kit was from Lonza ( Switzerland ) . Reagents for protein purification , nuclear and cytosolic extraction , and immunoanalysis were from Pierce ( Rockford , IL ) . Primers , dNTP mix , TRIzol LS reagent , one-step RT-PCR system , and competent cells were from Invitrogen ( Carlsbad , CA ) . Protein A/G beads , anti-α-tubulin , anti-hnRNP A2/B1 , rabbit anti-hnRNP L , and anti-GAPDH antibodies were from Santa Cruz ( Santa Cruz , CA ) . Mouse monoclonal anti-hnRNP L antibody was from Novus ( Littleton , CO ) . Anti-HDAC1 and anti-β-actin antibodies were from Biovision ( Mountain View , CA ) . Anti-c-Myc , anti-HA , goat anti-rabbit/mouse IgG ( Alexa Fluor® 488 Conjugate ) , streptavidin-HRP , and anti-ubiquitin antibodies were from Cell Signalling Technology ( Danvers , MA ) . Anti-DRBP76 antibody was from Biorbyt ( Cambridge , UK ) . GST monoclonal antibody was from Thermo Scientific ( West Palm Beach , FL ) . Anti-VHL antibody was from GeneTex ( San Antonio , TX ) . Anti-hydroxyproline antibody was from Abcam ( Cambridge , MA ) . Anti-rabbit IgG , anti-mouse IgG , and random-primer labeling kit were from GE healthcare ( UK ) . Translation grade [35S]methionine was from NEN-Dupont ( Boston , MA ) , α-[32P]CTP was from PerkinElmer ( Boston , MA ) , and [32P]orthophosphoric acid was from MP Biomedicals ( Solon , OH ) . Actinomycin-D , DMOG , and L-Mimosine were from Sigma ( St . Louis , MO ) . In vitro ubiquitination assay kit and ubiquitin were from Biomol ( Plymouth Meeting , PA ) and Boston Biochem ( Cambridge , MA ) , respectively . Human U937 monocytic cells ( ATCC , Rockville , MD ) were cultured in RPMI 1640 medium containing 10% heat-inactivated fetal bovine serum ( FBS ) , 2 mM glutamine , and 100 U/ml of penicillin and streptomycin at 37°C and 5% CO2 . PBM from healthy clinical donors were isolated by leukapheresis and countercurrent centrifugal elutriation under a Cleveland Clinic Institutional Review Board–approved protocol that adhered to American Association of Blood Bank guidelines . For preparation of cytosolic extracts , the cells were incubated for 1 h in medium containing 0 . 5% FBS and then with ( or without ) IFN-γ ( 500 units/ml ) in presence of hypoxia ( 1% O2 ) for an additional 8 or 24 h . Cell lysates were prepared in Phosphosafe extraction buffer containing protease inhibitor cocktail . To knock down endogenous hnRNP L , DRBP76 , hnRNP A2/B1 , or VHL , U937 cells were transfected with appropriate concentration of ( 100–200 nM ) gene-specific siRNA or a scrambled control siRNA using human monocyte nucleofactor kit . hnRNP L siRNAs containing 3 oligomers targeting the 3′UTR or ORF were from Origene . siRNA against DRBP76 , hnRNP A2/B1 , and VHL were from Santa Cruz . The bacterial expression plasmid pRSET-hnRNP L was generated using pcDNA3-hnRNPL-c-Myc as template and cloned between BamHI and EcoRI restriction sites in the pRSET-A vector for expression and purification of His-tagged hnRNP L . HNRNPL ORF was subcloned into pGEX-4T-1 vector and the plasmid transformed into E . coli BL21 ( DE3 ) for expression and purification of GST-tagged hnRNP L . hnRNP L cDNA was subcloned into pcDNA3-c-Myc between BamHI and EcoRI restriction sites and expressed in human U937 cells as described [20] . The pcDNA3-based hnRNP L Tyr-to-Ala , -Asp , and -Phe mutants were prepared using GENEART Site-Directed Mutagenesis System ( Invitrogen ) according to the manufacturer's instructions . The mutation was confirmed by DNA sequencing . DRBP76 ORF was cloned into pET28-a vector between NdeI and EcoRI restriction sites . Expression of GST-tagged proteins was induced with 500 nM isopropyl-β-D-thiogalactopyranoside ( IPTG ) at 30°C for 6 h with 50 µg/ml ampicillin . Soluble protein was extracted and purified with B-PER GST purification kit ( Thermo Fisher ) . His-tagged DRBP76 was generated in vitro using rabbit reticulocyte lysate in vitro translation system ( Promega ) , and purified with MagneHis Protein Purification System ( Promega ) . His-tagged wild-type hnRNP L and phospho-mimetic hnRNP L were expressed in E . coli BL21 ( DE3 ) with IPTG induction and in rabbit reticulocyte lysate in vitro translation system , respectively , and purified with Ni-NTA resin ( Qiagen ) . Recombinant GST-hnRNP A2/B1 and hnRNP A2/B1 were from Novus Biologicals and Origene , respectively . S100 extracts ( 4 mg ) from U937 cells cultured in normoxia or hypoxia were pre-cleared by incubation for 30 min at 4°C with 2 µg 5-biotinylated , mutant antisense CARE-E RNA oligomer ( 5′-biotin-UCUGUGUGGGUGGGUGUAUGUAUGUAAAUA-3′ ) , added to 200 µl of μMACs magnetic streptavidin microbeads for 10 min , and applied to μMACS separator . The cleared lysate was incubated with 2 µg of 5′-biotinylated , wild-type CARE-E RNA oligomer ( 5′-biotin-AGACACACCCACCCACAUACAUACAUUUAU-3′ ) , and then with streptavidin microbeads and applied to μMACS separator as above . The column was rinsed with 100 µl protein equilibration buffer and twice with 100 µl of lysis buffer . The bound material was applied to the column and washed 4 times with 100 µl of lysis buffer to decrease nonspecific binding . 200 µl of buffer containing 300 mM NaCl was applied to the column to elute bound protein . The eluate was desalted and concentrated using Centrifugal Filter Unit ( Microcon YM-3K , Millipore , Billerica , MA ) . Eluates were subjected to SDS-PAGE and Coomassie stain . Bands enriched only in hypoxia-treated sample were trypsinized and peptides mapped by capillary column LC-tandem MS ( LTQ-linear ion trap MS system , ThermoFinnigan , San Jose , CA ) . The data were analyzed with Mascot using CID spectra to search the human reference sequence database . Matching spectra were verified by manual interpretation aided by additional searches using the Sequest and Blast . Most IP experiments were done with Co-Immunoprecipitation kit ( Pierce ) following the manufacturer's instruction to eliminate antibody contamination of IP products . For some IP experiments , traditional method was used . Cells were lysed in Phospho-safe extraction buffer , and 500 µl of cell lysate was combined with 50 µl protein A/G agarose beads ( 50% bead slurry ) and pre-cleared at 4°C for 60 min . The samples were centrifuged at 13 , 000 rpm for 10 min at 4°C and the supernatant added to 50 µl of protein A/G beads and 2 µg of antibody , and rotated for 4 h at 4°C . The beads were washed 5 times with 1 ml cold lysis buffer . Protein gel loading dye ( 100 µl ) was added , and the samples boiled and loaded onto the gel . To avoid interference from IgG , rabbit-derived secondary antibody was used against mouse-derived primary antibody . GST and GST-hnRNP L were generated from E . coli BL21 ( DE3 ) transformants containing pGEX-4T-1 and pGEX-4T-1-hnRNP L , respectively . Cells were sonicated and the supernatant collected after high-speed centrifugation . GST and GST-hnRNP L ( 1 µg of each ) were incubated separately with glutathione-agarose beads for 30 min . After washing the agarose beads 4 times with 1 ml of PBS , 1 µg of recombinant DRBP76 and hnRNP A2/B1 were diluted in binding buffer ( 20 mM HEPES , pH 7 . 5 , 200 mM KCl , 5 mM MgCl2 , 0 . 2% bovine serum albumin , 10% glycerol , 0 . 1% Nonidet P-40 , 1 mM phenylmethylsulfonyl fluoride , and complete protease inhibitor mixture ) , combined , and incubated at 4°C for 2 h . The agarose beads were washed 5 times with binding buffer ( without bovine serum albumin and glycerol ) , and bound protein eluted by boiling in SDS loading buffer . Cycloheximide ( 50 µg/ml ) was added to 8×106 U937 cells in 4 ml RPMI1640 medium . Cells were harvested and lysed . Immunoblot was done using anti-hnRNP L antibody and the band intensity quantified and normalized by the initial value at 0-h time point . In vitro reconstitution of hnRNP L ubiquitination was performed as described [55] . Purified His-tagged hnRNP L ( 0 . 5 µg ) was preincubated with U937 cell lysate , and then incubated with a mixture of E1 and E2 enzymes , biotin-ubiquitin , and cell lysate as a source of hnRNP L E3 ligase . Recombinant hnRNP L was immunoprecipitated with anti-His tag antibody , and biotin-ubiquitin was detected by blotting with streptavidin-HRP . The metabolic labeling assay was performed as described previously [12] . U937 cells ( 8×106 cells ) in 4 ml RPMI 1640 medium were collected by centrifugation , re-suspended in phosphate-free medium , and metabolically labeled with a 4-h pulse of 32P-orthophosphate . The cells were collected by centrifugation and lysed with Phospho-safe extraction buffer containing protease inhibitor cocktail . hnRNP L was immunoprecipitated from lysates using mouse anti-hnRNP L antibody and protein A/G-agarose in cell lysis buffer . Proteins were resolved by 12% SDS-PAGE , and the gel was dried and applied to Phospho-screen for determination of radiolabeling . In vitro transcribed , 32P-labeled full-length HSR or truncated HSR RNA ( 20 fmol ) was incubated for 30 min at 4°C with purified recombinant proteins ( 0 . 2 µg ) in 20 µl of buffer containing 20 mM HEPES ( pH 7 . 5 ) , 5 mM MgCl2 , 50 mM KCl , 1 mM DTT , protease inhibitor cocktail , 0 . 1% Triton X-100 , 0 . 1 mg/ml yeast total tRNA , 40 U RNasin , and 10% glycerol . The mixture was crosslinked by 15 min exposure to ultraviolet light ( 1 , 800 J/cm2 ) on ice in a UV crosslinker . The protein-RNA complex was incubated with 1 µl of RNase A for 20 min at 25°C . Samples were denatured in SDS-PAGE buffer under reducing conditions , and complexes analyzed by 10% SDS-PAGE and autoradiography . The RIP assay was performed as described previously [13] . Protein A/G beads ( 50 µl ) were incubated with 500 µl of cell lysate ( 4 mg protein ) for 1 h at 4°C with rotation to pre-clear . The cell lysate was centrifuged and the supernatant collected . Mouse anti-hnRNP L antibody ( 2 µg ) was added ( mouse pre-immune IgG was used as negative control ) and the mixture incubated at 4°C overnight with rotation . Protein A/G beads ( 50 µl ) were added and incubated at 4°C for 4 h . The beads were washed five times with 1 ml of lysis buffer with rotation at 4°C . Total immunoprecipitated RNA was extracted with Trizol . Total RNA from the lysate was extracted and used as a positive control for RT-PCR . Immunoprecipitated RNA ( 3 µl ) and 1 µg of total RNA were used in reverse transcriptase reaction and subsequent PCR with Taq DNA polymerase . The PCR reaction ( 5 out of 20 µl ) was visualized by 1 . 5% agarose gel . The primers for semi-quantitative RT-PCR were as follows: RT_βactin-f: 5′-ATGGATGATGATATCGCCGCG-3′; RT_βactin-r: 5′-CTAGAAGCATTTGCGGTGGAC-3′; RT_VEGF-f: 5′-ACAGAACGATCGATACAGAA-3′; RT_VEGF-r: 5′-AAAGATCATGCCAGAGTCTC-3′; RT_hnRNPL-f: 5′-GAGTCCCATCTGAGCAGGAA-3′; and RT_hnRNPL-r: 5′-CAATTTTATTGAAATGTGCC-3′ . Polysome profiling was done as described [13] . CHX ( 100 µg/ml ) was added to cells for 15 min and then collected and washed two times with CHX-containing , ice-cold PBS . 107 cells were suspended in 350 µl TMK lysis buffer and incubated on ice for 5 min . The lysates were centrifuged at 12 , 000 rpm for 10 min and the supernatants collected . RNase inhibitor ( 2 µl , 40 U/µl ) and CHX ( 50 µl , 100 µg/µl ) were added in 50 ml each of freshly prepared 10% and 50% sucrose gradient solutions just before use . Cytosolic lysates were loaded on the sucrose gradient and centrifuged at 29 , 000 rpm for 4 h , and 8 fractions of about 1 ml were collected and combined; light RNP , 40S , 60S , and 80S formed the translationally inactive pool , and heavy polysome fractions formed the translationally active pool . Total RNA was isolated from both combined fractions by extraction with Trizol reagent and purified by RNeasy minikit ( Qiagen , Valencia , CA ) following the manufacturer's procedure . The RNA was quantitated and purity determined by agarose formaldehyde gel , and used for real-time PCR analyses . Capped , poly ( A ) -tailed template mRNAs was prepared using mMESSAGE mMACHINE SP6 and T7 kits ( Ambion ) . Firefly-Luc-VEGFA GAIT element-poly ( A ) ( 200 ng ) and Renilla-Luc ( 200 ng ) reporter RNAs were incubated with U937 cytosolic lysates ( 500 ng of protein ) from IFN-γ-treated U937 cells in the presence of 35 µl of wheat germ extract or rabbit reticulocyte lysate , and [35S]methionine . The translation reactions were performed for 90 min at 30°C and resolved by SDS-PAGE ( 10% polyacrylamide ) and visualized by phosphorimaging . In some experiments , the FLuc and RLuc activity was measured by chemiluminescence using luminator . U937 cells were transiently transfected with 5 µg of wild-type or mutant pCD-FLuc-VEGFA HSR using human monocyte nucleofactor kit . RLuc-expressing vector pRL-SV40 ( 1 µg ) was co-transfected for normalizing transfection efficiency . After 12 h , transfected cells were incubated with IFN-γ under Nmx . or Hpx . for up to 24 h , lysed , and lysate luciferase activities were measured using a dual luciferase assay kit ( Promega ) . The primers for semiquantitative RT-PCR of FLuc were as follows: RT_FLuc-f: 5′-GCCTGAAGTCTCTGATTAAGT-3′; RT_FLuc-r: 5′-ACACCTGCGTCGAAGT-3′; RT-RLuc-f: 5′-TGATTCAGAAAAACATGCAG-3′; RT-RLuc-r: 5′-ATATTTGTAATGATCAAGTA-3′ . Immunostaining of hnRNP L was as described [23] . U937 cells ( 106 cells/ml ) in 12-well plates with glass cover slip at the bottom were incubated in hypoxia or normoxia for 24 h . Cells were centrifuged for 5 min at 2 , 500 rpm and washed twice with PBS and then with 4% paraformaldehyde fixing solution for 20 min . Cells were washed twice with PBS , and incubated with rabbit anti-hnRNP L polyclonal antibody ( Santa Cruz , 1∶40 ) in blocking solution ( 2% BSA , 0 . 1% Triton X100 in PBS ) at room temperature for 2 h . Cells were washed twice with PBS and centrifuged at 1 , 500 rpm for 5 min . Alexa Fluor 488 goat anti-rabbit secondary antibody ( Invitrogen ) was added ( 1∶50 ) with phalloidin ( 1∶50 ) in blocking solution for 1 h . Cells were washed with PBS three times . DAPI dye was mixed in the mounting solution and the slides imaged . | Many cells of our body , particularly cells such as monocyte/macrophages involved in host immunity , are exposed to diverse and constantly changing environments . These cells require mechanisms by which they can rapidly respond to multiple , sometimes conflicting , environmental cues to generate appropriate responses . The 3′ untranslated regions ( UTRs ) , i . e . the noncoding tail of messenger RNAs , often contain multiple protein- and RNA-binding elements , thereby making it an ideal setting for receiving multiple such environmental cues , which can then be integrated into a single response that regulates the gene's expression . Monocytic cells exposed to inflammation and hypoxia produce vascular endothelial growth factor ( VEGF ) -A , a critical factor in blood vessel formation . VEGF-A expression is regulated under these conditions via a complex regulatory mechanism that involves its 3′UTR . Here we show how this regulatory switch works . Inflammation induces formation of a four-protein complex that binds an RNA element present in the VEGFA 3′UTR and blocks translation . Hypoxia , however , triggers the assembly of a completely different three-protein complex ( termed “HILDA” ) that coordinates the function of three neighboring RNA elements to “flip” the RNA conformation in such a way that prevents the first complex from binding , thereby allowing VEGF-A expression . We hypothesize that the VEGFA switch might function to ensure appropriate angiogenesis and tissue oxygenation when cells are exposed to conflicting signals from combined inflammation and hypoxia conditions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"biology"
] | 2013 | The HILDA Complex Coordinates a Conditional Switch in the 3′-Untranslated Region of the VEGFA mRNA |
One of the hurdles to understanding the role of viral quasispecies in RNA virus cross-species transmission ( CST ) events is the need to analyze a densely sampled outbreak using deep sequencing in order to measure the amount of mutation occurring on a small time scale . In 2009 , the California Department of Public Health reported a dramatic increase ( 350 ) in the number of gray foxes infected with a rabies virus variant for which striped skunks serve as a reservoir host in Humboldt County . To better understand the evolution of rabies , deep-sequencing was applied to 40 unpassaged rabies virus samples from the Humboldt outbreak . For each sample , approximately 11 kb of the 12 kb genome was amplified and sequenced using the Illumina platform . Average coverage was 17 , 448 and this allowed characterization of the rabies virus population present in each sample at unprecedented depths . Phylogenetic analysis of the consensus sequence data demonstrated that samples clustered according to date ( 1995 vs . 2009 ) and geographic location ( northern vs . southern ) . A single amino acid change in the G protein distinguished a subset of northern foxes from a haplotype present in both foxes and skunks , suggesting this mutation may have played a role in the observed increased transmission among foxes in this region . Deep-sequencing data indicated that many genetic changes associated with the CST event occurred prior to 2009 since several nonsynonymous mutations that were present in the consensus sequences of skunk and fox rabies samples obtained from 20032010 were present at the sub-consensus level ( as rare variants in the viral population ) in skunk and fox samples from 1995 . These results suggest that analysis of rare variants within a viral population may yield clues to ancestral genomes and identify rare variants that have the potential to be selected for if environment conditions change .
Rabies virus ( RABV ) is one of the most deadly pathogens known and is able to infect a wide variety of mammalian hosts . RABV is present on all continents except for Antarctica and has reservoirs in terrestrial species as well as bats ( Chiroptera ) . Although vaccination and antibody therapy is effective in treating known exposures to RABV , an estimated 55 , 000 human deaths occur annually mostly in developing countries [1] . RABV is a member of the Lyssavirus genus , family Rhabdoviridae . The genome is composed of negative-sense single-stranded RNA , about 12 kb in size which codes for five proteins- nucleoprotein ( N ) , phosphoprotein ( P ) , matrix protein ( M ) , glycoprotein ( G ) and polymerase ( L ) . Like other RNA viruses , RABV has a high mutation rate due to the high error rate of the polymerase , thus populations of RABV exist as a mutant swarm , or quasispecies [2] . RABV evolution is believed to be driven predominantly by purifying selection and RABV is not known to recombine [3]–[6] . ;Different RABV variants are associated with different reservoir hosts and geographical locations . Typically , interspecies transmission of rabies virus from reservoir to non-reservoir host produces a single fatal spillover event secondary transmission has rarely been observed [7] . For example , a bat variant may infect and cause disease in skunks , but it does not transmit efficiently within the skunk population and skunks would be considered a dead-end host for this variant . The exception to this would be the case of cross-species transmission ( CST ) where the variant from one species adapts to transmission by a new species [8] . For example , in 2001 , bat variant rabies adapted to transmission within the skunk population in Flagstaff Arizona [9] , and in 2009 this variant adapted to transmission by foxes [7] . These events demonstrate the capacity of rabies virus for CST which may lead to increased exposure of humans to the pathogen and increase the geographical range of the virus . Greater than 90 of North American rabies cases occur in wildlife [7] , [9] , and striped skunks ( Mephitis mephitis ) serve as the most frequent source of terrestrial rabies cases in California [10] . Rabies in striped skunks was first documented in California in 1899 and skunk rabies has been considered enzootic since the 1950s [11] . The Northern Pacific coast region ( which includes Humboldt Co . ) is unusual in that this is the only region of CA where large numbers of gray foxes ( Urocyon cinereoargenteus ) are known to be infected with the skunk rabies variant [11] . In 2009 , the number of rabid foxes in Humboldt County infected with the CA skunk variant increased 356 from an average of 12 per year in the preceding 15 years to 7 in the latter months of 2008 to 38 in 2009 ( Annual Reports from California Department of Public Health , Veterinary Public Health Section ) . In 2009 , only 2 skunks were reported rabid in Humboldt County suggesting that rabies infections in foxes had fundamentally shifted from a typical pattern of spillover from skunks to foxes to one resulting from fox-to-fox transmission . The reported numbers underestimate the extent of the outbreak since additional foxes exhibiting unusual or aggressive behavior were euthanized but not tested ( S . Chandler , USDA , personal communication ) . This epizootic of rabies in Northern California raised concerns not only because the primary species involved was gray foxes ( Urocyon cinereoargenteus ) and not striped skunks which are the terrestrial reservoir species in this region , but also because this led to a significant spike in attacks by rabid animals on humans and their pets [12] . The apparent sustained fox-to-fox transmission in this outbreak suggests that CST occurred and enabled this epizootic . We hypothesized that molecular changes in the viral genome would be associated with this event . While phylogenetic data support that rabies viruses have jumped species boundaries historically [5] , it is rare and has never been subject to comprehensive genetic analysis at the intra-host population level . To test our hypothesis and better understand the evolution of rabies , we applied deep-sequencing to 44 unpassaged rabies virus samples from the Humboldt epizootic . Sequence data were generated by two different platforms ( Illumina and 454 ) and by three different commercial services to determine reproducibility . For 40 of the samples , approximately 11 kb of the 12 kb genome was amplified and sequenced using the Illumina platform ( the remaining 4 samples were sequenced using the 454 platform only ) . Average coverage was 17 , 448 and this allowed characterization of the rabies virus population present in each sample at unprecedented depths .
The tissue samples used in this study were obtained from the archived collection of California Department of Public Health , Viral and Rickettsial Disease Laboratory ( CDPH-VRDL ) . Gray foxes ( Urocyon cinereoargenteus ) and striped skunks ( Mephitis mephitis ) displaying symptoms of rabies were submitted for rabies testing in the Humboldt Co . Public Health Laboratory between March 2009 and January 2010 . Brain tissue samples that were laboratory confirmed to be infected with rabies virus were forwarded to CDPH-VRDL for genetic characterization . Other earlier skunk and fox tissue samples from Humboldt Co . were also available from the CDPH-VRDL archives . As part of routine rabies surveillance in California , the VRDL genotypes rabies-positive samples received from local public health laboratories by RT-PCR and performs sequence analysis on RT-PCR products using forward primer 1066 deg ′′5-GARAGAAGATTCTTCAGRGA-3 and reverse primer 304 targeting a portion of the nucleoprotein ( N ) gene as described in Trimarchi and Smith ( 2002 ) and Velasco-Villa , et al . ( 2006 ) [13]–[15] . Approximately 1 gram of brain tissue from foxes and skunks infected with the California skunk rabies virus variant were placed in TRIzol LS Reagent ( Invitrogen , Carlsbad , CA ) and sent to LLNL for further analysis . RNA was extracted from the tissue sample following the manufacturers protocol . Approximately 11 kb of the 12 kb rabies virus genome was amplified using degenerate primers ( Table S1 ) . Primers were designed to be as sensitive to target strain variants as possible , while still being specific enough to not cross-react with non-targets . Sensitivity was achieved by targeting regions of high sequence similarity , identified through a Multiple Sequence Alignment ( MSA ) of the target sequences . Specificity was achieved by targeting regions that do not appear to be similar to any other organisms , determined by searching a database of known genome sequences . Primer candidates were selected based on the combined results of the MSA and sequence searches . This technique is a modified version of the approach outlined in Slezak et . al . [16] , which accommodates degenerate primer design for diverse target genomes , and places a lower relative priority on primer uniqueness as compared to other known genomes . For rabies virus , which lacks perfect primer-length conservation around the genomic regions of interest , it was necessary to identify degenerate primers for many non-conserved primer regions . From the identified primer candidate regions , which included both perfectly conserved regions and degenerate regions , individual primer pairs were selected which provided overlapping coverage of the DNA being sequenced . Final checks were performed which helped avoid hybridization problems such as primer dimerization . Reverse transcription was performed using random hexamers and the Superscript III RT reverse transcriptase kit ( Invitrogen ) . The rabies virus cDNA templates were amplified using the Phusion polymerase kit ( New England BioLabs , Ipswich , MA ) , following manufacturers instructions . PCR conditions consisted of 98C for 30 s , followed by 40 cycles of 98C for 15 s , 64C for 20 s , and 72C for 1 . 2 min . The final cycle was 72C for 10 min . A plasmid control was generated to determine the error rate of the PCR and sequencing steps as described previously [17] . PCR products were prepared for sequencing using the QIAquick PCR Purification kit ( Qiagen , Valencia , CA ) . Sequencing of an aliquot of a subset of 40 samples was carried out by Eureka Genomics , Hercules , CA using an Illumina Genome Analyzer IIx . Another aliquot of the same samples plus an additional 4 samples were set for 454 sequencing at the Brigham Young University DNA Sequencing Center . Sequencing was performed as described previously [17] , [18] . For all samples sequenced by Illumina ( paired-end read technology ) , overlapping read pairs ( ORPs ) , generated by combining short fragment libraries with long sequencing reads , was used to reduce sequencing errors and improve rare variant detection accuracy . Quality filtering procedure was also described in [17] , [18] . The open source read mapping software SHRiMP2 , which was shown to have high read mapping sensitivity [19] was chosen for the tools ability to map as many reads as possible in the face of individual errors within each read [20] . All rabies reads were initially mapped to GenBank rabies reference sequence GI:260063801 . This reference sequence was used as the common coordinate system for comparing samples and identifying coding frames . Based on a later observation that this newly sequenced rabies virus genome could differ by approximately 9 relative to our selected previously sequenced reference fox rabies sequence , we checked to see if observed error rates would increase by introducing random mutations at 9 of the control reference sequence , however , no noticeable increase in error rates were observed , suggesting that read mapping parameters were able to tolerate this rate of divergence . The Binomial error model defines the expected number of non consensus bases that should occur given the assumed PCR and sequencing error rate for a given number of observed reads , using a preset P-value ( set to 0 . 01 with a Bonferonni correction ) . Non-consensus base calls were made when the number of reads with the rare variant exceeded the expected count threshold [17] . The sequencing data used in this study including reads and the analysis files used to make all base calls is available at NCBIs archive BioProject # PRJNA216100 . Data analysis include 44 samples sequenced using 454 across 10 , 330 genome positions and 40 samples sequenced using Illumina across 10 , 379 genome positions . The minimum coverage cutoff was 50 . After quality filtering , the mean coverage for the sequencing data was 980 for 454 data and 17 , 448 for Illumina ORP data , the median coverage was 777 for 454 data and 15 , 758 for Illumina ORP data ( Fig . S1 ) . In total , 10 , 451 positions of the rabies genome were sequenced by either 454 or Illumina , of these , 10330 positions ( 98 . 8 of 10 , 451 loci ) were covered by both platforms , though not necessarily for all samples an additional 36 positions ( loci 51995206 , 5215 , 5216 , 52185220 , 5224 , 95429563 ) were covered only by 454 , and 85 positions ( loci 183267 ) were covered only by Illumina , also not necessarily for all samples . The few disagreements in consensus base calls between 454 and Illumina were resolved either by taking the base call with far superior coverage or omitting the base call from data analysis entirely . In most cases the disagreement was due to low coverage of the loci by one platform ( just above the 50 cutoff ) compared to coverage by the other platform ( 500 ) . Hence the final consensus sequences of the two data sets contain no disagreement and can be considered accurate with high confidence . To differentiate rare variants from sequencing errors , methodologies were developed to measure and control for sequencing and PCR errors and described in Fig . S1 legend [17] , [18] . Briefly , all mismatched read pairs in the ORPs were identified as sequencing errors and removed from analysis . Erroneous matching read pairs in the plasmid control were used to estimate the overall PCR error rate [17] , [18] . Rates of these two types of errors were then combined in a position-dependent bionomial error model to make variant calls .
Among all 10 , 451 genome positions sequenced , 243 positions contained more than one consensus nucleotide across the samples , and 4 of these positions showed 3 different consensus nucleotides across the samples . These consensus-level variations across the samples occurred in all five genes of the rabies genome as well as four intergenic regions ( Fig . S2 ) . The intergenic regions tended have higher rates of consensus-level mutations compared to the 5 genes , with the region between G and L being the most variable ( 0 . 06 mutations per nucleotide , Table 1 , Fig . S2 B ) . Separating outbreak samples from earlier samples showed that 67 of these positions remained variable at the consensus level across the samples collected during 2009 and 2010 ( Table 1 , Fig . S2 C , D ) . The intergenic region between G and L remained the most variable , followed by the intergenic region between P and M . The M protein had the highest rate of consensus-level variation across all samples , but the G protein had the highest rate of consensus-level variation in the outbreak samples . Reads from the noncoding regions were concatenated and consensus data from 454 and Illumina sequencing were compared ( Fig . S7 ) . Most samples clustered according to date and location . Noncoding nucleotide data from representative samples from 1995 , 2000 , 2003 , 2007 , and 2010 were Blasted to identify similarity to samples in GenBank . Data from all samples had the best match to CA982 ( CA skunk from 1994 , accession no . JQ685894 ) with all samples 97 similar except for the sample from 2000 , which was 98 similar to CA982 .
Although rabies virus has jumped species multiple times in the past [5] , [8] , [9] , [25] , the event is relatively rare and deep genome sequence analysis has never been applied to examine the role of intra-host viral populations in such an event . Importantly , the rabies outbreak samples collected by the CDPH were accompanied by epidemiologically important documentation such as exact date and location for most of the samples . Additionally the CDPHs ongoing surveillance efforts provided a unique repository of samples for previous rabies host jumping events , which failed to be efficiently transmitted within the gray fox population . Next generation sequencing technology has recently been used to examine viral heterogeneity of rabies genomes present in infected tissues but has not yet been optimized for detection of rare genotypes ( less than 1 ) [26] . Deep genome sequencing of these recent and past samples allowed us to define the viral mutational dynamics that were associated with a skunk rabies virus variant that efficiently transmitted within a population of gray foxes , suggesting possible adaptation to a novel host species . Historically , the skunk rabies virus variant present in Humboldt Co . has been detected in foxes more frequently than in any other region of the state but not until 2009 has transmission shifted so disproportionally to the fox population [11] . Our data indicate that the outbreak haplotype responsible is able to be transmitted readily by both skunks and foxes since no genetic changes viral sequence differentiated the skunks and the foxes from the epizootic . These results are similar to those from a study describing the genetic changes associated with the Arizona CST events in that the rabies genotype from the donor species ( bats ) could not be differentiated from that found in the recipient species ( skunks or foxes ) [8] . All of the Humboldt samples had an unusual sequence , ETGL , as the final four amino acids at the carboxyl end of G protein . A recent study demonstrated that the last four amino acids in this region impact the virulence of the virus [27]; if the final sequence is ETRL then the virus is attenuated due to induction of neuronal apoptosis . According to this study , virulent , wild type RABV haplotypes have QTRL as the terminal sequence , and do not cause apoptosis of the host cell . The impact of the ETGL haplotype on viral virulence is unclear , although it did not perceptibly impact virulence in skunks and foxes . No amino acid changes were unique to the 2003 samples as compared to the 2009/10 outbreak samples from the Eureka area and further south ( Table 2; ) however distinctive nucleotide changes were present in the noncoding and coding regions . Temporal and genetic data indicate that the outbreak began in south Humboldt Co . and spread north to Arcata and three amino acid changes characterized samples from Arcata area and further north ( Table 2 ) . Whether or not these genetic changes contributed to the explosive increase in fox rabies that occurred primarily in Arcata during 2009 would require further study ( Fig . 2 ) . It seems likely that a subset of the foxes infected during 2009/10 were part of a fox-to-fox transmission cycle , with limited skunk-to-fox transmission occurring as well . This is supported by the fact that the CDPH collected 24 rabid fox samples from the city of Arcata from October 2008 to January 2010 but only 4 rabid fox samples from Eureka during this time period . During this time there were also 2 samples obtained from rabid skunks , one from Eureka in October 2008 and one from Arcata in April 2009 . The collection of 1 skunk to 4 fox samples seen in Eureka is not exceptional there was 1 skunk sample and 4 fox samples collected in Humboldt Co . in 2007 ( a nonepizootic year ) . In 2003 , 4 skunk samples were collected along with 10 fox samples . Thus collection of 1 skunk sample and 21 fox samples from Arcata during 2009 is exceptional and one may speculate that this was driven primarily by fox-to-fox transmission . In support of this notion , the online postings of local papers ( i . e . Times-Standard , Humboldt Sentinel , Arcata Eye ) were searched for articles relating to rabid skunks and foxes during 2009 . This search yielded mention of 14 rabid foxes , all were from Arcata . Although relatively few amino acid changes were associated with the 2003/10 host jump , it is possible that one or more of these changes may have been required for efficient transmission of the virus in the local gray fox population . Despite an extensive search of the rabies virus literature , none of these amino acid changes were described by other studies as being associated with a change in viral phenotype . Transmission studies using reverse genetics are required to identify which genetic changes are responsible for increased transmissibility . Information from this type of analysis may provide important information on the risk of a similar host jump occurring in other regions , including regions where gray foxes overlap with populations of mesocarnivores that have threatened or endangered status . Both consensus and deep-sequencing data indicate that the haplotype associated with sustained fox-to-fox transmission during the 2009 outbreak occurred prior to 2009 since several nonsynonymous mutations that were present in the consensus sequences of skunk and fox rabies samples obtained from 20032010 were present at the sub-consensus level ( as rare variants in the viral population ) in skunk and fox samples from 1995 ( Figure 5 ) . Analysis of the Illumina ultra-deep sequencing data supported the hypothesis that variants that were later enriched to become consensus could be detected at higher frequencies than variants that did not . In particular , all of the mutations that distinguish the 1995/96 haplotype from the 2003/10 haplotype were present as rare variants in Fx 31 , the only 2003 sample for which there is deep sequencing Illumina data available . These results suggest that analysis of rare variants within a viral population may yield clues to ancestral haplotypes and identify rare haplotypes that have the potential to be selected for if environment conditions change . | Understanding the role of genetic variants within a viral population is a necessary step toward predicting and treating emerging infectious diseases . The high mutation rate of RNA viruses increases the ability of these viruses to adapt to diverse hosts and cause new human and zoonotic diseases . The genetic diversity of a viral population within a host may allow the virus to adapt to a diverse array of selective pressures and enable cross-species transmission events . In 2009 a large outbreak of rabies in Northern California involved a skunk rabies virus variant that efficiently transmitted within a population of gray foxes , suggesting possible adaptation to a novel host species . To better understand the evolution of rabies virus that enabled this host jump , we applied deep-sequencing analysis to rabies virus samples from the outbreak . Deep-sequencing data indicated that many of the genetic changes associated with host jump occurred prior to 2009 , and these mutations were present at very low frequencies in viral populations from samples dating back to 1995 . These results suggest deep sequencing is useful for characterization of viral populations , and may provide insight to ancestral genomes and role of rare variants in viral emergence . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Ultra-Deep Sequencing of Intra-host Rabies Virus Populations during Cross-species Transmission |
Despite the importance of G-protein coupled receptors ( GPCRs ) their biogenesis is poorly understood . Like vertebrates , C . elegans uses a large family of GPCRs as chemoreceptors . A subset of these receptors , such as ODR-10 , requires the odr-4 and odr-8 genes to be appropriately localized to sensory cilia . The odr-4 gene encodes a conserved tail-anchored transmembrane protein; the molecular identity of odr-8 is unknown . Here , we show that odr-8 encodes the C . elegans ortholog of Ufm1-specific protease 2 ( UfSP2 ) . UfSPs are cysteine proteases identified biochemically by their ability to liberate the ubiquitin-like modifier Ufm1 from its pro-form and protein conjugates . ODR-8/UfSP2 and ODR-4 are expressed in the same set of twelve chemosensory neurons , and physically interact at the ER membrane . ODR-4 also binds ODR-10 , suggesting that an ODR-4/ODR-8 complex promotes GPCR folding , maturation , or export from the ER . The physical interaction between human ODR4 and UfSP2 suggests that this complex's role in GPCR biogenesis may be evolutionarily conserved . Unexpectedly , mutant versions of ODR-8/UfSP2 lacking catalytic residues required for protease activity can rescue all odr-8 mutant phenotypes tested . Moreover , deleting C . elegans ufm-1 does not alter chemoreceptor traffic to cilia , either in wild type or in odr-8 mutants . Thus , UfSP2 proteins have protease- and Ufm1-independent functions in GPCR biogenesis .
Molecular chaperones ensure the correct folding , assembly , quality control , traffic , and sub-cellular targeting of newly made proteins . Failure of these processes results in protein aggregation , with potential pathological consequences [1] . In the nervous system a growing number of chaperones have been identified that are specialized to facilitate the biogenesis of specific molecules [2]–[3] . Together with general chaperones , these molecules prevent accumulation of protein aggregates in neurons and provide protection against neurodegeneration [4] . GPCRs form a large family of polytopic transmembrane proteins that share a common fold [5] . Despite their importance , the biogenesis of GPCRs is poorly understood . GPCRs are co-translationally targeted to and inserted into the endoplasmic reticulum ( ER ) membrane using the canonical translocon machinery [6] . In the ER , GPCRs are often N-glycosylated and fold with the help of chaperones [7]–[9] . General chaperones implicated in GPCR folding include calnexin and calreticulin . More specific chaperones have also been identified for a few receptors , including RTP1 and RTP2 for some mammalian olfactory receptors [10] , XPORT and NinaA for fly rhodopsin [11] , and DRiP78 for the D1 dopamine receptor [12] . GPCR assembly in the ER is monitored by poorly defined quality control systems , and improperly folded receptors are targeted for refolding or degradation by the ubiquitin-proteasome system [9] . Molecules passing ER quality control are exported to downstream compartments of the secretory pathway . Motifs involved in GPCR ER export have been defined for some receptors , [12] , although some of these may be involved in correct GPCR folding , a pre-requisite for export [13] . In the Golgi and trans-Golgi network GPCRs are further glycosylated , and assembled into vesicles appropriate for traffic to their appropriate sub-cellular localization . In neurons these destinations can include dendrites , axons or synapses . While the molecular machineries mediating many steps in GPCR biogenesis are unknown , it is clear that ER retention of GPCRs due to misfolding leads to disease , including retinitis pigmentosa [14] , nephrogenic diabetes insipidus [15] , and hypogonadotropic hypogonadism [16] . In the nematode C . elegans , like in vertebrates , GPCRs play central roles in chemoreception [17]–[19] . The C . elegans genome encodes more than 1300 predicted GPCR chemoreceptors of widely divergent sequence [18] . Many of these GPCRs appear to be expressed in one or more of 12 gustatory and olfactory neuron types [17] . The most intensively studied of these receptors , ODR-10 , mediates C . elegans chemotaxis to the volatile odor diacetyl . ODR-10 is specifically expressed in the AWA olfactory neuron , and localizes to sensory cilia [19] . Screens for odortaxis defective mutants have identified two loci that disrupt ODR-10 localization to cilia , odr-4 and odr-8 [20] . In these mutants ODR-10-GFP is retained in endomembranes in the cell body . odr-4 and odr-8 are only required for localization of a subset of GPCR chemoreceptors , and do not affect localization of other signal transduction components , such as G-proteins and ion channels , to cilia [20] . These genetic data suggest that odr-4 and odr-8 encode accessory proteins involved in the maturation , traffic or localization of GPCRs . As well as being required for olfactory responses , odr-4 and odr-8 are required for npr-1 animals to aggregate efficiently [21] and to respond to variation in ambient oxygen [22] . Our interest in the molecular basis of aggregation behaviour motivated us to understand ODR-4 and ODR-8 . Earlier work had established that odr-4 encodes a tail-anchored transmembrane protein selectively expressed in 12 C . elegans chemosensory neurons and localized to unidentified intracellular membrane compartments [20] . The molecular identity of odr-8 has remained unknown . Here , we identify the gene encoded by odr-8 and provide a characterization of its relationship to odr-4 and its role in GPCR biogenesis .
We mutagenized npr-1 animals , isolated mutants that failed to aggregate , and sequenced their genomes . At the same time we sequenced two strains bearing alleles of odr-8 , ky173 and ky31 , which we previously showed disrupt aggregation behavior [21] . Genome sequence analysis identified only one open reading frame that had lesions in ky31 and ky173 alleles and mapped to the interval where odr-8 maps genetically [20] . Three other aggregation-defective strains in our sequenced collection also had mutations in this gene , which is called F38A5 . 1 . Of the five mutations in F38A5 . 1 , three introduced stop codons ( Figure 1A ) . To establish if defects in F38A5 . 1 caused odr-8 phenotypes we carried out transgenic rescue experiments . Genomic DNA bearing the F38A5 . 1 open reading frame rescued odr-8 defects in odortaxis , aggregation behavior , and responses to oxygen ( Figure 1B–D ) , confirming the molecular identity of odr-8 . F38A5 . 1 encodes the C . elegans ortholog of Ufm1-specific protease 2 ( UfSP2 ) . Ufm1 ( Ubiquitin fold modifier ) is a ubiquitin-like post-translational modifier [23] . Its tertiary structure is very similar to that of ubiquitin ( Ub ) although its primary sequence shows no obvious sequence similarity to Ub [24] . Like Ub , Ufm1 is synthesized as an inactive precursor that is proteolytically cleaved to expose a glycine C-terminal residue . This glycine can then be conjugated to target proteins at lysine residues , although only one Ufm1 target has been identified to date , the ER protein C20orf116 [25] . In vitro studies show that mouse UfSP2 can both activate pro-Ufm1 and cleave it from native protein conjugates , a process termed de-ufmylation ( Figure S1 ) [25]–[27] . Ufm1 and UfSP2 are highly conserved across eukaryotes ( Figure S1B and Figure S2 ) , but their in vivo roles are unknown . To determine where ODR-8/UfSP2 is expressed , we fused DNA encoding the fluorescent protein mCherry upstream of the odr-8 open reading frame and expressed this transgene from the odr-8 promoter . The construct rescued all the defects observed in odr-8 mutants , and revealed mCherry-ODR-8 expression in a small number of head neurons and two tail neurons . The phenotypes of odr-8 mutants closely resemble those of odr-4 defective animals [20] . ODR-4 is a conserved tail-anchored transmembrane protein ( Figure S3 ) . To examine if odr-4 or odr-8 were expressed in the same cells , we made animals that carried both podr-4::odr-4::gfp and podr-8::mCherry::odr-8 transgenes . Green and red fluorescence in these animals overlapped , consistent with ODR-4 and ODR-8 functioning in the same neurons ( Figure 2A ) . We explicitly identified odr-8 expressing neurons using the stereotyped position of C . elegans neurons and DiO dye-filling and a podr-3::gfp transgene as fiduciary markers . We observed mcherry-ODR-8 expression in 10 head neurons , the amphid neurons ADL , ASI , ASH , ASJ , ASG , ADF , ASK , AWA , AWB , AWC , and in two tail neurons , the phasmid neurons PHA and PHB . These are all the cells previously reported to express odr-4 [20] . We did not detect expression in any additional neurons . ODR-4 functions in the AWA neurons to promote chemotaxis to the odor diacetyl [20] , and in the ADL neurons to promote aggregation [21] . Expressing odr-8 specifically in AWA neurons , using the odr-10 promoter , rescued odr-8 odortaxis to diacetyl but not to benzaldehyde , which is sensed by the AWC neurons ( Figure 2B ) . Similarly , expressing odr-8 in ADL , using the srh-220 promoter , restored aggregation behavior ( Figure 2C ) and oxygen responses ( Figure 2D ) . Expressing odr-8 from the odr-3 promoter , which drives expression in the AWC olfactory neuron and a small number of other neurons including AWA , rescued odr-8 chemotaxis to both benzaldehyde and diacetyl ( Figure 2B ) . These data indicate that ODR-8 and ODR-4 function in the same neurons to promote specific behavioral responses . The behavioral defects of odr-8 and odr-4 mutants appear to reflect failure in the localization of a subset of G-protein coupled receptors ( GPCR ) to sensory cilia [20] . ODR-10 is an olfactory GPCR that mediates attraction to diacetyl [19] . In wild type animals ODR-10-GFP is localized to the sensory cilia of AWA olfactory neurons . In odr-4 or odr-8 mutants , ODR-10-GFP is predominantly found in unidentified endomembrane compartments [20] . We observed similar defects when we compared localization of two other GPCRs of the STR olfactory receptor family , STR-112 and STR-113 , in the AWA neurons of wild type and odr-8 mutants ( Figure 3A–B and Figure S4 ) . To confirm that defects in C . elegans UfSP2 caused the ODR-10-GFP trafficking phenotype , we performed transgenic rescue experiments . ODR-10-GFP was correctly localized to AWA cilia in odr-8 mutants bearing the F38A5 . 1 transgene , whereas in non-transgenic siblings ODR-10-GFP was localized to the cell body ( Figure 3C ) . Expressing the F38A5 . 1a open reading frame specifically in the AWA neurons restored cilia localization of ODR-10-GFP , whereas expression in the ADL neurons did not , suggesting that ODR-8 functions cell-autonomously ( Figure 3C and data not shown ) . The small size of C . elegans neurons makes sub-cellular localization of fluorescently-tagged proteins challenging . Nevertheless , to investigate the fate of ODR-10-GFP in odr-4 and odr-8 mutants , we co-localized it with mCherry-tagged markers for different membrane compartments that we expressed specifically in AWA ( Figure 4A ) . We included in our studies the rough ER marker TRAM-1 , the Golgi marker alpha-mannosidase II [28] , the ER exit site markers SEC-16 and SEC-23 [29] , the early endosome marker RAB-5 [30] , the lysosomal marker LMP-1 [31] [32] , and RAB-8 which in sensory neurons marks vesicles destined for traffic to the cilia [33] . As expected , in wild type animals , ODR-10-GFP was predominantly localized to cilia , with some fluorescence in the cell body ( Figure 4A and Figure S5 ) . The cell body fluorescence co-localized most extensively with mCherry-RAB-8 , in foci that were closely apposed to Golgi ( Figure 4A and Figure S5 ) . We also observed some co-localization with the lysosomal marker LMP-1 ( Figure 4A and Figure S5 ) . Co-localization of ODR-10-GFP with mCherry-RAB-8 has been reported previously when the receptor was expressed heterologously in the AWB and phasmid neurons [33] . In odr-8 mutants co-localization of ODR-10-GFP and mCherry-RAB-8 was lost; instead ODR-10-GFP co-localized most extensively with markers for the ER ( Figure 4A and Figure S5 ) . These data suggest that in odr-8 mutants the ODR-10-GFP receptor either does not exit the ER , or undergoes retrograde traffic back to the ER , or is degraded following ER exit . We sought to establish the sub-cellular compartments to which ODR-4 and ODR-8 proteins predominantly localized to in AWA neurons . We made transgenes expressing ODR-4-GFP or mCherry-ODR-8 in AWA , and crossed them into animals in which different compartments were highlighted in AWA neurons using mCherry-tagged markers . ODR-4-GFP co-localized with the ER marker TRAM-1 ( Figure 4B ) , suggesting ODR-4 functions in this compartment . mCherry-ODR-8 showed some co-localization with ODR-4 , although we also observed diffuse staining consistent with cytoplasmic localization ( Figure 4C ) . We sought to design genetic experiments that shed light on where ODR-4 and ODR-8 function in the secretory pathway . Previous work indicates that post-Golgi sorting of ODR-10-GFP involves formation of clathrin-coated vesicles via adapter complex 1 ( AP-1 ) [33]–[34] . Mutants lacking unc-101 , which encodes the mu1 subunit of AP1 , fail to traffic ODR-10-GFP to sensory cilia and instead traffic the receptor to the cell membrane . We reasoned that if odr-8 mutations disrupt ODR-10-GFP receptor maturation prior to its reaching the Golgi , then disrupting unc-101 should not alter ODR-10-GFP localization in odr-8 mutants . Consistent with this , we observed no difference in ODR-10-GFP localization between odr-8 and odr-8; unc-101 animals ( Figure 5A–B ) . To extend our studies we imaged ODR-10-GFP localization in animals defective in or expressing dominant negative or dominant active version of the RAB GTPases RAB-1 , RAB-2 , RAB-6 . 1 and RAB-6 . 2 . These RABs play roles at different steps in the secretory pathway [35] , and some have previously been implicated in GPCR traffic [36] . In yeast , Rab1 is thought to be important for ER exit and traffic to the Golgi , Rab2 is implicated in retrograde traffic to the ER from the Golgi , and Rab6 plays a role in intra-Golgi traffic . For rab-1 , rab-6 . 1 and rab-6 . 2 we expressed dominant negative and dominant active transgenes in AWA using the odr-10 promoter ( see methods ) . For rab-2 we used the deletion allele nu415 , which is predicted to be null , and the n777 allele , which is associated with a S149F mutation that constitutively activates RAB-2 [37] . None of the perturbations of RAB protein function completely blocked traffic of ODR-10-GFP to AWA cilia in wild type animals , although there were clear and significant reductions of ODR-10 cilia localization in all cases except when we expressed rab-6 . 1 ( Q70L ) ( Figure S6 ) . In rab-2 ( n777 ) mutants , ODR-10 was not correctly localized to AWA cilia in 6 out of 29 animals ( Figure 5C–E ) , and more strikingly , a significant amount of ODR-10 was mis-sorted to axons ( Figure 5C–E ) . Previous work in C . elegans has implicated RAB-2 in biogenesis of dense core vesicles [37] , [38] and in post-endocytic trafficking [39] , suggesting a more complex role for this RAB protein . Axonal mislocalization of ODR-10 was not observed in rab-2 ( n777 ) ; odr-8 double mutants , consistent with odr-8 being required early in the secretory pathway ( Figure 5C–E ) . For the remaining rab gene perturbations , odr-8; rab double mutants , and odr-8 animals expressing mutant rab transgenes resulted in an ODR-10-GFP localization phenotype that was indistinguishable from that of odr-8 mutants . Together , these observations and the unc-101 data suggest that ODR-8 functions at an early step in GPCR trafficking . Biochemical studies indicate that mouse UfSP2 de-ufmylase can activate pro-Ufm1 and cleave it from native protein conjugates [26] [25] . We therefore examined whether disrupting the C . elegans orthologs of Uba5 or Ufc1 , the only known E1 and E2 ligases implicated in ufmylation [23] ( Figure S1 ) , or deleting C . elegans ufm-1 itself , altered ODR-10 localization . C . elegans UBA-5 is encoded by T03F1 . 1 [40] whereas C . elegans UFC-1 is encoded by C40H1 . 6 . UFM-1 is encoded by tag-277 . Deletion mutations that are likely to be complete loss-of-function alleles are available for all three genes [40] ( Figure S7A and Figure 6 ) . If the mutant phenotypes of odr-8/ufsp2 reflected a failure to activate pro-ufm1 , then disrupting ufmylation should recapitulate odr-8 phenotypes . Conversely , if the primary defect of odr-8/ufsp2 is failure to cleave UFM-1 from native protein conjugates , then disrupting ufm-1 or inhibiting ufmylation might suppress odr-8 phenotypes . Deleting Ce-uba-5 or Ce-ufc-1 did not affect viability , and had no visible effect on diacetyl chemotaxis or on ODR-10-GFP localization at AWA cilia in an otherwise wild type background ( Figure S7B–C ) . No change in the phenotype of an odr-8 mutant was observed upon further deletion of Ce-uba-5 or Ce-ufc-1 ( Figure S7B–C ) . The only available allele that deletes Ce-ufm-1 , gk379 , also removes part of an operon that includes the ubiquinone biosynthetic enzyme coq-5 , and results in embryonic lethality ( Figure 6A ) . We suspected that lethality reflected disruption of the coq-5 operon , rather than loss of ufm-1 . Consistent with this , we could rescue the gk379 embryonic lethality with a 9 . 1 kb genomic fragment of the region in which we truncated the ufm-1 open reading frame by a premature stop codon or a frame shift mutation ( Figure 6A ) . This allowed us to examine ODR-10-GFP localization at AWA cilia in adult animals lacking UFM-1 . We did not observe any significant change in ODR-10-GFP localization , either in odr-8 ( + ) or in odr-8 ( − ) animals that also lacked ufm-1 ( Figure 6B ) . Furthermore , we could not detect UFM-1 expression in AWA neurons ( Figure S7D–E ) . To confirm our observations , we generated three independent ufm-1 loss-of-function mutations using CRISPR-Cas gene-editing ( Figure 6C ) [41] . Two mutations caused frameshifts of the ufm-1 open reading frame , and presumably are null alleles ( Figure 6C ) . None of the mutations disrupted ODR-10-GFP cilia localization in odr-8 ( + ) animals , or rescued the ODR-10-GFP cilia localization defect in an odr-8 mutant background ( Figure 6D–E ) . These data suggest that the primary role of ODR-8 UfSP2 in biogenesis of ODR-10-GFP is independent of ufmylation . Biochemistry and crystallography have highlighted the active sites of the UfSP1 and UfSP2 proteases [25]–[27] . Like in several de-ubiquitinases , the active sites of these enzymes include highly conserved cysteine and histidine boxes ( Figure 7A ) . The cysteine box contains the catalytic cysteine residue that is thought to undergo de-protonation , enabling nucleophilic attack of the carbonyl carbon of incoming substrates . The histidine in the histidine box assists this de-protonation . In vitro biochemical studies show that mutating the active site cysteine residue to a serine , or the histidine to an alanine , abolishes UfSP1 and UfSP2 proteolytic activity [25] , [27] . To examine if an intact active site was required for ODR-8/UfSP2 in vivo functions , we made a transgene encoding an ODR-8 in which the active site cysteine was mutated to a serine . Unexpectedly , this transgene rescued all the behavioral defects of odr-8 mutants we tested , and restored ODR-10-GFP localization to AWA cilia ( Figure 7B–D ) . Since serine residues can engage in nucleophilic attack , we also created a transgene encoding an ODR-8 in which the active site cysteine was mutated to an alanine . This Cys-to-Ala active site mutant also rescued the behavioral and cell biological defects of odr-8 mutants ( Figure 7D and data not shown ) . We next mutated the active site histidine residue to an alanine , and tested the biological function of the protein in transgenic animals . The His-to-Ala odr-8 transgene also substantially although not completely rescued the behavioral and cell biological phenotypes of odr-8 mutants . Incomplete rescue could be due to effects of this mutation on protein structure ( Figure 7B–D ) . Together , these data suggest that the protease activity of ODR-8 UfSP2 is not essential for its in vivo function in GPCR maturation . The expression patterns of ODR-8 and ODR-4 , the subcellular localizations of the two proteins , and the phenotypes displayed by odr-8 and odr-4 mutants all suggested that these two proteins function together to promote ODR-10 trafficking to the cilia . However , the small size of AWA neurons and the limits of light microscopy prevented high-resolution co-localization of ODR-4 and ODR-8 . In addition , biochemical analyses of ODR-4/ODR-8/ODR-10 protein interactions in worm extracts were precluded by the restricted expression patterns of these proteins in vivo . To circumvent these obstacles , we heterologously expressed ODR-4 , ODR-8 , and ODR-10 in cultured mammalian cells and analyzed their biochemical interactions and sub-cellular localization . Epitope-tagged ODR-8 and ODR-4 ( two isoforms , ODR-4a and ODR-4b ) were expressed either individually or in combination in HEK293 cells and analyzed by immunoprecipitation ( IP ) . HA-ODR-8 co-precipitated with either ODR-4a-FLAG or ODR-4b-FLAG , regardless of which component was subjected to IP ( Figure 8A ) . Immunoprecipitation of ODR-4 could also co-precipitate ODR-10-GFP ( Figure 8B ) , while ODR-8 did not seem to directly interact with ODR-10 ( Data not shown ) . Since ODR-4 and ODR-8 are conserved , we asked if their human orthologs also interacted biochemically . Immunoprecipitation of hODR4-FLAG brought with it HA-UfSP2 ( Figure 8C ) , suggesting that like the C . elegans proteins , the human orthologs can form a complex in human cells . Immunofluorescence studies localized the site of these interactions to the ER . Here , we found that ODR-4 co-localized with the ER resident protein TRAPα ( Figure 8D ) . By contrast , ODR-8 was diffusely cytosolic when expressed alone , but somewhat reticular when co-expressed with ODR-4 ( Figure 8E ) . This suggested that ODR-8 was recruited to the ER via its physical interaction with ODR-4 . This was confirmed by demonstrating that ODR-8 was efficiently released from cells by permeabilization of the plasma membrane , but partially retained when co-expressed with ODR-4 ( Figure 8F ) . The retained ODR-4 and ODR-8 were co-localized , consistent with their physical interaction ( Figure 8E ) . Thus , ODR-4 and ODR-8 form a physical complex at the ER membrane . ODR-10-GFP was also ER-localized in HeLa cells ( Figure S8A ) . This localization remained qualitatively similar in cells co-expressing ODR-4 and/or ODR-8 ( Figure S8B ) . This suggests that while ODR-4 and ODR-8 are clearly necessary for proper ODR-10 surface trafficking in AWA neurons , they may not be sufficient in the heterologous HeLa cell system . This remains to be investigated further . Nevertheless , the analysis in cells , together with the genetic and ( lower resolution ) localization data in AWA neurons , indicates that ODR-4 and ODR-8 form an ER-localized complex that facilitates ODR-10 surface expression .
ODR-8 , the C . elegans ortholog of Ufm1 specific protease 2 , is involved in the biogenesis of GPCRs via a non-catalytic mechanism that does not require Ufm1 . odr-8/UfSP2 mutants , like odr-4 mutants , fail to localize a subset of olfactory receptors to sensory cilia . This defect occurs at an early step in secretory pathway , most likely prior to exit from the endoplasmic reticulum . Consistent with this , ODR-4 , ODR-8 and ODR-10 form a complex in the ER of HEK cells . UfSP2 and its relative UfSP1 were first identified biochemically in mouse by virtue of their ability to process the ubiquitin-like protein Ufm1 [26] . UfSP2 is found throughout eukaryotes , from plants to man . UfSP1 is found in mouse , man , and flies but not C . elegans . Crystallography shows that UfSP1 and UfSP2 have a papain-like domain with a Cysteine-Aspartate-Histidine catalytic triad [25] , [27] . The fold structure is most similar to that of the autophagy gene Atg4 , which de-conjugates Atg8 ubiquitin-like proteins from phosphatidylethanolamine and promotes autophagosome formation . Mutagenesis of active site residues confirms the importance of these residues for in vitro UfSP protease activity: UfSP1 and UfSP2 are catalytically dead if the catalytic cysteine is mutated to a serine . The residues comprising the papain fold , including the catalytic residues , are highly conserved across UfSP orthologs , with 40% identity from grape vine to man . Unexpectedly , we find that active site residues of ODR-8 UfSP2 are not required for any of its known in vivo functions . Mutating the catalytic cysteine or histidine residues of C . elegans UfSP2 does not abolish these functions . Thus , at least in C . elegans , and perhaps more generally , the major functions of UfSP2 do not require proteolytic activity . In addition , deleting Ce-ufm1 does not perturb localization of the ODR-10-GFP GPCR in AWA neurons , either in wild-type or in odr-8 mutants . This suggests that ODR-8 UfSP2 and UFM-1 do not necessarily have to function together in vivo . ODR-4 localizes predominantly to the ER in vivo , and the soluble ODR-8 UfSP2 protein also shows some ER localization . Previous work showed that mouse UfSP2 is recruited to the ER when expressed with its client protein C20orf116 [25] . We show that ODR-4 can also recruit ODR-8 UfSP2 to the ER in HeLa cells . Disrupting either ODR-4 or ODR-8 results in some olfactory GPCRs , including ODR-10 , being retained in the ER rather than trafficking in RAB-8-containing vesicles to sensory cilia . Neither ODR-4 nor ODR-8 accumulates preferentially at ER exit sites , and ODR-10-GFP does not appear to accumulate at ER exit sites in odr-4 or odr-8 mutants , suggesting these proteins are not involved in ER exit per se . ODR-4 and ODR-8 , and ODR-4 and ODR-10 interact biochemically in HEK cells , A simple interpretation of all our data is that ODR-4 and ODR-8/UfSP2 form a complex in the ER with some olfactory receptors , including ODR-10 , that allows these GPCRs to mature and become competent to exit the ER . In the absence of maturation , these GPCRs are retained in the ER . Quality control of folding in the ER is thought to involve cycles of ubiquitination and de-ubiquitination [42]–[44] . One highly speculative model is ODR-8/UfSP2 binds to ubiquitin or a ubiquitin-like molecule to regulate maturation of GPCRs . In C . elegans ODR-4 and ODR-8/UfSP2 are expressed in the same small set of chemosensory neurons , and their function appears limited to maturation of a subset of GPCR chemoreceptors . Orthologues of ODR-8/UfSP2 and ODR-4 are found in organisms that do not have a nervous system , such as plants . Moreover , the mouse orthologs of ODR-4 and ODR-8 are expressed in many non-neuronal tissues . A simple explanation is that outside C . elegans these proteins participate in maturation of non-olfactory GPCRs . Interestingly , human ODR-4 and human UfSP2 form a complex in HEK cells , suggesting they also function together in humans .
Animals were grown under standard conditions [45] . Strains used are listed in Table S1 . Aggregation behavior , chemotaxis , and responses to changes in oxygen levels were measured as described previously [46]–[48] . Whole genome sequencing was carried out on the Illumina HiSeq 2000 platform . Plasmid construction: Plasmids were constructed using the multi-site Gateway system ( Invitrogen ) . Promoters used include podr-10 ( 1 . 2 kb ) , podr-8 ( 3 . 1 kb ) , psrh-220 ( 2 . 1 kb ) , pstr-1 ( 4 kb ) , podr-3 ( 2 . 7 kb ) , and pufm-1 ( 4 . 1 kb ) ; numbers refer to DNA upstream of the initiation codon . Promoters were cloned into the first position of the Gateway system , genes of interest were cloned at the second position , and the unc-54 3′UTR or the SL2::mCherry sequence at the third position . To rescue the odr-8 mutant phenotypes , we used a genomic fragment encoding F38A5 . 1a ( Figure S9 ) . To generate in frame C-terminal GFP or mCherry fusions , genes of interest were cloned in the second Gateway position without a stop codon , and the third position was occupied by GFP or mCherry DNA . For constructs encoding N-terminus GFP or mCherry fusion proteins , GFP or mCherry sequences without stop codons were placed at the second position and genes of interest at the third position . When the third position was occupied by sequences other than unc-54 3′UTR , we used a destination vector that placed unc-54 3′UTR sequences downstream of the 3rd Gateway insert . We rescued the lethality associated with the gk379 deletion using a ∼9 . 1 kb genomic fragment centered on the ufm-1 gene . This fragment was PCR amplified using primers flanked with attB1 and attB2 sites and cloned into pDonr221 using the BP reaction . To target the ufm-1 gene by CRISPR , a ufm-1 gene-specific sequence was cloned into the sgRNA vector bearing rpr-1 promoter using the Gibson assembly kit , as described [41] . Site-specific mutagenesis: Mutagenesis was performed using the Quikchange II XL kit from Agilent . Changes made include: odr-8 ( C421S: TGT to TCT ) ; odr-8 ( C421A: TGT to GCT ) ; odr-8 ( H547A: CAT to GCT ) ; ufm-1 ( S2stop: TCG to TAA ) ; ufm-1 ( S2stop:TCG to TAA; nucleotide A at position 12 deleted ) ; rab-1 ( GDP bound S25N: TCG to AAT ) ; rab-1 ( GTP bound Q70L: CAG to CTG ) ; rab-6 . 1 ( GDP bound T25N: ACT to AAT ) ; rab-6 . 2 ( GTP bound Q70L: CAG to CTG ) ; rab-6 . 2 ( GDP bound T24N: ACC to AAC ) and rab-6 . 2 ( GTP bound Q69L: CAG to CTG ) . Transgenic strains: the ufm-1 rescue plasmid and fusion genes that express fluorescent markers of sub-cellular compartments were injected at 2–4 ng/ul together with 50 ng/ul of coelomocyte marker and 1 kb ladder . Other constructs were microinjected at 50 ng/ul . Animals were mounted on 2% agarose pads containing 50 mM sodium azide . Confocal images were taken on a Nikon Eclipse Ti inverted microscope coupled to the Andor Ixon EMCCD camera and spinning disk confocal unit . GFP signals were captured with a 100 ms exposure time at an EMCCD Gain of 300 , except for the ODR-4-4 images , which were captured with 5 ms exposure times . The mCherry signals of subcellular markers were imaged with different exposure times , according to their signal intensities . The subcellular localization images were generated by averaging 32 images . Co-localization of ODR-8 with various markers was quantified as described [39] with slight modifications . We used the co-localization analysis package in Huygens software ( Figure S5 ) . A region of interest was cropped around the AWA cell body . The pixel intensity representing the brightest 4% of pixels in each image was extracted for both green and red channels . The pixel intensity values corresponding to the cut-off of the brightest 4% of pixels were used to threshold each image . The thresholded images were used to calculate the percentage of co-localization between ODR-8 and different markers . To quantify the amount of ODR-10::GFP in the cilia and cell body , z-stack images were taken on a spinning disk confocal microscope using a 100× lens and 100 ms exposure time . The 3-D images were reconstituted with the aid of IMARIS software ( Figure S4 ) . GFP pixel intensities brighter than 1200 were cropped by creating a surface with 0 . 25 µm details . The total pixel intensities inside the surface were calculated . Frameshift mutations in ufm-1 were obtained using the CRISPR-Cas9 system , as described [41] . Briefly , Cas9 was expressed from the eft-3 promoter , while the sgRNA targeting the ufm-1 gene was expressed from the rpr-1 promoter . The Cas9 and sgRNA constructs were injected into the worm gonad with co-injection marker ( cc::GFP ) . 38 ccGFP-positive F1 animals were singled , and genotyped after the eggs were laid . HeLa cells grown on coverslips were washed with PBS and fixed in 3 . 7% formaldehyde in PBS for 15 min . For semi-permeabilization , cells were treated with 50 µg/ml digitonin in KHM buffer ( 110 mM KAc , 20 mM HEPES ( pH 7 . 4 ) and 2 mM MgCl2 ) for 5 min at 4°C before fixation . Fixed cells were permeabilized with 0 . 1% Triton ×100 in PBS for 5 min , blocked with 10% fetal bovine serum in PBS for 30 min , and incubated with primary antibodies for 1 h . After washing , cells were incubated with AlexaFluor 488-conjugated goat anti-mouse IgG and/or AlexaFluor 564-conjugated goat anti-rabbit IgG secondary antibodies ( Invitrogen ) for 60 min . Plasma membrane was stained using 0 . 01% PKH26 , according to the manufacturer's instructions ( Sigma ) . Images were acquired on a confocal laser microscope ( LSM 780 , Zeiss ) using a 63× oil-immersion objective lens with a numerical aperture ( NA ) of 1 . 42 . Cell lysates were prepared in a lysis buffer ( 50 mM Tris-HCl ( pH 7 . 5 ) , 150 mM NaCl , 1% Triton X-100 , 1 mM phenylmethanesulfonyl fluoride ( PMSF ) and protease inhibitor cocktail ( complete EDTA-free protease inhibitor , Roche ) ) . The lysates were clarified by centrifugation at 15 , 000 rpm and subjected to immunoprecipitation using anti-Flag M2 affinity gel and anti-HA agarose ( Sigma ) . Precipitated immunocomplexes were washed five times in a washing buffer ( 50 mM Tris-HCl ( pH 7 . 5 ) , 150 mM NaCl , 1 mM EDTA , and 1% Triton X-100 ) and boiled in sample buffer . Samples were subsequently separated by SDS-PAGE and transferred to Nitrocellulose membranes ( Biorad ) . Immunoblot analysis was performed with anti-GFP , anti-Flag M2 antibody and anti-HA ( clone 16B12 ) , and visualized with Super Signal West Pico Chemiluminescent substrate ( Pierce ) . | Despite the importance of G-protein coupled receptors ( GPCRs ) , we know little about their biogenesis . Olfactory receptors form a large and divergent group of GPCRs . We investigate their biogenesis in C . elegans . We show that maturation of a subset of these GPCRs , including the diacetyl receptor ODR-10 , requires Ufm1 specific protease 2 ( UfSP2 ) , which corresponds to odr-8 . Biochemical studies suggest mouse UfSP2 activates the Ubiquitin-like molecule Ufm1 and cleaves it from protein conjugates . However , neither the protease active site nor ufm-1 is required for UfSP2/ODR-8 to promote ODR-10 maturation . C . elegans UfSP2 is expressed in the same chemosensory neurons as ODR-4 , a tail-anchored transmembrane protein also required for ODR-10 maturation . ODR-4 resides in the endoplasmic reticulum ( ER ) ; UfSP2 is cytosolic but associates with ER membranes . In odr-4 and odr-8 mutants ODR-10-GFP is retained in the ER , suggesting these genes are required to fold GPCRs or traffic them out of the ER . ODR-4 interacts biochemically with ODR-8 and ODR-10 to form an ER complex . ODR-4 and UfSP2 are conserved from plants to man , and human ODR4 can bind human UfSP2 and recruit it to ER membranes . Both proteins are expressed widely in mammals , suggesting a broader role in GPCR biogenesis . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"animal",
"models",
"molecular",
"neuroscience",
"caenorhabditis",
"elegans",
"behavioral",
"neuroscience",
"model",
"organisms",
"cellular",
"neuroscience",
"gustatory",
"system",
"olfactory",
"system",
"genetics",
"membranes",
"and",
"sorting",
"molecular",
"genetics",
"signaling",
"pathways",
"biology",
"sensory",
"systems",
"molecular",
"cell",
"biology",
"neuroscience",
"gene",
"function"
] | 2014 | An ER Complex of ODR-4 and ODR-8/Ufm1 Specific Protease 2 Promotes GPCR Maturation by a Ufm1-Independent Mechanism |
Prior GWAS have identified loci associated with red blood cell ( RBC ) traits in populations of European , African , and Asian ancestry . These studies have not included individuals with an Amerindian ancestral background , such as Hispanics/Latinos , nor evaluated the full spectrum of genomic variation beyond single nucleotide variants . Using a custom genotyping array enriched for Amerindian ancestral content and 1000 Genomes imputation , we performed GWAS in 12 , 502 participants of Hispanic Community Health Study and Study of Latinos ( HCHS/SOL ) for hematocrit , hemoglobin , RBC count , RBC distribution width ( RDW ) , and RBC indices . Approximately 60% of previously reported RBC trait loci generalized to HCHS/SOL Hispanics/Latinos , including African ancestral alpha- and beta-globin gene variants . In addition to the known 3 . 8kb alpha-globin copy number variant , we identified an Amerindian ancestral association in an alpha-globin regulatory region on chromosome 16p13 . 3 for mean corpuscular volume and mean corpuscular hemoglobin . We also discovered and replicated three genome-wide significant variants in previously unreported loci for RDW ( SLC12A2 rs17764730 , PSMB5 rs941718 ) , and hematocrit ( PROX1 rs3754140 ) . Among the proxy variants at the SLC12A2 locus we identified rs3812049 , located in a bi-directional promoter between SLC12A2 ( which encodes a red cell membrane ion-transport protein ) and an upstream anti-sense long-noncoding RNA , LINC01184 , as the likely causal variant . We further demonstrate that disruption of the regulatory element harboring rs3812049 affects transcription of SLC12A2 and LINC01184 in human erythroid progenitor cells . Together , these results reinforce the importance of genetic study of diverse ancestral populations , in particular Hispanics/Latinos .
Red blood cell ( RBC ) development and maintenance are critical for transport of oxygen to tissues throughout the body . Several parameters commonly measured in clinical blood count evaluations are used to characterize RBC: hematocrit ( HCT ) , hemoglobin ( HGB ) , RBC count , mean corpuscular hemoglobin ( MCH ) , mean corpuscular hemoglobin concentration ( MCHC ) , mean corpuscular volume ( MCV ) , and red cell distribution width ( RDW ) ( detailed trait description provided in S1 Table ) . RBC traits differ by self-reported ancestry , and both genetic ( e . g . , inherited hemoglobin variants ) and acquired ( e . g . , iron deficiency , kidney disease ) factors contribute to these ethnic differences[1 , 2] . Quantitative RBC parameters are also polygenic traits that exhibit moderate to high heritability ( trait-specific h2 between 40% and 90% ) [3–5] . Over 80 genomic regions have been associated with one or more RBC traits through genome-wide association studies ( GWAS ) , performed primarily in European- and , to a lesser extent , Asian- and African-descent populations[6–14] . Hispanics/Latinos are ethnically heterogeneous , with admixture of European , West African , and Amerindian ancestral populations . In general , RBC trait values among Hispanics/Latinos have been reported to be similar to those among non-Hispanic whites , though certain types of congenital and acquired anemias are more common among Hispanics/Latinos[15–19] . As with most complex traits , GWAS for discovery or generalization of RBC trait loci has yet to be performed in Hispanics/Latinos or other populations with Amerindian ancestry . In the current study , we performed genome-wide association analysis of seven quantitative RBC traits in 12 , 502 participants ascertained by the Hispanic Community Health Study/Study of Latinos ( HCHS/SOL ) and replicated any new association findings discovered in HCHS/SOL in three independent samples of Hispanic/Latino Americans .
Of the 24 genomic regions harboring variants that reach genome-wide significance for association with RBC traits in HCHS/SOL , 17 have been previously found to associate with RBC traits either through GWAS and/or Mendelian RBC disorders . Genomic regions and variants previously implicated in Mendelian RBC disorders include the African ancestral alleles for sickle cell trait/anemia or hemoglobin S ( HBB rs334 ) ; hemoglobin C ( HBB rs33930165 ) ; the common African form of G6PD A- deficiency ( rs1050828 ) ; the 3 . 8kb alpha-globin gene deletion responsible for alpha-thalassemia trait ( esv2676630 ) ; and a proxy SNP ( rs2032451 ) for the European hereditary hemochromatosis ( HFE ) p . H63D allele . At 13 of the 17 previously reported RBC loci , the lead variant for the trait detected in HCHS/SOL Hispanics was the same as the previously reported index SNP in European- , African- , or Asian-descent individuals or a strong linkage disequilibrium ( LD ) proxy ( r2 >0 . 8 ) for the variant , where LD was measured in the relevant ancestral population in 1000 Genomes . There were four cases in which the lead variant in HCHS/SOL was not an LD equivalent to the reported index SNP . The first , rs607203 ( MAF = 0 . 07 ) , is a lead SNP for MCH and MCV association loci located within a DNaseI hypersensitive region on chromosome 6q24 approximately 146kb upstream of CITED2 . Rs607203 is not in strong LD ( HCHS/SOL r2 between 0 . 06 and 0 . 11 ) with any of the previously reported CITED2 European or Japanese index SNPs ( rs590856 , rs643381 , rs628751 , rs668459 , rs632057 ) , and therefore appears to represent an independent signal in the CITED2 locus . Among 1000 Genomes super-populations , the frequency of rs607203 is highest in African ( AFR ) ( MAF = 0 . 14 ) populations; uncommon in European ( EUR ) , American admixed ( AMR ) , and South Asian ( SAS ) ( MAF<0 . 05 ) populations; and monomorphic in East Asian ( EAS ) populations . A second exception is rs4714548 , an intronic SNP of CCND3 associated with MCV . This HCHS/SOL lead SNP exhibits weak or no LD ( HCHS/SOL r2<0 . 1 ) with any of the CCND3 index SNPs previously reported in Europeans ( rs9349204 , rs9349205 ) or Japanese ( rs3218097 ) populations . Additionally , we report novel associations for two of the variants significantly associated with RDW in HCHS/SOL: SLC12A7 rs4565255 and TMPRSS6 rs855791 . SLC12A7 rs4565255 is a proxy for rs4580814 , which was previously associated with MCHC in Japanese populations[9] . TMPRSS6 rs855791 has been previously associated with multiple red cell and iron-related phenotypes , but not with RDW[6 , 8 , 9] . To formally assess whether variants previously associated with RBC traits in populations of European , Asian , and African ancestry generalized to HCHS/SOL Hispanics/Latinos , we used a directional FDR approach . Of 251 unique published SNP associations with any of the seven RBC traits , 146 ( 58% ) generalized to HCHS/SOL ( S4 Table ) . The proportion of loci generalized varied by RBC trait: 5 of 13 HCT variants generalized ( 38% of SNPs , 42% of loci ) ; 17 of 42 HGB variants generalized ( 40% of SNPs , 37% of loci ) ; 24 of 33 RBC variants generalized ( 73% of SNPs , 61% of loci ) ; 38 of 61 MCH variants generalized ( 62% of SNPs , 61% of loci ) ; 12 of 25 MCHC variants generalized ( 48% of SNPs , 33% of loci ) ; 49 of 76 MCV variants generalized ( 64% of SNPs , 58% of loci ) ; and the only variant previously associated with RDW generalized . The seven remaining genome-wide significant variants in the HCHS/SOL discovery sample were at previously undetected loci ( Table 1 ) , and three of these variants replicated in a meta-analysis of three independent Hispanic/Latino samples ( Table 2 , S2 Table ) . The replicated loci are ( 1 ) chromosome 1q32 . 3 PROX1 rs3754140 ( MAF = 0 . 39 , replication p = 5 . 2x10-3 ) associated with HCT; ( 2 ) chromosome 5q23 . 3 SLC12A2 rs17764730 ( MAF = 0 . 18 , replication p = 1 . 6x10-3 ) associated with RDW; and ( 3 ) chromosome 14q11 . 2 PSMB5 rs7147308 ( MAF = 0 . 30 , replication p = 1 . 4x10-5 ) associated with RDW . The four loci that did not meet the Bonferroni-corrected replication threshold ( P< 0 . 0071 ) are ( 1 ) RBFOX3 rs76539504 associated with RBC count ( MAF = 0 . 04 , replication p = 0 . 31 ) ; ( 2 ) MCTP2 rs111473449 ( MAF = 0 . 03 , replication p = 0 . 037 ) ; ( 3 ) an intergenic variant on chromosome 1q31 ( rs6685034 , MAF = 0 . 41 , replication p = 0 . 26 ) associated with RDW; and ( 4 ) IDO2 rs141848064 ( MAF = 0 . 02 , replication p = 0 . 72 ) associated with MCV . At each of the three replicated discovery RBC-associated loci , we evaluated the functional genomic annotation and regulatory potential of the lead variant and any proxy variants ( r2≥0 . 8 in HCHS/SOL ) in erythroid cells to determine the most likely causal variant ( s ) . We identified the following variants as the most likely functional candidates: three intronic SNPs of PROX1 ( rs7541039 , rs7517701 , and rs4282786 ) located within the same erythroid enhancer; one SNP 3’ of PSMB5 ( rs11846575 ) ; and rs3812049 , which is located in a bi-directional promoter between SLC12A2 and an anti-sense long noncoding RNA , LINC01184 ( S5 and S6 Tables ) . We next performed mutagenesis analysis of the regions containing the PROX1 , PSMB5 , and SLC12A2 candidate causal variants using CRISPR-Cas9 genome editing to disrupt the respective putative regulatory elements in human umbilical cord-derived erythroid progenitor ( HUDEP-2 ) cells ( oligonucleotide sequences described in S7 Table ) . At the SLC12A2 locus , a single guide RNA was expressed along with Cas9 to produce indels surrounding the predicted functional SNP rs3812049 . These edits resulted in a substantial decrease in expression of both SLC12A2 and LINC01184 ( Fig 1 ) . Differentiation of erythroid cells was not obviously affected by disruption of the bi-directional promoter site . In a separate mutagenesis experiment , deletion of the third exon of LINC01184 resulted in a 3-fold reduction in LINC01184 expression , but did not appear to exhibit substantial cis effects on SLC12A2 expression ( S3 Fig ) . While the candidate regulatory region of PROX1 is located within an erythroid enhancer , PROX1 itself is not expressed in human erythroid cells including HUDEP-2 , suggesting that the enhancer element might regulate a distal target . However , a 700 base-pair biallelic deletion of the PROX1 intronic region containing rs7541039 , rs7517701 , and rs4282786 did not show any effect on HUDEP-2 cell maturation or on expression of neighboring genes SMYD2 and CENPF , both located within 300 kb of the putative enhancer element ( S4 Fig ) . Similarly , deletion of the putative enhancer downstream of PSMB5 did not significantly alter expression of PSMB5 or neighboring genes ( PRMT5 , HAUS4 , C14ORF93 , and ACIN1 ) that are both expressed in erythroid precursors and located within the same topologically associated domain of K562 cells ( S4 Fig ) . Since the quality of structural variants imputed from 1000 Genomes may be lower than single nucleotide variants , we applied a specialized copy number variant ( CNV ) calling algorithm to re-type the key 3 . 8kb alpha-globin structural variant using raw probe intensity data from the custom 2 . 5M Illumina genotyping array used in HCHS/SOL , as described under Methods . Comparison of the CNV genotype calls to those for esv2676630 imputed from 1000 Genomes revealed that genotype calling using imputation appears to result in “under-calling” of the 3 . 8kb deletion , especially homozygous deletions ( S8 Table ) . In addition , there are a number of individuals in HCHS/SOL who carry a 3 . 8kb duplication ( 3 or 4 copies of the structural variant ) , which are mis-called by 1000 Genomes imputation as wild-type . Notably , the improvement in genotype accuracy with the CNV calling algorithm resulted in a nearly two-fold increase in effect size for MCH and MCV ( Table 3 ) compared to 1000 Genomes imputation ( S9 Table ) . Therefore conditional association analyses were performed using alpha-globin deletion/duplication genotypes derived from the CNV calling algorithm . To identify additional independent association signals at known or novel RBC-associated loci , we performed step-wise conditional regression analyses in which we adjusted for the index variant at each genome-wide significant locus . The analysis was repeated with adjustment for each independently associated single variant or structural variant until no further independent signals were identified within that genomic region . Using a significance threshold of α = 5x10-8 , we identified additional independent variants associated with one or more RBC traits ( Table 3 ) in two genomic regions . At the beta-globin locus on chromosome 11p15 containing the index SNP rs334 ( sickle cell variant ) , there was an additional intergenic variant ( rs113342804 ) independently associated with MCV . At the terminal region of chromosome 16p13 containing the alpha-globin locus , we identified two additional low-frequency variants—HBM-HBA2 rs145546625 ( or its proxy HBM rs148323035 for MCH and MCV ) and the 3 . 8kb alpha-globin duplication ( for MCV ) —independently of the 3 . 8kb alpha-globin deletion . Several variants associated with RBC traits in the HCHS/SOL population are highly differentiated across ancestral populations . The HBB rs334 , HBB rs33930165 , esv2676630 alpha-globin 3 . 8kb gene deletion , and G6PD rs1050828 lead variants are derived from an African ancestral background , while the HFE hemochromatosis variant rs2032451 ( proxy of rs1799945 p . H63D ) is common among Europeans and Amerindian populations and much less common among Asians and West Africans . In addition , we note that the two newly reported independent association signals at the chromosome 16 alpha-globin locus—rs148323035/rs145546625 ( Table 3 ) and the 3 . 8kb duplication—appear to be more common among populations of Amerindian ancestry[20 , 21] . To assess whether any additional genomic regions might contain ancestrally differentiated SNPs associated with RBC traits , we performed a genome-wide admixture-mapping scan in HCHS/SOL for discovery analysis in each RBC trait . Admixture mapping in HCHS/SOL only detected associations already reported in the initial association testing: the chromosome 11p15 beta-globin region ( for MCV ) ; the chromosome 16p13 alpha-globin region ( for RBC , HGB , MCV , MCH , MCHC , and RDW ) ; and the RDW association on chromosome 14q11 , which corresponds to the PSMB5 association signal discovered in the HCHS/SOL GWAS ( S5 Fig ) . The PSMB5 index SNP shows large inter-continental allele frequency differences ( rs7147308 T allele frequency is 0 . 87 in AFR , 0 . 40 in SAS , 0 . 30 in EUR , 0 . 21 in AMR , and 0 . 06 in EAS 1000 Genomes populations ) .
We performed a GWAS of seven red blood cell traits in a diverse subsample of approximately 12 , 500 Hispanic/Latino participants of HCHS/SOL from across the continental U . S . We discovered and replicated three genome-wide significant variants ( SLC12A2 rs17764730 and PSMB5 rs941718 for RDW , and PROX1 rs3754140 for HCT ) . We also showed that common African ancestral hemoglobin variants ( beta-globin Hb S and Hb C missense variants rs334 and rs33930165 , and alpha-globin 3 . 8kb thalassemia structural variant ) and the African G6PD A- variant are associated with variation in RBC traits among the U . S . Hispanic/Latino population . Overall , 58% of previously identified GWAS loci for RBC traits generalized to HCHS/SOL . We additionally provide a more detailed characterization of allelic heterogeneity at the alpha- and beta-globin loci , including a newly identified Amerindian ancestral variant that overlaps a known regulatory region of the alpha-globin gene cluster . The HCT index SNP rs3754140 is located within a putative enhancer region positioned in the second intron of PROX1 and is in high LD ( r2 >0 . 8 ) with approximately 30 other intronic PROX1 variants ( S5 and S6 Tables ) . Some of these intronic proxy SNPs ( rs7541039 , rs7517701 , and rs4282786 ) occur within putative regulatory regions in erythroleukemia or proerythroblast cells , have CADD phred score >10 , and therefore represent likely functional candidates . All three of these proxy SNPs are located in a putative enhancer element that exhibits DNaseI hypersensitivity in fetal proerythroblasts and K562 cells . Although enhancers can have distal target genes , a potential target is the enhancer-harboring gene PROX1 , which has been reported as a negative regulator of hematopoietic stem cell renewal and for which mutations have been found in hematopoietic cell lines and primary blood malignancies[22 , 23] . PROX1 encodes Prospero Homeobox 1 , a widely expressed transcription factor involved in the development and differentiation of tissues such as endothelial lymphatic vessels , liver , retina , and pancreas[24] . Several PROX1 variants ( e . g . , rs340874 , rs340839 ) located in the 5’ UTR of PROX1 or adjacent antisense noncoding RNA have been associated with metabolic traits such as fasting glucose , insulin resistance , diabetes , and triglyceride levels[25–27] . The HCT-associated signal we detected in Hispanics/Latino is independent of the previously reported PROX1 metabolic trait association signal . Molecular analysis , including biallelic deletion of a 700bp region surrounding rs7541039 in the second intron of PROX1 , showed no effect on transcription of PROX1—which does not appear to be expressed in human erythroid precursors—or neighboring genes SMYD2 and CENPF[28] . In light of this information , further investigation of the role of the putative PROX1 intronic regulatory region and associated genetic variants in hematopoiesis—specifically RBC production—is warranted . The RDW-associated locus on chromosome 14q11 is located in a gene-rich region . The lead SNP rs941718 and several LD proxies are non-coding variants within or near PSMB5 , which encodes a 20S core proteasome subunit . From the standpoint of RBC biology , the ubiquitin proteasomal system may be particularly important during erythroid maturation and hemoglobin synthesis to control globin-chain balance and limit potential toxicities of unstable free globin chains[29] . The lead SNP rs941718 is also a blood cis-eQTL for nearby genes HAUS4 , MRPL52 , PRMT5-AS1 , and PRMT5 [30 , 31] and has a CADD phred score of 15 . 8 ( S5 and S6 Tables ) . PRMT5 encodes an arginine methyltransferase involved in binding to the γ-globin promoter and silencing fetal hemoglobin expression , and therefore represents an additional potential mechanism for influencing RBC phenotype[32 , 33] . The LD proxy rs11846575 , located just 3’ of PSMB5 , is proximal to a highly tissue-specific erythroid enhancer[34–36] and therefore merits further functional experimentation in the context of erythroid development and hemoglobin synthesis . The other newly reported RDW-association signal is located on chromosome 5q23 and spans ~100kb including SLC12A2 and an upstream long non-coding RNA ( LINC01184 ) on the antisense strand . SLC12A2 ( which codes for the protein NKCC1 ) is a sodium- , potassium- , and chloride-ion transporter membrane protein involved in cell-volume regulation and maintenance in kidney , RBC , and other cell types[37] . Genetic variation in other RBC membrane ion-transport proteins ( e . g . , PIEZO1 , SLC4A1 ) has been associated with inter-individual variability in RBC traits[13] . The lead SNP at the SLC12A2 locus ( rs17764730 ) lies within an exon of LINC01184 . RNA-Seq data indicates that both SLC12A2 and LINC01184 are expressed in erythroblasts[35] . The lead SNP is in high LD ( r2 >0 . 8 ) with 23 other variants spanning SLC12A2 and LINC01184 ( S5 and S6 Tables ) . The strongest functional candidate SNP ( rs3812049 , imputation quality score 1 . 006 , r2 to lead SNP = 0 . 89 ) is located within a bi-directional promoter region between the 5' ends of SLC12A2 and LINC01184 . Rs3812049 is also positioned within an erythroid DNaseI hypersensitive region and is occupied by multiple transcription factors , including the erythropoietic transcription factors GATA1 and TAL1 in erythroblasts and EGR1 in K562 cells . These observations suggest the possibility that the antisense transcript may be involved in erythrocyte maturation or maintenance by regulating SLC12A2 in erythrocytes . While this paper was under review , additional variants in the region of SLC12A2 and LINC01184 were reported to be associated with RDW in a predominantly European samples[38 , 39] . In human erythroid progenitor cells , we showed that small deletions in the bi-directional promoter region , including directly overlapping the position of rs3812049 , lead to reduced expression of both SLC12A2 and LINC01184 . Although formally demonstrating the function of the underlying element , these results could be consistent with a model in which rs3812049 alleles differentially modulate promoter activity . While disruption of the bi-directional promoter element did not reveal any differences in erythroid development , in vitro conditions may incompletely model a complex trait like RDW that appears highly dependent on appropriate RBC maturation and clearance in vivo . Finally , it is interesting to note both the large allele-frequency differences of the SLC12A2 index variant between African and non-African populations ( Table 1 ) and a report of lower erythrocyte NKCC1 protein activity in African Americans compared to whites[40] . This is particularly noteworthy given the established role of NKCC1 in blood pressure regulation , kidney function , and RBC-volume maintenance , and ethnic differences among these traits[37] . Based on our preliminary molecular results , both SLC12A2 and LINC01184 should be examined further for their potential roles in erythrocyte and non-erythroid traits . The HCHS/SOL cohort represents a diverse subsample of Hispanics/Latinos across the U . S . , with varying admixture proportions of three continental ancestry groups: Amerindians , Africans , and Europeans . The beta-globin hemoglobin S and hemoglobin C variants , alpha-globin 3 . 8kb deletion , and G6PD A- variant have previously been shown to contribute to RBC phenotypic variance among U . S . African Americans[41 , 42] . Here , we establish that these same common African ancestral hemoglobin and G6PD gene variants are associated with quantitative RBC phenotypes among U . S . Hispanics/Latinos . The heterozygous states of each of these inherited RBC conditions are prevalent in populations in Africa , Asia , southern Europe , and South and Central America , and confer a survival advantage against malaria[43] . Even though carriers are generally without clinical sequelae , the heterozygous state of Hb C can induce RBC dehydration , resulting in a higher MCHC[44] . Alpha-globin deletion carriers[1] and sickle cell trait carriers[45] may have lower levels of HCT , MCV , and MCH , and higher RBC counts , due to ineffective erythropoiesis . We also show that the HFE p . H63D variant ( rs1799945 ) is associated with RBC phenotypes in Hispanics . Both C282Y and H63D hemochromatosis mutations are prevalent in Northern Europeans , while H63D appears more broadly in North Africa , the Middle East , and less commonly in Asia . Emigration from Europe over the past 500 years likely introduced C282Y and H63D to Americas and Oceania , leading to a frequency of H63D in Amerindians and Hispanics/Latinos exceeds that of East and South Asians[46 , 47] . At the alpha-globin locus , the 3 . 8kb deletion and duplication generally arise as a result of misalignment of homologous sequences within HBA1 and HBA2 and unequal crossing over during recombination . In U . S . Hispanics/Latinos , we observed that the 3 . 8kb alpha-globin duplication was significantly associated with lower MCV independently of the 3 . 8kb deletion . This may be due to imbalanced alpha/beta globin-chain synthesis , which may be exacerbated by co-inheritance of other globin gene mutations[48] . Nonetheless , given the caveats of structural variant calling from genotype data , this finding requires additional validation using other molecular techniques . We observed additional allelic heterogeneity at the alpha-globin locus , a novel association signal for MCV and MCH with two Amerindian ancestral variants in high LD ( r2>0 . 99 ) : the HBM splice-site variant rs148323035 , and rs145546625 , located ~2 kb upstream of HBA2 . HBM encodes hemoglobin mu , a globin chain similar to the oxygen high-affinity delta-globin found in reptiles and birds that is transcribed in a tightly regulated fashion in erythroid cells , particularly during the terminal differentiation stage[49] . The HBM splice donor variant rs148323035 overlaps with a putative regulatory region that spans the transcription start site and first intron of HBM and is DNase hypersensitive , occupied by GATA1 and TAL1 in pro-erythroblasts[49] . Overall , generalization analysis revealed that 58% of RBC trait associations identified in GWAS of European- , Asian- , or African-descent populations generalized to HCHS/SOL Hispanics/Latinos . Nearly half of the previously reported genomic regions associated with RBC traits also had at least one variant associated one or more RBC traits in the HCHS/SOL , and 79% of individual SNPs previously reported as significant for more than one RBC trait generalized to HCHS/SOL for at least one of the previously reported traits . These results demonstrate that the same loci are likely involved in RBC trait biology across global populations , whether the functional variants are shared with or differ between ancestral groups . Failure to generalize can occur for one of several reasons , including but not limited to: ( 1 ) coverage of the relevant locus on the genotyping array is insufficient for the study population; ( 2 ) the originally published variant was a false positive and that locus is not associated with the relevant trait; ( 3 ) the power for generalization in HCHS/SOL is low due to the HCHS/SOL study population size; or ( 4 ) the power for generalization in HCHS/SOL may be low due to allelic frequency differences between populations . In summary , we report three novel loci associated with RBC traits in Hispanics/Latinos as well as independent signals within two RBC trait-associated regions previously identified in African descent populations . This includes an Amerindian ancestral variant at the alpha-globin gene cluster that overlaps a known alpha-globin regulatory region . This particular variant is monomorphic among European , Asian , and African ancestral populations . Other Amerindian-specific loci for platelet count or diabetes have been identified among Hispanics/Latinos[50 , 51] . These findings emphasize the importance of performing genetic studies in Hispanic/Latino populations .
The HCHS/SOL is a cohort of 16 , 415 self-identified Hispanic/Latino persons aged 18–74 years who were selected from households and census block groups in Chicago , IL , Miami , FL , Bronx , NY , and San Diego , CA , as previously described[52] . Study participants self-identified as having Hispanic/Latino background in one of six sub-groups , with the total study population including 6 , 471 participants identifying as having a Mexican background , 2 , 728 as Puerto Rican , 2 , 348 as Cuban , 1 , 730 as Central American , 1 , 460 as Dominican , and 1 , 068 as South American . Individuals were recruited to HCHS/SOL between 2008 and 2011 , and underwent a baseline clinical exam that included clinical , lifestyle , and sociodemographic assessment[53] . Based on kinship coefficient among the genotyped individuals , the HCHS/SOL sample includes 204 parent-offspring trios , 1 , 042 parent-offspring duos , 699 full-sibling pairs , and numerous second- and third-degree relatives . The IRB committees for the HCHS Coordinating Center at UNC Chapel Hill , San Diego State University , University of Illinois at Chicago , University of Miami , and Yeshiva University-Albert Einstein College of Medicine have all reviewed and approved the informed consent documents and study protocol . Written and signed informed consents in the language preferred by the participants are administered and archived at each of the participating field centers . All participants in this publication from HCHS/SOL have consented to use of their genetic and non-genetic data . Anyone not providing consent has been excluded from this analysis . Demographic characteristics and RBC trait descriptive statistics for included study populations are presented in S2 Table . Whole blood ( approximately 58 to 76ml ) was collected at Visit 1 for all consenting HCHS/SOL participants by certified technicians trained at their respective field-center institutions . Supplies and procedures were standardized across all field centers; 4ml of whole blood for complete blood count ( hemogram ) was collected in a tube containing EDTA as an anticoagulant . CBC values were measured from whole blood using an automated hematology analyzer ( Sysmex XE-2100 , Sysmex America , Inc . , Mundelein , IL 60060 ) at the central laboratory at the University of Minnesota Medical Center , Fairview , in Minneapolis . Of the 16 , 415 individuals in the HCHS/SOL cohort study , 12 , 803 consented to genotyping and passed QC . Several individuals from the genotyped subset were excluded from the analysis , including individuals with predominantly Asian ancestry ( n = 19 ) , pregnant women ( n = 8 ) , participants with >5% immature granulocytes ( n = 2 ) , end-stage kidney disease ( n = 46 ) , hematologic cancer ( n = 28 ) , or those undergoing cancer chemotherapy ( n = 54 ) . After exclusions , a total 12 , 502 participants were included for HCT , HGB , RBC , MCH , and MCV; 12 , 501 for RDW; and 12 , 500 for MCHC . HCHS/SOL subjects who consented to genetic studies had DNA extracted from whole blood , which was genotyped on the Illumina SOL HCHS Custom 15041502 B3 array . This array comprised the Illumina Omni 2 . 5M array ( HumanOmni2 . 5-8v1-1 ) and additional custom content[51 , 54] . In order to capture more Amerindian variation , the Omni2 . 5M array was modified by the addition of custom content comprised of ~150K SNPs selected from the CLM , MXL , and PUR 1000 Genomes Phase I samples for higher informativeness to identify Amerindian continental ancestry and for higher frequency in Amerindian genomic segments . Standard quality assurance/quality control ( QA/QC ) methods for SNP- and sample-level quality were applied . Quality metrics used to filter SNPs included Illumina/LA Biomed assay-failure indicator , missing call rate ( >2% ) , deviation from Hardy-Weinberg equilibrium ( p<10−5 ) , Mendelian errors ( >3 in 1343 trios or duos ) , and duplicate sample discordance ( >2 in 291 sample pairs ) . Following genotyping QA/QC procedures , there were 12 , 803 unique study participants and 2 , 232 , 944 SNPs available for imputation . For imputation , we used 1000 Genomes Project phase 1 reference panel and IMPUTE2 software . Genotypes were initially pre-phased using SHAPEIT2 ( v2 . r644 , www . shapeit . fr ) , and subsequently imputed using IMPUTE2 software ( v2 . 3 . 0 , https://mathgen . stats . ox . ac . uk/impute/impute_v2 . html , last accessed Dec 2016 ) [54] . Only variants with at least two copies of the minor allele present in any of the four 1000 Genomes continental panels were imputed , yielding a total of 25 , 568 , 744 imputed variants ( SNPs and indels ) . Imputed genotype dosages were modeled on a continuous scale from 0 to 2 in order to account for genotype uncertainty . Oevar is an imputation quality metric , defined as the ratio of the observed variance of imputed dosage to the expected binomial variance . Variants with an oevar <0 . 3 were considered low quality and excluded from analysis . Additional information about imputation and quality metrics is found in Conomos , et al[55] . The SOL Illumina Omni 2 . 5M array contains five variants ( rs2362744 , rs4021971 , rs4021965 , rs11639532 , rs2858942 ) within the 3 , 811bp alpha-globin structural variant that can be used for determining copy number . Raw probe intensity data ( normalized X and Y values ) were exported from GenomeStudio as FinalReport files and then imported into the Genvisis software package ( http://genvisis . org , last accessed Jan 2017 ) in order to use its specialized CNV calling algorithm . The first step in the process is to re-compute the Log R ratios ( LRRs ) using centroids derived from only high-quality samples ( standard deviation of the autosomal LRRs <0 . 32 and genotype call rate >98% ) . LRRs from a set of ~50 , 000 curated markers were included in a principal components analysis ( PCA ) to capture DNA quality , DNA quantity , and batch effects . After regressing out 60 PCs from the raw intensity data , we recomputed LRRs and determined the median LRR value for the five markers in the alpha-globin region . Copy-number ( 0 , 1 , 2 , 3 , or 4 ) calling for the structural variant was then performed after visual inspection of the cluster boundaries with median LRR on the x-axis and median absolute difference on the y-axis . For RBC phenotype association analyses , genotypes were then coded and analyzed separately for the presence of the 3 . 8kb alpha-globin deletion ( 0 , 1 , or 2 copies ) and the presence of the 3 . 8kb alpha-globin duplication ( 0 , 1 , or 2 copies ) . For replication of discovery associations in HCHS/SOL , 1000 Genomes Project phase 1-imputed GWAS data were utilized from three Hispanic/Latino study populations . These included the Women's Health Initiative ( WHI ) SNP Health Association Resource ( SHARe ) project ( n = 3 , 454 ) , the Multi-Ethnic Study of Atherosclerosis ( MESA ) cohort ( n = 782 ) , and Mount Sinai BioMe biobank ( n = 2 , 854 ) [56] . Genotyping in WHI-SHARe and MESA was performed using Affymetrix 6 . 0 array and imputation was performed with MaCH software[57] . BioMe was genotyped using the Illumina OmniExpressExome beadchip array , phasing was performed using ShapeIt Version 2 release 644 and imputation with Impute version 2 . 3 using the All 1000 Genomes Project phase 1 integrated variant set ( Aug 2012 ) as the reference . All outcomes were analyzed using linear mixed-effect models ( LMMs ) , with random effects accounting for inter-individual correlation ( due to either relatedness , shared household , or census block group ) . The covariates ( fixed effects ) included age , sex , five principal components , recruitment center , current cigarette smoking , sampling weight , and genetic analysis group ( Cuban , Dominican , Puerto Rican , Mexican , Central American and South American ) [54] . When performing analysis on the X chromosome , we included the first two X chromosome-specific principal components as covariates . Additionally , pairwise genetic relatedness as estimated from the X chromosome was included as a random effect along with the autosomal genetic relatedness matrix . Additionally , since males have only one copy of X chromosome , genotypes on the X chromosome were coded 0 , 1 , 2 for females and 0 , 2 for males . We also conducted three additional analyses for the known G6PD locus on the X chromosome: ( 1 ) sex-stratified analysis ( S10 Table ) ; ( 2 ) genotype-specific analysis in women , since there is evidence for skewed X chromosome-inactivation with age[58] ( S11 Table ) ; and ( 3 ) age-genotype interaction analysis ( S12 Table ) . More information about the principal components , kinship matrix computation , and the genetic analysis groups , is provided in Conomos , et al[54] . Potential inflation was assessed using quantile-quantile plots of the test statistics against the standard normal distribution , and a calculated inflation factor λgc . We report genome-wide significant results at significance threshold of p-value ≤5 . 0x10-8 and suggestive significance threshold of p-value <1 . 0x10-7 in the HCHS/SOL discovery sample for all variants with MAF = 0 . 01 and imputation oevar >0 . 3 All SNPs exceeding genome-wide significance threshold of p-value <1x10-7 are described in S9 Table . Local ancestry estimates were previously inferred in the HCHS/SOL[59] . A genome-wide admixture mapping scan was performed using a linear mixed model with covariates and random effects described above , jointly testing the three ancestries ( European , African , Amerindian ) at each available locus . On the basis of previous simulation results , a nominal p-value of 5 . 7x10-5 yielded a genome-wide type I error of 0 . 05 . There are currently no well-developed , validated methods available for local ancestry estimation on the X chromosome . Hence , we performed admixture mapping analysis only on the autosomes . Association testing was performed in each of the three Hispanic replication data sets ( WHI , BioMe , MESA ) using linear regression and the same RBC trait transformation as the discovery samples , adjusted for age , sex , and principal components . Meta-analysis of results from the 3 independent Hispanic replication study samples was performed using the inverse-variance-weighted method implemented in METAL ( http://csg . sph . umich . edu//abecasis/Metal , last accessed Dec 2016 ) . We defined novel , replicated loci as those which exceeded a Bonferroni-corrected significance threshold of p <0 . 05/7 , or 0 . 0071 ( accounting for 7 SNPs carried forward for replication ) and are located >1 megabase ( Mb ) from a previously reported genome-wide significant association signal . We performed step-wise conditional analysis for each RBC phenotype to identify secondary , independent association signals within 500kb of known and newly discovered GWAS loci . In the first round of the conditional analysis for each trait , we used the same regression model as in the discovery GWAS , with additional adjustment for previously reported or novel variants identified in this study . The list of variants used in conditional analysis of each trait is provided in S13 Table . The significance threshold for discovering new , independent association signals was the same as the genome-wide discovery threshold ( α = 5 . 0x10-8 ) as well as MAF≥0 . 01 . Subsequent rounds of conditional analysis were repeated for each genomic region , adding the strongest genome-wide significant variant from the previous round as a covariate in the regression model , until no further genome-wide significant variants satisfying the MAF threshold remained in that region after covariate adjustment . The full models for each trait used in the final round of conditional analysis are listed in S14 Table . After obtaining probe intensity-based CNV calls for the 3 . 8kb alpha-globin CNV , we conducted conditional analysis on chromosome 16 using the calls from the re-typed CNV . The full models for each trait used in these conditional analysis are also listed in S14 Table . Conditional analysis with the re-typed 3 . 8kb alpha-globin CNV was conducted on the subset of 12 , 390 individuals for whom the re-typed CNV calls were available . Variants used in generalization analyses were identified by one of two inclusion methods: ( 1 ) any variant listed as genome-wide significant for any of the seven RBC traits in our study in the European Bioinformatics Institute GWAS catalog ( http://www . ebi . ac . uk , last accessed Jan 2017 ) ; or ( 2 ) any RBC trait genome-wide-significant variants published in the main text or supplement of an English-language GWAS indexed in PubMed prior to December 2016 . ( Of note , we did not identify any GWAS published in a language other than English , hence we expect our list of variants identified using these methods to be complete prior to 2017 . ) We tested each published RBC-associated variant to see whether that association generalized to Hispanics/Latinos . The directional generalization null hypothesis is rejected if there is enough evidence that the published variant is directionally consistent and associated with the outcome in both the discovery study and HCHS/SOL . We evaluated for generalization all available signals previously reported in any GWAS published in English , for all seven traits evaluated in this paper ( S4 Table ) . Most of these SNPs were reported in studies of adults of European ancestry , but we also generalized associations from African- and Japanese-ancestry populations . No variants identified in Danjou , 2015 , were included in our genotyped or imputed dataset and hence these variants could not be evaluated for generalization[60] . To test the generalization null hypotheses , we computed directional FDR r-values for each of the tested SNPs . Directional r-values were calculated based on one-sided p-values from both the “discovery” study ( reported in the literature ) and the HCHS/SOL , and based on the number of tests performed in the discovery study , in order to properly account for multiple testing . A SNP was considered generalized if its r-value was <0 . 05[61] . In generalizing associations reported by Ganesh et al . ( 2009 ) , we did not employ directional control since Ganesh , et al . , ( 2009 ) did not report effect sizes or directions[8] . The implication is a slight loss of power . Generalization analysis was performed by looking up reported SNPs in HCHS/SOL results , in an analysis that mimics the analysis reported in the discovery study . For example , if a trait was reported as an association analysis with the natural-log-transformed trait , we performed the analysis with the same transformation in the HCHS/SOL population . In some cases , as with Kamatani , et al . ( 2010 ) , we also matched effect-size reporting methods ( standard deviations ) for ease of comparison . Transformations , when applicable , are described in S1 Table . Since the same SNP-trait association may be reported by multiple studies , we counted only unique SNP-trait associations . In instances where more than one study reported associations for the same SNPs and trait , but used different trait transformations , we selected the results from the generalization analysis in which the trait transformation matched our primary analysis . Since some SNPs are associated with more than one RBC trait , and some genomic regions contain multiple SNPs associated with multiple traits , we summarize the generalization results as follows . Overall , we summarize the number of generalized unique trait-SNP associations ( the same SNP may be counted more than once , if associated with more than one trait ) . Then , for each trait , we summarized ( 1 ) the number of unique SNPs , and ( 2 ) the number of unique genomic regions . To define genomic regions , we identified specific SNPs , and a 1Mb genomic region around them . Other SNPs within these regions were clumped together . We say that a genomic region generalized for a specific trait if at least one SNP in the region was associated with the trait . We assessed any novel , replicated red blood cell associated loci to determine potentially causal variants . At each locus , we determined if the lead or proxy variants ( r2 ≥ 0 . 8 ) were located within putative erythroid regulatory elements , defined on the basis of enrichment for various histone-modification and ChIP-Seq signals in either erythroblasts or the erythroleukemia cell line K562[34–36] . We defined these regulatory regions as follows: enrichment for histone H3K4me1 as an enhancer , enrichment for histone H3K4me3 as a promoter . Variants located within a putative promoter or enhancer , and that overlapped a DNaseI hypersensitive site in proerythroblasts or K562 cells , were prioritized as putatively functional [34 , 36 , 62] . Regulatory elements often are bound by transcription factors and hence we report ChIP-Seq peak overlaps of key erythroid transcription factors ( GATA1 , TAL1 ) , and others in proerythroblasts and K562 cells to provide further support for the functional role of putative regulatory elements in erythroid cells[34 , 36 , 62] . The ENCODE and BLUEPRINT datasets were accessed through the ENCODE analysis Hub and Blueprint Hub respectively via the UCSC genome browser[63 , 64] . Datasets from Xu , et al , were accessed from codex ( http://codex . stemcells . cam . ac . uk , last accessed Dec 2016 ) [62 , 65] . To hypothesize likely mode of action via which the causal variants influence the trait , we report eQTL targets and or motifs disrupted by prioritized variants using HaploReg v4 . 1[66] . All the datasets used for functional annotation were mapped to Human GRCh37/hg19 assembly . Functional annotation is summarized in S5 Table . We also used in silico prediction algorithms to annotate variants . These included RegulomeDB , the Combined Annotation Dependent Depletion ( CADD ) phred score , GWAVA , and deltaSVM[67–70] . These annotations are summarized in S6 Table . The CRISPR/Cas9 system was used to mutagenize individual variants or small regions of interest identified during discovery analysis and subsequent bioinformatics interrogation . All oligonucleotide sequences used in CRISPR-Cas9 genome editing experiments are listed in S7 Table . The human umbilical cord blood derived erythroid progenitor cell line #2 ( HUDEP-2 ) was cultured and used for genome editing as previously described[71] . Individual and tandem pairs of single chimeric guide RNAs were cloned to lentiviral expression vectors ( lentiGuide-Puro , Addgene plasmid 52963 ) . Cells were transduced and selected for lentiviral integrants by antibiotic selection ( 10 μg/ml blasticidin for lentiCas9-Blast [Addgene plasmid 52962] , 1 μg/ml puromycin for lentiGuide-Puro ) . For SLC12A2 individual sgRNA promoter editing , indel frequencies were assessed after 7 days by nested PCR followed by amplicon deep sequencing . For SLC12A2-LINC01184 , PSMB5 , and PROX1 interstitial deletions , cells were plated at limiting dilution to isolate clones 7 days after transduction with tandem sgRNAs . Clones with biallelic deletions were characterized by presence of gap PCR amplification with primers outside the deleted segment and absence of PCR amplification from inside the deleted segment . Expression of mRNA of genes of interest was compared to GAPDH expression using quantitative reverse transcription PCR ( RT-qPCR ) in control and edited HUDEP-2 cells . For SLC12A2 individual sgRNA promoter editing , the total population of edited cells was evaluated in bulk by RT-qPCR . For SLC12A2-LINC01184 , PSMB5 , and PROX1 interstitial deletions , clones were first identified by PCR screening and then evaluated by RT-qPCR . For differentiation experiments , control and edited HUDEP-2 cells were cultured separately for 4 days in Erythroid Differentiation Media ( EDM ) with Iscove’s Modified Dulbecco’s Medium ( IMDM ) ( Life Technologies ) supplemented with 330 mg/ml holo-transferrin ( Sigma ) , 10 mg/ml recombinant human insulin ( Sigma ) , 2 IU/ml heparin ( Sigma ) , 5% human solvent detergent pooled plasma AB ( Rhode Island Blood Center ) , 3 IU/ml erythropoietin , 100 ng/ml human SCF , ( R&D ) , 1 mg/ml doxycycline , 1% L-glutamine , and 2% penicillin/streptomycin . Subsequently the cells were cultured an additional 4 days in EDM lacking SCF , and then an additional 4 days in EDM lacking both SCF and doxycycline . Erythroid maturation was evaluated by flow cytometry staining with CD71 ( eBiosciences ) , CD235a ( eBiosciences ) , CD49f ( Miltenyi ) , and DRAQ5 ( eBiosciences ) as well as morphology by May-Grunwald-Giemsa staining , Student's t-tests were used for statistical analysis of results . Genotype data and GWAS results of discovery analysis of all the seven RBC traits can be requested via dbGaP study accession phs000880 . Phenotype data can be requested via dbGaP study accession phs000810
Supplemental data includes five figures , nine tables , and five Excel spreadsheets . | Red blood cells ( RBC ) are important for transport of oxygen to tissues throughout the body . Distribution of RBC traits differs by ethnicity and gender , and both genetic and acquired factors likely contribute to these differences . Prior genetic studies have identified physical regions of the genome associated with RBC traits in populations with European , African , and Asian ancestry . These studies have not included individuals with ancestry from the American continents ( Amerindian ancestry ) , such as Hispanics/Latinos . In an analysis of RBC traits in up to 19 , 608 Hispanics/Latinos , we identified an Amerindian-ancestry genetic association in a known alpha-globin regulatory region . We also identified three new RBC trait associations , including a regulatory variant of SLC12A2 that encodes a RBC membrane ion-transport protein . Experimental disruption of this regulatory element led to reduced expression of both SLC12A2 and an adjacent long non-coding RNA in human erythroid progenitor cells . These results contribute to understanding the physiology of red blood cells and reinforce the importance of genetic study of diverse ancestry populations , in particular Hispanics/Latinos . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods",
"Supplemental",
"data"
] | [
"blood",
"cells",
"genome-wide",
"association",
"studies",
"quantitative",
"trait",
"loci",
"alleles",
"red",
"blood",
"cells",
"genome",
"analysis",
"hemoglobin",
"sex",
"chromosomes",
"animal",
"cells",
"proteins",
"chromosome",
"biology",
"x",
"chromosomes",
"genetic",
"loci",
"biochemistry",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"genomics",
"computational",
"biology",
"chromosomes",
"human",
"genetics"
] | 2017 | Genome-wide association study of red blood cell traits in Hispanics/Latinos: The Hispanic Community Health Study/Study of Latinos |
Hookworm-related cutaneous larva migrans ( HrCLM ) is a neglected tropical skin disease associated with significant clinical pathology . Little knowledge exists about prevalence and risk factors of HrCLM in endemic regions . To understand the epidemiology of HrCLM in Amazonia , we conducted a cross-sectional study in a resource-poor township in Manaus , Brazil . HrCLM was diagnosed in 8 . 2% ( 95% CI , 6 . 3–10 . 1% ) of the study population ( N = 806 ) with a peak prevalence of 18 . 2% ( 95% CI , 9 . 3–27 . 1% ) in children aged 10–14 . Most of the tracks ( 62 . 4% ) were located on the feet , and 10 . 6% were superinfected . HrCLM was associated independently with age under 15 , male sex , presence of animal faeces on the compound , walking barefoot on sandy ground and poverty . HrCLM is common in resource-poor communities in Amazonia and is related to poverty . To reduce the disease burden caused by HrCLM , living conditions have to be improved .
Hookworm-related cutaneous larva migrans ( HrCLM ) is a parasitic skin disease caused by the penetration of feline or canine hookworm larvae into the human epidermis . The most frequent species are Ancylostoma braziliense , Ancylostoma caninum and Uncinaria stenocephala [1–3] . In humans , the larva is unable to cross the basal membrane of the epidermis and migrates in the compartment of the epidermis until it dies spontaneously after a few weeks to several months [1 , 4 , 5] . The migration of animal hookworm larvae causes a typical elevated erythematous linear or serpiginous track known as “creeping eruption” [6] . HrCLM is associated with intense pruritus and significantly impairs the quality of life [7] . The resulting scratching leads to denudation of the skin , which facilitates bacterial superinfection of the lesion [1 , 8 , 9] . Additional skin injury may be caused by inappropriate surgical manipulation of the lesion and treatment with toxic substances [10] . Whereas animal hookworm species parasitize dogs and cats worldwide [11] , HrCLM is mainly seen in tropical and subtropical areas in South America , the Caribbean , Africa and South-East Asia [11–14] . Sporadic cases have been reported for Europe [15–20] . In semi-arid north-eastern Brazil , prevalence ranged from 0 . 2% to 4 . 4% in the general population and from 0% to 14 . 9% in children <5 years [21–23] . No population based data exists for other endemic areas . Known risk factors are male sex , young age , living in a house without a solid floor and barefoot walking [8 , 23] . An association with low income has been suspected [23] . In order to investigate the epidemiology of HrCLM in Amazonia and to develop sustainable means of control , in a first step we determined prevalence and risk factors in a resource-poor community in the outskirts of Manaus . Data of a spatial analysis will be published separately .
The study was conducted in Manaus , capital of Amazonas State , North Brazil . Manaus is situated at 03°06' south latitude and has a hot humid climate . The average annual precipitation is 2307mm and the mean annual temperature is 26 . 7°C ( International Institute of Meteorology of Brazil , http://www . inmet . gov . br/portal/index . php ? r=clima/normaisclimatologicas ) . The study area is part of Nova Vitoria , a resource-poor neighbourhood at the outskirts of Manaus . The boundaries of the study area are defined on three sides by an igarapé , a small affluent of the Amazon River . On the fourth side a paved road separates the study area from Grande Vitoria , another resource-poor community . The study area is characterized by unpaved roads , absence of public health facilities , kindergartens or public schools . There was no sewage disposal system at the time of the study . Electricity was available but only half of the households were legally connected to the grid; the other half used hand-made wire connections . Drinking water was distributed via rubber hoses , which often flooded the streets . Many cats and dogs strayed around in the streets and gardens . Children usually played on the compound of the house , in the streets or on improvised football fields . Hence , the study area was representative for the innumerable poor neighbourhoods at the periphery of Manaus . As a first step into a comprehensive series of investigations on the epidemiology of HrCLM in Amazonia , we conducted a cross-sectional study in Nova Vitoria in April 2009 , at the end of the rainy season . First , a census of all households and inhabitants was performed . During a door-to-door survey , households were GPS-mapped and environmental , socio-economic and behaviour-related risk factors were documented using a pre-tested , structured questionnaire . Inclusion criteria were residency in the study area for more than two months and provision of an informed , written consent . All participants were examined clinically for HrCLM . The examination took place in the house where the family lived , in a room where privacy was guaranteed . The genital area was spared in case of absence of symptoms such as itching . HrCLM was diagnosed clinically by two investigators ( DP and FR ) when the characteristic slow-moving , elevated linear or serpiginous tracks were present [1 , 6 , 7 , 11–13 , 24] . Lesions were counted and the appearance and location of the tracks were documented . Each track was defined as a single lesion . Bacterial superinfection was diagnosed when pustules or suppuration were visible . The study was approved by the Ethical Committee of the Fundação de Medicina Tropical- Amazonas ( FMT-AM ) . Informed , written consent was obtained from each participant or in the case of minors from their legal guardian . Each affected inhabitant of Nova Vitoria was offered free treatment independently of the participation in the study . Treatment consisted of ivermectin ( Ivermec , Uci-farma , São Paulo , Brazil ) given as single oral dose ( 200 μg/kg ) or—in the case of children <5 years or <15 kg and women with suspected or confirmed pregnancy—of topically applied thiabendazole ( 5%; Tiadol , Bunker Indústria Farmacêutica Ltda . , São Paulo , Brazil ) 3 times a day for one week . Data were entered in Microsoft Office Access 2007 , cleaned for entering errors and analysed using PASW Statistics Version 18 . 0 ( SPSS Inc . , Chicago , USA ) . Missing data were assumed to be missing at random and flagged up in the analysis . Only complete cases were analysed . An asset index was formed using principal component analysis ( PCA ) to categorize households according to socio-economic status . First , a set of assets that reflect wealth was identified . From this set of assets , we selected items with a high inequity in distribution among the households and a high eigenvalue [25] . Included assets were presence of a car , television , fridge , type of house construction , legal connection to electricity and monthly mobile phone costs . Using these assets , an index ( “wealth score” ) was built based on the respective value of each item in the PCA [25] . Households were ranked and divided into tertiles representing a high , intermediate or low socio-economic status . Income was categorized into three categories with the official minimum wage ( R$ 465 per month in 2009 ) as a reference . A knowledge score was derived out of six questions concerning the etiology of HrCLM . Every correct answer added one point to the score . The knowledge score values were categorized in tertiles representing households with little knowledge ( 0–3 correct answers ) , moderate knowledge ( 4 correct answers ) and high knowledge ( 5–6 correct answers ) . Age groups were formed similar to previous population-based studies on HrCLM to allow comparison of the results [8 , 21 , 23] . For bivariable risk factor analysis , odds ratios ( OR ) were calculated together with 95% confidence intervals ( 95% CI ) . Statistical analysis consisted of χ²-test or Fisher-exact-test to compare relative frequencies and logistic regression for non-binary variables . For multivariable risk factor analysis , all variables that showed weak evidence of an association with HrCLM ( p<0 . 1 ) were entered into a stepwise logistic regression . We observed standard errors and 95% CI to identify multicollinearity and removed variables where necessary . A random effects model was used to control for clustering on household level .
According to the census 412 households existed in the study area , 127 of which were found without a resident present . Of the remaining 285 households , 5 ( 2% ) did not match the inclusion criteria and 18 ( 6% ) refused to participate . The remaining 262 households ( 92% ) were inhabited by a total of 1104 people out of whom 806 ( 73% ) were present during sampling and were included in the study . Seventy-eight per cent of the adults were unemployed or working in the informal sector . Fifty-eight per cent of the households had one minimum wage ( R$ 465 per month ) or less at their disposition . The proportion of illiteracy in adults was at least 27% . Only 11 . 5% of the households had been visited by a community health worker within the last 12 months . Thirty-one per cent of the households stated that at least one case of HrCLM had occurred in household members within the last 12 months . ( Table 1 ) The median age was 13 years ( range 0–72 ) . The majority of the participants were females ( 59 . 3% ) . Sixty-six persons ( 8 . 2%; 95% CI , 6 . 3–10 . 1% ) had HrCLM with a total of 117 lesions . Clinical characteristics of the infected study participants are presented in Table 2 . Children aged 10–14 had the highest prevalence ( 18 . 2%; 95% CI , 9 . 3–27 . 1%; Fig 1 ) . In all age groups of children , boys were significantly more affected than girls ( p<0 . 001 ) . The feet were the most common localisation of HrCLM . Previous episodes of HrCLM were remembered of 18 . 7% of the participants . Following anamnestic information 39 . 7% had suffered of pediculosis capitis , 26 . 8% of tungiasis and 5 . 7% of scabies in the past year . Bivariable risk factor analysis showed that male sex , age younger than 15 , low family income , a low wealth score , playing football , practicing sport barefoot and presence of animal faeces on the compound were significantly associated with a high risk of HrCLM ( Table 3 ) . Those who reported to have had HrCLM in the last year had a significantly higher risk to be diagnosed with HrCLM in the cross-sectional study ( OR = 15; 95% CI , 8 . 5–26 . 7 ) . The highest risk was associated with the habit of always walking barefoot on sandy ground or soil ( OR = 23 . 4; 95% CI , 8 . 0–68 . 6 ) . Multivariable risk factor analysis ( Table 4 ) revealed that always walking barefoot on sandy ground or soil was the most important independent risk factor . Male sex , young age and presence of animal faeces on the compound remained independent risk factors for the presence of HrCLM . Obviously , HrCLM was significantly associated with poverty: A low wealth score of a household showed an adjusted odds ratio of 2 . 5 ( 95% CI , 1 . 1–5 . 8 ) .
Clinical features were similar to those reported by others [12 , 13] . Most of the tracks ( 62 . 4% ) were located on the feet , which reflects the fact that many people walked barefoot . This is consistent with our previous population-based study in rural Northeast Brazil [22] . The percentage of superinfected tracks was 10 . 6% . Previous studies in endemic areas by us and others reported similar proportions between 8 and 28% [8 , 21 , 22 , 30] . Unhygienic living conditions and practices as well as limited access to healthcare may explain the higher proportion of superinfected HrCLM in our study than usually seen in travellers [10 , 11] . The overall prevalence of 8 . 2% ( 95% CI 6 . 3–10 . 1% ) found in this study is the highest ever documented in a population-based study . Previous population-based studies in Northeast Brazil showed an overall prevalence between 0 . 2% and 4 . 4% during the dry and the raining season , respectively [8 , 21 , 23] . Similar to previous studies , the prevalence differed by age group and sex with a peak prevalence of 25 . 6% in 10–14 year old boys ( Fig 1 ) [8 , 23] . Whether there is a seasonal variation in HrCLM prevalence in the Amazonas region , where the climate is hot and humid throughout the whole year , remains to be clarified . Outside Brazil only one prevalence study has been conducted on devotees of a temple in Sri Lanka . Fifty-eight per cent of the devotees were found to have HrCLM; however , it is doubtful whether this finding reflects the true overall prevalence in that area since the participants were examined after a special religious ritual increasing the odds for exposure [30] . The extremely high prevalence found in our study indicates excellent conditions for the completion of the off-host cycle of animal hookworm in Nova Vitoria . First , many stray dogs and cats roam in the community and act as animal reservoirs . There is no public veterinary service at all and pets are not treated against intestinal helminths . Animal faeces were present on 11 . 8% of all compounds , and faecal material littered many public areas . Second , hookworm eggs require an environment that protects them from desiccation to evolve into infective third stage larvae [31] . Manaus is located in the middle of the Amazon basin . The precipitation in the month preceding the study was around 230 mm with 20 days of rain ( International Institute of Meteorology of Brazil ( INMET ) ) . All streets and most of the compounds in Nova Vitoria were unpaved and became muddy after heavy rainfall . Furthermore , the average temperature never falls below 25°C . This means that the environmental conditions are exceptionally favourable for the propagation of animal hookworm larvae [5] . And third , risky behaviour with prolonged contact to contaminated soil was frequent . Many children did not go to school but roamed through the streets and compounds the whole day , the majority walking barefoot at least part of the time . The multivariable model showed a complex pattern of risk factors with walking barefoot on sandy soil being most significant . This corroborates our previous findings from a semi-arid area of Brazil , where the lacking use of footwear was an independent risk factor [23] . For the first time we could show that the odds differed by the frequency protective footwear was used . Participants who always used shoes ran a lower risk of acquiring HrCLM than those wearing shoes sometimes ( Table 4 ) . Even the commonly used flip-flops ( plastic sandals , which consist of a thin rubber sole with a single string ) provided significant protection . However , closed shoes were worn regularly only by seven individuals . Obviously , HrCLM was predominantly acquired outdoors . Neither walking barefoot indoors , even if the floor consisted of sand or soil , nor owning a cat or dog were identified as independent risk factors . Assumedly , animal hookworm larvae were unable to complete the life cycle indoors because the floors were usually dry and accidentally dropped animal excrements were rapidly removed . It remains uncertain whether the infections predominantly took place peridomestically or in public areas , such as parks , as suspected in some outbreak investigations [32–34] . Our findings that the presence of faeces on the compound was an independent risk factor and that playing football on improvised playgrounds was not an independent risk factor indicate that peridomestic transmission is important . This study shows for the first time that low income and poverty-related living conditions are crucial risk factors for HrCLM . Hitherto , a low family income has been identified as a risk factor but didn´t reach statistical significance in the mulitvariate analysis . The concept of an asset index as a long-term indicator of the socio-economic status of the household has never been applied in earlier studies [8 , 23] . Even within a poor population , as in the community of Nova Vitoria , the relative level of poverty predicted the risk of acquiring HrCLM . A household income of one minimum wage or less was associated with a high risk of acquiring HrCLM . Also , a low wealth score was an independent risk factor . Hence , the poorest of the poor are the most vulnerable part of the population , which corroborates our hypothesis that occurrence of HrCLM is a proxy of the economic situation in a country [35] . Many neglected tropical diseases are considered to be associated with poverty [36 , 37] but HrCLM is particular in the sense that it affects the poorest of the poor . In contrast to other soil-transmitted helminths , HrCLM has a pure animal reservoir and thus treating the human population cannot influence the incidence of HrCLM . Veterinary anthelmintic therapy can be effective [38] but is hard to realise in areas lacking basic infrastructure even for human health . Therefore , disease control strategies have to point towards improvement of living conditions , environmental factors and protective behaviour . Preventing access of cats and dogs to playgrounds and informing the public about pet-associated health risks and protective shoewear will be essential to reduce the parasite burden in humans as long as infrastructure remains precarious [32–34 , 39 , 40] . For safety reasons Nova Vitoria could only be visited during daylight hours . Thus , there may have been a selection bias in favour of women and children staying at home versus adult males being at work . By means of an exhaustive sampling strategy , we still obtained a high participation and a representative sample of the daytime population . We have no reasons to believe that study participants with missing data differed from those without missing data and hence any missing observation reduced statistical power but is unlikely to have biased the results [41] . Confusion of HrCLM with other conditions that present as a creeping skin eruption such as gnathostoma , Strongyloides stercoralis ( larva currens ) , fly maggots ( migratory myiasis ) and scabies is theoretically possible [1 , 6 , 24] . However , a slightly elevated linear or serpiginous track and the slow velocity of progression with several millimetres to few centimetres per day are pathognonomic [6 , 42] . We therefore assume that all participants were correctly diagnosed . The study revealed the highest prevalence of HrCLM in a representative population sample known to date and showed transmission in peridomestic areas . We could prove that HrCLM is a disease of the poorest of the poor . It is therefore plausible that for elimination of HrCLM as a public health threat , it is necessary to improve the living conditions . | Hookworm-related cutaneous larva migrans ( HrCLM ) is a parasitic skin disease caused by the penetration of animal hookworm larvae into the human skin . In this compartment the larvae cannot pass the basal membrane and reproduce , but migrate in the outer skin layer for several weeks , causing skin inflammation and intense itching . Thus , humans are a biological impasse . Although HrCLM is a common skin disease in tropical and subtropical regions , studies on prevalence and risk factors are scarce . We clinically examined the population of a resource-poor neighbourhood in Manaus , capital of Amazonas State , Brazil , and investigated HrCLM-associated risk factors . HrCLM was very common with an overall prevalence of 8 . 2% . Children in general , and boys in particular , were most frequently infected . We could confirm that walking barefoot on sandy ground is a significant risk factor , and we identified the presence of animal faeces on the compound as another important predictive factor . Clearly , HrCLM was associated with low income and poverty-related living conditions . The poorest of the poor were identified as the most vulnerable population group . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"recreation",
"children",
"medicine",
"and",
"health",
"sciences",
"salaries",
"pathology",
"and",
"laboratory",
"medicine",
"sports",
"cutaneous",
"larva",
"migrans",
"geographical",
"locations",
"tropical",
"diseases",
"social",
"sciences",
"vertebrates",
"parasitic",
"diseases",
"animals",
"mammals",
"dogs",
"age",
"groups",
"developmental",
"biology",
"signs",
"and",
"symptoms",
"neglected",
"tropical",
"diseases",
"families",
"minimum",
"wage",
"larva",
"migrans",
"labor",
"economics",
"south",
"america",
"behavior",
"lesions",
"economics",
"brazil",
"people",
"and",
"places",
"helminth",
"infections",
"population",
"groupings",
"biology",
"and",
"life",
"sciences",
"sports",
"science",
"metamorphosis",
"larvae",
"organisms"
] | 2016 | Prevalence and Risk Factors of Hookworm-Related Cutaneous Larva Migrans (HrCLM) in a Resource-Poor Community in Manaus, Brazil |
Longquan City , Zhejiang province , China , has been seriously affected by hemorrhagic fever with renal syndrome ( HFRS ) since the first cases were registered in 1974 . To understand the epidemiology and emergence of HFRS in Longquan , which may be indicative of large parts of rural China , we studied long-term incidence patterns and performed a molecular epidemiological investigation of the causative hantaviruses in human and rodent populations . During 1974–2011 , 1866 cases of HFRS were recorded in Longquan , including 20 deaths . In 2011 , the incidence of HFRS remained high , with 19 . 61 cases/100 , 000 population , despite the onset of vaccination in 1997 . During 1974–1998 , HFRS cases in Longquan occurred mainly in winter , while in the past decade the peak of HFRS has shifted to the spring . Notably , the concurrent prevalence of rodent-borne hantaviruses in the region was also high . Phylogenetic analyses of viral sequences recovered from rodents in Longquan revealed the presence of novel genetic variants of Gou virus ( GOUV ) in Rattus sp . rats and Hantaan virus ( HTNV ) in the stripe field mice , respectively . Strikingly , viral sequences sampled from infected humans were very closely related to those from rodents . HFRS represents an important public health problem in Longquan even after years of preventive measures . Our data suggest that continual spillover of the novel genetic variant of GOUV and the new genetic lineage of HTNV are responsible for the high prevalence of HFRS in humans . In addition , this is the first report of GOUV associated with human HFRS cases , and our data suggest that GOUV is now the major cause of HFRS in this region .
Hantaviruses are important zoonotic pathogens . Although they can establish a persistent and asymptomatic infection in their natural rodent reservoirs [1] , in humans hantaviruses can cause two severe diseases: hemorrhagic fever with renal syndrome ( HFRS ) and hantavirus ( cardio ) pulmonary syndrome ( HPS ) [2] . In Eurasia HFRS is associated with Hantaan virus ( HTNV ) , Seoul virus ( SEOV ) , Amur/Soochong virus ( ASV ) , Dobrava-Belgrade virus ( DOBV ) , Saaremaa virus ( SAAV ) , Sochi virus , and Puumala virus ( PUUV ) , whereas HPS is due to the infection of Sin Nombre virus ( SNV ) , Andes virus ( ANDV ) , and other viruses in the Americas [2] , [3] . The clinical severity of HFRS is related to the etiologic agents involved [4]–[8] , with DOBV and HTNV being the most dangerous representatives , with fatality rates of up to 15% [4]–[7] . In contrast , SEOV usually causes a milder form of HFRS with a mortality rate of approximately 1% [6] , [7] . PUUV causes a mild disease referred to as nephropathia epidemica ( NE ) with a mortality rate ranging from 0 . 1% to 0 . 3% in Europe 5 , 8 . HFRS cases caused by HTNV mainly occur in the winter , while the HFRS cases caused by SEOV peak in the spring and summer [9] , and which likely reflects occupation-connected differences in exposure to rodents in different seasons . Following the implementation of comprehensive preventive measures and socioeconomic development , the numbers of HFRS cases and fatalities in China have decreased dramatically , although remain the highest globally [7] . In China , the most prevalent hantaviruses are HTNV and SEOV carried , respectively , by the striped field mouse ( Apodemus agrarius ) and Norway ( or brown ) rat ( Rattus norvegicus ) [6] , [7] , [9] , [10] . To date , only these two viruses have been identified to cause HFRS in China . However , hantaviruses from bats , insectivores , and rodents ( e . g . Dabieshan virus ( DBSV ) , Gou virus ( GOUV ) , Longquan virus ( LQUV ) , Thottapalayam virus ( TPMV ) ) have also been documented [7] , [11]–[15] , although whether they are associated with human disease is unclear . Longquan is a county-level city located in the southwestern part of Zhejiang Province . It includes both urban and rural areas , with a population of approximately 280 , 000 . More than 90% of the Longquan's total area is mountainous . In 1974 , the first HFRS case was recorded in Longquan . Since that time , Longquan has been one of the most severely affected regions in Zhejiang and in China as a whole . However , little is known about the epidemiology and etiologic agents of HFRS in this region . Our recent surveys in Longquan revealed at least nine species of rodents and insectivores , with A . agrarius and R . norvegicus dominant in rural and residential areas , respectively [16] . Herein we report the changing incidence of HFRS in Longquan , the genetic characterization of the etiologic agents ( hantaviruses ) circulating in local rodents , and their connection to the human population .
This study was reviewed and approved by the ethics committee of National Institute for Communicable Disease Control and Prevention , Chinese Center for Disease Control and Prevention ( Chinese CDC ) . All animals were treated in strict according to the guidelines for Laboratory Animal Use and Care from the Chinese CDC and the Rules for the Medical Laboratory Animal ( 1998 ) from the Ministry of Health , China . These protocols were approved by the National Institute for Communicable Disease Control and Prevention of the China CDC . All surgery was performed under ether anesthesia , and all efforts were made to minimize suffering . Collecting human serum samples from HFRS patients was also approved by the ethics committee of National Institute of Communicable Disease Control and Prevention of the China CDC , according to the medical research regulations of Ministry of Health , China . A signed individual written informed consent was obtained from each of five patients when their blood samples were collected . Records for HFRS cases occurring during 1974–2011 were obtained from the Longquan Center for Disease Control and Prevention . Until 1982 , HFRS cases were defined according to the national standard of clinical criteria , and confirmed by detection of hantavirus-specific IgM and IgG antibodies against HTNV or SEOV . From 1982 clinical cases were confirmed by a four-fold or greater titer increase of IgG antibodies in paired sera , as well as a IgM antibody titer >1∶20 in single serum as scored positive by an indirect immunofluorescent assay ( IFA ) ( see below ) [9] . The reaction pattern of positive serum was characterized as scattered and green granular cytoplasmic fluorescence in hantavirus-infected Vero E6 cells . The incidence rates of HFRS during 1974–2011 were calculated according to the population census number for each year . Small mammals were trapped in fields and residential areas in Longquan during 2008–2011 . Cages with a treadle release mechanism were used for live trapping according to the protocols described previously [17] . Traps were set in the same fields or residential areas during each season . Lung and kidney samples were collected from the trapped animals and stored in liquid nitrogen . All surgery was performed under ether anesthesia to reduce suffering . Ethanol-cleaned instruments were used for each animal . Serum samples collected from five patients who suffered from acute HFRS during 2009–2011 were also studied . These serum samples were tested by IFA using HTNV-infected or GOUV-infected Vero-E6 cells as antigens [18] . The secondary antibody used was fluorescei-isothiocyanate-conjugated goat anti-human IgG or IgM ( Southern Biotech , Birmingham , Alabama , USA ) . Hantavirus antigen in lung or kidney tissues from rodents and insectivores was detected by IFA as described previously [18] , with rabbit antibodies against the mixed antigens of HTNV/76-118 and SEOV/L99 prepared in this laboratory as the primary antibodies and FITC-labeled goat anti-rabbit IgG antibodies used as the secondary antibodies ( Sigma , St . Louis , MO , US ) . Generally , lung tissues were tested first , and kidney tissues were tested if lung tissues were found to be negative . Total RNA was extracted from hantavirus antigen-positive lung and kidney tissues , and human serum samples , using the TRIzol reagent ( Invitrogen , San Diego , CA ) according to the manufacturer's instructions . cDNA of the Small ( S ) and Medium ( M ) segments of the hantavirus genome was prepared with AMV transcriptase ( Promega , Beijing , China ) in the presence of primer P14 [19] . Partial or complete sequences of the S and the M segments were amplified as described previously [10] , [20] , [21] . All voles and insectivores were also screened for hantaviruses using RT-PCR as described previously [22] . DNA products were purified by a QIAquick Gel Extraction kit ( QIAGEN , Beijing , China ) and subjected to direct sequencing using the ABI-PRISM Dye Termination Cycle Sequencing ready reaction kit and a ABI-PRISM3730 genetic analyzer ( Applied Biosystems , Carlsbad , CA , USA ) . The genome sequences of hantaviruses were aligned using the ClustalW method implemented in the Lasergene program , version 5 ( DNASTAR , Inc . , Madison , WI ) . Nucleotide ( nt ) and amino acid ( aa ) sequence similarities were calculated using DNAStar . Phylogenetic trees for each segment were inferred using the Bayesian method implemented in MrBayes 3 . 1 [23] and the Maximum likelihood ( ML ) method available in the RAxML Blackbox webserver [24] , employing the best-fit GTR+I+Γ model of nucleotide substitution as determined using jModeltest [25] . Trees were visualized with the TreeView software [26] . The GenBank accession numbers for the sequences obtained here are JQ912697 to JQ912907 , and KC344236 to KC344269 ( Table S2 ) .
The first clinical HFRS case in Longquan was reported in 1974 ( Figure 1 ) . During the 38-year period between 1974 and 2011 , a total of 1 , 866 HFRS cases were registered in this city . Only nine cases were recorded in the 1970s , such that the annual incidence of HFRS increased dramatically during 1980s and 1990s . A peak of 138 cases ( 51 . 2 cases/100 , 000 population ) was reached in 1998 , after which it decreased , likely in part due to the onset of hantavirus vaccination in 1997 and the intense rodent control efforts undertaken in China [7] . In total , more than 63 , 000 people have been vaccinated either by inactivated vaccines ( Youerjian , Tianyuan Bio-Pharma , Hangzhou , China ) for HTNV ( during 1997–2000 ) or purified bivalent vaccine for HTNV and SEOV cultured in sand rat renal cells ( Youerjian , Tianyuan Bio-Pharma , Hangzhou , China ) or Vero cells ( Royal , Royal ( Wuxi ) Bio- Pharmaceutical , Wuxi , China ) ( during 2001–2011 ) . However , despite this vaccination the incidence of HFRS remained relatively high during 1999–2011 , with between 11 . 15 and 23 . 6 cases/100 , 000 population . During 1974–2011 , a total of 20 patients died of HFRS in Longquan , with an average fatality rate of 1 . 07% . The highest fatality rates were observed during the first 10 year period ( 1974–1983 ) , and reached 11% ( 10 fatal cases of 91 cases ) . Notably , all fatal cases occurred in autumn and winter . Additional fatalities were recorded in 1985 ( 1 ) , 1986 ( 2 ) , 1989 ( 1 ) , 1992 ( 1 ) , 1997 ( 1 ) , 1998 ( 1 ) , 2002 ( 1 ) , and 2006 ( 2 ) : these cases occurred in autumn and winter , with the exception of one death in March 2002 . No patients have died of HFRS since 2007 , likely reflecting improvements in disease treatment . The seasonality of HFRS noted above may provide important clues to its cause [9] . We therefore analyzed the seasonality of HFRS in Longquan for different time periods during 1974–2011 . HFRS cases occurred in winter ( November to January ) and in spring/summer ( May to July ) at respective frequencies of 49 . 66% and 14 . 51% during 1974–1990 , 38 . 18% and 24 . 18% in 1991–2000 , and 36 . 20% and 31 . 31% in 2001–2011 ( Figure 2 ) . As the peak of HFRS associated with rats occurred in the spring , whereas HFRS associated with mice occurred mainly in the winter [9] , the recent increase in cases in spring/summer suggests a rising disease toll due to rat-associated hantavirus ( es ) in Longquan . A similar seasonal shift , from mice-dominated to rat-dominated transmission , has been reported in other HFRS endemic regions [27] . To analyze genetic diversity in the natural hantavirus reservoir and its relationship to those viruses found in humans , a total of 2 , 652 small mammals , representing 10 species of rodents and 3 species of insectivores ( Table 1 ) , were captured in Longquan during 2008–2011 . A . agrarius mice and M . fortis voles were the dominant field species , accounting for 41 . 82% ( 1109 ) and 18 . 17% ( 482 ) of all small mammals collected , respectively . However , in residential areas the dominant species were rats of the family Rattus including 425 R . losea ( 16 . 02% ) , 372 R . norvegicus ( 14 . 03% ) , and 201 R . flavipectus ( 7 . 58% ) . Using IFA and RT-PCR , hantavirus antigens were detected in a total of 118 rodents including 78 A . agrarius ( 7 . 03% ) , 5 M . fortis ( 1 . 04% ) , 32 R . norvegicus ( 8 . 60% ) , and 3 R . flavipectus ( 1 . 49% ) . No hantaviruses were found in insectivores . Thus , the etiologic agents of HFRS cases in Longquan were likely hantaviruses carried by A . agrarius mice and R . norvegicus rats . Serum samples from five human patients were collected on day 1 of hospitalization . Samples were tested for IgM and IgG antibodies by IFA using GOUV- or HTNV-infected cells ( Table S1 ) . Three serum samples showed higher IgM and IgG titers in HTNV- specific IFA , and one in GOUV-specific IFA . One sample showed higher IgM titers in HTNV-specific IFA , but with the same titers in HTNV- or GOUV specific IFA , suggesting cross-reactivity of HTNV with GOUV . To further characterize the etiologic agents of human infection in Longquan , complete or partial hantavirus M segment sequences were recovered from 118 hantavirus antigen-positive rodent samples ( Table S2 ) . In addition , complete S segment sequences were amplified from all 118 hantavirus antigen-positive rodent lung tissues , and partial S segment sequences were recovered from the five human serum samples collected from patients with acute HFRS . Notably , the sequences recovered from 78 A . agrarius , 5 M . fortis , and 4 human samples were very closely related to each other , with 98 . 2–100% nt and 98 . 4–100% aa sequence identities in the M segment and 98 . 6–100%/99 . 3–100% identities in the S segment . This similarity is indicative of direct viral transmission from rodents to humans . To determine the phylogenetic relationships among the viruses described here and to known hantaviruses , phylogenetic trees were estimating using the M and S segment sequences ( in which Bayesian and ML methods produced similar topologies ) . Most notably , the sequences sampled from A . agrarius mice , M . fortis voles , and humans clustered together and formed a distinct and well-supported lineage in both trees ( Figures 3–4 ) . Interestingly , these strains also exhibited a close evolutionary relationship to strains HTNV and Z5 previously isolated from A . agrarius mice in Zhejiang Province [28] . All hantavirus sequences recovered from rats were very closely related to each other , with 97 . 5–100% nt and 98 . 5–100% aa sequence identities in the M segment and 98 . 5–100%/98 . 6–100% identities in the S segment . The partial S segment sequence recovered from the serum of a human patient in which high titers of IgG antibodies against GOUV had been detected ( Table S1 ) was very closely related to sequences recovered from rats ( 98 . 9–99 . 5%/99 . 2–100% ) . Remarkably , these hantavirus sequences were closely related to previously described variants of GOUV – Gou3 , ZJ5 , YongjiaRf45 and YongjiaRn14 – but more distant from variants of SEOV . In phylogenetic trees of the M or S segments those strains from Rattus rats ( R . flavipectus and R . norvegicus ) and human clustered together , forming a distinct and strongly supported cluster ( posterior probabilities of 1 . 0 for both M and S sequences ) within the broader group of GOUV sequences ( Figures 3–4 ) . This phylogenetic pattern is indicative of a new genetic variant of GOUV in Longquan .
HFRS was a serious problem in China during the 1980 and 1990s [7] , [9] . As a result of comprehensive preventive measures and improved living conditions , the incidence of HFRS in China has declined dramatically during the last decade [7] . Because of favorable ecological conditions and low socioeconomic status in rural areas , farmers have frequently been the major victims of HFRS , both inside and outside of China [5] , [7] , [9] , [29]–[31] . The annual number of registered HFRS cases in Longquan has decreased , from 138 in 1998 to 46 in 2011 , with a similar pattern observed in other parts of China [7] , [32] . However , the incidence rate ( >10 cases/100 , 000 population ) in this region is still the highest in China despite ongoing vaccination . Considering the dramatic decrease in the rural population of Longquan in recent years ( at least 30% of the rural population had moved into cities or towns by the end of 1990s ) , the real incidence rate of HFRS in rural areas may be much higher than reported here . In addition , the prevalence of hantavirus infection is high in rodents in both rural and residential areas in Longquan , especially GOUV in Norway rats ( >8% ) . Thus , hantavirus infection will likely remain a major public health problem for the foreseeable future in Longquan city . GOUV was first isolated from R . rattus captured in Zhejiang Province in 2000 [11] , and initially considered as a variant of SEOV [10] , [11] . However , GOUV is distinct from SEOV both serologically and genetically [11] and is found in a different rat species ( R . rattus , R . flavipectus ) . Such distinction means that GOUV is currently defined as a tentative hantavirus species by the International Committee on Taxomony of Viruses ( ICTV ) [14] . In this study hantavirus variants originating from rats ( R . flavipectus and R . norvegicus ) from Longquan were most closely related to GOUV , forming a distinct and strongly supported lineage in both the M and S segment trees . Hence , these data suggest that the hantavirus variants carried by Rattus rats in Longquan represent a new genetic variant of GOUV . As no SEOV or other hantaviruses have been found in Rattus rats from Longquan and the sequences recovered from one patient in Longquan belonged to GOUV , our data clearly indicate that GOUV carried by Rattus rats ( R . flavipectus and R . norvegicus ) can cause human disease , and that there is ongoing spillover from the rodent reservoir to the human population . This is the first report of GOUV being associated with human HFRS cases since its discovery in 2000 [11] . Accordingly , further studies are needed to determine the pathogenicity and severity of GOUV in humans , as well as the possibility of human-to-human transmission . Similar to other hantaviruses [2] , [33] , HTNV exhibits considerable genetic diversity and displays a geographic clustering of genetic variants , especially in mountainous regions [34] . To date , at least nine genetic lineages of HTNV have been found in Apodemus mice in Eastern Asia [34] . In this study , the virus sequences recovered from Apodemus mice and Microtus voles in Longquan formed a distinct lineage within HTNV in both the M and S segment trees , suggesting that a new genetic variant of HTNV is circulating in Longquan . Our earlier studies in the northeast and central parts of China documented Yuanjiang virus ( YUJV ) and Vladivostok virus ( VLAV ) in M . fortis voles , respectively [35] . However , these viruses were not detected in the Microtus voles from Longquan . Additional study is needed to determine if these viruses are present and whether they are pathogenic to humans in the HFRS-affected region . Previous investigations revealed that HFRS caused by HTNV transmitted by Apodemus mice occurred mainly in winter , while the peak of HFRS caused by hantaviruse ( s ) transmitted by Rattus rats was in spring [9] , [36] , [37] . The seasonal analyses of HFRS cases performed here indicated that most of the HFRS cases registered in Longquan during 1974–1998 occurred in winter , in turn suggesting that human infections were due to HTNV . However , during the last decade the peak of HFRS has shifted to the spring in Longquan , with a similar pattern observed in other HFRS endemic regions [27] . In addition , GOUV was highly prevalent in R . norvegicus in residential areas , and more so than HNTV in Apodemus mice ( prevalences of 8 . 60% and 7 . 03% , respectively ) . In sum , these data suggest that hantavirus ( es ) carried by rats may have become the major cause of HFRS in Longquan city over the past decade . In conclusion , we have shown that hantavirus infection is endemic in both humans and rodents in Longquan , with the latter acting as a major reservoir for the former . Epidemiological and phylogenetic analyses indicate that GOUV and HTNV are circulating in local rodents and have a direct connection to the human population . As rats ( Rattus species ) are more mobile than the hosts of other hantaviruses [10] , this study strongly reinforces the need for vigilance in preventing the spillover of GOUV from rats in China . | Hemorrhagic fever with renal syndrome ( HFRS ) is a major public health problem in China despite human vaccination . We investigated the epidemiology and emergence of HFRS in Longquan ( Zhejiang Province ) , a rural area with a high incidence of HFRS . During 1974–2011 , a total of 1866 cases of HFRS were recorded in Longquan , including 20 deaths . Strikingly , phylogenetic analyses of viral sequences sampled from local rodents in Longquan revealed the presence of novel variants of Gou virus ( GOUV ) in Rattus sp . rats and Hantaan virus ( HTNV ) in the stripe field mice , respectively . Moreover , viral sequences sampled from infected humans in Longquan were very closely related to those from rodents . Overall , these data indicate that there is a continual spillover GOUV and HTNV from rodents to humans in Longquan , and this might be responsible for the high prevalence of HFRS . As well as highlighting the importance of the human-animal interface , these data also suggest that GOUV is now the major cause of HFRS in this region . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [] | 2013 | Ongoing Spillover of Hantaan and Gou Hantaviruses from Rodents Is Associated with Hemorrhagic Fever with Renal Syndrome (HFRS) in China |
Systems neuroscience has identified a set of canonical large-scale networks in humans . These have predominantly been characterized by resting-state analyses of the task-unconstrained , mind-wandering brain . Their explicit relationship to defined task performance is largely unknown and remains challenging . The present work contributes a multivariate statistical learning approach that can extract the major brain networks and quantify their configuration during various psychological tasks . The method is validated in two extensive datasets ( n = 500 and n = 81 ) by model-based generation of synthetic activity maps from recombination of shared network topographies . To study a use case , we formally revisited the poorly understood difference between neural activity underlying idling versus goal-directed behavior . We demonstrate that task-specific neural activity patterns can be explained by plausible combinations of resting-state networks . The possibility of decomposing a mental task into the relative contributions of major brain networks , the "network co-occurrence architecture" of a given task , opens an alternative access to the neural substrates of human cognition .
There is uncertainty about pertinent concepts of functional brain architecture . Systems neuroscience has established the existence of a set of fluctuating yet robust brain networks in humans [1 , 2] . It however remains elusive how these neurophysiological phenomena relate to the repertoire of mental operations of an individual . This calls for methodological approaches that go beyond computing linear correlations ( e . g . , [3 , 4 , 5] ) or independent components ( e . g . , [6 , 7 , 8] ) in the "resting" human brain without controlled task modulation . The central hypothesis of the present work is that network patterns can effectively describe fMRI data in both a resting mind-wandering and goal-directed task context . We introduce a methodological approach that enables formal assessment of this task-rest correspondance . On the one hand , the mostly descriptive statistical analyses used by many previous neuroimaging studies are extended in the present work by introducing an inferential statistical approach for the network involvement during task and at rest . On the other hand , the proposed approach combines deriving and predicting mutually overlapping network patterns ( i . e . , "network co-occurrence modeling" ) , whereas existing neuroimaging methods frequently focus on non-overlapping voxel and region patterns . Identical neural networks have repeatedly been observed across cognitive domains using diverging neuroscientific methods . These observations prompted widely-adopted notions , including for instance the “default-mode network” [9] , “salience network” [10] , and “dorsal attention network” [11] . Such "networks" ( i . e . , spatiotemporally coherent signal patterns ) are likely manifestations of electrophysiological oscillation structure [12 , 13] . Developmentally , large-scale networks emerge during late fetal growth [8] , before cognitive capacities mature in childhood . In adults , nodes of a same cohesive network probably have more similar functional profiles than nodes from different networks [14] . Indeed , resting-state fluctuations between large-scale networks were observed to be less stable than coupling between regions of each network when assessed by intra-class correlation [15] . Between-network connections were also less stable than intra-network connections when assessed by Kendall’s coefficient of concordance [15] . During task-unrelated random thought , the dynamical engagement of major brain networks therefore appear to be more volatile across participants and brain scans than intra-network dynamics [15–17] . While it is currently unknown "how global network architectures self-organize or reconfigure for specific tasks" [18] , this hints at the constellation of relative network involvements as an under-appreciated unit of functional brain organization . In line with this contention , the onset of a given cognitive task might induce characteristic changes in functional coupling of large-scale networks . For instance , the salience network and dorsal attention network tend to display blood-oxygen-level-dependent ( BOLD ) signal increases due to experimental stimulation , while the default-mode network often decreases across a wide range of tasks [19] . Whether stimulus-evoked compositions of such networks explain the majority [6] or only a fraction [4] of overall task activity is currently unresolved . For instance , a working-memory task entailed increase in BOLD activity in dorsal attention network regions but decrease in default-mode regions [20] . Notably , the functional connectivity did not change significantly within either dorsal attention network or default-mode network during this neuroimaging task . During auditory event transitions in another experimental fMRI study , both dorsal attention network and salience network increased in activity , whereas the default-mode network decreased in activity [21] . These changes of network constellation are probably mechanistically relevant for unfolding behavior [7 , 22–24] . This idea is supported by evidence that proportional default-mode network recruitment impairs task performance , which is believed to be subserved by other large-scale networks [25 , 26] . The mediation between canonical networks was tentatively proposed to involve the right anterior insula [21] and right temporo-parietal junction [27] . Moreover , the relevance of network engagement architectures possibly extends to psychiatric and neurological disorders [28 , 29] . For instance , reciprocal coupling between default-mode network decreases and task-recruited networks were found to be absent in autism [30] , reduced in schizophrenia [31] and major depression [32] , as well as frequency-altered in attention deficit hyperactivity disorder [33] . The neuroarchitectural difference underlying the task behaviors and the idling brain has been investigated by what one can summarize as “dichotomy” and “manifold” hypotheses . The dichotomic view advocates functional antagonism between so-called “task-positive” and “task-negative” neural networks ( e . g . , [35] ) . Task-positive networks are believed to instantiate exteroceptive , environment-oriented mind sets to maintain task-constrained stimulus evaluation and response . Task-negative networks are believed to instantiate interoceptive , environment-detached mind sets to maintain adaptive mental imagery . The dichotomic view receives support from the following observations: a ) the default-mode network consistently decreased in neural activity during many ( not all ) neuroimaging tasks [19 , 34] , b ) activity in task-positive brain regions was consistently anti-correlated with activity in task-negative regions [35] , and c ) the spectrum of activity patterns in subcortical , limbic , and primary sensorimotor areas was found to be richer during task than at rest [4] . Hence , the dichotomic view predicts that extracting network components from resting-state data will yield a dictionary of network definitions insufficient to delineate task-specific network compositions . In contrast , the “manifold” view advocates a fluctuating equilibrium of functionally distinct large-scale networks that is perturbed at task onset ( e . g . , [66] ) . This favors an identical ensemble of neural networks underlying functional brain architecture in focused and resting brain states . This view , in turn , receives support from the following observations: a ) seed-region-based resting-state correlations frequently recovered task-typical networks [36] , b ) separate decomposition of task and rest activity maps yielded a number of topographically similar networks [6] , and c ) only 11% of whole-brain connectivity patterns always shifted at onset of different tasks [3] . Hence , the manifold view predicts that network components extracted from either rest or task data will perform similarly in capturing task-specific neural activity . While the present investigation underlines the functional integration account of human brain organization , it will also be explicitly contrasted with the functional segregation account more frequently embraced by previous fMRI studies . Functional integration emphasizes brain function as an emergent property of complex connections between distinct brain regions [37 , 38] . Functional specialization , in turn , emphasizes that microscopically distinguishable brain regions are responsible for solving distinct classes of computational processes [39 , 40] . For instance , single-cell recordings and microscopic examination revealed the anatomical segregation of the occipital visual cortex into specialized V1 , V2 , V3 , V3A/B , and V4 areas [41 , 42] . Tissue lesion of the mid-fusiform gyrus of the visual system , as another example , is known to impair identity recognition from others' faces [43] . In neuroimaging research , specialized brain regions are frequently revealed by clustering algorithms [44 , 45] , such as k-means or hierarchical ( ward ) clustering . Functional brain network , however , are often extracted via matrix factorization methods [46] , such as independent component analysis ( ICA ) and principal component analysis ( PCA ) . It is important to appreciate that functional segregation findings necessitate neuroscientific interpretation according to non-overlapping , discrete region compartments , whereas network integration findings are to be interpreted by embracing cross-regional integration by overlapping network compartments . These considerations motivate the central question of the present study: To what extent do task-evoked fMRI signals lend themselves more to encoding and reconstruction in a network space rather than in a region space ? The present study systematically evalutes the possibility of analyzing fMRI tasks as co-recruitment of the entire set of major networks using multivariate statistical learning . A multi-step framework capitalized on two independent , large datasets ( n = 500 and n = 81 ) with a total of 36 typical neuroimaging tasks . Each of these two task batteries attempted to cover the diversity of human cognitive processes . In an unsupervised learning step ( i . e . , naive to task labels ) , we first derived a small number of representative brain networks from these extensive neuroimaging resources . This first step yielded explicit models of possible network patterns with minimal statistical assumptions and without recourse to cognitive theory . In a supervised learning step ( i . e . , based on task labels ) , the dimensionality-reduced neuroimaging data were then submitted to an 18-task classification problem . For each task , this second step determined a plausible combination of the modes of variation in fMRI signals from the brain . This quantitative association with traditional psychological concepts made the data-driven results human-interpretable . In a validation step , we recovered task activation patterns from the respective component loadings in the learned statistical models . This third step evaluated how well formal network models can generate realistic task activation patterns . To show the usefulness of network co-occurrence modeling , we quantitatively revisited to what extent network activity underlying idling brain states can explain the variation of neural activity patterns during engagement in task performance .
The analyses , results , and figures are divided into two different sections ( see schematic in Fig 1 ) . In the first section ( Figs 2–5 , entitled "network co-occurrence modeling" ) , we statistically tested whether neural activity patterns measured with fMRI in humans can be largely explained by changes in cohesive network units . The term “network” henceforth refers to spatiotemporal modes of variation extracted from time series of fMRI activity whose weighted linear combination sum up to whole-brain activity patterns [6 , 47 , 48] . Whole-brain activity maps were expressed as a linear combination of 40 spatiotemporally coherent patterns ( i . e . , network components ) . The distributed BOLD signals from voxel space were thus reduced to 40 component loadings in a network space . The network engagement pattern is shown to successfully classify and restore neural activity from the 18 psychological tasks across two large datasets . In the second section ( Figs 6–8 , entitled "task-rest correspondence" ) , this possibility to compress neural activity maps into low-dimensional summaries of brain network involvements was used to revisit the relationship between the functional brain architecture during defined psychological tasks and unconstrained task-free mental activity . While the first section used a same task dataset ( either HCP or ARCHI ) for both unsupervised network discovery and supervised multi-task classification , the applied second section probed diverging ways of deriving the network dictionary for subsequent classification . In particular , networks from i ) the same task data dataset , ii ) the respective other task dataset , iii ) rest data , and iv ) Gaussian noise were quantitatively assessed for their commonalities and differences in restoring task activity patterns . We first assessed the predictive accuracy of network co-occurrence models across methodological choices and datasets ( Figs 2–5 ) . Without using any rest maps , ICA and sparse PCA were applied to 50% of the task maps to then translate the remaining 50% of task activity maps into sets of network loadings . l1-penalized support vector machines ( SVM ) with C-hyperparameter tuning on ICA and sparse PCA loadings ( i . e . , second 50% data ) correctly detected 18 tasks 90% to 93% of the time in both HCP and ARCHI datasets ( Fig 2 ) . Note that l1 penalization in these and below classification model estimations increased interpretability by introducing zeros into the model weights corresponding to network loadings ( cf . methods section ) . Note further that chance level is at 5 . 6% in an 18-class scenario . To evaluate diagnostic metrics typically used in machine learning [49] for ICA and sparse PCA , the recall ( How many brain images labeled as a cognitive task were correctly recognized to belong to that class ? ) ranged between 90% and 93% , while the precision ( How many brain images recognized to belong to a certain class were really labeled as that class ? ) ranged between 87% and 90% . We deemphasized results based on PCA and factor analysis whose lower accuracy scores ranged between 87% and 92% . Additionally , ICA and sparse PCA also yielded higher model sparsity ( induced by l1-norm penalization to bias the SVM estimation ) than PCA and factor analysis , computed by sparsity = ∑ [||weights||1 / ||weights||F] based on the vectorized SVM weights of all 18 classes . This parsimony-based metric of model "effectiveness" ranged between 8 . 55 and 13 . 19 for ICA and sparse PCA ( a lower number indicates a sparser weights of the SVM model ) , whereas it ranged between 13 . 33 and 14 . 51 for PCA and factor analysis . We thus found that activity maps can be described more correctly ( according to model performance metrics ) and more effectively ( according to a model sparsity metric ) by the learned network sets based on ICA and sparse PCA compared to components obtained from PCA and factor analysis . More generally , the observation of many zeros among the averaged model weights suggests that the “true” decomposition of task activity maps is a linear combination of few network components . Indeed , 18 diverse cognitive tasks could be very well distinguished solely based on 40 loadings of large-scale networks . The approach has been validated using four different decomposition methods in both HCP [50] and ARCHI [51] datasets . In sum , using task-derived network dictionaries ( from an identical dataset ) , the extraction of and projection into network sets from ICA and sparse PCA achieved the highest prediction accuracies , precision and recall scores , as well as the most parsimonious model parameters . ICA and sparse PCA hence identified hidden networks underlying task settings that allowed the most efficient quantification of distinctive neural activity aspects . After validating the proportional implication of large-scale networks as a salient property of task activity , we aimed at the interpretability of the network co-occurrence models from task maps ( without any recourse to rest maps ) . The 18-task classification problem was solved by l2-penalized SVM of the selected k most important network loadings for each task ( k = 40 , 20 , 10 , 5 , and 1 ) ( Fig 3 ) . As an exception to the generally employed l1-regularization of the SVM models , l2 penalization is used at this point because variable selection is already induced by the preliminary ANOVA-based selection ( cf . methods section ) . The k best predictors for each task were inferred by classical univariate ANOVA tests for the network features that explain most variance between each task and the respective other tasks ( Fig 4 ) . Importantly , the determined single most important ICA or sparse PCA network loading per task yielded classification scores between 83% ( ICA , standard deviation [SD] computed across participant-wise data folds in cross-validation scheme: 1 . 3 ) and 46% ( Sparse PCA , across-fold SD = 1 . 7 ) for HCP as well as between 72% ( ICA , SD = 3 . 0 ) and 38% ( sparse PCA , SD = 4 . 7 ) for ARCHI . Increasing the number of k discriminative networks per task rapidly saturated the predictive accuracy . In both datasets , the classification accuracy was virtually identical when knowing all 40 or only the 20 most distinctive network loadings , which in turn was comparable to knowing only 10 and 5 network loadings . In sum , when assuming existence of 40 major networks , experimentally evoked neural activity patterns can be well discriminated by the relative implication of five task-activity-derived large-scale networks per experiment . We then inspected the learned classification models based on task-derived network dictionaries as to whether they were fit for purpose [52] . The fitted model weights were unboxed by back-projection into whole-brain space ( cf . methods section ) . It was thus quantified to what extent the winning explicit models capture genuine properties of fMRI task activity . This provided a real-world face-validity criterion to disambiguate whether task-immanent aspects of neural activity or arbitrary discriminative aspects ( e . g . , structured noise , participant/scanner-related idiosyncracies ) explain the discriminative performance of a statistical model ( Fig 5 ) . Whole-brain activity maps were thus generated from the network co-occurrence models for each task ( S6 and S7 Figs ) . The model-derived activity maps were then Pearson correlated with the mean first-level activity maps as a measure of recovery performance . For ICA decomposition , the mean linear correlation across 18 tasks reached r = 0 . 81 ( HCP , standard deviation computed across tasks = 0 . 20 ) and r = 0 . 88 ( ARCHI , SD = 0 . 07 ) . For sparse PCA decomposition , in turn , mean correlations reached r = 0 . 69 ( HCP , SD = 0 . 21 ) and r = 0 . 70 ( ARCHI , SD = 0 . 15 ) . As a side note , these findings were confirmed against the negative tests by pseudo-network priors derived from decomposition of Gaussian noise ( i . e . , 1000 noise maps with smoothed random activations ) . The corresponding correlation analyses between first-level task maps and model-derived task maps ranged between r = 0 . 32 ( across-task SD = 0 . 07 ) and r = 0 . 25 ( SD = 0 . 06 ) . In sum , fMRI-task-activity derived neurobiological networks allowed significantly better reconstruction into the original activity space of >60 . 000 voxels than a dictionary of Gaussian random network templates . As an alternative to network co-occurrence models , whole-brain task activity was also captured by region co-occurrence models ( Figs 5 , S9 and S10 ) . Capturing task activity in local cluster units , rather than distributed network units , was much less successful as clearly indicated by the acid test of reconstructing task-specific whole-brain activity from a few numbers alone ( Figs 5 , S9 and S10 ) . Indeed , building a feature space that summarizes neural activity based on a ward cluster segregation of gray matter ( out-of-sample accuracy 85% for HCP and 90% for ARCHI ) , the mean linear correlation between average class images and class reconstruction across 18 tasks reached only r = 0 . 11 ( HCP , across-task SD = 0 . 12 ) and r = 0 . 26 ( ARCHI , SD = 0 . 13 ) . Summarizing neural activiy in k-means clusters ( out-of-sample accuracy 86% for HCP and 89% for ARCHI ) , the mean linear correlation reached only r = 0 . 32 ( HCP , across-task SD = 0 . 18 ) and r = 0 . 27 ( ARCHI , across-task SD = 0 . 16 ) . Finally , the network- and region-based reconstruction performances were compared to fitting the same l1-penalized support vector machine models on all 61 , 472 gray-matter voxels without the initial feature-learning step , that is , without recourse to global spatiotemporal components or local voxel clusters . Neurobiologically uninformed support vector machines applied to the raw voxel features ( 89% out-of-sample performance for HCP and 93% for ARCHI ) , otherwise identical to above classifier training by multi-class , one-versus-rest , and leave-one-participant-out design , achieved correlations between z-scored class average maps and z-scored model weights of r = 0 . 24 ( across-task SD = 0 . 04 ) for HCP and r = 0 . 11 ( across-task SD = 0 . 05 ) for ARCHI . Consequently , cluster-based and voxel-based multivariate predictive models were less successful in capturing the original task activity patterns than network co-occurrence models . After evaluating network co-occurrence models , they were put into practice by testing explicit hypotheses about the difference between functional brain architecture of the human brain during tasks and at rest . In contrast to the above results , these experiments also performed regions/network discovery step and region/network-based task classification step based on non-identical data: HCP task maps , ARCHI task maps , and rest maps ( Figs 6–8 ) . We formally tested whether given task activity patterns can be accounted for by network sets obtained from a diverging task battery and from the task-free resting brain ( Figs 8 , S11 and S12 ) . To this end , recovery performance was compared between network sets learned from non-identical task data and from rest data . The rationale is that if functional brain architecture during task and at rest is similarly rich in variation patterns as measured by fMRI , then a rest-derived coordinate system should be able to effectively encode and reconstruct task activity maps as expressions on its 40 axes with little loss of information . How much information on the actual 3D activation patterns is lost by information compression in a network space is measured by the recovery performance . Learning network co-occurrence models from a non-identical task battery ( i . e . , networks from HCP task data and classification in ARCHI task data , or vice versa ) yielded recovery performances between r = 0 . 50 ( across-task SD = 0 . 17 ) and r = 0 . 43 ( SD = 0 . 17 ) across two decompositions and datasets . Importantly , learning network co-occurrences models from task-unrelated resting-state correlations yielded very similar recovery performances between r = 0 . 51 ( SD = 0 . 15 ) and r = 0 . 46 ( SD = 0 . 11 ) . Assessed by independent t-tests , recovery performance based on network sets from different task data was in no instance significantly better than recovery from networks discovered in task-unrelated rest data . This suggests that the network ecosystem of the mind-wandering brain as measured with fMRI scanning is recruited in a characteristic fashion in response to environmental challenges . The application of network co-occurrence models to the task-rest correspondence thus indicates that the directions of variation observed in the human brain at rest are sufficiently rich to explain the directions of variation observed in the human brain during task performance .
Only recently , systems neuroscience has transitioned the interpretational focus from regional segregation to network integration [18 , 28 , 37 , 72] . Many neuroimaging studies then performed ICA without relation to human cognitive processes or conducted resting-state correlations yielding unknown mixtures of constituant brain networks . The present study may be the first to quantify the composition of large-scale network recruitments during defined psychological tasks and to thus qualify a meaningful mechanism underlying cognitive neuroscience experiments . Task-specific network compositions might thus intimately relate to psychological notions of mental operations , although fine-grained local effects are ignored . This organizational principle of the human brain might extend to other species given existence of large-scale networks in monkeys [73] and rats [74] . Moreover , clinical research has corroborated default-mode network dysfunction in various psychiatric and neurological disorders [75 , 76] . As a tempting alternative hypothesis for future research , not disturbance in the default-mode network itself but its relation to other canonical networks might be specific to brain disorders . Ultimately , the present investigation exposes network co-occurrence architecture as an important neural mechanism that complements regional brain responses in maintaining human cognition .
Data folding and model selection were performed in the following fashion . In the first step , one half of the task datasets ( i . e . , HCP and ARCHI ) and the entire rest dataset were used for unsupervised ( i . e . , label-independent ) discovery of latent structure by matrix decomposition and clustering methods . The components of variation identified in the data allowed for feature engineering from biological structures . In the second step , we applied supervised ( i . e . , label-dependent ) classification algorithms to the other half of the activity maps to predict cognitive tasks . The winning models for classifying 18 cognitive tasks were compared by cross-validation . This inferential statistical framework is the gold standard to obtain an unbiased estimate of how well a trained classifier generalizes beyond the data samples at hand [80 , 81] . The previously unseen half of the task data ( i . e . , 4325 maps from HCP and 702 maps from ARCHI] were split into as many data folds as participants ( i . e . , 498 for HCP and 78 for ARCHI] . In each fold , all task maps of a given participant were left out as the test set for assessment of out-of-sample performance , while the task maps from the remaining participants served as the training set for model estimation . This leave-one-participant-out cross-validation scheme ensured model fitting of task effects , rather than interindividual differences . In the third and last step , the averaged winning models were back-projected into realistic task activity maps as face validity for the reduced representation of task activity patterns . Python was selected as scientific computing engine . Capitalizing on its open-source ecosystem helps enhance replicability , reusability , and provenance tracking . Nipy [95] performed basic analyses of functional neuroimaging data ( http://nipy . org/ ) . Scikit-learn [96] provided efficient , unit-tested implementations of state-of-the-art statistical learning algorithms ( http://scikit-learn . org ) . This general-purpose machine-learning library [97] was interfaced with the neuroimaging-specific nilearn library for high-dimensional neuroimaging datasets ( http://github . com/nilearn/nilearn ) . 3D visualization of brain maps was performed using PySurfer ( http://pysurfer . github . io/ ) . All analysis scripts of the present study are readily accessible to the reader online ( http://github . com/banilo/taskrest2016 ) . | Assuming the central importance of canonical brain networks for realizing human cognitive processes , the present work demonstrates the quantifiability of relative neural networks involvements during psychological tasks . This is achieved by a machine-learning approach that combines exploratory network discovery and inferential task prediction . We show that activity levels of network sets can be automatically derived from task batteries of two large reference datasets . The evidence supports the often-held suspicion that task-specific neural activity might be due in large part to distinct recombinations of the same underlying brain network units . The results further discourage the frequently embraced dichotomy between exteroceptive task-associated versus interoceptive task-unspecific brain systems . Standard fMRI brain scans can thus be used to reconstruct and quantitatively compare the entire set of major network engagements to test targeted hypotheses . In the future , such network co-occurrence signatures could perhaps be useful as biomarkers in psychiatric and neurological research . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"learning",
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"radiology",
"functional",
"magnetic",
"resonance",
"imaging",
"neural",
"networks",
"social",
"sciences",
"neuroscience",
"learning",
"and",
"memory",
"magnetic",
"resonance",
"imaging",
"multivariate",
"analysis",
"cognitive",
"psychology",
"mathematics",
"statistics",
"(mathematics)",
"artificial",
"intelligence",
"network",
"analysis",
"brain",
"mapping",
"cognition",
"neuroimaging",
"research",
"and",
"analysis",
"methods",
"computer",
"and",
"information",
"sciences",
"imaging",
"techniques",
"mathematical",
"and",
"statistical",
"techniques",
"support",
"vector",
"machines",
"principal",
"component",
"analysis",
"psychology",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"cognitive",
"science",
"statistical",
"methods",
"machine",
"learning"
] | 2016 | Formal Models of the Network Co-occurrence Underlying Mental Operations |
The secondary structure of a pre-mRNA influences a number of processing steps including alternative splicing . Since most splicing regulatory proteins bind to single-stranded RNA , the sequestration of RNA into double strands could prevent their binding . Here , we analyzed the secondary structure context of experimentally determined splicing enhancer and silencer motifs in their natural pre-mRNA context . We found that these splicing motifs are significantly more single-stranded than controls . These findings were validated by transfection experiments , where the effect of enhancer or silencer motifs on exon skipping was much more pronounced in single-stranded conformation . We also found that the structural context of predicted splicing motifs is under selection , suggesting a general importance of secondary structures on splicing and adding another level of evolutionary constraints on pre-mRNAs . Our results explain the action of mutations that affect splicing and indicate that the structural context of splicing motifs is part of the mRNA splicing code .
RNA molecules can adopt various conformations in solution by base pairings and hydrophobic interactions . For example , transfer RNA ( tRNA ) adopts a defined secondary and tertiary structure [1] , whereas most messenger RNAs ( mRNAs ) exhibit only local structures [2] . Defined RNA structures can be recognized by other molecules , as exemplified by aminoacyl-tRNA synthetases that contact with the minor groove of the tRNA acceptor stem [3] , which illustrates the functional importance of RNA structure . In contrast , most proteins regulating splice site selection recognize single-stranded , not base-paired RNA . For example , PUF ( Pumilio/FBF ) , Zn-binding , KH ( hnRNP K homology ) domains , and RRMs ( RNA recognition domains ) bind to 2–10 nucleotides of single-stranded RNA [4] . Frequently , the RNA part that binds to a protein is in a hairpin loop [5 , 6] . The sequence specificity of binding is achieved by hydrophobic interactions between RNA bases and amino acids on the surface of the protein , which explains why RNA binding proteins can bind with sequence specificity to unstructured RNA [4] . The large majority of human genes is alternatively spliced [7 , 8] , and the expression of more than one transcript from one gene represents an important mechanism to increase the diversity of a transcriptome and proteome from a limited number of genes [9 , 10] . In higher organisms , the splice sites do not contain all the information that is required for accurate intron recognition [11] . Additional enhancer and silencer signals located in exons and introns ( ESE , ESS , ISE , and ISS for short ) are essential for the alternative and constitutive splicing [12] . We refer to them collectively as splicing regulatory motifs . These splicing regulatory motifs bind to RNA binding proteins [4] or other RNAs [13] . There is emerging evidence that pre-mRNA secondary structure plays a role in alternative splicing ( reviewed in [14] ) . For example , a deletion in the mouse fibronectin EDA exon leads to a shift of a critical ESE from single- into double-stranded conformation , which causes exon skipping [15] . The skipping of exon 7 of the SMN2 gene is correlated with the stability of a stem structure that sequesters the donor splice site [16] . The splicing of mutually exclusive exons in the rat FGFR2 and Drosophila DSCAM gene is regulated by conserved secondary structures [17 , 18] . Moreover , it has been proposed that sequences surrounding alternative exons might form structures that loop out the exon and prevent its recognition [19 , 20] . Here , we analyze whether the structural context of binding sites for splicing regulatory proteins has a general importance . We first compiled a large set of experimentally verified splicing enhancer and silencer motifs with their natural sequence context . Calculating the probability of a motif to be single-stranded , we found that splicing motifs are located in a structural context that favors their sequence to be more single-stranded than expected from various controls . These results were confirmed by transfection experiments comparing the effect of known splicing motifs in single- and double-stranded conformation . Furthermore , we show that the structural context of predicted splicing regulatory motifs is under selection . Our findings suggest that pre-mRNA secondary structures are an integral part of splice site recognition .
To investigate the structural context of splicing motifs , we first developed an appropriate measurement of single-strandedness . Since in vivo mRNA secondary structures are mostly unknown [21] , we use the standard approach of predicting structures by energy minimization [22] . For measuring single-strandedness , we compute the probability that all bases in the motif are unpaired ( denoted as probability unpaired or PU value ) using the equilibrium partition function [23] . Higher PU values indicate higher single-strandedness of the motif . As PU values account for all possible structures , they include possible structural fluctuations and circumvent inaccuracies caused by considering only one energetically optimal structure or a limited number of suboptimal structures . PU values allow the direct comparison of the single-strandedness of splicing motifs . In vivo secondary structures of pre-mRNAs are likely to be local rather than global . Local RNA folding is influenced by the length of the flanking sequence context [24] , and several lines of evidence indicate that pre-mRNA folding windows are small . First , pre-mRNA is bound by numerous proteins , which influences their ability to fold freely . Second , the formation of secondary structures occurs cotranscriptionally , which favors short-range over long-range base pairing [24] . This model is consistent with the results of kinetic folding algorithms [25] . Third , experiments suggested that pre-mRNA folding is limited to a region of about 50 nt downstream of the transcribing polymerase [26] . For these reasons , we focused on local base pairing and considered all symmetrical context lengths from 11 up to 30 nt up- and downstream of the splicing motif . Thus , for a motif of length 6 nt , we considered sequences with a total length from 28 nt ( for context length 11 ) to length 66 nt ( for context length 30 ) . Smaller context lengths were not considered as the resulting sequences rarely form energetically stable structures . We computed the PU value of the splicing motif for all 20 context lengths . To obtain a single PU value for each splicing motif and to reduce a strong dependency on a single fixed context length , we averaged these 20 PU values . Figure 1A illustrates this approach for a binding site of the polypyrimidine tract binding protein . We used the AEdb motif database to obtain an experimentally verified set of high quality splicing regulatory motifs [27] . After filtering ( see Materials and Methods ) , this set comprises 77 exonic and intronic enhancers and silencers with a length up to 9 nt from human , mouse , rat , chicken , Drosophila , and several viruses ( Table S1 ) . For each motif , we computed the average PU value using the natural pre-mRNA sequence context as described above . To get an overall measure , we averaged these 77 PU values ( Table 1; Table S1 ) . To assess whether verified splicing motifs have a preference for single strands , we performed a statistical evaluation controlling for the motif length and the guanine-cytosine content ( GC ) content that influence the PU values ( Figures S1 and S2 ) . First , we randomly chose a new motif of the same length in the up- and downstream flanks of the natural sequence context for all 77 motifs ( control dataset 1 , Figure 1B ) . We repeated this 100 times to obtain 100 sets each with 77 randomly chosen motifs . The p-value was calculated as the fraction of random sets having a higher average PU value compared to the verified motifs ( for example , if all 100 random sets have a lower PU value , the p-value is less than 1/100 = 0 . 01 ) . This control accounts for possible biases in the selection of genes or exons since it uses the same sequences . We observed a substantial drop in the average PU value from 0 . 25 to 0 . 15 , indicating that verified motifs are significantly more single-stranded ( p < 0 . 01 , Table 1 ) . To also account for the sequence bias of the verified motifs , we used another control . We repeated the procedure of control 1 but copied the verified motif to a new position ( control dataset 2 , Figure 1B ) . All random sets exhibit a lower single-strandedness ( p < 0 . 01 ) . To further verify this effect , we used dinucleotide shuffling [28] to modify the up- and downstream flanks of verified motifs while preserving the motif sequence ( control dataset 3 , Figure 1B ) . Again , all 100 randomly generated sets have lower PU values ( p < 0 . 01 ) . As a final test , we randomly selected a motif in 10 , 000 exons and 10 , 000 introns ( control datasets 4 and 5 ) . Since exons and introns have differences in their nucleotide composition [29] , we split the 77 motifs into 50 exonic and 27 intronic ones according to their location in the exon-intron structure . As shown in Table 1 , we found a significantly higher single-strandedness for exonic and intronic verified motifs compared to exonic and intronic random motifs . To exclude the possibility that the maximal context length of 30 nt is inappropriate , we repeated the entire analysis with average values for context lengths 11–20 nt as well as 11–50 nt and found consistent results ( Table S2 ) . We also tested two other ways to measure the single-strandedness ( Text S1; Table S2 ) . It should be noted that even conservative controls with a lower GC content for control datasets 4 and 5 yield a higher single-strandedness for verified motifs ( Tables S2 and S3 ) . The consistent results observed for all tests led us to conclude that naturally occurring , experimentally verified splicing motifs have a significant preference to be single-stranded . Furthermore , the results from controls 2 and 3 , which keep the motif sequences , indicate that the higher single-strandedness is attributed to the flanks of the verified motifs rather than the motifs themselves . Next , we tested the hypothesis that the localization of a splicing regulatory element in a single- or double-stranded RNA structure influences splice site selection in an experimental system . We used the SXN-minigene [30] , which has been widely used to analyze the impact of sequences on pre-mRNA processing [31] . This minigene contains an artificial alternative exon between two constitutively spliced globin exons . We inserted splicing enhancer and silencer motifs located either in a single-stranded loop or in a double-stranded stem into this alternative exon ( Figure 2A ) . The single- and double-stranded motifs are located at similar positions in the exon to avoid positional effects [32] . The splicing pattern was analyzed by transfecting the constructs in HEK293 cells followed by detection of the splicing pattern by reverse transcription ( RT ) -PCR . As enhancer sequences , we used the experimentally well-characterized enhancer of the CD44 pre-mRNA ( CAACCACAA ) [33] and a pentamer ( CAAGG ) , which is the core of many computational predicted enhancers [34] . As silencers , we used the experimentally characterized hnRNP A1 ( TAGGGT ) silencer and a computationally predicted silencer ( GTAAGTGA ) [35] , which was previously experimentally verified in the HTR2C pre-mRNA [13] . As shown in Figure 2B and 2C , we observed that the function of these regulatory motifs strongly depends on their localization within an RNA conformation . Enhancers result in stronger exon inclusion when they are located in a loop compared with the location in the stem structure . Likewise , silencers located in a loop lead to stronger exon skipping compared to the loop structure . These data show that the pre-mRNA conformation influences the action of a splicing regulatory element . Next , we asked whether we could detect evolutionary selection on the structural context of computationally predicted splicing regulatory motifs . Predicted enhancers , silencers , and “splicing-neutral” hexamers were taken from Stadler et al . [36] . Our strategy to detect differences in single-strandedness of one hexamer was to compare the single-strandedness in enhancer-dependent and silencer-dependent regions . Human exons are dependent on enhancers , while the intronic regions between an authentic and a strong intronic decoy splice site ( decoy regions ) are dependent on silencers [37] . The single-strandedness of each hexamer was determined as the average PU value for up to 1 , 000 hexamer occurrences in these datasets . The direct comparison of the PU values of equal hexamers is complicated by a strong negative correlation between the GC content of the hexamer flanking regions and the PU values ( r = −0 . 64 , p < 0 . 0001 ) , as well as large differences in GC content between exons and decoy regions ( Figures S2 and S3 ) . To exclude this GC content bias , we focused only on those cases where selection is strong enough to overcome the overall correlation between PU and GC . Specifically , we considered a hexamer as selected to be single-stranded in exons if , despite more GC rich exonic flanks , the exonic PU value is higher compared to the decoy regions . Likewise , we considered a hexamer as selected to be single-stranded in decoy regions if , despite more GC rich flanks in decoy regions , the PU value is higher compared to exons . Using the fraction of splicing-neutral motifs that are selected to be single-stranded as the background fraction , we found that in exons enhancers are significantly more often selected for single-strandedness , while silencers are close to neutral motifs ( Figure 3; Table S4 ) . Strikingly , in decoy regions , silencers are significantly more often while enhancers are less often selected to be single-stranded . These results indicate that the high enhancer frequency in exons is further intensified by a tendency for single-strandedness . Likewise , silencers are abundant in decoy regions [35] , and in addition , their structural context has a tendency to be selected for single-strandedness . This is further supported by tests with other silencer-dependent regions: pseudo exons ( silent intronic regions bounded by strong splice sites [34] ) and intronic regions adjacent to authentic splice sites ( Figure 3 ) , as well as additional tests that compare enhancer/silencer/neutral motifs having the same number of GCs within and between datasets ( Tables S5 and S6 ) . Noteworthy , these differences are more pronounced for exons with weak splice sites and less pronounced for those with strong splice sites , consistent with the idea that the former have an even higher enhancer-dependency ( Figure S4 ) . These results indicate a widespread selection pressure on the structural context of splicing motifs .
It is well established that pre-mRNA sequences have enhancing or silencing activities on splice site selection , but the involvement of pre-mRNA secondary structure has not yet been systematically investigated . By analyzing the secondary structure of experimentally verified splicing motifs within their natural pre-mRNA context , we found a higher single-strandedness for these motifs . These results were confirmed by transfection experiments demonstrating that single-stranded splicing motifs exert a stronger effect on the exon usage . Finally , we found that the structural context of predicted splicing motifs is under selection in enhancer- and silencer-dependent regions . Exons , in particular those with weak splice sites , have a higher single-strandedness for enhancers , while silencer-dependent regions have higher single-strandedness for silencers . As most exons are dependent on several splicing regulatory motifs , these findings suggest a general importance of pre-mRNA secondary structures on the splicing outcome . They indicate that secondary structures are a part of the “mRNA splicing code” that determines exon recognition . Selection on secondary structures could also explain the observed constraints on synonymous codon sites [38] and the conservation of large intronic regions adjacent to alternative exons [39] . We propose that a coding exon is subjected to at least three different selection pressures: ( i ) preserving the coding sequence , ( ii ) preserving the sequence of splicing motifs , and ( iii ) preserving an appropriate structural context for these splicing motifs . Selection on the coding sequence is likely to be the strongest pressure . This is consistent with results that splicing motifs are interchangeable [36] and our finding that intronic motifs are more single-stranded than exonic ones ( Table 1 ) . Most splicing regulatory motifs are recognized by sequence-specific RNA binding proteins that make direct contacts with unpaired RNA bases [4] . This provides a mechanistic explanation for the inhibitory role of double strands on RNA–protein and mRNA–microRNA interactions [40 , 41] and for our observation that splicing regulatory motifs are located in single-stranded regions . In contrast to this general trend , a few individual motifs are located in a somewhat double-stranded conformation ( Table S1 ) . It remains to be determined which trans-acting factors bind to these motifs or whether a specific experimental situation influenced the results . It is also possible that proteins binding to the proximity of the double-stranded motifs change their single-strandedness or that RNA helicases unwind them [42] . Our findings can explain previously observed splicing effects of mutations that do not change splicing regulatory elements . For example , a silent mutation in human CFTR exon 12 that reduces exon inclusion from 80%–25% [43] does not create or destroy splicing motifs , but leads to a higher single-strandedness of existing ESSs and a lower single-strandedness of an existing ESE ( Figure 4A ) . Other examples for mutations in human HPRT1 exon 8 are shown in Figures S5 and S6 . Likewise , secondary structures can explain why mutations that change splicing motifs sometimes show no splicing effect . Most likely , the affected motifs are highly double-stranded in these cases; exemplified for motifs in rat beta-tropomyosin exon 8 ( Figure 4B ) , human MAPT alternative exon 10 ( Figure S7 ) , and human HPRT1 exon 8 ( Figures S8 and S9 ) . Thus , the structural context should be taken into account in mutagenesis experiments [44] , where observed splicing effects are usually interpreted as changes in regulatory motifs . To facilitate such analyses , we provide a web resource ( http://biwww2 . informatik . uni-freiburg . de/Software/NIPU ) that allows a comparison between PU values and ESE/ESS scores [36] . As the formation of secondary structures depends on the speed of transcription [24] , the effect of the promoter on splicing [45] might be partially explained by differential structure formation . Furthermore , secondary structures provide useful information in sequence motif finding [46] and will improve the computational detection of splicing motifs [31 , 34 , 47] and gene prediction algorithms [35] .
PU values were computed as described in [46] . Briefly , the PU value for the region a to b in an mRNA sequence is defined as where Eall is the free energy of the ensemble of all structures , Eunpaired is the free energy of the ensemble of all structures that have the complete region a to b unpaired , R is the universal gas constant , and T is the temperature . We computed Eall and Eunpaired using the partition function version of RNAfold [48] . For Eunpaired , we assure that the region a to b is unpaired by applying additional constraints ( RNAfold parameter-C ) . Other measurements of single-strandedness are described in the Text S1 . We carefully examined the literature for all splicing motifs listed in the motif database of AEDB ( http://www . ebi . ac . uk/asd/aedb/ ) [27] . We checked the consistency of the listed genes , species , and motif sequences using the respective publications . Only motifs that were demonstrated to influence splicing in their natural context were considered . We excluded motifs whose annotation was based solely on computational predictions , cases where the exact motif location is not given in the publication , or where we could not find the given motif in the respective gene . Motifs that presumably act by forming secondary structures instead of representing a protein binding site and motifs that were predicted on the basis of the splicing effect of a SNP were also excluded . Since three-dimensional structures of single-stranded RNA binding proteins indicate that they usually contact only a few residues [4] , we discarded all motifs with a length of more than 9 nt as these longer motifs are likely to contain a core binding site at a location that was not experimentally determined . We removed redundancy by requiring that each motif has a unique location within a gene . For control datasets 1–3 , we randomly generated 100 motif sets each with 77 random motifs by repeating the following procedures 100 times for each verified motif . Control dataset 1: We randomly selected a new position within the 150-nt up- and downstream flanking regions of the verified motif . The subsequence starting at this position with the same length as the verified motif was taken as the new motif . Control dataset 2: We used the same procedure as for control 1 but replaced the subsequence starting at the new position by the verified motif . Control dataset 3: Dinucleotide shuffling was done with Dishuffle [28] separately for the 300-nt up- and downstream flanks of the motif . The Dicodonshuffle program , which shuffles dicodons while preserving the coding sequence , could not be used because the output was not variable enough to yield 100 different sets . It should be noted that all control sequences were repeatedly derived from the same set of verified sequences . They are therefore not necessarily independent and the Wilcoxon rank-sum test cannot be used . To evaluate them statistically , we calculated the p-value as the fraction of random sets having a higher average PU value compared to the verified motifs . Control datasets 4 and 5: We downloaded from the University of California Santa Cruz Genome Browser the Human genome assembly ( hg17 , May 2004 ) as well as RefSeq transcript annotations ( refGene . txt . gz , January 2005 ) . Then , we extracted all internal exons and all 300-nt flanks from all introns . Redundancy was removed by using RSA-tools ( http://rsat . ulb . ac . be/rsat/purge-sequence_form . cgi ) with a minimal match length of 50 nt and at most three mismatches . We randomly selected 10 , 000 motifs within the exons ( introns ) while keeping the length distribution of the exonic ( intronic ) splicing motifs . All intronic motifs are at least 50 nt away from the splice site . The Wilcoxon rank-sum test was used to test whether verified and random motifs come from the same distribution . In all controls , we used the length distribution of the verified splicing motifs and controlled for the GC content ( Table S3 ) . Datasets with a lower and higher GC content for control 4 and 5 were generated by randomly discarding entries with a GC content of smaller and greater than 0 . 5 . Statistical tests ( Fisher's exact test , Wilcoxon rank-sum test ) were performed using R ( http://www . r-project . org/ ) . Splicing assays and cotransfection were performed as described [49] , employing the SXN derivatives . Briefly , 1 μg of indicated minigenes were transfected into HEK293 cells using the calcium phosphate method . RNA was isolated 24 h post-transfection by using the total RNA kit ( Qiagen ) . Reverse transcription was performed in a total volume of 8 . 5 μl by using 2 μl RNA , 1 μl oligo ( dT ) ( 0 . 5 mg/ml ) , 1 μl first-strand buffer ( Invitrogen ) , 10 mM DTT , 1 mM dNTP , 5 units SuperScript II RT ( Invitrogen ) at 42 °C for 60 min . To ensure that only plasmid-derived minigene transcripts were detected , subsequent amplification was performed using vector-specific primers ( fwd-CCATTTGACCATTCACCACA , rev-CACTCCTGATGCTGTTATGG ) . RT-PCR products were resolved on ethidium bromide-stained agarose gels . The ratio of exon inclusion to exon skipping was determined using the Image G program . To detect possible unspliced products , RNA samples were treated with DNAse and after reverse transcription , unspliced RNA was amplified using 4′ extension times . Using these conditions , no unspliced RNA could be observed . The exon set corresponds to the human exons used for control dataset 4 . The definition of pseudo exons was adopted from [34] as intronic region with a length between 50 and 250 nt , flanked by two splice sites having a score of at least 8 . 5 . We used a maximum entropy-based model for the quantification of the splice site strength as described in [50] , which we computed using the Web server http://genes . mit . edu/burgelab/maxent/Xmaxentscan_scoreseq . html . All pseudo exons that overlap with spliced ESTs were excluded . To search for “decoy regions , ” we scored all potential splice sites in the 100-nt intron flanks for introns longer than 400 nt using the maximum entropy model [37] . A decoy donor region is the region between a real donor site and a downstream decoy donor site ( located at least 6 nt downstream ) , which has a higher score . A decoy acceptor region is the region between a real acceptor site and an upstream decoy acceptor site ( located at least 15 nt upstream ) , which has a higher score . Decoy donor and acceptor regions were merged . The intron flank dataset consists of the regions +10 . . . +70 from a donor site and −80 . . . −20 from the acceptor site of the human exons . Redundancy in each dataset was removed using RSA-tools with a minimal match length of 30 nt and at most two mismatches . For each dataset , we randomly selected up to 1 , 000 occurrences for each of the 4 , 096 hexamers . Then , we determined the single-strandedness for one hexamer by averaging the PU values ( using context lengths 11–30 nt ) for each of the hexamer occurrences . The average GC content of one motif was computed for the 66-nt region ( 30 nt up/downstream context + 6 nt of the motif ) . The 4 , 096 hexamers were divided into exonic enhancers ( neighborhood inference ( NI ) score > 0 . 8 ) , silencers ( NI score < −0 . 8 ) , and splicing-neutral motifs ( −0 . 8 ≥ NI score ≤ 0 . 8 ) , as suggested in Stadler et al . [36] . | Almost all human protein-coding genes contain several exons and introns . Prior to translation , introns have to be removed and exons have to be joined , which happens in a processing step called splicing that generates the mature mRNA . For most genes , certain exons can be either included or excluded from the mature mRNA . It is currently not fully understood which signals are needed to accurately recognize the boundaries of exons in the intron-containing primary transcript . As in transcriptional regulation , enhancer and silencer sequence motifs are crucial for the correct recognition of exons . Splicing regulatory proteins identify these motifs in a sequence-specific manner . In general , these proteins bind to single-stranded RNA . Here , we analyzed local secondary structures of primary transcripts and found that known splicing motifs are preferentially located in a single-stranded context . Experimental tests demonstrated that motifs in single-stranded contexts have a stronger effect on splice site selection than those located in double-stranded regions . These results help to understand the action of human mutations that change the splicing pattern and indicate that local pre-mRNA secondary structures influence exon recognition . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"mammals",
"eukaryotes",
"computational",
"biology",
"molecular",
"biology",
"genetics",
"and",
"genomics"
] | 2007 | Pre-mRNA Secondary Structures Influence Exon Recognition |
Diagnosis of human African trypanosomiasis ( HAT ) remains a challenge both for active screening , which is critical in control of the disease , and in the point-of-care scenario where early and accurate diagnosis is essential . Recently , the first field deployment of a lateral flow rapid diagnostic test ( RDT ) for HAT , “SD BIOLINE HAT” has taken place . In this study , we evaluated the performance of “SD BIOLINE HAT” and two new prototype RDTs . The performance of “SD BIOLINE HAT” and 2 prototype RDTs was tested using archived plasma from 250 Trypanosoma brucei gambiense patients , and 250 endemic controls . As well as comparison of the sensitivity and specificity of each device , the performance of individual antigens was assessed and the hypothetical performance of novel antigen combinations extrapolated . Neither of the prototype devices were inferior in sensitivity or specificity to “SD BIOLINE HAT” ( sensitivity 0 . 82±0 . 01 , specificity 0 . 97±0 . 01 , 95% CI ) at the 5% margins , while one of the devices ( BBI ) had significantly superior sensitivity ( 0 . 88±0 . 03 ) . Analysis of the performance of individual antigens was used to model new antigen combinations to be explored in development of the next generation of HAT RDTs . The modelling showed that an RDT using two recombinant antigens ( rLiTat1 . 5 and rISG65 ) would give a performance similar to the best devices in this study , and would also offer the most robust performance under deteriorating field conditions . Both “SD BIOLINE HAT” and the prototype devices performed comparably well to one another and also to the published performance range of the card agglutination test for trypanosomiasis in sensitivity and specificity . The performance of individual antigens enabled us to predict that an all-recombinant antigen RDT can be developed with an accuracy equivalent to “ SD BIOLINE HAT . ” Such an RDT would have advantages in simplified manufacture , lower unit cost and assured reproducibility .
Human African trypanosomiasis ( HAT ) , otherwise known as sleeping sickness , is caused by infection with the haemoflagellate parasites Trypanosoma brucei gambiense ( in west and central Africa ) and T . b . rhodesiense ( in east and southern Africa ) [1] . Infection is initiated after the bite of an infected tsetse fly vector and progresses through an “early” stage when parasites proliferate in the haemo-lymphatic system causing a febrile illness , followed by a second or “late” stage of disease in which parasites invade the central nervous system ( CNS ) causing meningoencephalitis [2] . This latter stage is associated with neurological disturbances and ultimately death [3] . Overall , T . b . rhodesiense infections have an acute presentation with the onset of late stage and death within a few months of infection , while T . b . gambiense infections are chronic and may persist for several years although there is a spectrum of presentations within each sub-species [4] . Drug treatment , albeit with problems of toxicity , is available for both sub-species [5] , and this in combination with control programmes dealing with the vector and infections in zoonotic hosts have reduced the disease prevalence [6] . While the number of reported new cases is now less than 10 , 000 per year , it is likely that there is considerably greater burden of undiagnosed cases due to diagnostic challenges and inadequate surveillance . The clinical signs of HAT , especially in the early stages , are difficult to distinguish other infectious diseases such as malaria [7] . Initial screening of patients involves indirect diagnostic techniques , the most widely used of which is a serological test , the Card Agglutination Test for Trypanosomiasis ( CATT ) [7] . This must be followed by parasitological diagnosis , which is laborious , may require concentration techniques due to low parasitaemia , and must be carried out by skilled microscopists . The CATT is based on the agglutination by serum antibodies of lyophilized bloodstream forms of T . b . gambiense expressing variant surface glycoprotein type LiTat1 . 3 , which is expressed widely in T . b . gambiense isolates . Using undiluted blood , reported sensitivity varies between 0 . 688 and 1 and specificity between 0 . 835 and 0 . 993 [8] . Cases where specificity and sensitivity are lower are most likely due to exposure of the host to non-pathogenic trypanosomes [9] and infections with clones of T . b . gambiense that do not express LiTat1 . 3 [10] , respectively . Although a valuable diagnostic , the CATT does not meet the ASSURED criteria [11] due to a lack of robustness [8] , and the production process is also difficult to scale up . Yet , CATT is the only indirect diagnostic test that comes close to meeting the ASSURED criteria . Other immunological and molecular methods that perform well in a laboratory setting are expensive to conduct and require a combination of specialized equipment and skilled personnel ( reviewed in [8] ) . Thus , for the aims of eliminating HAT by 2020 as envisaged by the WHO Roadmap [12] and the London Declaration on NTDs [13] to be achieved , it will be essential to develop ASSURED compliant tests that are easy to produce at scale . Immunochromatographic lateral flow devices are capable of detecting low concentrations of antibodies to target antigens in biological fluids [14] , [15] . This technology may be used to develop rapid diagnostic tests ( RDTs ) that can detect anti-trypanosome antibodies in finger prick samples of human blood . RDTs based on lateral flow devices are simple to use , easy to read and have stability characteristics that allow distribution and availability in remote endemic areas . Recently the first RDT for HAT was deployed in the field . The test , developed by the Foundation for Innovative New Diagnostics ( FIND ) and Standard Diagnostics ( SD BIOLINE HAT ) , is based on a device using native variant surface glycoproteins ( VSG ) LiTat1 . 3 and LiTat1 . 5 to detect antibodies to trypanosomes [16] . A further lateral flow RDT based on these antigens ( HAT Sero-K-SeT ) has been described and developed by Coris Bioconcept [17] . In this paper we describe two further devices developed by SD and BBI Solutions ( UK ) . The first uses recombinant LiTat1 . 3 and LiTat1 . 5 antigens . While these are the same antigenic targets as used in the SD BIOLINE HAT , HAT Sero-K-SeT and CATT ( LiTat1 . 3 ) tests , the use of recombinant antigens has potential to simplify the production and reduce the costs of RDTs . The second prototype device , that uses the diagnostic potential of ISG65 [18] , is based on a combination of recombinant ISG65 and a native VSG MiTat1 . 4 [19] . ISG65 is one of two well-characterised moderately abundant invariant type-1 trans-membrane domain surface glycoproteins that is expressed in Trypanosoma brucei [20] . A summary of the three RDTs studied here is presented in Table 1 . The aim of this study was to evaluate the performance of the two new prototype RDTs in comparison to SD BIOLINE HAT in a side-by-side analysis using archived plasma samples from HAT patients and endemic controls .
This was a retrospective study . Clinical samples of heparinised plasma were obtained from 250 T . b . gambiense patients and 250 endemic controls . The sample size of infected and control groups was calculated to detect a 5% performance margin between devices at a power of 0 . 8 and confidence level of 0 . 95 . The samples were obtained from FIND-sponsored field studies in Angola , Central African Republic ( CAR ) and Uganda , and held in cryobanks in Makerere University ( Uganda ) and the University of Limoges ( France ) . Demographic details of the patient and control cohorts are presented in S1 Table . The infection status of patients was confirmed by observation of parasites in the blood , lymphatic system or cerebrospinal fluid , and this provided the reference standard . Patient samples were collected consecutively and there was no further selection for the purpose of this study . Controls were CATT negative and had no history of HAT or evidence of trypanosomes in blood when tested using the miniature anion exchange centrifugation test ( mAECT ) . After collection , samples from Angola and Uganda were kept in liquid N2 in the field and during transportation , and then stored frozen at −80°C . In CAR samples were kept at +4°C in the field and transferred to a central laboratory within 14 days where they were stored at −80°C . Samples were sent frozen on dry ice to the University of Dundee where they were blinded and randomised , and then to the University of Aberdeen for testing with the RDTs . The readers in Aberdeen were blind to the status of all samples . All clinical samples were obtained after written informed consent . Country-specific study protocols were approved by the following institutional review boards: Comissão de Ética do Instituto de Combate e Controlo das Tripanossomiases ( Angola , Meetings 12/02/08 and 12/07/11 ) , Comité scientifique chargé de la validation des proto-coles d'études et des résultats de la Faculté des sciences de la santé de l'Université de Bangui ( CAR , 9/UB/FACSS/CSCVPER/12 ) and Uganda National Council for Science and Technology ( HS 792 ) . Three RDTs were used in this study ( \ ) . Each device was run according to the manufacturers recommendations . Freshly thawed plasma ( 10 µl for SD devices , 5 µl for BBI devices ) was applied to the sample well , followed by chase buffer ( 120 µl for SD devices , 95 µl for BBI devices ) . Each plasma sample was applied to duplicate devices . The devices were incubated for 15 min ( NatSD and RecSD ) or 30 min ( BBI ) at room temperature . After the incubation period , each device was read by visual comparison using a 4-point lateral flow test standard ( Fig . 1 ) . The appearance of antigen bands ( bands 1 and 2 ) was scored as 0 , 1+ , 2+ , 3+ or 4+ depending on colour intensity . Additionally , where readers detected a faint band that was judged below the threshold of the 4-point standard ( +/- ) , the result was annotated with a score of 0 . 5 and used in a reanalysis of the accuracy of each device . The 3rd band on each device was a control band . Devices ( 4/1000 BBI devices and 1/1000 RecSD device ) where no control band was observed were discarded and the test repeated on a new device . Each RDT was scored independently by each of two readers . The readers were not aware of each other's scores until they had both been recorded . Primary and secondary readings took place within 5 minutes of each other . After all plasma samples had been run and scored , the raw data were sent to Dundee University for the sample codes to be un-blinded and identified as infected or control . Each score ( 0 , 1+ , 2+ , 3+ 4+ ) was represented by an integer between 0 and 4 . In a second run analysis we rescored all the bands that had been annotated as faint and below threshold as 1 . We established an arbitrary limit , L , to decide whether a score is positive or negative . A score was considered positive if it was greater than or equal to L . Unless otherwise stated in the results , sensitivity , specificity and accuracy were calculated at the cut-off level of L = 1 . When two antigen bands were read from a single device , the result was considered positive if either of the scores was positive . These positives and negatives were then compared with patient data and the total counts of true positives ( TP ) , false positives ( FP ) , true negatives ( TN ) and false negatives ( FN ) were summed across all patients for each reader and each duplicate device . The sensitivity , specificity and accuracy were defined as respectively . These can be alternatively defined as true and false positive rates , where TPR = Sen and FPR = 1− Spc , respectively . For the given reader and duplicate device , the errors on the above quantities were found as 95% confidence intervals of a proportion [21] . When reader data were combined , the counts of TP , FP , TN and FN were averaged across both readers and duplicate devices , which were considered as a set of 4 replicates . The mean and its 95% confidence interval were found across these replicates and errors were then propagated to Sen , Spc and Acc . The diagnostic results from each plasma sample using replicate devices or between reader 1 and 2 were tested for agreement using Cohen's kappa ( k ) . To compare the duplicate devices , we aggregated data from both readers and vice versa , to compare the readers , we aggregated data from both duplicate devices . Uncertainties of k were estimated following Fleiss et al . [22] . All errors quoted in this work are 95% confidence intervals . The difference between the means is assessed by a t-test ( assuming equal variance ) at a significance level of 0 . 05 .
Examples of each RDT used in this study , in which the different scores ( faint-sub-threshold , 1+ , 2+ , 3+ , 4+ ) and the difference between typical positive and negative results , are illustrated in Fig . 1 . Following scoring of the randomised and blinded groups of 250 HAT patient plasmas and 250 endemic control plasma samples , the sensitivity and specificity of each RDT was calculated for each reader and each duplicate test . The results presented in Fig . 2 demonstrate close agreement between readers and duplicate assays . The level of agreement was further quantified using Cohen's k ( Table 2 ) . A value of k ≥0 . 9 was found for all inter-duplicate and inter-reader agreements , which represents a very good level of agreement [23] . As there were no significant differences between the diagnostic results of the two readers using the minimum visual score of 1+ for a positive , the duplicate readings by each reader were re-analysed as four replicates and formed the basis of further analysis of RDT and individual antigen performance . The sensitivity , specificity and accuracy of each device are presented in Fig . 3a and Table 3 . Both prototype devices are not inferior to the NatSD RDT in any of these three parameters at the required 5% margin . The sensitivity of the BBI device ( 0 . 88±0 . 03 ) is however significantly superior to both NatSD and RecSD ( both 0 . 82±0 . 01 ) , with p = 9×10−4 and 5×10−4 , respectively . All devices show a performance similar to or better than the range of sensitivity ( ≥0 . 7 ) and specificity ( ≥0 . 8 ) reported for the CATT [8] . The specificity of the NatSD RDT is highest ( 0 . 97±0 . 01 ) but not significantly superior to the other devices . The accuracy of both the prototype BBI ( 0 . 91±0 . 02 ) and NatSD ( 0 . 898±0 . 009 ) devices are significantly higher than the accuracy of the RecSD device ( 0 . 884±0 . 008 ) , with p = 0 . 003 and 0 . 01 , respectively , and we found no evidence for the BBI test having different accuracy from the NatSD RDT ( p = 0 . 06 ) . When scoring the RDTs in this trial , each reader also made a record of any faint bands in the sub-threshold range ( +/- ) on the reference card ( Fig . 1a ) . These were given a nominal score of 0 . 5 , and were therefore below the cut off limit ( L = 1 ) for a positive result . In order to determine the effect of including such faint bands as positive they were rescored as 1+ . When this was done there was an increase in sensitivity for all devices with a loss of specificity ( Table 4 ) . This was most pronounced with the BBI device with a sensitivity of 0 . 96±0 . 03 ( an increase of 8% ) but a specificity of 0 . 79±0 . 15 ( a loss of 15% ) . While the recording of sub-threshold bands marginally increased inter-device agreement for duplicate devices , it led to a considerable reduction of inter-reader agreement ( Table 5 ) especially in the case of the BBI device . The scores recorded for each band on the 3 devices allowed the diagnostic potential of each of the 6 antigens to be evaluated ( Fig . 3 ) . This analysis reveals that NatSD2 ( LiTat1 . 5 ) provides the best diagnostic performance , with the highest sensitivity , specificity and accuracy . This is followed by RecSD2 ( rLiTat1 . 5 ) and RecBBI1 ( rISG65 ) , which had a comparable sensitivity , but a poorer specificity . The performance of each of the antigens was used to predict the theoretical performance of all combinations of 2 antigens on hypothetical new RDT formulations ( Fig . 3c ) . This analysis provides evidence that new antigen pairs have the potential for use in developing new improved RDTs . On examination , 6 novel combinations and the BBI device out-perform the NatSD RDT , providing significantly better sensitivity ( data above the dashed line in Fig . 3c ) . The top combination with the highest sensitivity and accuracy is NatSD2+RecBBI1 ( LiTat1 . 5+rISG65 ) , though neither its sensitivity ( 0 . 90±0 . 02 ) nor accuracy ( 0 . 92±0 . 01 ) is superior to the prototype BBI device ( p>0 . 1 in both cases ) . A combination of NatSD2+NatBBI2 ( LiTat1 . 5+MiTat1 . 4 ) is among those with the highest specificity ( 0 . 972±0 . 009 ) while retaining a high sensitivity ( 0 . 86±0 . 01 ) . The optimal pairing of recombinant antigens is RecSD2+RecBBI1 ( rLiTaT1 . 5+rISG65 ) , whose accuracy of 0 . 90±0 . 02 is not significantly different to the BBI device evaluated here . A similar analysis of performance of hypothetical 3 antigen band multiplex lateral flow devices for all three-way combinations of antigens was carried out and demonstrated no significant improvement in performance over the 2 antigen devices ( S1 Fig . ) . The scores provided by readers for each antigen band are not binary , but take into account the intensity of the band , comprising a scale between 0 and 4 . In the analysis so far , we converted them into positives and negatives using a fixed limit of L = 1 . In other words any band scored by matching the colour scale ( Fig . 1a ) as 1+ or greater is scored as positive . By increasing this limit we can study the effects of deteriorating field conditions such as poor lighting or reader eyesight in which weak bands may not be recognized . Fig . 4 shows the effect of varying L on sensitivity and specificity . The partial receiver operating characteristic ( ROC ) curves were calculated for the cut-off from L = 1 ( top right ) up to L = 4 ( bottom left ) . We note that due to a very limited range of specificity , we cannot reliably calculate the area under the curve . With increasing L ( corresponding to deteriorating field conditions ) there is an often dramatic drop in sensitivity , as fainter antigen bands are not spotted . On the other hand , there is a corresponding increase in specificity , as the faintest bands can create false positives . Of the individual antigens ( Fig . 4a ) , RecSD2 ( rLiTat1 . 5 ) and RecBBI1 ( rISG65 ) show the most moderate loss of sensitivity , down to ∼0 . 4 . The other four antigens drop in sensitivity below 0 . 2 in the limit of L = 4 . Fig . 4b shows that the NatSD RDT , while displaying consistently highest specificity , loses more sensitivity with deteriorating conditions than RecSD and BBI . Fig . 4c shows a selection of six hypothetical antigen combinations with highest sensitivity ( cf . Fig . 3c ) . The combination of the two antigens with the lowest loss of sensitivity is RecSD2+RecBBI1 ( rLiTat1 . 5 and rISG65 ) .
This study aimed to evaluate the performance of two novel prototypes and the commercially available SD BIOLINE HAT ( NatSD ) RDT in a side-by-side analysis using a panel of archived plasma samples from HAT patients and endemic controls . The sample size was designed using power analysis to be able to detect an inferiority margin of 5% . Evaluation of the RDTs was carried out by two readers in a blinded manner at a separate institution remote from where the un-blinded sample identities were held , and the two readers scored each device entirely independently of each other's readings . The samples were classified in the field at the time of collection as infected or control on the basis of robust criteria . For infected individuals , while initial identification of suspects was via the CATT test and presenting symptoms , all cases were confirmed parasitologically . All the controls had no symptoms , were negative with CATT , and had no detectable trypanosomes in the blood after the use of concentration techniques . It is possible that within this group there could have been sub-clinical cases with a very low parasitaemia , particularly if they were from parasites that did not express the CATT antigen ( LiTat1 . 3 ) or from individuals who were immunologically unresponsive to that antigen . We consider this unlikely , and indeed it may be predicted if that was the case then the RDT bands using a non-variant antigen ( BBI1/ISG65 ) and a non-CATT antigen ( LiTat1 . 5 , NatSD2 , RecSD2 ) would exhibit a higher specificity . The data ( Table 3 ) did not support this prediction . With all 3 RDTs a very high level of agreement ( Cohen's κ≥0 . 9 ) was obtained between readers and also between the duplicate RDTs used with each sample . Inter-reader agreement is in fact better than for CATT ( κ = 0 . 84 , [24] ) and also for a recent laboratory trial implementation of the loop-mediated isothermal amplification ( LAMP ) diagnostic [25] . On the basis of these results , duplicate readings by each reader were treated as replicates for the performance evaluation of each device . The sensitivity , specificity and accuracy were calculated for each device . All the devices performed well , and while the prototypes were not inferior at the 5% level in terms of sensitivity , specificity and accuracy in comparison with the NatSD , the BBI prototype was significantly more sensitive than the NatSD RDT . The sensitivity and specificity compared well to the range of published performance of the CATT ( sensitivity 0 . 69–1 . 0 , specificity 0 . 84–0 . 99 [8] ) and LAMP ( sensitivity 0 . 87–0 . 93 , specificity 0 . 93–0 . 96 [26] ) . When overall accuracy was calculated , there was no significant difference in performance between the NatSD RDT and the BBI prototype , but the RecSD prototype was significantly inferior to both . While sensitivity in this blinded study of both NatSD RDT and the BBI prototype were 0 . 82±0 . 01 and 0 . 88±0 . 03 respectively , a field trial study of the Coris Bioconcept HAT Sero-K-SeT lateral flow device [17] has recently been reported to give a sensitivity of and a specificity of ( 95% CI ) [27] . While different lateral flow platforms are used in NatSD and HAT Sero-K-SeT , they use the same antigens for detection . There are two possible reasons for the apparent discrepancy between the results for the RDTs presented here and those obtained with the HAT Sero-K-SeT . First , in this study archived plasma was used rather than whole blood . First , in this study archived plasma was used rather than whole blood . While there have been no published systematic side-by-side studies of the impact of this difference in immunodiagnostic assays for HAT , it is possible that performance of the tests would be improved when fresh blood samples are used . Secondly , there were important differences in the observation methodology . While in this study , all the results were scored completely blind , for fully described clinical and operational reasons [27] in the evaluation of HAT Sero-K-SeT about half of the samples were scored by readers already knowing a parasitological diagnosis or being aware of the clinical signs of the subjects . This has the potential to bias decisions on the reading of faint bands according to the known diagnosis or symptomatology of the subject , thus increasing the apparent sensitivity and specificity of the test . To model the effect of including all faint bands with the devices in this study , we reanalysed our data scoring every band that had been annotated sub-threshold as 1+ . This led to a significant increase in sensitivity with a performance of the BBI RDT that was not statistically different to the HAT Sero-K-SeT . In this case there was naturally a loss of specificity , as all faint bands were scored as positive . The scoring of faint bands led to a reduction of inter-reader agreement , and this is likely to be due to differences in visual acuity of different readers , given that both readers worked under identical lighting conditions . This reduction of inter-reader agreement justifies the use of the cut off of 0 . 5 ( L = 1 ) on the 4 point reference card , as sensitivity data obtained by scoring very faint bands as positive would not be reliably be duplicated by other readers . The current diagnostic procedure for HAT includes identification of suspects using a screening test , followed by parasitological confirmation [7] . This is essential , first to ensure that subjects who are false positive with the screening test do not undergo uncomfortable lumbar puncture during staging , and secondly to avoid exposing them to drug treatments that are associated with toxicity [5] . Therefore in assessing the performance of RDTs , the most important criterion is high sensitivity , as the false positives resulting from lower specificity may be excluded during parasitological confirmation . In this respect , the BBI prototype out-performed the others used in this study . It exhibited a higher sensitivity than both the RecSD and NatSD prototypes , despite a small loss of specificity ( less than 5% inferiority margin ) , and thus would be best placed to take forward for further development . This device has a further advantage over the other devices through its use of a non-variant antigen ( ISG65 ) that would be expressed in all isolates of T . b . gambiense , thus theoretically allowing higher sensitivity across a range of diverse T . b . gambiense foci . In comparison LiTat1 . 5 and LiTat1 . 3 , despite having been demonstrated to be very widely expressed [28] , will probably not be universally found in variant antigen repertoires as has been demonstrated in the field [10] . When the performance of individual antigens was analysed , the best antigen was NatSD2 ( LiTat1 . 5 ) , followed by RecSD2 ( rLiTat1 . 5 ) and then RecBBI1 ( rISG65 ) . Thus , at the individual antigen level , both the native and recombinant forms of LiTat1 . 5 were good diagnostic antigens . By increasing the cut-off limit at which an antigen band was considered positive , we demonstrated a deterioration of the performance of the antigens . This reflects the situation that could be encountered in the field if those performing the test are either not adequately trained , or they have other challenges . For example , if the results were read by a person with poor eyesight or the lighting is poor , the weakest antigen band may not be spotted , that otherwise would have been scored as 1 . In relation to this , we found that RecSD2 ( rLiTat1 . 5 ) and RecBBI1 ( rISG65 ) lose much less of their sensitivity than the other antigens . Based on the performance of the individual antigens , it was possible to predict the performance of all 2-way and 3-way combinations in hypothetical novel multiplex-RDTs [15] , based on the assumption that the antigens behave identically in performance in different combinations . This is a powerful approach to selection of antigens that should be exploited in development of the next generation of RDTs for HAT . Of the hypothetical devices , none of the 3-antigen combinations were superior to 2-antigen devices . Of the 2-antigen devices tested here , this analysis suggests that the combination in the BBI prototype ( LiTat1 . 5 and rISG65 ) is the best . However this device includes a native antigen , which presents some production and manufacturing difficulties . Yet , when we examined the hypothetical performance of devices with recombinant antigens only , it was apparent that one with rLiTat1 . 5 and rISG65 ( RecSD2+RecBBI1 ) would have a performance similar to the current best RDTs and would have an advantage of the smallest drop in sensitivity under deteriorating field conditions . Because recombinant antigens offer significant advantages in device manufacturing and reproducibility , we suggest these two antigens as important candidates for consideration in development of the next generation of RDTs for HAT . | The most prevalent species of trypanosome causing human African trypanosomiasis ( HAT ) , Trypanosoma brucei gambiense , presents a diagnostic challenge . While early diagnosis is essential for effective treatment and also to control transmission , symptoms are non-specific and parasitological diagnosis is laborious and technically difficult . Screening for HAT suspects has until now been done using the card agglutination test for trypanosomiasis ( CATT ) , which requires a cold chain and equipment , making it difficult to deploy . Thus there is an urgent need for sensitive point of care diagnostic tests that are suitable for use in rural areas in terms of stability , simplicity and cost . We describe the evaluation of 3 rapid diagnostic tests ( RDTs ) for HAT based on lateral flow devices that detect antibodies against defined parasite antigens in blood samples . We demonstrate that the SD BIOLINE HAT RDT currently being deployed in HAT endemic regions , as well as two new prototype devices , are accurate in screening for HAT . By analysing the sensitivity of each of the antigens used in the devices tested , we predict that a highly sensitive RDT based on recombinant antigens can be developed . An all-recombinant antigen RDT offers significant benefits in manufacturing reproducibility and cost , and would dramatically simplify HAT diagnosis . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"infectious",
"diseases",
"veterinary",
"diseases",
"zoonoses",
"medicine",
"and",
"health",
"sciences",
"diagnostic",
"medicine",
"biology",
"and",
"life",
"sciences",
"vector-borne",
"diseases",
"protozoan",
"infections",
"trypanosomiasis",
"parasitic",
"diseases",
"veterinary",
"science"
] | 2014 | Evaluation of the Diagnostic Accuracy of Prototype Rapid Tests for Human African Trypanosomiasis |
The extent and nature of epistatic interactions between mutations are issues of fundamental importance in evolutionary biology . However , they are difficult to study and their influence on adaptation remains poorly understood . Here , we use a systems-level approach to examine epistatic interactions that arose during the evolution of Escherichia coli in a defined environment . We used expression arrays to compare the effect on global patterns of gene expression of deleting a central regulatory gene , crp . Effects were measured in two lineages that had independently evolved for 20 , 000 generations and in their common ancestor . We found that deleting crp had a much more dramatic effect on the expression profile of the two evolved lines than on the ancestor . Because the sequence of the crp gene was unchanged during evolution , these differences indicate epistatic interactions between crp and mutations at other loci that accumulated during evolution . Moreover , a striking degree of parallelism was observed between the two independently evolved lines; 115 genes that were not crp-dependent in the ancestor became dependent on crp in both evolved lines . An analysis of changes in crp dependence of well-characterized regulons identified a number of regulatory genes as candidates for harboring beneficial mutations that could account for these parallel expression changes . Mutations within three of these genes have previously been found and shown to contribute to fitness . Overall , these findings indicate that epistasis has been important in the adaptive evolution of these lines , and they provide new insight into the types of genetic changes through which epistasis can evolve . More generally , we demonstrate that expression profiles can be profitably used to investigate epistatic interactions .
Epistatic interactions are revealed when the contribution of a mutation to an organism's phenotype depends on the genetic background in which it occurs . Epistasis plays an important role in many evolutionary theories , including those seeking to explain speciation [1] , the evolution of sex [2–5] , and adaptation [6–10] . In practice , however , epistatic interactions are usually difficult to study and their role in the evolution of organisms therefore remains unclear . Approaches based on quantitative-trait loci have been increasingly used to study epistasis [11–15] . Although these techniques have the advantage of being quite general , they suffer from some shortcomings including low statistical power , difficulty in detecting some types of epistatic interactions , and inapplicability to non-recombining organisms [11 , 16] . Recently , systems-level approaches have been developed that avoid some of these problems [17 , 18] . These approaches typically assess epistatic interactions by comparing the individual and pair-wise effects of large numbers of defined mutations , allowing the outline of functional biological modules and biochemical pathways to be determined [19–23] . To date , however , most systems-level studies have focused on deletion and other knockout mutations , and it is not clear whether findings of widespread epistasis are representative of mutations involved in adaptive evolution . Bacteria and viruses are ideal organisms with which to conduct controlled evolution experiments owing to their ease of culture and short generation times , as well as the capacity to store them in a non-evolving state from which they can later be revived to allow direct comparisons between ancestral and derived states ( reviewed in [24] ) . These experiments have allowed examination of many aspects of adaptation , including a variety of studies on the nature and extent of epistatic interactions that affect evolution [25–33] . One aspect in common to most of these studies is that they assess epistasis through the effects of mutations on fitness or some related high-level phenotype . However , at the biochemical level , it is easy to imagine that interactions might combine to create a non-linear mapping to fitness [34] . Moreover , inference of epistatic interactions from fitness alone does not usually give any insight into their underlying genetic and physiological causes . In this study , we combine a systems-level approach with a model experimental system to examine epistatic interactions that arose during the independent adaptation of two lines of E . coli to a glucose-limited minimal medium during 20 , 000 generations [35 , 36] . Specifically , we ask whether epistatic interactions occur between a key global regulatory gene , crp , and mutations that arose during the course of the experimental evolution; and , if so , to what extent these interactions evolved in parallel in the two lines . Parallel changes in independently evolving lines are of special interest because they are often characteristic of adaptive evolution , and thus call attention to those genes and phenotypic traits that have been important targets of selection [37–43] . We chose to examine interactions at the level of transcription and involving crp for several interrelated reasons . First , CRP ( cAMP receptor protein , previously known as catabolite activator protein ( CAP ) ) is a key hub in the E . coli transcriptional network . In fact , CRP is involved in more than 200 direct regulatory interactions [44–47] , which makes it a good candidate to have evolved interactions with mutations fixed during the evolution experiment . Consistent with this possibility , the evolved lines underwent substantial changes in their carbon-utilization profiles , and CRP is known to play a key role in regulating use of carbon sources in response to environmental signals including glucose concentration [48 , 49] . The crp gene has not itself acquired any new mutations during the evolution experiment , but new epistatic interactions may have arisen between crp and mutations that evolved in other genes [49] . Second , we have shown previously that transcription profiles changed during this evolution experiment and that at least some of these changes were caused by evolved differences in regulatory interactions [37] . Finally , mutations in regulatory genes known to depend on crp have been identified in the evolving populations [40 , 50] . In order to detect epistatic interactions , we compared the effect of deleting crp on the expression profiles of the ancestral and two evolved strains . Thus , we focused attention on evolutionary changes that affected the CRP regulon , even though the sequence of the crp gene itself did not change at all during the evolution experiment [37] . Expression profiles represent high-resolution phenotypes consisting of many low-level traits , thus allowing detailed comparisons of the effects of deleting crp . If the set of genes with which crp interacts remained constant during the evolution experiment , then its deletion should cause the same expression changes in the ancestral and evolved strains . However , if new epistatic interactions evolved , or if the strengths of existing interactions have changed , then we expect the crp deletion to have different effects in the evolved lines from those in the ancestor . Also , we can address the extent of parallelism in evolved changes in epistatic interactions by comparing the expression profiles of the crp mutants in the two independently evolved lines .
Deleting the crp gene caused significant reductions in the growth rate of the Ara+1 and Ara-1 evolved strains and in their ancestor ( paired t-tests , all p < 0 . 001 ) . The growth rate of the crp+ genotype was 1 . 35 ( SEM ± 0 . 03 ) times higher than that of the isogenic crp− deletion mutant in the ancestral genetic background . The corresponding growth rate differentials of the crp+ and crp− genotypes were 5 . 92 fold ( SEM ± 0 . 13 ) and 13 . 01 fold ( SEM ± 0 . 57 ) in the evolved Ara+1 and Ara-1 strains , respectively . In both cases , these differences represent significantly more deleterious effects of the crp deletion in the evolved strains than in the ancestor ( t-tests , both p < 0 . 0001 ) . These results are consistent with the possibility that interactions between crp and genes elsewhere in the genome were substantially altered during the evolution experiment . We can define the CRP regulon as the set of genes whose expression levels depend on crp . This set is defined operationally as all those genes that show significant differences in standardized expression levels depending on whether the wild-type crp+ allele or the crp– deletion allele is present . We do not attempt to distinguish whether the regulation by CRP is direct or indirect . The questions we aim to address are whether the CRP regulon changed during the evolution experiment and , if so , whether it changed in parallel ways in the two independently derived lines . To assist our interpretation of this large and complex dataset , we further define the core regulon as those genes whose expression depended on crp in all three strains , and the meta-regulon as those genes dependent on crp in any of the three strains . The Venn diagram in Figure 1 shows all the CRP regulons inferred from analyses of gene-expression data in the three strains . The small circle containing regions D , E , F and G represents the regulon inferred from the dependence of gene expression on crp in the ancestral strain . A total of 171 genes depended on crp in this strain , which is consistent with results of a previous study using a K-12 strain of E . coli [45] . The union of all seven regions , A-G , comprises the meta-regulon inferred by combining data from the ancestral and both evolved strains . The much smaller core regulon is represented by region E only , which is the intersection of the regulons independently inferred across all three strains . Scatter plots highlighting the relationships between crp-dependent genes across the three strains are shown in Figure 2 . Several features of this analysis are striking . First , the crp meta-regulon is very large , comprising some 1 , 089 genes ( about 25% of all genes ) . Second , the ancestral and core regulons are both much smaller , containing only 171 and 25 genes , respectively . Therefore , the evolved strains exhibit greatly expanded CRP regulons , compared to the ancestor . Third , and more subtly , there is much more overlap in the CRP regulons for the two evolved strains than for either evolved strain and its ancestor . The intersection for the two evolved lines is about 14% [= ( B + E ) / ( A + B + C + D + E + F ) ] , while the intersection for each evolved line and the ancestor is only 7% or 8% for Ara-1 and Ara+1 , respectively . Of course , the Venn diagram must be interpreted cautiously because each underlying datum – a test of the difference in gene expression between crp+ and crp− genotypes – is subject to statistical uncertainty . In the analyses that follow , we demonstrate that these striking features are real , and not mere artifacts , by performing a series of more constrained and rigorous statistical tests . Statistical analyses must , in general , consider two types of errors: false positives and false negatives . False positives arise when a test yields a nominally significant result , while in fact there is no real difference between the comparison groups . False negatives occur when there is a difference between comparison groups , but a statistical test indicates the observed difference is not significantly greater than expected by chance alone . False positives are especially important to consider when one performs numerous tests on large datasets , as in this study , where we compare expression levels between crp+ and crp− genotypes for several thousand genes in each of three genetic backgrounds . We now employ two different criteria to address whether the results summarized above and in Figure 1 could reflect false positives . Importantly , these tests do not focus on the narrow inferences of whether each particular gene is part of a CRP regulon as defined above , but instead these tests focus on the broad inference of whether the expression data , viewed as a whole , support the summary results stated above . The first test relies on a simple comparison between the total number of genes we identified as belonging to the CRP regulon and the number of false positives that might be included in the CRP regulon given the nominal significance level . That significance level , which was 5% in our tests , describes the likelihood of a false positive arising by chance . Given that there were 4290 genes in the analysis , one could expect 4290 × ( 1 − ( 1 − 0 . 05 ) 3 ) ) ≈ 612 false positives to be distributed approximately equally among the three genetic backgrounds that comprise the entire crp meta-regulon , where the exponent reflects the fact that independent tests were performed in three strains . This number is substantially less than the 1089 genes identified as belonging to the crp meta-regulon , and the discrepancy between observed and expected proportions is highly significant based on a binomial test ( p < 0 . 0001 ) . Therefore , while some false positives are undoubtedly included in the meta-regulon , the expression levels for many genes thus identified do indeed depend on whether the crp+ or crp− allele is present in one or more of the three strains . The second set of tests asks whether there is significant concordance across the strains in the directionality of differences in gene expression depending on which crp allele is present . Four such sign tests are possible; one test involves all three strains and the other three tests include the various pairs of strains ( Table 1 ) . Region E is the narrowly defined crp core regulon , which includes only the 25 genes whose expression level was significantly affected by the crp allele in all three strains . If this core regulon was a statistical anomaly , whereby measurement noise led to false positives , then we would expect half of the comparisons to show higher expression in the crp+ genotype than in the crp− genotype , and half to exhibit the opposite pattern . Given three strains , we expect only 0 . 53 × 2 = 25% of the genes to show the same directional effect of the crp− deletion across all three of them . In fact , 21 of the 25 genes identified as the core CRP regulon show the same directional effect of the crp deletion in all three strains , and this pattern is significantly different from the expected 25% based on a binomial test ( p < 0 . 0001 ) . We are confident , therefore , that the core regulon includes mostly genes that do , in fact , respond similarly across these strains to the loss of the crp-encoded function . Region B is of particular interest because it includes those genes that were not identified as belonging to the ancestral CRP regulon , but belong to the regulon in both independently evolved lines . Of the 117 genes in this region , 115 ( 98% ) had the same directional change in both lines with respect to the effects of the crp deletion . This parallelism is dramatically different from the 50% correspondence that one would expect if measurement noise had generated spurious associations that defined this region of overlap ( p < 0 . 0001 ) . Of the 115 genes that showed parallel changes in the two evolved lines , 78 ( 68% ) had higher expression in association with the functional crp+ allele , while 37 were more highly expressed when crp was deleted . Therefore , most genes that were recruited in parallel to the CRP regulon have increased reliance on the functional crp gene for their expression ( binomial test , p = 0 . 0002 ) . Using the same strategy outlined above , we can be confident that only one of the remaining two regions is biologically meaningful . Regions D and F represent the overlap between the regulons inferred for the ancestor and for the Ara-1 and Ara+1 evolved strains , respectively ( Figure 1 ) . Region D shows an excess of genes that change in the same direction; of the 30 genes in this region , 26 ( 86% ) change in parallel ( binomial test , p < 0 . 0001 , Table 1 ) . By contrast , the number of genes that changed in the same direction in region F was not greater than expected by chance alone ( 11 of 18 , 61% , binomial test , p = 0 . 2403 , Table 1 ) . This outcome does not mean that all of the genes in region F necessarily reflect spurious associations with the CRP regulon , but the statistical tests do not support that most of the associations are real based on the criterion of concordant directionality . It is clear that many genes were recruited to the CRP regulon in the evolved lines . But have other genes lost their dependence on crp in the evolved lines ? Of the 171 genes identified as the ancestral CRP regulon ( regions D , E , F , and G in Figure 1 ) , only 25 were significantly affected by the crp deletion in both evolved lines ( region E ) . This discrepancy may indicate that many genes became decoupled from the CRP regulon , but this inference is subject to the problem of false negatives ( i . e . , failing to confirm a genuine association owing to limited statistical power ) . We therefore reframed the question to ask whether there was a positive association between those genes that were dropped from the CRP regulon in one evolved line with those that were dropped in the other evolved line . We performed a contingency test using all the genes comprising the ancestral CRP regulon to measure this association . Indeed , there is a highly significant association between loss of a gene from the ancestral regulon in one evolved line and its loss in the other evolved line ( Fisher's exact test , p < 0 . 0001 ) . Of the genes in the ancestral regulon that were retained in the CRP regulon by the Ara+1 evolved line ( regions E and F ) , only about 42% ( 18/43 ) were dropped by the Ara-1 line . But of those genes in the ancestral regulon that were dropped by Ara+1 ( regions D and G ) , some 77% ( 98/128 ) of them were also dropped by Ara-1 . We conclude that the parallel recruitment of many genes to the CRP regulon in these evolved lines was accompanied by the parallel loss of other , albeit fewer , genes from the ancestral regulon . These analyses indicate that the CRP regulon was substantially altered in the evolved lines relative to its ancestral state , with many parallel changes in the two evolved lines . These changes could have occurred by at least three distinct mechanisms , including any combination thereof . First , both lines could have evolved changes in the pleiotropic action of CRP . These changes could occur through mutations in either crp or cyaA; the latter gene encodes adenylate cyclase , the enzyme responsible for synthesis of cAMP [49] . cAMP binds to CRP to make the active regulatory complex , cAMP-CRP . A mutation in either of these genes could change the affinity of the cAMP-CRP regulatory complex for target gene operator sequences . To test this possibility , we sequenced the upstream and coding regions of both crp and cyaA in all three strains . However , we found no mutations in either gene , indicating that the evolved changes to the CRP regulon were not caused by differences in the cAMP-CRP complex itself . A second possibility is that mutations were substituted in the evolved lines in each gene that was recruited to the CRP regulon so as to bring that gene's expression under the control of crp . Other mutations would be required for each gene that dropped out of the ancestral regulon . This hypothesis requires hundreds of mutations in each evolved line , on the order of the number of crp-dependent genes that were either recruited to or dropped from the CRP regulon during the evolution experiment . However , both theoretical and empirical evidence indicates that the number of mutations substituted in either of these evolved lines is less than 100 [36 , 51] , and thus much lower than the number of changes to the CRP regulon supported by our analyses ( Figure 1; Table 1 ) . Moreover , only some of the mutations that were substituted could be expected to interact with crp . Therefore , we consider this explanation to be insufficient to explain our finding of widespread evolved epistatic interactions with crp . A third hypothesis to account for the evolved changes to the CRP regulon is that mutations were substituted in a relatively few other regulatory genes that interact with one or more genes in the existing CRP regulon . The net effect of these few substitutions would be , in effect , to rewire the interacting gene-regulatory networks . Thus , under this model , genes under the control of these intermediaries would come under the influence of crp without requiring mutations at each of the affected loci . To explore the nature of the epistatic interactions between crp and the mutations in the evolved strains , we analyzed crp-dependent expression changes in the context of the characterized E . coli transcriptional network . The network that we used is based on a comprehensively curated database of interactions ( see Materials and Methods ) . This network comprises 1 , 217 genes , including 135 regulatory genes that mediate a total of 2 , 333 transcriptional interactions . If the increase in the number of crp-dependent genes was mediated by changes in the interactions between crp and other regulatory genes , then we expect the evolved changes to be modular , i . e . , the changes in expression should be concentrated in regulons under the control of some of those regulatory genes . To test this prediction , we quantified the expression changes within all the characterized regulons using a previously described method ( see Materials and Methods section for details ) [52] . As shown in Figure 3 , 20 of the 135 regulons we surveyed were significantly differentially regulated by crp in at least one of the evolved lines compared to the ancestor ( using a double Z-score cutoff criterion of 2 . 0 , which corresponds to p ≈ 0 . 05 [52] ) . Of these 20 regulons , 14 changed independently in both evolved lines , which is significantly more than expected by chance ( Fisher's exact test , p < 0 . 0001 ) . Moreover , all 14 of these regulons were affected in the same direction in both evolved lines ( binomial test , p < 0 . 0001 ) . Therefore , at the regulon level , as well as at the level of individual genes , epistatic changes were largely parallel . Twelve regulons – those controlled by crp , dgsA , dhaR , flhC , flhD , fliA , galS , glnL , hdfR , malT , mlc and rbsR – evolved to become less sensitive to the crp deletion . Only two regulons , controlled by lrp and ppGpp , a small molecule whose cellular concentration is controlled by the spoT and relA genes , became more sensitive to the crp deletion in the evolved lines than in the ancestor . Curiously , the numerical imbalance at the regulon level is opposite in direction to our earlier finding that many more genes were recruited in parallel to the CRP regulon by the evolved lines than were dropped from the ancestral regulon . In any case , this regulon-based approach confirms that not all genes were individually recruited to , or dropped from , the CRP regulon , but rather many of them came ‘bundled' in terms of their associations with other regulons , consistent with our third hypothesis . It is also interesting , and consistent with this interpretation , that three of the genes identified as candidates on the basis of this regulon approach have been shown previously to contain beneficial mutations in one ( spoT [37] ) or both ( malT [40]; and rbsR [50] ) of the evolved lines .
The large number of expression changes caused by the crp deletion makes it difficult , if not impossible , to determine the precise physiological basis for the much larger reduction in growth rate of the evolved lines with this deletion . In fact , there may well be multiple factors involved , stemming from the ‘bundling' of regulatory modules that appears to have occurred in the evolved lines as a large number of genes came under the control of crp . Here we suggest one possible mechanism , but we do not mean to exclude other possibilities . More highly connected computational and biological networks are thought to be more robust to the disruptive effects of environmental and genetic perturbations because they buffer against those perturbations [53–56] . Elimination of a network hub may thus reduce the network's capacity to resist perturbations . A recent analysis of the transcriptional connectivity of the E . coli genome predicted that deleting crp should substantially reduce that connectivity; on average , 3 . 96 transcriptional connections separate two randomly chosen genes in the wild-type network , increasing to 4 . 48 in an otherwise identical crp− network [56] . Experiments with the ancestral strain used here showed that disruptions of hub genes tended to reduce its robustness to environmental perturbations to a greater degree than did disruptions of other genes , although this pattern was not seen for robustness to mutations [56] . Theoretical models predict that beneficial mutations will tend to have deleterious pleiotropic effects , even if they confer a net benefit [57 , 58] . Therefore , a plausible explanation for the increased growth-rate sensitivity of the evolved lines to the crp deletion is that , without a functional crp gene , they are less well buffered to the pleiotropic side-effects of the otherwise beneficial mutations that arose during the evolution experiment . The individual gene and regulon based approaches we used to characterize the overall evolutionary changes to the CRP regulon gave rather different patterns . Considering the individual gene approach , genes recruited to the CRP regulon in at least one evolved line outnumbered losses in at least one line by 918 to 146 ( regions A + B + C vs . D + F + G , Figure 1; Table 1 ) . If new epistatic interactions involving regulatory genes were driving these changes , we expected that , in turn , we would see an overall tendency for regulons to show increased crp-dependence in the evolved lines . By contrast , only 7 regulons showed significantly increased crp-dependence in one or both evolved lines compared to 13 with reduced dependence ( Figure 3 ) . Regulons are simply groups of co-regulated genes , so how can it be that more genes , but fewer regulons , became more dependent on crp during evolution ? We consider three explanations for this apparent discrepancy . First , the fraction of crp-dependent genes that were recruited to the CRP regulon could differ between the subset of genes included in the regulon analysis and the complete set of genes . If so , changes in regulon-level crp-dependence would not be predictable from the overall pattern of gene recruitment and loss . To address this possibility , we repeated our earlier analysis , comparing the number of genes recruited to , and lost from , the CRP regulon during evolution , but considering only those genes that were included in the regulon analysis ( ∼28% of total genes ) . We found that the proportion of genes recruited to the CRP regulon was indeed lower in this subset ( Fisher's exact test , p = 0 . 0009 ) , but the number of genes recruited was still several-fold higher than the number that were lost ( 241 gains , 67 losses ) . Thus , this effect alone does not alter the expectation that an evolved increase in the number of crp-dependent genes should also increase the number of crp-dependent regulons . Second , the regulon analysis is limited to previously characterized regulons . Several mutations substituted during the evolution of the Ara-1 and Ara+1 lines have been shown to affect the physical topology , or supercoiling , of DNA [59] . Such changes are known to influence the accessibility of regulatory protein binding sites , which can thereby substantially alter the range of genes available to be regulated [60] . Indeed , one previous study has shown that there are a number of low-affinity CRP binding sites in the E . coli chromosome that can be influenced by topological changes [61] . If evolved topological changes altered the identity of genes that comprise a known regulon , it would make it harder to identify evolved changes in that regulon , because the actual set of co-regulated genes would have diverged from the presumptive regulon that is being tested . Third , the particular statistical method we used to assess the significance of expression change within regulons may itself be more sensitive to detecting loss rather than gain of crp-dependence in the evolved populations relative to the ancestor . The test compares the effect of deleting crp on gene expression within a regulon to a baseline covariance calculated by repeatedly sampling the same number of random genes ( chosen from among the 1217 genes included in our transcription regulation network ) . However , deleting crp had a much greater effect on the overall expression profiles of the evolved lines than on the ancestor's profile; thus , the distribution of covariances was broader in the evolved lines , making it more difficult to detect changes in regulon crp-dependence . Even so , we could identify significant changes in three regulons with known mutations in the evolved lines , which indicates that this analysis generated a biologically meaningful signal despite this constraint on statistical resolution . It is also difficult to know how many of the changes to the CRP regulon that occurred during the evolution of the Ara-1 and Ara+1 lines indicate new or altered molecular interactions among regulatory genes and their products , and how many reflect indirect effects of physiological changes on gene expression . In particular , it is certainly possible that some of the expression changes reflect the growth-rate differences between strains , especially given the more severe growth-rate impairment caused by the crp deletion in the evolved lines . One potential approach to test for growth-rate effects would be to compare the gene-expression profiles in these strains when they are all forced to grow at the same rate , which could be achieved by growing them in separate but identical chemostat vessels , where the dilution rate is slow enough to allow each strain to sustain itself and reach steady-state [62] . In any case , the fact that differences in growth rate may mediate some of the altered interactions demonstrated by our experiments does not alter the conclusion that these changes reflect widespread epistasis between crp and the different genetic backgrounds . In other words , it may be misguided to take a complex web of genes and their interactions , which collectively determine growth rate , and then attempt to remove the effect of growth rate in order to say that the genes do not interact . A key advantage of the regulon approach that we used is that it can identify specific genes that may mediate the new epistatic interactions . The overlap in the identity of the regulons affected by the deletion of crp marks their regulators as candidates for having substituted mutations in the evolved lines that interact epistatically with crp . Indeed , of the 14 regulatory genes controlling the regulons identified as having changed significantly in crp-dependence in both evolved genotypes , three of them – spoT , malT and rbsR – have been shown to contain beneficial mutations [37 , 40 , 50] . Two of these genes control well-characterized regulons: rbsR regulates the expression of six genes ( including itself ) that control ribose catabolism , and malT encodes a positive regulator of ten genes in two operons that control maltose catabolism . A cAMP-CRP binding site is located upstream of both of these regulatory genes [63] . The mutations in rbsR involve deletions of this gene , and the majority of the rbs operon , in both evolved lines , such that the operon is no longer induced by the cAMP-CRP complex in the low-glucose environment in which the cells were grown , explaining why the regulon became less responsive to the crp deletion . The mutations in malT produce amino-acid substitutions in the coding region of this gene in both evolved lines . Although the biochemical consequences of these mutations have not yet been determined , our results are consistent with previous findings that they reduce expression of the mal operons , making them less responsive to cAMP-CRP [40] . The underlying basis of the epistasis between spoT and crp is less clear; the spoT regulon is quite large and not as well characterized as the rbsR and malT regulons . The spoT gene influences cAMP-CRP activity by changing the intra-cellular concentration of ppGpp , thus providing a mechanism by which a mutation in this gene could affect the composition of the CRP-regulon [64] . However , there is no known reciprocal link between CRP-cAMP activity and ppGpp . Therefore , it is not clear how a mutation in spoT would change the dependence of the spoT regulon on crp . In future work , we will sequence regulatory genes in the 11 other regulons that changed their crp-dependence in both evolved lines , to determine whether they also contain mutations that might explain their altered interactions with crp . In this study , we deliberately used a mutation of large effect – a deletion of crp – to examine as fully as possible the evolved changes to the CRP regulon . However , it is also interesting to consider whether such widespread changes would result from ‘smaller' disruptions such as point mutations . Although it seems reasonable to imagine that the average effect of point mutations would be smaller , it is important to remember that the evolved epistatic interactions revealed by deleting crp were caused by mutations in other genes . We do not yet know the identity of all these mutations , but those that we have discovered include a mixture of point mutations , deletions , and insertions [37 , 39 , 40 , 50] . Targeted reversion of these mutations , including even ‘small' point mutations , are likely to have an effect on the extent of the evolved epistatic interactions . Two previous studies examining epistasis in the context of this same evolution experiment compared the effects of a set of 12 transposon-insertion mutations on the fitness of the ancestor and clones isolated from an evolved line at 1 , 000 and 10 , 000 generations [28 , 29] . Epistatic interactions between an insertion mutation and evolved mutations elsewhere in the genome were inferred when the effect of the insertion mutation on fitness was significantly different in the ancestral and evolved backgrounds . The ability of that approach to detect epistasis is limited by the resolution of fitness measurements . Moreover , fitness integrates effects over all underlying traits , such that epistatic interactions at lower levels may be obscured . The approach used here overcomes this limitation by treating the expression level of each gene as a phenotypic trait . The large number of such traits that can be surveyed simultaneously and with considerable precision may allow the detection of more subtle differences in the effect of a mutation in different genetic backgrounds , and this approach may thus provide more sensitive detection of epistatic interactions . Furthermore , the intimate relationship between genotype and expression levels offers greater opportunity for mechanistic insights into the nature of the epistatic interactions thus discovered . In conclusion , we have used expression arrays to analyze the extent of epistatic interactions occurring between crp and mutations that fixed during 20 , 000 generations of evolution in E . coli . These epistatic interactions are widespread , and many of them emerged in parallel in independently evolving lines , indicating that the derived mutations – and perhaps also the altered web of interactions – contribute to the bacteria's adaptation to the experimental environment . Indeed , some of the parallel changes in epistatic interactions involve known regulons in which key regulatory genes have been shown to harbor beneficial mutations in the evolved lines . Finally , the use of expression profiles to detect epistatic interactions has the advantage that it does not require a priori knowledge of the molecular and physiological bases of interactions that are discovered , although this approach can be integrated with such knowledge to inform our understanding of genotypic and phenotypic variation .
Twelve lines of E . coli B were founded from two ancestral types differing only by a selectively neutral marker . These lines were propagated at 37 °C in a glucose-limited minimal medium for 20 , 000 cell generations [35 , 36] . Expression profiles were obtained from crp+ clones isolated from two of these lines ( designated Ara-1 and Ara+1 ) after 20 , 000 generations of evolution and from the ancestor to one of those lines . These profiles were compared to otherwise isogenic crp− derivatives of these same three clones , constructed as described below . Only one ancestral type was examined here because no difference was detected between the expression profiles of the Ara- and Ara+ ancestors in a previous study [37] . During the evolution experiment , four of the twelve lines evolved mutator phenotypes; however , the mutation rate of both lines used in this study remained at the low ancestral level [51 , 65] . The crp ( cAMP receptor protein ) gene was deleted from the ancestral and evolved strains using a suicide plasmid-mediated approach described previously [50 , 66] . Briefly , a 1 , 016-bp PCR product containing a 591-bp in-frame crp deletion allele was cloned into pDS132 , a derivative of pCVD442 [50 , 66 , 67] . This plasmid was transferred by conjugation into recipient cells , in which the deletion would be introduced , and chloramphenicol-resistant cells ( resulting from chromosomal integration of the non-replicating plasmid ) were selected . Resistant cells were streaked onto LB + sucrose agar to select cells from which the plasmid had been lost through intra-chromosomal recombination ( pDS132 encodes the sacB gene conferring susceptibility to sucrose ) . These cells were then screened for the presence of the crp deletion allele by PCR and hybridization experiments . Several independent crp− strains were made in each background . All replicate crp− strains exhibited the same characteristic change in colony morphology relative to their respective crp+ progenitors . The crp− strains used in this study were further verified by sequencing the crp deletion allele and surrounding regions . This sequencing confirmed the deletion allele and ruled out the possible introduction of secondary mutations in the flanking regions of the introduced allele . Prior to growth rate assays , all strains were grown for one day under the same conditions that prevailed during the evolution experiment , in order to ensure they were similarly acclimated to the culture conditions ( 37 °C , minimal medium supplemented with 25 mg/L glucose ) . Following this pre-incubation , strains were diluted 1:100 into 300 μL of fresh media contained in a well of a standard 96-well microtiter plate . This plate was incubated with periodic shaking at 37 °C and the change in optical density at 450 nm ( OD450 ) was monitored using a Versamax Pro spectrophotometer . The maximal growth rate of each genotype was estimated by calculating the slope of the natural log of OD450 against time during the period of most rapid growth . Reported growth rates are the average of 10 independent estimates for each strain . Expression profiles were generated using Panorama E . coli cDNA macroarray membranes ( Sigma-Genosys ) . Prior to RNA extraction , all strains were acclimated to the same conditions used during the evolution experiment , then diluted 1:100 into fresh medium and grown to mid-exponential phase . Cells were harvested using 0 . 45 μm filter units ( Nalgene ) and re-suspended in a 1:1 mix of buffer and RNAlater RNA stabilizing solution ( Ambion ) . RNA was obtained using the Qiagen RNeasy system , including an additional step to remove contaminating genomic DNA using the Qiagen on-column DNAse kit . Subsequent cDNA production , labeling , and hybridization were performed according to the instructions of the macroarray membrane manufacturer . Following hybridization , membranes were washed as per manufacturer instructions and exposed to Kodak PhosphorImager screens for 24 h . Exposed screens were scanned on a STORM 840 PhosphorImager ( Molecular Dynamics ) . Image files were analyzed using Arrayvision software ( Version 6 . 0 , Imaging Research ) , and the output exported to Microsoft Excel for manipulation . Three independent mRNA preparations and hybridizations were performed for each strain . All RNA isolation and hybridization steps were done in complete blocks to control for any effects of day-to-day variation on results . The Panorama macroarray consists of 4 , 290 PCR products , each corresponding to an individual open reading frame from E . coli K-12 , spotted in duplicate onto a nylon membrane . To measure the expression level of each gene , we subtracted the average background from the mean of the two readings to calculate its adjusted expression level . To standardize expression and thereby account for possible differences between arrays , the adjusted expression levels were normalized to the median gene expression value obtained for the array . Standardized values were log10-transformed , and paired t-tests were performed using the transformed values from replicate arrays to identify genes whose expression had changed significantly . Expression values obtained from RNA samples isolated on the same day for crp+ and crp− derivatives of each strain were paired in this analysis . These tests were performed using the paired-data analysis option of the web-based Bayesian statistical program Cyber-T ( http://cybert . microarray . ics . uci . edu/ ) [68 , 69] . A sliding window of 201 genes and a confidence limit of 2 were used in this analysis . This program reduces the number of false-positive differences expected in a comparison of replicate expression profiles from two strains through the use of Bayesian estimates of variance among replicate gene measurements . The use of a formal statistical test avoids arbitrary cut-offs , but it is still expected to identify many false-positives . To account for this , additional tests were performed on subsets of the data , as described in the text , using Microsoft Excel and the SAS frequency procedure ( SAS Institute V8 , 2000 ) . These tests sought to compare observed distributions of crp-dependence to those expected by chance alone . Binomial tests were used to test the directionality of crp-dependent gene expression changes , reflecting our null hypothesis that genes identified by chance as being significantly changed in two genetic backgrounds would be equally likely to change in the same or in different directions . We also performed Fisher's exact tests to compare the numbers of genes changed in each direction to expectations given observed inequalities in the number of up- and down-regulated genes in the corresponding CRP regulons . In all cases , the outcomes of these Fisher's exact tests were qualitatively consistent with the reported Binomial tests ( data not shown ) . A transcription interaction network for E . coli was compiled from regulatory interactions downloaded from regulonDB [63] ( data were accessed and downloaded on June 22 , 2006 ) , and from relevant data presented elsewhere [70 , 71] . We define regulons as groups of genes directly regulated by the regulatory genes present in these datasets . Regulatory genes were defined as genes with outgoing links to at least two other genes . Because our aim was to increase the sensitivity of measuring regulatory changes caused by altered activity of any regulatory gene , those genes that were co-regulated within operons were considered as having separate links to any gene that controlled the transcription of that operon . The regulatory genes spoT and relA were considered together because they combine to control the level of ( p ) ppGpp , a key regulatory molecule that binds to RNA polymerase and alters its affinity for transcriptional start sites [72 , 73] . Although spoT and relA control the same set of genes through their action of modulating levels of ( p ) ppGpp , they do so in response to different environmental conditions [74] . Moreover , they are involved in distinct sets of protein-protein interactions [75 , 76] . Therefore , despite the apparent regulatory redundancy , we expect that a mutation altering the regulatory capacity of either gene would be discernable through changes in the expression of co-regulated genes . In all , our transcription regulation dataset comprised 2 , 333 interactions and included 1217 of the 4290 genes assayed on the Panorama macroarray . To measure the extent and significance of crp-dependent expression change within the 135 regulons identified in this dataset , we used an algorithm described by Balazsi et al . [52] . This algorithm calculates a double-Z score for each regulon based on the covariance of changes in expression caused by the crp deletion among genes in the regulon relative to an average background covariance calculated over 1 , 000 samples of an equal number of randomly chosen genes . This score therefore provides a measure of expression change within a regulon that controls for differences between paired crp+ and crp− strains in their overall measurement variability . We also considered more extensive interaction groups that included indirectly regulated genes ( origons in the terminology of ref . 52 ) . Results obtained using these groups were qualitatively similar to those obtained in the regulon analysis ( data not shown ) . | The effect of a genetic mutation can depend on the genotype of the organism in which it occurs . For example , a mutation that is beneficial in one genetic background might be neutral or even deleterious in another . The interactions between genes that cause this dependence—known as epistasis—play an important role in many evolutionary theories . However , they are difficult to study and remain poorly understood . We used a novel approach to examine the evolution of interactions arising between a key regulatory gene , crp , and mutations that occurred during the adaptation of a bacterium , Escherichia coli , to a laboratory environment . To do this , we measured the effect of deleting crp on the expression of all genes in the organism , providing a sensitive measure to identify new interactions involving this gene . We found that deleting crp had a dramatic and parallel effect on gene expression in two independently evolved populations , but much less effect in their ancestor . An analysis of these changes identified a number of regulatory genes as candidates for harboring beneficial mutations that could account for the parallel changes . These findings indicate that epistasis has played an important role in the evolution of these populations , and they provide insight into the types of genetic changes through which epistasis can evolve . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"evolutionary",
"biology",
"genetics",
"and",
"genomics",
"microbiology",
"eubacteria"
] | 2008 | Expression Profiles Reveal Parallel Evolution of Epistatic Interactions Involving the CRP Regulon in Escherichia coli |
Meiotic recombination is an essential biological process that generates genetic diversity and ensures proper segregation of chromosomes during meiosis . From a large USDA dairy cattle pedigree with over half a million genotyped animals , we extracted 186 , 927 three-generation families , identified over 8 . 5 million maternal and paternal recombination events , and constructed sex-specific recombination maps for 59 , 309 autosomal SNPs . The recombination map spans for 25 . 5 Morgans in males and 23 . 2 Morgans in females , for a total studied region of 2 , 516 Mb ( 986 kb/cM in males and 1 , 085 kb/cM in females ) . The male map is 10% longer than the female map and the sex difference is most pronounced in the subtelomeric regions . We identified 1 , 792 male and 1 , 885 female putative recombination hotspots , with 720 hotspots shared between sexes . These hotspots encompass 3% of the genome but account for 25% of the genome-wide recombination events in both sexes . During the past forty years , males showed a decreasing trend in recombination rate that coincided with the artificial selection for milk production . Sex-specific GWAS analyses identified PRDM9 and CPLX1 to have significant effects on genome-wide recombination rate in both sexes . Two novel loci , NEK9 and REC114 , were associated with recombination rate in both sexes , whereas three loci , MSH4 , SMC3 and CEP55 , affected recombination rate in females only . Among the multiple PRDM9 paralogues on the bovine genome , our GWAS of recombination hotspot usage together with linkage analysis identified the PRDM9 paralogue on chromosome 1 to be associated in the U . S . Holstein data . Given the largest sample size ever reported for such studies , our results reveal new insights into the understanding of cattle and mammalian recombination .
In eukaryotes , meiotic recombination through reciprocal crossovers is an essential biological process that ensures the proper segregation of homologous chromosomes during meiosis [1–4] . Any mistakes or aberrations during this process can result in aneuploidy , a potentially deleterious outcome [5 , 6] . Mechanisms of meiotic recombination are therefore conserved , and the location and frequency of meiotic crossovers are biologically regulated [7] . In addition , meiotic recombination contributes to genetic diversity by reshuffling maternal and paternal genetic alleles into the next generation , which provides novel combinations of genetic variants to selection and evolution [8] . Considerable variation in recombination rate among individuals has been documented from pedigree-based studies in humans and mice [9 , 10] . In humans , several genes have been identified to be associated with individual-level variation in recombination rate , including CPLX1 , RNF212 and PRDM9 [9 , 11–13] . Additionally , locations of recombination crossovers are not uniformly distributed along the genome , but are mainly regulated by the PRDM9 protein during the initiation of meiotic recombination [11–13] . Recombination hotspots , i . e . , short chromosome regions where crossovers occur more frequently than in other regions , have been identified in humans and mice [14–16] , and PRDM9 has been found to be associated with the percentage of crossovers in hotspots that is termed as ‘hotspot usage’ [9 , 12] . Furthermore , a recent study reported differences in locations of double-strand breaks between different PRDM9 alleles in humans [17] . While all these findings are restricted to humans and mice , studies in other mammalian species can provide comparative information for understanding recombination , especially in those that have the PRDM9 homologue such as cattle . Recombination rate varies considerably between the two sexes in many species , particularly in mammals [8 , 18 , 19] . In humans and mice , females have higher recombination rates or more crossovers than males [18 , 20–22] . In sheep , however , males tend to have more crossovers [23] . In cattle , previous studies have failed to find sex difference , as these studies were limited by the small to moderate sample sizes and numbers of genetic markers [24–28] . Two cattle studies on male recombination using the bovine 50K SNP chip were recently reported . Sandor et al . characterized cattle male meiotic recombination using 10 , 192 bulls from the Netherlands and 3783 bulls from New Zealand with 19 , 487 SNPs in common between the two groups . [29] . Weng et al . reported male recombination features and related genetic loci in beef cattle with a moderate sample size ( 2 , 778 Angus and 1 , 485 Limousin sire-offspring pairs ) [30] . While these studies provided insights into male recombination in cattle , neither study had information about female recombination to provide a female recombination map . Large scale study on sex differences in genome-wide recombination including the genetic control of female recombination remains unavailable . Sandor et al . ( 2012 ) reported an association between recombination hotspot usage and PRDM9 in bulls , but localized the gene to chromosome X . However , PRDM9 has four paralogues in the bovine genome and previous studies have found signals of positive selection associated with the copy on chromosome 1 [31] . This study provides clear evidence that the PRDM9 paralogue on chromosome 1 is associated with recombination hotspot usage in the U . S . Holstein population . Cattle are a uniparous species where the population structure generally lacks female recombination information that requires at least three generations . The United States Department of Agriculture ( USDA ) has received genotypes for over half a million Holstein cattle spanning several generations for genomic selection . This large multi-generational structure overcomes the problem of lacking female recombination information in cattle and provides a unique opportunity to study genome recombination in both females and males with unprecedented statistical power . Utilizing this large sample , the present study seeks to comprehensively survey the sex-specific patterns of meiotic recombination and to understand the genetic basis of individual differences in recombination in males and females . We also aim to generate the first SNP-based recombination maps in the two sexes and to evaluate the trend of meiotic recombination features that might be associated with the long-term artificial selection for dairy production .
The recombination map was calculated as the recombination rates between adjacent SNPs based on the AGIL SNP map . Using an EM algorithm [9] , we constructed cattle sex-specific recombination maps for all bovine autosomes , spanning 25 . 5 Morgans in males and 23 . 2 Morgans in females ( Supplemental File ) . These are the first such cattle recombination maps using genome-wide SNP markers . Also for the first time , we identified a significant sex difference in cattle recombination rate , with the male map being 2 . 3 Morgans ( 10% ) longer than the female map . Moreover , the male map was longer than the female map for every chromosome , with the difference ranging from 0 . 007 Morgans ( 1 . 4% ) for chromosome 27 to 0 . 188 Morgans ( 26 . 5% ) for chromosome 19 ( Fig 2 ) . The male and female recombination maps were positively correlated across the SNP intervals ( R = 0 . 636 ) , which is similar to the results in humans [9] . To evaluate whether SNP information measures differ between the two sexes , we compared the distribution of the number of informative SNPs in the two sexes and found no difference ( S1 Fig ) . The sex-specific recombination maps comprised of 59 , 309 SNP markers for all 29 bovine autosomes , with an average physical distance of 42 . 4 kb between adjacent SNPs and an average genetic distance of 0 . 043 cM in males and 0 . 039 cM in females , respectively . The 59 , 309 autosomal SNPs covered a total physical length of 2 , 516 Mb with 986 kb/cM in males and 1 , 085 kb/cM in females . The physical and genetic lengths of bovine autosomes had strong positive correlations of 0 . 960 in males and 0 . 985 in females across the 29 autosomes ( Fig 2 ) . Our estimated male map size of 25 . 5 Morgans for autosomes was consistent with a recent study using 10 , 106 cattle sperms and a 50K SNP chip that had an estimated genetic map length of 25 . 7 Morgans [29] . To evaluate the power of detecting crossovers in our study , we conducted simulations using the same settings , including a three-generation family structure and 50K SNP chip . The result showed that the power for identifying a crossover was 97 . 6% . Due to the large sample size of the study , our recombination maps extended far to the two ends of the chromosomes and an obvious decline in recombination rate was observed at a distance of 2 Mb . Cattle male and female recombination rates are unevenly distributed along the genome ( Fig 3 ) , consistent with the observations in humans and mice [9 , 11 , 12] . By defining hotspots as SNP intervals with recombination rate >2 . 5 standard deviations greater than the mean [29] , we identified 1 , 792 hotspots for males and 1 , 885 hotspots for females , with 720 of them shared between sexes ( i . e . , 40 . 2% for males and 38 . 2% for females were shared ) . The difference in recombination rate in subtelomeric regions between males and females largely explains the low sharing of hotspots between the two sexes ( Fig 3 ) . The male recombination hotspots covered 3 . 0% of the physical length of the autosomes but accounted for 25 . 1% of the total male recombination events . The female hotspots comprised of 3 . 2% of the autosomes but accounted for 25 . 6% of the total recombination . The 720 shared hotspots accounted for a similar amount of the total recombination events in males ( 11 . 2% ) and females ( 11 . 1% ) . The low sharing of hotspots between the two sexes ( 38 . 2% ~ 40 . 2% ) could have allowed opportunity for sire selection for combined genetic material not as easily obtainable in females , noting that sire selection has been the primary genetic selection in dairy cattle and has been highly efficient . Although our approach has a measure to minimize the effect of genotyping and genome assembly errors by requiring at least three informative markers for a crossover call , we caution that big chunks of genome assembly errors may still lead to spurious recombination hotspots [34] . To filter false-positive hotspots , we conducted pairwise linkage analysis using Locusmap [35] and checked the linkage disequilibrium ( LD ) pattern between each of the SNPs near a hotspot and all other SNPs on the same chromosome . As a result of this analysis , nine SNPs that showed suspicious linkage and LD patterns were removed from all analyses ( S2 Fig and S4 Table ) . These results were consistent with the observation that many of the hotspots with recombination rate greater than 0 . 01 were likely due to genome assembly errors [34] . We assessed the relationship between recombination rate and chromosomal locations , as recombination rates are known to differ considerably across chromosomal locations , including telomeres and centromeres . After removing the 2-Mb regions at the two ends of each chromosome where the power for identifying crossovers was reduced due to low SNP coverage [9] , we fitted a smooth spline model of recombination rate on relative chromosomal positions , to investigate how recombination rate changes along the chromosome in each sex separately . All cattle autosomes are acrocentric with the centromere located at the beginning and the telomere at the end of each chromosome [36] . Males had a considerably higher recombination rate than females in the subtelomeric regions , ~15% of the chromosome to the telomeric end ( Fig 4 ) . Consistently , a male-biased recombination near telomeres was observed for each of the 29 autosomes ( S3 Fig ) . More importantly , the subtelomeric regions accounted for all the sex differences in genome-wide recombination rate , showing a difference of 2 . 4 Morgans in recombination rate between males and females in the last 15% of the autosomal chromosomes near the telomere . Although a higher male recombination rate in subtelomeric regions has been shown in humans and mice [9 , 19 , 37] , this is the first such report in cattle . As expected , we also observed a very low recombination rate near the centromere , the beginning of each chromosome , for both males and females . Interestingly , the middle of a chromosome had a decreased recombination rate , although the centromere is far from the middle . This low recombination rate in the middle of a chromosome was not universal across all chromosomes , but more pronounced for chromosomes 9 , 10 , 11 , 13 , 15 , 16 , 19 and 23 ( S3 Fig ) . To evaluate whether crossover interference contributed to the bi-modal distribution of recombination events along the chromosome ( Fig 4 ) , we separated single-crossover and double-crossover chromosomes and then generated smooth-spline plots for recombination rate along chromosomal locations for these two sets of chromosomes separately ( S4 Fig ) . A lower recombination rate in the middle of chromosomes for double-crossover chromosomes than single-crossover chromosomes indicates the effect of positive crossover interference , consistent with the observation in mouse [38] . We conducted GWAS analysis of genome-wide recombination rates for 3 , 224 sires and 53 , 125 dams separately . We corrected for the effect of SNP number so that the average number of recombination events was the same regardless of the genotyping assay used ( S2 and S3 Tables ) . Due to the intensive use of artificial insemination , males had more progeny than females , resulting in more recombination measurements from each sires but a smaller number of sires than dams in the sample , even though the total numbers of meioses were the same between the two sexes . For each animal , the average number of recombination after correction across all meioses was used as a phenotype and the number of measurements/meioses was used as a weight . We tested the association between sex-specific genome-wide recombination rates and 310 , 790 imputed SNPs using a linear mixed model . We used variable residual variances that are inversely proportional to the weight and a genome-wide significance level of 1 . 6×10−7 from the Bonferroni correction . A total of thirteen loci were identified to have significant effects on recombination rate , four loci on male recombination rate and nine loci on female recombination rate , among which three loci were shared between the two sexes ( Table 1 and Fig 5 and S5 Fig ) . The three shared loci ( one on chromosome 6 and two on chromosome 10 ) were among the strongest associations ( Table 1 and Fig 5 ) . The top SNP at the chromosome 6 locus , rs110253089 ( Pfemale = 2 . 95×10−51; Pmale = 7 . 34×10−30 ) , was located in the intron of the CPLX1 gene , which was associated with genome-wide recombination rate in humans [39] . Using this SNP as a covariate in a conditional analysis , other originally associated SNPs at the same locus were no longer significantly associated with recombination rate , suggesting a potential single underlying QTL at this locus . We found two significantly associated loci on chromosome 10 . The associations at the first locus peaked at SNP rs137264867 ( Pfemale = 2 . 62×10−51; Pmale = 1 . 07×10−16 ) , which was located downstream of PABPN1 . A conditional analysis identified four independently associated SNPs at this locus , spanning a 9-Mb window that consisted of several meiosis-related genes , including REC8 , REC114 , and FMN1 ( Table 1 ) . The REC8 gene has been previously reported to associate with recombination rate in cattle [29] . The top associated SNP at the second locus on chromosome 10 was rs43640523 ( Pfemale = 8 . 96×10−23; Pmale = 9 . 10×10−13 ) . This SNP was located 10 kb downstream of NEK9 that was related to spindle organization and cell cycle progression during mouse oocyte formation [40] . A conditional analysis adjusting for the top SNP in this locus indicated a single underlying QTL in this region . Although the top three associated loci were shared between sexes , differences between sexes were observed among the less significant associations . We observed a trend of smaller P-values in females in general ( Table 1 ) , indicating a difference in statistical power between the two sexes . In total , we identified six loci associated only in females and one locus associated only in males ( Table 1 and Fig 5 ) . The female-biased association on chromosome 1 peaked at rs110661033 ( Pfemale = 4 . 14×10−26 ) . This SNP was also nominally associated in males ( Pmale = 0 . 013 ) , exhibiting the same direction of effect in both sexes . Taking into account the difference in power , this association is more likely to be shared between sexes than a female-specific effect . Moreover , this association was 35 kb downstream of the PRDM9 gene , which has been associated with both recombination rate and hotspot usage in the two sexes [9 , 11 , 12] . Similarly , the association at rs109665521 ( Pfemale = 1 . 53×10−13; Pmale = 6 . 68×10−5 ) in the first locus on chromosome 3 , showing the same effect direction in both sexes , are less likely to be female-specific . Potential sex-specific associations with recombination rate were found at five loci on chromosomes 3 , 26 and X , with one male-specific and four female-specific ( Table 1 ) . SNP rs137337293 on the X chromosome was associated only in the male ( Pmale = 1 . 27×10−7 ) . Among the four female-specific associations , two were inside or near genes closely related to the meiotic pathway . Rs136642773 ( Pfemale = 1 . 04×10−10 ) was located in the intron of MSH4 , which is a meiosis-specific MutS homologue that affects crossing over [41 , 42] . SNP rs133252805 ( Pfemale = 1 . 25×10−7 ) was upstream of SMC3 that encodes a protein related to meiotic chromosomes and synaptonemal complexes [43] . The other two female-specific associations were observed at rs109452965 near CEP55 ( Pfemale = 8 . 6×10−17 ) and rs42382307 on the X chromosome ( Pfemale = 2 . 8×10−8 ) . To comprehensively evaluate the associations and estimate their effects , we conducted a joint analysis by including all the significantly associated SNPs in one model as fixed effects ( Table 1 ) . As expected , the associated P-values become smaller from the joint analysis than in the single-marker analysis , for those independent associations in both males and females because of the reduced residual errors [44] . The largest difference/ratio between the P-values of the single-marker and joint analyses was 1014 for the association at rs137264867 in females . Based on the 1 , 792 male and 1 , 885 female recombination hotspots , we calculated the proportion of recombination events occurring in the hotspots genome-wide , i . e . , hotspot usage , for the sire and dam in a three-generation family . To increase accuracy , we only included the high-quality meioses , where the offspring , the parent and the grandsire were genotyped by the 50K SNP chip . We also used the average of multiple measurements of hotspot usage as the phenotype , resulting in a sample size of 1 , 772 and 12 , 756 for males and females respectively . We then tested the association between hotspot usage and each of the 310 , 442 imputed SNPs . To evaluate the effect of different definitions of hotspot , we tested a range of cutoff values , 2 , 2 . 5 , 3 , 5 , and 10 standard deviations , and found that the cutoff value of 2 . 5 standard deviations had the clearest signal for the association between PRDM9 and hotspot usage . The GWAS results indicated that recombination hotspot usage was much less polygenic than recombination rate , because we identified a single associated locus in both males and females for hotspot usage ( Table 2 and Fig 6 and S6 Fig ) and thirteen associated loci for recombination rate ( Table 1 ) . The top SNP was rs110661033 located 35 kb downstream of PRDM9 ( Pfemale = 2 . 20×10−134; Pmale = 6 . 59×10−13 ) . This SNP was also associated with genome-wide recombination rate ( Table 1 ) . Animals that carry one copy of the minor allele ( G ) of this SNP ( MAF = 0 . 09 ) , on average , showed a decrease of 2% and 1% in hotspot usage in females and males respectively ( Table 2 ) . However , the effect of the association with recombination rate was just the opposite: one copy of the major allele ( A ) had a decrease of 0 . 52 and 0 . 34 crossovers in female and male recombination rates respectively . By adjusting for the effect of rs110661033 , the conditional analysis identified a second , independent association at rs132965246 in males only ( Pmale = 1 . 03×10−15 ) . In females , this SNP was only nominally associated with hotspot usage from the joint analysis , but with an opposite effect ( Table 2 ) . Note that the locations of several associated SNPs near PRDM9 were different between the UMD3 . 1 assembly and the USDA-AGIL SNP coordinates: UMD 3 . 1 assembly placed PRDM9 near the middle of chromosome 1 while AGIL map moved these PRDM9-linked SNPs to the end of the chromosome ( Fig 6 ) . The Baylor Btau_4 . 6 . 1 genome assembly also placed the PRDM9 gene to the end of chromosome 1 [45] . To further validate the location of PRDM9 , we investigated the pairwise linkage disequilibrium patterns between the top associated SNP , rs110661033 , and all other SNPs on chromosome 1 . The results supported that the PRDM9 associated with cattle recombination is located to the end of chromosome 1 , because rs110661033 had strongest LD with SNPs at the end of the chromosome and lower LD with SNPs away from this mapped location ( S7 Fig ) . After adjustment for the effects of SNP array and inbreeding in both sexes and for number of offspring in bulls , we identified a decreasing trend in genome-wide recombination rate in males in the past decades ( S8 Fig ) . As described in previous sections , we estimated the genome-wide recombination rate and hotspot usage for bulls that were born between the years 1970 and 2012 and for cows born between 1990 and 2012 , allowing us to evaluate the trends of these recombination features over the years . Male recombination rate dropped from 27 . 1 to 24 . 7 Morgans from 1974 to 1990 and then continued the decrease but with a slower speed after 1990 . The decreasing trend in females is not as clear as that in males ( S8 Fig ) . This declining recombination rate with artificial selection is consistent with the recent empirical evidence that domestic animals exhibit lower recombination rate than their wild counterparts [46] . In addition , the decreasing trend in recombination rate partially explained the shorter recombination maps compared with existing maps because a major proportion of the cattle used in this study were born after 2000 . However , recombination hotspot usage showed a non-significant , but two-stage trend in both sexes: an increasing trend before 2006 and a reduction after 2006 ( S9 Fig ) .
The next decade is predicted to witness a substantial growth in global population and a possibly larger increase in demand for animal products due to spreading affluence . To meet the growing demand for meat and dairy products , cattle industries have begun to adopt alternative strategies for increasing production through genomic selection [47 , 48] . Understanding of the genomic features of cattle , including the mechanisms of meiotic recombination , genetic loci that are associated with recombination , and the high-resolution recombination maps , is directly relevant for genomic evaluation [49 , 50] . Based on SNP genotypes from over half a million Holstein cattle with pedigree information , the present study reported recombination maps for both males and females , identified recombination hotspots in each sex , provided in-depth insights into the genetic basis of individual differences in recombination , and demonstrated a decreasing trend over time in recombination rate that coincided with a period of steady selection response to artificial selection for milk production . This study reported cattle-specific features of recombination as well as features that are shared between cattle and other mammals . We provided compelling evidence of sex differences in recombination rate in cattle , which is consistent with the results that were previously reported for most of the mammalian species [9 , 18 , 51] . However , in striking contrast to humans and mice where the male recombination map is shorter , our results demonstrated that the male recombination map of cattle was over 10% longer than that of the females . We also showed that the higher recombination rate in males was most pronounced near the telomeres . Interestingly , higher male recombination rates in the subtelomeric regions have been consistently reported in humans and mice , despite the differences between cattle and other mammals in the overall patterns of sex-biased recombination rate [9 , 19 , 37 , 38] . In addition , a decrease in recombination rate at the centers of acrocentric chromosomes in cattle possibly due to crossover interference was also observed in humans ( S10 Fig ) and mice [38] . A further comparison between our GWAS results and two QTL mapping studies in mouse revealed a common QTL region encompassing MSH4 that was orthologous between cattle and mouse [52 , 53] . Although the biological significance for a longer male map in contrast to most mammalian species is unclear , we speculate that cattle domestication , which was estimated to have begun approximately 10 , 000 to 11 , 000 years ago [54] , and the intense artificial selection targeting specific traits thereafter , could be a plausible explanation . In the past , the breeding practices in dairy cattle have put more selection pressure on bulls than on cows . Based on several theories of recombination rate evolution , this male-biased selection may lead to a higher recombination rate in bulls if selection has a direct or indirect , positive effect on recombination [55–57] . Such a pattern of a longer male map was also observed in sheep [58] , which is presumed to have been domesticated during the same contemporary period as dairy was domesticated and then underwent similar male-biased selective breeding [59] . In contrast to domestic sheep , the female recombination rate of the wild bighorn sheep was reported to be 12% greater than that of the male [60] . Based on 59 , 309 autosomal SNPs , we constructed cattle male and female recombination maps , 25 . 5 and 23 . 2 Morgans in length . A previous study that exclusively used bulls also reported a similar length of the male map [29] . Compared to the previously documented cattle linkage maps that were based on a small number of markers with limited sample sizes [24 , 27 , 28 , 61] , our sex-specific maps were shorter in length . Such discrepancy could be due to several factors . Errors in the physical map or in the genotypes can inflate the number of identified recombination crossovers and increase the length of the genetic map [62] . Previously documented linkage maps were based on a smaller number of RFLP or microsatellite markers , which could potentially bias the estimates . Our simulation studies further validated the power of identifying crossovers in this study ( 97 . 6% ) and the accuracy of our estimates of the length of recombination maps . Moreover , the previous studies with a smaller number of markers were probably less powerful , potentially contributing to this difference . We found a significantly decreasing trend in recombination rate in males from the analyses of recombination in the past forty years . Such decline in recombination rate in the past forty years coincided with the steady increase in milk production and decrease in fertility , a result of the intensive artificial selection in cattle breeding [47 , 63 , 64] . Although recombination generally increases selection efficiency by providing more combinations of genetic alleles [3] , recombination likely was selected against in cattle breeding that predominantly occurred in males . In cattle breeding , bulls tended to carry more desired chromosomes so that a male progeny that inherited the most chromosome segments from an elite sire would have better performance and more chance to be selected for breeding . In other words , the cattle breeding favored paternal haplotypes that were not or less mixed with the maternal haplotypes during meiosis . Therefore , a sex-biased cattle breeding and selection could potentially decrease the number of recombination in a short period and likely explain the reduction of recombination rate in cattle , particularly in males . To evaluate whether the decrease in recombination rate is correlated with systematic changes in allele frequencies of associated genetic variants , we calculated the frequencies of the alleles that increase recombination rate for associated SNPs over years but found no clear patterns ( S11 Fig ) . Inbreeding decreases the power of identifying crossovers through reducing the number of heterozygote SNPs per individual , so we adjusted for the effect of inbreeding by including the genomic inbreeding coefficient of the individual and the numbers of informative ( phased heterozygote ) SNPs in both the parent and offspring in a linear model . As expected , we found a negative association between inbreeding coefficient and number of recombination events in both sexes . Our GWAS analyses identified several loci influencing genome-wide recombination rate . Some of these loci had significant influence in both sexes ( PRDM9 , GCLM , CPLX1 , PABP1 , REC114 , FMN1 , and NEK9 ) , and some of them were potentially sex-specific ( MSH4 , CEP55 and SMC3 ) . We also confirmed the putative role of PRDM9 in the genome-wide recombination rate in both sexes . From GWAS of hotspot usage , we confirmed the cattle PRDM9 gene in both sexes to be the paralogue on chromosome 1 in our population , although the cattle genome encompasses multiple paralogues of PRDM9 and a previous study localized the associated PRDM9 to chromosome X [29] . To better understand the sex difference in recombination rate in the subtelomeric regions , we conducted additional GWAS of subtelomeric recombination rates , in which the phenotype was the number of crossovers that occurred in the last 15% of each chromosome . Compared with the GWAS of genome-wide recombination rate , the subtelomere GWAS identified a smaller number of associations that have already been found from the GWAS of genome-wide recombination rate , including the loci on chromosomes 1 , 6 , 10 , and 26 ( S4 Table ) . While many of these associations showed a larger effect in males than in females , the association at PRDM9 exhibited the same effect size in the two sexes , suggesting a possible unique role of PRDM9 in the subtelomeric recombination . Interestingly , the effect size of the PRDM9 association with genome-wide and subtelomeric recombination rates was the same in males , which might be related to the large number of male recombination hotspots in subtelomeric regions . Recombination rate is positively correlated with physical distances between SNPs . In this study , we used the original recombination rate between two SNPs without adjusting for physical distance to define recombination hotspots for several reasons . First , the SNPs on genotyping chips were about evenly distributed . Second , our hotspot definition was supported by the identification of association between hotspot usage and PRDM9 , consistent with results in human and mouse . We tested a range of cutoff values to define hotspots from 2 to 10 standard deviations and the association between PRDM9 and hotspot usage was consistently identified . Third , the original recombination rate without adjustment for physical distance is unaffected by inaccurate physical distances in the genome assembly . To evaluate if our results were biased by the physical distances between SNPs , we standardized recombination rate by physical distance . With this correction , we noticed several spurious hotspots that had a very small physical length but a moderate recombination rate , suggesting the existence of potentially inaccurate physical distances . To eliminate biased correction for physical distance due to potentially inaccurate physical distances , we filtered all SNP intervals shorter than 500 bp and calculated a standardized recombination rate between SNP pairs by dividing the original recombination rate by its physical distance . With this correction , the pattern of recombination rate along the chromosome ( S12 Fig ) was similar to that without this correction ( Fig 4 ) . The standardized recombination rates between SNP intervals were less variable so we used a cutoff of 0 . 6 standard deviations to be able to identify 2 , 875 and 3 , 005 male and female recombination hotspots , respectively . Using these hotspots with the correction for physical distance , the GWAS of hotspot usage identified the association at PRDM9 in females ( S13 Fig ) , but with larger P-values ( less significant ) than those from the original GWAS ( Fig 6 ) . In males , the association at PRDM9 was only nominally significant ( S13 Fig ) . Taken together , we recommend using the original recombination rate for hotspot definition without adjustment for physical distances , and the quality of our results is evidenced by the shorter recombination maps and the confirmation of several known recombination genes including PRDM9 and CPLX1 . In conclusion , our large-sample study reveals new insights into the cattle meiotic recombination and its genetic basis by offering male and female recombination maps , a sex difference in recombination rate that predominantly occurred in subtelomeres , and genomic loci associated with recombination rate and hotspot usage in the two sexes . Our study clearly delineates that the genomic resources accumulated during many years of genetic evaluation in cattle provide valuable opportunities for understanding cattle genetics including genome recombination .
To infer recombination events and compare sex differences , we extracted a total of 185 , 917 three-generation families from a large pedigree of Holstein cattle maintained in the Animal Genomics and Improvement Laboratory ( AGIL ) at USDA , with one offspring , both parents and two grandsires with SNP genotypes in each family ( Fig 1 ) . In each of the 185 , 917 families , we inferred recombination events for a paternal meiosis from the sire/offspring pair and a maternal meiosis from the dam/offspring pair . For recombination map construction and GWAS of hotspot usage , we only included the highest-quality or most informative meioses where the offspring , the parent , and the grandsire were genotyped by 50K SNP chips , resulting in a total of 70 , 715 male and 61 , 616 female meioses . For GWAS analysis of genome-wide recombination rate , we included all paternal and maternal meioses from the 185 , 917 families regardless of the number of SNPs genotyped . The animals in the selected families were genotyped by various genotyping assays ( S1 Table ) , ranging from 3K to 770K SNPs [33] . The Illumina BovineSNP50 v1 chip with 56 , 947 SNPs , v2 chip with 54 , 609 SNPs , the high-density ( HD ) chip with 777 , 962 SNPs , and the GeneSeek HD chip with 77 , 068 SNPs are referred to as the 50K chip , as we used a combined set of >50K SNPs [65] . The Zoetis BovineLD chip with 10 , 555 SNPs , the GeneSeek Genomic Profiler v1 and v2 chips with 8 , 042 and 8 , 415 SNPs , the Illumina BovineLD BeadChip with 6 , 785 SNPs , the Illumina Bovine3K BeadChip with 2 , 708 SNPs are referred to herein as 10K , 8K , 7K , and 3K , respectively . The chips were designed as mostly nested with the higher-density chip including SNPs on the lower-density ones . Although the offspring and two parents were genotyped by various SNP chips , the grandsires were mostly genotyped by the 50K SNP chips . Note that the granddam was not necessarily genotyped in the selected three-generation families . Depending on the number of genotyped granddams , we collected 67 , 690 , 76 , 318 , and 41 , 909 families with two , one and zero genotyped granddams , respectively . Note that an animal may appear in more than one family based on the pedigree structure , especially for bulls that have hundreds of progeny . To study recombination , we included up to 59 , 309 genome-wide SNPs after quality control filtering and used the USDA-AGIL SNP coordinates that showed a higher quality than the UMD3 . 1 assembly [30 , 33] . Previously , several SNPs were relocated from the UMD3 . 1 assembly to the USDA-AIGL coordinates in cooperation with researchers from the University of Missouri ( R . D . Schnabel ) , and the University of Guelph ( M . Sargolzaei and J . Johnston ) . From pairwise linkage analysis in this study , we also removed nine suspicious SNPs that exhibited suspicious linkage disequilibrium ( LD ) patterns with SNPs on the same chromosome ( S2 Fig and S3 Table ) . Due to the low quality of the genome assembly , we excluded the X chromosome from recombination calculation in this study . To compare sex differences on an equal footing , we phased the raw genotypes for the paternal and maternal meioses within a three-generation family without using the possible additional information from multiple offspring [66] , because bulls generally had many more progeny than cows . In each of the three-generation families , we inferred the paternal and maternal haplotypes of the offspring based on the genotypes of the two parents , and also inferred the paternal and maternal haplotypes for both parents based on the genotypes of the grandparents [29 , 66] . Homozygous genotypes were phased trivially and the heterozygous genotypes were phased whenever the genotypes of the two parents are not heterozygous simultaneously . The parent-of-origin was then assigned to each allele after phasing to determine paternal and maternal haplotypes . After phasing the genotypes of the offspring and parents , we inferred recombination events in the paternal and maternal haplotypes of the offspring by comparing the offspring’s paternal haplotype to the two haplotypes of the sire as well as by comparing the offspring’s maternal haplotype to the two haplotypes of the dam . In the offspring haplotype , a recombination event was defined as a transition from the parent's paternal to maternal haplotype or vice versa . Note that the recombination events defined here were observed crossovers so that the number of observed crossovers could differ from the number of true crossovers when multiple crossover events occurred at the same site . This potential inconsistency between observed and true crossovers typically is addressed by a map function that translates a recombination frequency into a map distance in terms of crossovers . However , the physical distances between two adjacent SNPs were small so that the use of a map function virtually would not make a numerical difference . Therefore , our estimates of crossovers based on recombinants should be close to the true number of underlying crossovers . To further reduce false positives , we required a crossover call to be supported by at least three consecutive informative heterozygous SNPs [9] . In total , we identified ~4 . 5 million paternal and ~4 . 0 million maternal recombination events from the total of 185 , 917 paternal and maternal meioses , respectively . To ensure that our results do not depend strongly on the cutoffs , we repeated the analysis by using a different cutoff value of 5 consecutive informative markers and the number of identified crossovers was only reduced by 0 . 2% . A recombination event was assigned to a region spanned by two informative SNPs that may not be adjacent to each other . To construct a recombination map , we used an EM-algorithm to calculate the probability of crossing over per meiosis or recombination rate between each pair of consecutive SNPs based on the observed crossover regions [9] . After an initiation step to assign an expected count of 1/m to each of the m adjacent SNP intervals in a crossover region , the EM algorithm proceeded in the following iteration steps: 1 ) M-step: considering a total of n meioses , the overall expected count attributed to a SNP interval divided by n were the maximum likelihood estimate of the probability , and 2 ) E-step: for a crossover region , the expected count assigned to a SNP interval was estimated as proportional to the current estimate of the probability of crossover for that SNP interval . The M and E steps were iterated until convergence . We constructed the male and female recombination maps for 59 , 309 autosomal SNPs based on >1 . 8 million paternal and >1 . 4 million maternal recombination crossovers , which were identified from 70 , 715 male and 61 , 616 female meioses that are most informative where the offspring , the parent , and the grandsire were genotyped by 50K SNP chips . Note that some granddams were genotyped by a low density chip or even not genotyped . As described earlier , a correction for the number of SNPs of the granddam was employed in the two sexes separately such that the total number of crossovers after correction was the same regardless of the SNP numbers for granddams . We used the expected number of crossovers per meiosis or recombination rate between adjacent SNPs as the genetic distance in our recombination map , since one crossover event on average corresponds to a genetic distance of 1 Morgan and recombination rate is almost the same as genetic distance for small intervals [9 , 67] . Alternatively , using Haldane’s map function with crossover interference [68] , the male and female maps were slightly longer , 25 . 6 and 23 . 3 Morgans in length respectively . We defined recombination hotspots as the SNP intervals with a recombination rate >2 . 5 standard deviations from the genome-wide average in males and females separately , because 2 . 5 standard deviations are highly significant departures from the average recombination in cattle given our large sample sizes in both sexes , consistent with the observations from a recent recombination study in cattle [29] . We tested a range of cutoff values , 2 , 2 . 5 , 3 , 5 , and 10 standard deviations , and found that the cutoff value of 2 . 5 standard deviations showed the clearest signal for the association between PRDM9 and recombination hotspot usage . For validation purposes , we also defined recombination hotspots via using a standardized recombination rate that was calculated by dividing the original recombination rate between two SNPs by the physical length . After this adjustment of physical lengths , the standard recombination rates vary even less so we used a cutoff value of 0 . 6 standard deviations to define recombination hotspots . To ensure the quality of the hotspots identified , we calculated pairwise linkage disequilibrium ( LD ) statistics between each of the SNPs in or near a recombination hotspot and all other SNPs on the same chromosome using Locusmap [35] . The pairwise LD was evaluated by a LOD score and an estimated recombination rate . For a SNP with correct physical position , the LOD scores should peak near the SNP and decrease when moving away in the two directions . The recombination rate should follow the opposite pattern with small values near the SNP and increasing with the distance . Any obvious deviations from these expected LD patterns suggest a possible error in the SNP coordinate and thus a false recombination hotspot ( S2 Fig ) . Nine suspicious SNPs exhibiting unexpected LD patterns together with seven originally identified recombination hotspots were removed ( S5 Table ) . As expected , the power for identifying recombination events was affected by the number of genotyped SNPs . For male meiosis , the average number of crossovers identified varied from 25 . 9 to 17 . 7 depending on the number of measured SNPs of the animals in a three-generation family ( S2 Table ) . Overall , more crossovers were identified when the number of SNPs increased , except for a few categories that had a small sample size and thus large noise . A similar pattern was also observed for female meioses , where the number of crossovers ranged between 23 . 9 and 16 . 0 ( S3 Table To account for this effect of SNP numbers , we used the highest-quality meioses where all animals were genotyped by 50K SNP chip as a reference to correct the number of crossovers identified in other meioses . After correction , the average number of crossovers was equal to 25 . 5 in males and 23 . 2 in females . To evaluate the power for identifying crossovers using 50K SNP genotypes , we simulated 50K SNP genotypes for all animals in the real pedigree and used the same phasing and crossover identification procedures as described in previous sections . By defining a positive result as an identified crossover interval overlapping with the true location of a crossover , we calculated the power of crossover identification as the proportion of positive results across all simulated crossover events . As a result , our approach had a power of 97 . 6% for the three-generation families genotyped by 50K SNP chip , which means that on average only 2 . 4% crossovers were missed in our recombination map . To evaluate the relationship between recombination features and potentially related factors , we fitted a smooth spline model of the male and female recombination rates on relative chromosomal locations or time using the smooth . spline function implemented in R 3 . 1 . 1 [69] . We calculated a relative physical position for each of the SNP intervals by dividing the original physical position by the corresponding chromosome length . For the analysis of chromosome location effect on recombination , we used those meioses for which all the required individuals were genotyped by 50K SNP chip and removed 2-Mb regions to the end of all chromosomes where the power of identifying crossovers is low [9] . For the analysis of time trend , we used all the meioses and adjusted for the effects of SNP chips and inbreeding , as well as effect of influential bulls by accounting for the number of progeny for bulls . Specifically , the correction was conducted with a linear model for the number of crossovers by fitting fixed effects for the categories of SNP chips of the offspring , parent and two grandparents , genomic inbreeding coefficient of the parent , the number of phased heterozygous SNPs of the offspring and parent , quadratic and cubic terms of the two numbers of informative SNPs , and the interaction terms between them . As expected , we found a negative association between inbreeding coefficient and number of recombination events in both sexes ( Males: β = −0 . 28 and P-value = 0 . 006; Females: β = −0 . 11 and P-value = 3 . 2 × 10−12 ) . To investigate the common pattern across the 29 autosomes , we pooled all the autosomes together using the relative physical position and fitted a smooth spline model for all the data combined ( Fig 4 ) . We also fitted a smooth spline model for each of the chromosomes individually ( S3 Fig ) . A degree of freedom of five was used in all the smooth spline modeling . A similar smooth spline model was fitted for the analysis of time trend of recombination rate and hotspot usage ( S8 , S9 , S10 and S12 Figs ) . We estimated the number of recombination events for maternal and paternal meioses in each of the 185 , 917 three-generation families , which were then assigned to the sire and dam in the family . Each sire or dam may have multiple phenotypic measurements when appearing in more than one family , and we calculated the average of the multiple measurements as the phenotype for genome-wide recombination rate . A total of 3 , 224 bulls and 53 , 125 cows were included in the GWAS of genome-wide recombination rate . We corrected the originally estimated number of recombination events by the number of measured SNPs of the animals for each meiosis so that the average number of crossovers was the same regardless of the genotyping assays ( S2 and S3 Tables ) . We calculated the genome-wide proportions of crossovers occurring in the hotspots , i . e . hotspot usage , for the 70 , 715 male and 61 , 616 female meioses that are most informative where the offspring , the parent and the grandsire were genotyped by 50K SNP chips . After assigning the estimated hotspot usage to the sire or dam for each meiosis , we used the average of the multiple measurements as the phenotype , resulting in a sample size of 1 , 772 and 12 , 756 in males and females , respectively . Imputed genotypes of 777 , 962 SNPs on the Illumina BovineHD Genotyping BeadChip ( HD ) were obtained by running Findhap on measured genotypes with 3K to 50K SNP chips using a reference population of 2 , 433 animals directly measured with HD SNP chips [33 , 70] . After filtering SNPs exhibiting redundancy , very high linkage disequilibrium ( r > 0 . 95 ) , or small minor allele frequency ( MAF < 0 . 001 ) , we retained over 310K genome-wide SNPs in the association studies [33] . We tested for association between each SNP and a phenotype using a linear mixed model with variable residual variances that are inversely proportional to the number of repeated measures of the phenotype , i . e . , residual variance is smaller for individuals with more measurements of the phenotype . The model equation in matrix notation is y=Xg+Za+e where y = a vector of the phenotype , X = a design matrix of the fixed effects g , including a population mean and the additive effect of the candidate SNP , Z = a design matrix for a random animal effect a , and e = a vector of random residuals . We assume that a ~ N ( 0 , Aσa2 ) and e ~ N ( 0 , Rσe2 ) , where A is the genomic relationship matrix and R is a diagonal matrix with the ith diagonal element equal to 1/w , where w is the number of phenotypic measurements for the ith animal . This model has been implemented in the MMAP software package with optimized computing [71 , 72] , which can finish a GWAS analysis with 53 , 125 samples and 310K SNPs in hours using 32 CPU cores of the high-performance computer at USDA-AGIL . The model was empirically validated by observing no inflation in the quantile-quantile plots ( QQ-plot ) of the GWAS P-values for both recombination rate and hotspot usage in this study ( S5 and S6 Figs ) . | Previous studies on cattle recombination largely focused on males . Using a large Holstein sample from the USDA national database , we studied both male and female recombination by assembling paternal and maternal recombination events in at least three generations . This unique data set provides unprecedented statistical power to study cattle genome recombination in the two sexes: ( 1 ) We report for the first time that bulls have more recombination than cows , contrary to the common perception that females have more recombination than males as observed in many mammalian species including humans and mice , and that the sex difference in recombination primarily occurs near the subtelomeric regions of all bovine autosomes; ( 2 ) We identify several genes associated with cattle recombination in both females and males , and genes affecting female recombination only; ( 3 ) We define putative recombination hotspots and find the cattle PRDM9 gene to be associated with recombination hotspot usage . These results provide new insights for understanding cattle and mammalian genome recombination . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Cattle Sex-Specific Recombination and Genetic Control from a Large Pedigree Analysis |
Arenaviruses cause severe diseases in humans but establish asymptomatic , lifelong infections in rodent reservoirs . Persistently-infected rodents harbor high levels of defective interfering ( DI ) particles , which are thought to be important for establishing persistence and mitigating virus-induced cytopathic effect . Little is known about what drives the production of DI particles . We show that neither the PPXY late domain encoded within the lymphocytic choriomeningitis virus ( LCMV ) matrix protein nor a functional endosomal sorting complex transport ( ESCRT ) pathway is absolutely required for the generation of standard infectious virus particles . In contrast , DI particle release critically requires the PPXY late domain and is ESCRT-dependent . Additionally , the terminal tyrosine in the PPXY motif is reversibly phosphorylated and our findings indicate that this posttranslational modification may regulate DI particle formation . Thus we have uncovered a new role for the PPXY late domain and a possible mechanism for its regulation .
Arenaviruses are a family of rodent-borne viruses with a worldwide distribution . These viruses typically establish persistent , asymptomatic infections in rodent reservoir species [1] . In contrast , arenaviruses cause severe and often fatal diseases in humans . Several arenaviruses , including Lassa virus and Junín virus , cause hemorrhagic fever syndromes whereas infection with the prototypic arenavirus , lymphocytic choriomeningitis virus ( LCMV ) , can lead to aseptic meningitis in immunocompetent individuals , high lethality in immunocompromised individuals , or severe birth defects in the developing fetus [2 , 3] . U . S . Food and Drug Administration-approved vaccines do not exist for the prevention of arenavirus infection and effective antiviral therapies have been limited to the use of ribavirin for Lassa virus [4] or immune plasma for Junín virus [5] . Arenaviruses are enveloped viruses with a single-stranded , bi-segmented RNA genome that encodes four proteins in an ambisense manner . The small ( S ) segment encodes the nucleoprotein ( NP ) and glycoprotein ( GP ) while the large ( L ) segment encodes the RNA-dependent RNA polymerase ( L ) and the matrix protein ( Z ) [6] . Arenaviruses enter cells via receptor-mediated endocytosis [7] , undergo genomic replication and transcription in the cytoplasm [6] , and assemble and bud new particles at the plasma membrane [8] . The Z protein , which lines the luminal side of the viral membrane , is responsible for a number of critical functions in the virus life cycle , including driving the process of viral particle assembly and budding [9] . Accordingly , Z can form virus-like particles ( VLPs ) in the absence of other viral proteins and is thought to be both necessary and sufficient for driving the budding process [10 , 11] . Several VLP-based studies indicate that Z drives viral particle release by virtue of one or more encoded viral late domain ( s ) ( P ( S/T ) AP , YXXL , and/or PPXY ) , which can recruit proteins from the cellular endosomal sorting complex required for transport ( ESCRT ) pathway [10–12] . ESCRT machinery is required for most cellular membrane scission events that result in separation away from the cytosol including multivesicular body formation and cellular abscission [13–15] . Many enveloped viruses are known to hijack cellular ESCRT machinery via their late domains to complete the final membrane scission step required for virions to bud from host membranes ( for review see [16] ) . Viruses from diverse families , including arenaviruses , produce defective interfering ( DI ) particles in addition to standard , infectious virus during the normal course of infection [17] . DI particles are largely similar to standard virus particles in their appearance and viral protein content but cannot self-replicate , and interfere with the production of homologous standard virus [17] . In many cases , the primary difference between DI particles and standard virus is thought to be the presence of deletions in the viral genome [18] . With regard to LCMV , small deletions in the terminal untranslated regions of genomic and antigenomic RNAs have been observed , but it is not known whether these RNAs have interfering properties or are selectively incorporated in DI particles [19] . The interfering activity of arenavirus DI particles can be blocked by neutralizing antibodies but is maintained even after treatment with ultra-violet ( UV ) radiation , unlike standard particles , which are highly susceptible to both treatments [20] . Arenaviruses generate high levels of DI particles both in cell culture [21] and in host rodents [22] . It has long been postulated that arenavirus DIs are an important factor in the establishment of persistent infection [17 , 23 , 24] but a causal link between arenavirus DI particles and persistence has yet to be firmly established . There are several outstanding questions regarding the arenavirus matrix protein , including how its functionality is regulated and how , in the context of a fully replicating virus , encoded late domains contribute to the production of standard and DI particles . Herein we demonstrate that LCMV’s sole late domain , PPXY , is not required for standard virus budding but instead is the driving force of DI particle release . Further , standard virus appears to bud independently of ESCRT machinery while DI particle release is ESCRT-dependent . Finally , we show that the LCMV PPXY motif is tyrosine phosphorylated and that this post-translational modification appears to regulate DI particle formation .
The matrix protein plays a multifactorial role in the arenavirus life cycle yet little is known regarding how its various functions are regulated . Given the importance of phosphorylation for regulating the functionality of matrix proteins of other virus families [25–29] , we were interested in whether LCMV’s matrix protein might also be phosphorylated . LCMV strain Armstrong 53b particles grown in Vero E6 cells were purified via sucrose-banding ( Fig 1A ) and subjected to mass spectrometry . This analysis revealed a tyrosine phosphorylation site near the C-terminus of the LCMV Z protein at position 88 ( Y88 ) ( Fig 1B and S1 Fig ) , which lies within LCMV Z’s PPPY late domain ( Fig 1C ) . Both phosphorylated and unphosphorylated peptides containing this residue were observed at a ratio of 1 to 11 , respectively , which suggests that ~10% of the total Z protein in this virion preparation is phosphorylated ( Fig 1B and S1C Fig ) . Because the virion preparation contained a mixture of both standard infectious virus and DI particles , we were not able to determine whether the phosphorylated Z was derived from standard particles , DI particles , and/or both types of particles . To confirm the phosphorylation site , plasmids encoding either WT Z or a phenylalanine mutant ( Y88F ) that cannot be phosphorylated were transfected into HEK293T cells and 2 days later the cells were treated with either water or the tyrosine phosphatase inhibitor , hydrogen peroxide . WT Z and Y88F Z were affinity purified and probed with a phosphotyrosine-specific antibody . The phosphotyrosine signal detected from WT Z was greatly enhanced following inhibition of tyrosine phosphatases ( Fig 1D ) . Substitution of tyrosine 88 with phenylalanine , to prevent phosphorylation , resulted in a complete loss of detectable phosphotyrosine signal in both settings indicating that Y88 may be the only tyrosine of the 3 encoded in LCMV Z that is phosphorylated ( Fig 1D ) by endogenous kinases in these cells . To determine whether LCMV Z is tyrosine phosphorylated in the context of a relevant rodent cell line , we infected murine L929 cells with a rLCMV that encodes Z with a C-terminal streptavidin binding peptide ( SBP ) tag . Two days later , cells were either treated with hydrogen peroxide or not and Z was affinity purified from cell lysates for western blot analysis . As shown in Fig 1E , a phosphotyrosine signal was clearly detectable from Z and was enhanced following treatment with hydrogen peroxide . The finding that LCMV Z is phosphorylated at Y88 was intriguing as this residue is part of LCMV’s only late domain , PPPY . This motif is well conserved among most Old World arenavirus Z proteins ( Fig 2A ) and its importance for the budding activity of LCMV and Lassa virus Z in the context of VLP-budding assays has been well described [10 , 11] . To investigate the role of this late domain in the context of authentic virus and to determine whether tyrosine phosphorylation may regulate its function , we generated recombinant ( r ) LCMV encoding either phenylalanine or alanine at position 88 to prevent phosphorylation at this site or glutamic acid to mimic constitutive phosphorylation . The alanine mutant was included as a reference to previous studies on the function of this late domain for LCMV and Lassa virus Z , which used alanine substitutions at Y88 to assess the contribution of this late domain to Z’s budding efficiency in VLP assays [10 , 11] . Viruses containing all three mutations were recoverable despite the well-described defect in Z’s budding efficiency caused by mutation of this residue ( Fig 2B ) [10 , 11] . The growth kinetics of rLCMV Z-Y88F and Z-Y88A during the first 36 hours ( hr ) post-infection ( pi ) were nearly identical , but impaired ~15-fold compared to rLCMV WT ( P ≤ 0 . 0001; Fig 2B and 2C ) . The growth rate of the rLCMV Z-Y88E phosphomimetic was also attenuated compared to WT virus over this same time frame ( ~4-fold less PFU at 36 hr pi , P ≤ 0 . 05 , Fig 2B and 2C ) . However , the phosphomimetic virus grew to ~4–fold higher titers than the alanine or phenylalanine mutants ( P ≤ 0 . 01; Fig 2B and 2C ) . Additionally , the mean plaque size for rLCMV Z-Y88E was significantly increased compared to the Z-Y88F and Z-Y88A viruses ( 0 . 67 vs 0 . 53 or 0 . 52 mm2; P ≤ 0 . 01; Fig 2D ) , indicating that virus spread was partially restored in the phosphomimetic virus . Notably , each mutant virus eventually reached peak WT titers . Given the delayed kinetics observed in the mutant viruses , we tested for reversion mutations at 72 hr pi and confirmed that each virus retained its respective mutated residue at position 88 and its small plaque phenotype ( Fig 2D ) . Collectively , these results demonstrate that the PPXY late domain is not absolutely required for the formation and release of standard infectious particles . Further , phosphorylation of Y88 may have a positive regulatory impact on viral propagation . Point mutations made at Y88 suggested that dynamic phosphorylation of this residue was important for the function of the matrix protein . Given the important role of the LCMV matrix protein and its late domain motif in regulating viral budding [10 , 11] , we next investigated the specific effect these point mutations had on Z’s budding efficiency in a VLP release assay . Because the LCMV Z protein is sufficient for the production of VLPs in the absence of any other viral proteins [10 , 11] , we were able to assess the budding activity of plasmid-derived WT or Y88-mutant Z proteins . As a control , we also included the LCMV Z G2A mutant , which exhibits a pronounced defect in VLP formation due to its inability to be myristoylated at this glycine residue [30] . HEK293T cells were transfected with plasmids encoding WT or Y88 mutants and 1 day later the VLP-containing supernatant and cells were collected and analyzed by quantitative western blotting . The budding activity of all three Z Y88 mutants was significantly reduced compared to WT Z , indicating that mutations in this region reduce the efficiency of VLP release ( Fig 2E ) . In particular , the impaired VLP release exhibited by the Z Y88A mutant confirmed earlier findings by Perez et al . [11] . We did not observe a significant difference between the budding of the Z-Y88E phosphomimetic compared to Y88F and Y88A ( Fig 2E ) . This suggests that the partial gain of fitness observed with the phosphomimetic rLCMV-Z-Y88E virus in Fig 2B is not due to an increase in budding activity and as such Y88 phosphorylation does not appear to directly regulate the budding function of this late domain . However , because the VLP budding assay measures only the release of matrix protein , in the absence of other viral proteins , it is possible that this assay does not recapitulate all the facets of infectious virion production . To investigate the protein and genome composition of virions containing mutated late domains , an equivalent quantity of cell-free infectious virus particles from each rLCMV strain was concentrated for screening . Quantitative western blotting revealed substantial reductions in the total amount of NP , GP , and Z in the Y88 mutant particles relative to WT virus ( Fig 3A–3D ) . However , no difference was observed in the levels of these proteins among the three Y88 mutant viruses ( Fig 3A–3D ) . The quantity of Z protein detected in the Y88 mutant virus preparations was <3% of WT virus ( Fig 3D ) whereas NP and GP quantities were ~25% of WT virus ( Fig 3B and 3C ) . Viral genome content in particles was assessed by qRT-PCR . On a per PFU basis , the quantity of either L or S segment genomic RNA in the non-phosphorylatable mutants , Y88F and Y88A , was significantly reduced versus WT ( P ≤ 0 . 05 , Fig 3E and 3F ) . However , genome levels in the phosphomimetic virus , Y88E , were not significantly different than WT ( Fig 3E and 3F ) , which may explain a component of its partially restored growth kinetics ( Fig 2B ) . The observation of reduced viral proteins and/or genomes released from cells infected with the Y88 mutant viruses combined with the fact that WT LCMV is known to produce relatively large quantities of DI particles [21] led us to hypothesize that the PPXY mutants may have defects in their ability to generate DI particles , which could explain their greatly reduced levels of viral protein and genome relative to PFU . A substantial fraction of virus particles produced by LCMV are DI particles [31] . Accordingly , inoculation of LCMV at low multiplicities of infection ( MOI ) results in efficient production of standard virus and spread , while high MOIs do not . This seemingly contradictory phenomenon is caused by DI particles , which inhibit the propagation of standard virus and its ability to cause cytopathic effect with one hit kinetics [21 , 32] . Monolayers inoculated with high concentrations of standard infectious LCMV exhibit no cytopathic effect due to DI particle inhibition , but as the inoculum is diluted , standard virus particles that infect cells in the absence of a co-infecting DI particle will subsequently form plaques . We exploited this phenomenon to initially evaluate the relative amounts of DI particles generated by the PPXY mutant viruses . Equal infectious doses of WT virus and each Y88 mutant , spanning a range of 25 to 25 , 000 PFU , were applied to Vero E6 cell monolayers in a standard plaque assay ( Fig 4A and 4B ) . Evidence of possible DI particle interference is clearly seen in WT virus , where the most concentrated viral sample ( 25 , 000 PFU ) resulted in no cell death while in successive 10-fold dilutions ( 2 , 500 and 250 PFU ) the number of DI particles per cell is lowered allowing standard virus to enter cells in the absence of DI particles and form plaques ( Fig 4A ) . In contrast , the PPXY-mutant viruses exhibited a considerable increase in cytopathology ( Fig 4A , 25 , 000 and 2 , 500 PFU ) . Quantification of the observed cytopathology confirmed the striking phenotype and revealed significant differences between the mutant and WT viruses ( Fig 4B ) . Intriguingly , the cytopathology of the rLCMV Z-Y88E phosphomimetic at 25 , 000 PFU was significantly less than both Y88F or Y88A viruses and therefore more closely resembled WT virus ( Fig 4B ) . To confirm that the interfering activity observed in Fig 4 was indeed due to LCMV DI particles , we next established an assay to directly and quantitatively measure LCMV DI particle activity . At present , no consistent biochemical or genetic signature exists to distinguish LCMV DI particles from standard infectious particles [33 , 34] . In an attempt to uncover such a signature , we separated preparations of rLCMV WT or Y88 mutants via density ultracentrifugation . Similar to previous studies [33–37] , we were unable to isolate fractions containing pure DI particles as abundant levels of standard virus were detectable across all 15 fractions ( S2 Fig ) . Therefore , it was not possible to identify a DI particle-specific signature for screening purposes . Despite this limitation , several assays , including a yield reduction assay [21] , a plaque reduction assay [38] , and a focus interfering assay [32] have historically been used for accurate measurement of LCMV DI particle abundance and activity levels . Indeed , these assays were originally used to define LCMV DI particles . We utilized the plaque interference assay ( also known as the plaque reduction assay ) analogous to that used in [21] but also capitalized on the strong UV-resistance exhibited by LCMV DI particles , but not standard virus particles [38] . Briefly , cell-free virus preparations containing both standard and DI particles were treated with UV to neutralize standard virus particles while leaving the interfering properties of DI particles intact ( Fig 5A ) . It should be noted that standard virus particles treated with UV do not acquire detectable interfering properties ( Fig 5A ) [39] . Limiting dilutions of this UV-treated sample were applied to Vero E6 cells , followed by the addition of a fixed quantity of LCMV PFUs . As shown in Fig 5A , this allows for the determination of LCMV DI particle activity and is expressed as plaque interfering units50 ( PIU50 ) per mL of a given sample . Importantly , we recapitulated several key controls from previous studies to demonstrate the specificity of this assay for LCMV DI particles . In particular , UV-treated LCMV DI particle preparations only interfered , in a dose-dependent manner , with the growth of homologous LCMV , but not heterologous viruses such as vesicular stomatitis virus ( VSV ) or the New World arenavirus Junín virus Candid 1 ( JUNV C#1 ) , which rules out a nonspecific antiviral factor as a mediator of interference ( e . g . interferon ) ( Fig 5A ) . Further , passing LCMV DI particle-containing supernatant through a series of filters ( 0 . 45 μm , 0 . 2 μm , 30 kDa , 10 kDa ) showed that interference is not due to soluble factors that are smaller than 30 kDa ( e . g . cytokines ) or larger ( >0 . 2 μm ) membrane bound entities such as bacteria ( Fig 5B ) . When this assay was applied to the rLCMV WT and Y88 mutant samples examined in Fig 4 , it confirmed that the rLCMV WT samples exhibited substantial DI particle interfering activity ( mean 926 PIU50/mL ± 68 SEM ) , but that the mutant Y88 viruses had much less ( Fig 5C ) . There was no detectable DI activity for either the Y88F or Y88A viruses while the Y88E virus contained intermediate levels of interfering activity ( mean 131 PIU50/mL ± 64 SEM ) . Collectively , the findings in Figs 4 and 5 support the hypothesis that the LCMV PPXY late domain is required for the efficient formation of DI particles and that phosphorylation of Y88 may play a regulatory role in DI particle production and the inhibition of cytopathic effect . Viral late domains can drive virus budding by recruiting components of the cellular ESCRT pathway to complete the final membrane scission step . Given the important role that the LCMV PPXY late domain played in the production of DI particles ( Figs 4A , 4B and 5C ) , we hypothesized that this late domain might be recruiting the ESCRT pathway machinery to drive DI particle formation . To test this hypothesis , we utilized cell lines that lack a functional ESCRT pathway due to inducible expression of a dominant negative ( DN ) , E235Q point mutant , of VPS4 , an ATPase involved in the final stages of ESCRT pathway function [40–42] . Because the ESCRT pathway can also affect LCMV entry [43] , we first infected cells with LCMV for 48 hr to allow the entire monolayer to become infected before inducing expression of WT or DN VPS4 . The cells were washed and fresh media containing the induction agent was added to the cells 6 hr after initial induction ( 54 hr pi ) and the virus-containing media was collected 18 hr later ( 72 hr pi ) to determine levels of standard infectious particles and DI particles ( Fig 6 ) . Western blot analysis of protein lysates at 72 hr pi confirmed the strong induction of WT and DN VPS4B expression and examination of fixed coverslips showed that all cells were expressing both the induced VSP4B as well as LCMV NP ( Fig 6B ) . This infection protocol was chosen to ensure that we were examining virus that was produced in cells expressing the induced VPS4B proteins , while minimizing the effect that these proteins could exert on viral entry . Expression of DN VPS4B had no impact on the release of standard infectious LCMV ( P = 0 . 27; Fig 6B ) . In contrast , expression of DN VPS4B led to a marked reduction in the release of infectious VSV particles ( Fig 6A ) , which is consistent with previous studies [42] and confirms the specificity of our findings for LCMV . Measuring LCMV DI particle activity as described in Fig 4 revealed that WT LCMV produced considerably fewer DI particles per standard infectious virus particle in the DN VPS4B background when compared to cells expressing WT VPS4B ( Fig 6C and 6D ) . A similar trend for both LCMV infectious virus and DI particle activity was seen in cells expressing WT or DN VPS4A ( S3 Fig ) . We next used the assay described in Fig 5 to directly quantitate the LCMV DI particle activity in these samples . Consistent with the findings in Fig 6C and 6D , this demonstrated that significantly fewer DI particles are made in the context of the DN VPS4B background when compared to WT VPS4B ( mean 41 ± 6 SEM vs 1 , 491 ± 70 PIU50/mL , respectively; P = 0 . 0022 ) . Thus it appears that LCMV DI particle formation requires a functional ESCRT pathway ( Fig 6C–6E ) in addition to a canonical late domain ( Figs 4 and 5C ) while standard particles do not ( Fig 6B ) .
The ability of most arenavirus matrix proteins to drive viral budding is thought to be highly dependent upon one or more encoded late domains [10 , 11] . The arenavirus LCMV encodes a single late domain , PPPY . The PPXY motif is found in the matrix proteins of several families of enveloped RNA viruses and for many of these viruses is required for the release of infectious virions in an ESCRT-dependent fashion ( for review see [16] ) . We demonstrate here that the PPXY late domain encoded by LCMV is not absolutely required for infectious virus release . Further , our data suggest that infectious particle release can occur in the absence of a functional ESCRT pathway . Strikingly , we show that the formation of LCMV DI particles critically requires a functional PPXY late domain and that this process is ESCRT-dependent ( see Fig 7 for our proposed model ) . Last , our data demonstrate that the terminal tyrosine in the LCMV PPXY motif is phosphorylated and that this posttranslational modification may exert a regulatory effect on Z’s ability to drive DI particle release . Therefore , we have uncovered an unexpected role for the PPXY late domain and a possible mechanism for its regulation of DI particle production . Our findings raise the intriguing possibility that LCMV utilizes divergent pathways for the production of infectious and DI particles . Neither a functional PPXY motif nor ESCRT pathway were absolutely required for the release of standard infectious particles . These findings , combined with the fact that LCMV Z does not encode additional canonical late domains , strongly suggest that infectious LCMV release occurs through a novel , unknown mechanism . While the rLCMV PPXY mutants studied here initially displayed a slight lag in infectious virus release , each ultimately matched WT levels . Consistent with an earlier study by Perez et al . [11] , we observed that mutation of the PPXY domain impairs the ability of LCMV Z to form VLPs ( Fig 2E ) . This finding in the VLP system may accurately reflect the initial lag in infectious virus release seen for rLCMVs bearing the same PPXY mutations ( Fig 2B and 2C ) and/or their decreased ability to form DI particles ( Figs 4 and 5 ) . With regard to the importance of the ESCRT pathway for infectious virus production , expression of DN VPS4B had no impact on the ability of WT rLCMV to form standard infectious virus ( Fig 6B ) . Similarly , expression of DN VPS4B did not impair the ability of LCMV Z to form VLPs when compared to cells expressing WT VPS4B ( see S5 Fig ) . This VLP-based result does not agree with an earlier finding by Perez et al . whereby silencing expression of the ESCRT component TSG101 impaired the release of infectious LCMV VLPs [11] . This discrepancy may reflect differences in the particular VLP assays employed ( VLP release versus infectious VLP release and transduction ) or perhaps the format of the experiments ( siRNA silencing of TSG101 versus inducible expression of WT or DN VPS4B ) . The different VLP systems could also recapitulate different aspects of virus particle release , with one VLP assay perhaps mimicking infectious virus production while the other more closely resembles DI formation . In similar experiments featuring rhabdoviruses [44 , 45] , Ebola virus [46] , retroviruses [47–50] , and Hepatitis B virus [51] , loss of the PPXY motif and/or ESCRT resulted in standard virus growth that remained attenuated compared to WT . Interestingly , the New World arenavirus Pichinde remains attenuated following disruption of its encoded late domain ( PSAP ) [52] , while the matrix protein encoded by the New World arenavirus Tacaribe can form VLPs in the absence of its late domain ( YXXL ) , but requires VPS4 [53] . These findings suggest that arenaviruses have evolved diverse strategies to drive infectious virus release . One possible explanation for the observed late domain-independent generation of infectious LCMV particles is that the LCMV Z protein , either by itself or in combination with other viral structural proteins , may be sufficient to drive particle release in the absence of recruited host proteins . Alternatively , LCMV may contain additional sequence motifs , either in Z or the other structural proteins , that recruit novel host protein machinery to facilitate budding . In contrast to standard virus particles , both the PPXY late domain and the ESCRT pathway appear critical for the release of DI particles . To our knowledge , this is the first example of a virus utilizing a late domain to selectively drive the production of DI particles independently of standard virus . That LCMV has evolved such a mechanism likely reflects the presumed importance of DI particles for the successful establishment of an asymptomatic , persistent infection in reservoir rodents , which ultimately ensures the long term maintenance of LCMV in nature [19 , 22] . While the existence of arenavirus DI particles has long been realized , surprisingly little is known about their exact composition and properties . Our data demonstrates that cells infected with the rLCMV PPXY mutants release much less NP , GP , and Z per PFU of cell-free virus when compared to WT rLCMV . This is presumably due to reduced levels of DI particles being released by the PPXY mutant viruses . Interestingly , the degree of reduction was not equivalent among the viral proteins . In particular , Z was reduced to the greatest extent ( ~3% of WT ) when compared to NP or GP ( ~25% of WT ) , which could indicate that Z itself is enriched in DI particles and is critical for the ability of DI particles to interfere with the propagation of standard virus particles . Consistent with this idea , Z is able to render cells refractory to superinfection with homologous virus and treatment of DI particles with RNA-damaging levels of UV does not reduce their interfering ability [38 , 54] . Therefore , it is possible that particles containing high quantities of Z , or simply VLPs consisting primarily of Z , may represent a class of arenavirus DI particles . Arenavirus matrix proteins exhibit significant diversity in the type and number of late domains they encode . The PPXY domain is found in several Old World arenavirus matrix proteins but not in New World arenaviruses ( Fig 2A and [9 , 55] ) . We have observed that the New World arenavirus JUNV C#1 , which encodes both a PTAP and YXXL late domain , generates considerably fewer DI particles per standard infectious particle when compared to the PPXY-containing LCMV ( S4 Fig ) . While it is not known whether the PTAP and/or YXXL motifs contribute to DI particle formation in the case of JUNV C#1 , this observation may indicate that the PPXY domain is particularly strong in driving DI particle assembly and release . It is possible that individual arenaviruses require different rates of DI particle formation for optimal fitness and have evolved to encode particular late domain combinations to best meet those needs . We show that the LCMV Z protein is phosphorylated , which suggests that phosphorylation may be important for the regulation of one or more of Z’s functions . The fact that this modification occurs at the terminal tyrosine of the PPXY late domain and can be detected in virion-derived Z led us to hypothesize that it may influence Z’s budding function . To study the impact of this modification we generated rLCMV with mutations at tyrosine 88 that either prevented phosphorylation ( Y88F or Y88A ) or mimicked it ( Y88E ) . Relative to the mutants that cannot be phosphorylated , the Y88E phosphomimetic virus generated significantly more DI particles per infectious particle ( Figs 4 and 5C ) . This suggests that reversible phosphorylation of the PPXY motif may act as a rheostat to regulate the rate of DI particle production independent of standard virus , possibly through the recruitment of ESCRT machinery . Interestingly , the terminal tyrosine of the PPXY late domains encoded by Ebola virus and Marburg virus VP40 is also phosphorylated . Inhibiting phosphorylation of the Ebola virus VP40 PPXY motif reduced the release of infectious virus [25] whereas mutation of the Marburg virus VP40 PPXY motif to prevent phosphorylation did not impact infectious VLP release , but instead impaired the recruitment and incorporation of nucleocapsids into VLPs [29] . The role of the PPXY domain and/or its phosphorylation with regard to DI particle production in the filovirus model and other PPXY-containing virus families remains an open question . In the current model , it will be important to identify the host kinase responsible for phosphorylating the LCMV PPXY domain , although this may require a comprehensive tyrosine kinase screen as the flanking residues surrounding Y88 were not recognized by kinase motif prediction tools [56 , 57] . How does the PPXY late domain of LCMV promote the release of DI particles ? While it is well known that the PPXY motif can drive viral budding , the exact mechanism by which it recruits ESCRT proteins to promote this process is not fully understood . Several PPXY-dependent viruses also require NEDD4 family E3 ubiquitin ligases ( e . g . , NEDD4 , ITCH , or WWP1 ) , which can directly bind to the PPXY motif via their WW domains ( for review see [16] ) . Recruitment of ESCRT machinery could occur due to ubiquitination of Z by a NEDD4 E3 ubiquitin ligase , which in turn could recruit ESCRT components ( e . g . TSG101 ) that can bind ubiquitinated proteins [16] . Alternatively , arrestin-related trafficking adaptors ( ARTs ) may provide an ESCRT linkage as these proteins interact with both NEDD4 E3 ubiquitin ligases as well as ESCRT proteins and have been shown to influence both multivesicular body formation and viral budding [58] . Under this scenario , phosphorylation of the PPXY motif could regulate DI particle release by modifying the accessibility of the PPXY domain to NEDD4 E3 ubiquitin ligases . All viruses must strike a balance between pathogenesis and persistence , with the ultimate goal of ensuring their own maintenance in nature . It has long been postulated that DI particles may be one way in which LCMV is able to tip the scales towards persistence , thus lowering its immunological profile and fitness cost to its host . By identifying the specific cellular pathway required to form these DI particles , and the apparent importance of viral phosphorylation in accessing the pathway , our findings raise the possibility that arenaviruses can dynamically adjust DI particle production in response to external environmental factors . Further , the PPXY mutant viruses we have developed represent a new tool that will allow the field to formally test the importance of DI particles for the establishment of persistent LCMV infection in reservoir rodents . The ability of the PPXY late domain to drive the production of DI particles is a novel finding with important implications for understanding host-pathogen relationships and the design of vaccines and antivirals . In particular , PPXY-containing viruses such as Ebola virus and Lassa virus are known to persist for long periods following the resolution of acute human disease [59–61] . It is possible that DI particles play a significant role in allowing these viruses to persist in humans , similar to their presumed importance for infection in reservoir species . Therefore , targeting DI particle formation could be a promising approach to clear persistent infection in humans . Finally , the possibility that the PPXY late domain and ESCRT machinery could broadly drive DI particle production in other virus families represents an exciting area of future research .
Human embryonic kidney cells ( HEK-293T/17 ) ( CRL-11268 , American Type culture Collection , Manassas , VA ) ( referred to as HEK293T cells in the manuscript ) were maintained in Dulbecco’s Modified Eagle Medium ( DMEM ) ( 11965–092 ) supplemented with 10% fetal bovine serum ( FBS ) ( 16140–071 ) , 1% penicillin-streptomycin ( 15140–122 ) , 1% MEM Non-Essential Amino Acids Solution ( 11140–050 ) , 1% HEPES Buffer Solution ( 15630–130 ) , and 1% GlutaMAX ( 35050–061 ) purchased from Invitrogen ( Carlsbad , CA ) . L929 mouse fibroblast cells ( CCL-1 , American Type culture Collection ) were maintained in Minimum Essential Medium ( MEM ) ( 11095–080 ) supplemented with 10% FBS , 1% penicillin-streptomycin , 1% MEM Non-Essential Amino Acids Solution , 1% HEPES Buffer Solution , and 1% GlutaMAX . Baby hamster kidney cells ( BHK-21 ) were kindly provided by M . J . Buchmeier ( University of California , Irvine ) and grown in DMEM supplemented with 10% FBS , 1% penicillin-streptomycin , and 1% GlutaMAX . African green monkey kidney cells ( Vero E6 ) were kindly provided by J . L . Whitton ( The Scripps Research Institute , La Jolla ) and grown in DMEM supplemented with 10% FBS , 1% penicillin-streptomycin , and 1% HEPES Buffer Solution . T-Rex HEK293 cells stably transduced with a tetracycline-inducible plasmid encoding WT or dominant negative EQ mutant vacuolar protein sorting 4A ( VPS4A ) or VPS4B as described in [40–42] were generously provided by M . Kielian ( Albert Einstein College of Medicine , Bronx ) and were maintained in DMEM supplemented with 10% FBS , 1% penicillin-streptomycin , 1% MEM Non-Essential Amino Acids Solution , 1% HEPES Buffer Solution , 1% GlutaMAX , and 100 μg/mL Zeocin ( R250-01 , Invitrogen ) . VPS4 expression was induced by incubating cells in the above growth medium containing 1 μg/mL tetracycline as described [42] . All cell lines were grown at 37°C in a humidified incubator containing 5% CO2 . Lymphocytic choriomeningitis virus ( LCMV ) strain Armstrong 53b was kindly provided by J . L . Whitton . Wild-type vesicular stomatitis virus expressing green fluorescent protein ( VSV-GFP ) as described in [62] was kindly provided by J . Hiscott ( Vaccine and Gene Therapy Institute of Florida , Port St . Lucie ) and M . Shaw ( Icahn School of Medicine at Mount Sinai , New York ) . Junín virus ( JUNV ) C#1 , which is an attenuated vaccine strain derived from the virulent WT JUNV strain XJ as described in [63 , 64] , was kindly provided by M . Buchmeier ( University of California , Irvine ) and R . Tesh ( The University of Texas Medical Branch at Galveston ) . Working stocks of these viruses were amplified and titered ( via plaque assay ) on Vero E6 cells . See below under “Generation of Recombinant LCMV” for a description of the recombinant ( r ) LCMV strain Armstrong 53b that were generated for this study . The LCMV Armstrong 53b Z protein ( WT , Y88A , Y88E , or Y88F ) was subcloned into a modified pCAGGS expression vector [65] and different combinations of these plasmids were used to screen for the phosphorylation of Z ( Fig 1D ) or the budding efficiency of Z ( Fig 2E ) . The WT and Y88 mutant Z genes were fused to the streptavidin binding peptide ( SBP ) ( MDEKTTGWRGGHVVEGLAGELEQLRARLEHHPQGQREP ) through an 18 base pair linker at the C-terminus of Z to permit affinity purification and western blot detection of Z . The nucleotide sequence of the WT Z gene matches NCBI gene identifier number AY847351 while the translated amino acid sequence for the WT Z gene matches Protein Locus number AAX49343 . WT Z was amplified by PCR from the pT7-L ( + ) plasmid generously provided by J . C . de la Torre ( The Scripps Research Institute , La Jolla ) [66] using the forward primer LCMVZf ( 5’-ACAAGTTTGTACAAAAAAGCAGGCTGATATCGCCACCATGGGTCAAGGCAAGTCCAGA-3’ ) , which has a 5’ overhang containing Gateway AttB1 and Kozak sequences and the reverse primer LCMVZr ( 5’-ACCTCCACCTCCAGCTGCCTCTTCGTAGGGAGGTGGAGA-3’ ) , which has an overhang containing the linker sequence . The SBP tag was amplified from the pT7-FLAG-SBP-1 plasmid ( P3871 , Sigma-Aldrich , St . Louis , MO ) via PCR using the forward primer SBPf ( 5’- GCAGCTGGAGGTGGAGGTATGGACGAAAAAACCACCGGT-3’ ) , which has a 5’ overhang containing the linker sequence , and the reverse primer SBPr ( 5’-ACCACTTTGTACAAGAAAGCTGGGTCTTACGGTTCACGCTGACCCTGCGG-3’ ) , which contains a 3’ overhang with a stop codon preceding an AttB2 sequence . The two products were fused by PCR using the Z forward primer and the SBPr primer . The cassette was subcloned into the pCAGGS vector using the Gateway cloning system ( Invitrogen ) following the manufacturer’s instructions as has been previously described [65 , 67] . The plasmids pC-NP and pC-GP , which express the LCMV Armstrong 53b nucleoprotein ( NP ) and glycoprotein ( GP ) , respectively , and plasmids pol-I S and pol-I L , which express the LCMV L and S genome segments , respectively , were used to generate rLCMV . These reagents were generously provided by J . C . de la Torre and are described in [68] . Each of the Y88 mutant Z genes used in these studies were synthesized and subcloned into the pCAGGS or pol-I L vectors , respectively , by Biobasic , Inc . ( Markham , ON , Canada ) . A pol-I L vector containing an SBP-tag directly fused to the C-terminus of Z was also generated by Biobasic , Inc . All plasmid sequences were verified by DNA sequencing . To identify phosphorylation sites on LCMV Z via mass spectrometry , Vero E6 cells were infected with LCMV strain Armstrong 53b and 48 hr later cell-free virions were purified by sucrose-banding as described previously [69] . Purified virions were then lysed in Triton buffer ( 0 . 5% NP40 , 1% Triton X-100 , 140mM NaCl , and 25mM Tris-HCl containing a protease inhibitor cocktail ( 04693159001 , Roche Applied Science , Indianapolis , IN ) ) and mixed with Laemmli sample buffer ( 62 . 5 mM Tris-HCl , 10% glycerol , 2% sodium dodecyl sulfate and 0 . 01% bromophenol blue ( B392 , Fisher Scientific , Pittsburgh , PA ) ) containing 5% 2-mercaptoethanol . Virion protein lysates were separated on a 4–20% Tris-Glycine polyacrylamide gel ( EC60255 , Invitrogen ) . The gel was stained with Coomassie ( 40% methanol , 20% acetic acid , and 0 . 1% Brilliant Blue R ( B7920 , Sigma-Aldrich ) ) , destained with a solution of 30% methanol and 10% acetic acid , and then imaged using a Canon Canoscan 8800F scanner . For mass spectrometry , the protein band corresponding to the Z protein was excised and cut into 1 mm cubes and processed with chemicals from Fisher Scientific as follows . The gel pieces were rinsed with HPLC grade water and then incubated with destain solution ( 50 mM ammonium bicarbonate and 50% acetonitrile ) for 30 minutes at 37°C . The destain solution was removed and the gel pieces were dehydrated by incubating twice with 100% acetonitrile for 5 minutes . The gel pieces were reduced with 25 mM dithiothreitol in 50 mM ammonium bicarbonate for 30 minutes at 55°C . After cooling for 10 minutes at room temperature , the gel pieces were dehydrated by incubating with 100% acetonitrile for 5 minutes and then alkylated in the dark with 10 mM iodoacetamide in 50 mM ammonium biocarbonate for 45 minutes at room temperature . The gel pieces were washed two times in destain solution for 5 minutes , dehydrated with 100% acetonitrile , then rehydrated with water for 10 minutes . The gel pieces were further dehydrated with two 5 minute incubations in 100% acetonitrile before removing all liquid and drying the gel pieces at room temperature for 10 minutes . The gel pieces were rehydrated with a solution of 12 . 5 ng/μL sequencing grade chymotrypsin ( V1061 , Promega , Madison , WI ) or 12 . 5 ng/μL sequencing grade modified trypsin ( V5111 , Promega ) in 50 mM ammonium bicarbonate on ice for 30 minutes , before digesting overnight at 37°C . Peptides were extracted with a solution of 2 . 5% formic acid in 50% acetonitrile while spinning in a microcentrifuge at 13 , 000 rpm for 10 minutes . The supernatant was removed and saved while the gel pieces were subjected to further extraction and rinsing with 100% acetonitrile . The second extraction was combined with the initial extraction . All solvent was removed from the extracts using a vacuum centrifuge at 37°C . The peptides were resuspended in 2 . 5% formic acid , 2 . 5% acetonitrile prior to mass spectrometry analysis . Peptides were separated over 12 cm of Magic C18 , 5 μM , 200 Å reversed phase material ( PM5/66100/00 , Michrom Bioresources , Auburn , CA ) in a microcapillary column using a MicroAS autosampler ( Thermo Scientific , Pittsburgh , PA ) . Following 15 minutes of isocratic loading in 2 . 5% acetonitrile , 0 . 15% formic acid , the peptides were eluted from the column with a 5–35% gradient of acetonitrile with 0 . 15% formic acid over 40 minutes using a Surveyor Pump Plus HPLC ( Thermo Scientific ) . Mass spectra were acquired either in an LTQ-XL linear ion trap , or in a linear ion trap-orbitrap mass spectrometer ( Thermo Scientific ) as described previously [70] . Briefly , for most analyses 10 data-dependent MS/MS spectra followed each survey scan . However , in several cases after obtaining the initial spectra for phosphopeptides we followed up with targeted MS/MS spectra in order to increase fragment ion coverage . The IPI human forward and reverse concatenated database was used to search the raw data using SEQUEST software requiring tryptic peptides and either a 2 Da precursor mass tolerance ( for precursor data acquired in the LTQ ) or 20 PPM ( for precursor data acquired in the orbitrap ) . In the searches the following precursor mass differences were allowed: serine , threonine , and tyrosine residues ( +79 . 96633 Da ) ; methionine ( +15 . 99492 Da ) and cysteines ( +57 . 02146 Da or 71 . 0371 ) . To confirm that Z was phosphorylated in human cells as well as cells from rodent cells , in Fig 1D and 1E , plasmid-derived Z expressed in HEK293T cells and Z from rLCMV Z-SBP-infected cells were both probed for phosphotyrosine signal via western blot . For plasmid-derived Z , 2 x 105 HEK293T cells were seeded in a 12-well plate and transfected the next day with 0 . 8 μg per well of pLCMV-Z WT , pLCMV-Z Y88F , or an empty vector using 0 . 8 μL of a 1 mg/mL solution of polyethylenimine ( 23966 , Polysciences , Inc . , Warrington , PA ) per well . For Z derived from rLCMV Z-SBP-infected cells , 2 . 5 x 105 L929 cells were seeded in 6-well plates and infected the next day at an MOI of 0 . 01 . Two days following the transfection or infection , H202 at a final concentration of 8 . 8 mM or an equivalent volume of H20 was spiked into the appropriate wells containing HEK293T or L929 growth media . After a 15 minute incubation , the cells were lysed in Triton buffer containing a protease inhibitor cocktail and PhosStop phosphatase inhibitor cocktail ( 04906837001 , Roche Applied Science ) and the SBP-tagged Z proteins were affinity purified using magnetic streptavidin beads as previous described [67] . The purified proteins were separated via SDS-PAGE and screened for Z or tyrosine phosphorylated-Z via standard chemiluminescent western blotting and detected with film ( Fig 1D ) or with a LI-COR C-Digit digital imager ( LI-COR , Lincoln , NE ) ( Fig 1E ) . rLCMV WT , rLCMV Z-SBP and rLCMV containing Z-Y88 mutations ( Y88F , Y88E , Y88A ) were generated using the previously described reverse genetics system [68] . Briefly , 10 μL of Lipofectamine 2000 ( 52887 , Invitrogen ) was mixed with 100 μL of OptiMEM ( 31985 , Invitrogen ) and then added to a plasmid mixture consisting of 1 . 6 μg pC-NP , 2 . 0 μg pC-L , 1 . 6 μg pol-I S , and 2 . 8 μg pol-I L ( WT , Z-SBP or containing the described Y88 point mutations ) in 100 μL OptiMEM and incubated at room temperature for 25 minutes . 200 μL of this transfection mixture and 800 μL of OptiMEM was then added to 1 well of a 6-well plate which had been seeded the previous day with 3 . 5 x 105 BHK-21 cells and washed prior to transfection with 1 mL of OptiMEM . The cells were incubated with the transfection mixture for 4 hr after which the media was replaced with BHK-21 growth media diluted 5-fold in DMEM . Three days later the supernatant was collected , clarified by centrifugation at 1 , 200 RPM for 5 minutes at 4°C , and used to infect a fresh monolayer of 1 . 8 x 106 BHK-21 cells in a T-75 flask . Following a 1 hr absorption , the inoculum was removed and fresh BHK-21 growth media diluted 5-fold in DMEM was added to the cells . Three days later the supernatant of this flask was collected , clarified by centrifugation , and titered by plaque assay . To generate an expanded virus stock , Vero E6 cells were infected with this material at an MOI of 0 . 0001 and 48 or 72 hr later , supernatants were collected , clarified , and titered by plaque assay . A portion of the L segment ( most of the Z gene , the intergenic region , and part of the L gene ) of each rLCMV Y88 mutant was sequenced to ensure that these viruses had not reverted . The material used for this sequencing was derived from the 72 hr pi time point shown in Fig 2B . Viral RNA from clarified supernatants was isolated using the Qiagen Viral RNA mini kit ( 52906 , Qiagen , Valencia , CA ) according to the manufacturer’s protocol . Viral RNA was converted to cDNA using primer L 845- ( 5’- GCAGGACTTGAGGGCTATGA-3’ ) , Superscript III ( 18080–044 , Invitrogen ) , RNAse Out ( 10777–019 , Invitrogen ) , and 5 μL of RNA following the manufacturer’s protocol for first strand cDNA synthesis . A portion of the L-segment containing Z was amplified with 30–40 cycles of PCR using Platinum Pfx DNA polymerase ( 11708–013 , Invitrogen ) and primers L126+ ( 5’- ATAGTACAAACAGGGCCGAAATCC-3’ ) and L764- ( 5’- TTTGTTGGGTTCAGAGATAAGTGT-3’ ) following the manufacturer’s protocol . The PCR product was prepared for sequencing using ExoSAP-IT ( 78200 , Affymetrix , Santa Clara , CA ) following the manufacturer’s protocol and sequenced by the University of Vermont Cancer Center DNA Analysis Facility . Protein lysates were diluted in Laemmli sample buffer containing 5% 2-mercaptoethanol and separated on NuPAGE 4–12% Bis-Tris gels with MES buffer . Protein was transferred to nitrocellulose membranes using iBlot gel transfer stacks ( IB301001 or IB301002 , Invitrogen ) and the Invitrogen iBlot Device as directed by the manufacturer . Efficient protein transfer was confirmed by staining membranes with a solution containing 0 . 1% Ponceau S ( P3504 , Sigma-Aldrich ) and 5% acetic acid which was subsequently removed by washing with water . Two methods were used for protein detection: quantitative LI-COR-based detection or standard chemiluminescent-based detection . For quantitative LI-COR analysis , membranes were blocked with a solution of 5% milk in PBS for 1 hr and incubated overnight at room temperature with the indicated primary antibodies diluted in PBS containing 5% milk and 0 . 2% Tween 20 ( BP337 , Fisher Scientific ) . Following 5 washes in PBS with 0 . 5% IGEPAL CA-630 ( 198596 , MP Biomedicals , Solon , OH ) , the membranes were incubated for 1 hr at room temperature with secondary antibodies diluted in PBS containing 5% milk , 0 . 2% Tween 20 and 0 . 02% sodium dodecyl sulfate , washed 5 times in PBS with 0 . 5% IGEPAL CA-630 and 1 time with PBS , then imaged using the LI-COR Odyssey CLx imaging system . For quantitative LI-COR analysis of VPS4B in Figs 6A , 6B and S5 , membranes were probed using the iBind Flex western device ( SLF2000 , Thermo Scientific ) with the iBind Flex fluorescent detection solution kit ( SLF2019 , Thermo Scientific ) following the manufacturer’s instructions . For chemiluminescent-based detection of phosphorylated proteins , the same general procedure was used with the following exceptions: i ) membranes were blocked with either PBS containing 5% milk and 0 . 05% IGEPAL CA-630 or protein-free blocking buffer ( 37572 , Thermo Scientific ) , ii ) primary and secondary antibodies were diluted in PBS containing 5% milk , 0 . 05% IGEPAL CA-630 , and 3% fetal bovine serum or protein-free blocking buffer , and iii ) the secondary antibodies were incubated with the membrane for 2 hr . The following primary antibodies were used for western blotting ( at the indicated concentrations ) : mouse anti-streptavidin binding peptide ( MAB10764 , Millipore , Billerica , MA ) ( 1:10 , 000 ) , rabbit anti-actin ( A2066 , Sigma-Aldrich ) ( 1:10 , 000 ) , mouse anti-actin ( A5441 , Sigma-Aldrich ) ( 1:5 , 000 ) , rabbit anti-actin ( A2066 , Sigma-Aldrich ) ( 1:2 , 500 ) , mouse anti-phosphotyrosine ( clone 4G10 , Millipore ) ( 0 . 2 μg/mL ) , mouse anti-green fluorescent protein ( 632380 , Clontech , Mountain View , CA ) ( 1:1 , 000 ) , rabbit anti-LCMV Z ( 880 ) ( 1:500 ) , mouse anti-LCMV GP2 ( 33 . 6 ) ( 1:2 , 000 ) , and rabbit anti-LCMV nucleoprotein ( 2165 ) ( 1:5 , 000 ) . Antibodies 880 , 2165 , and 33 . 6 were generously provided by M . J . Buchmeier ( University of California , Irvine ) . For quantitative western blotting , the following secondary antibodies from LI-COR were used: IRDye 800CW goat anti-mouse ( 926–32210 ) for the Z release assay in Fig 2E at 1:20 , 000 and in Figs 6A , 6B and S5 at 1:3 , 000 ( for probing by iBind ) and IRDye 680LT goat anti-mouse ( 926–68020 ) and IRDye 800CW Goat anti-rabbit ( 926–32210 ) were used at 1:20 , 000 to detect proteins in Fig 3A–3D . IRDye 680LT goat anti-mouse was used at 1:3 , 000 ( for probing by iBind ) in S5 Fig to detect actin . A horseradish peroxidase-conjugated anti-mouse secondary antibody ( 71045 , EMD Millipore , Billerica , MA ) diluted 1:5 , 000 was used for chemiluminescent-based detection in Fig 1D and 1E . To determine the growth kinetics of rLCMV in Fig 2B , 6-well plates were seeded with 1 . 9 x 105 Vero E6 cells per well . The following day the cells were infected with each respective rLCMV at an MOI of 0 . 01 . Supernatants were collected at 12 , 24 , 36 , 48 , and 72 hr pi , clarified by centrifugation at 1 , 200 RPM for 5 minutes at 4°C , then titered by plaque assay . To determine the release efficiency of the Y88 mutant Z proteins in Figs 2E and S5 , 2 x 105 HEK293T cells or T-Rex HEK293 cells stably transduced with a tetracycline-inducible plasmid encoding WT or dominant negative EQ mutant VPS4B were seeded in a 12-well plate . The next day cells were transfected with 0 . 8 μg per well of pLCMV-Z WT , Z-G2A , -Z Y88F , -Z Y88E , or -Z Y88A using 0 . 8 μL of a 1 mg/mL solution of polyethylenimine per well . For the experiments shown in S5 Fig , VPS4B expression was induced with 1 μg/mL tetracycline at the time of transfection . The following day ( 24 hr post-transfection ) cells and VLP-containing media ( which had been clarified ) were collected , lysed with Triton lysis buffer , and subjected to quantitative western blotting . For detection of Z from VLPs produced in VPS4B WT or DN cell lines , SBP-tagged Z was affinity purified from lysed VLP-containing media using magnetic streptavidin beads prior to quantitative western blot analysis . To calculate the percent VLP release we first normalized each Z protein value ( from supernatants or cells ) by the sum of all Z protein bands on a particular gel as described in [71] . The percent VLP release was then calculated as the quotient of the Z protein quantity in VLPs divided by the quantity of Z in cells [ ( ZmutVLP / Zmutcells ) / ( ZWT VLP/ ZWT cells ) ] . To measure infectious virus titers , a standard plaque assay was employed as follows . Six-well plates were seeded with 1 x 105 ( LCMV and JUNV ) or 1 x 106 ( VSV ) Vero E6 cells per well and the following day inoculated with 10-fold serial dilutions of virus in a total volume of 0 . 5 mL of Vero E6 growth medium . Following a 90 minute absorption at 37°C , the cells were overlaid with a solution of 0 . 7% agarose ( 20–102 , Apex Industrial Chemicals , Aberdeen , United Kingdom ) in Vero E6 growth media . The plates were fixed 2 ( VSV ) or 4 ( LCMV and JUNV ) days later with a solution of 2 . 5% formaldehyde ( 1635-4L , Sigma ) in 3x PBS . Following removal of the agarose plugs , the fixed monolayers were stained with 0 . 1% crystal violet ( C581-100 , Fisher Scientific ) and 2 . 1% ethanol in water . To determine the plaque size of rLCMV in Fig 2D or the overall level of cytopathic effect induced by these viruses in Figs 4A , 4B , 6C and 6D , the wells were imaged with an Alpha Innotech digital camera paired to a Computar H6Z0812M motorized zoom lens . The area of each plaque as well as the mean pixel intensity of each well was calculated using ImageJ software . To determine the titer of LCMV DI particles , samples were transferred to clear snap cap tubes ( 21-402-904 , Thermo Scientific ) and irradiated for 2 minutes with UV light in a UVP CL-1000 ultraviolet crosslinker in to kill standard infectious virus . The samples were serially diluted in 5-fold increments and added to 24-well plates which had been seeded the previous day with 20 , 000 ( LCMV and JUNV C#1 ) or 100 , 000 ( VSV ) Vero E6 cells per well . Subsequently , 50 PFU per well of rLCMV WT ( or 50 PFU per well of JUNV C#1 or VSV in Fig 5A ) was added to each well containing UV-irradiated samples . UV-irradiated samples were also added to a second set of wells to which no standard virus was added to ensure that all infectious virus had been eliminated from the samples . After a 90 minute absorption period at 37°C , the cells were overlaid with a solution of 0 . 7% agarose in Vero growth media and left at 37°C . The plates were fixed and stained 2 ( VSV ) or 4 ( LCMV and JUNV C#1 ) days later as above for the plaque assay . The plaques were counted in each well and the plaque interfering unit 50 ( PIU50 ) was calculated using the plaque reduction statistical web tool ( https://exon . niaid . nih . gov/plaquereduction ) . Because a unique biochemical or genetic signature to differentiate standard infectious virus particles from DI particles has not been defined , the assay we employed relied on measurement of the interfering activity of DI particles as opposed to a direct physical measure of the particles themselves . For Fig 5B , rLCMV WT was filtered with either 0 . 45 μM ( 28145–481 , VWR , Radnor , PA ) or 0 . 2 μM ( 09-719C , Fisher Scientific ) syringe filters or Amicon 30K ( UFC903024 , Millipore ) or 10K ( UFC901024 , Millipore ) centrifugal filters prior to treatment with UV light and DI titering as above . To determine the role of the ESCRT pathway in LCMV release , 2 . 5 x 105 T-Rex HEK293 cells stably transduced with a tetracycline-inducible VPS4A or VPS4B ( WT or dominant negative EQ in each case ) were seeded in 6-well plates that were first coated with poly D-lysine ( P6407 , Sigma-Aldrich ) for 5 minutes then washed 3x with PBS . Cells were infected 24 hr later with rLCMV WT at an MOI of 0 . 001 . Forty-eight hr later ( when all cells were productively infected ) the cells were induced with growth medium containing 1 μg/mL tetracycline or a medium only control . Six hr after induction cells were washed 3x with PBS and fresh growth medium containing 1 μg/mL tetracycline or medium alone were added . Eighteen hr later the cells and supernatants were collected . In Fig 6A and 6B , the cell lysates were probed for VPS4B DN or WT protein ( via the GFP fusion tag on these proteins ) or actin expression by quantitative western blotting . Supernatants were titered by plaque assay for infectious virus and DI particle levels by measuring the cytopathic effect in a plaque assay with equal PFUs of virus in each well ( as described under plaque assay ) and/or by plaque interference assay . The role of VPS4B in VSV release was also tested . For the VSV challenge studies , 5 x 105 VPS4B WT or EQ cells were seeded in poly D-lysine treated wells and 24 hr later treated with either growth medium containing 1 μg/mL tetracycline or medium alone . One hr later , the cells were infected with VSV at an MOI of 10 . One hr following infection , the cells were washed 3x with PBS and fresh growth medium containing 1 μg/mL tetracycline or medium alone was added . Six hr later the cells and supernatants were collected and assessed by quantitative western blotting and plaque assays , respectively . In order to verify uniform VPS4B expression as well as rLCMV WT infection by microscopy , in parallel to the experiment described above , 5 x 104 cells were seeded on poly D-lysine-treated 12mm glass coverslips in 24-well plates . At the time of harvest ( 24 hr post-infection ) the coverslips were rinsed with PBS , fixed with 4% paraformaldehyde ( 15714 , Electron Microscopy Sciences , Hatfield , PA ) in PBS for 20 minutes , then washed 2x with PBS for 5 minutes . The cells were permeabilized with 0 . 1% Triton X-100 in 1% bovine serum albumin ( BSA ) in PBS , blocked with 10% normal goat serum ( 005-000-121 , Jackson , West Grove , PA ) in 1% BSA in PBS , and immunostained with anti-LCMV nucleoprotein antibody ( 1 . 3–3 ) ( 1:500 ) kindly provided by M . Buchmeier ( University of California , Irvine ) and secondary anti-mouse Alexafluor 555 ( A28180 , Thermo Scientific ) ( 1:1 , 000 ) each for 1 hr in 1% BSA in PBS . DNA was detected with 4’ , 6-diamidino-2-phenylindole hydrochloride ( DAPI ) ( D9542 , Sigma Aldrich ) in 1% BSA in PBS . Cells were washed with 1% BSA in PBS in between each step . Images were acquired on a Zeiss LSM 510 laser scanning confocal microscope using a 63X objective lens . Post-capture image processing was carried out in FIJI and Photoshop; the GFP fluorescence , NP staining , and DAPI signal are shown at equal exposures in all conditions . To determine the NP , GP , and Z protein content of rLCMV virions in Fig 3A–3D , 2 x 106 Vero E6 cells were seeded in a T-150 culture flask and infected the next day at an MOI of 0 . 01 , 0 . 001 , or 0 . 0001 . At 48 or 72 hr following inoculation , the supernatant was collected , clarified by centrifugation , and screened for infectious virus by plaque assay . An equal number of plaque forming units of each virus ( range 1 to 3x107 PFU per experiment ) were layered onto a solution of 20% sucrose in TNE buffer , pH 7 . 4 ( 10 mM Tris base , 1 mM EDTA , 0 . 2 M NaCl ) and centrifuged for 2 hr at 30 , 000 rpm at 4°C in a Thermo-Scientific Sorval WX Ultra 80 ultra centrifuge equipped with a Sorval Surespin 630 rotor . The resulting virus pellet was resuspended in 2X-concentrated Laemmli buffer containing 5% 2-mercaptoethanol , then analyzed by SDS-PAGE and quantitative western blotting . To separate rLCMV by gradient centrifugation in S2 Fig , 2 x 106 Vero E6 cells were seeded in a T-150 culture flask and infected the next day at an MOI of 0 . 0001 . At 72 hr following inoculation , the supernatant was collected and clarified by centrifugation . The clarified supernatants were added to 50 mL tubes ( 430290 , Corning ) containing polyethylene glycol ( PEG ) 8000 ( 81268 , Sigma-Aldrich ) and sodium chloride such that the final concentrations were 10% and 1% , respectively . The solutions were incubated at 4°C on a rotating platform for 2 hr then were centrifuged for 30 minutes at 10 , 000 rpm at 4°C in a Thermo-Scientific Sorval Legend RT+ centrifuge equipped with a Sorval Fiberlite F15-8x50cy rotor . The supernatant was removed and the virus-PEG pellet was resuspended in TNE buffer and screened for infectious virus by plaque assay . Density gradients were prepared by layering solutions of 7% , 10% , 13% , 16% , and 19% optiprep ( D1556 , Sigma-Aldrich ) diluted in PBS in 36 mL tubes ( 03141 , Thermo Scientific ) then leaving overnight at 4°C to allow a continuous gradient to form . An equal number of plaque forming units of each virus ( range 4 x 107 to 1 x 108 PFU per experiment ) was layered onto the continuous gradient and centrifuged for 12 hr at 30 , 000 rpm at 4°C in a Thermo-Scientific Sorval WX Ultra 80 ultracentrifuge equipped with a Sorval Surespin 630 rotor . The resulting separated virus was collected in 2 mL fractions using a New Era NE-9000G programmable peristaltic pump and titered via plaque assay . To enumerate copies of LCMV S and L segment genomic RNA contained in virions for Fig 3E and 3F , viral RNA was extracted from cell-free virions using the Qiagen Viral RNA mini kit according to the manufacturer’s instructions and subjected to quantitative RT-PCR as previous described [72] . Briefly , cDNA was generated in a 50 μL RT reaction containing 5 μL of viral RNA , 0 . 2 μM of the gene specific primer S 2865- ( 5’-CAGGGTGCAAGTGGTGTGGTAAGA-3’ ) or L 5906- ( 5’- TGGGACTGAGTTTCGAGCATTACG-3’ ) , which are complementary to the S or L segment genomic RNA , 5 μL of 10x PCR Buffer II ( #E12874 , Applied Biosystems , Carlsbad , CA ) , 5 μL of 10 mM dNTP mix ( 362275 , Applied Biosystems ) , 1 μL RNase inhibitor ( N808-0119 , Applied Biosystems ) , and 1 . 25 μL of Multiscribe reverse transcriptase ( 4308228 , Applied Biosystems ) . RT reaction conditions were 25°C for 10 minutes , 48°C for 30 minutes , and 95°C for 5 minutes . Quantitative PCR was then performed in a 25 μL reaction volume consisting of 5 μL of cDNA , 0 . 9 μM each of the forward primer S 2275+ ( 5’-CGCTGGCCTGGGTGAAT-3’ ) or L 5517+ ( 5’-GGCCTTGTATGGAGTAGCACCTT-3’ ) and reverse primer S 2338- ( 5’-ATGGGAAAACACAACAATTGATCTC-3’ ) or L 5645- ( 5’-GGTCTGTGAGATATCAAGTGGTAGAATG-3’ ) , 0 . 2 μM of the TaqMan probe S 2295+ ( 5’-6FAM-CTGCAGGTTTCTCGC-MGBNFQ-3’ ) or L 5582- ( 5’-6FAM-CTGAAGAATACCACCTATTATACCA-MGBNFQ-3’ ) , and 12 . 5 μL of the TaqMan Universal PCR Master Mix ( 4326614 , Life Technologies , Grand Island , NY ) . Reaction conditions were 95°C for 10 minutes followed by 40 cycles of 95°C for 15 seconds and 60°C for 1 minute . Copy numbers of LCMV S or L segment genomic RNAs were calculated by comparison with a series of standard dilutions of the pT7-S or pT7-L plasmids as described [72] . Data was generated on an Applied Biosystems StepOnePlus Real-Time PCR System and analyzed with the provided software . Statistical analysis was performed using GraphPad Prism software . For the virus growth curves in Fig 2B , the data was first log transformed , then a two-way analysis of variance ( ANOVA ) was performed with a Holm-Sidak’s test for multiple comparisons to compare viruses at each time point . A one-way ANOVA with Holm-Sidak’s test for multiple comparisons was used to analyze the VLP release assay in Fig 2E , the viral protein levels in concentrated virions in Fig 3B–3D , and the S and L segment to PFU ratios in Fig 3E and 3F . To compare plaque area in Fig 2D , the data were first tested for normality using the D’Agostino and Pearson omnibus normality test , then the Kruskal-Wallis non-parametric test was used and multiple comparisons were made with Dunn’s multiple comparisons test . To analyze the cytopathic effect induced by rLCMV WT or Z-Y88 mutants ( Fig 4B ) or by rLCMV WT generated in VPS4B WT or dominant negative cells ( Fig 6D ) , a two-way ANOVA was performed with the Holm-Sidak’s test for multiple comparisons . To compare VSV or LCMV virus titers , LCMV DI particle titers , or Z VLP levels produced in VPS4B WT or EQ cells ( Fig 6A , 6B and 6E and S5 Fig ) a two-tailed unpaired t test with Welch’s correction was performed . To compare DI particle titers in Fig 5A–5C , a value of 19 PIU50/mL ( just below the limit of detection value of 20 PIU50/mL ) was substituted for samples that were below the limit of detection and then a one way ANOVA was performed . For all statistical analyses , the data utilized was generated from at least 3 independent experiments as indicated in each respective figure legend . | Arenaviruses cause severe and often fatal diseases in humans yet typically establish lifelong , asymptomatic infections in their rodent reservoirs . Several families of enveloped RNA viruses , including the arenaviruses , encode short amino acid motifs , called late domains , to hijack host proteins in the endosomal sorting complex required for transport ( ESCRT ) to drive the release of virus particles from the host cell’s outer membrane . Many late domain-containing viruses produce defective interfering ( DI ) particles in addition to standard , infectious virus . DI particles cannot self-replicate but interfere with the production of infectious virus and mitigate virus-induced cytopathic effect . Arenaviruses such as lymphocytic choriomeningitis virus ( LCMV ) generate high levels of DI particles , yet , the mechanism that drives their formation is not known . We show that LCMV’s only encoded late domain , PPXY , and a functional ESCRT pathway are critical for DI particle production , but in contrast , are not absolutely required for infectious virus production . We also demonstrate that the LCMV PPXY late domain is phosphorylated and that this modification may regulate DI particle production . In summary , we have discovered a new and unexpected role for a viral late domain in selectively driving the production of DI particles independently of standard infectious virus particles . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"phosphorylation",
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"vesicular",
"stomatitis",
"virus",
"pathology",
"and",
"laboratory",
"medicine",
"chemical",
"compounds",
"pathogens",
"drugs",
"microbiology",
"viral",
"structure",
"organic",
"compounds",
"viruses",
"tyrosine",
"rna",
"viruses",
"tetracyclines",
"antibiotics",
"amino",
"acids",
"pharmacology",
"aromatic",
"amino",
"acids",
"proteins",
"medical",
"microbiology",
"lymphocytic",
"choriomeningitis",
"virus",
"microbial",
"pathogens",
"chemistry",
"virions",
"biochemistry",
"rhabdoviruses",
"organic",
"chemistry",
"post-translational",
"modification",
"viral",
"persistence",
"and",
"latency",
"virology",
"viral",
"pathogens",
"arenaviruses",
"hydroxyl",
"amino",
"acids",
"biology",
"and",
"life",
"sciences",
"microbial",
"control",
"physical",
"sciences",
"organisms"
] | 2016 | The Lymphocytic Choriomeningitis Virus Matrix Protein PPXY Late Domain Drives the Production of Defective Interfering Particles |
Cutaneous leishmaniasis ( CL ) is a neglected worldwide , zoonotic , vector-borne , tropical disease that is a threat to public health . This threat may spread from endemic to non-endemic areas . Current research has exploited epidemiological , molecular and phylogenetical studies to determine the danger of an outbreak of CL in the borderline area between northern and central Iraq from 2014–2017 . For the first time , using sequence analysis of the cytochrome b gene , the occurrence of CL in the borderline area between northern and central Iraq was confirmed to be due to Leishmania major . The phylogenetic analysis indicated that it was closely related to the L . major MRHO/IR/75/ER strain in Iran . In conclusion , the genotype confirmation of the L . major strain will improve our understanding of the epidemiology of the disease . This is important for facilitating control programs to prevent the further spread of CL . Furthermore , this area could be considered as a model for further research on the risk of global CL epidemics in other non-endemic countries where both reservoir hosts and sandfly vectors are present .
Leishmaniasis is considered to be a neglected tropical and zoonotic disease that spreads via phlebotomine sandfly vectors [1] . Leishmaniasis is a parasitic disease caused by intracellular protozoa which in humans has four clinical forms including cutaneous ( CL ) , diffuse cutaneous ( DCL ) , visceral ( VL ) and mucocutaneous ( MCL ) leishmaniasis and it is endemic in different parts of the world [2] . The morbidity associated with human CL is up to 1 . 2 million cases distributed worldwide resulting in extensive integumentary lesions [3] . There are two groups of CL , New World and Old World leishmaniasis , with only the latter group identified in the Middle East and it includes three main species; L . major , L . tropica and L . infantum [4] . Recent studies showed a high prevalence of CL in Iran [5 , 6] , Turkey and Syria [7] . Although Iraq shares long borders with these countries and leishmaniasis is endemic , the World Health Organization has not classified it as a country with a high burden profile [8] . In Iraq , several studies have been performed to diagnose Leishmania parasites from skin lesions of human patients by using different methods including histopathological examinations , direct smears , cultures and serological tests [9 , 10] . Few studies have been conducted to exploit PCR in the characterization of the Leishmania strains in human cutaneous lesions [11] and VL-suspected patients [12] in central Iraq . Studies have been performed without conducting gene sequencing or phylogenetic analyses . However , in a US military base in Southern Iraq , a phylogenetic study investigated the prevalence of different Leishmania species in sandflies using molecular study and phylogenetic analysis [13] . Therefore , the aim of this study was to identify the genotype of the most prevalent CL strains in the region using cytochrome b gene amplification by PCR and sequencing .
An outbreak of leishmaniasis was clinically suspected for the first time in 2013 in areas belonging to the Kifri district in the Garmian Administration . The term ( Garmian ) is a Kurdish word which is used to denote a ‘hot and dry area’ indicating information about location and climate . The Garmian area is located in the southeast Kurdistan region of Iraq . It is in between the latitudes ( 34°15–33 = - 35° 11–05 = ) above the equator and the longitudes ( 44° 29–41 = - 45° 54–20 = ) of the eastern hemisphere . The Garmian includes the districts: Kalar , Kifri , and Khanaqin , and its total area is 6731 . 73 square kilometers . According to the official site of the general board of tourism of Kurdistan- Iraq in 2015 , the total population of the central town of Garmian , Kalar , is about 250 , 000 residents [14] . In this region , there is an increasing concern about the cutaneous form of leishmaniasis which is publicly known as “Baghdad sore” . Since 2014 leishmaniasis has been considered a notifiable disease and every new case with a clinical manifestation of cutaneous lesions of leishmaniasis should be recorded officially by local authority officers as a transmissible disease before the patient receives treatment . Northern and central Iraq have undergone economic and humanitarian crises due to the Iraqi civil war since 2014 . Moreover , the topography of both territories is different . In addition , there has been only one updated map up to 2008 based on the last report of CL incidence in Iraq by the WHO [3] . Thus , in this study , a spot map of CL cases was updated in the borderline region using Landsatlook viewer ( USGS Products , Data available from the U . S . Geological Survey ) . The map ( Fig 1a and 1b ) shows the outbreak of CL from an endemic area in the Kurdistan Region of Iraq ( KRI ) including Diyala province to a non-endemic area inside the KRI including the Garmian administrative region . Data of cutaneous leishmaniasis ( CL ) in Iraq were collected from the WHO website [8 , 15 , 16] , and a line graph showing annual numbers of CL cases from 1989–2015 was plotted using GraphPad Prism version 6 . 06 for Windows ( GraphPad Software , La Jolla California USA , www . graphpad . com ) . Furthermore , in the area of the study ( Garmian administration , Kurdistan Region , Iraq ) , new cases were referred by local health general practitioners to visit Kalar General Hospital in Garmian , Sulaimaniyah province to be clinically examined by dermatologists and receive Pentostam injection treatment ( sodium stibogluconate ) . The data of clinically examined patients collected by the Department of Transmissible Diseases in Garmian from 2014–2017 were also analyzed . Thirty samples were collected from lesions of new clinically suspected CL cases or from patients receiving early treatment in February , March and April 2017 . The sample collection was performed by cleaning the skin lesions with cotton soaked in 70% ethyl alcohol and left to dry . This was followed by injecting 0 . 1 ml sterile normal saline into the active borders of the skin lesions using a 25-gauge insulin needle and then aspirating the fluid into sterile 1 . 5 ml tubes . The samples were directly preserved in 0 . 4 ml absolute ethanol , labeled and stored at room temperature for molecular study . Later on , the samples were submitted to the molecular laboratories of the University of Garmian which is based in the Kalar district . Clinical samples were collected from patients who agreed to participate in this study and signed an informed consent form . The study was also approved by the Ethical Committee of the Department of Biology , College of Education , University of Garmian with permit number ( 85 , 18/04/2017 ) . After receiving permission from the General Directorate of Garmian Health ( permit number 1550 , 10/05/2017 ) , the samples were transported to the molecular biology lab of Garmian University . A pair of primers including Leishmania cytochrome b forward ( LCBF ) : GGTGTAGGTTTTAGTTTAGG , Leishmania cytochrome b reverse ( LCBR ) : CTACAATATACAAATCATAATATACAATT ( Macrogen Co . , Seoul , KR ) were exploited for amplification of the Leishmania cytochrome b gene with a product size of 866 bp as previously used for identification of almost all species of Leishmania by PCR and DNA sequencing [17] . Total genomic DNA of the ethanol-preserved samples was extracted by a PrimePrep Genomic DNA Extraction Kit ( from tissue ) . The ethanol was removed from the samples by using centrifugation and washing with normal saline . The pellets were mixed with 200 μl tissue lysis buffer ( TL buffer ) and 20 μl proteinase K and incubated at 56°C for approximately an hour until the samples were lysed . According to the manufacturer’s instructions , DNA was isolated using ethanol and buffers then eluted with 200 μl elution buffer ( TE ) provided by the company ( GENET BIO CO . , Daejeon , KR ) . Conventional PCR was performed individually for each sample in 20 μl reactions containing 1x Prime Taq premix ( 2x ) which contains Prime Taq DNA Polymerase 1 unit , 2x reaction buffer , 4 mM MgCl2 , enzyme stabilizer , sediment , loading dye , pH 9 . 0 and 0 . 5 mM each of dATP , dCTP , dGTP , dTTP and 0 . 5 μM final concentration from each of the LCBF and LCBR primers . The PCR reaction conditions were 94°C for 3 min; 40x at 94°C for 1 min , 60°C for 1 min , 72°C for 2 min; 72°C for 5 min using a thermal cycler ( Mastercycler nexus , Eppendorf AG , Hamburg , Germany ) . PCR products were run at 110 V for 50 min on a 1 . 5% agarose gel in 1x TBE ( 87 . 5 mM Tris base , 89 mM boric acid , 3 mM EDTA ) and stained with Prime safe dye ( GENET BIO CO . , Daejeon , KR ) . A total of 5 μl of PCR products from nine positive samples and 5 μl ( 5 pmoles ) of forward or reverse primers for forward or reverse sequencing , respectively , were sequenced using the Sanger method ( Macrogen Co . , Seoul , KR ) and edited by CodonCode Aligner ( CodonCode Corporation , 101 Victoria Street , Centerville , MA 02632 ) . To the best of our knowledge , no phylogenetic data were found exploring CL strains in the area along the borderline between the northern region ( Kurdistan Region ) and the middle part of the country based on the NCBI search engine [18] on 28/06/2017 using these keywords ( Leishmaniasis Iraq PCR ) . Seven high-quality sequences were submitted to the NCBI GenBank using Bankit [18] . The high-quality sequences were determined based on having high single peaks using CodonCode Aligner . The cytochrome b gene sequences together with those from representative strains were aligned with CLUSTAL W software and examined using the program MEGA ( Molecular Evolutionary Genetics Analysis ) version 7 . Phylogenetic trees were constructed by the neighbor-joining method with the distance algorithms available in MEGA version 7 . The distances were calculated using the Kimura 2-parameter method . Bootstrap values were determined with 1 , 000 replicates of the data sets [19] .
The CL outbreak occurred in areas between northern and central Iraq ( Fig 1a ) . The data showed that cases of CL seemed to have originated from endemic areas like Diyala , Saladin , Mosul , and Kirkuk provinces . This outbreak may have been caused by infected people traveling from the center to the north of Iraq . It is worth mentioning that before toppling down the previous government of Iraq , i . e . , before 2003 , people from the southern and central provinces rarely visited Kurdistan due to strictly controlled borders ( As shown in Fig 1b , interrupted blue line ) . After 2003 , the situation changed and there was free public access into the Kurdistan Region of Iraq ( KRI ) . Furthermore , since 2014 , people have moved from the war zones of the central provinces to the Garmian region . Thus , this area can be regarded as a typical model region to understand the risk of leishmaniasis spread from endemic to non-endemic areas . As shown on the map ( Fig 1b ) , CL spread from endemic areas of the KRI borderline such as Jabara , Kokiz , Qara Tapa , Khanaqin , Jalawla and Tuz Khurmatu to non-endemic areas including Kalar , Kifri , and Rizgari districts was due to the arrival of refugees . This may have led to spread of the disease in other parts of Kurdistan including Sulaimaniyah , Irbil , and Duhok . Even so , temperature and humidity could be related to the distribution of the vectors and reservoirs [20] . Therefore , further study on the sandfly and animal reservoirs from both the endemic and non-endemic areas will help uncover the epidemiology of the disease . According to Alvar et al . [3] , the WHO reported 1655 CL cases/ year from 2004–2008 in Iraq; the annual incidence rate was estimated from 8300 to 16 , 500 cases , although this number seems to be underestimated by the WHO . This could be due to a lack of diagnostic services , so the disease was not regarded as a major public health concern [21] . In addition , it is worth mentioning that not all cases were reported by the authorities for the following reasons: firstly , some infected people objected to receiving treatment due to painful intralesional injections . Secondly , some patients did not wish to visit hospitals since they use traditional medicine for treatment or believe that the lesions are self-curable . Finally , some places are far from health centers . Therefore , these data should be updated in Iraq and this will be important before introducing any CL control programs . In the current study , the data show the prevalence of CL in Iraq from 1989–2015 ( Fig 2 ) . As the country went through several wars , internal conflicts , economic crises , and sanctions over the previous 27 years , massive fluctuations in the number of reported CL cases can be noticed . There was a sharp increase in the number of cases after the second gulf war in 1990 . This trend remained high until 1997 , which could be due to discontinuation of control programs and lack of healthcare services due to the economic blockade . The number of cases remained low from 1998 until 2003 , possibly due to WHO interventions [15] and control programs or relative economic growth after the removal of the sanctions in 1997 . Again from 2009 the number of cases increased; however , it remained relatively low until 2014 , then a sharp rise was recorded with a peak in 2015 . The last increase in the number of CL cases could be due to the civil war in Iraq starting in 2014 as the war led to displacement of millions of people , especially from endemic areas to non-endemic areas , as well as a deterioration in health services . After the introduction of malaria control programs in Iraq , the number of CL cases decreased until their discontinuation in the mid-1960s [15] . Afterwards , massive fluctuations in the number of reported cases were noticed . The number of CL cases could be related to certain factors . One of the main factors is population displacement , which brings non-immune people to endemic areas and infected people to non-endemic areas . In addition , increased contact with reservoir animals and sandfly vectors , untreated patients , malnutrition , poor sanitation and environmental changes are other possible reasons for the increase in CL cases . Most of the districts of the Garmian region belong to the Sulaimaniyah governorate which is regarded as one of the non-endemic areas for CL [22] . However , after the start of the war from mid-2014 , large-scale emigration of people occurred in Iraq . Garmian , as one of the border regions , housed a large number of refugees from other parts of conflict zones in Iraq; especially people from the endemic areas of Diyala , Kirkuk , Saladin and Mosul arrived in the region . This may have changed the region from a non-endemic to endemic area . This argument is supported by the large increase in the number of CL cases over the last 4 years in the region as shown in ( Fig 3 ) . Further study is required to confirm whether the region has become endemic by recording new cases who have not visited any endemic areas . Regarding the monthly prevalence of CL in Garmian , the number of recorded CL cases started to increase from November ( Fig 4 ) . The maximum number of CL cases was recorded in January and February . The recorded numbers decreased from March and remained low until October . These findings agreed with those reported for other parts of Iraq [10] . As the incubation period of the disease ranged from two to four months , the majority of the cases recorded during winter months were probably bitten by insects from the summer to early autumn seasons . Of the thirty samples collected from suspected new cases and CL patients receiving early treatment , 15 samples were positive for PCR targeting the leishmanial cytochrome b gene on gel electrophoresis , showing single bands with a product size of about 850–900 bp ( Fig 5 ) . Of the 15 positive samples , sequences of seven samples showing strong signals were determined , and all the Leishmania parasites were identified as L . major with 100% similarity with the L . major strain MRHO/IR/75/ER cytochrome b gene ( GenBank accession number KU680828 ) [23] . This is the first molecular record of the L . major strain in Iraq using sequence analysis . Nonetheless , confirming only 7 cases out of a total 30 samples should be considered as a limitation of direct molecular tools for investigation of the L . major strain epidemiologically . Nucleotide sequence data reported will appear in the GenBank database under the accession numbers MF-370217-MF-370223 . The cytochrome b gene sequences obtained in this study were subjected to phylogenetic analysis together with those from representative Leishmania strains . The phylogenetic data indicated that Leishmania strains in the borderline area between northern and central Iraq were closely related to the Iranian MRHO/IR/75/ER strain and other L . major strains of Old World CL ( Fig 6 ) . This result is not surprising as the region shares its border with Iran which is the nearest country to the endemic area of central Iraq . Particularly , Diyala province has had a commercial relationship with Iran through the border close to the Khanaqin district since 2003 . Microsatellite analyses using specimens from geographically isolated areas , Central Asia , the Middle East and Africa showed L . major has little genetic variation when compared to L . tropica [25 , 26] . This may partly explain the slight variation in L . major identified in this study . Further large-scale genetic analysis using more sensitive methods such as microsatellite analysis will be necessary in the future . The exact origin of the parasite is unknown although there has been a history of the disease in Iraq [27] . In addition , there has been a long history of pilgrims visiting from Iran to sacred shrines in central and southern Iraq via either the Garmian region or central provinces . The reservoirs of L . major have not been identified in Iraq , but in Iran , four gerbil species were identified as main reservoirs including Rhombomys opimus , Meriones libycus , Meriones hurrianae and Tatera indica [28] . Nevertheless , we recently identified the L . major MRHO/IR/75/ER strain from cutaneous lesions of a dog in the study area [24] . Further studies will reveal the transmission vectors and reservoirs and aid in control of the outbreaks . To our knowledge , this is the first record of a phylogenetic study in Iraq concerning L . major causing CL prevalence . The findings are also significant for future creation of vaccines against the Leishmania strain in Iraq and it is important to understand the global prevalence and epidemiology of the Leishmania strains . In conclusion , we identified the L . major strain in Iraq for the first time using PCR and DNA sequencing . In addition , phylogenetic study revealed the main Leishmania genotype causing health problems in the borderline area between the non- endemic area in the north and the endemic region of central Iraq . The identified parasite was similar to the MRHO/IR/75/ER strain which is endemic in Iran . Furthermore , we reported new CL cases in the borderline area between central and northern Iraq , particularly in the Garmian region . This region can be regarded as a model for further study of epidemic CL outbreaks to other non-endemic areas . This study also suggests that researchers conduct more studies regarding the threat of CL which may spread globally to countries where both reservoirs and sandflies are present . | Leishmaniasis refers to a disease with three main types of clinical manifestation in infected individuals including cutaneous , mucocutaneous and visceral forms . It is caused by several species of a parasite belonging to the genus Leishmania and is transmitted by a small blood-sucking insect called a sandfly . The disease is mostly confined to the majority of the poorest countries worldwide , including Iraq , and is categorized as a low priority public health concern . The risk of the disease is exacerbated especially when suitable environments assist the sandfly and reservoir host to breed and spread and help the parasite to transfer from high incidence areas to places free from the disease . Therefore , we investigated the risk of the CL form of the disease after an outbreak in a borderline between northern and central of Iraq using the most sensitive diagnostic techniques including PCR and gene sequencing . The epidemiological , molecular and phylogenetic analyses of the parasites were studied , and we found that the parasite species Leishmania major was associated with the outbreak . Phylogeny analysis confirmed that the identified strain of the parasite matched an Iranian strain . These results indicate the risk of the disease spreading from endemic to non-endemic areas . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"discussion"
] | [
"taxonomy",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"geographical",
"locations",
"tropical",
"diseases",
"sand",
"flies",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"phylogenetics",
"data",
"management",
"protozoans",
"signs",
"and",
"symptoms",
"leishmania",
"phylogenetic",
"analysis",
"neglected",
"tropical",
"diseases",
"molecular",
"biology",
"techniques",
"insect",
"vectors",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"infectious",
"diseases",
"computer",
"and",
"information",
"sciences",
"zoonoses",
"artificial",
"gene",
"amplification",
"and",
"extension",
"bioinformatics",
"lesions",
"protozoan",
"infections",
"evolutionary",
"systematics",
"molecular",
"biology",
"disease",
"vectors",
"people",
"and",
"places",
"eukaryota",
"diagnostic",
"medicine",
"asia",
"polymerase",
"chain",
"reaction",
"database",
"and",
"informatics",
"methods",
"leishmaniasis",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"evolutionary",
"biology",
"iraq",
"organisms"
] | 2018 | An outbreak of Leishmania major from an endemic to a non-endemic region posed a public health threat in Iraq from 2014-2017: Epidemiological, molecular and phylogenetic studies |
Although Mycobacterium leprae ( M . leprae ) is usually found in macrophages and nerves of the dermis of patients with multibacillary leprosy , it is also present in all layers of the epidermis , basal , suprabasal , prickle cells , and keratin layers . However , the mechanism by which M . leprae invades the dermis remains unknown , whereas the underlying mechanism by which M . leprae invades peripheral nerves , especially Schwann cells , is well defined . M . leprae binds to the α-dystroglycan ( DG ) of Schwann cells via the interaction of α-DG and laminin ( LN ) -α2 in the basal lamina , thus permitting it to become attached to and invade peripheral nerves . In the current study , we investigated the issue of how M . leprae is phagocytosed by human epidermal keratinocytes , neonatal ( HEKn ) . LN-5 is the predominant form of laminin in the epidermis and allows the epidermis to be stably attached to the dermis via its interaction with α/β-DG as well as integrins that are produced by keratinocytes . We therefore focused on the role of LN-5 when M . leprae is internalized by HEKn cells . Our results show that M . leprae preferentially binds to LN-5-coated slides and this binding to LN-5 enhances its binding to HEKn cells . The findings also show that pre-treatment with an antibody against α-DG , integrin-β1 , or -β4 inhibited the binding of LN-5-coated M . leprae to HEKn cells . These results suggest that M . leprae binds to keratinocytes by taking advantage of the interaction of LN-5 in the basal lamina of the epidermis and a surface receptor of keratinocytes , such as α-DG , integrin-β1 , or -β4 .
Leprosy , Hansen’s disease , is a chronic granulomatous disease caused by the intracellular bacterium Mycobacterium leprae ( M . leprae ) . It mainly affects both the skin and peripheral nerves , resulting in the development of skin lesions , such as macules , plaques or nodules , and peripheral neuropathy [1] . M . leprae is usually found in macrophages and nerves of the dermal zone in patients with multibacillary leprosy [2] . In addition to the dermis , M . leprae can also be detected in the epidermis , sweat glands and hair follicles of patients with high bacteriological index ( BI>4+ ) multibacillary leprosy [3] . Although leprologists generally believe that M . leprae is transmitted through the respiratory tract , compared to the skin route , Job et al . [4] reported that M . leprae was also present in the superficial keratin layer of the skin of lepromatous leprosy patients , suggesting that M . leprae may be transmitted from the intact skin of patients with lepromatous leprosy . It has been suggested that M . leprae is transmitted to the epidermis from rapidly growing granuloma in the upper dermis of patients with lepromatous leprosy [5] . Moreover , Lyrio et al . [6] reported that HaCaT , a human keratinocyte cell line , phagocytoses M . leprae and the M . leprae-phagocytosed keratinocytes produce cathelicidin , an antimicrobial peptide , as well as tumor necrosis factor ( TNF ) -α . However , the mechanism responsible for the epidermis invasion by M . leprae is not known with certainty , whereas the underlying mechanism by which M . leprae invades peripheral nerves , especially Schwann cells , is well defined . M . leprae invades Schwann cells by binding to the alpha ( α ) -dystroglycan ( DG ) of Schwann cells via the interaction of α-DG and laminin ( LN ) –α2 in the basal lamina that surrounds the Schwann cell-axon unit [7] . The DG complex in Schwann cells consists of α-DG and β-DG . α-DG serves as a receptor on the Schwann cell that interacts with extracellular LN-α2 , and β-DG serves as a links between the extracellular matrix ( ECM ) and the intracellular cytoskeleton [8 , 9] . The basement membrane ( BM ) surrounding Schwann cells is composed of LNs , collagen IV , and proteoglycans [10] . LN-2 ( α2 , β1 , γ1 chains ) is the most common form of laminin in the basal lamina that surrounds Schwann cell-axon unit [11] . It has been reported that M . leprae simultaneously binds to the globular domain of LN-α2 and α-DG , a surface receptor , of Schwann cells , indicating that LN-α2 mediates the attachment and invasion of M . leprae to peripheral nerve cells [12] . Thus , we hypothesized that M . leprae uses components of the ECM , which is bound to a cell surface receptor , for the invasion of keratinocytes , as shown in Schwann cells . LN-5 ( α3 , β3 , γ2 chains ) is a major component of the basal lamina between the epidermis and dermis , and mediates the stable attachment of the epidermis to the dermis via the formation of hemidesmosomes [13] . Keratinocytes bind to LN-5 , collagen , and fibronectin via integrins including α2β1 , α3β1 and α6β4 [14 , 15] . In addition , α/β-DG is also expressed in keratinocytes that are present in all epidermal layers except for the corneal layer [16] . In the current study , we investigated the issue of how M . leprae is phagocytosed by human epidermal keratinocytes , neonatal ( HEKn ) . Our results show that M . leprae preferentially binds to LN-5 and that coating M . leprae with LN-5 enhanced its binding to HEKn cells . Our results also show that a pre-treatment with an antibody against α-DG , integrin-β1 , or -β4 inhibited the binding of LN-5-coated M . leprae to HEKn cells , suggesting that the binding of M . leprae to keratinocytes is assisted by the interaction of LN-5 in the basal lamina of the epidermis and a keratinocyte surface receptor , such as α-DG , integrin-β1 , or -β4 .
All procedures related to animal research were conducted in accordance with the Laboratory Animals Welfare Act , the Guide for the Care and Use of Laboratory Animals and the Guidelines and Policies for Rodent experiment provided by the IACUC ( Institutional Animal Care and Use Committee ) in school of medicine , the Catholic University of Korea ( Approval number: CUMC-2017-0091-02 ) . Human skin samples were obtained from patients who had upper lid blepharoplasties with no clinical evidence of inflammatory or immune diseases . These activities were undertaken after written informed consent was obtained from the donors , according to procedures approved by the Institutional Review Board of Seoul St . Mary’s Hospital ( KC10TISE0743 ) and the tenets of the Declaration of Helsinki . Auramine O , H2O2 , DAPI , Collagen IV and Fibronectin were obtained from Sigma-Aldrich ( St . Louis , MO ) . Laminin-α2 ( LN-α2 , LN211-02 ) and laminin-5 ( LN-5 , ab42326 ) proteins were obtained from BioLamina ( Matawan , NJ ) and Abcam ( Cambridge , MA ) , respectively . Antibodies against LN-5 ( ab102539 for immunohistochemistry ) , integrin-β1 ( ab24693 for immunocytochemistry and binding assay ) and -β4 ( ab133682 for immunocytochemistry and binding assay ) were obtained from Abcam ( Cambridge , MA ) . Antibodies against LN-α2 ( sc-55605 for immunohistochemistry ) , α-dystroglycan ( α-DG , sc-53987 for immunocytochemistry and binding assays ) , integrin-β2 ( sc-13548 for binding assays ) and–β3 ( sc-52589 for binding assays ) were obtained from Santa Cruz Biotechnology ( Santa Cruz , CA ) . CyTM5-conjugated secondary antibody and horseradish peroxidase-conjugated secondary antibody were obtained from Jackson ImmunoResearch ( West Grove , PA ) . We have cultivated M . leprae by using the mouse foot-pad technique [17] . M . leprae Thai-53 was donated by Dr . Kenji Kohsaka , Sasakawa Research Center , Soi Bamrasnaradoon Hospital , Thailand . BALB/c nude mice were inoculated in the hind foot pad with 3x105 M . leprae . BALB/c nude mice were obtained from Orient Bio ( Seong Nam , Gyunggi-do , Korea ) and were maintained under specific pathogen-free conditions in the Department of Laboratory Animals , the Catholic University of Korea . Standard mouse chow ( Ralston Purina , St Louis , MO ) and water were provided ad libitum . At 6 months after M . leprae inoculation , edematous changes in the M . leprae-inoculated hind foot were visually detected . The foot-pads of M . leprae-infected BALB/c nude mice were treated with a potadine solution and washed with ice-cold Dulbecco’s phosphate-buffered saline ( DPBS , Sigma-Aldrich Co . Ltd , MO ) to remove exogenous contamination . The foot-pads were excised , cut into small pieces , and homogenized with a MACs isolator ( Miltenyl Biotec , Teterow , Germany ) . The extract was filtered using a cell strainer ( BD Falcon , Durham , NC ) to remove tissue debris and the resulting solution was then centrifuged at 3 , 000 rpm ( Rotanta 460R , Hettich , Japan ) for 25 min at 4°C . The pellet was resuspended in 1 ml of ice-cold DPBS and treated with 2 N sodium hydroxide for 5 min . The reaction mixture was neutralized by adding 13 ml of ice-cold DPBS ( Sigma-Aldrich Co . Ltd , MO ) . After centrifugation and resuspension , acid-fast bacillus ( AFB ) staining was performed and the numbers of bacteria were counted by light microscopy under an oil immersion field using a procedure established by Shepard and McRae [18] . Human primary epidermal keratinocytes from neonatal foreskin ( HEKn ) cells were acquired from Invitrogen ( Carlsbad , CA ) and grown in EpiLife medium supplemented with 100 U/ml of penicillin , 100 mg/ml of streptomycin , 250 ng/ml of amphotericin B , 60 μM of calcium , and Human keratinocyte growth supplement ( HKGS , Cascade Biologics; Invitrogen , Carlsbad , CA ) . In the current study , to limit the differentiation of HEKn cells , we maintained HEKn cells in EpiLife medium supplemented with human keratinocyte growth supplement ( HKGS , Cascade Biologics; Invitrogen , Carlsbad , CA ) , and not in fetal bovine serum . These cells were maintained in a state of proliferation and non-differentiation . The cells were passaged with a gentle TrypLE select ( Invitrogen , Carlsbad , CA ) treatment followed by a trypsin neutralization solution ( Invitrogen , Carlsbad , CA ) . The cells were plated on 6-well plates at 1 x 105 cells/well or onto 4-channel chamber slides ( Lab-Tek II chamber slide , Thermo Fisher Scientific , Waltham , MA ) at 5 x 104 cells/well and were grown until reaching 70% confluence . The HEKn cells were cultured on coverslide in a 6-well plate . M . leprae was pre-incubated with LN-α2 ( 10 μg/ml ) or LN-5 ( 2 μg/ml ) in DPBS for 2 h at 37°C , followed by washing . The cells were incubated with M . leprae at multiplicity of infection ( MOI ) of 10:1 , 20:1 , 50:1 and 100:1 for 1 h at 37°C . After removing extracellular M . leprae by washing with phosphate-buffered saline ( PBS ) , M . leprae were stained with the AFB stain or Auramine O , and examined in an oil immersion field of a light microscopy or fluorescence microscopy . Immunofluorescent staining of paraffin-embedded skin tissues and cells was performed using standard methods with the minor modifications [19 , 20] . Skin samples were fixed in 4% formaldehyde for 4 h at room temperature prior to being embedded in paraffin and 4 μm thick sections were dewaxed and rehydrated in a series of graded alcohol solutions . The sections were incubated in 0 . 3% sodium citrate buffer ( pH 6 . 0 ) for 10 min at 100°C and 3% hydrogen peroxide ( H2O2 ) for 10 min after which , they were rinsed with PBS and incubated in blocking solution [5% goat serum and 0 . 001% Tween-20 in tris-buffered saline ( TBS ) ] for 20 min . The sections were then incubated overnight with an antibody against LN-α2 or LN-5 in an incubation solution ( 5% goat serum and 0 . 1% Tween-20 in TBS ) at 4°C . After washing with PBS , the sections were incubated with a mouse CyTM5- or a rabbit CyTM5-conjugated secondary antibody at room temperature for 2 h . HEKn cells were fixed in 4% paraformaldehyde in PBS . The resulting fixed cells were then rinsed with PBS and incubated in blocking solution ( 5% goat serum and 0 . 001% Tween-20 in TBS ) for 20 min . The cells were then incubated overnight with an antibody against α-DG , integrin-β1 , or -β4 in an incubation solution ( 5% goat serum and 0 . 1% Tween-20 in TBS ) at 4°C . After washing with PBS , the cells were incubated with a mouse CyTM5- or a rabbit CyTM5-conjugated secondary antibody at room temperature for 2 h . After staining with the secondary antibody , the tissue sections and cells were washed with PBS . Nuclei were counterstained for 5 min with DAPI ( Sigma-Aldrich Co . Ltd , MO ) . The negative control was processed in the absence of the primary antibody . Immunofluorescence was visualized by confocal microscopy ( LSM 510 Meta , Zeiss , Germany ) . In the assay for the binding of M . leprae to the ECM-coated culture plate , 4-channel chamber slides were coated , as described in a previous report [21] . The slides were coated with 0 . 1 μg/ml of LNs , type IV collagen , or fibronectin at room temperature overnight . Saline was used as a negative control . Nonspecific binding was blocked with 5% bovine serum albumin ( BSA ) for 3 h at 37°C and the sample was then washed 5 times with DPBS . Ten microliters of a suspension of M . leprae ( 5 x 108 bacteria/ml ) was added to each well , followed by incubation for 1 h at 37°C . Unbound bacteria were removed by washing 5 times with DPBS . After fixation with 2% paraformaldehyde for 10 min , the bacteria were stained with Auramine O . The level of Auramine O-labeled M . leprae that was bound to slide was determined using the ZEN program ( Zeiss , Oberkochen , Germany ) under a LSM 510 Meta confocal microscopy ( Zeiss , Oberkochen , Germany ) . For assaying the binding of M . leprae to HEKn cells , the HEKn cells were cultured in 4-channel chamber slides and incubated overnight at 37°C under 5% CO2 . For determining the M . leprae that was bound to HEKn cells , M . leprae was pre-incubated with 10 μg/ml LN-α2 or 2 μg/ml LN-5 for 2 h at 37°C before inoculation at MOI of 10:1 , 20:1 , 50:1 and 100:1 . For the binding inhibition assay , HEKn cells were pre-incubated with an antibody against α-DG , integrin-β1 , -β2 , -β3 or–β4 for 2 h at 37°C before inoculation with M . leprae at an MOI of 100:1 . After incubating the HEKn cells with M . leprae for 1 h at 37°C in 5% CO2 , extracellular M . leprae were removed by washing 5 times with PBS and fixing in 2% paraformaldehyde for 30 min . M . leprae were labeled with the AFB stain and examined in the oil immersion field of a light microscopy .
We initially investigated the issue of whether M . leprae is phagocytosed by HEKn cells . HEKn cells were incubated with M . leprae at MOI of 10:1 , 20:1 , 50:1 and 100:1 , respectively , for 6 h at 37°C . At an MOI of 100:1 , the percentage of M . leprae-phagocytosed cells was 77 . 4% and the average number of M . leprae per cell was determined to be 3 ( Fig 1 ) . We examined the expression pattern of LN-α2 and LN-5 in human skin . Consistent with previous reports [13] , LN-5 , but not LN-α2 , was expressed in the basal lamina between the epidermis and dermis ( Fig 2 ) . We then examined the expression patterns of cell surface receptors in HEKn cells . As shown in Fig 3 , HEKn cells expressed α-DG , integrin-β1 and -β4 on the cell surface . We then investigated the issue of whether M . leprae adheres to the immobilized extracellular matrix LN-5 , collagen IV and fibronectin using a solid-phase bacterial-adherence assay . We used LN-α2 as a positive control since LN-α2 in Schwann cells basal lamina is known to be the primary target molecule for M . leprae [21] . The level of M . leprae binding was increased in the LN-α2- as well as the LN-5-coated slides , compared to collagen IV- and fibronectin-coated slides ( Fig 4 ) . We also examined the binding ability of LN-α2- or LN-5-coated M . leprae to HEKn cells . As shown in Fig 5 , the coating of M . leprae with LN-α2 or LN-5 resulted in an increase in the number of M . leprae that had adhered HEKn cells ( average number of adherent M . leprae to HEKn cells per 100 HEKn cells; 69 . 3±5 . 7 in LN-α2-coated M . leprae and 44 . 0±2 . 4 in LN-5-coated M . leprae in comparison with 35 . 0±3 . 2 in non-treated M . leprae ) . Rambukkana et al . [7] reported that when M . lerpae , that had been coated with the recombinant globular domain of LN-α2 ( LN-α2G ) , were pre-incubated with recombinant α-DG , the LN-α2G/α-DG-mediated M . leprae binding to rat Schwann cells was competitively inhibited , suggesting the existence of a linkage between LN-α2 and α-DG in the interaction of M . leprae with Schwann cells . In the current study , although LN-α2 is not expressed in skin , we employed the LN-α2/α-DG-mediated M . leprae binding to cells as a positive control in the binding assay . Consistent with Rambukkana et al . ’s results [7] , our result also showed that the pre-treatment of HEKn cells with an anti-α-DG antibody inhibited the binding of LN-α2-coated M . leprae to HEKn cells ( Fig 6A ) . In addition , the pre-treatment of HEKn cells with antibody against α-DG , integrin-β1 , or -β4 , all of which are expressed on the surface of HEKn cells ( Fig 3 ) , inhibited LN-5-coated M . leprae from binding to HEKn cells ( Fig 6B and 6C ) . However , pre-treatment with antibody against integrin-β2 or -β3 had no effect on inhibiting the binding of LN-5-coated M . leprae to HEKn cells ( Fig 6B ) . These results suggest that M . leprae binds to keratinocytes by taking advantage of the interaction of LN-5 in the basal lamina of the epidermis and a surface receptor of keratinocytes , such as α-DG , integrin-β1 , or -β4 .
ECM is an acellular proteinaceous fraction of the tissues . ECM proteins consist of collagen , elastin , fibrillin , LNs , fibronectin , vitronectin , thrombospondin , proteoglycans and hyaluronic acid . ECM is involved in the structural support of tissues as well as various cellular signaling processes , including cell adhesion , migration , growth , and differentiation [22] . Although pathogens need to breach and degrade ECM proteins in order to successfully invade a tissue , they also utilize ECM proteins to aid in their adhesion to host tissues . LNs and collagens are major target glycoproteins of various pathogens , such as bacteria , fungi , and viruses , for adhesion to cells of host tissue [23] . LNs are heterotrimeric glycoproteins that consist of α , β and γ chain . The chains , α , β and γ , which are connected to one another via disulfide bonds at their C-terminal regions , form a triple coiled-coil region , resulting in a ‘crucifix’-shaped structure [23] . There are currently five α chain , three β chain and three γ chain isoforms and 16 LN isoforms have been identified in humans [24] . LN isoforms are differentially distributed in human tissues or cells [23] . LN-2 ( α2 , β1 , γ1 chains ) is a predominant laminin associated with Schwann cells [11]; LN-5 ( α3 , β3 , γ2 chains ) is found in oral , intestinal and dermal epithelial cells [13 , 25 , 26]; LN-10/11 is expressed in the lung epithelium [27] . The interaction between ECM laminins and integrins of epithelial cells confers mechanical stability to tissues as well as an invasive mechanism for pathogens [23] . It has been reported that M . leprae binds to the globular domains ( LG1 , LG4 , and LG5 domains ) of LN-α2 chain and that the LN-α2 chain simultaneously binds to α-DG , a surface receptor , of Schwann cells , resulting in the attachment and invasion of M . leprae to Schwann cells [12] . Our results also show that coating M . leprae with LN-α2 enhanced the binding of M . leprae to HEKn cells ( Fig 5A ) and a pre-treatment with an antibody against α-DG inhibited the binding of LN-α2-coated M . leprae to HEKn cells ( Fig 6A ) . However , although LN-α2 ( α2 , β1 , γ1 chains ) mediates the attachment of M . leprae to HEKn cells , it was not detected in the skin ( Fig 2 ) , whereas LN-5 ( α3 , β3 , γ2 chains ) is a major form of laminins that is present between the epidermis and dermis [13] . Thus , in the current study , we focused on the role of LN-5 in the interaction of M . leprae with keratinocytes . It has been reported that LN-5 , which is expressed in the BM between the epidermis and dermis , has been reported to be a target molecule and mediator for the invasion of the Human papilloma virus ( HPV ) to keratinocytes [20] . HPV first infects keratinocytes in the basal layer of the epidermis and then replicates in a fully differentiating squamous epithelium [28] . Culp et al . [20] reported that the HPV capsid binds to LN-5 in the ECM of cultured keratinocytes . In that report , the authors reported that , when sections of cervical mucosa tissues were incubated with HPV , the HPV became bound to the suprabasal layer and BM of the cervical mucosa and that a pre-treatment with anti-LN-5 antibody blocked the binding of HPV to these cervical mucosa tissue sections . Our results also show that M . leprae preferentially bound to LN-5-coated slides , compared to collagen IV and fibronectin ( Fig 4 ) and that coating M . leprae with LN-5 enhanced the binding of M . leprae to HEKn cells ( Fig 5B ) , suggesting LN-5 mediates the attachment of M . leprae to HEKn cells . Although M . leprae can be detected in the all layers of the skin , it is more frequently detected in the suprabasal and basal layers of the epidermis of patients with multibacillary leprosy [3 , 29] . We conclude that the clinical findings support the conclusion that LN-5 in the BM of the epidermis mediates the attachment and invasion of M . leprae to non-differentiated , proliferating keratinocytes in the basal layer . It is well known that α-DG serves as a Schwann cell receptor for the LN-α2-mediated M . leprae invasion of Schwann cells [7] . In the skin , α/β-DG is present in the epidermal BM [30] . Thus , we hypothesized that α-DG is also involved in the LN-5-mediated M . leprae interaction with keratinocytes , as shown in the LN-2α-mediated M . leprae invasion of Schwann cells . As shown in Fig 6B and 6C , our results show that pre-treatment with an anti-α-DG antibody blocked the binding of LN-5-coated M . leprae to HEKn cells . LN-5 permits the stable attachment of the epidermis to the dermis via interaction with α/β-DG as well as integrins of keratinocytes [13–15 , 30] . In addition , the interaction of LN-5 with integrin α3β1 and α6β4 activates the adhesion and spreading of keratinocytes for wound healing [14 , 15] . These previous results indicate that LN-5/α3β1 or α6β4 may be involved in mediating the attachment of M . leprae to HEKn cells and their subsequent invasion . Consistent with these results , the findings reported herein show that a pre-treatment with anti-integrin β1 or β4 antibody blocked the binding of LN-5-coated M . leprae to HEKn cells ( Fig 6B and 6C ) . Although M . leprae is not frequently detected in the epidermis , studies have clearly shown that M . leprae is found in the epidermis of patients with multibacillary leprosy [3 , 4 , 29 , 31–33] . M . leprae was detected in all layers of the epidermis , basal , suprabasal , prickle cells , and keratin layers [3 , 4] . In addition , M . leprae was also reported to be distributed in sweat glands and hair follicles [3] . Job et al . [4] suggested that the transepidermal discharge of M . leprae may be attributed to the possibility that M . leprae is transferred to the keratin layer by travelling inside keratinocytes from the basal to the keratin layer and that M . leprae then exits from hair follicles or sebaceous glands . Satapathy et al . [29] also reported that a 49-year-old man with lepromatous leprosy after dapsone monotherapy for 12 years had a recurrence of leprosy and that his skin biopsies showed bacillary clumps in epidermis . In that report , the authors suggested that health workers in leprosy control should consider the possibility that leprosy can be transmitted through the skin and by skin to skin contact , since large numbers of M . leprae are shed , even through intact skin . The findings reported in this study suggest that M . leprae binds to non-differentiated , proliferating HEKn cells by taking advantage of the interaction of LN-5 in the basal lamina of the epidermis and a surface receptor on keratinocytes , such as α-DG , integrin-β1 , or -β4 . However , although our results show the possibility that the epidermis is a route for M . leprae transmission , the transmission of M . leprae through the skin has not yet been experimentally proved . The mechanism responsible for the transmission remains unknown . | In the current study , we investigated the issue of how M . leprae is phagocytosed by human epidermal keratinocytes , neonatal ( HEKn ) . We focused on the role of LN-5 , a predominant form of laminin of the epidermis , in the interaction of M . leprae with keratinocytes . Our results show that M . leprae preferentially binds to LN-5-coated slides and coating M . leprae with LN-5 enhanced its binding to HEKn cells . In addition , a pre-treatment with an antibody against α-DG , integrin-β1 or -β4 inhibited the binding of LN-5-coated M . leprae to HEKn cells . These results suggest that M . leprae binds to keratinocytes by taking advantage of the interaction of LN-5 in the basal lamina of the epidermis and a surface receptor of keratinocytes , such as α-DG , integrin-β1 , or -β4 . | [
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] | [
"skin",
"cell",
"binding",
"cell",
"physiology",
"mycobacterium",
"leprae",
"keratinocytes",
"medicine",
"and",
"health",
"sciences",
"integumentary",
"system",
"tropical",
"diseases",
"macroglial",
"cells",
"collagens",
"epithelial",
"cells",
"bacterial",
"diseases",
"dermis",
"schwann",
"cells",
"epidermis",
"neglected",
"tropical",
"diseases",
"bacteria",
"infectious",
"diseases",
"animal",
"cells",
"proteins",
"biological",
"tissue",
"actinobacteria",
"glial",
"cells",
"biochemistry",
"cell",
"biology",
"anatomy",
"epithelium",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"leprosy",
"organisms"
] | 2019 | M. leprae interacts with the human epidermal keratinocytes, neonatal (HEKn) via the binding of laminin-5 with α-dystroglycan, integrin-β1, or -β4 |
Tick-borne relapsing fever spirochetes are maintained in endemic foci that involve a diversity of small mammals and argasid ticks in the genus Ornithodoros . Most epidemiological studies of tick-borne relapsing fever in West Africa caused by Borrelia crocidurae have been conducted in Senegal . The risk for humans to acquire relapsing fever in Mali is uncertain , as only a few human cases have been identified . Given the high incidence of malaria in Mali , and the potential to confuse the clinical diagnosis of these two diseases , we initiated studies to determine if there were endemic foci of relapsing fever spirochetes that could pose a risk for human infection . We investigated 20 villages across southern Mali for the presence of relapsing fever spirochetes . Small mammals were captured , thin blood smears were examined microscopically for spirochetes , and serum samples were tested for antibodies to relapsing fever spirochetes . Ornithodoros sonrai ticks were collected and examined for spirochetal infection . In total , 11 . 0% of the 663 rodents and 14 . 3% of the 63 shrews tested were seropositive and 2 . 2% of the animals had active spirochete infections when captured . In the Bandiagara region , the prevalence of infection was higher with 35% of the animals seropositive and 10% infected . Here also Ornithodoros sonrai were abundant and 17 . 3% of 278 individual ticks tested were infected with Borrelia crocidurae . Fifteen isolates of B . crocidurae were established and characterized by multi-locus sequence typing . The potential for human tick-borne relapsing fever exists in many areas of southern Mali .
The epidemiology of tick-borne relapsing fever was founded on the works of several independent investigators working across central Africa during the first decade of the 20th century . David Livingstone is credited with the first written account in 1857 of a malady associated with the bite of soft ticks in areas now known as Angola and Mozambique [1] , although the identity and route of transmission of the etiological agent were not discovered for another 45 years . Then in several closely dated publications of clinical and field observations , spirochetes were reported in the blood of acutely ill patients [2]–[6] , and the soft tick Ornithodoros moubata was identified as the vector , transmitting the bacteria when these naturally infected ticks were fed experimentally on cercopithecus monkeys [3] , [7]–[9] . The seminal work was done by J . Everett Dutton and John Todd while working in the eastern region of the Congo Free State ( now the Democratic Republic of Congo ) [10] . Both men contracted the infection while performing autopsies , and Dutton died there on February 27 , 1905 [3] . One year later , the spirochete that caused tick-borne relapsing fever across central Africa was named Spirillum Duttoni [11] , now named Borrelia duttonii [12] , to honor Dutton's contributions and sacrifice while working on tick fever and other tropical diseases including malaria and trypanosomiasis . Borrelia duttonii has no known nonhuman animal reservoir , although many investigations have tried to demonstrate that such associations with wild and domestic animals exist . This spirochete is transmitted from person to person by the bite or coxal fluid of O . moubata [13] . However , all other species of tick-borne relapsing fever spirochetes are maintained in enzootic foci that involve a diversity of small mammals [14] . The first spirochete reported in a wild African mammal was Spirochaeta crocidurae , which was found in a shrew Crocidura stampflii in Dakar , Senegal [15] . This spirochete , now named Borrelia crocidurae , is likely widespread across much of Africa north of the equator , from Egypt to Senegal and north to Tunisia [16]–[18] . The spirochete infects a variety of wild and peridomestic rodents and shrews , and is transmitted by two species of soft ticks , Ornithodoros erraticus and Ornithodoros sonrai . These ticks were previously considered two varieties , the large and small form , respectively , of O . erraticus [19] . Sautet and Witkowski [20] named the small form O . sonrai , in honor of the ancient Sonrai Empire centered at Gao , Mali , from where the ticks were described . This species of tick is the primary vector of B . crocidurae in sub-Saharan Africa [21] , [22] . Most ecological and epidemiological studies of tick-borne relapsing fever caused by B . crocidurae in West Africa have been done in Senegal . In 1989 , a series of investigations were begun soon after a French child living there contracted a recurrent febrile illness that was originally thought to be malaria , but after three months and seven acute episodes the illness was diagnosed as a borrelia infection [23] . Tick-borne relapsing fever of humans is prevalent in Senegal , and O . sonrai and B . crocidurae are associated with numerous species of small mammals in many regions of the country [18] , [22]–[26] . The potential risk for humans to acquire relapsing fever infection in Mali immediately to the east of Senegal is not well known , and only a few human cases have been reported from there . Some forays into Mali by Senegalese-based investigators found O . sonrai in the burrows of small mammals , and some of the ticks were infected with B . crocidurae based on PCR and detection of spirochetal DNA [18] , [21] , [27]; however , previous efforts directed at tick-borne relapsing fever in Mali are not clear . Rodhain et al . [28] identified two human cases in southwestern Mali in 1977 and 1988 , and two more recent human cases were diagnosed in France soon after the patients arrived there from Mali where they had become infected [29] , [30] . Clearly , tick-borne relapsing fever has occurred in Mali and the illness may be confused with malaria , as was suspected in Togo [31] . As stated in Manson's Tropical Diseases for the diagnosis of relapsing fever: “This fever is most usually confounded with subtertian malaria , from which it may be indistinguishable on clinical grounds” [32] . Therefore , given the high incidence of malaria in Mali [33] , and the lack of information regarding the prevalence of tick-borne relapsing fever there , we initiated studies to determine if there were endemic foci that involved small mammals , ticks and spirochetes . Here we identify several areas with evidence of infection , and discuss one region in particular that has a high prevalence of infected small mammals and ticks that live in close association with humans . In these villages in south central Mali , the potential risk for humans to acquire tick-borne relapsing fever is significant .
The Rocky Mountain Laboratories , NIAID , NIH , Animal Care and Use Committee approved study protocols #2008-1 and #2011-48 to perform the animal field studies , and protocols #2009-32 and #2009-87 for the feeding of ticks , mouse infection and isolation of relapsing fever spirochetes . All work in our study was conducted adhering to the institution's guidelines for animal husbandry , and followed the guidelines and basic principals in the United States Public Health Service Policy on Humane Care and Use of Laboratory Animals , and the Guide for the Care and Use of Laboratory Animals . Residents in the villages gave informed consent prior to our setting traps and collecting ticks in their houses . We collected small mammals in 20 villages across southern Mali from December 2007 to October 2011 . The locations varied from latitude 10° 35′ 20 . 8″ to 15° 01′ 25 . 3″ N and longitude 2° 50′ 55 . 0″ to 9° 58′ 32 . 5″ W ( Table 1 ) ( Figure 1 ) . The areas sampled ranged from the drier Sahel in the north to the moister wooded savannah in the south . The small mammals were captured alive in Sherman live traps ( H . B . Sherman Traps , Tallahassee , FL ) . During the first two field efforts ( December 2007 and January 2009 ) , we used both small and large traps: small trap size was 5 . 2×6 . 4×16 . 5 cm; large trap size was 7 . 6×8 . 9×22 . 8 cm . However , the larger traps were much more productive at capturing animals and thereafter we used only them . Traps were set in the late afternoon with bait comprised of locally acquired crushed peanuts , chopped onions and occasionally pieces of dried fish . Traps were placed inside and outside houses and collected early the next morning , at which time the animals were processed . The outside location of traps varied among the villages from immediately adjacent to the walls of houses to community gardens on the outskirts of the village . The animals were euthanized by the inhalation of isoflurane , and a terminal blood sample was collected via intracardiac puncture with a 1 ml tuberculin syringe and a 26-gauge 3/8-inch needle . Thin blood smears were made on glass microscope slides . The animals were examined for ectoparasites , which if found were collected in 70% ethanol . The animals were tentatively identified to genus or species in the field based on external characters [34]–[36] . Their body weight was measured in grams with a Pesola spring scale ( PESOLA AG , Baar , Switzerland ) , gender determined , and lengths of the head & body and tail were measured in centimeters . Each animal was photographed with a digital camera for future reference . One of us ( TGS ) visited the Smithsonian Institution's African mammal collection to examine specimens collected from various locations in West Africa to examine the external characters and skulls to assist in the identifications . Skulls from 60 animals collected in Mali were prepared for museum voucher specimens following standard curatorial procedures [37] and compared to illustrations and keys [36] . The nomenclature and taxonomic status of the animals were based on currently accepted names [38] , [39] . One external ear pinna was collected from every animal and preserved in 70% ethanol for DNA extraction and molecular identification of the species . A 3-mm round skin punch biopsy was extracted later from each external ear sample and DNA was purified with the DNeasy Blood and Tissue Kit , 96-well format ( QIAGEN Sciences , Inc . , Germantown , MD ) following the manufacturer's instructions . The mitochondrial ( mt ) cytochrome b ( cyt-b ) DNA was amplified with the PCR primers L14723 and H15915 [40] ( Table 2 ) and the Go Taq Flexi DNA Polymerase kit ( Promega Corp . , Madison , WI ) with an initial denaturation at 96°C for 3 min , followed by 35 cycles of 94°C for 30 sec , 55°C for 30 sec , 72°C for 2 . 5 min , and a final heating at 72°C for 7 min . PCR products were purified with the QIAquick PCR Purification Kit ( QIAGEN Sciences ) following the manufacturer's Spin Protocol . DNA sequences of the amplicons were produced with the BigDye Terminator v3 . 1 Cycle Sequencing Kit ( Applied Biosystems , Foster City , CA ) with reactions run for 45 cycles of 95°C for 10 sec , 50°C for 5 sec , and 60°C for 4 min . The sequence reaction products were cleaned with the BigDye XTerminator Purification Kit ( Applied Biosystems ) and sequenced with an Applied Biosystem's 3730xl DNA Sequencer . These sequences were submitted to GenBank at the NCBI using BLASTn [41] to determine the identification of each individual . GenBank accession numbers representative for each species we captured are in the results . Thin blood smears were fixed with 100% methanol and stained with the QUICK III statpak kit ( Astral Diagnostics Inc . , West Deptford , NJ ) . Fifty fields on each slide were examined for stained spirochetes with a Nikon Eclipse E800 microscope ( Nikon Instruments Inc . , Melville , NY ) at 600× magnification and oil immersion objective lens . Serum samples from most of the animals captured were tested by immunoblot for antibodies to relapsing fever spirochetes . Briefly , whole-cell lysates of Borrelia duttonii CR2A or B . crocidurae DOU ( isolated during this study ) , and purified heterologous GlpQ from Borrelia recurrentis were prepared as described [42] , [43] and electrophoresed in adjacent lanes by SDS-PAGE in Novex 12% Tris-Glycine Mini Gels-1 mm ( Life Technologies , Carlsbad , CA ) . The proteins were transferred onto nitrocellulose using the iBlot Dry Blotting System and iBlot Transfer Stacks ( Life Technologies ) or the Mini-PROTEAN II Cell and 0 . 45 µm membrane ( BioRad , Life Sciences Research , Hercules , CA ) . Each serum sample was tested at 1∶100 dilution ( two low-volume samples were diluted 1∶200 ) with the two-lane nitrocellulose panel . Bound primary antibodies were detected with HRP-rec Protein A ( Life Technologies ) and the Amersham ECL Western Blotting Detection Reagents ( GE Healthcare , Piscataway , NJ ) or the SuperSignal West Pico Chemiluminescent Substrate ( Thermo Scientific , Rockford , IL ) . Chemiluminescence of bound antibodies was detected with Amersham Hyperfilm ECL ( GE Healthcare Ltd . , Buckinghamshire , UK ) . A sample was considered positive if it contained antibodies reactive to 5 or more proteins in the whole-cell lysate and to the purified GlpQ protein . Attempts to collect the argasid tick vector , Ornithodoros sonrai , from small mammal burrows were done with a Craftsman , Incredi-Pull , 4-Cycle Blower/Vacuum 29 cc ( Sears , Roebuck and Co . , Hoffman Estates , IL ) . A tube was attached to the air intake aperture and connected to a hose ( modified from a version first described by Butler et al . ) [44] . One end of the hose was attached to the mouth of the intake , while the other end was inserted into the opening of a mammal burrow within the peoples' houses . A fine mesh screen was inserted into the air pathway to capture material that was aspirated into the hose . This burrow material was then removed from the device , screened again by hand and examined for live ticks . Some ticks were preserved in 70% ethanol while others were kept alive , and all specimens were taken to the laboratory and examined with a Zeiss Stemi SV 11 Stereomicroscope ( Carl Zeiss MicroImaging , Inc . , Thornwood NY ) , and identified to species [20] , [45] , stage and sex . DNA samples from 208 individual ticks preserved in ethanol were purified with the DNeasy Blood and Tissue Kit , following the manufacturer's Mini Spin Column Protocol ( QIAGEN Sciences ) . Prior to the extractions , each tick was placed in a 1 . 5 ml microfuge tube with the pestle and frozen together in liquid nitrogen to increase the efficiency for grinding the ticks into powder . The purified DNA was used as template for PCR that targeted the tick mt 16S rDNA and several borrelia genes ( see below ) . The tick mt 16S rDNA was amplified and both strands were sequenced using the primers listed ( Table 2 ) and the methods described above for the mammalian cyt-b gene . The chromatograms for each sequence were aligned to assure the correct identification of each base and ambiguous sequences at the ends were deleted . Live ticks were fed on laboratory mice , Mus musculus . For this procedure , the mice were anesthetized with pentobarbital ( 0 . 5 mg/10 g body wt ) via an intraperitoneal injection , the hair on the abdomen of the mouse was cut with electric clippers , and the ticks were placed on the abdomen and allowed to feed . Ticks typically took 30 to 60 minutes to feed , after which time they were placed into ventilated plastic tubes and kept within jars at 28°C and a saturated solution of KCl to maintain a relative humidity of 85% [46] . Those mice that were fed upon by ticks were caged individually and examined for spirochete infections for up to 10 days after the ticks had fed . Mice were examined for infection by nicking the tip of the tail and expressing a drop of blood ( approximately 5 µl ) from the tail vein onto a microscope slide , placing a cover slip over the blood , and examining the fresh sample with a Nikon Eclipse E200 dark-field microscope ( Nikon Instruments Inc . ) at 400× magnification . When spirochetes were detected , a terminal blood sample was collected from the anesthetized mouse via intracardiac puncture and 50–100 µl were inoculated into 5 ml of BSK-H medium ( Sigma-Aldrich Co . , St . Louis , MO ) or mBSK-c medium [47] supplemented with 12% rabbit serum . Cultures were held at 33–35°C and examined every few days for viable spirochetes . Cultures that grew successfully were passed into 100 ml of medium and allowed to grow to late exponential phase . Aliquots of the cultures were frozen with 10%/volume dimethyl sulfoxide ( DMSO ) at −80°C , and DNA was purified from the remainder of the sample as described previously [48] . Genomic DNA samples were analyzed by gel electrophoresis and multilocus sequence typing ( MLST ) . Intact plasmids were resolved from chromosomal DNA in 1% agarose reverse-field gels [49] and 0 . 35% agarose 2-dimension gels [50] to identify linear and circular molecules . Four borrelia chromosomal loci were targeted by PCR to amplify DNA for sequencing: the intergenic spacer ( IGS ) between the 16S rDNA and 23S rDNA , the 16S rDNA , the flagellin flaB gene , and the glycerophosphodiester phosphodiesterase gene ( glpQ ) ( primers and references in Table 2 ) . The methods and parameters for PCR and sequencing were as described above for the cyt-b gene . The nucleotide sequences were submitted to NCBI using BLASTn for comparison to homologous sequences in the database . The new sequences were analyzed with those sequences retrieved from GenBank using the MacVector 10 . 1 software package ( Accelrys , San Diego , CA ) and multiple sequence alignments were done with CLUSTAL W [51] . Phylograms were constructed with several algorithms but those presented were built with the UPGMA and Jukes-Cantor methods provided in the MacVector package .
We set traps on 36 nights during the study in 20 villages with a total effort of 2 , 909 trap-nights . We captured 744 animals for an overall trap success of 26% that included 14 species of rodents and shrews; however , seven species comprised 96% ( 717 of 744 total ) of the individuals ( Table 3 ) . Mastomys natalensis , Mastomys erythroleucus , and Mastomys huberti together comprised 76% of all captures ( 565 of 744 total ) . Mastomys natalensis was captured most frequently and this animal was the most ubiquitous , as we captured these rats at 17 of the 20 villages we sampled ( Table S1 ) . Praomys daltoni and the shrew , Crocidura olivieri , were also captured in many of the villages . In contrast , M . huberti , which is restricted to the Inland Delta of the Niger and Bani Rivers , was captured only in and near Belenikegny . In Soromba and the other nearby villages in the southern-most region we worked , M . natalensis was the only species we captured . Individuals of some species were trapped much more frequently within houses than outside ( Table 3 ) . While most of the M . natalensis were captured indoors , the opposite was true for M . erythroleucus and M . huberti . The cosmopolitan roof rat , Rattus rattus , was found mostly in Belenikegny and all but one of these animals was captured indoors . Crocidura olivieri was also captured more frequently within houses than outside . Sixty skulls were prepared from nine of the species captured ( Table S2 ) and these samples were deposited as voucher specimens in the Smithsonian Institution , National Museum of Natural History , Division of Mammals , Collection of African Mammals , Washington DC . The specimens currently have RML numbers ranging from M#-168 to M#-568 and are awaiting museum accession numbers . Identifications of 717 of the 744 animals captured were also supported by mitochondrial cyt-b DNA sequences , and 36 representative sequences for 12 species are deposited in GenBank with accession numbers JX292860–JX292895 ( see Table S2 for accession numbers associated with voucher skull specimens ) . No O . sonrai were found on any of the small mammals , which was not surprising given the short feeding time and nidicolous nature of these argasid ticks . Numerous fleas ( Xenopsylla species ) and mesostigmatid mites were collected , as were a few ixodid ticks and sucking lice . None of these arthropods are germane to the present study and will not be discussed further . Serum samples from 726 animals were tested for antibodies to investigate prior infection with relapsing fever spirochetes . For all locations , 82 animals ( 11 . 3% ) were seropositive by immunoblot analysis with antibodies binding to multiple proteins in the borrelia whole-cell lysate and to the purified GlpQ ( Table 4 ) ( Figure 2 ) . One or more of the animals captured in 14 of the 20 villages were seropositive ( Table 5 ) ( Table S1 ) . Animals that contained antibodies to relapsing fever spirochetes were distributed from Djougounte to Petaka , the most westward and eastward locations , respectively , that we sampled . However , more than half of the seropositive animals were captured in two villages near Bandiagara: Kalibombo and Doucombo . Here , 45 of 128 of the animals ( 35% ) were seropositive , and included M . natalensis , P . daltoni and C . olivieri . These three species that showed evidence of prior infection were captured in houses and lived in close proximity to humans . Thin blood smears were stained and examined for spirochetes from 724 of 744 animals captured . We detected spirochetes in 16 animals ( 2 shrews and 14 rodents ( Table 6 ) ( Figure 3 ) ; actively infected animals were captured in five villages . The number of spirochetes observed varied from 1 to 264 spirochetes in the 50 microscopic fields examined . The majority of infected animals were captured in Kalibombo and Doucombo , the same two villages with the highest prevalence of seropositive animals . In these two villages , 13 of 130 animals ( 10% ) had detectable spirochetes when captured , including M . natalensis ( 11 individuals ) , P . daltoni ( 1 individual ) and C . olivieri ( 1 individual ) . Serum samples from three slide-positive M . natalensis contained live spirochetes and these samples were inoculated into laboratory mice . We isolated spirochetes from two of the mice ( DOU-686 and DOU-690 ) and characterized them by multi-locus sequence typing ( see below ) . We focused our tick collecting efforts in Doucombo and Kalibombo because of the higher percentage of infected and seropositive animals there compared to other locations . During April 2011 , September–October 2011 , and January 2012 , we found O . sonrai ticks in small mammal burrows inside 24 houses in Doucombo and 14 houses in Kalibombo . In total , we collected 734 O . sonrai , which included 501 nymphs , 146 males and 87 females . From this total , the 208 ticks collected during April 2011 were preserved in 70% ethanol and examined individually by PCR for spirochete infection by targeting only the IGS locus . The mitochondrial 16S rDNA sequence was also determined for five of these ticks , which confirmed their identity as O . sonrai ( GenBank accession numbers JX292854–JX292859 ) . In this group of 208 ticks , borrelia DNA was detected in 37 ( 17 . 8% ) of them ( 21 nymphs , 10 males , 6 females ) . DNA sequences for the IGS locus were of two types with 99 . 2% identity . The two sequences were submitted to GenBank for comparisons to other available sequences and both sequences aligned closest to the IGS sequence of the Achema strain of B . crocidurae that originated from O . sonrai ticks collected in Mauritania [52] . Ticks collected during September–October 2011 and January 2012 were kept alive and shipped to the Rocky Mountain Laboratories , NIAID , NIH , Hamilton , Montana ( CDC Permit #2011-08-41 and USDA Veterinary Permit #117004 ) . Five pools of ticks ( 2–11 ticks per pool ) and 70 single ticks ( 18 nymphs , 28 males , 24 females ) were fed on individual mice . Two tick pools and 11 individual ticks ( 2 nymphs , 7 males , 2 females ) transmitted spirochetes and infected blood from these mice produced 13 isolates of spirochetes in mBSK-c medium . DNA samples from the 15 isolates that originated from ticks and M . natalensis were analyzed by reverse field and 2-dimensional gel electrophoresis and MLST . Six distinct plasmid profiles ( I–VI ) were found among the undigested borrelia DNA samples from the isolates ( Figure 4 ) . Like all borreliae , these spirochetes contained numerous linear plasmids ( at least ten ) that ranged in size from approximately 12 . 5 to 100 kilobases and circular plasmids of undetermined size . The differences in the plasmid profiles for the isolates grouped them closely with the DNA sequence data presented below . MLST segregated the 15 spirochete isolates into four primary groups ( A–D ) ( Table 7 ) . Each group had identical 16S rDNA and flaB sequences that were unique from members of the other groups , while glpQ sequences segregated the spirochetes into the same groups for 13 of the 15 isolates . Four IGS sequence types were found that included the two sequence types we identified in the ethanol-preserved ticks . The IGS sequences varied among the groups but there was not strict congruence . For example , IGS sequence type 1 was shared among all group A and four of the five group D spirochetes . Overall , three of the four groups of spirochetes ( A , B , and C ) were distinguished by their unique sequences and plasmid profiles . The fourth group ( D ) was distinct from the others but also displayed some heterogeneity in the glpQ and IGS sequences . The group D isolate DOS-6 was unique from all other isolates by its glpQ and IGS sequences , and its unique plasmid profile . Sixty-two borrelia DNA sequences have been deposited in GenBank with the following accession numbers: 16S rDNA ( JX292896–JX292910 ) ; flaB ( JX292911–JX292925 ) ; glpQ ( JX292926–JX292940 ) ; IGS ( JX292941–JX292957 ) . Phylograms derived from multiple alignments for each locus including our isolates and sequences in the database all grouped the Malian spirochetes with B . crocidurae ( data not shown ) , with one exception . The partial DNA sequence of the 16S rDNA ( 1 , 262 bp ) for isolate DOS-2 was identical to the same length of sequence for B . duttonii Ly . We present two phylograms based on the IGS locus ( Figure 5 ) and the concatenated sequence comprised of the 16S rDNA ( 1 , 262 bp ) , flaB ( 990 bp ) , and glpQ ( 1 , 002 bp ) ( 3 , 254 bp total ) ( Figure 6 ) . The four unique IGS sequences from all our Malian B . crocidurae aligned closest with IGS sequences for B . crocidurae from Mauritania ( Achema strain ) and Tunisia ( 7-10TO47 and 12TO38 DNA samples from infected ticks ) . This locus clearly grouped the B . crocidurae samples and distinguished them from the other Old World species ( B . duttonii , B . recurrentis , B . persica and B . hispanica ) ( Figure 5 ) . The percentage identity values for the IGS sequences among the seven B . crocidurae samples in heterologous matches ranged from 98 . 2 to 99 . 8% but these sequences had identity values of 65 to 91 . 9% when compared to the other Old World species ( Table S3 ) . The identity values were considerably less ( 49 . 7 to 57 . 3% ) when the B . crocidurae samples were compared to the New World relapsing fever spirochete species Borrelia hermsii , Borrelia turicatae , and Borrelia parkeri . The phylogram constructed with the concatenated sequences ( Figure 6 ) contained fewer Old World species of Borrelia than did the IGS analysis ( Figure 5 ) , because fewer sequences were available . We included the Achema strain of B . crocidurae because complete sequences were available for the three loci ( CP003465 ) [53] . This analysis again identified the Malian spirochetes as B . crocidurae , which were clearly separated from the two very closely related species B . duttonii and B . recurrentis . Percentage identity values among the 15 Malian isolates ( six unique sequences ) were high at 99 . 5 to 99 . 9% ( Table S4 ) .
One of our primary methods to determine the presence of relapsing fever spirochetes in the small mammals was done indirectly with serological tests for anti-relapsing fever spirochete antibodies that identified animals that had been previously infected . Serological approaches for the surveillance of other vector-borne pathogens have been used for many years , such as wild carnivore serology for plague [54] and the use of sentinel chickens to monitor seroconversion for seasonal activity of numerous mosquito-borne viruses [55] . Serological surveys for relapsing fever have rarely been used for field studies [56] , [57] and have never been utilized for the studies of enzootic foci of relapsing fever in Senegal or other regions of Africa . Earlier concerns for the specificity of such tests for relapsing fever antibodies were directed at the extreme antigenic variation known for these bacteria during infection [58] ( hence what antigens should or could be used ) , and the antigenic relatedness among different Borrelia species , which results in the lack of specificity of the antibodies detected [59] . Such serological cross reactivity meant that people having antibodies reactive to antigens of Borrelia burgdorferi , the cause of Lyme borreliosis , may have actually been infected with relapsing fever spirochetes , with the reverse also being true . This dilemma was rectified to a large extent with the identification of an immunogenic protein in the relapsing fever spirochetes , glycerophosphodiester phosphodiesterase ( GlpQ ) , which is absent in the agents of Lyme borreliosis [42] , [43] , [60] . For studies in North America where both Lyme borreliosis and relapsing fever exist , the application of GlpQ has helped to serologically discriminate people and wild mammals that were infected previously with relapsing fever spirochetes and not B . burgdorferi . While the presence of Lyme borreliosis spirochetes throughout West Africa is unknown and doubtful given the ecological requirement of the Ixodes species of ticks [61] , our test would discriminate between such prior infections . The strength of a specific and sensitive serological test for relapsing fever compared to a blood smear taken from the same animal lies in the temporal persistence of antibodies after infection compared to the brief and transient time when spirochetes are detectable in the blood . Therefore , in a population or community of animals susceptible to a bacterial infection , the proportion of individuals that are seropositive should increase seasonally whereas the number of animals actively infected at any one time may not . Our data demonstrate this utility of serology over active infection quite convincingly for spirochete activity . In the 20 villages we sampled , 14 villages had seropositive animals while only 5 villages had animals with detectable spirochetemias . These results were strengthened by the fact that for the six villages where no seropositive animals were found , neither was any animal found with active infection . Overall , 11 . 3% of the animals tested from all villages were seropositive while only 2 . 2% of the animals had spirochetes seen in their blood . Additionally , serology implicated eight species of mammals as hosts for spirochetes ( Table 4 ) while the examination of blood smears found spirochetes in just five species ( Table 6 ) . The serological results were supported again by the blood smears as only those species that had seropositive individuals also had animals with spirochetes detected by microscopy . We examined 50 microscopic fields for each blood smear to examine the mammals for spirochete infection . Investigators in Senegal typically examined 200 microscopic fields while looking for spirochetes in blood smears [18] , [24] , [26] . Therefore , our approach would have only a 25% chance at detecting a positive smear having only one spirochete in 200 fields , if the area of one microscope field and the volume of blood in each field were the same . However , as stated above we relied on serology to increase the sensitivity of our surveillance . We realize that microscopy is not the most sensitive method to detect relapsing fever spirochetes in the peripheral blood of an infected animal , and other investigators have on occasion used animal inoculation for studies on relapsing fever in Senegal . Diatta and colleagues examined 82 rodents comprised of three species collected in Dielmo , Senegal , and compared the success of examining blood smears to inoculating their blood and brain suspensions in laboratory mice to detect infection in the wild animals [24] . Mastomys erythroleucus comprised 89% ( 73 of 82 total ) of the animals examined and from them only one blood slide was positive while five of the blood samples produced spirochetemias in mice , and brain tissues from 10 animals yielded laboratory infections . Thus brain inoculations were ten-times more sensitive at detecting infections in the wild rats compared to the microscopic examination of stained blood smears . More recently , Vial and colleagues expanded the studies in the same village in Senegal , and again they found that the inoculation of brain suspensions from the wild mammals into laboratory mice resulted in 12% infection compared to only 0 . 74% prevalence of infection based on the examination of blood smears [18] . Nordstand and colleagues also detected B . crocidurae and B . duttonii in patient blood samples by PCR when no spirochetes were observed by microscopy [31] . We probably missed some active infections in the animals we captured by examining only 50 microscopic fields and not utilizing PCR . We relied on serological surveillance to complement our microscopic examinations to increase our ability to detect the presence of spirochetes circulating in the numerous locations and species we sampled . Our efforts across southern Mali demonstrated that many of the villages had spirochetes infecting several species of rodents and shrews , however the prevalence of infection was low . We began searching for ticks in a few villages that had higher seroprevalence rates , such as Belenikegny on the Bani River . Our initial attempts to find O . sonrai ticks there were unsuccessful . Then in late September 2010 , nearly the entire village was flooded when the Bani River overflowed its banks . We redirected our efforts to the Bandiagara region , where our attempts to find ticks were successful . Additional trapping of the small mammals there demonstrated that in two nearby villages , Doucombo and Kalibombo , 35% of the rodents and shrews were seropositive and 10% of the animals had positive blood smears at the time of capture . The spirochetemias in some of the animals were quite high ( Table 6 ) with three of the infected M . natalensis having 112 , 140 , and 264 spirochetes observed in the 50 fields examined . The studies in Senegal did not report the numbers of spirochetes seen in the blood of wild mammals , although for clinical investigations with human blood the densities were low; 75% of the smears contained less than 20 spirochetes in 200 fields [23] . Ornithodoros sonrai ticks were difficult to find until we intensified our efforts at Doucombo and Kalibombo . Here , the ticks were abundant and present in burrows in most of the houses we sampled . Ticks had a prevalence of spirochete infection of 17–18% . The estimates of infection were strikingly similar based on the PCR assays of alcohol-preserved ticks and when live ticks were fed on mice and transmitted spirochetes to them . In and around Dielmo , Senegal , the prevalence of B . crocidurae infection in O . sonrai ticks varied between 21 to 66% , based on PCR and DNA sequencing the flaB gene [18] . In transect surveys in Senegal , Mauritania and Mali , O . sonrai ticks were found in 26 of 30 villages sampled [18] , although the publication does not specifically state what was found in Mali . However , Trape reported elsewhere that O . sonrai ticks were found in burrows in Djougounte , Sama , Molibana and Gao , Mali [27] . A survey of small mammal burrows in Tunisia found 15 . 1% of the O . erraticus ticks were infected with borrelia [16] . The uncultured spirochetes were identified as B . crocidurae by sequencing the DNA of amplicons of the 16S rDNA , flaB , and IGS loci . Much earlier surveys in Egypt , which predated the development of PCR and a culture medium for borrelia , demonstrated that 76 of 215 pools of O . erraticus ticks collected from rodent burrows transmitted spirochetes , assumed to be B . crocidurae , when fed on laboratory mice [17] . We established 15 novel in vitro isolates of B . crocidurae from Malian ticks and rodents . These isolates allowed us to get a preliminary characterization of the genetic diversity from this group of spirochetes , and to compare our molecular data to the results of previous investigations . However , while reviewing the literature we soon realized that very few in vitro cultures of B . crocidurae existed prior to our work . van Dam and colleagues claimed to be the first to isolate B . crocidurae in culture from the blood of two patients infected in The Gambia and Senegal in 1997 [62] . Yet , three years earlier Fukunaga and colleagues included two strains of B . crocidurae ( ORI and one isolate not designated ) in their phylogenetic analysis of Borrelia species that had been grown in BSK-II medium [63] . Previous work to characterize Old World relapsing fever borrelia utilized spirochetes that were isolated and maintained by serial passage in mice [28] , [64] . Ras and colleagues performed the first large scale phylogenetic analysis of what they called “noncultivatable” relapsing fever spirochetes by utilizing PCR to amplify the 16S rDNA from spirochetes in the blood of infected laboratory mice [64] . Their analysis included nine in vivo isolates of B . crocidurae that originated from ticks , human blood , and rodents from Mauritania , Senegal , Morocco and Mali . The two in vivo isolates from Mali , BAR and SIS , were those spirochetes from human patients infected in 1977 and 1988 , first reported by Rodchain et al . [28] . In spite of the various locations and biological sources for the nine in vivo isolates of B . crocidurae examined by Ras and colleagues , the 16S rDNA sequences were identical [64] . This is in contrast to what we found among our 15 isolates of B . crocidurae from Doucombo and Kalibombo . At these two nearby villages , we identified four 16S rDNA sequences , one of which ( from isolate DOS-2 ) was identical to the sequence for the Achema strain of B . crocidurae studied by Ras and colleagues [64] and two other research groups [53] , [65] . During the epidemiological investigations of tick-borne relapsing fever in Senegal , no attempts were made to culture or identify the spirochetes observed in humans , wild mammals or ticks [23]–[26] . Naming the spirochetes as B . crocidurae during these studies was based on the identity of the tick vector , O . sonrai , which is not known to transmit any other species of relapsing fever spirochete . More recently , Trape and his colleagues reported the incidence of human relapsing fever in Dielmo , Senegal , for 14 consecutive years ( 1990–2003 ) [18] . Small mammals and ticks were also collected and examined for spirochete infection . Spirochetes detected in human blood and small mammals were not isolated or identified but O . sonrai ticks were assayed by PCR . Partial internal fragments of flaB were sequenced from infected O . sonrai ticks collected in Senegal and Mauritania [18] , [22] . All sequences ( number of samples not stated ) were identical but the one partial sequence deposited ( 284 bp; DQ234749 ) varied by 1 bp from our flaB sequences within the 284 bp that could be compared . The trend in recent years to identify B . crocidurae has been to use as little sequence data as possible via PCR using one or more partial coding or non-coding targets . Most of these approaches have been used to identify spirochetes in people living in or having traveled to endemic areas of West Africa [29] , [66]–[70] . The clinical diagnostic approach has merit but eliminates the potential to gain more genetic and biological information had these spirochetes been isolated in culture . For example , through the many efforts of Cutler and her collaborators , the borrelia research arena has benefited tremendously by having many isolates of B . recurrentis and B . duttonii established in vitro [71]–[74] . These isolates have provided the basis for a greater understanding of the genetic diversity and molecular biology of African relapsing fever spirochetes , and they provided the material for a whole genome comparison of these two important louse- and tick-borne pathogens [53] . Our isolates of B . crocidurae demonstrated a rather striking amount of genetic diversity in the plasmid content and DNA sequences of highly conserved genes . In 1986 , Hyde and Johnson first reported that B . crocidurae harbored plasmids [75] . However , we found nothing in the literature to which we could compare our findings , which is the variation in number and size of plasmids among different isolates . The genome of the Achema strain of B . crocidurae has been determined ( CP003426–CP003465 ) [53] , although the plasmid-associated contigs were not assembled into their full-length native molecules . Our estimate of at least 10 linear and one or more circular plasmids in our isolates of B . crocidurae may be an underestimate . B . duttonii Ly and B . recurrentis A1 contain 16 and 7 plasmids , respectively [53] , although the number and size of plasmids varies among isolates for both species [72] , [73] , as we observed for B . crocidurae ( Figure 4 ) . Our MLST method identified the spirochete isolates as B . crocidurae and identified four distinct genomic groups . We and other collaborators have used this approach to characterize the North American relapsing fever spirochetes B . hermsii , B . turicatae and B . parkeri [49] , [76] , [77] . Toledo and colleagues applied MLST to identify an isolate of relapsing fever spirochete from Spain as Borrelia hispanica [65] . Recently , the chromosomes of B . recurrentis A1 , B . duttonii Ly and B . crocidurae Achema were aligned to identify homologous non-coding , intergenic spacer sequences ( not the IGS locus ) that were used to develop a PCR – DNA sequence typing scheme to distinguish these three species of spirochetes [53] . This multispacer sequence typing utilized the concatenated sequence of five intergenic spacers that totaled approximately 2 , 300 bp . The method was applied to 60 samples of infected human blood from relapsing fever patients from Ethiopia ( 30 B . recurrentis samples ) , Tanzania ( 17 B . duttonii samples ) and Senegal ( 13 B . crocidurae samples ) . The method clearly distinguished the three species of spirochetes with no overlap , which was not the case when using the IGS region [78] , [79] . The method also demonstrated seven types among the 13 samples of B . crocidurae compared to only five types among the 47 samples representing the other two species [53] . Clearly , a pattern is emerging from our efforts and those of Elbir and colleagues [53] that shows a much more diverse population structure for B . crocidurae , an enzootic pathogen with multiple vertebrate hosts , than has yet been demonstrated for either B . recurrentis or B . duttonii , neither of which has a nonhuman vertebrate host . Herein we present evidence for the widespread occurrence of relapsing fever spirochetes infecting small mammals across southern Mali . In the Bandiagara area , we identified two villages where the essential triangle for an enzootic focus for a vector-borne disease exists . The susceptible peridomestic rodents and shrews , the tick vector O . sonrai , and the spirochetal agent B . crocidurae all coexist with the human population . Therefore , we conclude that the potential for human infections is present in Mali , as has been described in Senegal [18] . During the course of our studies , a young girl who visited Mopti , approximately 75 km west-northwest of Bandiagara , was diagnosed with a B . crocidurae infection when she returned to France [29] . While her blood smears were negative for plasmodia , spirochetes were eventually detected retrospectively after antibiotic treatment . Just as it was over 100 years ago and remains today , distinguishing tick-borne relapsing fever from human malaria remains a clinical and diagnostic challenge [2] , [31] . We hope that our efforts described herein will stimulate future investigations to determine the extent to which tick-borne relapsing fever is a human public health problem in Mali . | Tick-borne relapsing fever is a spirochete-caused , recurrent illness acquired by the bite of fast-feeding ticks . In Mali , the potential for people to acquire relapsing fever is unknown although a few human cases have been reported there . Human malaria is also abundant in Mali , and could be complicating the diagnosis of relapsing fever . The relapsing fever spirochete , Borrelia crocidurae , is maintained in natural cycles involving small mammals and its tick vector Ornithodoros sonrai . Therefore , we investigated 20 villages across southern Mali to determine if relapsing fever spirochetes were circulating in small mammals and ticks that lived with people . We found that 11 . 3% of the 726 mammals tested showed evidence of prior infection , while 2 . 2% of the animals were actively infected . The tick vector was abundant in two villages we sampled , and overall 17 . 3% of the individual ticks tested were infected with spirochetes . We also isolated the spirochetes , Borrelia crocidurae , from rodents and ticks and compared their genetic makeup to other species of African spirochetes . We conclude that in some areas of Mali , people are at risk of acquiring tick-borne relapsing fever . Therefore , we recommend that blood smears from acutely ill patients be examined microscopically for spirochetes . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"public",
"health",
"and",
"epidemiology",
"mammalogy",
"zoology",
"ecology",
"vector",
"biology",
"epidemiology",
"global",
"health",
"biology",
"microbiology",
"host-pathogen",
"interaction",
"public",
"health",
"bacterial",
"pathogens",
"parasitology"
] | 2012 | Endemic Foci of the Tick-Borne Relapsing Fever Spirochete Borrelia crocidurae in Mali, West Africa, and the Potential for Human Infection |
We present methods to construct phylogenetic models of tumor progression at the cellular level that include copy number changes at the scale of single genes , entire chromosomes , and the whole genome . The methods are designed for data collected by fluorescence in situ hybridization ( FISH ) , an experimental technique especially well suited to characterizing intratumor heterogeneity using counts of probes to genetic regions frequently gained or lost in tumor development . Here , we develop new provably optimal methods for computing an edit distance between the copy number states of two cells given evolution by copy number changes of single probes , all probes on a chromosome , or all probes in the genome . We then apply this theory to develop a practical heuristic algorithm , implemented in publicly available software , for inferring tumor phylogenies on data from potentially hundreds of single cells by this evolutionary model . We demonstrate and validate the methods on simulated data and published FISH data from cervical cancers and breast cancers . Our computational experiments show that the new model and algorithm lead to more parsimonious trees than prior methods for single-tumor phylogenetics and to improved performance on various classification tasks , such as distinguishing primary tumors from metastases obtained from the same patient population .
In this paper , we develop new methods to advance the theory of phylogenetic inference for reconstructing evolutionary histories of cell populations in solid tumors . The work is specifically designed for use in tracking tumor evolution by gain and loss of genomic regions as assessed by multicolor fluorescence in situ hybridization ( FISH ) , which measures the copy numbers of targeted genes and chromosomes in potentially hundreds of individual cells of a tumor . This technology was the basis of the earliest methods for phylogenetic reconstruction of single tumors [1] , [2] . FISH remains uniquely valuable for such studies because the large number of cells that FISH can profile makes it possible to collect data on enough tumors in enough detail to build cell-by-cell phylogenies for populations of tumors and begin to study the common features of these phylogenies . In the present work , we specifically extend our previously developed inference algorithms to encompass a more complicated but realistic model of evolution of FISH probe counts , accounting for gain and loss of genetic material at the level of single gene probes , multiple probes on a single chromosome , or a probe set distributed across the whole genome . We demonstrate the value of these algorithmic improvements to more accurate phylogenetic inference and improved effectiveness of the resulting phylogenies in downstream prediction tasks . The present work adds to the growing list of phylogenetic methods in cancer modeling , which were reviewed through 2008 in [3] . These include methods for analyzing comparative genomic hybridization ( CGH ) or other genetic gain/loss data in a single tumor type [4]–[11] , for defining the cell type lineage of single tumors [1] , [2] , [12] , , for organizing a taxonomy of tumor types [14] , for reconstructing a partial order of genetic changes in multiple samples from one patient [15] , and for reconstructing progression from cell types inferred from bulk genomic assays [16] . Recent high-throughput sequencing studies have also used ad hoc phylogenetic methods to infer putative tumor progression scenarios , e . g . , [17]–[20] . Like many of these methods , the present work is aimed at building tree models that provide a proposed partial order on the observed cell states , a strategy motivated originally by the work of Fearon and Vogelstein , proposing a linear order for four types of events in colorectal cancer and associating each event with a tumor stage [21] . Other ordering methods have been proposed , mostly for CGH or breakpoint data [15] , [22]–[28] and , more recently , sequencing data [29] , [30] . The present work specifically advances the reconstruction of phylogenetic histories of single tumors from intratumor cellular heterogeneity data . The use of phylogenetic methods to reconstruct histories of single tumors was first developed in our prior work [1] , [2] by taking advantage of the ability of FISH to profile genetic changes in large numbers of single cells , allowing one to survey hundreds of cells per tumor in populations of tens of tumors [31] . This early work showed that even small numbers of markers could reveal numerous genetically distinct cell populations in single tumors , which could be resolved by phylogenetic inference to reveal multiple distinct pathways of progression between tumors and even within single tumors . Numerous studies since then , using multicolor FISH [2] , [31]–[36] and , more recently , single-cell sequencing [19] , [37]–[39] have greatly increased our ability to identify distinct cell populations and , in the process , revealed far more extensive intratumor heterogeneity than had been suspected prior to 2010 ( reviewed in [40] ) . The repeated observation of intratumor heterogeneity has necessitated a reconsideration of Nowell's [41] theory that tumors evolve clonally , showing that a tumor may contain many subpopulations relevant to the clinical prognosis of the patient [42] and that rare subpopulations may be more relevant to prognosis than the most common ones [43] . Furthermore , a simulation study has suggested that methods based on average copy number data perform poorly when there is substantial intratumor heterogeneity [44] . Such findings suggest a need for improved methods for organizing the dozens or hundreds of observed cell states in single tumors to infer the evolutionary processes that produced them . Despite extensive work on tumor phylogenetics , however , the study of algorithms for reconstructing tumor evolution from large numbers of single cells has lagged far behind advances in data generation . The standard in practice for single-cell tumor phylogenetics remains the use of simple generic phylogeny algorithms ( e . g . , neighbor-joining [45] ) that are not designed to model the patterns of copy number changes one would expect from evolution by chromosome abnormalities that largely drive tumor evolution . Until recently , algorithms designed specifically for inferring phylogenies of single tumors from FISH data have been limited to just a few probes per cell and lacked robust , publicly available software implementations [1] , [2] , [34] . In prior work [46] , we developed algorithms to find copy-number phylogenies for in principle arbitrary numbers of probes and cells . That work , however , was itself limited to a simple model in which tumor cells evolve by events of gain or loss of a single copy number of a single probe at each mutation step . In real tumors , gene copy numbers can change due to a variety of mechanisms , including: These events are illustrated schematically in Figure 1 . While more complex probabilistic models of tumor evolution have been developed for inference of small phylogenies , with approximately ten taxa per tumor corresponding to distinct biopsies ( e . g . , [47] ) , the class of inference algorithms such models require would not be expected to scale to phylogenies of hundreds of single cells per tumor such as those examined in the present work . The work presented here seeks to fill this need for scalable phylogenetic algorithms capable of fitting more realistic models of tumor-like evolution to data sets of hundreds of single cells per tumor . We improve on our prior work for inferring tumor evolutionary models considering only SD events [46] to now include CD and GD events , which are also frequently observed in tumor progression . We specifically focus on the problem of accurately inferring evolutionary distances between distinct cells in terms of maximum parsimony combinations of SD , CD , and GD events . The major contributions of the work are: The new methods are implemented in version 2 of our software FISHtrees ( ftp://ftp . ncbi . nlm . nih . gov/pub/FISHtrees ) . The work addresses a critical need in modern cancer research for algorithms capable of inferring evolutionary trajectories of hundreds of single cells per tumor under plausible models of evolution including both gene-specific and chromosome abnormalities that are central drivers of true tumor evolution .
To measure accuracy of the methods for FISH datasets with a known ground truth , we generated a dataset of trees with six probes , two of which were treated as being on the same chromosome . Each tree was generated by starting from a diploid root node and executing a branching process in which each node was recursively assigned a number of children drawn from a geometrically distributed random variable with mean . Each child was distinguished from its parent by selecting an SD , CD , or GD event with probability for each of the six possible SD events , of a CD event , and of a GD event . This process terminated when all leaf nodes had been assigned zero children by the sampling . We then generated simulated FISH data for each tree by uniformly sampling cells from the nodes in this topology . The simulated data corresponds to counts of probes for each sampled cell in the tree . We applied Algorithm 3 ( see Methods ) to find a minimum-cost tree for each of four event models: ( i ) SD only , ( ii ) SD and CD , ( iii ) SD and GD , and ( iv ) SD , CD and GD . We quantified the accuracy of tree inference by comparing each simulated true tree to its corresponding inferred tree derived from the sampled cells . This assessment was performed at the level of accuracy of tree edges by the following procedure: Intuitively , this formula measures the fractional agreement between bipartitions of the trees relative to the total number of bipartitions . We use a matching-based formula , rather than the more familiar Robinson-Foulds metric [53] , both because of its greater sensitivity to small changes in trees and because the Robinson-Foulds measure is not defined for trees with different node sets . We also note that we use a different normalization factor than in our prior work [46] , normalizing essentially by the total number of edges between the two trees , to control properly for the fact that different inference methods may infer different numbers of tree edges . The reconstruction error ranges in value from , if the real and inferred trees are isomorphic , to an upper bound of in the limit of complete disagreement . To illustrate the meanings of the terms of the equation for , we present a simple example using a hypothetical ground truth and an inferred tree presented in Figure 3 ( A ) and Figure 3 ( B ) , respectively . The set of nontrivial bipartitions in the ground truth are and the nontrivial bipartitions in the inferred tree are If we apply the matching algorithm on these two sets of bipartitions , the first and second bipartitions in the ground truth tree are matched with the first and second bipartitions in the inferred tree , respectively . The weight of the matching is . The number of common taxa between these two datasets is . The total number of nontrivial bipartitions in the real and inferred trees are and . Plugging these values into the equation for , we calculate . A comparison of the four models is presented in Figure 4 . The SD model showed reconstruction error with standard deviation ( s . d . ) of across the trees . The SD+CD model yielded error with s . d . . SD+GD yielded error with s . d . . The full SD+CD+GD model yielded error with s . d . . Collectively , the results suggest that one can reconstruct reasonably accurate trees even from the SD-only model , despite the fact that the trees were generated from a model of all three event types , although accuracy improves with each event type added . Accounting for GD events made a larger difference in accuracy than accounting for CD events , presumably because a missed GD event might require many SD or CD events to explain it , while a missed CD event could be explained with just two SD events . The reconstruction error for the full model is reduced by more than 1 . 7-fold relative to the SD-only model considered in our prior work . We further compared these results to those derived using generic phylogenetic methods that have been used in much of the single tumor phylogenetics work to date [16] , [54] . We tested the accuracy of reconstruction of the simulated trees described above using generic neighbor joining ( NJ ) with Euclidean distance and pure maximum parsimony ( MP ) treating copy numbers as arbitrary characters , approaches chosen because they have been the primary alternatives to our specialized algorithms in the single-tumor phylogeny literature . We omit here comparison to more complicated Bayesian phylogenetic models ( e . g . , [47] ) because such approaches are not scalable to the numbers of cells we examine . We then used the weighted matching based similarity method , described above , to calculate the mean percentage reconstruction error between the inferred and the ground truth trees . The mean reconstruction errors for NJ and MP were ( s . d . ) and ( s . d . ) , respectively , in contrast to the error of ( s . d . ) for the SD+CD+GD algorithm proposed here . The test thus demonstrates that when the underlying evolutionary process includes cancer-like chromosome abnormalities , errors are substantially reduced by using an algorithm designed for that model relative to standard off-the-shelf algorithms still widely used for single-tumor phylogenetics work . We performed additional experiments to evaluate the effects of different evolutionary parameters on the accuracy of inference of tumor progression trees by FISHtrees . For this experiment , we selected five different combinations of probabilities of SD , CD and GD events for generating the ground truth trees and then used SD , SD+CD , SD+GD and SD+CD+GD models to infer the tumor phylogenies . These data sets again each used six probes with two of the six on a common chromosome . The selected five combinations of ( SD , CD , GD ) event probabilities are: , , , and . These combinations of event probabilities were chosen to yield trees of comparable complexity to the real data while producing test sets enriched in distinct combinations of the three event types . They thus allow us to consider how robust our algorithms are to contributions from each of the three event types , singly or in combination . We report the reconstruction error for trees for each of these combinations of event probabilities in Table 1 . These results again show that accuracy improves with each event type added . When the probability of SD events is high ( as in combination 3 ) , the SD model results in highly accurate trees ( mean reconstruction error of with s . d . ) . Accounting for GD events in combination with SD events always result in larger improvement in the reconstruction error in comparison to the SD+CD models , even when the CD events are very frequent ( as in combinations 2 and 4 ) . Finally , accounting for GD events in combination with SD and CD events results in the largest improvements when the probability ratio of GD events to SD+CD events is highest , as can be seen from comparison of parameter sets 1 and 2 . Next , we performed simulation tests to evaluate the effects of non-uniform distributions of cells across different levels of the trees on the performance of our tree inference method . In our initial simulation experiments described above , we assumed that observed cells were sampled uniformly across clones . In real tumors , the distribution of cells would not typically be uniform due to differences in age and fitness of clones . In order to test robustness of our method to non-uniformity of clone frequencies , we sampled the cells following a non-uniform model in which the sampling frequency of a clone varies geometrically with its depth in the tree with a parameter . We used values of and for in our experiments . When , of the total cells are located in the first three levels of the trees , while for , this fraction is . We generated trees in each case with probabilities of SD , CD and GD events fixed at and . We again used SD , SD+CD , SD+GD and SD+CD+GD models to infer the tumor progression trees . We present the results from this experiment in Table 2 , where we also show the results from the uniform sampling of the cells . Additionally , we report the results on the trees inferred using NJ and MP for these three different cell distributions . From the table , we can see that the reconstruction error increases with increasing for all methods . The SD+CD+GD model , however , shows the best performance among all the models for all three values of and the least loss of performance with increasing . Finally , we performed simulation experiments to understand the effects of varying the numbers of chromosomes with multiple probes . We created a simulated dataset of trees with eight probes where two pairs of probes each reside on two different chromosomes and the remaining four probes reside on four separate chromosomes . The probabilities of each of the SD , CD and GD events were fixed at , and , respectively . We report the results from this experiment in Table 3 , which compares the results from this experiment with our earlier result using only a single chromosome with two probes and four other probes located on separate chromosomes . The table shows that inclusion of the extra possible CD event results in higher accuracy for all the models except for the SD only model . The performance drop in the SD model is expected , as it would require more SD events to explain a greater number of missed CD events . The highest gain in performance is observed for SD+CD+GD model . These results show that our algorithm will tend to yield comparatively more advantage over the earlier work with more complicated scenarios of sharing probes across chromosomes , suggesting its utility will increase as improvements in technology allow for larger probe sets . We applied the algorithm to two sets of real data: Among the eight genes in the BC dataset , DBC2 and MYC reside on chromosome and HER-2 and TP53 reside on chromosome . The other four genes belong to distinct chromosomes . The oncogene Cyclin D1 ( CCND1 ) , which plays a role in many solid tumor types , is in both the BC and CC datasets . However , in some other tumor types , such as oral cancer , CCND1 is part of a larger region with recurrent copy number gains on chromosome and other nearby genes have also been suggested to play a role in oncogenesis [66] . We evaluated the SD+CD+GD method by its effectiveness in reducing the parsimony score ( total number of mutation events ) of the resulting trees relative to the prior SD-only model . With the primary CC samples , the SD+CD+GD method found a lower-cost tree in of cases , a tree of equal weight in cases , and a higher-cost tree in cases . In each case of increased weight , the increase was by and appears to result from the subtree regrafting heuristic used in handling GD events ( see Methods ) . These results suggest that the heuristic tree search may more often yield a suboptimal result for the SD+CD+GD model than it does for the SD-only model . The benefit of the more realistic model , however , outweighs the cost of this suboptimality in a large majority of instances . For trees derived from metastatic samples , of trees had lower weight for the full SD+CD+GD model and the remainder all had equal weight for the two models . Metastatic data sets tend to have fewer distinct cell types than do primary trees and thus may represent an easier optimization challenge . For the BC samples , of DCIS ( samples 1–13 ) and of IDC ( samples 14–26 ) had lower weight for the full model , with the remaining one sample having equal weight . Parsimony scores by tree are provided in Figures 5 and 6 . We next evaluated effects of the improved model on overall tree topology , based on results of our prior work [46] that tree topology can significantly distinguish trees drawn from distinct progression stages of a given tumor type , with possible implications for the varying balance of diversification and selection acting on different stages of tumor progression . Figure 7 quantifies the topology for each sample set based on fractions of cells inferred at each tree depth from to . The figure shows similar qualitative trends for both SD and SD+CD+GD methods , although with small quantitative differences . For example , both SD and SD+CD+GD trees recapitulate a tendency for CC primary trees to show relatively broad topology ( Figure 7 ( A ) ) while CC metastatic trees prune rapidly beyond the first few tree levels ( Figure 7 ( B ) ) . There is , however , an overall shift to lower depth in the SD+CD+GD trees . For CC primary trees , of cells are located in the first tree levels for SD versus for SD+CD+GD . For CC metastatic , of cells are located in the first tree levels for SD versus for SD+CD+GD . For BC , the comparable numbers of cells in depths are for SD versus for SD+CD+GD in DCIS and for SD versus for SD+CD+GD . These results suggest that the overall tree topology is not greatly sensitive to the combination of event types , although there is a noticeable shift towards lower depth in the full model . An additional evaluation was possible for the BC trees , because for the BC data , a probabilistic model and expert annotation based on two additional centromere probes made it possible to estimate the cell ploidy [36] , which we define as the mode among the number of copies of the twenty-two autosomal chromosomes in a cell . Each cell in that dataset is thus annotated with an expert-curated overall ploidy estimate . We used these ploidy estimates to validate our inference of GD events based on whether edges assigned to GD events in our trees correspond to doubling of annotated ploidy . The percentage agreement by edge between GD events and annotated doubling in ploidy is across DCIS trees and across IDC trees . In of all inferred GD events , at least one endpoint of the corresponding edge is a Steiner node , and the uncertainty among whether a GD event occurred prior to or after the emergence of the Steiner node may explain why the per-edge agreement is not higher . Nonetheless , the data support the conclusion that inferred GD events are correct in a majority of cases . As a final step , we repeated an approach developed in our prior work [46] to both validate the biological relevance of the trees and develop a practical application of them by treating the trees as sources of features for classification tasks applied to the CC data . For this purpose , we developed several sets of quantitative features based on inferred trees as well as comparative features derived from raw FISH probe counts . We used the following set of tree-based features: We omitted a third feature set , bin count , used in our prior work because it is not easily comparable between SD and SD+CD+GD trees . We compared these features to four features derived directly from FISH probe counts without reference to the trees: We used each feature set as input to the Matlab support vector machine ( SVM ) classifier with a quadratic kernel using rounds of bootstrap replicates per test with leave-one-out cross-validation to compute mean and standard deviation of accuracy . We used Matlab functions “svmtrain” and “svmclassify” for training and testing of the SVM classifier . We then applied these methods for three classification tasks: ( i ) distinguishing primary samples that progressed to metastasis from their paired metastatic samples , ( ii ) distinguishing all primary samples from all metastatic samples , and ( iii ) distinguishing primary samples that metastasized from primary samples that did not metastasize . The first two tasks are relevant to identifying features that help us understand the differences in evolutionary mechanisms of primary and metastatic samples . The third is intended to model an important practical problem in cancer treatment: determining whether a given primary tumor will metastasize . Figure 8 shows results on each task . For task ( i ) , allowing SD+CD+GD events increased accuracy relative to SD trees from to for edge counts and from to for tree level cell count . The SD+CD+GD tree level cell count was the most effective of all features , tree-based or not . For task ( ii ) , we similarly saw a substantial improvement in prediction accuracy for SD+CD+GD trees relative to SD trees . Classification accuracy improved from to for edge count features and from to for tree level features . In this case , both SD+CD+GD tree feature sets outperformed all other features sets , tree-based or otherwise . These results provide an indirect validation that using a more general tree model gets closer to the biological ground truth . For task ( iii ) , we saw no improvement , with identical results for SD and SD+CD+GD trees for either feature set . All tree-based feature sets significantly outperformed all non-tree-based feature sets for this task . We conclude that the more realistic evolutionary models appear not to reveal any more information to the classifiers for predicting which primary samples will go on to metastasize than the SD trees , which were already quite effective for that task . A key advantage of FISH for profiling tumor heterogeneity is that it makes it cost-effective to profile much larger numbers of cells than alternatives such as single-cell sequencing . To assess the practical importance of this advantage , we asked two related questions: ( 1 ) how many cells do we need per tumor to accurately reconstruct single-cell phylogenies and ( 2 ) how many tumors do we need to examine to identify reproducible , statistically significant features across trees . We first assessed the number of cells needed per tumor by using our first simulated dataset of trees described above with subsamples of varying numbers of cells per tumor , measuring reconstruction error of our SD+CD+GD algorithm with the weighted matching algorithm . The mean reconstruction errors calculated across cases for subsamples of , , , and cells were ( s . d . ) , ( ) , ( ) , ( ) , and ( ) respectively . We can thus conclude that accuracy improves noticeably with increasing numbers of cells to at least cells per tumor before plateauing at approximately error . We next assessed numbers of tumors needed to identify meaningful statistically significant properties of tumor classes by analysis of the CC paired and primary samples . We randomly subsampled from among the pairs and , for each subsample , calculated the following three tree statistics on progression trees inferred from our SD+CD+GD algorithm: We then compared distributions of each statistic on primary vs . metastatic trees by a Wilcoxon signed rank test . As the samples were selected randomly , no ordering among the samples was considered . Figure 9 shows the 1-sided p-values of the three statistical tests when the number of randomly selected samples are increased from to . The figure shows that ability to distinguish the two tumor subsets improves with increasing number of tumors . While the threshold for significance varies by statistic , each reaches weak significance ( p0 . 05 ) between and tumors . We can thus conclude that finding reproducible features distinguishing the tree types requires on the order of tens of tumors , at least for the candidate probe sets examined here . Taken together , these two results demonstrate that building accurate trees on a large enough scale to distinguish meaningfully primary from metastatic trees requires data sets with roughly the order of thousands of single cells ( hundreds of cells per tumor for tens of tumors ) , a scale of data that has so far been achieved only by FISH studies of tumor heterogeneity . We note , however , that one would expect these numbers to vary depending on the degree of tumor heterogeneity , the classes of trees one wishes to distinguish , and the specific markers examined .
This paper has presented novel theory and algorithms for reconstructing evolutionary trajectories of gene copy numbers in solid tumors in terms of a model of tumor evolution incorporating changes at the scale of single gene probes , full chromosomes , or all probes in the genome . We have derived algorithms to reconstruct maximum parsimony sequences of events , and thus estimates of evolutionary distance , between pairs of cells assayed by FISH probes . We have further incorporated these inferences into a method for building phylogenies of hundreds of cells in single tumors . These methods have been added to FISHtrees [46] , our software for inferring tumor phylogenies from single-cell copy number data . Experimental results on simulated data confirm the ability of the new methods to improve phylogenetic inference accuracy relative to simpler models by adding CD and GD events that model chromosome-scale and whole-genome copy number changes that are frequently observed in tumor evolution . Application to observed human tumor data shows that these extended evolutionary models are able to yield more parsimonious tree reconstructions and that the resulting trees lead to improved accuracy in prediction tasks related to diagnosis and prognosis . In future work , we hope to extend the theory developed here to handle even more realistic models and more challenging data types . One important direction will be advancing the theory developed here to improve upon the heuristic approximations used in the Steiner tree inference to better approach the goal of finding globally optimal trees for the most computationally challenging FISH data sets . The evolutionary models , likewise , might be further extended to go beyond the three mutational event types considered here to better approximate the numerous distinct mutational mechanisms by which copy number profiles of tumor cells might evolve . The data sets studied here do not include geographical information about locations of individual cells in the tumor , but other data sets for analyzing tumor heterogeneity do include such geographical information [38] , [68] . We expect it would be interesting to construct phylogenies with distance functions that combine spatial distance in three dimensions with combinatorial distance measures between the cell count patterns , as we have studied here . Further , while FISH for the moment retains a unique advantage in the large number of cells it can profile , one can reasonably anticipate that single-cell sequencing will eventually become practical for comparable cross-tumor studies . There would thus be value in extending the theory developed here to single-cell sequencing data , a goal that would pose substantial algorithmic challenges due to the much larger number and variety of markers it can reveal as well as the more complicated error models it would entail . Finally , we hope to make more use of these single-tumor phylogenetic models in clinically relevant prediction tasks and further explore the biological insights one can gain from more accurate tumor phylogenies .
We develop the theory for inference of the Steiner ( unsampled or extinct cell configurations ) nodes in the paths formed by the sequence of gene copy number gains and losses from an initial configuration to a final configuration . We first extend the prior theory to account for SD and CD events . Our model assumes that on division of a tumor cell , the configuration can change either by gain or loss of one copy of a single gene ( SD event ) or by gain or loss of one copy of each gene on a single chromosome ( CD event ) . For example , a configuration of four genes with the first two genes on the same chromosome might evolve in a single mutational event to by an SD event or to by a CD event . We propose Algorithm 1 , provided in Figure 10 , to calculate the minimum number of steps required to transform into considering SD and CD events , where , without loss of generality , we assume that the genes on a common chromosome have consecutive indices in . Algorithm 1 also identifies a minimum-length sequence of events , although this sequence is not necessarily unique . For example , if there are four genes on one chromosome and we want to get from configuration to configuration , then a shortest sequence of SD and CD events would be CD to , SD to , SD to , and SD to . Other orders of the same four events are also possible . The above example focuses on a single chromosome because as explained below , the problem of finding the shortest SD+CD path can be solved one chromosome at a time . We begin by establishing the following lemmas: We now extend the theory from the prior section to include SD , CD , and GD events . We assume in the proofs and discussion below that , where denotes lexicographical ordering . This assumption reduces the number of cases in several proofs . If instead , , the proofs are identical or symmetric except that GD events may be used in the wrong direction ( halving instead of doubling ) . The use of halving events is corrected heuristically by a procedure of subtree pruning and regrafting at line 24 of the pseudocode of Algorithm 3 , described below , and in FISHtrees . We will produce the complete proof by deriving a series of lemmas for three cases that together will cover all possible and : We provide an upper bound on the runtime of Algorithm 2 as a function of the number of genes and their copy numbers . Considering all three events , where , the maximum number of doublings required is , where denotes the copy number of the first gene where and . At each stage of the algorithm , the maximum number of nodes generated as a result of a operation is . SD and CD events are used to create each of those nodes in the case of an odd configuration . So , the maximum number of required operations is . Therefore , the number of operations performed during the execution of Algorithm 2 is . We implemented Algorithm 2 and integrated it with our approximate median-joining-based algorithm from our prior SD-only FISHtrees [46] code . The key steps of this algorithm are summarized in Algorithm 3 ( Figure 12 ) , which we describe at a high level here . The phylogeny algorithm first relies on Algorithm 2 to derive a matrix of pairwise distances between observed cell configurations , which are treated as states on a truncated integer lattice of dimension with a maximum value ( UB ) set to 9 in the current code . It then repeatedly samples triplets of nodes , identifying as potential Steiner nodes those that agree in each dimension with at least one of the triplet . Those Steiner nodes that lead to reduced minimum spanning tree cost are added to the node set , with the process is repeated until there is no further improvement . Finally a series of post-processing steps are performed to prune Steiner nodes that are not needed for the final tree and to apply subtree regrafting to correct for a potential source of suboptimality arising from the fact that the core phylogeny algorithm assumes symmetric distances but GD operations are asymmetric . Neighbor Joining ( NJ ) and Maximum Parsimony ( MP ) methods have been commonly used for building single-tumor phylogenies [16] , [54] and we therefore compared their accuracy to that of our own methods in inferring copy number phylogenies . We applied these two traditional phylogenetic tree building methods to build tumor progression trees using the individual copy number profiles as taxa and compared them with the trees built using our algorithms . We used implementations of both approaches in MEGA version 6 [69] . For NJ , we used Euclidean distances between cell copy number profiles to build the pairwise distance matrix . For MP , we treated copy number profiles of the genes in individual cells as sequences of arbitrary phylogenetic characters . We used the “Close-Neighbor-Interchange on Random Trees” search method . For the parameters “Number of Initial Trees” and “MP search level” , we used values of and respectively . | Cancer is an evolutionary system whose growth and development is attributed to aberrations in well-known genes and to cancer-type specific genomic imbalances . Here , we present methods for reconstructing the evolution of individual tumors based on cell-to-cell variations between copy numbers of targeted regions of the genome . The methods are designed to work with fluorescence in situ hybridization ( FISH ) , a technique that allows one to profile copy number changes in potentially thousands of single cells per study . Our work advances the prior art by developing theory and practical algorithms for building evolutionary trees of single tumors that can model gain or loss of genetic regions at the scale of single genes , whole chromosomes , or the entire genome , all common events in tumor evolution . We apply these methods on simulated and real tumor data to demonstrate substantial improvements in tree-building accuracy and in our ability to accurately classify tumors from their inferred evolutionary models . The newly developed algorithms have been released through our publicly available software , FISHtrees . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"algorithms",
"computer",
"and",
"information",
"sciences",
"cancer",
"genetics",
"mathematics",
"evolutionary",
"modeling",
"genetics",
"applied",
"mathematics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"evolutionary",
"biology",
"evolutionary",
"processes"
] | 2014 | Algorithms to Model Single Gene, Single Chromosome, and Whole Genome Copy Number Changes Jointly in Tumor Phylogenetics |
8-oxoG is one of the most common and mutagenic DNA base lesions caused by oxidative damage . However , it has not been possible to study the replication of a known 8-oxoG base in vivo in order to determine the accuracy of its replication , the influence of various components on that accuracy , and the extent to which an 8-oxoG might present a barrier to replication . We have been able to place a single 8-oxoG into the Saccharomyces cerevisiae chromosome in a defined location using single-strand oligonucleotide transformation and to study its replication in a fully normal chromosome context . During replication , 8-oxoG is recognized as a lesion and triggers a switch to translesion synthesis by Pol η , which replicates 8-oxoG with an accuracy ( insertion of a C opposite the 8-oxoG ) of approximately 94% . In the absence of Pol η , template switching to the newly synthesized sister chromatid is observed at least one third of the time; replication of the 8-oxoG in the absence of Pol η is less than 40% accurate . The mismatch repair ( MMR ) system plays an important role in 8-oxoG replication . Template switching is blocked by MMR and replication accuracy even in the absence of Pol η is approximately 95% when MMR is active . These findings indicate that in light of the overlapping mechanisms by which errors in 8-oxoG replication can be avoided in the cell , the mutagenic threat of 8-oxoG is due more to its abundance than the effect of a single lesion . In addition , the methods used here should be applicable to the study of any lesion that can be stably incorporated into synthetic oligonucleotides .
All DNA bases are subject to a variety of different types of damage due to reactive oxygen species ( ROS ) [1] . Among the most common and most mutagenic is 7 , 8-dihydro-8-oxoguanine , or 8-oxoG , which is mutagenic because of its tendency to pair with adenine and thus create GC to TA transversion mutations [2] , [3] . In the yeast Saccharomyces cerevisiae there are several mechanisms either to repair 8-oxoG lesions or to prevent 8-oxoG-induced mutations . 8-oxoG lesions opposite C , which would be formed by oxidative damage of double-stranded DNA , are removed by the glycosylase Ogg1 [4] , [5] , which has little if any activity on 8-oxoG paired with other bases [6] . Mismatch repair ( MMR ) plays an important role in preventing mutations due to oxidative damage [7] , and it has been shown that yeast MutSα , consisting of the Msh2 and Msh6 subunits , recognizes A replicated opposite an 8-oxoG lesion and thereby prevents mutations [8] . Thus in S . cerevisiae , MMR appears to replace the function of MutY , which is absent [8] , [9] . MutSβ , consisting of the Msh2 and Msh3 subunits appears to play no role in 8-oxoG repair [8] . For 8-oxoG lesions that are not removed prior to replication , translesion DNA synthesis ( TLS ) is importantly involved in bypass , with Pol η playing the major role in yeast . A variety of biochemical experiments using oligonucleotide templates with an 8-oxoG lesion have demonstrated that Pol η replicates through an 8-oxoG lesion , usually inserting a C [10]–[12] . This accuracy is explained by structural studies that show Pol η with a template containing an 8-oxoG lesion can hold the lesion in an anti conformation , permitting a C to be inserted [13] . In contrast , Pol δ is ten-fold less accurate and efficient in bypassing 8-oxoG [14] and Pol ε does not bypass 8-oxoG at normal dNTP concentrations , but does , inaccurately , at damage-induced levels of dNTPs [15] . Genetic studies are more complicated because the existence of an 8-oxoG lesion can be inferred only by its mutation signature , generally in an ogg1 background that greatly increases the amount of 8-oxoG in DNA . Such studies were used to show the involvement of MMR in preventing mutations due to 8-oxoG [8] , the role of Pol η in accurate replication of 8-oxoG [11] , [16] , and the lack of a significant role for Pol ζ [16] , [17] . The interplay of TLS and MMR is not completely clear . It was proposed that MMR was responsible for recruiting Pol η for bypass [18] but a detailed study of 8-oxoG bypass and repair concluded that Pol η acted independently of MMR [19] . It appears that monoubiquitination of PCNA is necessary for most TLS and in yeast this step is carried out by the Rad6-Rad18 heterodimer [20] , [21] . Genetic studies implicate RAD6 and RAD18 as well as RAD30 ( the gene encoding Pol η ) but not REV3 ( the gene encoding the catalytic subunit of Pol ζ ) in 8-oxoG tolerance [16] . Most TLS is assumed to occur at the replication fork [20] , [21] , although it can occur after S phase [22] . Furthermore , as there appear to be different replicative polymerases on the leading and lagging strands of replication [23] , one might expect 8-oxoG tolerance could exhibit strand differences . There are only a limited number of such studies . Using a reversion analysis of a URA3 mutation , it was found that 8-oxoG was preferentially repaired on the lagging strand of replication [24]; most of the differential repair was ascribed to the preferential activity of MMR on the lagging strand [25] . Using a mutation analysis of ogg1-dependent mutations in a SUP4-o reporter assay , the lagging strand bias of MutSα was observed , as well as a lagging strand bias for accurate Pol η bypass [19] . It is not clear what effect an 8-oxoG lesion has on replication . Some work has suggested that an 8-oxoG lesion has no effect on replication [10] , [18] , whereas a stall site was observed in vitro at a nucleotide prior to an 8-oxoG lesion with Pol δ but not Pol η [11] . An in vivo study inferred replication stalling or blockage from a mutational analysis [19] . Lesions that block or stall replication forks appear to be tolerated , especially in yeast , by homologous recombination [26] . Recent interest has focused on tolerance mechanisms by template switching in which a blocked 3′ end invades the replicating sister strand , either by a fork regression or strand invasion [26] . Such mechanisms of template switching appear to be dependent on polyubiquitination of PCNA by a complex of Ubc13-Mms2-Rad5 [20] . Because the substrate of Ubc13-Mms2-Rad5 is PCNA monoubiquitinated by Rad6-Rad18 , template switching would also be expected to be dependent on Rad6 and Rad18 [20] . In addition to its role in polyubiquitination , the helicase function of Rad5 may also be important in template switching [27] , [28] . Rather than using an ogg1 mutant background , a more direct method of analyzing 8-oxoG bypass in vivo would be to introduce DNA containing a defined lesion directly into cells . Plasmids containing a single-strand gap with an 8-oxoG or 8-oxoG in duplex DNA have been introduced into E . coli and mammalian cells [29]–[32] and a plasmid treated with methylene blue to induce oxidative damage was introduced into yeast for analysis [33] . The problem with the use of plasmids for analysis , in addition to the difficulty of substrate construction , is that the mechanism of replication may differ from that within the chromosome and various forms of recombinational bypass may also differ . Another approach would be to transform cells with single-stranded oligonucleotides ( oligos ) containing an 8-oxoG lesion . Transformation of yeast with oligos was first performed in Fred Sherman's laboratory [34] , [35] and the method has subsequently been used to study various lesions carried into yeast by oligos [36]–[41] . However , in most cases the lesion itself was responsible for generating a phenotype and with one exception [36] the mechanism of transformation with oligos was not fully understood . In order to study 8-oxoG bypass , we wanted to introduce the 8-oxoG on an oligo that would create a selectable phenotype that would be independent of the presence of the 8-oxoG lesion . Such an experimental design allows us to study both replication across the 8-oxoG and bypass of the 8-oxoG by template switching outside of a context of overall increased oxidative damage in the cell . As detailed below , we find evidence that the 8-oxoG lesion does stall replication; that only Pol η is able to replicate 8-oxoG accurately; that template switching is invoked frequently in the absence of Pol η; and that MMR strongly influences the outcome of 8-oxoG replication .
We had initially hoped to investigate damaged base bypass by transforming with an oligo which contained one normal base to revert the Trp- phenotype and another damaged base placed in a silent position where any base incorporation would be tolerated . However , our prior experiments [43] as well as a number of preliminary experiments indicated that we needed a method to mark incorporation of bases on both sides of the damaged base in order to be sure that we were observing bypass , and not partial incorporation of the relevant region of the oligo . These goals were accomplished by transforming with Oligos G and GO ( Figure 1 ) . The C at position 20 , highlighted in yellow , creates a Trp+ phenotype upon incorporation; the G at position 12 , highlighted in blue , if incorporated , creates a new SphI site . The G at position 15 is an 8-oxodG in Oligo GO and is highlighted in red , forming an 8-oxoG-G mismatch with the trp5-G148Cm sequence . Oligo G is identical , with a G instead of an 8-oxodG at position 15 . This mismatch was deliberately chosen , as one of the main glycosylases processing 8-oxodG , Ogg1 , should have little or no activity on an 8-oxodG-G mismatch [44] , [45] , and the efficiency of its removal by another glycosylase , Ntg1 , is low , if it exists [46] . In addition , 8-oxodG , when bypassed , is very unlikely to template a G , so if the original sequence at that position is maintained , that would be strong evidence either of removal of the 8-oxodG before replication , or a failure to bypass the 8-oxodG . The expectation for 8-oxodG bypass is that either a C or A is incorporated . If an A is incorporated opposite the 8-oxodG , a BfaI site is created , thus allowing a simple restriction digestion to indicate a mutagenic bypass of the 8-oxodG . In summary , at the site in question , a G on the coding strand indicates that 8-oxoG was either not used as a template for replication or was removed before replication , a C indicates that 8-oxoG was bypassed accurately , and an A indicates inaccurate replication of 8-oxoG . The overall design of the assay system and its expected results are illustrated in Figure 2A . Incorporation of the oligo can be selected by the Trp+ phenotype , and given that only 7 nt separate the base creating the Trp+ phenotype and the base creating an SphI site , we initially expected that all Trp+ cells should contain a new SphI site . What we found as analysis proceeded is that a fraction of oligos , even those containing all normal bases , exhibited “partial removal” as indicated in Figure 2A: in the presence of MMR , a substantial fraction of cells ( as much as 30% or more ) transformed by Oligo G ( containing only normal bases ) were Trp+ but did not contain an SphI site [43] . Those results were explained by a failure of MMR to recognize the C-C mismatch created by the oligo during MMR-directed excision from the 5′ end of the oligo [43] . Such results were seen only in the presence of MMR and with Oligo G and Oligo GO , but not with Oligo UG or UGO , as will be detailed below . It is in the second round in which the oligo sequence , now fully incorporated into the genome , is replicated for the first time . In Trp+ cells that were transformed by Oligo GO and contain the SphI site , DNA synthesis must have used the 8-oxoG for a template , and the base inserted can be subsequently analyzed . In the absence of MMR , all Trp+ cells would be expected to contain an SphI site , and that is true for cells transformed by Oligo G but not for all strains transformed by Oligo GO . The failure of cells transformed by Oligo GO to contain an SphI site could be explained by the process of template switching , in which the replication fork switches to use the newly replicated strand of the sister chromatid [26] . Strains with a variety of different genotypes were transformed by Oligo G and Oligo GO and assayed for the presence of an SphI site . The results for strains of the R orientation are shown in Figure 3A . Results for strains of the F orientation are shown in Figure S1 and the numbers of colonies analyzed for each strain are given in Table S1 . Because of the problem of partial oligo removal discussed above , strains with an active MMR cannot be analyzed for template switching ( i . e . Trp+ transformants lacking an SphI site ) with Oligo GO . In MMR-defective strains , with the exception of rad30 msh6 strains lacking both MMR and Pol η , the SphI site is created in almost all Oligo GO transformants . If the lack of the SphI site in that background is due to template switching , it should be blocked by loss of Rad5 , Mms2 , or Rad18 [20] , [27] , [47] , [48] . That is seen to be true , as rad5 rad30 msh6 , mms2 rad30 msh6 , and rad18 rad30 msh6 strains show minimal loss of the SphI site . Our previous results had suggested that placing a base creating an additional mismatch 3′ of the C-C mismatch in Oligo G would prevent the partial removal of the oligo illustrated in Figure 2A [43] . Therefore , we used Oligo UG and Oligo UGO ( Figure 1 ) to repeat a subset of the experiments shown in Figures 3A and S1 . The results assaying presence of the SphI site in both R and F strains are presented in Figure 3B . As observed with Oligo GO , Oligo UGO displays template switching in the absence of both MMR and Pol η ( Figure 3B ) . Loss of the SphI site is suppressed in rad5 rad30 msh6 strains . Oligo UG transformants in the presence of MMR showed little loss of the SphI site ( Figure 3B ) . Oligo UGO transformants in rad30 strains also demonstrated little SphI site loss , and were significantly reduced in template switching compared to rad30 msh6 strains ( Figure 3B ) . Therefore we can conclude that most template switching is suppressed by MMR . It appears in Figure 3B that the level of SphI site loss in rad30 Oligo UGO transformants is somewhat elevated compared to wild-type strains . The difference is not statistically significant in strains with the F orientation , and is only marginally significant ( P = 0 . 03 ) in the R orientation . 8-oxodG is considered to be extremely mutagenic due to the frequency of misreplication , with an A inserted opposite the 8-oxodG . In order to measure the bypass accuracy of the introduced 8-oxodG , we selected only those revertants that were both Trp+ and contained an SphI site , as all of those revertants should have incorporated the intervening 8-oxodG into the genome . As illustrated in Figure 2 , the 8-oxodG in oligos GO and UGO was placed opposite a G in the genome; thus removal of the 8-oxoG lesion would have resulted in retention of the original sequence at that point . Replication of the 8-oxoG lesion would be expected to yield only a C for accurate bypass or an A for inaccurate bypass , both leading to a change of sequence at that position . The replication accuracy could have been directly determined by sequencing each one of the revertants . However , as indicated in Figure 1 , inaccurate replication with an A creates a novel BfaI site , allowing a direct measurement of accuracy without sequencing . In order to assess the validity of this approach , we sequenced 129 Trp+ revertants that contained the introduced SphI site but lacked the BfaI restriction site and found 126 C , 1 T and 2 G at that position , thus confirming the utility of the restriction site assay and the assumption that a C would be found in such cases; 22 out of 22 sequences that contained the BfaI site had an A as expected . The result of this assay in a variety of genetic backgrounds using Oligo GO is shown for strains of R orientation in Figure 4A and for the F orientation in Figure S2 . All numbers are given in Table S1 . To our surprise , not only was replication extremely accurate in wild-type cells , it was also highly accurate in the absence of MMR , averaging 94% in MMR-defective strains of both orientations compared to 97% in wild-type strains; the difference is not statistically significant . The source of the accurate replication was clearly Pol η , as in MMR-deficient strains in the absence of Pol η , the accuracy dropped to 36% in R orientation and 44% in F orientation ( Figures 4A and S2; Table S1 ) . The resulting 8-oxoG-A mismatch was efficiently recognized and corrected by MMR , as the replication accuracy in rad30 strains was 92% in R and 93% in F ( Figures 4A and S2; Table S1 ) , neither of which was significantly different from that measured in wild type or MMR-defective strains . In order to confirm our results with Oligo GO , we conducted a reduced set of experiments with Oligo UGO ( Figure 4B; Table S1 ) . The accuracies measured in either rad30 or msh6 strains were not significantly different from each other . The double mutant combinations of msh6 rad30 were significantly lower , at 44% in F and 36% in R orientation . The graphs of accuracy in Figures 4 and S2 demonstrate that there are basically two categories of strains: those strains with deficient Pol η and MMR , and those that have at least one of the two pathways intact ( as discussed above , Rad18 is thought to be necessary for Pol η function , which is consistent with our results ) . In general within each group there is no statistically significant difference among strains , and there is a significant difference between strains in the two groups . It appears that rad5 strains could be an exception . In both the R ( Figure 4A ) and F ( Figure S2 ) orientations , accuracy in rad5 strains is lower than in wild-type , and accuracy in rad5 msh6 strains is lower than in msh6 strains . The P values are marginal , ranging from 0 . 01 to 0 . 04 , but the pattern is consistent in the four comparisons . Another question is whether there are differences between the accuracies observed in the two orientations of the TRP5 gene . The measured accuracy in R strains is lower than in F strains for Oligo GO in msh2 , msh6 , msh3 msh6 , and for Oligo UGO in msh6 strains . However , only by combining the results in msh2 , msh3 , and msh3 msh6 strains for Oligo GO does the difference approach statistical significance ( P = 0 . 05 ) . Because the 8-oxoG is replicated in the second round , replication of the 8-oxoG would be on the leading strand in strains with the R orientation .
We know that Trp+ revertants containing a new SphI site must have resulted from replication past the 8-oxoG and can therefore measure the accuracy of that bypass . 8-oxoG is considered to be a very mutagenic lesion; therefore it was somewhat of a surprise that in wild-type cells it was replicated quite accurately . The accuracy of replication was 98% in F strains ( replication on the lagging strand ) and 97% in R strains . Even more surprising was the accuracy in MMR-deficient cells; pooling data from all genotypes lacking MutSα ( msh2 , msh6 , msh3 msh6 ) gave 96% accuracy in F strains and 91% in R . As noted above , this difference is of marginal statistical significance ( P = 0 . 05 ) , but it is in agreement with experiments that showed a lagging strand bias for Pol η [19] . One important distinction between our measurement and other in vivo measurements is that our strains contain one 8-oxoG lesion above the background level of such lesions , whereas most other measurements have been made in Ogg1-deficient strains in which one would expect large numbers of additional 8-oxoG lesions . Given that the amount of MutSα in cells is low ( one estimate is 1230 molecules of Msh2p and 5330 molecules of Msh6p per cell [49] ) , elevated levels of 8-oxoG in the cell could potentially titrate out MutSα . The source of accurate replication of 8-oxoG in MutSα-deficient strains is clearly Pol η , as in cells deleted for Pol η , 8-oxoG is replicated accurately only 40% of the time ( Figure 4 and Table S1 ) . Pol η is also in relatively low abundance in yeast ( an estimated 1860 molecules per cell [49] ) again suggesting a potentially misleading picture of 8-oxoG replication in cells with elevated levels of 8-oxoG . Thus our measurements indicate the levels of accuracy to be expected for repair of spontaneous levels of oxidative damage in normal conditions , but might not apply for cells under oxidative stress or with reduced levels of MMR or Pol η . Given that the accuracy of bypass in the absence of MMR was due to Pol η , our results support the independence of MMR and Pol η in maintenance of accuracy , as previously observed [19] , and are not consistent with a model in which MMR is required to recruit Pol η [18] . Previous in vitro measurements had determined that yeast Pol η could replicate an 8-oxoG much more accurately than could Pol δ [11] , [14] , but it is impossible to extrapolate from such in vitro experiments using single DNA polymerases to an in vivo situation in a chromosomal context with multiple DNA polymerases available . The low accuracy observed in MMR- and Pol η-defective strains ( 40% ) suggests that no other DNA polymerase in the cell is able to replicate 8-oxoG accurately . Therefore the high accuracy of 8-oxoG replication observed in MMR-defective strains indicates that most of the 8-oxoG replication must be due to Pol η . ( Suppose Pol η is used for 75% of 8-oxoG bypass , and other polymerases with only 40% accuracy are used the rest of the time; the overall accuracy of bypass in the absence of MMR would be approximately 85% , considerably lower than what we observe . ) In vitro experiments found a strong stall site with Pol δ just before replication of an 8-oxoG [11]; such a stall is likely the signal responsible on either replication strand for switching to synthesis by Pol η or switching templates . Particularly because of the different abilities of Pol δ and Pol ε to bypass an 8-oxoG in vitro [14] , [15] , one might have expected a difference in replication fidelity due to replication of leading and lagging strands by different DNA polymerases [23] , although we see little evidence for that here . The similarity in accuracy of leading and lagging strands in the absence of Pol η is surprising , particularly given the expected difference in the ability of Pol δ and Pol ε to bypass 8-oxoG . Our data suggest that the 8-oxoG lesion causes a stall; it has been hypothesized that a lesion on the leading strand could induce a switch to continued synthesis on the leading strand by Pol δ [50] and so it is possible that the synthesis observed on either strand across 8-oxoG in the absence of Pol η could be due primarily to Pol δ . The design of these experiments made it possible to observe template switching , in which the replicating DNA polymerase uses DNA from the replicating sister chromatid as a source of template [26] . In strains deficient in MMR , template switching was observed only in strains lacking Pol η ( Rad30 ) , and in those cases , occurred in about half of the replication events on both the leading and lagging strands of replication ( Figures 3 and S1 ) . As expected for template switching , these events were not observed in strains lacking Rad5 , Mms2 , or Rad18 . Template switching was measured as Trp+ revertants that did not contain the SphI site introduced by the oligo . As explained above , experiments with Oligo GO could not examine possible template switching in the presence of MMR because of the loss of part of the oligo during transformation in some Trp+ colonies ( Figure 2A ) . Such partial loss was not observed with Oligo UGO , and those experiments showed that MMR suppresses template switching , as such events were significantly lower in rad30 strains compared to rad30 msh6 strains in both orientations ( Figure 3B ) . It is possible that some template switching events were missed in these assays , as only 4 nucleotides separate the 8-oxoG from the C needed to produce Trp+ cells . If in template switching , there is loss of more than 4 bases from the 3′ invading end , the resulting strain would be Trp- and therefore not observed . Template switching could occur via fork regression on the leading strand [26] , but on the lagging strand must occur via a mechanism involving homologous recombination [47] . What triggers template switching and how does MMR suppress template switching ? A stall by the replicative DNA polymerase in advance of the 8-oxoG is a strong candidate for a template switching signal . As postulated above , as a replicative DNA polymerase encountered an 8-oxoG , it would stall and either induce a switch to Pol η replication or the replicative polymerase would switch templates and bypass the lesion in an error-free manner . In the presence of MMR , presumably the same template switching would occur but when the DNA copied from the sister chromatid was brought back to pair with the template strand , MMR would recognize the 8-oxoG-G mispair and initiate removal of the newly synthesized DNA , thus abolishing the effect of the template switch . It is evident from Figures 3A and S1 that deletion of either RAD5 or MMS2 blocks template switching . However , deletion of RAD5 in either wild-type or msh6 strains appears to somewhat decrease accuracy in Oligo GO transformants , whereas deletion of MMS2 in the same strains does not ( Figures 4A and S2 ) . That result is consistent with a role for Rad5 in addition to template switching [51] . A rad5 strain is more sensitive to UV damage than an mms2 strain [52] and a role for Rad5 was observed in TLS of UV damage independent of Mms2-Ubc13 [53] . Although on many substrates the Rad5-dependent TLS might be mutagenic [54] , it would appear from our work that the Rad5-mediated events are accurate in replicating 8-oxoG . The sequence context of a lesion can affect its fate . The trp5-G148Cm strains we have made could accommodate a lesion at several different positions , and thus somewhat different sequence contexts . Another option would be to place a lesion in a sequence context of choice and integrate it into a strain that would select for the loop integration [36] . The potential difficulty with such methods is that the spontaneous background of such reversion events , particularly in the absence of MMR , is considerably higher than in the trp5-G148Cm strains . The use of oligos to place defined DNA damage at unique places in the chromosome is potentially very informative . With proper markers in the oligos , we have shown that the fate of a defined lesion can be measured in a completely normal chromosome context in a variety of genetic backgrounds .
The trp5-G148Cm mutation was created as described for the other trp5 mutations [42] using delitto perfetto [55] and created the sequence CGATGTTATCCAACTGGGA starting at position 138 of TRP5 with mutated bases underlined . The lys2CT1265GA mutation was similarly created by delitto perfetto . The genotypes of strains used in these experiments are given in Table S2 . All gene deletions were created by one-step disruption with PCR generated fragments . In general gene deletions were made from a PCR fragment generated from the collection of yeast gene deletions [56] . The kanMX4 resistance marker was changed to hphMX4 or natMX4 by transformation with a fragment from pAG32 or pAG25 , respectively [57] . For the msh6Δ::loxP deletion , the PCR fragment was from a strain in which MSH6 had been disrupted by a loxP-kanMX-loxP fragment that was subsequently excised by Cre expression [58] . Transformation was a modification of the method used previously [43] , [59] . An overnight culture of a strain was diluted 1∶50 in YPAD [60] , incubated with shaking at 30° to an OD600 of 1 . 3–1 . 5 , washed twice with cold H2O , and once with cold 1 M sorbitol . After the final centrifugation , all solution was removed from the cells and a volume of cold 1 M sorbitol equal to that of the cell pellet added to resuspend the cells . For a typical transformation , 200 pmol of a Trp oligo and 200 pmol of LYS2TCARev40 ( used to revert the lys2CT1265GA mutation ) was added to 200 µl of this cell suspension in a 2-mm gap electroporation cuvette , and the mixture electroporated at 1 . 55 kV , 200 Ω , and 25 µF ( BTX Harvard Apparatus ECM 630 ) . Immediately after electroporation , the cell suspension was added to a volume of YPAD equal to that of the initial culture , and the cells incubated at 30° with shaking for 2 h . Cells were then centrifuged , washed with H2O , and plated on synthetic dextrose ( SD ) medium lacking either tryptophan or lysine [60] to select transformants . The number of Lys+ transformants served as a useful guide that a particular transformation experiment had worked , but was not correlated well enough with the number of Trp+ transformants to be used as an internal control ( results not shown ) . Individual Trp+ revertants were picked into 200 µl SD-Trp medium in 96-well deep well plates , grown overnight at 30° with shaking , a small aliquot of each transferred to fresh SD-Trp medium with a Boekel Microplate Replicator and grown overnight , and finally transferred with the replicator to another deep well plate for overnight growth in 300 µl YPAD . Cells were then transferred with the replicator to a PCR microplate containing 15 µl per well of 2 mg/mL Zymolyase 20T ( USBiological ) in 0 . 1 M Phosphate Buffer pH 7 . 4 and incubated at 37° for 30 min and 95° for 10 min . After incubation , 85 µl H2O was added to each well . PCR was performed using 5 µl of the lysate in a total volume of 50 µl of the recommended buffer with 0 . 3 µM trpseq2 and trpseq8 primers [43] and 0 . 5 µl Takara e2TAK DNA polymerase for 30 cycles . For restriction digestion , 5 µl of the PCR reaction was incubated with 2 units of either BfaI or SphI ( New England Biolabs ) in the recommended buffer in a total volume of 15 µl at 37° overnight and analyzed by gel electrophoresis . The number of revertant colonies analyzed for each strain and oligo are given in Table S1 . For each combination , usually 48 colonies were analyzed; in some instances experiments were repeated multiple times . For experiments repeated three or more times , means and standard deviations are shown in the figures . For comparison of results between strains , all data from a given strain and oligo were combined . Statistical calculations were performed using the VassarStats website ( http://www . vassarstats . net/ ) . In most cases , the number of samples was large enough for use of a chi-square test; in the remainder of cases , a Fisher's exact test was used . P values are given for each comparison in the relevant figure . | In the course of normal cellular functions , many types of reactive oxygen species are produced that can lead to oxidative damage in the cell . DNA bases are subject to the formation of various oxidative lesions; one of the most common is the production of 8-oxoG which can pair relatively well with A instead of C , leading to GC→TA transversion mutations . In this work , we have been able to place a single 8-oxoG in the yeast chromosome and observe its replication . We find that in a wild-type cell its replication is surprisingly accurate due primarily to two components: DNA mismatch repair , which recognizes an A inserted opposite the 8-oxoG and initiates its removal and subsequent re-replication; and translesion synthesis in which the existence of the 8-oxoG induces a switch from a normal replicating DNA polymerase to a specialized DNA polymerase , Pol η , that can accurately replicate an 8-oxoG . We also find that the 8-oxoG can cause the replicating strand to shift to the newly replicated strand in the sister chromatid , thus avoiding the 8-oxoG lesion . These findings indicate not only how cells deal with a known DNA lesion , but also demonstrate how other such lesions can be studied in the future . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"mutagenesis",
"biochemistry",
"model",
"organisms",
"genetic",
"mutation",
"dna",
"replication",
"nucleic",
"acids",
"genetics",
"dna",
"dna",
"repair",
"biology",
"yeast",
"and",
"fungal",
"models",
"saccharomyces",
"cerevisiae"
] | 2013 | In Vivo Bypass of 8-oxodG |
In evolutionary games , reproductive success is determined by payoffs . Weak selection means that even large differences in game outcomes translate into small fitness differences . Many results have been derived using weak selection approximations , in which perturbation analysis facilitates the derivation of analytical results . Here , we ask whether results derived under weak selection are also qualitatively valid for intermediate and strong selection . By “qualitatively valid” we mean that the ranking of strategies induced by an evolutionary process does not change when the intensity of selection increases . For two-strategy games , we show that the ranking obtained under weak selection cannot be carried over to higher selection intensity if the number of players exceeds two . For games with three ( or more ) strategies , previous examples for multiplayer games have shown that the ranking of strategies can change with the intensity of selection . In particular , rank changes imply that the most abundant strategy at one intensity of selection can become the least abundant for another . We show that this applies already to pairwise interactions for a broad class of evolutionary processes . Even when both weak and strong selection limits lead to consistent predictions , rank changes can occur for intermediate intensities of selection . To analyze how common such games are , we show numerically that for randomly drawn two-player games with three or more strategies , rank changes frequently occur and their likelihood increases rapidly with the number of strategies . In particular , rank changes are almost certain for , which jeopardizes the predictive power of results derived for weak selection .
In evolutionary theory , weak selection means that differences in reproductive success are small . If fitness differences are close enough to zero , perturbation analysis allows to derive analytical results in models of population dynamics . This approach has a long standing history in population genetics , where selection is typically frequency independent [1]–[4] . More recently , weak selection has been introduced into evolutionary game theory [5] . If selection is weak , the outcomes of the game have only a small impact on fitness . Possible interpretations of this assumption include that the effects of the game under consideration are small or that it represents only one of many factors influencing reproductive success . A number of important analytical results have been derived using weak selection as well as rare mutations in finite populations [6]–[8] . In infinitely large populations , the intensity of selection merely results in a rescaling of time , but does not affect the outcome of the evolutionary dynamics [9] , [10] . This means that long-term results under weak selection equally hold for arbitrary intensities of selection , provided the population is infinitely large . For finite populations it has been suggested that results obtained under weak selection may remain valid when the selection intensity is no longer weak [6] , [8] . Here , we show that in general this is not the case . If population size is finite , the intensity of selection plays a decisive role and can qualitatively change the outcome . Let us illustrate this idea with an example . Consider the public goods game discussed in [11] . Therein , individuals are chosen from a population of size to play a public goods game . Individuals choose whether to contribute a fixed amount to a common pool at a cost . The amount in the common pool is multiplied by a positive factor ( ) and distributed amongst all participants . The game considers three strategies: Cooperators , who contribute a fixed amount to a common pool , defectors , who do not contribute but benefit from the contributions of others , and punishers , who contribute and pay a cost to impose a fine upon defectors . The game is devised to inspect the emergence of altruistic punishment , a behavior commonly found in human subjects [12] . The model assumes a standard Moran process [5] , in which one individual is chosen proportional to fitness to reproduce and its offspring replaces a randomly chosen individual . Fitness is an increasing function of the payoff from the game , , where is the intensity of selection [13] . In addition , there is a small rate of mutations , such that a new mutant either goes extinct or reaches fixation before the next one occurs [14] , [15] . This allows to approximate the dynamics by an embedded Markov chain on the monomorphic states , with fixation probabilities describing the transitions between those monomorphic states . The stationary distribution of this Markov chain allows to infer the relative abundance of different strategies . This approach is used frequently to describe evolutionary games in finite populations with more than two strategies [11] , [16]–[19] . Figure 1 shows the strategy abundance in an imitation process for this public goods game with punishment . Panel illustrates the outcome when payoffs are mapped onto fitness with an exponential function , , where is the intensity of selection [20] . For weak selection altruistic punishment is the strategy most favored by selection , but this is not true for stronger selection . Moderate intensities of selection change the picture in favor of defection . This also holds when payoffs are mapped into fitness with the linear function [5] , as shown in Panel . Changes in the ranking of strategies also occur for larger strategy sets [11] , [17]–[19] but for a concise illustration of our point three strategies are sufficient . In the example above , focusing only on the weak selection leads to results that do not even qualitatively hold for higher intensities of selection . The change in the order of strategies shows that , in this case , the predictive power of weak selection to higher intensities of selection is limited . However , many results on the selection of strategies are based on weak selection [8] , [21] . In particular , in the context of the evolution of cooperation , simple analytical results derived under weak selection are popular [6] , [7] , [22]–[25] . However , based on the above example a number of questions arise: Are changes in the ranking of the frequencies of strategies a common occurrence as selection increases ? What facilitates the change of ranks ? The number of players ? Or the number of strategies ? Or does it depend on specific assumptions on the evolutionary dynamics ? To answer these questions , we formally study imitation processes in symmetric games . Our results show that for games with two strategies , the ranking in strategy abundance can change with the intensity of selection , provided the number of players is more than two . Moreover , rank changes also arise in pairwise games with more than two strategies , and it is even highly likely in games with many strategies .
The abundance ranking of strategies is invariant under changes of the selection intensity in games for any imitation process [30] . However , for the Moran process with arbitrary payoff-to-fitness mappings , this does not necessarily hold: For example , in a Moran process with linear payoff-to-fitness mapping , such effects can appear in games with negative payoff entries when the intensity of selection approaches its maximal value , as the transition probabilities can approach zero in this case , leading to rapid changes of the fixation probability . Figure 2 shows that rank changes can readily arise for simple imitation processes in games with three players , i . e . , the minimal group size of multiplayer games . In this example , the ranking derived under weak selection carries over to any selection intensity for the Fermi imitation function , . But for the rescaled error function , which represents a qualitatively similar imitation function , the ranking changes . It turns out that for any two-strategy game , the ranking invariance holds for the imitation function , as a result of the special property . For the imitation function , however , the criterion to determine that strategy 1 is more abundant under weak selection differs from that under strong selection . Section 3 of the SI shows technical details of these results . Why do similar functions lead to radically different results when selection is not weak ? The intuition behind is as follows: As shown in the SI , the stationary distribution depends only on the product . Here is the payoff difference between strategy 1 and strategy 2 , where is the number of strategy 1 individuals in the population . The ranking can change when both the product in the enumerator and the product in the denominator converge to zero with increasing intensity of selection . In this case , not the imitation function , but its first derivative or potentially its higher derivates far from zero matter , based on L'Hopital's rule . In the SI , we show that even the monotonicity in the payoff difference cannot ensure the invariance of ranking for any two-strategy game and any imitation function ( see Section 3 in SI ) . Yet this monotonicity applies for all games , where the invariance property holds for any imitation function [30] . Therefore , we conclude that in general , ranking invariance does not hold for two-strategy games with arbitrary imitation processes . Since such multiplayer games have only become popular recently [31]–[36] , this result may not be particularly surprising . However , in the next Section we show that even for games between two players , ranking changes can occur . For games with more than two strategies , i . e . , , the problem is harder to tackle , because the stationary distribution does no longer depend on a single ratio of fixation probabilities , but becomes a more intricate rational function of all fixation probabilities , see e . g . [11] . At first , we restrict ourselves to games and show that weak selection results do not carry over to stronger selection . Numerically we establish that this phenomenon occurs very often in the case of games with randomly drawn payoff matrices . An example in which the ranking of strategies changes with the intensity of selection was already provided in the introduction . To go one step further , we provide a theorem for a more challenging constraint in which the limits of both weak and strong selection are identical , yet rank changes occur at intermediate selection strengths . Theorem 1 Consider any imitation process with a strictly increasing , twice differentiable imitation function . For a sufficiently large population size and any selection intensity , there exists a payoff matrix with the following two properties: Theorem 1 states that weak selection results cannot be extrapolated to non-weak selection for games ( for a proof by construction see Section 4 of the SI ) . This implies that the ranking of strategies under weak selection has limited predictive value for higher intensity of selection . The theorem also shows that even if both weak selection and strong selection limits lead to the same evolutionary outcome , the ranking of strategies can still change at an intermediate selection intensity . This precludes the robustness of conclusions based on both the weak selection approximation and the strong selection . In order to determine how frequent such rank changes occur or how generic these games are , we analyze changes in the ranking of strategies in random games [36]–[39] . In particular , we compute the probability that rank changes occur and determine the number of changes in the rank of strategies , see Figure 3 . The numerical procedure generates a random matrix , where each entry is drawn independently from a Gaussian distribution with zero mean and variance one or a uniform distribution over the interval . Strictly speaking , our numerical results are restricted to these two sampling distributions for the payoffs . However , the results suggest that the distribution has only a small influence on the number of rank changes as shown in Figure 3B . We compute the strategy abundances for an imitation process in the interval for in , where is chosen maximally while preventing numerical overflows . We then count the number of rank changes between all pairs of strategies . Note that the proof of Theorem 1 shows implicitly that in random games a simultaneous rank change of all three strategies occurs with probability measure zero . This is because these games are located on a subspace with a lower dimension than the space of games with intersections of pairs of strategies ( see Section 4 . 1 in SI ) . Figure 3A shows an example where a randomly generated game results in four rank changes . This illustrates that the ranking obtained for weak selection cannot be used to extrapolate to non-weak selection . The commonness of rank changes is estimated based Monte Carlo simulations , see Figure 3B . With a probability greater than at least one rank change occurs but the likelihood decreases rapidly with the number of rank changes . The numerical approach shows that the construction provided in Theorem 1 is relevant for a substantial fraction of random games and does not merely represent a non-generic , special case . It also shows that a larger number of rank changes may occur as illustrated in Figure 3 . Theorem 1 states that games exist in which the strategies change their ranking in abundance . Naturally this also holds for games with more strategies . To determine the probability and numbers of such rank changes in random games [36]–[39] , we generalize the numerical procedure described above . Games with more than strategies increase in complexity and , as expected , increasing leads to more rank changes . Let be the probability that at least changes in the abundance ranking occur in random games . Figure 4 shows that increases rapidly with the number of strategies if we assume that the entries of the payoff matrix are sampled from either a uniform or a Gaussian distribution . For , the probability that the ranking derived under weak selection is not valid for higher selection intensity already exceeds one half . For , it is almost . The numerical investigation of random games shows that with many available strategies , the stationary distribution computed for weak selection can be very different from the stationary distribution obtained for larger intensities of selection . This is of particular relevance in applications where behavioral diversity is important [18] , [26] . Similarly , the expected number of rank changes for random games also increases with , see Figure 4 . In particular , for , the expected number of rank changes is already more than one . Hence , for games with many strategies it is very likely that the stationary distribution obtained under weak selection is qualitatively quite different from the stationary distribution obtained for stronger selection .
For two-strategy multiplayer games in well-mixed populations under small mutation rates [14] , [15] , we have shown that the ranking of the average strategy abundance derived for weak selection may change when increasing selection strength . Moreover , the ranking is sensitive to the details of the evolutionary process , such as the choice of imitation functions . In evolutionary games in finite populations the assumption that mutation rates are sufficiently rare to consider pairwise invasions between strategies is popular [11] , [17] , [26]–[29] , [40] and often the only analytically feasible approach . However , it remains challenging to interpret the analytical results for the stationary distribution for all selection intensities [11] . Therefore , weak selection approximations [5] , [22] or strong selection limits [17] , [19] , [27] are often used to obtain simpler analytical results that are easier to interpret . Here , we have shown that already for games , attempts to extrapolate results derived in one of those simplifying cases may often fail because even the qualitative features of the stationary distribution , i . e . the ranking of strategy abundances , may change as a function of the selection strength . In particular , the strategy with highest abundance may change with the intensity of selection . In fact , even considering the two limiting cases of the selection intensity together is not enough . Our results show that even if weak and strong selection limits lead to the same ranking , other rankings can still arise for intermediates selection intensities . Thus , we conclude that even though these extreme cases are insightful , abundances at intermediate selection intensity levels have to be considered as well to establish the generality of the results and robustness of the conclusions . An intuitive reason for changes in the abundance ranking of strategies for games in both the weak and strong selection is based on risk dominance . For strong selection , the pairwise probability current always flows towards the risk dominant strategy [14] , whereas for weak selection , the average abundance is based on the sum of the risk dominance conditions between all different strategies [41] . We have focused solely on well-mixed populations and our analytical considerations cannot easily be generalized to structured populations . However , several papers on the evolution of cooperation have shown that the ranking of the average abundance of strategies can change in structured populations even in games [42]–[45] . Thus , this issue is also of interest in structured populations , where the weak selection approximation is particularly powerful [21] , [24] , [25] , [46] , but for example fails to predict the potential decrease of cooperation in the spatial snowdrift game [42] . Our results have been obtained for imitation processes , i . e . processes in which one individual probabilistically compares its performance to another one and tends to adopt strategies of better performing members of the population . The results derived for three or more strategies assume rare mutations such that the transition matrix of the embedded Markov chain only depends on the fixation probabilities of pairs of strategies . Therefore , all our results immediately carry over to the Moran process with exponential payoff-to-fitness mapping [20] , [47] , because such a Moran process has the same fixation probability as the imitation process with imitation function for any intensity of selection [20] . This fact illustrates that the existence of such rank changes do not depend on the details of the microscopic evolutionary process , but are a generic feature of evolutionary games in finite populations .
We assume a finite well-mixed population of size . For two player games , individuals interact in pairs according to a symmetric game given by the matrix where denotes the number of strategies . A player with strategy playing against strategy gets payoff . Payoffs are computed for every individual assuming everyone interacts with everyone else in the population . For the multiplayer case , we follow the notation from [36] . Selection acts by comparing the payoffs of two randomly chosen individuals . Individual with payoff adopts the strategy of individual with payoff with probability , where is called the imitation function . In an evolutionary process individuals must be more likely to imitate a strategy that performs better and hence we assume that is increasing , for all , and for technical reasons we require that is continuously differentiable ( with the exception of the proof of Theorem 1 , which requires twice continuous differentiability ) . This implies that strategies achieving higher payoffs have a higher probability of being represented in the next generation . The intensity of selection is . If approaches zero , payoff differences have vanishingly small effects on selection . We also assume that and , which means that for infinite intensity of selection , only the sign of matters . Variation in the population is generated by mutations . That is , the imitation step described above happens with probability . With probability , a mutation occurs and the focal individual adopts a uniformly chosen strategy . Without mutations it is possible to compute the fixation probability , of a mutant playing strategy in a population of individuals playing [5] . For small , the dynamics is approximated by an embedded Markov chain [14] , [15] with an transition matrix that is fully determined by the different fixation probabilities ( Section 2 SI ) . In a large class of evolutionary processes ( where all transitions between states are possible ) , the transition matrix has a unique stationary distribution for every . More precisely , the stationary distribution is unique whenever the Markov chain is irreducible and aperiodic , and characterizes the average abundance of each strategy in the long run [48] , [49] . The evolutionary outcome we are interested in is the ranking based on the average abundance on the set of strategies . The stationary distribution of is the the uniform distribution for . For weak selection , we obtain the ranking over the strategies by ordering the derivatives of the components of the stationary distribution at . For the computational results , we determine the strategy abundances for a random game as a function of the selection strength , , over the interval , where is dynamically adjusted to avoid arithmetic overflow . We then count the number of rank changes between any pair of strategies , i . e . changes in their relative abundance for in . Random games are constructed by sampling payoffs from an independent identical distribution , which is either uniform or Gaussian . Averages are taken over samples in all cases . Our source code in Python is publicly available on figshare ( http://dx . doi . org/10 . 6084/m9 . figshare . 814470 ) . | In evolutionary game dynamics in finite populations , selection intensity plays a key role in determining the impact of the game on reproductive success . Weak selection is often employed to obtain analytical results in evolutionary game theory . We investigate the validity of weak selection predictions for stronger intensities of selection . We prove that in general qualitative results obtained under weak selection fail to extend even to moderate selection strengths for games with either more than two strategies or more than two players . In particular , we find that even in pairwise interactions qualitative changes with changing selection intensity arise almost certainly in the case of a large number of strategies . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [] | 2013 | Extrapolating Weak Selection in Evolutionary Games |
Infection with Schistosoma spp . affects more than 258 million people worldwide . Current treatment strategies are mainly based on the anthelmintic Praziquantel , which is effective against adult worms but neither prevents re-infection nor cures severe liver damage . The best long-term strategy to control schistosomiasis may be to develop an immunization . Therefore , we designed a two-step Schistosoma mansoni infection model to study the immune-stimulating effect of a primary infection with either male or female cercariae , measured on the basis of TH1/TH2-response , granuloma size and hepatic fibrosis after a secondary bisexual S . mansoni challenge . As a first step , mice were infected with exclusively female , exclusively male , or a mixture of male and female S . mansoni cercariae . 11 weeks later they were secondarily infected with male and female S . mansoni cercariae . At week 19 , infection burden , granuloma size , collagen deposition , serum cytokine profiles and the expression of inflammatory genes were analyzed . Mice initially infected with female S . mansoni cercariae displayed smaller hepatic granulomas , livers and spleens , less hepatic fibrosis and higher expression of Ctla4 . In contrast , a prior infection with male or male and female S . mansoni did not mitigate disease progression after a bisexual challenge . Our findings provide evidence that an immunization against S . mansoni is achievable by exploiting gender-specific differences between schistosomes .
The blood flukes of the genus Schistosoma spp . are among the world's most prevalent human helminthic parasites . According to the WHO over 258 million people are currently receiving preventive therapy , mostly to avoid severe long-term liver damage [1] . During their life-span of up to 15 years , schistosomes produce a myriad of tissue-damaging eggs [2] . Entrapped within the intestinal wall and small liver sinusoids they provoke an inflammatory , granulomatous reaction that is mainly caused by CD4+ T cells of the subtype 2 and alternatively activated macrophages [3] . This repair response suppresses initial TH1 inflammation but results in hepatic fibrosis ( e . g . Symmer's pipe stem fibrosis ) , portal hypertension and its clinical sequelae , ascites and esophageal varices [4] . S . mansoni infection triggers a transient T-helper-1 ( TH1 ) cell reaction mediated by proinflammatory cytokines such as IFN-γ , TNF-α , IL-12 , and iNOS . Following the onset of egg production , the inflammatory TH1-response shifts towards a profibrotic TH2-response mediated by IL-4 and IL-13 [5–7] . However , this TH1/TH2 dogma is not as stringent as formerly supposed , since it has been shown that isolated S . mansoni eggs and soluble egg antigens suffice to induce a TH2-response in mice [7–9] . Moreover , the discovery of IL-4-responsive macrophages before the onset of egg production indicates the presence of a TH2-dominant milieu from early on [10] . Loss of worm integrity leads to a strong release of antigens and thus results in a certain resistance to re-infection . This protection can be hastened factitiously by the killing of adult worms with Praziquantel , but also occurs when worms die naturally [11 , 12] . A number of studies have looked into the generation of immunity against adult schistosomes using sterile , unisexual infection models . It is known that soluble S . mansoni worm antigens ( SWA ) sensitize mice to granuloma formation , and when injected into the tail vein of naive mice , parasite eggs form perioval granulomas that are smaller in size and which differ in cellular composition to the granulomas found in mice pre-infected with either male or female cercariae [13] . In the studies in question , single-sex infection with male Schistosoma cercariae led to pronounced organ changes ( increased liver and spleen weight ) , delayed-type hypersensitivity and higher numbers of peripheral blood cells in mice [14] , whereas single-sex infection with female Schistosoma cercariae increased antibody response in baboons [15] . In contrast , cell-mediated immunity was observed in splenocytes isolated from mice infected with both sexes of cercariae compared to single sex infected mice [16] . However , protection after unisexual infection was not achievable when reinfection was performed more than 6 weeks later [17] . These discrepancies led us to revisit the issue of unisexual infection , placing special emphasis on the expression of proinflammatory and profibrotic markers . To this end we designed a two-step Schistosoma mansoni infection model measuring TH1/TH2-response , granuloma size and hepatic fibrosis .
All experiments were performed according to German animal protection regulations and approved by the local committee on animal care and use ( 7221 . 3–1 . 1-008/13 ) . Different stages of Schistosoma mansoni ( Belo Horizonte strain ) were kept using Biomphalaria glabrata freshwater snails ( B . glabrata , Brazilian strain ) as intermediate hosts and 6–8 week-old female NMRI mice as definitive hosts , as previously described . In brief , cercariae were obtained by using light exposure to trigger mass shedding , and the number of cercariae/ml was determined using a conventional light microscope ( 100-fold magnification ) . Mice were kept on a 12:12 hour light/dark cycle and given standard mouse chow ( SSNIFF , Soest , Netherlands ) and water ad libitum . B . glabrata were kept in aquarium water at 25°C on a lettuce diet [18] . Separated B . glabrata were exposed to single S . mansoni miracidia to obtain either female or male cercariae from each snail 6 weeks later [19] . The sex of the cercariae was determined by DNA amplification of sex-related chromosome segments using female-specific W1 and W2 primers and male/female specific Sm23 primers as a positive control ( Table 1 ) . Eleven weeks after the primary infection , mice ( groups mf/mf , m/mf and f/mf ) were percutaneously infected a second time with 50 S . mansoni cercariae of both sexes , and an additional control group ( -/mf ) infected for the first time . The naive control group was not infected . Blood was sampled for the second time and an assessment of signs of disease progression performed at week 19 , when mice were sacrificed via cervical dislocation under ketamine/xylazine anesthesia ( Fig 1A ) . The experiments were then repeated with an infection dose of 150 S . mansoni cercariae to ensure there was a high number of adult worms and eggs in all mouse livers ( Fig 1B ) . In the primary infection step ( week 0 ) , mice were percutaneously infected with 100 male and female ( mf/mf ) S . mansoni cercariae , 100 male ( m/mf ) or 100 female ( f/mf ) S . mansoni cercariae , or not infected ( naive control ) . Blood sampling was performed for the first time at week 8 . Luminex analysis was performed using ProcartaPlex™ Multiplex Immunoassay ( eBioscience , Germany ) according to the manufacturer’s instructions . Serum from all time points was assayed for the murine cytokines IL-4 , IL-10 , IL-12p70 , IL-13 , TNF-α and IFN-γ . The samples and standards were measured using Bio-Plex® 200 System . At week 8 a SWAP ( S . mansoni adult worm soluble antigen ) –in-house ELISA [23] was performed on all mice to detect immunoglobulin G ( IgG ) against worms . To determine infection burden , total egg numbers were assessed by microscopically evaluation ( at 100-fold magnification ) in defined , weighted liver fractions ( squash slides ) . Adult S . mansoni worm pairs were counted using the liver perfusion technique , as described elsewhere [24] . At week 19 the weights of spleens and livers were determined and expressed as ratio of the respective organ to body weight . The extent of liver damage was assessed macroscopically on the basis of infection-related changes in liver color , stiffness and the prevalence of nodules compared to healthy controls . For histological evaluation one half of the right liver lobe was fixed in 10% neutral buffered formalin solution ( Sigma Aldrich , Germany ) and embedded in paraffin . Thin sections of 5 μm were stained with either hematoxylin/eosin ( HE ) or Sirius red ( SR ) . Granuloma size was determined using ImageJ software ( v1 . 47v; National Institutes of Health , USA ) . The extent of hepatic fibrosis was analyzed on the basis of thin sections stained for collagen ( Sirius red , SR ) . The SR-positive areas were assessed using ImageJ software ( v1 . 47v; National Institutes of Health , USA ) . Liver specimens were snap-frozen in nitrogen-cooled methylene butane and then stored in liquid nitrogen until RNA analysis . Total RNA was isolated ( RNeasy Plus Mini Kit , Qiagen , Germany ) and reversely transcribed into cDNA using High-capacity cDNA Reverse Transcriptase Kit ( ThermoFisher , Germany ) according to the manufacturer’s instructions . Real-time PCR ( RT-PCR ) was performed in triplicates using the following TaqMan Gene Expression Assays: Foxp3 Mm00475162; Ctla4 Mm00486849; Arg1 Mm00475988; Retnla Mm00445109 ( ThermoFisher , Germany ) . The reaction was performed on the 7900HT Fast Real-Time PCR System under the following reaction conditions: thermal cycling conditions were 50°C for 2 minutes followed by 95°C for 10 minutes , 45 cycles at 95°C for 15 seconds , and at 60°C for 1 minute . Gene expression values were normalized to the endogenous reference gene GAPDH ( Rodent GAPDH control reagent , ThermoFisher , Germany ) and presented as normalized expression values relative to naive controls . Serum biochemistry for alanine aminotransferase ( ALT ) , aspartate aminotransferase ( AST ) and alkaline phosphatase ( AP ) was performed using UniCel® DxC 800 Synchron® Clinical System ( Beckman Coulter GmbH ) . Statistical analysis was performed using the GraphPad Prism 4 . 0 ( GraphPad Software , San Diego , CA ) . Values are expressed as mean ± standard deviation ( SD ) . Groups were compared using ANOVA ( with Tukey post-hoc ) and , in the event of non-normality , using the Kruskal-Wallis test ( with subsequent Mann–Whitney U-tests , pairwise ) . Normal distribution was checked using the Kolmogorov-Smirnov test . P≤0 . 05 ( Bonferroni-adjusted for multiple testing ) was considered to be statistically significant .
At week 19 macroscopic evaluation of the livers of f/mf mice revealed smooth surfaces with no macroscopically visible nodules and a mean weight of 1 . 28 ± 0 . 14 g , while the livers of m/mf mice were enlarged ( 1 . 34 ± 0 . 07 g ) with rough surfaces and clearly visible nodules . These alterations were even more prominent in group mf/mf , which presented greyish , nodular livers and a mean liver weight of 1 . 92 ± 0 . 15 g ( Fig 2A ) . Group f/mf had lower liver and spleen ratios than the other infected groups ( Fig 2B ) . Serum ALT and AST levels were significantly elevated in all infected groups compared to the naive controls , while levels of AP were not ( Fig 2C ) . Parasite load in groups f/mf and m/mf reached comparable levels ( hepatic eggs: f/mf 7775 ± 4537 and m/mf 8522 ± 3668; worm pairs: f/mf 22 ± 8 and m/mf 28 ± 9 , p> 0 . 05 ) . The highest infection burden was detected in group mf/mf and -/mf ( hepatic eggs: mf/mf 14921 ± 4401 and -/mf 12357± 2368; worm pairs: mf/mf 35 ± 6 and -/mf 39 ± 14 , p> 0 . 05 ) ( Fig 2D ) . Hepatic granulomas were detectable in all infected groups ( Fig 3A ) . Primary infection with female S . mansoni cercariae ( f/mf ) led to significantly smaller granulomas than in the other infected groups ( Fig 3B ) . The largest granulomas , accompanied by a pronounced formation of fibrotic septa , were found in group mf/mf . Mice livers from the groups m/mf and -/mf displayed comparable granuloma sizes and early signs of porto-portal bridging . Quantification of Sirius red-positive areas revealed the highest collagen deposition in group mf/mf and significantly less pronounced hepatic fibrosis in group f/mf ( Fig 3B ) . At week 8 , infection was verified using SWAP-ELISA , which detected antibodies against S . mansoni worm antigen in all infected groups ( Fig 4 ) . TH1-cytokines ( TNF-α , IFN-γ , IL-12p70 ) and TH2-cytokines ( IL-4 , IL-10 , IL-13 ) were measured 8 weeks after initial infection . TNF-α and IFN-γ serum levels were significantly higher in groups f/mf , m/mf and mf/mf than in the naive control group , with the highest levels of TNF-α found in group mf/mf ( Fig 5A ) . This indicates a proinflammatory TH1-reaction . IL-4 and IL-13 were solely detectable in group mf/mf ( Fig 5B ) . Levels of IL-10 did not differ between the groups ( f/mf , m/mf , mf/mf , naive ) . At week 19 the numbers of surviving mice per group were: naive = 6 of 6 , f/mf = 5 of 6 , m/mf = 5 of 6 , mf/mf = 6 of 6 and -/mf = 4 of 6 . Groups f/mf , m/mf , mf/mf and -/mf displayed comparable levels of TNF-α and IFN-γ that were higher than those in the naive controls , indicating that a second infection does not influence TH1-response ( Fig 6A ) . Expression of the TH2 cytokines IL-13 and IL-4 was lower in group f/mf than in groups m/mf , mf/mf and -/mf ( Fig 6B ) . Arg1 and Retnla mRNA expression was measured in order to detect potential disturbances in TH2 polarization . Arg1 was expressed in all infected groups , with the highest values found in group mf/mf . Retnla expression was significantly lower in groups f/mf and m/mf than in -/mf and mf/mf . Foxp3 expression was measured in order to detect the presence of Foxp3 positive regulatory T-cells ( Tregs ) . Foxp3 expression was highest in group mf/mf , while no differences were found between the other infected groups . Ctla4 expression , which inhibits TH2 response , was higher in group f/mf than in m/mf and -/mf ( Fig 7 ) .
We demonstrated that primary single-sex infection with female but not male S . mansoni cercariae decreases granuloma size and hepatic fibrosis , and is accompanied by a Ctla4-mediated suppression of TH2 hyperreactivity . There are two basic options for reducing egg-induced pathology during Schistosoma spp . infection: 1 ) achieve resistance to reinfection , or 2 ) dampen down granulomatous hyperreactivity [4] . The notion that infection with different developmental stages of Schistosoma spp . could have an immunizing effect was based on the observation in both human and animal models that recurrent Schistosoma spp . infections or repeated Praziquantel treatment lead to the development of resistance to reinfection [15 , 25–28] . In our study , a primary infection with male and female or exclusively male or exclusively female cercariae 11 weeks prior to a bisexual challenge did not induce resistance to reinfection , as demonstrated by comparable numbers of worm pairs and hepatic eggs . Previous studies , however , have shown that after a prior bisexual infection with S . japonicum , mice become significantly resistant to reinfection within 6 weeks , and that this resistance peaks at 8 weeks [17] . In yet another study , mice harboring light male infections of the Philippine strain of S . japonicum were not resistant when challenged 7 and 10 weeks after infection [29] . In our study , contrary to expectations , slightly higher numbers of worm pairs were detected at week 19 in group -/mf than in the doubly infected group mf/mf . We assume that worms stemming from the first infection ( 19 weeks ) in group mf/mf may be located deeper within the mesenteric veins and could therefore be more difficult to retrieve using liver perfusion techniques . A de facto higher worm load in group mf/mf is most probable since hepatic egg load in this group exceeds numbers compared to the other groups . Although parasite burden was not influenced by a primary infection , TH2-associated liver damage was significantly less in mice pre-infected with female S . mansoni cercariae . Besides displaying smaller hepatic granulomas and less extensive hepatic fibrosis , group f/mf also had the lowest liver and spleen indexes , indicating a slower rate of disease progression . In contrast to our results , a study from 1997 found no difference between the size of perioval granulomas within the lungs of mice subject to prior male single-sex infection and those in mice subject to prior female single-sex infection when S . mansoni eggs were injected intravenously nine weeks later [13] . Since both the infection steps in our experiment were percutaneous , the involvement of specific worm antibodies in the modulation of immunopathology cannot be ruled out , but comparable IgG titers in all infected groups speak against this hypothesis . In accordance with the smaller granuloma size and less extensive hepatic fibrosis in group f/mf at week 19 , lower serum cytokine levels of pro-fibrotic IL-4 and IL-13 were measured . A combined knock-out of IL-4 and IL-13 leads to an increase in TH1 regulated inflammation accompanied by necrotic tissue destruction and higher mortality [30 , 31] . However , we did not observe higher levels of TH1 regulated inflammation in group f/mf in our setting . In addition , the expression of Arg1 in the livers of all infected mice confirmed an adequate TH2 response . This ties in with our previous findings that in murine Schistosomiasis , bile acid treatment mediates a reduction in TH2 response and hepatic fibrosis without augmenting TH1 response [32] . TH1 and TH2 responses have been shown to be reciprocally regulated to a certain degree [33 , 34] . In our experiments , however , TH1 and TH2 responses seem to be regulated independently . The deposition of S . mansoni hemozoin pigment in mouse livers is associated with the presence of Arg1-positive macrophages that specifically lack Retnla ( also known as Relmα and Fizz1 ) [35] . Retnla is known to be involved in the downmodulation of TH2 response ( via a negative feedback regulation ) [36] . In our experiments Retnla was significantly lower in group f/mf exclusively . Unmated female schistosomes from unisexual infections are developmentally stunted and do not enter mesenteric veins as unmated male schistosomes or worm pairs do [37] . On the basis of our own and others' observations , it seems that the majority of virgin female worms are entrapped within the liver [38] . This might result in larger amounts of hemozoin pigment within the liver , accompanied by a Retnla-mediated suppression of TH2 response . During Schistosoma infection Foxp3 positive T cells ( Tregs ) are key regulators of immune homeostasis . The cytotoxic T-lymphocyte-associated protein 4 ( Ctla4 , also known as protein receptor CD152 ) is constitutively expressed on Foxp3 positive Tregs and constitutes a further potent inhibitor of TH2 response by mediating T-cell anergy and tolerance [39] . In our setting Ctla4 was highest in group f/mf , while Foxp3 was uniformly expressed in all infected groups . As shown recently , blocking Ctla4 during the acute stage of Schistosoma infection results in an exaggeration of TH2 response , suggesting that high levels of Ctla4 might be involved in a reduction of TH2 response and of hepatic fibrosis . Though , this suppression of initial TH2 response is not to be confused with downmodulation of TH2 in persisting infections . Downmodulation of TH2 was observed in group mf/mf and might be an explanation for the reduction of IL-4 and IL-13 levels in group mf/mf with a longer time span of egg deposition ( in total 13 weeks ) versus -/mf ( in total 2 weeks ) [40] . In addition , Foxp3 positive Tregs can produce IL-10 , which also inhibits TH2 response . Due to high variation in the values within the infected groups in our setting , we cannot rule out that IL-10 did indeed play a role [41] . In conclusion , a primary infection with female S . mansoni cercariae has a protective effect on granuloma size , hepatic fibrosis and disease progression in bisexually challenged mice . This protection is potentially associated with Ctla4-mediated TH2 suppression but not with a reduction in parasite load . Our findings provide evidence that protection against egg-induced granulomatous hyperreactivity is achievable by exploiting gender-specific differences between schistosomes . | Schistosomiasis remains a major cause of morbidity and mortality , and in the tropics and subtropics in particular , infection rates are high . The efficacy of anthelmintic therapy is limited since it has no effect on immature parasite stages and does not prevent re-infection . The root cause of disease is a granulomatous hypersensitivity reaction to parasite eggs entrapped within the intestinal wall and small liver sinusoids . This reaction is mainly caused by CD4+ T cells of the subtype 2 and alternatively activated macrophages . As a repair response it suppresses inflammation but results in hepatic fibrosis ( e . g . Symmer's pipe stem fibrosis ) , portal hypertension and its clinical sequelae , ascites and esophageal varices . We demonstrated herein that a primary infection with female S . mansoni cercariae leads to the suppression of TH2-mediated granuloma size , hepatic fibrosis and disease progression in bisexually challenged mice . This protection is associated with Retnla- and Ctla4-mediated TH2 suppression but not with a reduction in parasite load . Our findings provide evidence that protection against egg-induced granulomatous hyperreactivity is achievable by exploiting gender-specific differences between schistosomes . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"schistosoma",
"invertebrates",
"schistosoma",
"mansoni",
"innate",
"immune",
"system",
"medicine",
"and",
"health",
"sciences",
"liver",
"immune",
"cells",
"immune",
"physiology",
"cytokines",
"helminths",
"granulomas",
"spleen",
"immunology",
"fibrosis",
"animals",
"developmental",
"biology",
"molecular",
"development",
"animal",
"cells",
"bisexuals",
"immune",
"system",
"people",
"and",
"places",
"cell",
"biology",
"anatomy",
"physiology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"population",
"groupings",
"sexuality",
"groupings",
"organisms"
] | 2017 | Single-sex infection with female Schistosoma mansoni cercariae mitigates hepatic fibrosis after secondary infection |
Theory of Mind ( ToM ) , i . e . the ability to understand others' mental states , endows humans with highly adaptive social skills such as teaching or deceiving . Candidate evolutionary explanations have been proposed for the unique sophistication of human ToM among primates . For example , the Machiavellian intelligence hypothesis states that the increasing complexity of social networks may have induced a demand for sophisticated ToM . This type of scenario ignores neurocognitive constraints that may eventually be crucial limiting factors for ToM evolution . In contradistinction , the cognitive scaffolding hypothesis asserts that a species' opportunity to develop sophisticated ToM is mostly determined by its general cognitive capacity ( on which ToM is scaffolded ) . However , the actual relationships between ToM sophistication and either brain volume ( a proxy for general cognitive capacity ) or social group size ( a proxy for social network complexity ) are unclear . Here , we let 39 individuals sampled from seven non-human primate species ( lemurs , macaques , mangabeys , orangutans , gorillas and chimpanzees ) engage in simple dyadic games against artificial ToM players ( via a familiar human caregiver ) . Using computational analyses of primates' choice sequences , we found that the probability of exhibiting a ToM-compatible learning style is mainly driven by species' brain volume ( rather than by social group size ) . Moreover , primates' social cognitive sophistication culminates in a precursor form of ToM , which still falls short of human fully-developed ToM abilities .
How do you know what others think or feel ? Theory of Mind ( ToM ) , i . e . the ability to identify covert mental states from others’ overt behaviour , is a crucial component of human social intelligence . Although ToM endows humans with highly adaptive skills such as bonding , teaching or deceiving , its contribution to the cognitive toolkit of other animal species , including primates , is debated [1–3] . Thus , a few theories have been concurrently proposed as candidate explanations for why humans have evolved such unusually sophisticated ToM . For example , the "social brain hypothesis" posits that the complexity of primates' societies is the primary driver of primates' cognitive skills [4 , 5] . The existence of a statistical relationship across primate species between social group size ( a proxy for social network complexity ) and brain volume ( a proxy for general cognitive capacity ) is typically taken as evidence in support of this idea [6 , 7] . Critical here is the notion that the adaptive fitness of social cognitive skills may overcompensate the metabolic cost incurred by large brains [8 , 9] if the typical species' social organization is complex enough . Recent theoretical work demonstrated that such cost-benefit competition can explain the evolutionary dynamics of "Machiavellian intelligence" [10] , i . e . a specific subset of cognitive skills geared towards achieving social success [11] . In short , sophisticated ToM would have evolved mostly as an "on-demand" response to social challenges posed by big herds . However , increases in brain volume may have arisen from other forms of selective pressure ( e . g . , unpredictable and dispersed food resources ) , eventually favouring non-social cognitive skills that endow primates with , e . g . , innovative tool uses or foraging strategies [12–15] . In turn , the causal relationship may be reversed , i . e . larger brains may have eventually enabled species to build and maintain bigger social networks . Under this view , social intelligence is a byproduct of evolutionary pressure on brain volume , which has opened a window of opportunity for sophisticated ToM to emerge [16] . In other terms , the evolution of ToM would be mainly determined by neurobiological limiting factors such as the species' "cognitive reservoir" [17 , 18] . This idea is in line with developmental studies in humans that show that sophisticated ToM is , at least partially , "scaffolded" on domain-general cognitive improvement [19 , 20] . In what follows , we refer to this idea as the "scaffolding hypothesis" [4] . To date , discriminating between these evolutionary hypotheses has not been possible because it requires the difficult combination of ( i ) an operational definition of ToM sophistication that is amenable to behavioural testing in non-human primates , and ( ii ) a balanced comparison of ToM sophistication in primate species that differ in terms of sociobiological features such as group size and brain volume . These are the issues we address in this work , using combined experimental and computational means . Most non-human primates typically engage in diverse and complex social interactions , exhibiting seemingly deceptive and manipulative behaviour [21 , 22] . Following early experimental investigations [23] , positive evidence has supported the idea that chimpanzees—arguably the smartest non-human primate species and the phylogenetically closest to humans- understand what conspecifics know [24] , want [25] or learn [26] . This line of investigation , however , has been challenged by negative results regarding , e . g . , the ability to understand what others perceive [27–29] or to distinguish between one’s own belief and others’ [30–32] . In retrospect , positive evidence might simply have neglected simpler behaviorist explanations of animal policies in social contexts , such as flexible forms of stimulus-response associative learning [2] . Furthermore , notwithstanding a few recent studies on non-ape species—mostly about macaques or other old world monkeys—yielding similarly inconsistent results [33–36] , no systematic comparative study of ToM across primate species has been conducted . This eventually raised profound methodological and theoretical concerns regarding theories of ToM's evolutionary foundations based on existing ethological studies [2 , 37–40] . Taking inspiration from recent advances in machine learning and cognitive psychology [41 , 42] we suggest an operational definition of ToM that departs from previous qualitative ToM investigations . We start with the premise that ToM solves a specific evolutionary challenge , namely: predicting others' overt behaviour from learned associations with social cues ( including past behaviour ) . Critical here is the notion that primate species may differ with respect to their learning styles , whose sophistication may depend upon their innate cognitive structure [16] . Arguably , somewhere at the end of the spectrum lie human learning styles that derive from so-called metarepresentational ToM [43] , whose sophistication increases with the depth of recursive beliefs ( as in "I believe that you believe that I believe…" ) . These highly sophisticated forms of ToM possess adaptive value in the context of strategic social interactions , in which individuals can learn about each other [44–46] . Nevertheless , learning in such contexts can take less sophisticated forms , ranging from simple heuristics , to trial-and-error learning , to cognitive precursors of ToM that simply care about others' overt reaction to one's own actions [47] . Critically , mathematical modelling can be used to turn a given learning style into a learning rule ( i . e . the precise way in which agents adapt to the history of past actions and feedbacks ) , whose cognitive sophistication is formally defined in terms of the computational complexity of information processing [48] . In appropriate experimental contexts ( e . g . , dyadic games ) , this endows learning styles with a specific behavioural signature that can be disclosed from quantitative analyses of trial-by-trial choice sequences . In turn , the cognitive sophistication of learning styles can be inferred from observed overt behaviour , and eventually compared across species . We have previously validated this computational approach by showing that when engaging in mentalizing , human adults’ learning styles are specifically captured by second-order recursive belief updating schemes [49] . We now extend this approach to a comparison of non-human primate species , and ask which of the above hypotheses is the most likely explanation for the evolution of social intelligence . We let 39 individuals from seven non-human primate species with different phylogenetic distances from humans ( including lemurs , macaques , mangabeys , orangutans , gorillas and chimpanzees ) play simple repeated games with familiar zookeepers who followed the instructions of ( on-line ) learning algorithms endowed with calibrated ToM sophistication . Fig 1 below depicts the statistical relationship between endocranial volume ( ECV ) and social group size ( in the wild ) across primate species . Critically , although ECV and social group size are correlated across the full range of primate species ( r = 0 . 62 , p<10−4; see graphical inset in Fig 1 ) , the sample correlation is very weak across the seven tested species ( r = -0 . 37 , p = 0 . 41; see Fig 1 ) . This enables us to evaluate the evidence for candidate evolutionary scenarios by identifying the ensuing statistical relationships existing between social group size , brain volume and ToM sophistication , across tested species . Note that there is an ongoing debate regarding which sociobiological feature of primate species is appropriate for such type of analysis ( see first section of S1 Text ) . We will comment on this and related issues in the Discussion section .
Our main task consisted of multiple sessions of a so-called "hide and seek" game ( 60 trials each ) against three distinct opponents ( below ) . To succeed , primates had to anticipate and predict the behaviour of their opponent , who hid a fruit in one out of two possible locations ( left/right hand ) at each trial ( see Fig 2 below ) . Opponents either followed a predetermined pseudo-random sequence with a 65% bias for one hand ( condition RB ) , or attempted to deceive the primates from learned anticipations of their behaviour ( conditions 0-ToM and 1-ToM ) . The difference between 0-ToM and 1-ToM lies in how they learn from the past history of primates’ actions . In brief , 0-ToM does not try to interpret the primates' action sequence in terms of a strategic attempt to win . Rather , it simply assumes that abrupt changes in the primates' behaviour are a priori unlikely . It thus tracks the evolving frequency of primates’ actions , and chooses to hide the reward where it predicts the primate will not seek . It is an extension of “fictitious play” learning [50] , which can exploit primates' tendency to repeat their recent actions . In contrast , 1-ToM is equipped with ( limited ) artificial mentalizing , i . e . it attributes simple beliefs and desires to primates . More precisely , it assumes that primates’ actions originate from the strategic response of a 0-ToM agent that attempts to predict his own actions . Note that the computational sophistication of artificial mentalizing is not trivial , since 1-ToM has to explicitly represent and update its ( recursive ) belief about its opponents' beliefs . In turn , 1-ToM learning essentially consists in an on-line estimation of 0-ToM’s parameters ( i . e . : learning rate and behavioural temperature; see Methods ) given the past history of both players’ actions . This makes 1-ToM a so-called “meta-Bayesian” agent [49 , 51] that can outwit strategic opponents that do not mentalize when competing in the game ( such as 0-ToM ) . Critically , primates were not cued about opponent conditions . This implies that they had to adapt their behaviour according to their understanding of the history of past actions and outcomes . In addition , except in the control ( RB ) condition , there is no possibility to learn the correct answer from simple reinforcement . This is because 0-ToM and 1-ToM artificial learners exhibit no systematic bias in their response . Further details regarding the experimental protocol ( including animal training ) as well as k-ToM artificial agents can be found in the methods section below . As we will see below , one cannot unambiguously recognize primates' ToM sophistication from their pattern of performance across task conditions . Rather , one has to decompose action sequences and identify learning styles . Nevertheless , let us start with a simple summary of performance results . Fig 3A below shows the net rate of correct answers ( averaged across individuals within species ) , after adjustment for non-specific session effects ( see Methods section ) . One can see that , on average , primates seem to perform reasonably well in the control condition ( RB ) , which means that they have understood the basic tasks' rules . We performed a random-effect analysis to test for the effects of opponent' sophistication and species onto performance ( see Methods ) . At the group-level , we found a significant main effect of opponent ( F[2 , 58] = 14 . 0 , R2 = 32 . 6% , p<10−4 ) and a trend for a main effect of species ( F[6 , 58] = 2 . 17 , R2 = 18 . 3% , p = 0 . 06 ) . No interaction between species and opponent was found ( F[12 , 58] = 1 . 0 , R2 = 17 . 1% , p = 0 . 46 ) . Moreover , when further testing inter-species differences , we found that the ECV predicted overall performance ( F[1 , 58] = 5 . 2 , R2 = 32 . 6% , p = 0 . 026 ) whereas group size did not ( F[1 , 58] = 0 . 3 , R2 = 0 . 5% , p = 0 . 14 ) . Intriguingly , the effect of ECV went in the opposite direction of what could be intuitively expected , in that having a larger brain actually yields worse performance on average . As will be clearer below , this is due to the non-trivial effect of ToM sophistication on performance in this task . This issue will be addressed later , using model-based analyses of action sequences . Now eyeballing the opponent's effect on performance ( see Fig 3B ) reveals the following pattern: overall , primates win in the control ( RB ) condition , whereas they tend to lose similarly against 0-ToM and 1-ToM . This strongly contrasts with the results of our previous experiment on healthy human participants [49] , who win against 0-ToM and 1-ToM , most likely by relying on sophisticated mentalizing akin to competitive 2-ToM learning ( see Fig 3E ) . In fact , two classes of learning styles would be qualitatively compatible with the pattern of primates’ performances across conditions . On the one hand , numerical simulations show that simple non-mentalizing learning schemes such as 0-ToM show a gradual performance decrement with opponent's sophistication ( see Fig 3C ) . On the other hand , cooperative strategies based upon sophisticated mentalizing ( e . g . , 1-ToM or 2-ToM ) eventually win against RB and lose against 0-ToM and 1-ToM ( see Fig 3D ) . Thus , discriminative evidence for or against mentalizing can only be derived from quantitative analyses of trial-by-trial choice sequences . As we will see , these are in fact much more sensitive and informative than model-free performance analyses . Our second step of analysis thus consisted of Volterra decompositions [52] of primates' choice sequences , i . e . we looked at how much trial-by-trial variance in choice sequences can be concurrently explained by the past history of both players actions ( see Methods ) . This decomposition enables us to capture learning styles in terms of model-free mixtures of imitative and perseverative tendencies [49] . Fig 4 below summarizes the mean magnitudes of each species' Volterra kernels , for all conditions . One can see that , on average , primates tend to imitate their opponents' choices ( positive impact of past opponent's choice , Aop ) , which is a good strategy when playing against RB because this , on average , yields reward more often than chance . Although this is reminiscent of a "win-stay/lose-switch" heuristic strategy , we will see below that other learning styles may eventually exhibit this tendency . In addition , they also seem to perseverate , i . e . they tend to repeat their own past choices ( positive Aself on average ) . However , the relative magnitudes of imitative and perseverative tendencies seem to differ across species and conditions . Thus , we performed a random-effect analysis to test for the effects of opponent' sophistication and species onto perseverative ( Aself ) and imitative ( Aop ) tendencies . We found a main effect of opponent for Aself ( F[2 , 58] = 9 . 8 , R2 = 25 . 3% , p = 2×10−4 ) but not for Aop ( F[2 , 58] = 2 . 3 , R2 = 7 . 3% , p = 0 . 1 ) . This is important , since this is a sign of a ( moderate ) strategic adaptation to opponents , such that primates persevere less against 0-ToM than in the other conditions . In addition , we found a strong effect of species on both Aself ( F[6 , 58] = 22 . 0 , R2 = 69 . 5% , p<10−4 ) and Aop ( F[6 , 58] = 19 . 0 , R2 = 66 . 3% , p<10−4 ) , and no interaction ( p = 0 . 7 for Aself and p = 0 . 5 for Aop ) . Note that , when further investigating inter-species differences , we found that both imitative and perseverative tendencies increased with EVC and network size ( all p<10−4 ) . At this point , we asked whether the effect of species and opponent onto performance were mediated by changes in learning styles . We thus computed the correlation between the estimated Volterra kernel of each individual's choice sequence in each condition and that of the corresponding optimal learning style ( namely: 0-ToM against RB , 1-ToM against 0-ToM and 2-ToM against 1-ToM ) . Classical Sobel mediation tests [53] then confirmed that primates' similarity to optimal learning styles mediated the effect of opponent ( p = 0 . 025 ) , ECV ( p = 4×10−4 ) and group size ( p = 5×10−4 ) onto performance . We refer the interested reader to the Methods section for methodological details regarding Volterra analyses . These results are important , because they indicate that performance variations are likely to be driven by differences in species-specific learning styles . For example , a tendency to perseverate may signal a strategic behavioural response relying on sophisticated ToM inference , based on a cooperative interpretation of the game . Intuitively , if primates believe that the goal of the zoo keeper ( the opponent ) is aligned with their own ( e . g . , that he wants to feed them ) , then repeating their own choices is instrumental ( it serves the purpose of achieving coordination ) . Fig 5 below illustrates how different the Volterra kernels of cooperative learning styles and non-mentalizing learning styles can be . We also included a summary of the Volterra results from our previous experiment in humans , which will serve as a reference point . To begin with , note how Volterra kernels of human subjects differ from those of non-human primate species . Critical here is the fact that they adapt their imitative and perseverative tendencies in a quasi-optimal manner . In particular , humans correctly repress their imitative tendency when playing ( unknowingly ) against 1-ToM ( as competitive 2-ToM learners do ) . No non-human primate species exhibits such adaptive flexibility . As one can see on Fig 5 , primates' Volterra kernels are in fact more compatible with either non-mentalizing agents ( 0-ToM ) or cooperative agents with mild sophistication ( 1-ToM ) . More precisely , the strong and rigid imitative tendency of most primate species is similar to 0-ToM’s , while the moderate flexibility of their perseverative tendencies is rather reminiscent of cooperative 1-ToM learning ( cf . U-shaped perseverative kernels across opponent conditions ) . Taken together , we have found strong inter-species differences in Volterra kernels , and some of these variations may be compatible with mentalizing learning styles . One cannot , however , directly interpret quantitative changes in Volterra kernels across species in terms of differences in , e . g . , cooperativeness or learning style . Evidence for the latter can only be derived from direct quantitative comparisons of primates’ trial-by-trial choices sequences and predictions derived from learning models . In what follows , we report the results of a statistical ( Bayesian ) model comparison that quantifies , for each species , the evidence in favour or against ToM-compatible learning styles , given primates' trial-by-trial choice sequences . We considered a set of candidate learning models that differ in terms of their sophistication , ranging from simple behavioural heuristics , to mildly sophisticated learning schemes , to ToM-based ( meta-Bayesian ) recursive belief update schemes . This model set first consists of a family of four different non-ToM models , namely: BN ( biased Nash ) , WS ( "win-stay/lose-switch" heuristic ) , RL ( reinforcement learning ) and 0-ToM . In addition , we included a family of six ToM models , namely: Inf ( cooperative and competitive "influence learning" ) , 1-ToM ( cooperative and competitive ) and 2-ToM ( cooperative and competitive ) . Each of these computational models provides a probabilistic prediction of observed primates' trial-by-trial choice sequences , given the past history of players' actions and specific unknown parameters controlling e . g . , biases and learning rates [49] . Note that the essential difference between ToM and non-ToM models is that only the former assume that observed responses are intentional actions . We fitted these models on primates’ trial-by-trial choice sequences and evaluated their marginal likelihood . We then derived a species-specific estimate of the probability pToM of exhibiting a ToM-compatible learning style . We refer the interested reader to the Methods section for details regarding computational models and the ensuing statistical model comparison procedure , the result of which is summarized on Fig 6 below . We are now in a position to directly compare our two main hypotheses . Recall that under the Machiavelian intelligence hypothesis , ToM sophistication should mostly align with social group size , whereas , under the cognitive scaffolding hypothesis , it should rather align with brain volume ( ECV ) . We can directly test these predictions by asking whether inter-species differences in pToM are best predicted by either group size or brain volume . The result of this procedure is summarized on Fig 6 below . Fig 6A reports the estimated probability of exhibiting a ToM-compatible learning style ( pToM ) . One can see that this probability varies greatly across species , ranging from pToM = 0 . 25 ± 0 . 12 ( mangabeys ) to pToM = 0 . 81 ± 0 . 09 ( chimpanzees ) . Fig 6B summarizes the statistical relationship between group size and pToM ( across species ) . One can see that the pairwise correlation between the two variables is very weak and does not reach statistical significance ( r = -0 . 22 , p = 0 . 69 ) . Now Fig 6C summarizes the statistical relationship between ECV and pToM . Here , there is a strong and significant pairwise correlation between the two variables ( r = 0 . 75 , p = 0 . 03 ) . Note that this result remains statistically significant when accounting for the structured phylogenic relationships between these species ( p = 0 . 04 for a one-sided test on the correlation; cf . S1 Text ) . In addition , ECV is marginally better than group size at predicting inter-species variability in pToM ( p = 0 . 07 ) . These qualitative results are left unchanged if one assumes that inter-species variability in pToM results from a linear mixture of inter-species variability in group size and ECV . Indeed , when regressing pToM concurrently against both ECV and group size , we find that the effect of ECV is significant ( t[4] = 2 . 18 , adjusted R2 = 54 . 3% , p = 0 . 047 ) whereas group size is not ( t[4] = 0 . 28 , adjusted R2 = 2 . 0% , p = 0 . 39 ) . This holds true even if we account for the interaction between ECV and group size ( ECV: p = 0 . 02 , group size: p = 0 . 45 , [ECV x group size]: p = 0 . 13 ) , or if we include the human species in the analysis ( ECV: p = 0 . 02 , group size: p = 0 . 95; assuming pToM[humans] = 1 ) . Let us now ask which learning style ( among the ten candidate models considered here ) best captures choice sequences within species with either small or large brains ( according to a median-split on ECV ) . Note that , using a between-groups Bayesian model comparison [54] , we find that the posterior probability that species with large brains have evolved a more ToM-sophisticated learning style than species with small brains is P = 0 . 99 . Additional details regarding this procedure can be found in S1 Text . Fig 7 below shows the estimated frequency of all learning models for each subgroup of species . One can see that the two subgroups of species strongly differ in terms of learning styles prevalence . More precisely , the learning style that best captures choice sequences of primate species with large brains is the cooperative "influence learning" model ( estimated frequency = 50% ) , whereas species with small brains seem to mostly rely on either reinforcement learning ( estimated frequency = 33% ) or "win-stay/lose-switch" strategies ( estimated frequency = 28% ) . These results are qualitatively consistent with the previous model-free analyses , essentially because "influence learning" exhibit performance and Volterra patterns that are similar to those of 1-ToM . Importantly , none of these subsets of species matches our previous estimate of human ToM sophistication , which was dominated by 2-ToM learning styles [49] . This signals an evolutionary gap between apes and humans , given that "influence learning" is much less sophisticated than 2-ToM learning . We will comment on the computational distinction between "influence learning" and k-ToM learning in the discussion section below .
In this work , we have performed a comparison of non-human primate species playing simple competitive games against human opponents . Using computational analyses of primates' choices sequences , we found that inter-species differences in ToM sophistication are predicted by differences in brain volume but not by differences in social group size . Moreover , we identified an evolutionary gap between great apes and humans , in terms of the sophistication of their respective ToM skills . Our results provide evidence against the common-sense notion that selective pressure favoured sophisticated ToM in species that lived in bigger herds . They are in line with studies showing that , e . g . , the prevalence of social learning ( e . g . , imitative behaviours ) is correlated with neocortex ratio but not with social group size [14] . This immediately raises the following question: given the biological cost of brain tissue , what then endows social intelligence and , in particular , ToM sophistication , with adaptive fitness ? One possibility is that , when it comes to comparing social cognitive skills , social group size is a poorly reliable proxy for the complexity of primates' societies . This has led some authors to rather focus on field reports of , e . g . , "animal culture" , which would be operationally defined as the within-species heterogeneity of socially-acquired behaviour [55] . Alternatively , the adaptive fitness of ToM sophistication may depend in a non-trivial manner on the nature of within-species interactions . For example , it has been shown , using evolutionary game theory , that cooperative interactions promote ToM sophistication to a much lesser extent than competitive interactions , essentially because less sophisticated phenotypes can benefit from the sophistication of cooperative partners [44] . Yet another perspective is that complex primate societies may endow ToM with adaptive fitness only when in conjunction with other socially-relevant skills such as , e . g . , intentional communication [56 , 57] , empathy [58] or reputation management [59] . None of the above suggestions actually challenge the idea that ToM has been selected because it addressed some of the specific challenges posed by complex ( primate ) societies . But this neglects the fact that , in most primate species , combinations of rigid social norms and/or hierarchies with mundane though expedient heuristics have proven sufficient to solve most social challenges [60–63] . An intriguing alternative however , is that sophisticated ToM derived its adaptive fitness from its contribution to solving non-social challenges . For example , social skills such as ToM may enable group members to "distribute their cognition by storing information into other minds" [4 , 64] . Humans , in particular , have reached an unprecedented level of "distributed cognition" , anecdotally culminating in unique forms of collective memory [65] . Under this view , if equipped with the "cognitive reservoir" necessary to scaffold sophisticated ToM , a species can bypass the cognitive limitations of its constituent individuals . Although highly speculative , this perspective is interesting because it explains how a moderate though critical ToM gap between apes and humans can eventually trigger the remarkable evolutionary success of the human species [39] . We will further discuss this notion on computational grounds below . Let us now discuss a few striking aspects of our inter-species comparison . This work corroborates the existing body of studies that provide evidence for a rudimentary form of ToM in apes , as opposed to prosimians and monkeys [16 , 35 , 66] . This is perhaps best exemplified on Fig 7 , which shows the relative frequencies of learning styles for apes and monkeys , respectively: the former learn the influence they may have on others , whereas the latter engage in some form of trial-and-error learning heuristics . Our results are in line with field studies reporting that , e . g . , monkeys show some evidence of imitative behaviours , but to a much lesser extent than apes [67 , 68] . This resonates with the quote that "apes are good psychologists—in that they are good at reading minds—whereas monkeys are good ethologists—in that they are good at reading behaviour—" [69] . Note that one may be surprised by the relatively disappointing results of orangutans , whose estimated ToM sophistication does not quite live up to one's expectations . This deserves a few clarifying comments . First , our estimate of orangutans' ToM sophistication ( worse than other apes but better than most monkey and prosimian species ) may in fact be deemed quite consistent with what would be expected from their position in the primates' phylogenic tree ( see S1 Text ) . Second , there is in fact very few published studies on orangutans' ToM , and these yield quite inconsistent results [30 , 70 , 71] . Third , we tested orangutans in different zoos . This implies that tested individuals are not coming from a single population , which increases the chance that our results are generalizable . Finally , one of the orangutans is somehow special in that she is showing characteristic signs of primates’ Down syndrome [72] . Interestingly , her pToM score is zero , which may have decreased our empirical estimate of orangutans ToM sophistication ( pToM = 0 . 51±0 . 14 if this individual is excluded ) . Note that the results of our analyses are left qualitatively unchanged if we exclude this individual from the data sample . Recall that our experiment aimed at revealing the sophistication of learning styles by observing the patterns of primates' response to the history of choices from artificial agents endowed with calibrated ToM sophistication . We had originally designed the experiment using competitive agents mostly because it yielded the best expected discriminability between learning styles [44 , 49] . However , despite careful training sessions ( see methods ) , primates seem to have partly misinterpreted the human opponent's intentions . In particular , those primate species that display a ToM-compatible learning style behave as if they were engaging in a cooperative game . This may be seen as an unavoidable consequence of the fact that primates were playing with their usual ( human ) caregivers , who are feeding them on a daily basis . One may thus wonder whether this non-ecological aspect of our experimental paradigm may have influenced our analyses . For example , one may think that this may have somehow impeded on their pragmatic understanding of the task . However , primates perform well above chance level against RB , which indicates that they have at least understood the game's contingency between their choice and the reward they get . In fact , primates also perform below chance level against 0-ToM and 1-ToM , which should count as evidence that their learning style was consistent enough to be exploited by artificial competitive agents . On a similar line , one could argue that observed inter-species differences may be confounded by variations in domain-general cognitive competence , which would eventually determine learning efficiency . The intuition here is that , with sufficient training , animals could eventually learn the best response to their opponent , without having to mentalize . We agree that this is in principle possible , since k-ToM artificial agents are reducible ( up to about 80% accuracy ) to a linear convolution of past competing players' actions [49] . Thus , known specificities of species cognitive skills ( such as , e . g . , working memory or attention ) could in principle make a difference . To begin with , note that our stopping criterion for the training/habituation phase was based upon performance , i . e . all species engaged the main protocol with an identical understanding of the task ( see methods section below ) . Now , irrespective of any potential performance improvement across session repetitions , the evidence in favour of ToM-compatible learning styles correlates negatively with performance ( cf . main effect of ECV ) . Finally , in contrast to Volterra kernel magnitudes , we found no difference in Volterra decay rates across species . This means that the effective number of past trials that was impacting on subjects' behavioural responses was the same for all species . In other terms , all species learned from the same amount of past remembered/attended actions and outcomes , but they differed in how they learned . Taken together , this makes domain-general cognitive competence an unlikely confounding factor for our computational results . One may also question the robustness and/or efficiency of our computational approach . First , recall that Bayesian inference is immune to the statistical criticisms that have been raised against the use of p-values in classical inference [73–75] . Nevertheless , one may wonder whether our model-based Bayesian data analysis may not be somehow biased towards ToM-compatible models , eventually yielding artefactual results . This is highly unlikely however , given the differences in model comparison results for species with small and large brains ( cf . S1 Text ) . In brief , it is difficult to think of a statistical bias ( favouring either more or less sophisticated models ) that would be inconsistently expressed in two different groups of subjects . Second , one may ask how reliable our model-based results are , given the apparent complexity of the Bayesian statistical procedure . Beyond authoritative arguments , we are committed to provide pragmatic demonstrations of our methodological rigor . First , we performed a statistical confusion analysis , which confirmed that candidate models were well identifiable under our experimental design ( see S1 Text for details ) . This means that the potential algorithmic imperfections of our statistical procedure do not compromise the interpretation of our results . Second , although less sensitive , the results of performance and Volterra analyses are consistent with our model-based conclusions ( cf . Figs 4 and 5 ) . This provides construct validity to our computational approach . Finally , one may argue that our sample of selected species is too small for drawing any definitive conclusion . We acknowledge that , in statistical terms , our sample size is arguably limited ( n = 7 primate species and about 5 individuals per species ) . However , it is largely exceeding the standards in the field , in which data availability is a known issue [76 , 77] . Besides , it is in fact remarkable that we detect our effect of interest in the context of such small-powered study . Equipped with computational means for discriminating learning styles , we have separated learning styles that do rely on mentalizing from learning styles that do not . This effectively induced some sophistication cut-off between those behavioural patterns that are likely to be based upon ToM and those that are not . We used this to assess the evidence in favour of a statistical relationship between ToM sophistication and either brain volume or group size . This raises a number of related comments . First , one may ask how robust to changes in species' sociobiological features our results really are . The relevance of such concern is at least twofold . First , we used ECV as a proxy for some measure of "cognitive reservoir" , which ToM could eventually be scaffolded upon . However , ECV also grows with "non-cognitive" brain mass ( e . g . , cerebellum , basal forebrain , etc… ) , which is why other measures such as relative neocortex volume have been sometimes preferred . Although the two measures are known to correlate with each other [14 , 78 , 79] , considering relative neocortex volume instead of ECV may make a difference for , e . g . , gorillas , which have a relatively small neocortex given their total brain volume . Second , field estimates of group size in the wild are notoriously debated for orangutans species , which may evolve in so-called "fission-fusion societies" [80] . In our context , this calls for a critical reappraisal of their semi-solitary status ( see S2 Text ) , eventually revising their estimated community size by one order of magnitude . Having said this , it turns out that the conclusion of our analyses does not change if we regress ToM sophistication against relative neocortex volume instead of ECV ( neocortex ratio: p = 0 . 03 , group size: p = 0 . 97 ) , even if we modify orangutans' group size estimate ( neocortex ratio: p = 0 . 04 , group size: p = 0 . 95 ) . In addition , we acknowledge that other important factors may eventually determine primates' social cognitive skills . Examples include , but are not limited to: flexibility of social hierarchies [81] or dietary constraints on foraging strategies [82] . The issue with considering such sociobiological constraints is twofold . Whether and how they complement or moderate simpler features such as ECV or group size cannot be predicted from first ( evolutionary ) principles [16] . In fact , this may critically depend on how they are operationally defined . More pragmatically speaking , exploring these dimensions would require testing a huge amount of species in order to compensate for likely statistical correlations between explanatory variables . Taken together , we think it is beyond the scope of the present study to commit to such an exhaustive assessment of the candidate social and biological determinants of animal cognitive skills . Second , one may challenge our computational definition of ToM , whose least sophisticated form simply cares about others' instrumental reaction to one's actions [47] . Recall that the algorithmic complexity of such "influence learning" scheme lies somewhere between that of 0-ToM and 1-ToM . Interestingly , although it is in principle possible to augment the "influence learning" rule with higher-order adjustment terms ( cf . Eq 5 in the Methods section ) , this does not bring any significant behavioural change [49] . This contrasts with k-ToM learners , whose depth of recursive beliefs critically determines the expected outcome of social interactions [44] . Note that , in our previous investigation of ToM sophistication in healthy human adults , we found that people mostly behave as either 1-ToM ( estimated frequency = 26% ) or 2-ToM ( estimated frequency = 59% ) meta-Bayesian agents [49] . We found no strong evidence for such recursive ToM belief update schemes in non-human primates . This implies that meta-Bayesian recursive belief updating schemes may be the hallmark of human social cognition . As we have discussed earlier , the lack of evidence for meta-Bayesian learning in monkeys and apes is in line with the notion of an evolutionary gap between human and non-human minds [39] . But this is not to say that apes lack anything remotely resembling ToM . This is because they behave as if they were adjusting their estimate of others' likely responses to their own actions . Recall that this adjustment depends upon others' covert ( cooperative or competitive ) intentions . Although it is beyond the grasp of such "influence learning" to realize that others may be using ToM themselves ( in contrast to , e . g . , 2-ToM ) , we argue that it should be seen as a precursor form of ToM in its own right . In conclusion , although this work does not resolve the debate regarding whether ToM is a uniquely human cognitive skill , it provides an unprecedented computational insight onto the evolutionary roots of social intelligence . In particular , we provide empirical evidence against an orthodox variant of the Machiavellian intelligence hypothesis , which would state that sophisticated ToM evolved mostly as an "on-demand" response to complex societies . Rather , the evolution of sophisticated ToM seems to be mainly determined by neurobiological limiting factors such as the species' "cognitive reservoir" . Importantly also , the sophistication of non-human primates' ToM culminates in some form of cognitive precursor of human ToM , or proto-ToM . These results are compatible with the idea that ToM may be a byproduct of evolutionary pressure on non-social cognitive skills , which , in conjunction with rigid social norms and/or hierarchies , may otherwise be sufficient to solve most social challenges in most primate species .
Animals' care and behavioural assessment was performed in accordance with institutional ethical guidelines . The experiments were carried out in four different institutions: the Institut du Cerveau et de la Moelle épinière ( Paris , France ) , the Ménagerie du Jardin des Plantes ( Paris , France ) , the St Martin-la-Plaine zoo ( France ) and the Bioparco ( Roma , Italy ) . Seven primate species were sampled as follows: N = 7 orangutans ( Pongo pygmeus ) , N = 6 chimpanzees ( Pan troglodytes ) , N = 5 western gorillas ( Gorilla gorilla ) , N = 4 lion-tailed macaques ( Macaca silenus ) , N = 5 rhesus macaques ( Macaca mulatta ) , N = 9 sooty mangabeys ( Cercocebus atys lunulatus ) and N = 4 ring-tailed lemurs ( Lemurs catta ) . This gives an average of about 5 . 7 ± 1 . 8 individuals per species . We refer the interested reader to S1 Text for additional information regarding individual characteristics ( e . g . , sex , age , rearing ) these and species' sociobiological features ( social group size and ECV ) . The protocol consisted in two phases: a habituation/training and an experimental phase , which occurred right before the daily food delivery to keep animals motivated . The food reward was matched to the animal body size ( e . g . , one or two pieces of dried grapes or papaya ) and was kept constant across the entire protocol . The experimenter ( a familiar caregiver ) always faced the animal in front of the cage ( through which the animal could pass their hands or fingers ) and positioned his two hands symmetrically ( to avoid postural biases ) . To prevent any olfactory detection of the hiding hand , the caregiver carefully rubbed both hands with the food reward before each test . The habituation phase was introduced to teach the animal that the reward was hidden in one hand only ( before their choice ) , that a trial begins by the presentation of the caregiver's closed hands , and that it would obtain the content of the hand it would touch or point at . It consisted of two distinct steps . In the first step , the caregiver placed the food reward in one hand and a small stone in the other . Then , he presented both open hands to the animal , such that both contents were clearly visible . The animal received the food only when it touched or clearly pointed uniquely the hand containing the reward . Rewarded side was counterbalanced across trials according to a pseudorandom sequence . This first step was considered successful once the animal reached 10 consecutive correct answers . The second step consisted of a series of three sequences of five trials each: ( i ) the caregiver first showed both open hands ( while attended by the animal ) but then closed the non-rewarded hand , ( ii ) he first showed both open hands and then closed the rewarded hand , and ( iii ) he first showed both open hands and then closed both hands . In all cases , the individual had to choose the correct hand to obtain the reward . The second step was considered successful once the individual made no error through the entire set of trials ( if unsuccessful , the three steps were repeated ) . The proper experimental phase began after successful habituation/training , and was grouped into 4x3 = 12 daily sessions of 60 trials each . The order of the three game conditions ( RB , 0-ToM or 1-ToM ) were counterbalanced across the 12 sessions , but each game condition was performed by a specific caregiver ( counterbalanced across subjects ) . All sessions were video-recorded . If a daily session was interrupted for more than 10 minutes ( because of , e . g . , frustration or attentional distraction ) , the session was terminated and possibly restarted on another day . Only sessions longer than 20 trials were included in the final analysis . At each trial , the caregiver presented his two hands closed after having hidden the food reward and the stone out of the animal's sight . If the animal chose the correct hand , he was allowed to take and eat the food reward . Otherwise , the caregiver acted as if he was eating the food while exaggerating chewing , vocalizing pleasure and staring at the animal . The reward location was instructed by the algorithm corresponding to the game condition ( RB , 0-ToM or 1-ToM ) . This required the presence of a co-experimenter who entered the individual's response into a laptop computer at each trial , enabling the model to compute on-line the reward location at the next trial . In this section , we give a brief overview of the set of candidate learning models , with a particular emphasis on k-ToM models ( because these are also used as on-line algorithms during the experimental phase ) . We will consider repeated dyadic ( two-players ) games , in which only two actions are available for each player ( the animal and the caregiver ) . Hereafter , the action of a given agent ( resp . , his opponent ) is denoted by aself ( resp . , aop ) . A game is defined in terms of its payoff table , whose entries are the player-specific utility U ( aself , aop ) of any combination of players' actions at each trial . In particular , competitive ( resp . , cooperative ) social interactions simply reduce to anti-symmetric ( resp . symmetric ) players’ payoff tables ( see tables S3 and S4 in S1 Text ) . By convention , actions aop and aself take binary values encoding the first ( a = 1 ) and the second ( a = 0 ) available options . According to Bayesian decision theory , agents aim at maximising expected payoff V = E[U ( aself , aop ) ] , where the expectation is defined in relation to the agent's uncertain predictions about his opponent's next move . This implies that the form of the decision policy is the same for all agents , irrespective of their ToM sophistication . Here , we consider that choices may exhibit small deviations from the rational decision rule , i . e . we assume agents employ the so-called "softmax" probabilistic policy: P ( aself=1 ) =11+exp ( −ΔVβ ) ( 1 ) where P ( aself = 1 ) is the probability that the agent chooses the action aself = 1 , ΔV is the expected payoff difference ( between actions aself = 1 and aself = 0 ) , and β is the so-called behavioural "temperature" ( which controls the magnitude of deviations from rationality ) . The sigmoidal form of Eq 1 simply says that the probability of choosing the action aself = 1 increases with the expected payoff difference ΔV , which is given by: ΔV=pop ( U ( 1 , 1 ) −U ( 0 , 1 ) ) + ( 1−pop ) ( U ( 1 , 0 ) −U ( 0 , 0 ) ) ( 2 ) where pop is the probability that the opponent will choose the action aop = 1 . This prediction is critical , in that it provides the agent with prospective action values . For example , if one believes that the opponent is likely to pick action aop = 1 ( i . e . if pop ≈ 1 ) , then the expected payoff reduces to ΔV = U ( 1 , 1 ) −U ( 0 , 1 ) , which directly determine the incentive towards choosing either aself = 1 or aself = 0 . In our context , animals are rewarded for choosing the hand in which the caregiver has hidden the food reward , which is simply written as: U ( 1 , 1 ) −U ( 0 , 1 ) = U ( 0 , 0 ) −U ( 1 , 0 ) = 1 ⇒ ΔV = 2pop −1 . Let us first disclose the intuition behind k-ToM models , which essentially differ in how they estimate pop . We refer the interested reader to S1 Text for a more detailed mathematical description . In brief , the repeated observation of his opponent's behaviour ( aop ) gives the agent the opportunity to learn his opponent's behavioural tendency pop . Theory of Mind comes into play when agents consider that pop is driven by the opponent's hidden beliefs and desires . More precisely , k-ToM agents consider that the opponent is himself a Bayesian agent , whose decision policy pop = P ( aop = 1 ) is formally similar to Eq 1 . In this situation , k-ToM agents have to track their opponent's prediction pself about their own actions . In line with [42] , this meta-Bayesian inference is based upon recursive belief updating ( "I believe that you believe that I believe…" ) . The recursion depth k induces distinct ToM sophistication levels , which differ in how they update their subjective prediction pop , hence k-ToM . More formally , k-ToM learning agents are defined recursively , starting with 0-ToM . By convention , a 0-ToM agent does not attribute mental states to his opponent , but rather tracks his overt behavioural tendency without mentalizing . More precisely , 0-ToM agents simply assume that their opponents choose the action aop = 1 with probability pop = s ( xt ) , where the log-odds xt varies across trials t with a certain volatility σ0 ( and s is the sigmoid function ) . Observing his opponent's choices gives 0-ToM information about the hidden state x , which can be updated trial after trial using Bayes rule , as follows: μt0≈μt−10+Σt0 ( atop−s ( μt−10 ) ) Σt0≈11Σt−10+σ0+s ( μt−10 ) ( 1−s ( μt−10 ) ) ( 3 ) where μt0 ( resp . Σt0 ) is the approximate mean ( resp . variance ) of 0-ToM's posterior distribution p ( xt0|a1:top ) . Inserting p^t+1op=E[s ( xt+1 ) |a1:top] into Eq 1 now yields 0-ToM's decision rule . Here , the effective learning rate is the subjective uncertainty ∑0 , which is controlled by the volatility σ0 . At the limit σ0 → 0 , Eq 3 converges towards the ( stationary ) opponent's choice frequency and 0-ToM essentially reproduce "fictitious play" strategies [50 , 83] . 0-ToM's learning rule is the starting point for a 1-ToM agent , who considers that she is facing a 0-ToM agent . This means that 1-ToM has to predict 0-ToM's next move , given his beliefs and the choices' payoffs . The issue here is that 0-ToM's parameters ( volatility σ0 and exploration temperature β ) are unknown to 1-ToM and have to be learned , through their non-trivial effect on 0-ToM's choices . At trial t + 1 , a 1-ToM agent predicts that 0-ToM will chose the action aop = 1 with probability pt+1op , 0=s∘v0 ( xt0 , a→t ) , where the hidden states xt0 lumps σ0 and β together and the mapping v0 is derived from inserting 0-ToM's learning rule ( Eq 3 ) into Eqs 1 and 2 . Similarly to 0-ToM agents , 1-ToM assumes that the hidden states xt0 vary across trials with a certain volatility σ1 , which yields a meta-Bayesian learning rule similar in form to 0-ToM's , but relying on first-order meta-beliefs ( i . e . beliefs about beliefs ) . In brief , 1-ToM eventually learns how her ( 0-ToM ) opponent learns about herself , and acts accordingly ( cf . Eqs 1 and 2 ) . 1-ToM agents are well equipped to deal with situations of observational learning . However , when it comes to reciprocal social interactions , one may benefit from considering that others are also using ToM . This calls for learning styles that rely upon higher-order meta-beliefs . By construction , k-ToM agents ( k ≥ 2 ) consider that their opponent is a κ-ToM agent with a lower ToM sophistication level ( i . e . : κ < k ) . Importantly , the sophistication level κ of k-ToM's opponent has to be learned , in addition to the hidden states xκ that control the opponent's learning and decision making . The difficulty for a k-ToM agent is that she needs to consider different scenarios: each of her opponent's possible sophistication level κ yields a specific probability pt+1op , κ=s∘vκ ( xtκ , a→t ) that she will choose action aop = 1 . The ensuing meta-Bayesian learning rule entails updating k-ToM's uncertain belief about her opponent's sophistication level κ and hidden states xκ: λtk , κ≈[λt−1k , κptop , κ∑κ'<kλt−1k , κ'ptop , κ']atop[λt−1k , κ ( 1−ptop , κ ) ∑κ'<kλt−1k , κ' ( 1−ptop , κ' ) ]1−atopμtk , κ≈μt−1k , κ+λtκΣtk , κWt−1κ ( atop−s∘vκ ( μt−1k , κ ) ) ∑tk , κ≈[ ( ∑t−1k , κ+σk ) −1+s'∘vκ ( μt−1k , κ ) λtκWt−1κTWt−1κ]−1 ( 4 ) where λtk , κ is k-ToM's posterior probability that her opponent is κ-ToM , and Wκ is the gradient of vκ with respect to the hidden states xκ . Note that although the dimensionality of k-ToM's beliefs increases with k , k-ToM models do not differ in terms of the number of their free parameters . More precisely , k-ToM’s learning and decision rules are entirely specified by their prior volatility σk and behavioural temperature β . Finally , the only difference between "competitive" and "cooperative" k-ToM learners lies in the specification of the utility table U ( aself , aop ) . Although it is held constant across trials , it can induce profound changes in the effective learning style of k-ToM agents [44 , 49] . We refer the interested reader to the S1 Text for mathematical details regarding k-ToM learning models . Critically , only k-ToM agents with k≥1 are learning about others' covert mental states ( by updating meta-beliefs ) . This would suggest a clear sophistication cut-off for discriminating ToM and no-ToM learning styles . But in fact , we will also consider a hybrid ( non Bayesian ) model that somehow lies in between 0-ToM and 1-ToM , and still qualifies for ToM . We refer the interested reader to [47] for a mathematical derivation of the "influence learning" model . In brief , it is essentially a 0-ToM learner that heuristically adjusts his learning rule to account for how her own actions influence her opponent’s strategy: pt+1op=ptop+η ( atop−ptop ) ⏟predictionerror−λptop ( 1−ptop ) ( 2atself+ ( 2Icomp−1 ) βs−1 ( ptop ) +Icomp ) ⏟“influence”adjustmentterm ( 5 ) where η ( resp . λ ) controls the relative weight of its prediction error ( resp . the “influence” adjustment term ) , and Icomp is a binary indicator variable for the type of social interaction ( competition: Icomp = 1 , cooperation: Icomp = 0 ) . In contrast to 1-ToM , this learning rule bypasses any form of recursive belief update . However , Inf explicitly depends upon the other player's covert ( competitive or cooperative ) intention , which is beyond the grasp of 0-ToM . In analogy with k-ToM models , it is in principle possible to augment Eq 5 with higher-order adjustment terms . This , however , has little effect on the way the algorithm learns [49] . In addition , numerical simulations show that , in a competitive game setting , Inf wins over 0-ToM but loses against 1-ToM . This is why , altogether , we think of "influence learning" as some form of proto-ToM . With the exception of 0-ToM , we so far only described sophisticated learning models that are capable of ( artificial ) ToM . But even 0-ToM can be considered too sophisticated for some primate species . In the aim of assessing the evidence for ToM sophistication ( from primates' choice sequences ) , we thus have to benchmark the above models against simpler learning styles that involve even fewer cognitive resources . We will describe three of these "unsophisticated" learning models below . First , animals may learn by trial and error , eventually reinforcing the actions that led to a reward . Such learning style is the essence of classical conditioning , which is typically modelled using reinforcement learning or RL [84] . In this perspective , animals would directly learn the value of alternative actions , which bypasses Eq 2 . More precisely , an RL agent would update the value of the chosen option in proportion to the reward prediction error , as follows: {Vt+1i=Vti+α ( Rt−Vti ) ifactionatself=iwaschosenVt+1i=Vtiotherwise ( 6 ) where Rt=U ( atself , atop ) is the last reward outcome and α is the ( unknown ) learning rate . At the time of choice , animals simply tend to pick the most valuable option ( cf . Eq 1 ) . Second , an even simpler way of adapting one's behaviour in operant contexts such as this one is to repeat one's last choice if it was successful and alternate otherwise . This can be modeled by the following update in action values: {Vt+1i=Rtactionatself=iwaschosenVt+1i=−Rtotherwise ( 7 ) This strategy is called win-stay/lose-switch ( WS ) , and is almost identical to the above RL model when the learning rate is α = 1 . Despite its simplicity , WS can be shown to have remarkable adaptive properties [85] . Last , the agent may simply act randomly , which can be modeled by fixing the value difference to zero ( ΔV = 0 ) . Although embarrassingly simple , this probabilistic policy eventually prevents one's opponent from controlling one's expected earnings . It thus minimizes the risk of being exploited at the cost of providing chance-level expected earnings . It is the so-called "Nash equilibrium" of our "hide and seek" game [86] . Since we augment this chance model with a potential bias for one of the two alternative options ( as all the above learning models ) , we refer to it as biased Nash or BN . Our statistical data analysis proceeds in three steps of increasing specificity , namely: multiple regression of behavioural performances , Volterra decompositions of trial-by-trial choice sequences and Bayesian model comparison . All statistical analyses were performed using the VBA toolbox [87] . First , let us summarize our random-effect analysis of performance . As a preliminary stage , we regressed out the effect of session repetition and time elapsed since the last experimental session from measured individual performances . We then reported the adjusted individual performance scores per opponent at the group level . We regressed performance against the effect of species , opponent ( conditions RB , 0-ToM and 1-ToM ) , and their interaction . In addition to subject-specific intercepts , we also included the interactions of the opponent effect with age ( normalized by species-specific life time expectancy in the wild ) , sex and rearing ( wild vs captivity ) . In turn , statistical tests for effects of species and opponent assess significance above and beyond these potential inter-individual differences . The specific effects of ECV and group size were tested using weighted linear contrasts . Second , we performed Volterra decompositions of trial-by-trial choice sequences using session-specific Bayesian logistic regressions , as follows: p ( aself|ω ) =∏tqt ( ω ) atself ( 1−qt ( ω ) ) 1−atselfqt ( ω ) =s ( ω0+∑τωτop ( 2at−τop−1 ) +∑τωτself ( 2at−τself−1 ) ) ( 8 ) where qt ( ω ) =p ( atself=1|ω ) is the probability that the agent chooses the first option at trial t , τ is some arbitrary time lag and ω is the so-called Volterra kernel ( ω0 is a potential bias for one of the alternative options ) . Volterra kernels ωop ( resp . ωself ) capture the impact of lagged opponent's ( resp . own ) actions aop ( resp . aself ) onto primates’ choice probability . For the sake of efficiency , we further reduce the Volterra kernels to parameterized exponential mappings , i . e . : ωτ = Aexp ( −λτ ) , where A ( resp . λ ) is the kernel's magnitude ( resp . temporal decay ) . For each individual and each session , we fit the resulting model and report the kernels' magnitudes Aop and Aself at the group level . The ensuing random-effect analyses are identical to the above performance scores . Third , we performed statistical ( Bayesian ) model comparisons . For each subject , we fitted the above ten learning models on trial-by-trial action sequences using a variational-Laplace approach [88 , 89] . Different sessions of the same opponent condition were pooled together , allowing us to constrain the model parameters to be identical across sessions ( but not across opponents ) . Eventually , we obtained 10x35 = 350 model evidences ( 10 models and 35 individuals; the 3 opponent conditions were lumped together for model inversions ) . These model evidences were partitioned into ToM ( 1-ToM , 2-ToM and Inf ) and no-ToM ( all other models ) families , to obtain within-subject posterior probabilities pToM of exhibiting a ToM-compatible learning style . These scores were then averaged across individuals within species to yield the variable pToM , for further analyses ( see Fig 5 ) . In addition , we performed a group-level random-effect Bayesian model comparison [54 , 90] . In particular , this analysis enabled us to estimate the frequency profiles of learning models within species with high versus low ECV . We refer the interested reader to S1 Text for additional statistical details regarding the Bayesian model comparison . | The contribution of Theory of Mind ( ToM ) , i . e . the ability to understand others' mental states , to the cognitive toolkit of non-human animal species ( including primates ) , is fiercely disputed . We contribute to this debate by ( i ) proposing a computational definition of ToM sophistication that is amenable to behavioural testing in non-human primates ( which we had previously validated in humans ) , and ( ii ) performing a balanced comparison of seven primates species ( from lemurs to monkeys to great apes ) . In turn , our study provides an unprecedented computational insight into the evolutionary roots of human social intelligence . In particular , we provide empirical evidence against the common-sense idea that sophisticated ToM evolved mostly as an "on-demand" response to social challenges posed by big herds . Rather , the evolution of sophisticated ToM seems to be mainly determined by neurobiological limiting factors such as the species' "cognitive reservoir" . En passant , we identify an evolutionary gap between great apes and humans , in terms of the sophistication of their respective ToM skills . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] | [
"learning",
"medicine",
"and",
"health",
"sciences",
"vertebrates",
"social",
"sciences",
"limbs",
"(anatomy)",
"neuroscience",
"theory",
"of",
"mind",
"mammals",
"learning",
"and",
"memory",
"animals",
"primates",
"cognitive",
"psychology",
"animal",
"behavior",
"cognition",
"animal",
"management",
"zoology",
"animal",
"cognition",
"animal",
"performance",
"musculoskeletal",
"system",
"hands",
"human",
"learning",
"behavior",
"agriculture",
"arms",
"psychology",
"eukaryota",
"anatomy",
"biology",
"and",
"life",
"sciences",
"cognitive",
"science",
"amniotes",
"organisms"
] | 2017 | Reading wild minds: A computational assay of Theory of Mind sophistication across seven primate species |
Many members of the phylum of Apicomplexa have adopted an obligate intracellular life style and critically depend on active invasion and egress from the infected cells to complete their lytic cycle . Toxoplasma gondii belongs to the coccidian subgroup of the Apicomplexa , and as such , the invasive tachyzoite contains an organelle termed the conoid at its extreme apex . This motile organelle consists of a unique polymer of tubulin fibres and protrudes in both gliding and invading parasites . The class XIV myosin A , which is conserved across the Apicomplexa phylum , is known to critically contribute to motility , invasion and egress from infected cells . The MyoA-glideosome is anchored to the inner membrane complex ( IMC ) and is assumed to translocate the components of the circular junction secreted by the micronemes and rhoptries , to the rear of the parasite . Here we comprehensively characterise the class XIV myosin H ( MyoH ) and its associated light chains . We show that the 3 alpha-tubulin suppressor domains , located in MyoH tail , are necessary to anchor this motor to the conoid . Despite the presence of an intact MyoA-glideosome , conditional disruption of TgMyoH severely compromises parasite motility , invasion and egress from infected cells . We demonstrate that MyoH is necessary for the translocation of the circular junction from the tip of the parasite , where secretory organelles exocytosis occurs , to the apical position where the IMC starts . This study attributes for the first time a direct function of the conoid in motility and invasion , and establishes the indispensable role of MyoH in initiating the first step of motility along this unique organelle , which is subsequently relayed by MyoA to enact effective gliding and invasion .
The phylum of Apicomplexa includes numerous human and animal pathogens that have adopted an obligate intracellular life style and consequently critically depend on active invasion and egress from the infected cells to ensure survival and propagation . Host cell entry is initiated by the attachment and reorientation of the polarized parasites , such that the apical secretory organelles ( micronemes and rhoptries ) sequentially discharge their contents at the point of contact with the host cell plasma membrane . Both host cell entry and exit are driven by gliding motility , a process involving conserved machinery termed the glideosome , located in the space between the inner membrane complex ( IMC ) and the parasite plasma membrane where adhesins are translocated from the apical to the posterior pole [1] . Adhesins and other proteins are secreted apically by the micronemes and among them , AMA1 forms a complex with a set of rhoptry neck proteins ( RONs ) to establish a tight apposition between the parasite and the host cell membrane [2] . This attachment zone forms a ring-like structure called the moving junction [3] and through which the parasite enters the host cell [4] . It is referred here to the circular junction ( CJ ) . Toxoplasma gondii is among the most successful invaders with a third of the world’s human population chronically infected , as well as a broad range of warm-blooded animals . This parasite is responsible for toxoplasmosis , which can lead to severe neurological defects in situations of immunosuppression and in case of congenital infection [5] . The gliding motility of T . gondii tachyzoites has been experimentally dissected and deconstructed into three forms of motion that include circular and helical gliding and stationary twirling [6] . This has recently been further and more accurately assessed through use of a 3-dimensional , matrigel-based motility assay [7] . As a member of the coccidian subgroup of the Apicomplexa , T . gondii harbours an additional apical motile organelle termed the conoid , which is connected to two preconoidal rings ( PCR ) at the top and an apical polar ring ( APR ) at the bottom that serves as a polarized microtubule organizing centre forming all together the apical complex [8] . This organelle protrudes during gliding , invasion and egress and is formed by a unique polymer of tubulin fibers arranged as a set of counter-clockwise spiralling filaments , ultimately creating a cone-shaped structure [9] . A similar structure is present in other members of the group of Alveolata [10] and has also been recently studied in the Gregarine genus [11] . In T . gondii , twenty-two subpellicular microtubules ( MTs ) emerge from the APR , extend in a helical fashion over two thirds of the parasite length and contribute to the overall shape , rigidity and polarity of the parasite [9 , 12] . More than 170 proteins have been identified in the conoid enriched fraction including dynein , calcium binding proteins , MORN-domain-containing proteins , centrin , and also some myosin heavy and light chains [8] . More specifically , two proteins , the lysine methyltransferase ( AKMT ) and the ring protein RNG2 have been localised to the apical complex and functionally associated to parasite motility [13 , 14] . The conoid is the first organelle generated during daughter cell scaffold formation [12 , 15] and its extrusion was thought to contribute to motility and invasion [16] . Intriguingly , a screen for small molecules inhibiting invasion identified Inhibitor 6 as a compound able to block conoid protrusion without affecting parasite motility [17] . Protrusion is induced by the calcium ionophore A23187 or ethanol , and inhibited by Ca2+ chelation ( BAPTA-AM ) , suggesting that Ca2+ release from internal stores may act as a key signal to trigger this movement [16 , 18] . Additionally sensitivity to cytochalasin D ( CD ) and to the myosin ATPase inhibitor 2 , 3-butanedione monoxime ( BDM ) implicates a role for a myosin motor in conoid protrusion [18] . Micronemes and rhoptries presumably secrete their contents at the tip of the apex . To discharge and inject their contents into host cells , the neck of the rhoptries traverses the apical complex and reaches a point of contact at the tip of the parasite referred to as the porosome [19] . It is unclear whether full conoid protrusion is necessary for rhoptry discharge , in contrast it does not appear to be a prerequisite for the release of microneme content [17] . T . gondii possesses eleven unconventional myosin heavy chains , many of which are yet to be assigned a role [20] . MyoA is highly conserved across the phylum of Apicomplexa and plays a central role in powering motility [21] . As part of the glideosome , MyoA is anchored to the IMC via its interaction with the Gliding Associated Protein 70 ( GAP70 ) at the level of the apical cap , and with GAP45 throughout the remaining length of the parasite . An additional glideosome based on the class XIV myosin C ( MyoC ) is restricted to the posterior pole of the parasite and interacts with GAP80 and IMC-associated protein 1 ( IAP1 ) [22] . Although recent work demonstrated that myoA gene is dispensable for tachyzoite survival [23] , MyoC was later shown to compensate for the absence MyoA and deletion of the two myosin genes was not tolerated [24 , 25] . The essential nature of these two myosins during gliding and invasion , points to an important role for actin during these events . Conditional excision of the unique actin gene in T . gondii ( ACT1 ) resulted in a severe but not complete block in invasion . This led to the postulation of an alternative model for invasion [24 , 26] . Despite this , an additional in depth investigation of ACT1 depletion consolidated the importance of actin during invasion [27] . Here we have characterized T . gondii myosin H ( MyoH , XP_002366825 ) , an indispensable actin-based motor associated to the microtubules of the conoid . The tail region of MyoH possesses alpha-tubulin suppressor 1 domains ( ATS1 ) that are a prerequisite for the anchoring of this motor to the conoid . Despite the presence of an intact MyoA-glideosome , depletion of MyoH dramatically impacts on motility , invasion and egress from the infected cells . Importantly , microneme secretion and conoid protrusion remain unaffected in the absence of MyoH . These results highlight a key contribution of the conoid with MyoH acting presumably as a prerequisite initiator of motility at the parasite tip , which is then relayed by MyoA at the level of the IMC .
MyoH belongs to the class XIVc of myosins and is found restricted to the coccidian subgroup of Apicomplexa [20] . This myosin heavy chain of 1513 amino acids is composed of a head , which maintains ATPase and actin-binding activities , linked via the neck to the tail domain . The unusually large neck domain comprises 8 IQ motifs that presumably serve as binding sites for myosin light chains ( MLCs ) in order to fulfil structural and regulatory functions . Lastly , the tail domain includes 3 α-tubulin suppressor 1 ( ATS1 ) or RCC1 ( Regulator of chromosome condensation 1 ) domains previously reported to play a role in the regulation of microtubule coordination during the cell cycle of Saccharomyces cerevisiae [28] ( S1 Fig ) . To determine the subcellular localization of MyoH , the gene was modified at the endogenous locus by the insertion of 3 epitope tags ( Ty ) via single homologous recombination to produce MyoH-3Ty . In accordance with its predicted amino acid sequence , MyoH-3Ty has an apparent molecular weight of 170 kDa by western blot analysis ( WB ) ( Fig 1A ) . Indirect immunofluorescence assay ( IFA ) revealed that this motor is localized to the apical pole , at the tip of both intracellular replicating and extracellular motile parasites and is produced early during parasite division when the daughter IMCs began to emerge ( Fig 1B and 1C ) . In addition , the same localization was observed when the endogenous locus was modified to introduce a myc tag at the N-terminus of MyoH under the control of a tetracycline ( tet ) -regulatable promoter , without impacting on targeting and function . Based on a series of IFAs with markers of the tachyzoite apex , we concluded that MyoH is positioned at the tubular core of the conoid . RNG1 , a marker of the APR [29] , co-localized with myc-MyoH in extracellular parasites but gave a distinct signal upon conoid protrusion ( Fig 1D ) . Partial co-localization was observed between calmodulin 1 ( CAM1 [8] ) and MyoH ( Fig 1D ) . In contrast , no co-localization was observed between MyoH and centrin 2 ( CEN2 ) which is located at the PCR [8] . Moreover , MyoH-3Ty was shown to form a ring at the top of the parasite by using three-dimensional reconstruction of the MyoH-3Ty staining by super-resolution microscopy ( SIM ) ( S1 Movie ) . Taken together , these results indicate that the MyoH ring is restricted to the upper part of the conoid , close to the PCR , as depicted in the scheme of Fig 1D . TgMyoH was refractory to gene disruption by single homologous recombination in the RHΔku80 ( Δku80 ) strain , suggestive of the crucial nature of this motor . We subsequently undertook conditional disruption approaches and readily obtained an inducible knockdown strain by swapping the endogenous promoter with a tet-repressive promoter ( S2A Fig ) . The resulting strain ( myoH-iKD ) was confirmed by genomic PCR analyses ( S2B Fig ) . WB and IFAs performed on parasites incubated for 48 h in presence of anhydrotetracycline ( ATc ) revealed a tight regulation of the inducible Myc-MyoH ( Fig 2A and 2B ) . To explore the phenotypic consequences of MyoH depletion , plaque assays were performed with myoH-iKD parasites in the presence or absence of ATc for seven days . The absence of detectable plaques upon ATc treatment was indicative of a severe defect in one or more steps of the lytic cycle ( S2C Fig ) . To pinpoint the step ( s ) affected by MyoH depletion , a series of targeted assays were performed . Parasite replication was unaltered , as scored by intracellular growth assay ( Fig 2C ) . The biogenesis and positioning of the secretory organelles implicated in motility and/or invasion were not affected ( S2D Fig ) . In contrast , egress induced by addition of the calcium ionophore A23187 and host cell invasion were severely impaired in the presence of ATc when compared with untreated parasites falling to 12% and 10% , respectively ( Fig 2D and 2E , S2 and S3 Movies ) . Since invasion and egress critically depend on parasite motility , gliding assays were performed using myoH-iKD and uncovered a dramatic defect ( Fig 2F ) . Assessment of the number of trails revealed a phenotype as severe as treatment with CD which was previously demonstrated to abrogate gliding motility [30] ( Fig 2G ) . To determine if MyoH had an impact on conoid protrusion , both the reversible ethanol inducer [18] and the stronger A23187 inducer [31] were compared . Despite the suggestive localization of MyoH , its depletion revealed no significant reduction of conoid protrusion upon either stimulation ( Fig 3A ) . The essential nature of MyoH gene and the contribution of this motor in invasion and egress was confirmed by a parallel strategy leading to the transient excision of the 3’UTR of the gene and the concomitant destabilization of the residual transcript as previously described [32] . To this aim the endogenous locus of MyoH 3’UTR was replaced by an excisable 3’UTR followed by U1 sequences to generate MyoH-3Ty-floxU1 strain which was readily obtained . As anticipated , transient transfection of these parasites with a Cre-GFP expressing vector led to the disappearance of MyoH by IFA ( S2E Fig ) . Concordantly , the parasites expressing transiently Cre-GFP ( lacking MyoH expression ) were significantly impaired in egress and invasion ( S2F and S2G Fig ) . Depletion of MyoH recapitulates the same phenotypic consequences as in the previously described mutants affecting microneme discharge [17 , 33] . Consequently , MyoH depleted parasites were carefully scrutinized for microneme proteins secretion . Subsequent assays revealed that upon ethanol stimulation , the microneme proteins AMA1 , MIC6 and MIC4 were normally secreted in the absence of MyoH as indicated by their presence in the excretory-secretory antigen ( ESA ) fraction of myoH-iKD parasites ( Fig 3B ) . The same results were obtained with MIC2 ( Fig 3C ) whereas its secretion was completely blocked when parasites were treated with compound 2 ( S2H Fig ) . Like compound 1 , compound 2 inhibits cGMP-dependent protein kinase , which critically contributes to the signalling cascade leading to microneme secretion [34 , 35] . Discharge of MIC2 at the parasite surface did not accumulate at the tip in absence of MyoH ( S2I Fig ) . Concordant with an unaltered microneme secretion , host cell attachment performed on MyoH depleted parasites was not affected when compared with GFP expressing wild type parasites ( Fig 3D ) . The impact of MyoH depletion was also tested in a rhoptry secretion assay whereby formation of empty vacuoles ( e-vacuoles ) in the presence of CD was monitored as an indicator of rhoptry discharge [36] . CD pre-treated myoH-iKD parasites +/- ATc showed no defect in e-vacuole formation ( Fig 3E ) , indicating that rhoptry secretion is not affected in myoH-iKD parasites . Members of the Apicomplexa possess a unique AKMT at the conoid that rapidly re-localizes in the presence of egress-stimulating signals in tachyzoites [14] . Interestingly , akmt-KO parasites share with myoH-iKD the same severe defect in invasion , egress and gliding motility while the substrate ( s ) and function of this intriguing enzyme are not known [14 , 37] . To determine if MyoH function is linked to that of AKMT , we investigated the A23187-dependent cytoplasmic redistribution of AKMT in the absence of MyoH . AKMT re-localization was not affected in myoH-iKD parasites treated ± ATc ( Fig 3F ) however some residual AKMT protein could still be observed in the A23187-induced parasites depleted for MyoH . This suggests either an independent role for MyoH in gliding , or a role likely downstream of the AKMT signalling cascade . Finally , the assembly and composition of the MyoA-glideosome was examined to ensure that depletion of MyoH did not indirectly impact on glideosome functioning . Both GAP45 and MLC1 were distributed evenly along the IMC in myoH-iKD ± ATc parasites ( S2D and S2J Fig ) . Immunoprecipitation of the MyoA-glideosome revealed no alteration of its composition in the absence of MyoH ( Fig 3G ) , however we cannot entirely rule out that this motor would fail to be activated in the absence of MyoH . Parasites impaired in gliding motility such the myoA-iKD strain loose virulence in the mouse model [21] . Similarly , animals infected with myoH-iKD and supplemented with ATc in their drinking water survived infection whereas in the absence of ATc the mice succumbed to infection 8 days post-inoculation ( S3A Fig ) . However , and in contrast to MyoA depletion , infection with parasites depleted in MyoH led to a more radical phenotype in vivo since the surviving mice failed to seroconvert after 20 days ( S3B Fig ) and concordantly showed no protection against a subsequent challenge . To glean further insight into MyoH functioning , we determined its solubility using detergent partitioning and carbonate extraction ( Fig 4A ) . A large portion of MyoH is insoluble and likely tightly associated with the parasite cytoskeleton . Treatment of extracellular parasites with deoxycholate ( DOC ) prior to IFA preserves the association of MyoH with the intact MTs of the conoid ( Fig 4B ) . Since myosins traditionally bind to actin filaments , we artificially induced apical actin polymerization using jasplakinolide ( JAS ) . This assay revealed that MyoH remained strongly anchored to the conoid , again attesting for the robust direct or indirect interaction with MTs ( Fig 4B ) . To determine which part of MyoH interacts with the MTs , we generated transgenic parasites in which a Shield-regulated destabilisation domain ( DD , [38] ) was appended to extra-copies of truncated forms of MyoH ( Fig 4C and 4D ) . Given the absence of a coiled-coil domain , MyoH is not predicted to operate as a dimer and hence expression of the neck and tail region ( DD-MyoH-NT ) , neck and tail region without ATS1 domains ( DD-MyoH-NT-ΔATS1 ) or tail domain only ( DD-MyoH-T ) were not anticipated to interact with endogenous full length MyoH . However , overexpression of DD-MyoH-NT , which localized to the conoid , exhibited a strong dominant negative effect ( Fig 4F ) . In sharp contrast , DD-MyoH-T remained cytosolic and its expression was neutral for parasite survival ( Fig 4E and 4F ) . Insertion of a GFP in DD-MyoH-NT ( DD-Myc-GFP-MyoH-NT ) did not perturb the localization or the expression of the truncated MyoH ( S3C and S3D Fig ) . Metabolic labelling followed by co-IPs of DD-Myc-GFP-MyoH-NT showed no band detected at the wt size of MyoH ( 170 kDa ) suggesting that no deleterious heterodimer is formed with endogenous MyoH ( S3E Fig ) . Since the phenotype of DD-MyoH-NT overexpression perturbs organelle biogenesis ( S3F Fig ) these results led us to conclude that the numerous IQ motifs could titrate out MLC ( s ) shared by other myosin motors , compromising their function and hence explaining the toxic effects observed . Concordantly , DD-MyoH-NT-ΔATS1 failed to localize to the conoid but still exhibited a dramatic effect on parasite growth again likely attributable to the over-expression of the neck domain ( Fig 4E and 4F ) . Taken together these results suggest that the ATS1 domains are either directly involved in the association of MyoH with the conoid MTs or indirectly by interacting with conoid specific microtubule-associated proteins ( MAPs ) . Additionally , the neck domain of MyoH appears to participate in bringing the motor to its site of action . MyoH is predicted to possess 8 putative IQ motifs that are presumed to bind to one or more MLC ( s ) . T . gondii possesses at least seven MLCs that have been previously localized by expression of tagged second copies and only MLC5 ( TGME49_311260 ) was reported to localise to the conoid [39] . However , expression of second copies could lead to localization artefacts due to over or improper timing of expression . The MLCs were therefore epitope-tagged via single homologous recombination at the endogenous locus and in addition to MLC5 , MLC3 ( S3G Fig ) and MLC7 were found to localize exclusively to the conoid ( Fig 5A ) . Immunoprecipitation of MLC5 and MLC7 revealed the co-IP of a protein with a size corresponding to the MyoH ( S3H Fig ) . Due to its insolubility , immunoprecipitation of MLC3 could be not investigated . The identity of MyoH was confirmed by mass spectrometry with 11 and 43 unique peptides corresponding to MyoH in the MLC5 and MLC7 co-IPs , respectively ( S3I Fig and S1 Dataset ) . Given the poor solubility of MyoH , the identification of potentially loosely bound partners interacting with this motor would be hampered due to the stringent co-IP conditions utilised in these assays . In light of these findings we subsequently Ty-tagged MLC5 and MLC7 at their endogenous locus in the myoH-iKD background strain in order to investigate their fate in the absence of MyoH . Upon MyoH depletion , MLC5 became cytosolic ( Fig 5B ) although its expression level quantified by WB was unchanged even after 72 h of ATc treatment ( Fig 5D ) . Interestingly , upon MyoH depletion , MLC7 was still present at the conoid ( Fig 5C ) but its level of expression decreased after 24 h and stabilized after 48 h of ATc treatment ( Fig 5E ) . To further investigate the contribution of MLC5 and MLC7 in MyoH function , the corresponding genes were individually and concomitantly deleted using CRISPR/Cas9 [40] to introduce out of frame mutations in RH strain parasites . Gene disruption mutants were confirmed by sequencing of the corresponding locus ( S4A Fig ) . In the absence of both MLC5 and MLC7 , there was no significant loss in parasite fitness as revealed by plaque assay ( Fig 5F ) , indicating that these accessory light chains are dispensable for MyoH localization to the conoid ( Fig 5G ) and for its function . Metabolic labelling followed by co-IPs using MyoH-3Ty and MyoH-3Ty-mlc5&7-KO revealed a band at the size of MLC1 ( Fig 5H ) . This interaction was confirmed by co-IPs using MLC1 antibodies and demonstrated that MyoH interacts with MLC1 not only in mlc5&7-KO where MyoH was subsequently endogenously tagged but also in wt situation ( S4B Fig ) . No detectable change in MLC1 distribution at the parasite tip was observed when comparing IFAs performed on MyoH-3Ty and MyoH-3Ty-mlc5&7-KO strains ( S4C Fig ) . To define the step at which MyoH participates in invasion , we performed pulses of invasion and documented the events by IFA using anti-RON4 as a marker of the CJ . In addition and prior to permeabilization with saponin , the accessible extracellular part of the parasites was stained with anti-SAG1 ( major surface antigen 1 ) . The myoH-iKD parasites under ATc treatment were compared to wild type and myoA-KO parasites for their respective abilities to invade host cells within 7 min ( Fig 6A ) . While the majority of wild type parasites successfully completed invasion within this short time , most parasites depleted in MyoH were simply attached to the host cell by their tip with the CJ forming only a dot ( Fig 6A and 6B ) . In sharp contrast , myoA-KO parasites successfully inserted their conoid into the host cell , however it must be noted that the CJ is arrested at the beginning of the cap where MyoA is anchored via its interaction with GAP70 [22] ( Fig 6C ) . To confirm that only the tip of the parasite including the conoid entered into the host cell , myoA-KO parasites were stained with a marker of the apical cap , anti-ISP1 [41] following pulse invasion ( Fig 6D ) . In myoA-KO parasites , SAG1 labelling is interrupted at the level of ISP1 confirming that only the tip of the parasite has penetrated into the host cell .
The conoid is a mysterious organelle , originally identified as 13 internal microtubule protofilaments of α-tubulin first described in T . gondii by De Souza in 1974 [9 , 42] . Since then , a number of studies have shed light on its structure and its calcium-dependent protrusion/retraction ( Fig 7A ) , however none have assigned any definitive function to this enigmatic and dynamic organelle [8 , 18 , 43] . Here we demonstrate that MyoH is distributed to the upper part of the conoid , close to the pre-conoidal ring and probably outside of the conoid structure as suggested by the MyoH distribution visualized in the S1 Movie . This myosin of the class XIV is predominantly insoluble in the presence of detergent and remains attached to the conoid upon deoxycholate extraction , strongly indicating a direct or indirect interaction with the MTs constituting this organelle . This association of MyoH to the conoid MTs could involve either the recognition of a specific conformation or posttranslational modification ( such as methylation ) of the MTs or alternatively implicates some conoid specific MAPs . Expression of the neck and tail regions of MyoH in the presence or absence of the ATS1 domains points toward their involvement in binding MTs either directly or indirectly . With its tail anchored to the conoid , MyoH is presumably facing the plasma membrane in a similar orientation as MyoA , which is anchored via MLC1 and GAP70 to the apical cap , and via MLC1 and GAP45 to the rest of the IMC [22 , 25] . Although such an orientation is not formally demonstrated for MyoH , it would be plausibly positioned like MyoA to interact with actin filaments underneath the plasma membrane . MyoH is associated with MLC1 , MLC5 and MLC7 . Although MLC5 was previously shown to be at the conoid , localization of MLC7 was overlooked , likely due to the overexpression of a second copy driven by a constitutive promoter [39] . Intriguingly and quite unexpectedly , MLC5 and MCL7 are simultaneously dispensable and hence not absolutely required to sustain or regulate MyoH function . In contrast , MLC1 , which is also shared by MyoA and MyoC [25] , is not dispensable [24] , hampering investigation of its role in MyoH function . MLC5 is exclusively associated with MyoH at the conoid since it is no longer detectable at this location upon depletion of MyoH . In contrast , MLC7 is still present at the conoid following MyoH depletion . In this context it is worth mentioning that other yet uncharacterized myosins could be present in the same location . Conditional depletion of MyoH resulted in a block in motility , invasion and egress in tissue culture that translated into a complete avirulence in the murine infection model , indicating that the absence of MyoH is deleterious during the asexual part of the T . gondii life cycle . Further phenotypic investigations of myoH-iKD revealed that neither the secretion of the contents from the apical secretory organelles , nor protrusion of the conoid were impacted by the absence of MyoH . Taken together these results led us to conclude that a direct impairment of this actomyosin system accounts for the incapacity of the parasites to invade or egress . A closer examination of the fate of the CJ using anti-RON4 antibodies in parasites lacking either MyoH or MyoA was tremendously informative . In the absence of MyoA , the CJ is blocked at the beginning of the apical cap whereas in the absence of MyoH the CJ fails to form and progress , resulting in the presence of a dot at the extreme apical tip of the parasite probably blocked at the PCRs level . In light of these findings , we propose a model whereby MyoH acts as the initiator of gliding motility , potentially bringing the adhesin/receptor complexes to the cap where MyoA takes the relay and translocates the complexes to the posterior pole of the parasites ( Fig 7B ) . Additionally the helical organization of the MTs of the conoid plausibly confers the helical movement to the parasites ( Fig 7C ) , which is likely relayed at the level of the IMC by the array of inner membranous particles that are connected to the helical subpellicular MTs [44] . Finally , to explain the complete block in motility observed in absence of MyoH , it is tempting to speculate that this motor serves as a gatekeeper and allows polymerized actin to enter the glideosome space between the plasma membrane and the IMC . Numerous questions still remain unanswered . Upon calcium stimulation the conoid protrudes , but in a MyoH-independent manner . The role of this protrusion and the motor generating it are unknown . A previous study reported the identification of an inhibitor ( number 6 ) that affects conoid protrusion but not parasite motility , indicative of a role in host cell entry only [17] . Interestingly , a physical link between the APR and the rest of the conoid is established by RNG2 , a protein recently shown to link the polar ring to the conoid [13] . Intriguingly , RNG2 localizes to the APR and flips when the conoid is protruded and could potentially ensure the relay between MyoH and MyoA . This hypothesis is supported by the phenotype observed in the rng2-iKD parasites , i . e . an impairment in gliding motility , which might not be solely explained by the reported decrease in microneme secretion [13] . AKMT is located at the conoid and its depletion displays a similar phenotype as the absence of MyoH [14] . Here we show that MyoH depletion does not prevent the cytosolic AKMT re-localization suggesting that the motor acts either independently or downstream of AKMT . Speculatively , AKMT could be involved in the methylation of the α-tubulin forming the conoid , given that T . gondii α-tubulin has been shown to be methylated [45] . In this proposed model MyoH , MyoA and MyoC participate in gliding motility that crucially depends on a physical connection between surface adhesins and the actomyosin system to generate power [46] . The central contribution of MyoH in motility and invasion reinforces the fundamental role of actin in establishment of infection . The connection between F-actin and the tail of the adhesins was thought to be mediated by aldolase , however recent studies ruled out a role of this glycolytic enzyme in gliding motility and invasion and hence the identification of a physiologically relevant connector is necessary to validate the overall model [47] . In addition to the coccidian subgroup of Apicomplexa , the Gregarines also harbour a closed conoid however we failed so far to identify a MyoH homologue in the available genome sequences of Gregarina niphandrodes ( http://eupathdb . org ) . Moreover MyoH phylogeny based on the head domain showed a heterogeneous distribution within Apicomplexa with homologues also found in Eimeria , Cryptosporidium and Piroplasmida ( Babesia and Theileria ) [20] . It is unclear if these closely related motors preserved a similar function or evolved after divergence of these lineages . In this context it is worth mentioning that the invasive stages of the malaria parasites exhibit an electron dense zone at the tip of the zoite where the class XIV PfMyoB has recently been localized [48] . It is tempting to speculate that PfMyoB might fulfil at least in part a similar role as MyoH . PfMyoB is not anchored to MTs but is instead associated with an unusual light chain , which might bring this indispensable motor to its site of action , as reported for MyoA [22 , 48] .
E . coli XL-10 Gold chemocompetent bacteria were used for all recombinant DNA experiments . T . gondii tachyzoites ( RHΔhxgprt , RHΔhxgprtΔKu80 and derivatives ) were grown in human foreskin fibroblasts ( HFF ) or in Vero cells ( African green monkey kidney cells ) maintained in Dulbecco’s Modified Eagle’s Medium ( DMEM , Gibco ) supplemented with 5% foetal calf serum ( FCS ) , 2 mM glutamine and 25 μg/ml gentamicin . Conditional expression of the different transgenes was performed with 0 . 5 μM Shld-1 for DD-fusion stabilization [38] and with 1 μg/ml anhydrotetracycline ( ATc ) for the Tet-inducible system [21] . Genomic DNA was isolated with the Wizard SV genomic DNA purification system ( Promega ) . RNA was isolated using Trizol ( Invitrogen ) extraction . Total cDNA was generated by RT-PCR using the Superscript II reverse transcriptase ( Invitrogen ) according to manufacturer’s protocol . TgMyoH GenBank accession number: XP_002366825 . TgMyoH-NT and TgMyoH-T fragments were amplified on cDNA using primer pairs MyoH-3747-11F/MyoH-3749-12R and MyoH-3748-13F/MyoH-3749-12R ( S1 Table ) , respectively and cloned into the pTub8DDmycFH2-HX [49] vector between NsiI and PacI sites . The GFP coding fragment was digested with NsiI from the pT8DDmycGFPTgMyoF-tail-HX-Ble vector [50] and sub-cloned into the unique NsiI site of DD-MyoH-NT to generate DD-GFP-TgMyoH-NT vector . TgMyoH-NTΔATS1 fragment was amplified from TgMyoH-NT using MyoH-5395-ΔATS1-18F/MyoH-5396-ΔATS1-19R primers following Q5 Site-Directed Mutagenesis Kit ( NEB ) manufacturer’s recommendations . To generate TgMyoH-3Ty , a gDNA fragment was amplified with primers MyoH-2748-1F/MyoH-2749-2R , digested with KpnI and PstI restriction enzymes and cloned into pT8-TgMIC13-3Ty-HX [51] between KpnI and NsiI sites . To generate 5’MyoH-TATi1-HX-tetOpS1MycNtMyoH vector , the 5’MyoH fragment was amplified using primers MyoH-3964-3F/MyoH-3965-4R and cloned between NcoI and BamHI in 5’MyoF-TATi1-HX-tetO7S1MycNtMyoF [50] . NtMyoH was amplified with primers MyoH-3966-5F/MyoH-3967-6R and cloned between BglII and SpeI in the same plasmid . To generate TgMLC5-3Ty , TgMLC7-3Ty and MLC3-3Ty , gDNA fragments were amplified with primers MLC5-4129-19F/MLC5-1374-20R , MLC7-4403-24F/MLC7-2348-25R and MLC3-4402-31F/MLC3-1910-32R respectively and cloned into pT8-TgMIC13-3Ty-HX [51] between KpnI and NsiI sites . To generate MyoH-3Ty-LoxP-3’UTR-LoxP-U1 , the C-terminal sequence from the TgMyoH-3Ty was digested with KpnI and PstI and sub-cloned into the KpnI and NsiI sites of pG152-3Ty-LoxP-3’UTRSag1-HXGPRT-LoxP-U1 [32] after modification to introduce a unique KpnI site . To generate vectors with gRNA specific of MLC5 , MLC7 , and MyoH , a directed mutagenesis was performed on the CRISPR/Cas9-GFP vectors [40] using the Q5 Site-Directed Mutagenesis Kit Protocol and the primer pairs gRNA-Rv-4883-33R/MLC5-gRNA-5087-21 , gRNA-Rv-4883-33R/MLC7-gRNA-5068-26 and gRNA-Rv-4883-33R/MyoH-gRNA-5389-29 following the manufacturer recommendations . pTub5-Cre vector [52] has been modified by insertion of a GFP coding sequence in frame with the Cre recombinase to create pTub5-Cre-GFP . T . gondii tachyzoites were transfected by electroporation as previously described [53] . Δku80 [54 , 55] strain was transfected with 40 μg of the plasmids: 5’MyoH-TATi1-HX-tetO7S1MycNtMyoH ( linearized with SbfI/SpeI ) or TgMyoH-3Ty ( linearized with XhoI ) or MyoH-3Ty-LoxP-3’UTR-LoxPU1 ( linearized with NsiI ) . Stable transfectants were selected for hypoxanthine-xanthine-guanine-phosphoribosyltransferase ( HXGPRT ) expression in the presence of mycophenolic acid and xanthine as described earlier [56] . Parasites were cloned by limiting dilution in 96 well plates and analysed for the expression of the transgenes by indirect immunofluorescence assay ( IFA ) . Transient transfection of CAM1-GFP and EGFP-TgCentrin2 , provided by Ke Hu ( Indiana University ) , was performed by using 40 μg of each plasmid as previously described [53] . 40 μg of the pTub5-Cre-GFP was transiently transfected in the Δku80 and MyoH-3Ty-floxU1 strains 30 h and 48 h prior to performing egress and invasion assays , respectively . To disrupt MLC5 and MLC7 locus , 30 μg of the MLC5 gRNA-specific CRISPR/Cas9-GFP vector was transfected in RH parasites . After confirmation of gene disruptions , the MLC7 gRNA-specific CRISPR/Cas9-GFP vector was transfected in the mlc5-KO background or in wt RH to obtain respectively the mlc5/7-KOs and mlc7-KO . TgMyoH was subsequently endogenously tagged using 30 μg of the gRNA-specific CRISPR/Cas9-GFP co-transfected with the MyoH-Ki-2748-1F/ MyoH-Ki-RH-5390-30R PCR product amplified from the TgMyoH-3Ty vector . 48 h after transfections , parasites were FACS sorted and cloned into 96-well plates . mAbs α-Ty tag BB2 , α-MLC1 [57] and GFP-trap ( Chromotek ) antibodies were used to performed co-IPs . Antibodies described here were used for indirect immunofluorescence assay ( IFA ) and western blot analysis . The mAbs α-Ty tag BB2 , α-Myc tag 9E10 , α-SAG1 T4-1E5 , α-MIC3 T42F3 , α-ROP2 T3-4A7 , as well as the polyclonal Abs αMIC4 , α-PRF , α-GAP45 , α-GAP40 , α-CAT were previously described [22 , 58–62] . Monoclonal α-RNG1 and polyclonal α-AKMT Ab were kindly provided by Dr . Naomi Morrissette ( California University ) and Dr . Ke Hu ( Indiana University ) respectively . α-tubulin antibody ( #32–2500 , Invitrogen ) was used . For western blot analysis , secondary peroxidase conjugated goat α-rabbit and α-mouse Abs ( Molecular Probes , Paisley , UK ) were used . For IFAs , parasite-infected HFF cells seeded on cover slips were fixed with 4% paraformaldehyde ( PFA ) or 4% PFA/0 . 05% glutaraldehyde ( PFA/GA ) in PBS , depending of the antigen to be labelled . Fixed cells were then processed as previously described [63] . Confocal images were collected with a Zeiss microscope ( LSM700 , objective apochromat 63x /1 . 4 oil ) at the Bioimaging core facility of the Faculty of Medicine , University of Geneva . Stacks of sections were processed with ImageJ and projected using the maximum projection tool . Crude extracts of T . gondii tachyzoites were subjected to SDS-PAGE . Western blot analysis was carried out using polyacrylamide gels under reducing conditions . Proteins were transferred to hybond ECL nitrocellulose . Primary and secondary antibodies are diluted in PBS , 0 . 05% Tween20 , 5% skimmed milk . Bound antibodies were visualized using the ECL system ( Amersham ) . Cells were infected with about 50 parasites and let to develop for 7 days . Fixation and staining were performed as previously described [58] . Similarly to plaque assays HFF were infected with parasites but fixation occurred after 24h with PFA/GA and stained with anti-TgGAP45 . The number of parasites per vacuole was determined by counting the parasites in 100 vacuoles in duplicate for three independent experiments . Freshly egressed T . gondii tachyzoites were pelleted and re-suspended in either: 20mM HEPES , 5 mM CaCl2 for calcium ionophore A23187 ( 3 μM ) stimulation , or 20 mM HEPES , 138 mM NaCl , 2 . 7 mM KCl , 10% FCS , pH 7 . 2 for EtOH ( 2% ) stimulation and same volume of DMSO or water was added as controls respectively . Parasites were then incubated for 8 min at 37°C on Poly-L-lysine coated coverslips and fixed for 30 min with PFA/GA . Protruded and non-protruded parasites were counted using the 63x magnification under DIC condition . The average number of protruded parasites was determined by counting 100 parasites for each condition for three independent experiments . For Δku80 and myoH-iKD , parasites were pre-treated for 48 h ± ATc . Data are presented as mean values ± SD from three independent experiments . The significance of the data was evaluated using a parametric paired t-test . The red/green assay was performed as previously described [64] . The number of intracellular and extracellular parasites was determined by counting 100 parasites in triplicate for three independent experiments . For Δku80 and myoH-iKD , parasites were pre-treated for 48 h ± ATc . Attachment of T . gondii parasites to HFF monolayers was assessed as previously described [65] . Briefly , extracellular wt expressing GFP and myoH-iKD cell lines pre-treated 48 h ± ATc were mixed 50/50 in Endo buffer ( 44 . 7 mM K2SO4 , 10 mM Mg2SO4 , 106 mM sucrose , 5 mM glucose , 20 mM Tris , 0 . 35% wt/vol BSA , pH 8 . 2 ) containing 1 μM cytochalasin D ( Sigma-Aldrich ) . Incubate at room temperature for 10 min . 500 μL was added to a coverslip coated with an HFF monolayer ( assay ) or Poly-L-lysine ( control ) and centrifuged for 1 min at 1000 rpm . Controls samples were fixed with PFA/GA for 7 min at RT . Then , medium was replace with pre-warmed DMEM 5% FCS containing 1 μM cytochalasin D to prevent invasion , incubated for 15 min at 37°C and fixed with PFA/GA for 7 min . Immunofluorescence was performed using α-GAP45 and Alexa647 ( “red” ) as secondary antibodies . Ratio between red/green ( wt ) and red only ( myoH-iKD ) attached parasites was counted in triplicate ( minimum of 100 parasites screened ) for three independent experiments either for controls or assays . E-vacuole assays utilizing cytochalasin D were performed as described [65] . Freshly released parasites were inoculated on new host cells , centrifuged for 1 min at 1000 g and allowed to invade for 7 minutes before fixation with PFA/GA for 5 minutes . IFAs were performed first without permeabilization with α-SAG1 antibodies then fixed 10 min with 1% formaldehyde and permeabilized with 0 . 1% Saponin/PBS for 20 min at RT and stained with α-RON4 . Only parasites harbouring a RON4 positive signal were counted . Ratio between the different conformations ( attached—rhoptry content secreted , conoid invaded or post conoid invaded ) were counted in triplicate ( minimum of 100 parasites screened ) for three independent experiments . Data are presented as mean values ± SD from three independent experiments . Δku80 and myoH-iKD were grown for 18 h ±ATc . Freshly egressed tachyzoites were added to a new monolayer of HFF and grown for 30 h ±ATc . The infected HFF were then incubated for 5 min at 37°C with DMEM containing either 3 μM of the Ca2+ ionophore A23187 ( from Streptomyces chartreusensis , Calbiochem ) or DMSO as a negative control . Host cells were fixed with PFA/GA and IFA were performed using α-GAP45 antibodies . The average number of egressed vacuoles was determined by counting 200 vacuoles per strain and per condition . Data are presented as mean values ± SD from three independent experiments . The significance of the data was evaluated using a parametric paired t-test and the two-tailed p-value is written on the corresponding graph . Freshly egressed tachyzoites were harvested , washed in PBS and then re-suspended into PBS , PBS/ 1% Triton-X-100 , PBS/ 1 M NaCl , or PBS/ 0 . 1 M Na2CO3 , pH 11 . 5 . Parasites were lysed by freezing and thawing followed by sonication on ice . Pellet and soluble fractions were separated by centrifugation at 14 000 rpm for 1 h at 4°C . Metabolic labelling of the tachyzoites was done with 50 mCi [35S]-labeled methionine/cysteine ( Hartmann analytic GmbH ) per ml for 4 h at 37°C followed by co-IP as previously described [25] . Infected host cells were washed with 0 . 1 M phosphate buffer pH 7 . 4 and were fixed with 2 . 5% glutaraldehyde in 0 . 1 M phosphate buffer pH 7 . 4 , post-fixed in osmium tetroxide , dehydrated in ethanol and treated with propylene oxide prior to embedding in Spurr’s epoxy resin . Thin sections were stained with uranyl acetate and lead citrate prior to examination using a Technai 20 electron microscope ( FEI Company ) . Samples for EM were prepared twice independently and multiple thin sections for each sample were examined . Trail deposition assays have been performed as described before [6] . Briefly , freshly released parasites were allowed to glide on Poly-L-Lysine-coated glass slides for 15 min at 37°C before they were fixed with PFA/GA and stained with SAG1 . myoH-iKD parasites were treated for 48 h ± ATc . The percentage of produced trails was determined by counting the number of anti-SAG1 labelled trails for 1000 parasites . Data are presented as percentage mean values ± SD from three independent experiments . The significance of the data was evaluated using a parametric paired t-test and the two-tailed p-value is written on the corresponding graph . Freshly egressed parasites were attached to Poly-L-Lysine-coated coverslips and either treated with 10 mM deoxycholate for 10 min at room temperature . Parasites were then fixed with PFA/GA for 10 min and then proceeded as for IFAs . DD-MyoH-NT parasites were pre-treated for 24 h with Shield-1 . T . gondii tachyzoites freshly egressed from host cells were harvested by centrifugation at 240 g , RT for 10 min and the pellet washed twice in intracellular buffer ( 5 mM NaCl , 142 mM KCl , 1 mM MgCl2 , 2mM EGTA , 5 . 6 mM glucose and 25 mM HEPES , pH 7 . 2 ) prewarmed to 37°C . Control parasites were treated 30 min at room temperature with 1 μM of compound 2 [34 , 35 , 66] to prevent microneme secretion . Pellets were resuspended in egress buffer ( 141 . 8 mM NaCl , 5 . 8 mM KCl , 1 mM MgCl2 , 1mM CaCl2 , 5 . 6 mM glucose and 25 mM HEPES , pH 7 . 2 ) ± ethanol ( 2% ) , incubated 30 min at 37°C . Parasites were centrifuged at 1000g for 5 min , 4°C . Pellets were washed once in PBS and stored before Western blot . Supernatants were centrifuged at 2000 g for 5 min , 4°C and supernatant was used as ESA ( excreted/secreted antigen ) . Pellets and ESA samples were analysed for AMA1 , MIC2 , MIC6 , MIC4 , Myc , catalase and dense granule ( GRA1 ) by Western blot . For Δku80 and myoh-iKD , parasites were pre-treated for 48 h ± ATc . | The Apicomplexa phylum groups important pathogens that infect humans and animals . Host cell invasion and egress from infected cells are key events in the lytic cycle of these obligate intracellular parasites . Host cell entry is powered by gliding motility and initiated by the discharge of apical secretory organelles at the site of contact with the host cell . Anchored to the parasite pellicle , the glideosome composed of myosin A and the gliding associated proteins is the molecular machine which translocates the secreted adhesins from the apical to the posterior pole of the parasite and hence propels the parasite into the host cell . Toxoplasma gondii exhibits a helical form of gliding motility and as member of the coccidian-subgroup of Apicomplexa possesses an apical organelle called the conoid , which protrudes during invasion and egress and consists in helically organized polymer of tubulin fibers . We have deciphered here the function of a novel myosin associated to the microtubules composing the conoid . Myosin H is essential and prerequisite for motility , invasion and egress from infected cells . This unusual motor links actin- and tubulin-based cytoskeletons and uncovers a direct role of the conoid in motility and invasion . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2016 | The Conoid Associated Motor MyoH Is Indispensable for Toxoplasma gondii Entry and Exit from Host Cells |
Dengue virus affects millions of people worldwide each year . To date , there is no drug for the treatment of dengue-associated disease . Nucleosides are effective antivirals and work by inhibiting the accurate replication of the viral genome . Nucleobases offer a cheaper alternative to nucleosides for broad antiviral applications . Metabolic activation of nucleobases involves condensation with 5-phosphoribosyl-1-pyrophosphate to give the corresponding nucleoside-5’-monophosphate . This could provide an alternative to phosphorylation of a nucleoside , a step that is often rate limiting and inefficient in activation of nucleosides . We evaluated more than 30 nucleobases and corresponding nucleosides for their antiviral activity against dengue virus . Five nucleobases and two nucleosides were found to induce potent antiviral effects not previously described . Our studies further revealed that nucleobases were usually more active with a better tissue culture therapeutic index than their corresponding nucleosides . The development of viral lethal mutagenesis , an antiviral approach that takes into account the quasispecies behavior of RNA viruses , represents an exciting prospect not yet studied in the context of dengue replication . Passage of the virus in the presence of the nucleobase 3a ( T-1105 ) and corresponding nucleoside 3b ( T-1106 ) , favipiravir derivatives , induced an increase in apparent mutations , indicating lethal mutagenesis as a possible antiviral mechanism . A more concerted and widespread screening of nucleobase libraries is a very promising approach to identify dengue virus inhibitors including those that may act as viral mutagens .
Dengue virus ( DENV ) is a worldwide health threat , with hundreds of millions of people infected yearly in more than 100 countries [1] . There are four known DENV serotypes and a first infection with one serotype followed by a second infection with another serotype may result in severe disease [2 , 3] . For these and other issues , vaccines designed for a pan-serotype protection , including the commercial dengue vaccine approved and used in a few countries , have yielded mixed results [4 , 5] . Safety and partial efficacy concerns in addition to cost , storage and delivery issues may hinder implementation of vaccines in many countries . There are currently no approved drugs to treat DENV infection . Thus far , classical antiviral approaches ( e . g . NS5 polymerase inhibitors , entry inhibitors , protease inhibitors , etc . ) have yet to provide treatments for DENV infection and therefore the investigation of new antiviral strategies is warranted [6–8] . One such strategy to explore is lethal mutagenesis [9] . The idea of viral lethal mutagenesis is to exploit the natural tendency of RNA viruses to mutate in order to favor the accumulation of deleterious mutations in the newly formed viruses , eventually leading to viral extinction ( for review see [10] ) . DENV and other RNA viruses display a high mutation rate ( 10−4 to 10−6 mutations per bp per generation ) [11 , 12] as an evolutionary characteristic allowing these viruses to escape host immune defense mechanisms and adapt rapidly to new stress conditions [13 , 14] . An error-prone viral polymerase combined with a high replication rate are considered to be the main sources of mutations . It is this critical source of viral adaptability ( e . g . the virus high mutation rate ) that makes RNA viruses a target of choice for antiviral lethal mutagenesis strategies [15–17] . RNA viruses maintain a delicate balance between their need to adapt and their need to preserve a level of genetic integrity put at risk by deleterious mutations [18] . Modifying this fragile equilibrium by increasing the viral mutation rate with mutagens has been proposed as an antiviral strategy [15] . The well-known antiviral nucleoside drug ribavirin induces lethal mutagenesis for different viruses [19–22] . The discovery of new nucleosides as antiviral mutagens has been impaired by several hurdles including the toxicity of the potential drugs as well as synthetic challenges . In addition , potentially mutagenic nucleoside analogues frequently suffer from poor metabolic conversion to the active triphosphate form required by the viral polymerase . The first phosphorylation of the nucleoside analogue is often the rate limiting step to obtain the active mutagenic nucleoside triphosphate used by the viral polymerase [23 , 24] . In order to overcome the potential first phosphorylation difficulty , we propose to use nucleobases ( purine or pyrimidine base without the ribose or phosphate moieties of a nucleoside ) . Enzyme mediated condensation of nucleobases with 5-phosphoribosyl-1-pyrophosphate to give the corresponding nucleoside-5’-monophosphate can provide an alternative pathway . Thus , for a nucleoside where the first phosphorylation is inefficient then using its corresponding nucleobase could allow metabolic conversion to the corresponding nucleoside triphosphate , thereby providing a more efficient metabolic conversion to the triphosphate . Studies on nucleobases have been limited [25] , with the notable exceptions of 5-fluorouracil and T-705 ( favipiravir ) [26–32] . However , in the context of targeting viruses that affect developing countries , mutagenic nucleobases present key advantages over mutagenic nucleosides . In addition to their different metabolic activation pathways , nucleobase analogues are considerably cheaper , more diverse and commercially available in higher numbers compared to corresponding nucleoside analogues . The chemical synthesis of a nucleobase is faster and simpler than the synthesis of the corresponding nucleoside . Similarly to nucleosides , nucleobases possess their own cellular transporters [33 , 34] . In this study , we describe the identification of five nucleobases and three corresponding nucleosides that possess potent anti-DENV activity . These compounds have not been previously described to have anti-DENV activity , except for the nucleoside 1b ( ribavirin ) [35 , 36] . We compared the antiviral activities and toxicities of the nucleobases with their corresponding nucleosides . For virus passaged in the presence of a nucleobase 3a ( T-1105 ) or nucleoside 3b ( T-1106 ) , we detected an increase in mutations compared to virus passaged in DMSO indicating a possible reduction in virus titer via increased mutagenesis . To our knowledge , our study is the first to fully compare the antiviral mechanisms and efficacies of a nucleobase and its corresponding nucleoside , highlighting the differences , similarities and potential advantages of nucleobases versus nucleosides . Our study also highlights the potential of lethal mutagenesis induction during DENV replication as an alternative to classical antiviral strategies .
5a ( T-705 ) ( CAS# 259793-96-9 , 6-fluoro-3-hydroxypyrazine-2-carboxamide ) was purchased from ASTA Tech . 3a ( T-1105 ) ( CAS# 55321-99-8 , 3-Hydroxy-2-pyrazinecarboxamide ) was purchased from Alfa Aesar . 3b ( T-1106 ) was synthesized according to known procedures ( Preparation of nucleosides with non-natural bases as anti-viral agents Can . Pat . Appl . ( 2006 ) , 149pp . CODEN:CPXXEB; CA2600359 ) . 1a ( ribavirin base ) ( CAS# 3641-08-5 , 1 , 2 , 4-Triazole-3-carboxamide ) was purchased from Ark Pharm . 1b ( ribavirin ) ( CAS# 36791-04-5 ) was purchased from Carbosynth . 2a ( mizoribine base ) ( CAS# 56973-26-3 , 5-Hydroxy-1H-imidazole-4-carboxamide ) was purchased from Ark Pharm . 2b ( mizoribine ) ( CAS# 50924-49-7 ) was purchased from Carbosynth . 4a ( diaminopurine ) ( CAS# 1904-98-9 , 2 , 6-diaminopurine ) was purchased from Sigma-Aldrich . 4b ( diaminopurine riboside ) ( CAS# 2096-10-8 , 2-Aminoadenosine ) was purchased from Berry and Associates . 6 ( mycophenolic acid ) ( CAS# 24280-93-1 ) was purchased from Sigma-Aldrich . The hepatocyte-derived cellular carcinoma cell line Huh-7 [37] was used for DENV infection and drug treatment . The African green monkey kidney Vero cell line ( ATCC CRL-81 ) was used to titer DENV via plaque assay . The baby hamster kidney cell line carrying a DENV subgenomic replicon , BHK pD2-hRucPac-2ATG30 [38] ( obtained from Dr . M . Diamond , Washington University , School of Medicine ) , was used for DENV replicon assay . All cell lines were maintained in Dulbecco’s modified Eagle’s ( DME ) medium supplemented with 10% fetal bovine serum ( FBS ) , 100 IU streptomycin/penicillin per ml and 10 μg/mL plasmocin ( InvivoGen ) at 37°C in a 5% CO2 incubator . DENV replicon cells were supplemented with 3 μg/mL puromycin ( Life Technologies ) . DENV-2 stocks from New Guinea C strain ( ATCC VR-1584 ) were generated from C6/36 mosquito cell cultures ( ATCC CRL-1660 ) grown in Minimum Essential Medium ( MEM ) supplemented with 10% FBS , 1% non-essential amino acids and 1% sodium pyruvate at 28°C with 5% CO2 . The C6/36 cells on T-150 flasks were inoculated with virus and the supernatant harvested after complete cytopathic effects . Viral stock titers were determined by plaque assay on Vero cells . The sensitivity of the cell lines to the compounds was examined using the 3- ( 4 , 5-dimethylthiazol-2-yl ) -5- ( 3- carboxymethoxyphenyl ) -2- ( 4-sulfophenyl ) -2H-tetrazolium ( MTS ) -based tetrazolium reduction CellTiter 96 Aqueous Non-Radioactive cell proliferation assay ( Promega G5430 ) . The compounds were initially tested at 10 and 50 μM final concentrations . Each plate also contained DMSO alone , medium alone , and an inhibitory compound , 6 . DENV replicon or Huh-7 cells were plated at a density of 1 , 500 or 8 × 103 cells , respectively , per well in 96-well plates containing 100 μl of culture medium overnight . Compounds were added to triplicate wells in culture medium and incubated for an additional 72 h . MTS reagent was then added to each well and incubated at 37°C in a humidified 5% CO2 atmosphere . The plates were read at various time points at a wavelength of 490 nm using a Molecular Devices M5e plate reader . Mean values of triplicate wells were determined and compared to the mean value for the wells that received DMSO alone . For compounds selected for dose-response experiments , the CC50 was determined by comparing cell viability for eight serial dilutions of the compound and DMSO treated cells using GraphPad Prism software . The CC50 value was defined as the compound concentration resulting in a 50% reduction readout compared with the DMSO . Compounds were evaluated for antiviral properties using BHK cells containing a DENV-2 viral replicon . 1 . 5 × 103 replicon-containing cells per well were plated in white opaque 96-well plates in the absence of antibiotic selection and the next day , compounds dissolved in DMSO were added to triplicate wells in culture medium . The compounds were initially tested at 10 and 50 μM final concentrations and each plate also contained DMSO alone , medium alone , and 6 . Three days later , medium was replaced with a 1:1000 dilution of ViVi-Ren Live Cell Substrate ( Promega ) in DME minus phenol red and 10% FBS . Luminescence was measured with a Molecular Devices M5e plate reader . Mean values of triplicate wells were determined and compared to the mean value for the wells that received DMSO alone . For compounds selected for dose response experiments , the concentration of compound that reduced luciferase activity by 50% was defined as the 50% effective concentration ( EC50 ) . The EC50 was determined by comparing luciferase activity for eight serial dilutions of the compound and DMSO treated cells using GraphPad Prism software . Huh-7 cells were seeded in 12-well plates at a density of 4×105 cells per well in 1 mL culture medium . The next day , cells were washed and inoculated with DENV at a multiplicity of infection ( m . o . i . ) of 0 . 2 in 500μl infection medium ( MEM containing 2% FBS and 10 mM HEPES ) . The inoculum was removed after 1h , cells were washed with PBS and then incubated in 1 mL MEM , 2% FBS , 1% pen/strep plus compound for 72 h . Viral supernatants were clarified by centrifugation for 5 min at 1500×g and aliquoted and stored at -80°C . Viral titers were determined using a plaque assay on Vero cells . Briefly , confluent Vero cell monolayers in 24-well plates were incubated at 37°C for 1 h with duplicate 300 μl samples of 10-fold serial dilutions of viral supernatants . The cells were then washed to remove unbound viral particles and overlaid with 500ul MEM containing 1 . 3% methylcellulose , 5% FBS and 10mM HEPES . After 5 days of incubation at 37°C and 5% CO2 , cells were washed with PBS , fixed , and stained using 1% Giemsa . Infectious virus titer ( pfu/mL ) was determined using the following formula: number of plaques × dilution factor × ( 1/inoculation volume ) . The viral titer was presented as the mean of duplicate samples from a dilution yielding approximately 20–50 plaques per well . For qPCR determination of viral genome copy number , viral RNA was isolated from 140 μL of drug-treated or DMSO-treated infected culture supernatant using QIAamp Viral RNA mini kit ( Qiagen ) , following manufacturer’s protocol . Viral RNA was quantified using the TaqMan RNA-to-CT 1-Step qPCR Kit ( Applied Biosystems ) . Primers used for qPCR were 5’-CATGATGGGAAAAAGAGAGAAGAAGCT-3’ ( forward ) and 5’-GGCTCTGCTGCCTTTTGC-3’ ( reverse ) amplifying a region numbering 8928–8988 in the genome ( numbering starting from the beginning of genome , accession number KM204118 ) . The qPCR FAM probe sequence is 5’-TTGCCGAACTCCCC-3’ . Serial 10-fold dilutions of plasmid containing the NS5 gene of DENV were used to generate a standard curve for the quantification of viral RNA genome copy number based on cycle threshold ( CT ) values . The limit of detection for NS5 plasmid dilutions was 30 copies ( S1 Fig ) . One-way ANOVA was performed to determine statistical significance of mean genome copy numbers among treatments at each virus passage and Tukey’s honestly significant difference ( HSD ) used to determine statistical significant ( p < 0 . 05 ) between DMSO and 3a or 3b at each passage . Statistical analysis via GraphPad Prism 5 software . To obtain sequence data from viral RNA isolated at each passage , cDNA was generated via M-MLV reverse transcriptase and random hexamers ( New England Biolab Inc . ) per manufacturer’s instructions . An approximately 1600-base fragment covering membrane protein ( prM ) and envelop protein gene ( E ) of DENV-2 viral genome was amplified using PfuUltra II Fusion HS DNA polymerase ( Agilent ) with primers prMEfor ( 5'- AACTCAGAATTCTTCCATTTAACCACACGTAAC-3' ) and prMErev ( 5'- AACTCAGAATTCTCCTTTCTTAAACCAGTTGAG -3' ) . PCR products were purified using Qiaquick PCR purification kit ( Qiagen ) and then digested with EcoRI and ligated into pcDNA3 . 1 for sequencing . Sequence for approximately 40–50 individual clones per sample was obtained from the University of Minnesota Genomics Center . Sequences were aligned over a 980-base region that had adequate quality sequencing reads for all clones . Only mutations present in both the forward and reverse reads of a clone were counted . All incidence-based determinants of mutation frequency ( Mf min , Mf max , Mfe ) were calculated as described [39] . A two-tailed Mann-Whitney U test ( GraphPad Prism 5 software ) was used to determine if there were statistically significant differences for the mean number of mutations per clone between DMSO-treated and each drug-treated virus passage . The DENV-2 viral RNA was isolated from 140 μL of lab stock DENV-2 using QIAamp Viral RNA mini kit ( Qiagen ) . cDNAs corresponding to viral RNAs were generated with random hexamers ( New England Biolab Inc . ) . A 2600-base fragment of NS5 cDNA was amplified using PfuUltra II Fusion HS DNA polymerase ( Agilent ) with primers NS5for ( 5'- GGCCAGTGCCAAGCTTGAACTGGCAACATAGGAAGAACGC-3' ) and NS5Rev ( 5'- CCGGGGATCCTCTAGACCACAGGACTCCTGCCTCTT -3' ) . PCR product were purified using Qiaquick PCR purification kit and inserted into a XbaI and HindIII digested pUC18 vector using In-Fusion HD Cloning Kit ( Clontech ) . A positive clone was identified and the nucleic acid sequence of NS5 confirmed by sequencing at the University of Minnesota Genomics Center . Virus was passaged on Huh-7 cells supplemented with 200 μM 3a , 500 μM 3b or DMSO ( 0 . 5% ) . Huh-7 cells were seeded in 12-well plates and inoculated with DENV-2 as described for Titer reduction experiments above except an m . o . i . of 0 . 01 was used . After 3 days of compound treatment , 50 μL of the harvested supernatant was used to inoculate fresh Huh-7 cells in the continued presence of compound . The virus titer in the harvested supernatant was determined by plating ten-fold serial dilutions onto single wells of a 24-well plate . The wells were washed and overlaid as described in the Titer reduction assay above . If no plaques were obtained in any of the harvested supernatant dilutions , the undiluted supernatant was used . If plaques were not detectable , the virus titer was considered to be at the limit of detection , 1 plaque ( 3 . 3 pfu/mL ) . Supernatant titer was determined from a single well where there were approximately 20–50 plaques when possible . This experiment was repeated three times at the compound concentrations listed .
Antiviral nucleoside identification can be hindered by difficulties in chemical synthesis and poor conversion of the nucleoside to the active triphosphate form . To circumvent these issues , we propose to use nucleobases in our initial screen for antiviral agents because of their different activation pathway to the active nucleotide ( Fig 1 ) , their low cost and ready commercial availability . Phosphoribosyl transferases of the cellular nucleotide salvage pathway directly convert some nucleobases to the corresponding nucleoside monophosphate and therefore the corresponding nucleoside analogue need not be an efficient substrate for a nucleoside kinase ( Fig 1 ) [40 , 41] . In that regard , 3a and analogue 5a ( Fig 2 ) are substrates of human phosphoribosyl transferases and are converted in one step to the corresponding nucleoside monophosphate [41] . Our strategy to identify nucleobase and nucleoside DENV inhibitors was to screen for activity and toxicity of selected compounds at 10 μM and 50 μM using a luciferase-reporting DENV replicon cell line , BHK pD2-hRucPac-2ATG30 [38] . Compounds that demonstrated inhibitory activity against the replicon cell line were used in dose-response analysis to assign EC50 and CC50 values . The nucleobases were generally more active with a higher tissue culture therapeutic index ( CC50/EC50 ) than their corresponding nucleosides ( Fig 2 ) . The EC50 values of the active nucleobases range from 2 . 4 to 110μM , comparable to the EC50 values of the active nucleosides that range from 1 . 3 to 113μM ( Fig 2 ) . Nucleobase 3a is 5 times more active than nucleobase 5a ( favipiravir ) . The CC50 values of the nucleobases 1a , 3a and 5a were beyond 665μM ( Fig 2 ) . Remarkably , nucleobase 1a did not show cytotoxicity at 1000 μM compared to 1b nucleoside ( Fig 2 ) where the CC50 was 20 μM . 2a nucleobase displayed a clear antiviral effect at 2 . 4 μM . Representative EC50 curves for 2a plus corresponding nucleoside 2b and for 3a plus corresponding nucleoside 3b are shown in Fig 3 . Inactive nucleobases at initial screening are listed in S1 Table in the supplementary material . We focused on the antiviral mechanism of nucleobase 3a and corresponding nucleoside 3b for the following reasons: 1 ) among the 5 active nucleobases , 5a was already known to induce viral lethal mutagenesis but its corresponding nucleoside was not available and unable to be synthesized for our study; 2 ) 1a and 2a are likely to possess a complex mechanism of action due to their probable inhibitory effect of inosine monophosphate dehydrogenase ( IMPDH ) after conversion to the nucleotide form; 3 ) 4b was too toxic to perform a full study comparing the nucleobase to the nucleoside ( Fig 2 ) . Therefore , we chose 3a , known to be substrate of human phosphoribosyl transferase [41] , and its corresponding nucleoside 3b to be the best candidates for antiviral mechanism of action studies and to compare the effects of nucleobase and corresponding nucleoside . Our approach to study mechanism of action and possible lethal mutagenesis was to passage virus in Huh-7 cells in the presence of a compound and determine the compound’s effect on virus titer , genome copy number and genome sequence . We had initially identified inhibitors using a DENV replicon BHK cell line so we wanted to verify inhibitory activity in Huh-7 cells using replication competent DENV . We chose Huh-7 cells because they are a human cell line , they are susceptible to DENV infection and they produce readily detectable infectious virus . The BHK replicon cells are a convenient tool to identify initial DENV inhibitors . We used human cells for more detailed studies for these compounds that require conversion to the active form by cellular enzymes . We repeated the dose-response experiments for 3a and 3b using a titer-reduction assay with Huh-7 cells as previously described [42] . The values obtained for the compounds ( Table 1 ) were consistent with those from the replicon assay . Based upon the data in Table 1 , we used non-toxic levels of 3a ( 200 μM ) and 3b ( 500 μM ) that we empirically determined ( S2 Fig ) would reduce virus replication steadily during multiple passages and ultimately lead to undetectable levels of infectious virus . The results for virus passage in the presence of 3a ( 200 μM ) , 3b ( 500 μM ) or DMSO ( 0 . 5% ) are shown in Fig 4 . Compounds were added to Huh-7 cells shortly after inoculation . After 3 days incubation in compound , the supernatant was collected and a fixed volume ( 50 μL ) of supernatant was used to inoculate fresh cells . The level of infectious virus present in the supernatant for each passage was determined by plaque assay and the number of viral genomes by RT-qPCR . For the RT-qPCR , we chose primers within the NS5 gene to increase likelihood of detecting full-length genomes . Both compounds induced a steady decline in infectious virus production and supernatant genome copy number over repeated passages ( Fig 4 ) . In the presence of 3b , infectious virus was undetectable at passages 4 and 5 and genomic RNA was undetectable at passage 5 ( Fig 4 ) . For 3a , no infectious virus was detected at passages 5 and 6 and no genomic RNA was detected at passage 6 ( Fig 4 ) . The mean genome copy numbers ( Fig 4B ) for the three treatments were statistically different at each passage ( p < 0 . 001 , One Way ANOVA ) with statistically significant differences between 3a and DMSO and between 3b and DMSO at each passage ( p < 0 . 05 , Tukey’s HSD ) . Both compounds displayed a reduction in the ratio of infectious virus to RNA genome copy number ( sometimes referred to as RNA specific infectivity ) when compared to DMSO ( Fig 4 ) . This delay in reduction of genomic RNA compared to infectious virus is a hallmark of a mutagenesis-based antiviral activity [43–46] . These observations , along with the knowledge that 5a ( analogue of 3a and 3b ) induces mutagenesis in influenza A , norovirus and the flavivirus West Nile virus [27–29] , motivated us to determine if a mechanism of action for 3a or 3b included an enhanced mutagenesis of the viral genome . We hypothesized that an increase in mutations induced by a particular nucleobase or nucleoside would be detected by analyzing the viral genome sequence at a passage near where the titer was significantly reduced or undetectable . Therefore , we amplified a region containing the pre-membrane ( prM ) and envelope ( E ) genes from viral genomic cDNA derived from passage 3 3b-treated cells and passage 4 3a-treated cells because those passages were just before viral titer was undetectable . We amplified the prM/E region of the viral RNAs instead of the downstream NS5 gene to increase the likelihood we would obtain PCR products when viral titers were greatly reduced . The presence of the NS5 gene would require an almost complete genomic RNA whereas the prM/E region was closer to the 5’ end of the viral genomic RNA its presence would not require a complete viral genome . The amplified products were inserted into a cloning vector and the nucleotide sequence was analyzed for the resulting 35–50 independent clones . We sequenced at least 31 , 000 nucleotides for each set of clones similar to other published studies of viral lethal mutagenesis [27 , 43] and compared the sequences to that of the consensus sequence obtained from DMSO-treated cell supernatants at passages 3 and 4 . An initial plot and analysis of the number of mutations per clone for 3a treatment passage 4 and 3b passage 3 clearly indicate there was a significant difference ( p < 0 . 0001 ) in the number of mutations per clone for sequence obtained from compound-passaged virus compared to DMSO ( Fig 5A ) . We next calculated a number of incidence-based mutation indices to better determine possible heightened mutagenesis as recommended in a recent review [39] . There was an increase in every index of mutation frequency and sequence diversity for 3a- and 3b-treated clones compared to those for DMSO ( Table 2 ) . These include , the minimum mutation frequency ( Mf min , mutation at a given nucleotide counted only once ) , maximum mutation frequency ( Mf max , all mutations counted ) and the entity level mutation frequency ( Mfe , mutations with respect to sequence of dominant haplotype of drug-treated clones ) . In addition , the number of haplotypes , number of different mutations , number of total mutations and number of clones with a mutation ( Table 2 ) were all higher for 3a and 3b passage sequences . We noted fewer mutations in the DMSO passage 4 sequence than seen for other DMSO passages ( Table 2 and Fig 5B ) . That does not change our interpretation of the 3a passage 4 results . The mean number of mutations per clone for 3a passage 4 was significantly higher than the mean mutations per clone for any DMSO passage ( ranging from p < 0 . 01 to p < 0 . 03 ) . In addition , all advanced indices of mutation frequency and sequence diversity were greater for 3a passage 4 when compared to all DMSO passages ( Table 2 ) . Lastly , we examined sequence data from other 3a- and 3b-treated passages . Sequence from passage 2 also showed higher numbers of mutations per clone and an increase in all advanced mutation indices for 3a- and 3b-passaged virus compared to DMSO ( Fig 5 and Table 2 ) . The sequence data from 3a passage 3 yielded mixed results with certain measures of mutation frequency similar to DMSO and certain others higher than those for DMSO ( Table 2 and Fig 5A ) . We were unable to obtain sequence data for 3b passage 4 or 5 . Taken together , these results indicate an enhanced mutagenesis as a likely significant contributor to the antiviral mechanism of action for 3a and 3b . In the sequence analysis , we excluded nucleotides 952–4 ( numbering starting from the beginning of genome , accession number KM204118 ) in analysis of mutations . This codon , located in the E gene , changes in response to repeated DENV passaging in cultured cells [47 , 48] . The sequence for nucleotides 952–4 of our stock virus was ATA . However , nearly all the clones from virus passaged with DMSO , 3a and 3b displayed changes in that sequence ( S2 Table ) . We also excluded mutations that were present in a majority of clones and therefore were consensus sequence changes . These included one nucleotide position ( 1043 ) in the 3a clones and three positions ( 960 , 1057 and 1493 ) in the 3b clones ( S2 Table ) . A hallmark of lethal mutagenesis is an unchanged consensus sequence despite an increase in single mutations [44 , 45] and so it is possible those changes were introduced during the RT-PCR amplification of the viral RNA genomes . An examination of the types of mutations induced by nucleoside 3b revealed an almost exclusive increase in transition mutations for purines and pyrimidines ( Table 2 ) . For nucleobase 3a , a similar increase in mostly transition mutations was observed ( Table 2 ) . These results are also consistent with data observed with related compound 5a [27 , 28] .
Effective and inexpensive drug therapies for DENV infection are urgently needed . So far , classical antiviral strategies have failed to identify small molecules to treat DENV infection . In addition , developing broad antiviral strategies is critical to face future outbreaks of emerging viruses . The adaptability of viruses to changing environments results in the generation of a heterogeneous population of closely related yet different viral variants during infection . The diversity present within this population may result in viruses escaping drug treatments or host immune defenses . Lethal mutagenesis aims to generate deleterious viral mutations that would prevent viral adaptation and drive the viral population to collapse . Induction of lethal mutagenesis has largely been based on using nucleoside mutagens to target the error prone viral polymerase , a main source of viral mutations . However , limitations in the chemical synthesis of nucleosides and poor metabolism to the active triphosphate form have limited the discovery of new viral mutagens . Here , we have identified nucleobases and nucleosides with significant antiviral activity against DENV , some of which appear to act through lethal mutagenesis . Along with nucleosides , nucleobases may be valuable molecules to induce viral lethal mutagenesis or other antiviral effects . For DENV antiviral therapy , nucleobases are cheaper and display a better therapeutic index in cell culture compared to nucleosides . This result warrants further exploration . Overall , our study supports antiviral lethal mutagenesis as a potentially effective strategy to target DENV . In this study , we have identified five nucleobases and three nucleosides that possess anti-DENV activity . Nucleobase 3a and corresponding nucleoside 3b were selected for antiviral mechanism-of-action studies . 3a and 3b both increased mutations . Although the antiviral mechanisms of 5a , 1b and 4b have not been explored in this study , these compounds are known antiviral mutagens [22 , 25 , 28] supporting our result that viral lethal mutagenesis may be an effective antiviral strategy against DENV . The strong antiviral effects of 2b ( EC50 15μM ) and its nucleobase 2a ( EC50 2 . 4μM ) suggest a critical role of IMPDH for DENV replication , yet the fact that 2a and 2b possess a rotatable amide bond ( theoretically able to mimic either adenosine or guanosine ) indicates a component of the antiviral effect could occur through lethal mutagenesis . To our knowledge , this is the first description of 2a as an antiviral agent . Interestingly , in our study , the active nucleobases and nucleosides ( Fig 2 ) share common structural characteristics such as 1 ) a purine analog structure and/or 2 ) a rotatable amide bond . Some of these nucleobases have been described to be converted metabolically to the active nucleotide form [41] . Our study is the first to compare a nucleobase side by side with its corresponding nucleoside for their abilities to induce viral genome mutagenesis as an antiviral mechanism of action . The way by which mutagenic nucleobases/nucleosides and resulting mutagenic nucleotides induce mutations is usually due to ambiguous base pairing misinterpreted by the viral polymerase when replicating the viral genome [49] . Typically , ambiguous base pairing originates from rotational or tautomeric forms of the base that result in the mutagenic nucleoside triphosphate resembling more than one natural nucleoside triphosphate . In that matter , 5a , an analogue of 3a and 3b , was found to be converted to the active triphosphate form and then used by influenza polymerase as an A or G mimic , likely by rotation of its amide bond [50] . In a similar fashion , 1b ( ribavirin ) possesses a rotatable amide bond and is widely accepted as an A or G mimic . Overall , 3a and 3b in this study and 5a and 1b in previous studies all essentially induce transition mutations [22 , 27] , a result compatible with these molecules resembling A or G . The 3b triphosphate would manifest as an ATP or GTP mimic and the rotation of its amide bond would allow base pairing with a U or C , respectively ( Fig 6 ) . We did not observe a clear preference for A to G and U to C transition mutations versus the G to A and C to U mutations indicating that the DENV polymerase may not have a preference for using 3b triphosphate . The proportions of A to G or G to A transitions induced by these mutagens can vary from virus to virus [27 , 28] . The affinity of the different viral polymerases for one conformation of a mutagen rather than another may explain the observed variance . A mutagen such as 1b also inhibits IMPDH , an enzyme involved in nucleotide biosynthesis , resulting in an imbalance in endogenous nucleotide pools which may impact the induced mutation spectrum compared to a purely mutagenic compound . In our case , we did not observe major trends for one particular transition mutation versus another . Exploring molecules capable of inducing transversion mutations may be of interest to induce a stronger antiviral effect . | Dengue virus is a world-wide public health menace estimated to infect hundreds of millions of people per year . Vaccines to prevent dengue virus infection have had limited success due in part to the requirement to elicit effective immune responses against the four dengue serotypes . There is an urgent unmet need for anti-dengue virus therapies . Nucleosides are effective antiviral small molecules which usually work by inhibiting the accurate replication of the viral genome . Typically , nucleosides must be converted within the cell to their triphosphate form to inhibit virus replication , thus inefficient phosphorylation often leads to suboptimal activity . We screened a small library of nucleobases that require an activation pathway different from nucleosides to achieve the same active form . We identified some known and previously undescribed dengue virus nucleobase inhibitors and their corresponding nucleosides . Our investigation of the mechanism of action of one nucleobase and its corresponding nucleoside found evidence for enhanced mutagenesis of the dengue virus genome in the presence of the compounds in cell culture . A wide screening of nucleobases libraries is a promising strategy to discover dengue virus inhibitors including potential viral mutagens . | [
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] | [
"dengue",
"virus",
"glycosylamines",
"microbial",
"mutation",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"nucleobases",
"pathogens",
"microbiology",
"nucleotides",
"viruses",
"rna",
"viruses",
"viral",
"genome",
"microbial",
"genomics",
"viral",
"genomics",
"medical",
"microbiology",
"microbial",
"pathogens",
"mutagenesis",
"biochemistry",
"flaviviruses",
"virology",
"nucleosides",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"glycobiology",
"organisms"
] | 2018 | Nucleobases and corresponding nucleosides display potent antiviral activities against dengue virus possibly through viral lethal mutagenesis |
Local and spontaneous calcium signals play important roles in neurons and neuronal networks . Spontaneous or cell-autonomous calcium signals may be difficult to assess because they appear in an unpredictable spatiotemporal pattern and in very small neuronal loci of axons or dendrites . We developed an open source bioinformatics tool for an unbiased assessment of calcium signals in x , y-t imaging series . The tool bases its algorithm on a continuous wavelet transform-guided peak detection to identify calcium signal candidates . The highly sensitive calcium event definition is based on identification of peaks in 1D data through analysis of a 2D wavelet transform surface . For spatial analysis , the tool uses a grid to separate the x , y-image field in independently analyzed grid windows . A document containing a graphical summary of the data is automatically created and displays the loci of activity for a wide range of signal intensities . Furthermore , the number of activity events is summed up to create an estimated total activity value , which can be used to compare different experimental situations , such as calcium activity before or after an experimental treatment . All traces and data of active loci become documented . The tool can also compute the signal variance in a sliding window to visualize activity-dependent signal fluctuations . We applied the calcium signal detector to monitor activity states of cultured mouse neurons . Our data show that both the total activity value and the variance area created by a sliding window can distinguish experimental manipulations of neuronal activity states . Notably , the tool is powerful enough to compute local calcium events and ‘signal-close-to-noise’ activity in small loci of distal neurites of neurons , which remain during pharmacological blockade of neuronal activity with inhibitors such as tetrodotoxin , to block action potential firing , or inhibitors of ionotropic glutamate receptors . The tool can also offer information about local homeostatic calcium activity events in neurites .
Calcium ions mediate fast signaling to regulate neuronal development , synaptic transmission , and synaptic plasticity [1–5] . Calcium imaging has become a standard technique for mapping neuronal activity in neurons in vitro and in vivo [6 , 7] . Although calcium signaling of neurons is well investigated , calcium-dependent mechanisms underlying spontaneous or cell-autonomous excitability are not well understood [8–11] . Investigating the molecular mechanisms underlying spontaneous calcium influx revealed two principle mechanisms of how spontaneous excitation is initiated . Either spontaneous excitation is ligand-dependent and caused by the non-synaptic release of transmitters such as glutamate or GABA [12 , 13] . Alternatively , excitability is part of a developmental program and is triggered by the neuron itself , meaning by cell-autonomous excitation using subthreshold active ion channels , or is caused by self-enhancement of intrinsic excitability through autocrine signaling [10 , 14 , 15] . Furthermore , homeostatic calcium fluxes are cell-autonomously controlled and occur in cellular subdomains [16] . Three major explanations of why signaling of spontaneous or cell-autonomous excitability is not well investigated can be proposed: ( 1 ) Calcium transients can appear at very local places , and their spatiotemporal footprint is not predictable . ( 2 ) Proteins involved in neuronal excitation show a high functional diversity , depending on their locus of action . ( 3 ) Neuronal signals can be very fast and ‘small’ , thus making it difficult to identify real signaling events due to the unavoidable measurement noise . Techniques to extract calcium signals from imaging data are commonly region of interest ( ROI ) analyses . ROIs are typically selected manually or in a semi-automated manner [6 , 17 , 18] . This approach is time-consuming in assessing the spatiotemporal activity pattern of neurons . Moreover , the selection of ROIs and the criteria to define activity events is rather heuristic and largely user-dependent [17 , 18] . Signal identification , not only in calcium imaging data , but also in mass spectrometric data , is often performed using noise filtering and peak identification algorithms . However , this strategy creates a high false-positive rate , especially when low amplitude peaks are analyzed [19–21] . Calcium signal identification and signal source separation are two separate problems in calcium signal analysis . To support the analysis of imaging data many helpful automated or semi-automated computational tools have been developed [18 , 21–28] . Therefore , advanced computational tools are helpful to localize , extract and sort cellular signals to detect neuronal calcium activity [18] or to strengthen the detection of calcium activity by signal demixing and denoising [22 , 23 , 26] . Multiple arithmetical strategies to read out calcium signals from raw data exist , and each of these methods has its advantages [18 , 21 , 23 , 29] . However , strategies to identify local activity close to the noise level with high sensitivity are not well developed . There is a certain risk that activity events , which are not spike-like , or are very close to the measurement noise , may be ignored by computational tools using denoising , or feature-dependent methods for calcium time series analysis . This can be of advantage , as this strategy creates robust data to describe spike-like activity in neurons . However , those strategies might ignore activity below the spike threshold , or activity that is local and not synchronous with the firing pattern , or not triggered by voltage-dependent mechanisms . Another problem is that local calcium transients are commonly not shaped as calcium spikes ( fast onset followed by a slower decay of the signal ) and are filtered out by algorithms that expect a calcium spike or are filtered out because of computational strategies that define noise levels or noise signals [23 , 27] . Many biologically relevant calcium signals are very small and close to the measurement noise . Here , we tackle the problem of computing “signal-close-to-noise” activity , which is a calcium signal under pharmacological blockade of neuronal activity , in the absence of any exogenous stimulation or treatment . Here we define an activity event ( signal ) as a calcium event carrying information , while a computed noise event does not carry biological information . Therefore , we assume that homeostatic calcium fluxes of resting neurons [30] should create calcium events . In certain research questions , the analysis of local calcium transients offers important functional information . Local calcium activity regulates functions such as neurotransmitter specification , neurite extension , growth cone dynamics , activity-dependent axon growth , network wiring , as well as synaptic scaling [10 , 12 , 31–36] . While we know much about functions of local calcium signaling , the molecules triggering local or spontaneous activity are not well understood . Here , we introduce an intuitive computational approach to assess the activity state of neurons and to visualize signal-close-to-noise activity . The motivation was to create this tool for an unbiased computing of local activity in neurites , to monitor local excitability events , and to visualize homeostatic calcium activity . The bioinformatical tool is a refined Continuous Wavelet Transform ( CWT ) -based algorithm which identifies peak candidates in calcium activity signals occurring in the x , y-t imaging data set . As local calcium transients are hardly predictable or regularly shaped , we create a grid-based activity pattern over the x , y-field of the raw image data . CWT-guided peak detection was used to compute calcium signals , because we found that it targets critical problems in signal-close-to-noise computing: ( 1 . ) Baseline removal and signal smoothing is not needed and even small signals are accurately computed . ( 2 . ) The algorithm can be directly applied to raw data . ( 3 . ) Low amplitude peaks can still be real , and peak detection through local signal-to-noise ratio computation above a heuristic threshold may become an insufficient criterion to detect small signals , or to discriminate real signals from high amplitudes of noise signals . Therefore , instead of directly detecting peaks in the calcium signal , the algorithm identifies ridges in a CWT coefficient matrix and utilizes these coefficients to create an effective SNR for peak identification . These coefficients effectively guide the search for peaks that correspond to real calcium activity on the neurons . This improves the robustness of peak detection and keeps the false positive rate low [20] . Here we confirm that our strategy enables the identification of calcium activity events independent of segmentation parameters and shape parameters . The tool is easy to handle , powerful over the complete intensity value range , and ignores baseline shifts . More importantly , the computation identifies local activity events , even under pharmacological blockade of neuronal activity , and gives new information on calcium activity close to the signal noise . The tool is an open source tool .
Imaging can be performed with any imaging system that creates intensity values . Here , we used a CCD camera ( 8-bit ) at 2 . 5 Hz ( motoneurons ) , or 10–20 Hz ( hippocampal neurons ) ( Fig 1a ) . Next , the image series is uploaded to the activity detector tool . After opening the image series , the program asks for three tuning parameters: the grid window size ( WS ) , the signal-to-noise ratio ( SNR ) and the signal average threshold ( SAT ) ( Fig 1b and 1c; see S1 Manual ) . The window size ( WS ) will create a grid of ROIs with a defined size ( e . g . 8x8 pixel , Fig 1d ) . The grid is used to localize activity events . The signal-to-noise ratio ( SNR ) is the most important value and defines the stringency of the calcium event detector . SNR values of 1 . 5 to 4 ( or even 1 . 3 ) provide robust results on typical datasets acquired with the help of fluorescent calcium indicators . In our experience , computing of high signal-to-noise datasets , such as those obtained with modern genetically encoded calcium indicators [23 , 37] , can be performed with much higher SNR values . The signal average threshold ( SAT ) can be used to threshold the signals or to remove background signals . The threshold is a mean fluorescence value in a selected ROI , which can be taken either from the cell-free background or from any low-activity area in the x , y-image series . The SAT can be determined by a simple rule; SAT is the rounded threshold value . Note that the SAT can be set close to the black level . The reason is that the tool works with a CWT-based detection algorithm [20] and therefore background removal with the help of a threshold value has only a minor influence on the final result . Calcium activity event identification is then performed by two operations: ( 1 ) Process image stack to read out the signal intensity values in each grid window; ( 2 ) Detect activity to perform a wavelet transformation and to detect activity peaks . A pdf-document is automatically generated and put into the result folder of the activity tool ( S1 Fig; computing WS8 , SNR2 , SAT7 , MAC1 ) . This document shows on page 1 an image of the time series , the grid , and loci of calcium activity indicated by red circles within the grid window . The diameter of the circles is bigger when the tool finds more activity events . All activity events are summed up to give a computed value , the ‘total activity’ value ( Fig 1d ) . This value represents all calcium events in the whole x , y-t images series . The value represents the activity state under a specific experimental condition . Furthermore , the resulting pdf document shows all traces in which a calcium activity event was found , thus enabling fast access to the raw data and the interpretation of activity events by the tool . Activity events are marked and counted ( Fig 1c and 1d ) . Finally , a txt-file is generated that shows the calculated numbers of activity per grid window in a x , y-table structure ( S2 Fig ) . We also offer the possibility to increase the stringency of the tool . The user can select to count only those signal traces with more than one activity event ( Minimum activity counts; MAC ) . Furthermore , the tool is able to analyze signal fluctuations ( Include variance ) . With this function , the tool offers a value called variance area . The idea is that signal fluctuations are bigger , when , for instance , homeostatic calcium fluxes are pronounced . This is very useful to experimentally test whether fast fluctuations in a calcium signal trace are activity-dependent or not . The tool also offers the possibility to estimate a point of change in the general behavior of the signal in order to compute states of long-lasting activity , for instance after stimulation of neurons with an agonist . This feature of the tool is based on a confidence test that recursively searches for the point in the signal where the biggest change in variance and average occurs . We also implemented a ROI tool , which allows the re-analysis of loci of interest . ROIs can be selected with the help of a ROI manager , an ImageJ function , and can be analyzed as described above . ROI information can also be imported to the ROI manager . Calculation of 1300 images needed just about 60 seconds with the help of a standard desktop computer ( Windows X64 Intel core i-5 machine with 4 Gigabyte of RAM memory ) . This short computing time allows fast testing of user-dependent tuning parameters , for instance to compute the same signals with different SNR values . The tool processes the calcium imaging videos in ImageJ [38] for signal extraction . The data are automatically transferred to ‘R’ ( https://www . r-project . org/ ) for the statistical computing ( Fig 1e; details are explained in the methods section ) . For some applications , pre-processing of the videos might be necessary . For this purpose , it is of advantage that the presented tool is embedded in ImageJ , which gives the user access to a wide range of image processing tools and interactive access to the data . In Table 1 the application features of our tool are compared to alternative approaches , delineating the areas where we find that our tool is used advantageously .
Firstly , we asked whether the tuning parameters ( SAT and SNR; Fig 1b ) are robust enough to assess the activity state of a motoneuron . For convenience , this is again explained based on the single motoneuron ( Fig 2 ) . The movie was split into a low activity state ( frame 1–1300 ) and a high activity state ( frame 1301–2600 ) . We compared the low activity state with the high activity state of this globally firing motoneuron over a wide range of SNR values between 1 . 5 and 4 ( Fig 2b ) . Increasing the SNR increased the stringency of the tool , meaning that the tool preferentially identified stronger changes in fluorescence , or spike-like activity events ( Fig 2b ) . The SAT can be determined according to a rule , but little changes in the SAT only have a minor influence on the total number of activity events ( Fig 2b ) . As the calculation time is relatively short , the user can easily find a SNR , which fits best to the individual experimental condition or experimental setup . In this example , the high activity state of the motoneuron is defined by the total activity value of 613 ( WS: 8; SNR: 2; SAT: 6; MAC: 1; Fig 2c ) . The low activity state is represented by a total activity value of 102 ( S3 Fig ) . Although the underlying intensity signals within the grid windows are based on an almost 10-fold difference in the intensity values between the soma and the growth cone , the tool is able to compute relevant changes in signal intensity in the grid windows that superimpose the motoneuron ( Fig 2d ) . To test the performance and reliability of our computation , we performed simultaneous whole-cell patch clamp recording and calcium imaging . First , we offered the calcium indicator dye with the intracellular patch clamp solution ( Fig 3a ) and stimulated the cells by twelve current injections for 10 , 100 , 200 , or 500 ms to induce action potentials ( AP ) and AP-induced calcium transients . We extracted the calcium signals and determined whether the AP-induced calcium transients can be detected by our computation . Our computational strategy detected the AP-induced calcium event ( true positive response , TPR ) with high precision ( Fig 3b and 3c ) . On average , at an imaging speed of one image per 50 ms , the time between the AP and the time point of computed AP-induced calcium peaks was 320 ms ± 65 ms ( n = 192 events from 4 cells , mean value ± variance ) . False-positive events ( FPR ) were a minor problem under these circumstances . Other event detection methods also provided reliable results ( Fig 3b ) . While a deconvolution method marked several false-positive events , template-matching strategy and a strategy to extract significant signals above a computed baseline [27] labelled the AP-induced calcium signal with high precision [21] . Next , we asked whether our computation could also detect spontaneous activity in long-term cultured hippocampal neurons ( DIV 24 ) ( Fig 3d–3f ) . APs generated by spontaneous activity were inducing typical calcium spikes , and these calcium events were detected by our computational approach ( Fig 3f ) . S4 Fig shows how our tool and other computations identify or count activity events in a calibration dataset showing simultaneous imaging with loose-seal cell-attached recording in four GCaMP expressing neurons [37] ( data taken from: https://crcns . org/data-sets/methods/cai-1/about-cai-1 ) . To see how the tool behaves under high signal-to-noise ratio conditions , we re-computed a recent dataset of typical calcium spikes generated by synchronously active neurons ( Fig 4a ) [40] . In this example , calcium signals of glia-derived neurons were investigated by confocal calcium imaging [40] . Under continuous perfusion , neurons were treated with 10 μM bicuculline to block gamma-aminobutyric acid ( GABA ) A receptors and to induce calcium spikes ( Fig 4a ) [40 , 41] . By using high stringency settings , the tool automatically extracts the spike signal ( Fig 4a , average of all computed loci ) and marks the grid windows in which the neurons show spike behavior ( Fig 4a , computed loci ) . Note that under these imaging conditions , calcium spikes are preferentially observed in the neuronal somata . Next , we computed spontaneously spiking primary hippocampal neurons ( Fig 4b ) . Here , the bioinformatics tool is powerful enough to automatically mark typical calcium spikes and to count 790 activity events ( Fig 4b , middle ) . Acute blockade of the calcium spikes with the inhibitors tetrodotoxin ( TTX ) to inhibit TTX-sensitive voltage-gated sodium channels , and CNQX ( 6-cyano-7-nitroquinoxaline-2 , 3-dione ) to inhibit ionotropic glutamate receptors , blocked the generation of calcium spikes . A more than 10-fold lower total activity value and a much lower number of active loci illustrate the change in activity ( Fig 4b , right ) . The decline in the cytosolic calcium signal after activity blockade may be due to the low activity of voltage-gated calcium channels and the inability of neurons to maintain high calcium in the ER and cytosol [4 , 30 , 42] . The experiment shows that the activity detector is powerful enough to identify global spike-activity over a broad range of intensity signals . Next we asked how the tool behaves under more complex experimental conditions such as chemical LTP induction . Next , we performed calcium imaging with hippocampal neurons at a speed of 20 Hz and acquired thousands of images ( up to 10 , 000 ) per experiment under continuous perfusion . To induce increased activity , neurons were stimulated with an extracellular solution for chemical LTP induction ( Fig 5 ) . Two situations are compared; the control image sequence shows the time window before stimulation ( Fig 5b and 5c , left panel ) , while the cLTP stimulation phase is an equally long image series to describe the acute stimulation ( Fig 5b and 5c , right panel ) . Under control conditions , regions of high activity were preferentially identified in the neuronal periphery over neurites ( Fig 5b ) . Many of the computed signals were very small and showed no typical shape of a classic calcium spike ( Fig 5c; S5 Fig; WS8 , SAT11 , SNR 2 . 5; MAC2 ) . Local activity with typical synchronous calcium spikes showed up at multiple positions of the x , y-field ( Fig 5d , yellow and blue marks ) . A look at the underlying raw image data showed that this spiking activity is caused by multiple varicosities of an axon-like structure . During chemical LTP induction , neurons showed increased spike-like activity ( S6 Fig; WS8 , SAT11 , SNR 2 . 5; MAC2 ) , and even developed global synchronous activity over cell somata and neurites ( Fig 5b ) . The increase in activity is described by the total activity values ( 2135 activity events versus 4785 activity events ) . The tool marked many activity events with a noise-like character ( Fig 5c , example traces on the left side ) . During stimulation with the cLTP-inducing solution , the same neuronal structures increase their cytosolic calcium levels , some oscillated in their calcium signal , and also developed typical calcium spikes ( Fig 5c , example traces on the right side ) , and the tool calculated an increased number of activity events . The stringency of the tool depends on the wavelet algorithm and the chosen SNR . In Fig 5d , five signal traces are marked and show synchronous activity in the periphery . Purple arrows point to calcium spikes which were not counted by the activity detector , while the red arrow points to a local event that is only present in this trace ( grid window 20/32 ) . To show how the tool operates real noise in relation to noise-like activity , we compared the camera signal with real imaging data ( Fig 6 ) . For simplicity , this is explained based on the data in Fig 5 . We selected 12 grids and the underlying signal traces showing synchronous activity ( as in Fig 5d ) . In these traces , 23 calcium spikes can be identified in all of these grid windows . Grid windows of four representative loci are marked in Fig 6a , corresponding signal traces in Fig 6c . We put the SAT close to the black level to include all grid windows in the analysis . Next we determined the spike detection precision under different SNR values ( SNR 1 . 5 –SNR 3 . 0 ) ( Fig 6e ) . The tool computed all 23 events with a SNR of 1 . 5 , but finds fewer events with an SNR of 3 . 0 , which reflects the small change in signal intensity in these regions . For instance , in Fig 6c , grid window 22/23 shows a high sensitivity to high SNR values ( ‘worst case’ ) . Here , the mean intensity signal lies between 12–14 . 5 arbitrary units ( de facto reflecting the mean of the raw bit values ) , and therefore high SNR values ignore some calcium spikes . However , in the best case , in grid window 1/13 , the spike detection precision is constant over a wide range of SNR values , because the signal is very clear . Here , the mean raw values of the signal shifts from about 10–16 arbitrary units during a calcium spike . An auto-correction removes border effects caused by the calculation itself . The missing calcium events in Fig 5d ( purple arrows ) were excluded because of a too stringent definition of the SNR . Furthermore , it shows that it can be useful to analyze signal traces of the same video under different SNR conditions in order to describe different experimental conditions , as , for instance , before or after activity blockade . Next , we asked whether the tool computes too much noise . We imaged a homogenous light source , a fluorescent paper , with identical camera settings ( Fig 6b ) and computed the noise signal with different SNR values . The SAT was again set at 2 , meaning close to the black level , while the mean intensity value in the image field was 13 . All activity events were counted ( MAC of 1 ) . Representative signal traces are shown in Fig 6d . With a SNR of 1 . 5 , 5524 events were computed in the noise video , while 11145 events were found in with neurons ( Fig 6f ) . However , with an SNR of 2 . 5 , the noise signals ( 53 ) were in a good balance compared to the signals found in neurons ( 4092 ) . One has to consider that the program was computing 1452 grid windows of 1030 images . For this specific setting and imaging situation , this indicates that about 1 . 3% of the computed neuronal activity events reflected camera noise . Furthermore , this indicates that even with an SNR of 1 . 5 , more than 5500 activity events in the whole x , y-t imaging raw data might be regarded as neural activity . The camera noise alone cannot be responsible for the high number of signal-close-to-noise activity events detected by the algorithm . However , this experiment does not prove whether very small signals identified by the algorithm reflect a biological phenomenon or not . To target this question , we performed the following experiments . First , we imaged hippocampal neurons in presence of specific inhibitors of neuronal activity ( Fig 7 ) . Furthermore , we investigated activity events and signal fluctuations of homeostatic calcium fluxes during acute withdrawal and re-addition of extracellular calcium under activity blockade [30] . To prove whether the activity detector marks too many false-positive events , we tested the tool on imaging data of spontaneously firing neurons , and blocked activity and excitatory neurotransmission with TTX , a powerful inhibitor of action potential firing , CNQX , and APV ( D-2-amino-5-phosphonovalerate ) , to inhibit ionotropic glutamate receptors . Before activity blockade , the activity detection tool could well monitor the calcium-spike activity of a neuron ( Fig 7 ) , but also found many small and non-spike like activity events ( Fig 7c , grid 28/9 ) . Here , for simplicity , we use the term calcium spike to describe the typical waveform of AP-induced calcium spikes , with a fast onset and a slower decay of the signal . Acute perfusion with the activity inhibitor cocktail stopped the appearance of the classical calcium spike ( Fig 7c ) , hence , less activity events were computed . However , many small activity signals were found in the neuronal periphery . This non-spiking activity was typically local , meaning not a global calcium signal , or signal component of the complete neuron . Many of these signals were not shaped like typical calcium spikes with a fast onset and a slow decay , and were not uniformly shaped ( Fig 7d , purple rectangle to indicate a local activity hot spot ) . Next , we asked whether activity events close to the baseline are random fluctuations in the signal , or whether these small activity events reflect calcium fluxes or activity close to the baseline . Theoretically , when a neuron shows fluctuations in the calcium signal and many small calcium transients , but no typical calcium spikes , then the variance of the signal should be bigger than under conditions when activity is blocked . We again imaged hippocampal neurons ( DIV10; 10 . 000 images , 20Hz ) , and compared neuronal activity under control conditions and after acute activity blockade with TTX and CNQX ( Fig 8 ) . Typical calcium spikes were detected ( Fig 8a and 8b; left trace ) and the pharmacological treatment was sufficient to block the typical calcium spikes with a fast onset followed by a decaying transient ( Fig 8b , right trace ) . Then we calculated the variability of each signal using a sliding window of 30 frames and determined the variance and the mean of the signal ( computed with the tool: WS8 , SAT7 , SNR 3; Include variance: 30 ) . Finally , we determined the area of the variance around the mean in the signal trace ( yellow band in Fig 8c–8e ) and determined the number of activity events in the same traces with the activity detection tool . We call this value VA30 , for variance area in a sliding window of 30 frames ( S7 Fig , shortened example , 12 of 289 pages ) . In grid windows with a low number of activity events ( none to two ) , the variance area value VA30 was almost identical between control conditions ( Fig 8c , left trace , VA30 = 294 , 2 events ) and after TTX/CNQX block of neuronal activity ( Fig 8c , right trace; VA30 = 297 , 1 event ) . However , in another grid window ( Fig 8d ) , the variance area shifted during activity blockade from VA30 = 532 to VA30 = 370 , and activity events were reduced from 11 to 1 , showing that many of the small activity events in the control condition ( left trace ) are sensitive against inhibitors of neural activity , and might be real calcium signal events . In Fig 8e , a high variance area value ( VA30 = 1208 ) correlated with a high number of activity events ( 20 activity events ) . Activity blockade with TTX and CNQX reduced the variance area to VA30 = 559 , indicating reduced calcium fluxes after activity blockade . However , despite the strong reduction in signal fluctuation , the activity detector identified a local ( not global ) TTX/CNQX-independent , long-lasting transient with 9 activity events ( Fig 8e , lower trace , frames 1100–1700 ) . This shows that the computational assay is able to describe small loci of activity with two different values , the activity event number and the variance area . We controlled that our protocols , meaning imaging over several minutes and blocking of typical calcium spike behavior is reversible and that calcium spike activity recovers upon wash out of the corresponding inhibitors ( exemplarily shown in S8 Fig ) . Zooming in on corresponding loci ( Fig 9 ) shows that growth cone-like structures ( Fig 9c , trace 1–3 ) , neuritic elements ( Fig 9c , trace 4 ) , or varicosity-like structures ( Fig 9c , trace 5 ) form these local activity hotspots . Computed local activity events exhibit either a calcium spike-like shape ( Fig 9c , trace 1 , 2 ) , or reflect a phase of increased fluctuations ( e . g . Fig 9c , trace 4: images 1000–2000 ) , or a sudden jump in the activity state ( Fig 9c , trace 5: image 2300–3000 ) . The local activity events with the shape of a ‘broad’ or ‘elongated’ calcium spike are reminiscent of typical cell-autonomous local calcium spikes in motoneurons , which are mediated by local activity of voltage-gated calcium channels . In motoneurons , these local calcium events are triggered by local activity of TTX-insensitive voltage-gated sodium channels and are mediated by the local activation of high-voltage activated calcium channels [10 , 14 , 32] . The data confirm that many ‘small’ activity events in an active locus of a neuron are not solely a misinterpretation of the measurement noise . We term these events in calcium imaging data as ‘signals-close-to-noise’ activity . While some activity events under blockade of neuronal activity with TTX , CNQX , and APV can easily be regarded as real activity events ( e . g . in Fig 8d or Fig 9e ) , many computed signals cannot easily been distinguished from noise . We asked whether statistical computing of activity events could be used to describe the biological phenomenon of homeostatic calcium fluxes in resting neurons [30] . Under conditions of activity blockade ( TTX , CNQX , APV , 50 μM NiCl2 ) , resting neurons show pronounced homeostatic calcium fluxes between the endoplasmic reticulum and the extracellular space , as recently shown by direct ER calcium imaging [30] . To compensate unavoidable loss of ER calcium over the plasma membrane , resting neurons balance ER calcium levels through a continuous calcium influx mechanism with properties of store-operated calcium entry ( SOCE ) [30 , 42–44] . To compute this calcium activity , we imaged calcium fluxes in resting hippocampal neurons and acquired 6600 images at a frequency of 10 Hz ( Fig 10a ) . After one minute under steady-state conditions , neuronal activity was blocked with a high amount of TTX ( 500 nM ) , CNQX and APV ( each 20 μM ) . Then extracellular calcium was withdrawn for more than three minutes , before extracellular calcium was re-added to stimulate neuronal SOCE . Computational analysis was then performed under medial stringency conditions ( SAT 2 , SNR 2 . 5 , MAC2 ) . As shown in the signal traces in Fig 10b and 10c , removal of extracellular calcium led to a brief decline in the cytosolic calcium signal ( from bit value 7 to bit value 3 ) , which was accompanied by a decline in computed activity events ( Fig 10d and 10e ) . This was observed for individual grid windows ( for instance Fig 10b , yellow ROI in a ) , or the whole image plane ( Fig 10c , magenta ROI ) . Re-addition of extracellular calcium also restored the intensity of cytosolic calcium signals and more activity events were computed ( Fig 10f ) . We asked then whether the number of activity events or the change in the variance area is best suited to describe the changes in signal-close-to noise activity . We summed up the activity events ( Fig 10g ) and the computed VA30 area ( Fig 10h ) of all computed grid windows . The computed activity event value was well suited to distinguish signal-close-to-noise activity in the presence or absence of extracellular calcium . The tool identifies the calcium activity when the power of the signal is high enough in comparison to the intensity values in the neighborhood . Signal traces showing a decrease in cytosolic calcium concentration , as it happens to neurons under low calcium conditions ( see Fig 10b , frames 2000–4000 ) , will affect the CWT computation of the trace . Nevertheless , this has a minor influence in the activity event detection , because the algorithm constructs the ridges according to local maxima and estimates the power of an activity peak according to its close vicinity . The variance area value could also statistically distinguish between the two activity states , albeit with much less discriminatory power . The example shown here is representative of 15 imaging series performed with five independent neuron preparations . These results suggest that signal-close-to-noise activity can be regarded as a biological phenomenon , here triggered by homeostatic calcium fluxes [30] , and can be visualized in an unbiased way with the bioinformatics tool described here and both computed values , the total activity value and the variance area . Our tool identified a region of high activity in the neurites of the neuron . We used this ROI information ( yellow ROI in Fig 10a ) to show how other calcium event identification strategies analyze the signal ( Fig 10i ) . A deconvolution strategy offered many events under calcium withdrawal conditions , suggesting that these signals may be false positive . Notably , template-matching [21] ( green labels in Fig 10i ) and our CWT/pruning strategy found signal events before calcium withdrawal and after re-addition of calcium , indicating that both computational approaches can be used to define phases of homeostatic calcium activity in neurites . Signal extraction with the help of the ‘Romano toolbox’ [27] failed to extract and define a significant signal above a computed background noise signal in phases of higher homeostatic activity in the presence of calcium and is therefore not marked in Fig 10i . Finally , we analyzed this calcium movie with a tool for component segmentation and signal extraction [23] . This tool extracted five signal components from the movie ( S9 Fig ) . The typical signal pattern of calcium withdrawal and calcium re-addition was mirrored in four of five components . Notably , the computation indicates that the phase of calcium withdrawal is a phase of low activity ( indicated by a straight line in the signal components 1 , 3 , 4; S9 Fig ) . This suggests that the computation underlying this signal demixing strategy is useful to extract and interpret homeostatic activity from neurites . However , tens of loci of local homeostatic calcium activity were overseen and not identified as a signal component ( compare Fig 10d–10f with S9 Fig ) . Furthermore , the calcium withdrawal effect appears in the background signal and is erroneously considered to be a noise signal ( S9 Fig ) . Here , we assume that an unbiased grid strategy might help to offer an initial segmentation of the image data , before signals are extracted and demixed from the corresponding structures in the grid . In summary , CWT-based computation and signal pruning is suited to identify calcium activity components in loci of homeostatic calcium activity and the data can be used to count activity events for comparison of activity states . After minimal parameter tuning , two other strategies , template matching according to Patel et al . [21] and signal component analysis according to Pnevmatikakis et al . [23] would also be suited to compute signal-close-to-noise calcium signals . Here , some computational adaptations and tuning parameters might be computed and tested in the future . We found that counting of activity events in grid windows is useful to describe , in consideration of the pros and cons , local homeostatic calcium activity .
Specific parameter settings were important to achieve optimal performance; examples are given in the results ( Figs 1 , 2 and 5 ) . Here we discuss this more in general . To find a balance between two extremes , namely overestimation of activity through the computation of signals within noise , and disregarding of real activity close to the noise , we included simple tuning parameters . These tuning parameters , the signal-average threshold , the signal-to-noise ratio , the minimal activity count number , and the window size to define an x , y-grid of ROIs , enable robust , transparent , and quantitative results on individual , comparative data sets . The most important tuning parameter is the signal-to-noise-ratio ( SNR ) . The SNR defines the stringency of the detection tool . Analysis of our imaging data on cultured neurons with synthetic fluorescent calcium indicators ( OGB1 ) shows that an SNR of 1 . 5 to 2 . 5 is useful in finding activity events close to the noise level . SNR values of 3 to 4 focus on calcium transients with higher amplitude , and tend to discard ‘smaller’ transient-like signals . The quite short computing time in combination with the flexibility of the tool makes it easy to adapt the parameters for personal use . When we analyzed calcium imaging data from neurites of hippocampal neurons , it was obvious that some traces exhibited more and higher signal variations than other neuronal loci . For simplicity , we term this common phenomenon as signal fluctuation . To describe signal fluctuations , we calculated the variance in a sliding window of 30 frames , and found that fluctuations in the calcium signal became smaller when we blocked calcium spikes . The tool can be tuned to use longer or shorter sliding windows . In loci with intensive signal fluctuations , the wavelet analysis found a certain number of activity events . We therefore suggest using two signal trace features to compare neuronal activity mediated by signals close to the noise level: ( 1 . ) the number of total activity events in a grid window , and ( 2 . ) the variance area in a sliding window of a calcium trace . As shown in Fig 10 , even activity caused by homeostatic calcium fluxes can be described by the computed activity number and variance area in a signal trace . Both values are ideally taken from a dataset before and after pharmacological treatment , or from a wildtype neuron in comparison to genetically modified neurons , or can be determined in comparison to a control region in the same image series . Powerful strategies to extract cellular calcium signals are based on multiplication of two matrices to encode the spatial and temporal signal information [18 , 23 , 47] . A critical problem with this is to identify the spatial footprint of activity loci , meaning how to select the region of interest ( ROI ) or image segment for the subsequent calcium signal analysis . Independent component analysis ( ICA ) for matrix factorization [18 , 21] , non-negative matrix factorization ( NMF ) [47] , and constrained non-negative matrix factorization ( CNMF ) are powerful methods for extracting cells’ location and automated signal extraction . To target the problem of overlapping loci of activity , CNMF was shown to be eminently powerful [23] . While these strategies are all excellent in defining somatic calcium events and global activity of neurons , the tools may disregard small and local calcium events , which happen outside of the computed image component or ROI or have different time signatures . Furthermore , to better define calcium spikes , the true calcium signal traces are commonly ‘smoothed’ and therefore activity events close to the measurement noise may be ignored , or defined as noise [18 , 21 , 23] . We found that CNMF can be powerful to extract homeostatic signals from neuronal segments ( S9 Fig ) . However , new tuning criteria may be needed for improved separation of local activity from background signals . Maybe a grid strategy would be useful to support this analysis , thus simply separating a neuron in unbiased subcompartments . Furthermore , an event definition and a counter would be helpful to offer a natural number representing calcium activity . In our approach here , we intended to go one step back and to pattern the x , y-field with a pixel-wise grid , to avoid trace smoothing , and to paint an activity pattern over the x , y-grid field , thus making the activity detection independent of a user- or computation-based ROI definition . The user gets an idea of how activity is distributed over the image field and gets new information about local calcium activity events . Theoretically , it is a disadvantage that the locus of analysis is not drawn by the underlying neuronal morphology and therefore only partially fits the underlying structure . It is also a fact that the SNR of the extracted signal in the grid is not ideal , since the temporal traces are extracted not only from pixels belonging to one compartment , but also from pixels from another structure in the same grid window . The grid idea also causes that the same activity event , e . g . a spike , is counted in many grid windows if the activity spreads over a region bigger than the grid size . However , for this tool and our research question it is a powerful and robust strategy to identify activity loci and to count a computed activity event . This is because spontaneous activity is astonishingly diverse , occurs at very small loci , and shows signals which do not appear in the close neighborhood . These events reflect microdomain activity . The grid concept is a simple way to compute total activity patterns of all loci . The output image creates an activity image , which rebuilds the gross structure of the underlying activity centers ( e . g . Figs 3 , 4 and 9 ) . The implemented ROI tool and the window size parameters allow pixel-wise re-analysis of image segments . The simplistic concept makes the tool user-friendly and robust . Calcium transient identification can be computed with different algorithms and wavelet-based detection methods are useful for peak detection , even though variants of the algorithm produce a high false positive rate and introduce small phase-shifts [21] , while other algorithms are well suited to reduce the false-positive rate [20 , 48] . We show by analyzing real data that the computing is quite precise when the tool is well tuned . Knowledge-based activity event detection ( template-matching ) uses a database of calcium signal waveforms or synthetic datasets and proves whether there is a similarity between a waveform and the more noisy true calcium signal trace [21 , 25] . We avoided knowledge-based strategies on the basis of pre-defined calcium signal waveforms , as we cannot easily create knowledge-based masks for the non-spiking activity observed in calcium imaging data of spontaneously active neurons or in the homeostatic situation . At the same time our data suggest that template-matching [21] is suited to detect signal-close-to-noise activity by homeostatic calcium fluxes . Future tools might include both strategies and the CWT computing might support the creation of templates for template-matching . One cannot easily decide whether computed calcium signals close to the noise are ‘real’ signals , or wrongly identified signals . Our tool accepts this problem and decides that it is better to label event candidates , than to ignore them . As already mentioned above , the tool tries to find calcium activity events . It is important to notice that the wavelet transform provides a set of point candidates that could be possibly labeled as activity peaks . However , this labeling depends on the pruning done to the tree formed with the set of candidates . This pruning process is done following the branches on the activity tree and such branches are formed according to the ridges of the wavelet transform . The pruning consists in discarding such points which are not present in more than one scale ( meaning that they are just local maxima ) and those points that do not have the minimally required SNR level . At the end of the pruning , all the remaining points are tagged as activity peaks . For the CWT computation , we used a symmetric kernel . For this reason , the asymmetry of spike-induced calcium transients is a disadvantageous condition for our algorithm , not because of the wavelet transform algorithm itself , but because of the selection of the symmetric second derivative Gaussian as the mother wavelet . However , the second derivative Gaussian has the important property that it has an almost zero phase shift . This allows us to mark the point in time of the detected peak . Our results in Figs 3–5 confirm that the symmetric mother wavelet is suited to compute spike-induced calcium signals with high precision . The tool is an open source tool , based on ImageJ and ‘R’ , and does not need any commercial , license-protected computing environment . The tool is easily operated . A user manual is available ( S1 Manual ) . Based on our imaging data , we defined a minimum distance between peaks , which is a distance of five data points . When neurons show high activity rates , one can either increase the speed of image acquisition , or one can test beneficial properties of a specific calcium indicator to improve the temporal resolution or the signal-to-noise ratio [6 , 49] . In the manual we show how the tool can be tuned to better label higher event frequencies ( S1 Manual , Chapter VIII ) . To compare different experimental conditions , we recommend starting with a bigger grid window , the SAT according to the provided thresholding rule , or close to the black level ( see above ) and an SNR of 2 . 5 as the default setting . The grid window and the SNR affect the computing time and for a first assessment of raw data these default settings worked very well in our hands with synthetic high affinity calcium indicators . For instance , using a standard desktop computer , 3 , 000 images ( 348 x 260 pixel ) are computed for one minute to process the image stack , and for another three minutes to complete the wavelet transform and the generation of the documentation pdf . One thousand images are computed in less than two minutes . Initially , we normally check the data with an SNR of 2 , 2 . 5 , and 3 , and a grid window size that is small enough to cover the neurites . A second computation with another SNR value , e . g . to separate high SNR signals and low SNR signals , needs just the resetting of the computed values , the change in the SNR value and the restart of the “Detect activity” computing , to get the next signal documentation . This needs some seconds . After a first assessment , it is easy to tune the parameters for the final analysis or a specific image acquisition setup . Small grid windows will offer an intuitive image of the neuronal loci where activity events took place , but at the cost of a longer computing time . This tool shifts the focus from spike analysis of cell bodies to local signal-close-to-noise analysis in any area of a neuron . Small and local calcium signals mediate important biological functions [9 , 12 , 32 , 50 , 51] and may be caused by a multiplicity of signaling mechanisms [1 , 2 , 4 , 52] . The molecules underlying local calcium signals are not well described , but are relevant target factors for potential protective and functionally restorative treatments in psychiatric and neurological disorders . One example is the identification of target factors to treat motoneuron diseases [53–55] . Motoneurons show cell-autonomous spontaneous calcium transients , which appear in an unpredictable spatiotemporal on-off , and high versus low frequency pattern [10 , 14] . By using calcium imaging and subsequent assessment of spontaneous calcium events , disease-relevant molecular factors have been identified . For instance , the local excitability pattern of motoneurons is triggered by a subthreshold active ion channel , the sodium channel NaV1 . 9 [32 , 56] , and mediated by the N-type calcium channel CaV2 . 2 [14] . This excitability signal cascade is disturbed in motoneurons from mouse models for spinal muscular atrophy [10 , 14 , 32 , 54] , the most common genetic cause for infant mortality [57] . Screening-like approaches based on calcium imaging and automated excitability analysis with bioinformatics may offer new information on the role of these genetic factors in motoneurons and patient-derived induced neurons . Furthermore , not much is known about the role of the excitability factors in neurons of the brain or within neural networks , which are triggering neural network oscillations [58 , 59] . Locally acting signaling factors such as neuropeptides , neurotrophins , or the contribution of subthreshold voltage changes on excitability are not well integrated in a functional concept of synaptic communication [60] . Furthermore , local tuning and scaling of excitability by calcium-dependent pathways plays an important role in synaptic development [9] , and one has to consider that the neuronal excitability is also affected by homeostatic calcium fluxes at rest , which are maintained by a store-operated calcium entry mechanism [30 , 43 , 44] . The software tool presented here operates on a wide range of neuronal activity detection tasks including very different signal intensities and assesses a broad visual field . It is able to offer an unbiased analysis of spontaneous and local neuronal activity . As it is not focused on calcium spikes or a computed ROI , it is able to offer initial insights into the total calcium activity pattern under different experimental conditions . The idea is to use an unbiased x , y-grid , to include all pixels in a imaging video and to offer an activity map . The tool avoids a preferential look at one activity mode of a neuron , e . g . the spiking behavior of neuronal somata , but instead tries to find any activity by detecting signals-close-to-noise .
The animal welfare committee of the University of Würzburg , in accordance with European Union guidelines , approved all experimental procedures . Hippocampal neurons were prepared from CD1 mice of either sex , as described earlier [39] . Hippocampi of newborn mice were collected in Hank’s buffered saline solution ( HBSS ) . Trypsin ( Worthington ) was added to a final concentration of 0 . 1% and the tissue incubated for 15 min at 37°C . The protease digestion was stopped with 0 . 1% Trypsin inhibitor ( Sigma ) . After four steps of trituration in Neurobasal/B27 medium ( Life Technologies ) , cells were plated on poly-L-lysine-coated glass coverslips in Neurobasal , 1× B27 , 0 . 5% penicillin/streptomycin , 1% Glutamax , and 1× N2 supplement ( all Life Technologies ) and cultured at 37°C under an atmosphere of 5% CO2 . Calcium imaging experiments were performed after indicated days in vitro ( DIV ) . Primary motoneurons were prepared from spinal cord [61] . The lumbar spinal cord of mouse embryos was dissected at embryonic day 13 or 14 . Motoneurons were enriched by affinity-panning with antibodies against the p75NTR receptor , and plated at a density of 1 , 000–2 , 000 cells on 10mm glass coverslips coated with polyornithine and laminin-1 . Motoneurons were grown in Neurobasal/B27 medium ( Life Technologies ) , 2% horse serum , 10 nM β-mercaptoethanol , and 1x GlutaMax . The neurotrophic factors BDNF and CNTF were added at a concentration of 5ng/ml . One day after motoneuron isolation , 40% of the medium was replaced . Calcium imaging was performed at DIV 3 . Calcium indicator dye loading and Ca2+ imaging was performed in artificial cerebral spinal fluid ( ACSF ) . For motoneuron imaging ACSF contained ( in mM ) : 127 NaCl , 3 KCl , 2 . 5 NaH2PO4 , 2 CaCl2 , 1 MgCl2 , 23 NaHCO3 and 25 D-glucose , bubbled with 95% O2/5% CO2 . Hippocampal neurons were imaged under continuous perfusion with ( in mM ) : 135 NaCl , 6 KCl , 1 MgCl2 , 2 CaCl2 , 5 . 5 D-glucose , 10 HEPES ) . For chemical LTP stimulation the following buffer composition was used: 128 NaCl , 13 KCl , 3 CaCl2 , 5 . 5 D-glucose , 10 HEPES , 0 . 1 glycine ( in mM ) . Calcium-free ACSF contained 0 . 1 mM EGTA . The calcium indicator Oregon Green 488 BAPTA-1 , AM ( OGB; Invitrogen ) was prepared as 5mM stock solution in 20% Pluronic F-127 / DMSO ( Invitrogen ) . For dye loading , 0 . 5μl of the OGB/Pluronic mixture was mixed in 500μl of ACSF and neurons were labeled for 15 minutes at 37°C and 5% CO2 . Changes in OGB/calcium-fluorescence were monitored with the help of an upright microscope ( BXWI , Olympus , objective: Olympus 40x LUMPlanFI/IR , 0 . 8 W ) , in a heated imaging chamber ( Luigs & Neumann ) . Imaging was performed under continuous perfusion with ACSF . Images ( 8-bit ) were captured at the indicated speed in a streaming approach , with a Rolera-XR camera ( Qimaging ) and StreamPix 4 software ( Norpix ) under continuous illumination with a 470nm LED light source ( Visitron Systems ) . Fluorescence filters with the following parameters were used: excitation: 482 ± 35 nm; dichroic filter 506nm , emission filter 536 ± 40nm . The health status of the neurons is routinely tested by patch-clamp techniques , control stimuli , and washout controls . Current clamp recordings were performed in whole-cell configuration . Pipettes with 2 . 5–5 MΩ resistances were pulled from borosilicate glass ( GB 150-8P , Science Products ) with a P-97 micropipette puller ( Sutter Instruments ) . Data were acquired using a HEKA EPC-10 USB patch-clamp amplifier controlled by the PatchMaster software ( HEKA Electronic ) at 25°C . Raw data were continuously sampled at a frequency of 5 kHz and filtered at 2 . 9 kHz . Electrophysiological experiments with hippocampal neurons were performed with calcium imaging buffer as an external solution . For a steady flow of the external solution , a Minipuls 3 Peristaltic Pump ( Gilson ) was used . The internal solution contained 148 mM potassium gluconate , 10 mM HEPES , 10 mM NaCl , 0 . 5 mM MgCl2 , 4 mM Mg-ATP , 0 . 4 mM Na3-GTP ( pH 7 . 3 with KOH ) and Oregon Green™ 488 BAPTA-1 hexa potassium salt ( 2 μM ) . In whole-cell configuration , 200 pA current injections were given for either 500 ms , 200 ms , 100 ms or 10 ms for 12 times at intervals of 10 s . Calcium imaging and patch clamp recording was synchronized via TTL ( transistor-transistor-logic ) signals and controlled with the help of a TTL-induced 10 ms light pulse ( Thorlabs , T-cube LED driver LEDD1B ) . The light signal became visible in a single frame of the image series . Spontaneous activity was measured in the whole-cell current clamp configuration . Parallel calcium imaging was performed with a sample rate of 10 Hz . In some experiments , the calcium indicator loading was done with estered OGB , as described above . To compute calcium transients from raw image material the method was split into two stages . Signal extraction was computed on ImageJ [38] and activity events were calculated on ‘R’ ( https://www . r-project . org ) . Both computations were embedded in the Bio7 environment , an open platform ( http://bio7 . org/ ) . We call the presented application NA3 ( spoken: NA cubic ) . It is a powerful software tool , combining different routines and languages for optimal neuronal activity detection . The open source tool consists of around 800 lines of code and the core functionality is about 140 lines long . The application was developed on a Windows X64 Intel core i-7 machine with 16 Gigabyte of RAM memory . For time series analysis , a Windows X64 Intel core i-5 machine with 4 Gigabyte of RAM memory was used . The tool is available on: https://github . com/jpits30/NeuronActivityTool . The tool containing the ROI feature is available at: https://github . com/jpits30/NeuronActivityTool_ROI . The intensity signals are analyzed in order to detect the peaks that correspond to neuronal activity . The event detection uses a modified wavelet transform algorithm for peak detection [20 , 48] . For a more detailed explanation refer to this reference [62] . One advantage of the wavelet transform over other spectral analysis techniques , like the Fourier analysis , is its multiscale feature . The mother wavelet is scaled . Therefore , it can fit peaks of different sizes . Our algorithm sums the information obtained from matching the calcium activity peak with several scaled versions of the mother wavelet and decides based on that whether that part of the trace contains a peak of activity or not . The wavelet transform constructs a signal representation that changes with time and space . In order to construct such representation the signal is convolved with a pattern signal called mother wavelet . The mother wavelet is normalized such that ∥Ψ∥ = 1 . Using the mother wavelet Ψ , the wavelet transformation of a one dimensional function x ( t ) is defined in the equation ( Eq 1 ) . X ( a , b ) =1a∫−∞∞Ψ* ( t−ba ) x ( t ) dt ( 1 ) a denotes the scale parameter of the wavelet , b is the time shift parameter . The wavelet transformation can be understood as a matching procedure between the original signal and the mother wavelet , so with the objective of peak detection , the mother wavelet must be a signal with a clear peak and approximate symmetry . This computation uses a second derivative Gaussian as mother wavelet . This mother wavelet is appropriate for peak detection , as it enhances the peak and diminishes the neighboring signal values . The wavelet transform was chosen because the shifting and scaling of the basic function results in the time resolution property for peak detection within a noisy intensity signal , being able to recognize not only spiking neurons but also smaller peaks of calcium activity . The wavelet transformation maps the peak search into a search of ridges on the time-scale space , this space is smoother and easier to characterize . The wavelet transformation also enables that removal of the base line is not required because it is done intrinsically . We suppose , with respect to earlier studies [20] , that the calcium trace in the vicinity of a peak is composed of three components , the peak ( p ( t ) ) , the base line ( b ( t ) ) and the noise ( N ) as shown in equation ( Eq 2 ) xcal ( t ) =p ( t ) +b ( t ) +N ( 2 ) Then the wavelet transform of such signals would be ( Eq 3 ) Xcal ( a , b ) =1a ( ∫−∞∞Ψ* ( t−ba ) p ( t ) dt+∫−∞∞Ψ* ( t−ba ) b ( t ) dt+∫−∞∞Ψ* ( t−ba ) Ndt ) ( 3 ) The second and third integrals will be approximately zero because of the symmetry of the mother wavelet and because of the low level of matching between the base line and a mother wavelet such as the second derivative Gaussian . That means that the only term left is the convolution between the peak and the mother wavelet . As mentioned before , once the original calcium trace is transformed , the peak search problem turns into a problem of finding the ridges of the wavelet transformation and determining which of those ridges corresponds to peaks of interest for us . A ridge is conform with a subset of local maxima , which are nearby in time and belong to consecutive scales [63] . Fig 2a shows how the computation extracts a set of signal peaks to define calcium transient candidates . The wavelet transform has a set of ridges , which are used to construct a tree . Using that tree , the computation detects the peaks in the signal . In the coefficient space of a wavelet transform ( Fig 2a ) , a group of local maxima can be identified . In Fig 2a , the ridges are the brighter yellow areas ( almost white in some points ) under the blue vertical lines ( branches ) . The branches of the tree are assembled as follows . Starting at the highest scale ( top of Fig 2a ) , a local maximum within the coefficients is defined . This coefficient is set to be the starting point of a branch of a tree . This branch is elongated by adding it to the rest of the coefficients in the same ridge . This procedure is repeated , starting from the second largest scale and is then repeated until all the local maximum values in the wavelet transform were considered . Some scales need to be evaluated several times , because they contain several local maximum values . This procedure is done by using the wavCWTTree function from the wmtsa package in R [64] . At the end of this procedure , we have a set of branches , which are further pruned according to the length of the branch . The pruning of the branches is efficient and is performed as follows . Each branch ( the blue vertical lines ) extends over some scale values and has a particular length . Some of the branches are discarded if they do not meet certain criteria for their length . If a branch extends over less than one octave of the scale range , then it is discarded . The rest of the branches ( those long enough ) form the tree that we use to search for the peaks . In our experiments , we used one octave as the threshold . In the user manual , Chapter VIII ( S1 Manual ) , we show how this pruning can be modified by the user . Next , the tool searches each branch of the tree to find out whether it contains a peak . This search occurs according to two conditions . The first is the signal-to-noise ratio of the peak candidate and the second is whether the peak candidate lies on a scale higher than a certain value . The first condition , the SNR , is the most important and it is used in our algorithm as a tuning parameter . The signal-to-noise ratio is estimated with the help of the coefficients of the wavelet transform . The wavelet coefficients of the lowest scale near the branch are used to estimate the noise power . Similarly , the highest coefficient in the branch represents the power of the peak . Hence , the signal-to-noise ratio of the peak is simply the ratio of the power of the signal to the power of the noise . This signal-to-noise ratio has to be higher than the user-defined SNR; otherwise , the peak is discarded . This can be observed in Fig 2a , where some of the branches end up with an activity event ( peak ) , while other branches are not considered as an activity event . The tool has a border effect of three to five images , where it cannot find activity . To describe signal fluctuations in a signal trace , the mean of the signal with the corresponding variance is calculated in a sliding window . The number of window frames can be selected by the user . The variance window is visualized as a yellow band ( Fig 7 ) , whereas the yellow area is the variance area , describing the mean +/- of the variance , centered on the mean . This variance area is a virtual value to describe changes in signal fluctuations . The variance area value is used to compare two experimental conditions , for instance , the variance of the mean of a calcium imaging signal in a grid window before and after pharmacological treatments . The tool also offers the possibility to estimate a change in the general tendency of the signal . Such estimation is done with the help of the changepoint package in R . The estimated point of change in the general tendency is the point that splits the signal in the two parts with the most significant statistical difference in average and variance . It means that the signal before the point and after the point are the two most different parts of the signal which can be found , no other cut on the signal would produce a more different pair of signals . This tendency change estimation proves useful , for example when the researcher wants to see the time of action that an inhibitor needs to affect the behavior of cultures neurons .
We will test the option for automated constrained non-negative matrix factorization ( CNMF ) to see whether such algorithms will help to better define the signal source . Furthermore , we will test optional asymmetric kernels for CWT computation and , depending on the data , which combination of symmetric and asymmetric kernels is optimal . Numerous other options are available to further develop our tool and new applications will be investigated . Users and developers are invited to develop and test new options for NA3 and use our test data and code . | Calcium imaging has become a standard tool to investigate local , spontaneous , or cell-autonomous calcium signals in neurons . Some of these calcium signals are fast and ‘small’ , thus making it difficult to identify real signaling events due to an unavoidable signal noise . Therefore , it is difficult to assess the spatiotemporal activity footprint of individual neurons or a neuronal network . We developed this open source tool to automatically extract , count , and localize calcium signals from the whole x , y-t image series . As demonstrated here , the tool is useful for an unbiased comparison of activity states of neurons , helps to assess local calcium transients , and even visualizes local homeostatic calcium activity . The tool is powerful enough to visualize signal-close-to-noise calcium activity . | [
"Abstract",
"Introduction",
"Design",
"and",
"implementation",
"Results",
"Discussion",
"Materials",
"and",
"methods",
"Availability",
"and",
"future",
"directions"
] | [
"wavelet",
"transforms",
"action",
"potentials",
"medicine",
"and",
"health",
"sciences",
"engineering",
"and",
"technology",
"signal",
"processing",
"membrane",
"potential",
"electrophysiology",
"neuroscience",
"neurites",
"calcium",
"signaling",
"neuronal",
"dendrites",
"neuroimaging",
"research",
"and",
"analysis",
"methods",
"imaging",
"techniques",
"mathematical",
"functions",
"animal",
"cells",
"mathematical",
"and",
"statistical",
"techniques",
"metabolism",
"calcium",
"imaging",
"biochemistry",
"signal",
"transduction",
"cellular",
"neuroscience",
"cell",
"biology",
"physiology",
"neurons",
"signal",
"to",
"noise",
"ratio",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"cell",
"signaling",
"neurophysiology",
"bone",
"and",
"mineral",
"metabolism"
] | 2018 | An open source tool for automatic spatiotemporal assessment of calcium transients and local ‘signal-close-to-noise’ activity in calcium imaging data |
Dendritic cells ( DCs ) are the most potent antigen-presenting cells that link innate and adaptive immune responses , playing a pivotal role in triggering antigen-specific immunity . Antigen uptake by DCs induces maturational changes that include increased surface expression of major histocompatibility complex ( MHC ) and costimulatory molecules . In addition , DCs actively migrate to regional lymph nodes and activate antigen-specific naive T cells after capturing antigens . We characterize the functional changes of DCs infected with Orientia tsutsugamushi , the causative agent of scrub typhus , since there is limited knowledge of the role played by DCs in O . tsutsugamushi infection . O . tsutsugamushi efficiently infected bone marrow-derived DCs and induced surface expression of MHC II and costimulatory molecules . In addition , O . tsutsugamushi induced autophagy activation , but actively escaped from this innate defense system . Infected DCs also secreted cytokines and chemokines such as IL-6 , IL-12 , MCP5 , MIP-1α , and RANTES . Furthermore , in vitro migration of DCs in the presence of a CCL19 gradient within a 3D collagen matrix was drastically impaired when infected with O . tsutsugamushi . The infected cells migrated much less efficiently into lymphatic vessels of ear dermis ex vivo when compared to LPS-stimulated DCs . In vivo migration of O . tsutsugamushi-infected DCs to regional lymph nodes was significantly impaired and similar to that of immature DCs . Finally , we found that MAP kinases involved in chemotactic signaling were differentially activated in O . tsutsugamushi-infected DCs . These results suggest that O . tsutsugamushi can target DCs to exploit these sentinel cells as replication reservoirs and delay or impair the functional maturation of DCs during the bacterial infection in mammals .
Dendritic cells ( DCs ) are the most potent antigen-presenting cells ( APCs ) that initiate and orchestrate immune responses [1] . Upon pathogen infection , DCs capture foreign antigen and undergo maturational changes including increased surface expression of major histocompatibility complex ( MHC ) and costimulatory molecules , such as CD40 , CD80 , and CD86 . Moreover , they migrate from peripheral tissues via afferent lymphatic vessels into draining lymph nodes where they prime antigen-specific naive T cells [2] . Migration of DCs to regional lymph nodes is mainly regulated by changes in surface expression of chemokine receptors . Increased surface expression of CCR7 during DC maturation enables DCs to respond to the lymphoid chemokines , CCL19 and CCL21 , which are constitutively produced by lymphatic endothelial cells and secondary lymphoid organs . Therefore , surface expression of CCR7 , in addition to the expression of MHC and costimulatory molecules , is critical for initiating antigen-specific T cell responses in regional lymph nodes . Infectious microbial pathogens have established numerous strategies that disrupt and confound DC functions to survive and evade host immune antimicrobial mechanisms [3] . For example , secondary lymphoid organs of human immunodeficiency virus ( HIV ) -infected individuals have been shown to contain an accumulation of semi-mature dendritic cells that exhibit a lower expression of costimulatory molecules that support differentiation of CD4+ T cells into regulatory T cells and suppress effector functions [4] . Herpes simplex virus type 1 infection rapidly degrades cytohesin-interacting protein in DCs and impairs DC migration through increased integrin-mediated adhesion [5] . DCs infected with human respiratory syncytial virus do not efficiently increase CCR7 expression and hence displayed inefficient chemotatic migration toward a CCL19 gradient [6] . Filamentous hamagglutinin of Bordetella pertusis inhibits IL-12 and stimulates IL-10 production by DCs , which directs naive T cells to differentiate into regulatory subtypes [7] . These diverse hijacking strategies employed by microbial pathogens to utilize DCs for their own benefit may have been acquired during their eternal struggle for evolutionary survival . Orientia tsutsugamushi , the causative agent of scrub typhus , is an obligate intracellular bacterium [8] . The bacteria are transmitted from chigger mites to humans , after which O . tsutsugamushi invades cells in the dermis , causing an inflammatory lesion called an eschar [9] . A recent study using eschar skin biopsies from scrub typhus patients showed that O . tsutsugamushi has tropism for DCs and monocytes rather than endothelial cells , traditionally regarded to be the primary target of the bacterial pathogen [9] . Immunohistological analysis of eschar lesions revealed that DCs and macrophages predominantly infiltrate at the dermo-epidermal junction while the bacterial pathogen is mainly within Langerhan's cells , dermal DCs , and activated macrophages [9] . These results suggest that infection of dendritic cells and macrophages may be a potential route for dissemination of O . tsutsugamushi from the initial infection site and that cellular tropism may influence its interaction with host immune responses . Currently , there is limited knowledge of the role played by DCs in O . tsutsugamushi infection . Therefore , we investigated DC responses to O . tsutsugamushi infection for the first time . Although O . tsutsugamushi is capable of inducing humoral and cellular immune responses in vivo , it is not clear whether O . tsutsugamushi-infected DCs are successfully activated and migrate to secondary lymphoid organs to initiate anti-bacterial immune responses . This study was designed to evaluate the functional interaction of O . tsutsugamushi with DCs to better our understanding of the immunological pathogenesis of O . tsutsugamushi during the early phase of infection .
Animal experiments were approved by the Seoul National University Institutional Animal Care and Use Committee ( SNU IACUC , Permit No . SNU-100414-1 ) and performed in strict accordance with the recommendations in the National Guide Line for the care and use of laboratory animals . C57BL/10NAGCSnAi- ( KO ) Rag2 ( H-2b ) mice ( Taconic Farms , Germantown , NY ) and C57BL/6 mice ( Orient Bio , Seongnam , South Korea ) were housed and maintained in the specific pathogen-free facility at Seoul National University ( SNU ) College of Medicine . The Boryong strain [8] , [10] , [11] of O . tsutsugamushi was purified using a modified Percoll gradient purification method [12] . O . tsutsugamushi was propagated in L929 cells . At 3 to 4 days postinfection , infectivity was determined using an indirect immunofluorescence assay . When an infection rate of >90% was achieved , the cells were harvested by centrifugation at 6 , 000× g for 20 min . The cell pellet was resuspended with 6 . 5 ml of Tris-sucrose ( TS ) buffer ( 33 mM Tris-Cl [pH 7 . 4] , 0 . 25 M sucrose ) and the cells were homogenized using 100 strokes of a Polytron homogenizer ( Wheaton Inc . , Millville , NJ ) followed by centrifugation at 200× g for 5 min . The supernatant was then mixed with 40% Percoll ( Pharmacia Fine Chemicals , Uppsala , Sweden ) in TS buffer and centrifuged at 25 , 000× g for 60 min . The bacterial band was collected and centrifuged at 77 , 000× g for 30 min . The bacterial pellet was washed 3 times in TS buffer , resuspended in culture media , and stored in liquid nitrogen until use . The infectivity titer of the inoculum was determined as previously described [13] , [14] . For infection assays , 2 . 5×106 infected-cell counting unit ( ICU ) [14] of O . tsutsugamushi was to infect cell cultures in 24 well plate ( ∼10 bacteria/cell ) . The same amount of O . tsutsugamushi were also inactivated by exposing the bacteria to a 30-W UV lamp for 30 min in 6-well plates [15] or by heating the bacterial stocks at 100°C for 10 min [13] , and used as control infection . Unstimulated immature DCs were also used as negative control in each experiment . DCs were generated from the bone marrow of 6- to 12-week-old Rag2 knock-out mice . The bone marrow cells were flushed out of the femurs and tibias with serum-free Iscove's modified Eagle medium ( IMDM; Gibco Invitrogen , Grand Island , NY ) . The single cell suspension was then filtered through a nylon cell strainer ( 70-µm Nylon mesh; BD Biosciences ) , washed twice with complete IMDM [supplemented with 10% FBS , recombinant mouse GM-CSF ( 1 . 5 ng/ml; PeproTech , Rocky Hill , NJ ) and mouse IL-4 ( 1 . 5 ng/ml; PeproTech ) , penicillin ( 100 units/ml ) , streptomycin ( 100 µg/ml ) , gentamicin ( 50 µg/ml ) , L-glutamine ( 2 mM ) , and β-mercaptoethanol ( 50 nM; Gibco Invitrogen ) ] , and seeded at a concentration of 1×106 cells per well in a 24-well plate . Half of the medium was replaced every other day with an equal volume of complete IMDM medium for 6 days [16] . The immature DCs generated were stimulated with O . tsutsugamushi or 0 . 5 µg/ml Lipopolysaccharide ( LPS; Sigma Aldrich , St . Louis , MO ) for 20 h . In every infection study , we confirmed that more than 90% of the cells were infected with O . tsutsugamushi after 20 h of incubation . Immunofluorescence microscopy was used to visualize O . tsutsugamushi [12] . Briefly , infected cells were fixed in PBS containing 4% paraformaldehyde for 15 min at room temperature and permeabilized in 0 . 2% Triton X-100 for 15 min . Cells infected with O . tsutsugamushi were incubated with pooled sera from scrub typhus patients for 1 h , followed by incubation with AlexaFluor488-conjugated goat anti-human IgG ( Molecular Probes ) . For autophagy detection , cells were co-stained with anti-LC3 antibody ( NB100-2220; Novus Biologicals , Littleton , CO ) and AlexaFluor594-conjugated secondary antibody ( Molecular Probes ) . Immunostained cells were examined under an Olympus FV1000 laser scanning confocal microscope ( Olympus; Tokyo , Japan ) . All images were analyzed and processed using the Olympus Fluoview software ( Olympus ) . In order to investigate the responses of DCs after bacterial infection , immature DCs were incubated with live or inactivated O . tsutsugamushi ( ∼10 bacteria/cell ) , or E . coli LPS ( 0 . 5 µg/ml ) as a positive control . In some experiments , cells were infected with different doses ( 10 bacteria/cell or 20 bacteria/cell ) or stimulated with inactivated bacteria as mentioned above . At 20 h after stimulation , DCs were stained with antibodies against the indicated surface molecules or isotype control antibodies after blocking Fc receptors with anti-CD16/32 ( 2 . 4G2; BD Pharmingen ) . Allophycocyanin ( APC ) Cy7-conjugated anti-CD11c ( N418; Biolegend , San Diego , CA ) , isothiocyanate ( FITC ) -conjugated anti-I-A[b] ( AF6-120 . 1; BD Pharmingen , Franklin Lakes , NJ ) , phycoerythrin ( PE ) -conjugated anti-CD40 ( 3/23; BD Pharmingen ) , APC-conjugated anti-CD80 ( 1610A1; eBioscience , San Diego , CA ) , PE Cy7-conjugated anti-CD86 ( GL-1; eBioscience ) , PE-conjugated anti-CCR7 ( 4B12; eBioscience ) , and 7-AAD ( BD Pharmingen ) were used for flow cytometric analysis . Fluorescence intensities of the stained molecules were examined after gating 7-AAD-negative and CD11c-positive live DCs on a FACSCanto II flow cytometer ( BD Biosciences ) . Data were analyzed using Flowjo software ( Tree Star , Ashland , OR ) . DCs were infected with O . tsutsugamushi for 20 h and the culture supernatant was used for cytokine antibody array . The culture supernatant from unstimulated immature DCs was used as negative control . We used RayBio Mouse Cytokine Antibody Array III & 3 . 1 ( RayBiotech , Inc . , Norcross , GA ) , which can simultaneously detects 62 proteins ( Axl , BLC , CD30 L , CD30 , CD40 , CRG-2 , CTACK , CXCL16 , Eotaxin , Eotaxin-2 , Fas Ligand , Fractalkine , GCSF , GM-CSF , IFN-γ , IGFBP-3 , IGFBP-5 , IGFBP-6 , IL-1β , IL-10 , IL-12 p40/p70 , IL-12 p70 , IL-13 , IL-17 , IL-1α , IL-2 , IL-3 , IL-3 Rβ , IL-4 , IL-5 , IL-6 , IL-9 , KC/CXCL1 , Leptin/OB , Leptin R , LIX , L-Selectin , Lymphotactin , MCP1 , MCP-5 , M-CSF , MIG , MIP-1α , MIP-1γ , MIP-2 , MIP-3β , MIP-3α , PF-4 , P-Selectin , RANTES , SCF , SDF-1α , sTNF RI , sTNF RII , TARC , TCA-3 , TECK , TIMP-1 , TNF-α , Thrombopoietin , VCAM-1 , and VEGF-A ) , as recommended by the manufacturer's instructions . Briefly , each membrane was blocked in 2 ml of Blocking Buffer and incubated at room temperature ( RT ) for 30 min . Membranes were then incubated with 1 ml of culture supernatants at RT for 2 h , washed three times with 2 ml of Wash Buffer I at RT , and further washed twice with 2 ml of 1× Wash Buffer II at RT . 1∶250-diluted biotin-conjugated primary antibodies were added . The membranes were incubated at 4°C overnight , washed as described above , incubated with 1∶1000-diluted HRP-conjugated streptavidin at RT for 1 h , and washed with wash buffer three times . After incubation with Detection solution , the membrane images were analyzed by LAS-3000 ( Fujifilm , Tokyo , Japan ) . Signal intensities were analyzed by Quantity One software ( Bio-Rad , Hercules , CA ) . DCs unstimulated or stimulated with indicated agents were mixed with PureCol ( Advanced biomatrix , Poway , CA ) in 10× PBS , resulting in gels with a collagen concentration of 1 . 5 mg/ml . Final concentration of the cells in the assay was 1×107 cells/ml . The collagen-cell mixture was cast in 15 μ-slide VI flat ( ibidi , München , Germany ) and incubated at 37°C for 40 min . After assembly of the collagen fibers , recombinant chemokine CCL19 ( 1 . 2 µg/ml ) ( R&D Systems , Inc . , Minneapolis , MN ) diluted in IMDM containing 10% FBS and PureCol mixture were cast on the opposite side of the slide . For cell tracking , cells were visualized by time lapse imaging using a confocal microscope ( Olympus ) . Manual single cell tracking of samples was performed using Manual Tracking Plugin of Image J ( National Institute of Mental Health , Bethesda , MD ) . Cells were tracked every 2 min per frame for 4 h . Velocity , Euclidean distance and directionality parameters were calculated and visualized as plots [17] . Ears were obtained from sacrificed C57BL/6 mice . The ears were mechanically split into dorsal and ventral halves and mounted on 24 well plates with the dermal surface exposed [18] . Unstimulated immature DCs or DCs ( 106 cell/ml ) stimulated with O . tsutsugamushi or LPS for 20 h were labeled with 5- ( and 6 ) -carboxyfluorescein diacetate succinimidyl ester ( CFSE; Invitrogen ) . Cells were washed twice with PBS , incubated with CFSE ( 5 µM ) for 10 min at 37°C , and washed twice with PBS containing 0 . 5% of BSA . DCs were resuspended in culture medium , added on top of the dermis , and incubated at 37°C for 2 h . After gently washing away non-infiltrated DCs , the ears were fixed with 4% paraformaldehyde ( Sigma ) for 30 min . The ears were then incubated with a rat monoclonal anti-LYVE-1 antibody ( R&D Systems ) at 4°C overnight , washed with PBS and stained with Alexa fluor 647-conjugated anti-rat secondary antibody ( Molecular Probes ) . C57BL/6 mice were preinjected with LPS ( 0 . 3 µg/leg ) in the hind-leg footpad one day before injection of DCs [19] . DCs ( 106 cell/ml ) were stimulated with O . tsutsugamushi or LPS ( 0 . 5 µg/ml ) for 20 h , labeled with CFSE , and resuspended in PBS to a concentration of 108 cells/ml . Unstimulated immature DCs were also included as negative control . 30 µl of the cell solution was injected into the footpad of C57/BL6 mice . 48 h later , popliteal lymph nodes were collected and treated with 1 mg/ml of collagenase D ( Sigma ) at 37°C for 40 min to collect the cells from the lymph nodes . The percentage of migrated DCs in total lymph node cells was determined by FACS analysis [20] . DCs stimulated with CCL19 ( 200 ng/mL ) for the indicated time periods were washed twice with ice-cold PBS . Cells were lysed with lysis buffer ( 50 mM Tris-HCl [pH 7 . 5] , 150 mM NaCl , 1% Triton X-100 , 1% sodium deoxycholate , 0 . 1% sodium dodecyl sulphate ( SDS ) , and 2 mM EDTA ) . The cellular proteins separated by SDS–polyacrylamide gel electrophoresis were electrotransferred to PVDF membranes and subjected to immunoblot analysis using the indicated antibodies . anti-LC3 ( NB100-2220; Novus Biologicals , Littleton , CO ) , anti-p44/42 MAPK ( ERK1/2 ) ( 3A7; Cell Signaling Technology , Frankfurt , Germany ) , anti-phospho-ERK ( E-4; Santa Cruz ) , anti-p38α/β ( A-12; Santa Cruz ) , anti-phospho-p38 ( D-8; Santa Cruz ) , anti-GAPDH ( 6C5; Santa Cruz ) , Horseradish peroxidase–conjugated anti-mouse antibody ( Santa Cruz Biotechnology , Inc . , Santa Cruz , CA ) , were used for immunoblot analysis . The immune-reactive bands were detected using enhanced chemiluminescence reagents ( Ab frontier , Seoul , South Korea ) . Signal intensities were analyzed by Quantity One software ( Bio-Rad , Hercules , CA ) . Statistical analysis of all the experimental data was performed using the two-tailed Student's t-test with 95% confidence interval . Data are expressed as the mean ± standard deviation and a p value<0 . 05 was considered to be statistically significant . Statistical analyses were accomplished using GraphPad Prism 5 . 01 ( GraphPad Software Inc . , La Jolla , CA ) .
Since O . tsutsugamushi has been detected in DCs in eschars of human scrub typhus patients [9] , we first confirmed whether the bacteria are able to infect and replicate within DCs . Bone marrow-derived immature DCs were infected with O . tsutsugamushi and further incubated for one day . Efficient replication of O . tsutsugamushi in the perinuclear region [21] was observed using indirect immunofluorescence assay ( Figure 1A ) . In order to investigate the responses of DCs after bacterial infection , immature DCs were incubated with O . tsutsugamushi or E . coli LPS as a positive control . Surface expression of MHC II , CD40 , CD80 , and CD86 were significantly ( p<0 . 05 ) up-regulated at 20 h after infection with O . tsutsugamushi , even though the levels of expression were lower than those of cells stimulated with LPS ( Figure 1B and 1C ) . Mean fluorescence values of MHC II , CD40 , and CD80 in O . tsutsugamushi-infected DCs increased by more than 2 fold when compared to those of un-stimulated immature DCs , indicating that bacterial infection induced DC activation . Since the levels of surface expression of activation markers were lower than those of cells stimulated with LPS , we further examined whether this is dose-dependent by increasing the amount of bacterial inoculums . In addition , we also evaluated whether active bacterial replication was required for DC activation by exposing DCs to the same amount of heat or UV-inactivated bacteria . As shown in Figure 2 , surface expression of MHC II and CD86 did not significantly change with increased of live or inactivated bacteria , whereas CD80 expression in DCs incubated with inactivated bacteria was slightly reduced when compared to that of live O . tsutsugamushi . Possible explanations for the weaker induction of costimulatory molecules by O . tsutsugamushi infection compared to LPS stimulation could be a direct inhibition or insufficient stimulation by the intracellular pathogen . We therefore investigated whether the exposure of O . tsutsugamushi-infected DCs to a secondary stimulation with LPS would result in a more efficient induction of costimulatory molecules . DCs were infected with O . tsutsugamushi for 2 or 6 h , followed by a secondary stimulation with LPS . The surface expression levels of costimulatory molecules and MHC II were quantified at 20 h after infection ( Figure 1C , OT2/LPS and OT6/LPS ) . The secondary treatment with LPS induced a significant increase of costimulatory molecules on the surface of infected DCs , thus demonstrating that the relatively low levels of surface expression of costimulatory molecules on O . tsutsugamushi-infected DCs is due to an insufficient stimulation of DCs rather than a direct inhibition by the intracellular pathogen . O . tsutsugamushi also stimulates the production of proinflammatory cytokines and chemokines that are important in both innate and acquired immunity . In order to examine the inflammatory mediators released by O . tsutsugamushi-infected DCs , we compared the expression profiles of cytokines and chemokines of unstimulated and O . tsutsugamushi-infected DCs using a cytokine antibody array . We observed a more than 20% up-regulation of nine cytokines and chemokines in DCs infected with O . tsutsugamushi compared to unstimulated immature DCs ( Figure 3 ) . Among them , IL-6 , MIP-1α , and RANTES increased by more than 50% in O . tsutsugamushi-infected cells . These results further demonstrate that O . tsutsugamushi infection activates DCs . Autophagy is an innate defense mechanism against various intracellular pathogens . To survive within host cells , intracellular pathogens have evolved mechanisms to avoid elimination by autophagy [22] . To analyze whether infection with O . tsutsugamushi can stimulate autophagy in DCs , we infected cells with the bacteria and measured the conversion of LC3-I to LC3-II up to 4 h after infection ( Figure 4A and B ) . The conversion of LC3-1 to LC3-II gradually increased when DCs were infected with O . tsutsugamushi . Induction of autophagy was also monitored by confocal laser-scanning microscopy analysis in DCs . Infection of O . tsutsugamushi led to a drastic increase of intracellular autophagosomes ( Figure 4C ) . Interestingly , most of the intracellular bacteria did not co-localized with LC3-positive autophagosomes throughout the infection , indicating that few bacteria are actually captured . To further investigate whether O . tsutsugamushi actively evades autophagy , we treated DCs with UV-inactivated bacteria . Even though UV-treated bacteria induced autophagosome formation within DCs as efficiently as live bacteria , all the inactivated bacterial particles co-localized with LC3-positive autophagosomes in contrast to live bacteria . These results indicate that live O . tsutsugamushi actively evades the cellular autophagic system although the bacteria activate cellular autophagy upon infection regardless of its viability . The ability of activated DCs to migrate to secondary lymphoid organs where naive T cells reside is a crucial step in the generation of primary T cell responses . DC migration to regional lymph nodes is a complex process composed of multiple steps , including movement to the tissue interstitium , entry into lymphatic vessels , and extravasation from the lymphatic system into the lymph nodes [2] . In order to investigate the effect of O . tsutsugamushi infection on DC migration , we used 3D collagen gels to mimic the interstitial microenvironment and cells were exposed to a diffusion gradient of CCL19 . As shown in Figure 5 and Data S1 , S2 , S3 , S4 , and S5 , mature DCs stimulated with LPS migrated efficiently toward the chemokine source . However , the chemotactic response of DCs infected with O . tsutsugamushi was significantly impaired and similar to that of immature DCs ( p = 0 . 864 ) . Interestingly , directional migration was slightly increased in DCs stimulated with UV-inactivated bacteria ( p<0 . 001 vs . CNT , Figure 5A and B ) , suggesting that viability of the intracellular pathogen may affect its inhibition of DC migration . However , when we further stimulated O . tsutsugamushi-infected DCs with LPS , chemotatic migration recovered and was as efficient as LPS-stimulated DCs , suggesting that the impaired migration of infected DCs might not be due to an irreversible inhibition by the bacteria . In order to examine the entry of DCs into lymphatic vessels , we used crawl-in assays in which fluorescently labeled DCs were placed on the dermis of ear explants . After incubation for 2 h , the numbers of DCs localized within the LYVE-1+ lymphatic vessels were counted and compared ( Figure 6A and B ) . DCs stimulated with LPS were more efficiently localized within the lymphatic vessels than immature control DCs ( p<0 . 001 ) . In contrast , co-localization of O . tsutsugamushi-infected DCs was comparable to that of immature cells ( p = 0 . 627 ) . DCs stimulated with UV-inactivated bacteria showed a slight increase in the number of cells colocalized within lymphatic vessels , but this was statistically not significant ( p = 0 . 068 , Figure 6B ) . We further confirmed inefficient migration of DCs infected with O . tsutsugamushi in vivo ( Figure 6C ) . Fluorescently labeled DCs were injected in the footpads of mice and their popliteal lymph nodes were analyzed at 2 days after injection . Approximately three times more DCs were detected in the draining lymph nodes when stimulated with LPS compared to unstimulated immature DCs , whereas migration of O . tsutsugamushi-infected DCs was similar to unstimulated immature DCs ( p = 0 . 508 ) . Since the migration of DCs to regional lymph nodes requires the expression of CCR7 , the receptor for lymphoid chemokines CCL19 and CCL21 , we next analyzed the effect of O . tsutsugamushi infection on CCR7 surface expression in DCs . We found a portion ( 24 . 8±11 . 6% ) of immature DCs with surface CCR7 expression in our experimental setup and mature DCs stimulated with LPS showed a remarkable increase in CCR7 surface expression ( 60 . 0±5 . 1% ) ( Figure 7 ) . When DCs were infected with O . tsutsugamushi , CCR7-positive cells ( 52 . 2±2 . 8% ) were significantly increased compared to that of immature control cells . Secondary stimulation of O . tsutsugamushi-infected DCs with LPS further increased the population of CCR7-positive cells ( 63 . 6±1 . 8% ) . Since we detected two populations of DCs in terms of CCR7 expression level in O . tsutsugamushi-infected DCs , we examined whether this is dose-dependent . The upregulation of CCR7 expression was not significantly changed when the cells were infected with increasing amounts of bacteria nor UV-inactivated ones ( Figure 7B ) . Taken together , O . tsutsugamushi infection can induce surface expression of CCR7 in DCs and the impaired migration of DCs infected with O . tsutsugmushi might not be due to insufficient surface expression of CCR7 . It was previously reported that engagement of CCR7 by its ligands , such as CCL19 , activates MAP kinase members and this signaling pathway subsequently regulates chemotaxis of DCs [23] . Thus , we analyzed whether CCR7 induces activation of MAP kinases in O . tsutsugamushi-infected DCs upon exposure to CCL19 . DCs stimulated with O . tsutsugamushi or LPS were incubated with CCL19 ( 200 ng/ml ) for the indicated time periods . The cells were lysed , and analyzed by immunobloting using antibodies specific for the phosphorylated/active forms of MAP kinases , ERK and p38 ( Figure 8 ) . Treatment with CCL19 resulted in a rapid and potent activation of ERK in DCs . Phosphorylation of ERK reached a maximum after 5 to 10 min of incubation and returned to levels close to baseline by 60 min in immature control DCs and O . tsutsugamushi-infected DCs , but was sustained for 60 min in DCs stimulated with LPS . Interestingly , phosphorylation of p38 was barely detectable in control DCs and O . tsutsugamushi-infected DCs , in contrast to LPS-stimulated DCs which showed a transient activation of p38 after 5–10 min of incubation with CCL19 . These results suggest that a differential activation of MAP kinase members upon chemokine exposure may contribute to the inefficient chemotaxis observed in O . tsutsugamushi-infected DCs .
Recent studies affirm the immunostimulatory activities of DCs such as increased expression of MHC complexes , upregulation of costimulatory molecules , and secretion of cytokines needed for efficient T cell priming [24] . Similar to cells infected with other Rickettsia species , activation markers were significantly upregulated after DCs were stimulated with O . tsutsugamushi ( p<0 . 05 ) [25] . However , the level of DC activation , as measured by the surface expression of MHC II and costimulatory molecules , was significantly lower than those of cells stimulated with LPS ( p<0 . 05 ) . Previously , it was shown that O . tsutsugamushi lacks LPS and peptidoglycan in its cell wall [26] , [27] . In addition , the genes for the synthesis of such cell components are absent in the bacterial genome [8] , [27] . Thus , the absence of cell wall components , especially LPS , makes O . tsutsugamushi less stimulating to DCs and may explain the “semi-mature” phenotypes of DCs after bacterial infection . It could be postulated that the bacteria evolved to modify their envelope structures to be specific to the host to which they have adapted [8] , [27] , [28] . The secretion of chemokines and cytokines by activated DCs is critical in orchestrating T cell responses in regional lymph nodes . Previously , it was reported that O . tsutsugamushi induces various chemokines in macrophages and endothelial cells via the activation of NF-κB and AP-1 transcription factors [13] , [29] , [30] , [31] . In a macrophage cell line , mRNAs encoding MIP-1α/β , MIP-2 , and MCP-1 were increased in response to O . tsutsugamushi infection [13] . In addition , the expression of chemokines Ltn , RANTES , MIP-1α , MIP-1β , MIP-2 , and MCP-1 and cytokines LTβ , TNF-α , IL-6 , IFN-γ , TGF-β1 , and MIF was upregulated in mice infected with O . tsutsugamushi [32] . Here , we also observed increased secretion of IL-6 , IL-12 , MIP-1α , and RANTES in O . tsutsugamushi-infected DCs . These results suggest that even though O . tsutsugamushi lacks LPS , it is capable of activating DCs to a certain extent , inducing inflammatory responses potentially via the activation of NF-κB and AP-1 transcription factors . Given that the migration of DCs infected with O . tsutsugamushi was considerably impaired , preferential secretion of the CC chemokine subfamily , which includes RANTES and MIP-1α from O . tsutsugamushi-infected DCs at the infection site may contribute to the infiltration of monocytes and lymphocytes into the infection sites as observed in the eschars from scrub typhus patients [9] . Autophagy is recognized to be a bona fide immunological process with a wide array of roles in immunity against intracellular pathogens [33] . It is a specialized cytoplasmic system for the direct elimination of intracellular bacteria and a crucial contributor to intracellular antigen processing and MHC presentation of endogenously expressed antigens [34] , [35] , [36] , [37] . Recently , it was also shown that cellular autophagy is essential for innate cytokine production and APC functions in DCs infected with pathogenic viruses [38] , [39] , [40] . Therefore , autophagosomes induced by pathogenic infection play a pivotal role in both innate and adaptive host defenses against intracellular pathogens . Here , we show for the first time that live O . tsutsugamushi actively evades the cellular autophagy system , although it induces autophagosomes in DCs regardless of its viability ( Figure 4 ) . This active escape from induced autophagy may significantly affect the APC functions of DCs , which not only includes its ability to directly kill the intracellular pathogen via autophagosomal degradation , but also antigen processing and presentation , both essential for the subsequent induction of adaptive immunity in the infected host . Various intracellular bacteria have developed sophisticated mechanisms to escape from the autophagic machinery . Shigella IscB protein mediates active escape from autophagy by competitively inhibiting the interaction of VirG bacterial protein with host Atg5 , which is required for autophagy induction [41] . Listeria ActA and InlK proteins promote the recruitment of host cytosolic proteins to the bacterial surface in order to mask the pathogen from recognition by the autophagic system , thereby promoting intracellular survival [42] , [43] . Given that most live O . tsutsugamushi actively escape induced autophagosomes within the cytosol of DCs , the intracellular pathogen may also be equipped with an efficient mechanism of autophagy evasion that remains to be elucidated . Antigen-presenting DCs acquire foreign antigens in peripheral tissues . Efficient DC migration to draining lymph nodes through lymphatic vessels optimizes foreign antigen presentation to naive T cell [44] . DC migration into and along this conduit occurs through a series of steps including mobilization , detachment , interstitial migration , entry into the afferent lymphatics , and transit via lymph [24] . Within an artificial 3D matrix of collagen , DCs stimulated with LPS migrated along gradients of CCL19 and showed amoeboid morphology and velocities that were comparable to in vivo observations ( Figure 5 and Data S1 , S2 , S3 , S4 , and S5 ) [17] . However , the chemotactic response of DCs infected with O . tsutsugamushi was drastically impaired and similar to that of un-stimulated immature DCs . Moreover , the impaired migration of O . tsutsugamushi infected-DCs was consistent in both ex vivo and in vivo experimental settings ( Figure 6 ) . These results suggest that O . tsutsugamushi may actively inhibit the chemotactic migration of DCs or does not stimulate the cells as efficiently as LPS , a strong DC activator . To test this hypothesis , we stimulated O . tsutsugamushi-infected DCs with LPS . This secondary stimulation restored efficient chemotatic migration of DCs infected with O . tsutsugamushi ( Figure 5 ) as well as further upregulation of costimulatory molecules on DCs ( Figure 1 ) . The restoration of chemotatic migration by the secondary LPS stimulation was also observed even after extended incubation ( up to 8 h ) with O . tsutsugamushi ( Data S6 ) . Therefore , O . tsutsugamushi suboptimally stimulates DCs rather than irreversibly inhibits DC activation and migration . DC migration is primarily guided by the two chemokines , CCL19 and CCL21 , which are expressed in lymphatic endothelium and the T cell area of lymph nodes [17] . Antigen uptake by DC induces maturational changes that include decreased expression of the chemokine receptors CCR1 , CCR2 , and CCR5 that maintain DC residence in peripheral tissues , and increased expression of CCR7 that mediates the migration of antigen-bearing DC to lymphatic tissue via binding to the chemokines [6] . When we examined the surface expression of CCR7 on DCs infected with O . tsutsugamushi , the level of CCR7 expression was comparable to that of LPS-stimulated DCs ( Figure 7 ) , suggesting that suboptimal expression of CCR7 on DCs is not the cause of impaired chemotactic migration of O . tsutsugamushi-infected DCs . Various microbial pathogens have established diverse strategies to control DC migration . Human cytomegalovirus ( HCMV ) blocks the migration of infected monocyte-derived DCs toward lymphoid chemokines , CCL19 and CCL21 , by modulating the level of CCR7 expression [45] . HCMV very efficiently triggers the down regulation of CCR5 without inducing the expression of CCR7 in infected DCs , even following stimulation with LPS/TNF-α/IFN-γ , normally a potent stimulus for CCR7 induction [45] . In the case of human metapneumovirus ( HMPV ) and human respiratory syncytial virus ( HRSV ) , viral infection of monocyte-derived DCs did not efficiently decrease CCR1 , 2 , and 5 expression , and did not efficiently increase CCR7 expression [46] . The inefficient chemokine receptor modulation of DCs by viral infection results in poor migration toward the CCR7 ligand , CCL19 . HMPV- or HRSV-stimulated DCs responded to secondary stimulation with LPS or a cocktail of proinflammatory cytokines by increasing CCR7 and decreasing CCR1 , 2 and 5 expression , and by more efficient migration to CCL19 , suggesting that HMPV and HRSV suboptimally stimulate rather than irreversibly inhibit DC migration . When we analyzed the surface expression of chemokine receptors , CCR5 was efficiently downregulated ( data not shown ) and CCR7 was upregulated in O . tsutsugamushi-infected DCs , suggesting that decreased cheomotactic migration is not due to impaired chemokine receptor modulation in O . tsutsugamushi-infected DCs . Several studies have proposed that integrated signaling modules regulate chemotaxis in CCR7-stimulated DCs [23] , [47] . Various non-overlapping signaling modules , including Gαi-mediated activation of p38 and ERK1/2 , appear to regulate chemotactic migration of DCs induced by CCR7 chemokine ligands [47] . Recently , expression of CCR7 alone was shown to be insufficient for DC migration , as it can be expressed in a biologically insensitive state such that CCR7+ DCs either fail to undergo chemotaxis towards CCR7 ligands [48] or require a high concentration of CCR7 ligands before they respond [44] . The mechanisms by which these mediators alter CCR7 functionality is not known , but they probably trigger signaling events that , in turn , alter the signaling cascades that are engaged when CCR7 binds to its ligands [49] , [50] . In this study , phosphorylation of p38 was barely detectable in O . tsutsugamushi-infected DCs as well as in control DCs , in contrast to the LPS-stimulated DCs which showed a transient activation of p38 after 5–10 min of incubation with CCL19 . Given that phosphorylation of ERK occurs in both uninfected and infected DCs ( Figure 8 ) , the specific failure of p38 activation in O . tsutsugamushi-infected DCs may be the result of the “semi-mature” phenotypes and impaired migration of the infected DCs . Interestingly , inhibition of all these MAP kinases does not completely blunt CCR7-dependent chemotaxis [23] , [51] , suggesting that additional unidentified molecules or signaling pathways may also regulate chemotaxis of DCs [52] . In order to elucidate the precise molecular mechanisms that control DC migration during O . tsutsugamushi infection , further studies are needed . Recently , a paper reported that O . tsutsugamushi mainly infects “inflammatory” CD14/LSP-1/CD68 positive monocytes and CD1a/DCSIGN/S100/FXIIIa and CD163 positive dendritic cells in stained eschar skin biopsies from scrub typhus patients [9] . Not only that , innate APCs were highly accumulated in these pathologic lesions . The authors propose that infection of dendritic cells and activated inflammatory monocytes offers a potential route for dissemination of O . tsutsugamushi from the initial eschar site . The immunomodulatory effects of O . tsutsugamushi infection on local APCs in the eschar site could also interfere with downstream host immune responses . Our current results further suggest that O . tsutsugamushi infection might interfere with the active exit of DCs from the initial infection site and exploit these sentinel cells as a reservoir for bacterial replication . The functional exploitation of DCs by O . tsutsugamushi may contribute to bacterial pathogenesis during the early phase of infection and interfere with effective downstream adaptive immunity required for protection against bacterial infection . Further studies on the effect of O . tsutsugamushi infection on adaptive immune responses , especially on antigen-specific T cell immunity in conjunction with impaired DC functions by bacterial infection , may enhance our understanding of immunological pathogenesis in scrub typhus patients . | Scrub typhus is an acute febrile illness caused by Orientia tsutsugamushi infection and is one of the main causes of febrile illness in the Asia-Pacific region . If not properly treated with antibiotics , patients often develop severe vasculitis that affects multiple organs , and the mortality rate of untreated patients reaches up to 30% . To understand the pathogenic mechanisms of the infectious disease , we characterized the functional changes of O . tsutsugamushi–infected dendritic cells ( DCs ) , which play a pivotal role in orchestrating innate and adaptive immune responses . The obligate intracellular bacteria efficiently infected bone marrow-derived DCs and activated the cells as measured by induced surface expression of MHC II and costimulatory molecules , secretion of cytokines and chemokines , and autophagy induction . However , the live bacteria actively escaped from host autophagosomes and the migration of infected cells was severely impaired in vitro , ex vivo , and in vivo infection models . Finally , we found that MAP kinases involved in chemotactic signaling were differentially activated in O . tsutsugamushi-infected DCs . These results suggest that O . tsutsugamushi can target DCs to exploit these sentinel cells as replication reservoirs and delay or impair the functional maturation of DCs during the bacterial infection in mammals . | [
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] | [
"medicine",
"bacterial",
"diseases",
"infectious",
"diseases",
"immune",
"cells",
"antigen-presenting",
"cells",
"immunology",
"biology",
"scrub",
"typhus"
] | 2013 | Orientia tsutsugamushi Subverts Dendritic Cell Functions by Escaping from Autophagy and Impairing Their Migration |
Multicopy single-stranded DNAs ( msDNAs ) are hybrid RNA-DNA molecules encoded on retroelements called retrons and produced by the action of retron reverse transcriptases . Retrons are widespread in bacteria but the natural function of msDNA has remained elusive despite 30 years of study . The major roadblock to elucidation of the function of these unique molecules has been the lack of any identifiable phenotypes for mutants unable to make msDNA . We report that msDNA of the zoonotic pathogen Salmonella Typhimurium is necessary for colonization of the intestine . Similarly , we observed a defect in intestinal persistence in an enteropathogenic E . coli mutant lacking its retron reverse transcriptase . Under anaerobic conditions in the absence of msDNA , proteins of central anaerobic metabolism needed for Salmonella colonization of the intestine are dysregulated . We show that the msDNA-deficient mutant can utilize nitrate , but not other alternate electron acceptors in anaerobic conditions . Consistent with the availability of nitrate in the inflamed gut , a neutrophilic inflammatory response partially rescued the ability of a mutant lacking msDNA to colonize the intestine . These findings together indicate that the mechanistic basis of msDNA function during Salmonella colonization of the intestine is proper production of proteins needed for anaerobic metabolism . We further conclude that a natural function of msDNA is to regulate protein abundance , the first attributable function for any msDNA . Our data provide novel insight into the function of this mysterious molecule that likely represents a new class of regulatory molecules .
Retron reverse transcriptases ( RT ) in bacteria were first described in Myxococcus xanthus [1] and E . coli [2] in the 1980s and are now known to be widely distributed in the genomes of eubacteria and archaea ( reviewed in [3] ) . All retrons contain three regions essential for production of msDNA: msr ( RNA primer for reverse transcription ) , msd ( template sequence ) , and a reverse transcriptase ( RT ) . The retrons of pathogens , such as Salmonella Typhimurium ( STm ) , may also encode an additional ORF of unknown function [4] . The product of the ‘retron’ is a small covalently linked RNA-DNA hybrid molecule called multicopy single-stranded DNA ( msDNA ) that is predicted to form complex secondary structures [5] . The predicted secondary structures of msDNA from enteric pathogens including STm , enteropathogenic E . coli and Vibrio spp . are similar [4] but the reverse transcriptase amino acid sequence from these enteric pathogens share little identity . The location of the retron as well as the number of retrons in each species varies . These observations suggest that retrons have been horizontally acquired by convergent evolution to function in a fashion that is specific to the biology of the host bacterium . Although the molecular details of the production of msDNA have been heavily studied , no natural function has been attributed to this mysterious molecule despite 30 years of study ( reviewed in [3] ) . A critical obstacle to elucidating the natural function of msDNA was the lack of any phenotype for mutants unable to make this molecule . We have shown that the retron reverse transcriptase encoded by STM3846 is essential for Salmonella Typhimurium ( STm ) to colonize the calf intestine [6] , a natural model of enteric salmonellosis that recapitulates the earliest stages of human non-typhoidal Salmonella ( NTS ) infection . This was the first reported phenotype for a mutant lacking a retron reverse transcriptase . NTS are major threats to global animal and human health , causing more than 90 million cases of gastroenteritis in people worldwide [7] . Human enteric salmonellosis is characterized by inflammatory diarrhea containing primarily neutrophils . To efficiently colonize the host , NTS use the type 3-secretion system 1 ( T3SS-1 ) encoded on Salmonella Pathogenicity Island-1 ( SPI-1 ) to invade the intestinal epithelium [8 , 9] and to promote the characteristic neutrophilic inflammatory response . The host inflammatory response gives Salmonella a competitive advantage over resident microflora . Within the intestinal lumen , the product of the neutrophilic oxidative burst generates tetrathionate from oxidation of thiosulfate [10] . Salmonella uses tetrathionate as a terminal electron acceptor within the anaerobic conditions of the intestinal lumen to gain a competitive advantage over resident microflora . Effectors of the TTSS-1 may directly activate epithelial production of inducible nitric oxide synthase ( iNOS ) thereby creating nitrate , an additional terminal electron acceptor [11] . The relative importance of nitrate during infection is illustrated by the fact that it is a powerful chemoattractant for Salmonella during anaerobiosis [12] . In addition , Salmonella uses host-derived nutrients such as ethanolamine [13] during intestinal inflammation . These strategies facilitate the growth of Salmonella in the complex microbial community of the intestine . We used the enteric pathogen , Salmonella Typhimurium , to dissect the function of msDNA . In the work described here , we report that mutants lacking msDNA produced by the STM3846 reverse transcriptase are defective for colonization of the intestine using murine models of salmonellosis . This colonization defect is due , in part , to a growth defect for these mutants in anaerobic conditions . We show that mutants lacking msDNA have altered abundance of over 200 proteins in anaerobiosis , many of which are known to be required for growth in anaerobic conditions and for the pathogenesis of STm during enteric infection . Inappropriate abundance of proteins encoding alternate terminal electron acceptor reductases results in an inability of mutants lacking msDNA to utilize these compounds , inhibiting anaerobic growth in vitro . The mutants lacking msDNA can only utilize nitrate as an anaerobic terminal electron acceptor . Mutants lacking msDNA fail to colonize portions of the intestine lacking substantial neutrophilic inflammation , likely due to the ability to only utilize nitrate to support anaerobic growth . Finally , we report a similar defect in intestinal persistence for an enteropathogenic E . coli lacking its retron reverse transcriptase suggesting that msDNA is critical for enteric pathogens to thrive in the intestine of mammalian hosts . Thus , we report a role in regulating protein abundance for msDNA , the first reported natural function for any msDNA . msDNA may represent a new class of bacterial regulatory molecules .
Retron reverse transcriptases , including the STM3846 reverse transcriptase of the St-85 retron , use msr to prime reverse transcription of the msd template sequence to produce msDNA [14] ( Fig 1A ) . We generated a non-polar deletion of msd to establish that msDNA , and not some other potential product of the STM3846 RT , mediates STm colonization of the intestine . Neither the ΔSTM3846 mutant nor the Δmsd mutant produce msDNA and its production can be restored in both mutants by complementation in trans ( Fig 1B ) . The additional ORF , STM3845 , is dispensable for msDNA production . We used the murine colitis model [15] , which responds to NTS infection with profound neutrophilic inflammation in the cecum , to dissect the function of the retron in intestinal colonization . We confirmed the requirement for STM3846 in colonization of the inflamed intestine in this model ( Fig 1C ) . In addition , both the Δmsd and ΔSTM3846 mutants have indistinguishable phenotypes , suggesting that the effect of deletion of the RT is mediated by the msDNA itself . The ability of each of these mutants to colonize the intestine is rescued by complementation in trans ( Fig 1C and 1D ) . In cell culture , only the Δmsd mutant invades epithelial cells at a level mildly reduced compared to the isogenic wild type ( S1 Fig ) suggesting that reduced tissue invasion is unlikely to be the cause of the phenotype that we observed during infection of animal models . Our findings definitively link msDNA to the ability of Salmonella to colonize the intestine . The intestine is a specialized and highly diverse niche . Oxygen tensions within the lumen decline from the stomach to the colon [16 , 17] , and there is a gradient of increasing oxygen tension from the center of the lumen towards the epithelium [18] . Enteric pathogens must replicate in this hypoxic setting using both aerobic and anaerobic metabolic pathways [19 , 20] and express genes necessary for virulence in order to compete with resident microflora and colonize the host efficiently . To determine whether the intestinal colonization defect of the STm msDNA mutants could be due to an inability to grow in oxygen limited conditions , we measured the growth of our mutants in the absence of oxygen , a condition where the retron is highly expressed [21] . Both mutants unable to produce msDNA have severe growth defects in rich media in anaerobic conditions ( Fig 2A and 2B , S2 Fig ) , while the growth of these mutants in the presence of oxygen is similar to the isogenic WT in both rich and minimal media ( Fig 2C–2F ) . The necessity for msDNA during anaerobic growth is consistent with the inability of msDNA-deficient mutants to efficiently colonize the intestine . We hypothesized that msDNA might act as a trans regulator of gene expression for two reasons . First , small RNAs are well known to have regulatory properties through base pairing with DNA or mRNA transcripts [22] . Second , substantial over-expression of msDNA from one strain of E . coli in a heterologous strain lacking a retron resulted in small changes in the proteome [23] . To determine whether the msDNA produced by the St-85 retron might have regulatory properties , we evaluated the proteome of the WT and msDNA-deficient mutants ( ΔSTM3846 and Δmsd ) at late exponential phase , a time when the retron is expressed and msDNA is produced ( Fig 1B and [24] ) , in both the presence and absence of oxygen . Of the 1504 total proteins identified , no significant differences in protein abundance between the WT and mutants in the presence of oxygen were detected ( Fig 3 and S1 Table ) . This finding is consistent with previous findings that mutants lacking msDNA grow indistinguishably from the wild type organism in standard laboratory conditions ( Fig 2C–2F ) . In addition , we noted that very few proteins differ in abundance between the ΔSTM3846 and Δmsd mutants , consistent with the hypothesis that the reverse transcriptase and msDNA operate in the same biological pathway . In anaerobic conditions however , we identified 238 proteins that differed in abundance between the wild type and msDNA-deficient mutants ( Fig 3 and S1 Table ) . Forty-three percent of proteins with reduced abundance in the mutant were involved in amino acid and carbohydrate transport/metabolism and energy production/conversion ( Table 1 ) . Twenty-five percent of all proteins of altered abundance did not belong to a functional grouping ( Table 1 ) . The abundance of proteins encoded on SPI-1 was unchanged in the absence of msDNA ( Fig 3 ) . Proteins necessary for motility were increased in abundance in anaerobically grown msDNA-deficient strains ( Fig 3 ) . However , this apparent increase did not result in a change in swimming motility of these strains in anaerobic conditions compared with the WT ( S3 Fig ) . The abundance of numerous proteins known to be important for anaerobic growth and intestinal colonization was significantly reduced ( Fig 3 and S1 Table ) , including proteins for 1 , 2 propanediol utilization [25] , ethanolamine utilization [13] , anaerobic sn-glycerol-3-phosphate utilization [26] , anaerobic vitamin B12 biosynthesis [27] , and serine/threonine degradation [28] . Numerous proteins involved in reduction of anaerobic electron acceptors [29] were altered in abundance between msDNA mutants and wild type bacteria during anaerobic growth ( Fig 3 ) . Proteins important for the reduction of thiosulfate ( PhsAB ) and sulfide ( AsrC ) were of low abundance ( Fig 4 [adapted from [30]] and S1 Table ) . In addition , proteins necessary for the reduction of DMSO ( DmsA , STM4305 . s ) and fumarate ( FrdA ) were in low abundance in mutants lacking msDNA , although they did not meet our stringent criteria for statistical significance . Expression of genes necessary to utilize alternate electron acceptors is often induced by the presence of the electron acceptor [29] so the absence of a statistically significant reduction in some of these proteins is not surprising because these compounds were not present in the growth conditions we used . Interestingly , NapA , encoding the periplasmic nitrate reductase [29] , was one of the proteins that was present in increased abundance in msDNA deficient mutants compared to the WT , and there was no change in the abundance of NarGH , one of the two other nitrate reductase complexes ( Fig 4 and S1 Table ) . These data are consistent with the growth defect of our mutants in anaerobic conditions , and suggest that msDNA-deficient mutants have a severe dysregulation of proteins necessary for reduction of terminal electron acceptors needed during anaerobiosis . Our proteomic data predict that msDNA is critical for STm to produce proteins necessary for reduction of terminal electron acceptors critical for metabolism during anaerobic conditions . In order to confirm that the reduced abundance of anaerobic terminal electron acceptor reductases , as indicated by our proteomic data , has functional consequences , we tested the ability of the addition of various terminal electron acceptors to rescue anaerobic growth of the STM3846 mutant . We found that providing the alternate electron acceptors fumarate , DMSO , or thiosulfate to the culture media during anaerobic growth failed to restore growth of the strain lacking msDNA to WT levels ( Figs 5B–5F , 6B and 6C ) . This finding makes sense , as our proteomic data suggest that the enzymes that transfer electrons to these terminal electron acceptors during anaerobic growth , thiosulfate reductase , sulfide reductase , fumarate reductase , and two DMSO reductases , are reduced in abundance in mutants that lack msDNA . However , the addition of nitrate to culture medium rescued the anaerobic growth of the reverse transcriptase mutant ( Figs 5A and 6A ) . These data are consistent with our proteomic data showing that mutants lacking msDNA have adequate NarG and an increased amount of NapA allowing these strains to use nitrate as a terminal acceptor for electrons during anaerobic growth . In the presence of an intact T3SS-1 , NTS induce an inflammatory response that includes recruitment of luminal neutrophils and induction of inducible nitric oxide synthase as part of the inflammatory response [9 , 10 , 31] , resulting in generation of tetrathionate and nitrate as available terminal electron acceptors in the inflamed intestine . To determine whether the colonization defects we observed were dependent on a functional T3SS-1 and host neutrophilic inflammatory response , we performed competitive infection experiments between the virulent WT and the ΔSTM3846 mutant both in the presence and absence of SPI-1 ( Fig 7A ) . We observed that a ΔSTM3846 mutant colonizes the intestine poorly and associated organs . The modest colonization defect may be due to an inability to utilize carbon and amino acid sources within the inflamed intestine [13] , or due to poor growth compared with WT prior to the host inflammatory response . Interestingly , the colonization defect of the ΔSTM3846 mutant in the mouse cecum was exacerbated in the absence of a functional T3SS-1 , suggesting that a robust inflammatory response partially rescues mutants unable to produce msDNA ( Fig 7A ) . Consistent with this finding both the small and large intestines , which lack appreciable neutrophilic inflammation ( Fig 7C ) , are poorly colonized with the ΔSTM3846 mutant in mice inoculated with this strain alone ( Fig 7B ) . In murine models that do not develop a neutrophilic infiltrate in the intestine in response to infection ( murine typhoid model ) , the ΔSTM3846 mutant also colonizes poorly after oral infection ( Fig 8A and 8B ) . Our results suggest that STM3846 is essential for STm to colonize the intestine , a defect that is partially rescued in the presence of a profound host inflammatory response , supporting the necessity for intact anaerobic metabolic pathways in intestinal colonization . The msDNA of STm is similar in predicted secondary structure to msDNA of other enteric pathogens including enteropathogenic E . coli ( EPEC; [4] ) , a close relative of STm . EPEC attaches to the epithelial surface causing characteristic attaching and effacing lesions and a malabsorptive diarrhea [32] . Despite the fact that the pathology caused by NTS and EPEC is distinct , both organisms colonize the intestine and cause diarrheal illness in susceptible hosts . We hypothesized that the RT of EPEC O127:H6 , a serotype previously shown to produce msDNA [33] , is necessary for this organism to colonize the gut . To test this hypothesis , we generated a non-polar deletion of the retron RT ( ΔE2348C_3890 ) and performed competitive infections between this mutant and the WT . We found that an EPEC mutant lacking the RT fails to persist within the intestine of mice , both in the luminal contents and adherent to tissue ( Fig 9 ) . This defect was reversed by complementation in trans ( S4 Fig ) . These data suggest that the importance of retron reverse transcriptases during intestinal infection is not restricted to salmonellae , and thus are likely to be more broadly applicable to enteric pathogens .
The natural function of msDNA has remained elusive despite more than 30 years of study [1 , 2 , 5 , 34–38] . We describe the first phenotypes for any mutant lacking msDNA . Using the enteric pathogen S . Typhimurium , we show that msDNA produced by a retron reverse transcriptase is critical for efficient colonization of the mammalian intestine . In STm , msDNA is critical for the ability to grow in the absence of oxygen . Identification of these phenotypes creates the first opportunity for detailed studies of the molecular function of msDNA since the discovery of these unique molecules . We further showed that STm msDNA directs colonization of the intestine through regulation of the abundance of proteins necessary for central anaerobic metabolism . Thus , our data suggest that the natural function of msDNA may be to control protein abundance , the first natural function to be ascribed to any msDNA molecule . In STm , msDNA is produced by the STM3846 reverse transcriptase using msd as a template sequence and msr as a primer . The msDNA from STm has a predicted 85-nucleotide DNA stem with no mismatched base pairs and a 4-nucleotide loop , and an RNA portion with two predicted smaller imperfect stem loop structures [4] . The RNA and DNA portions of msDNA are covalently joined by a unique 2’5’ phosphodiester linkage on a conserved guanine [39] . It is unclear whether the entire msr RNA sequence remains in the mature STm msDNA . Consistent with prior reports [34 , 39] , we showed that both the RT and msd are requirements for production of msDNA . The intervening ORF , STM3845 , is dispensable for msDNA production . This is perhaps not surprising as the presence of another ORF in addition to the RT in retrons is relatively rare , and appears to be more common on retrons borne by pathogens [4 , 40 , 41] . It has been suggested that the retron RT could produce a variety of different cDNA molecules if the sequence of the mRNA transcript is identical to that of the 5’ end of msr [42] . However , we observed similar defects in intestinal colonization and anaerobic growth of mutants lacking either the RT or msd . These data suggest that it is msDNA , and not some other potential product of the RT , that mediates intestinal colonization of STm . Previous attempts to evaluate the function of msDNA have used artificial systems , failing to identify phenotypes for mutants lacking msDNA and to definitively identify the natural function of these molecules [23 , 43–47] . When an msDNA from one strain of E . coli with mismatched base pairs in the predicted DNA stem region is significantly overexpressed in a heterologous strain of E . coli lacking its own retron , the frequency of spontaneous mutation was increased due to sequestration of mismatch repair proteins [43 , 44 , 46] . Thus , the production of msDNA was thought to increase mutation frequency . However , no previous work has demonstrated that deletion of msDNA from a bacterium naturally producing msDNA decreases mutation frequency . We hypothesize that substantial over-expression of any mismatched DNA could increase mutation frequency by the same mechanism . Thus , this previous finding may not illuminate the true function of msDNA in the cell . When mutants lacking the ability to make msDNA are grown without oxygen , 15% of all proteins we could identify were in altered abundance . However , no dysregulated proteins were identified during aerobic growth , consistent with the lack of identifiable phenotypes in the presence of oxygen . Our proteomic data were generated using cultures grown for the same duration of time under varying growth conditions . Some of the differences in protein abundance may result because wild type and mutant that cannot make msDNA grow differently during anaerobic conditions . However , our growth data suggest that the growth phase of the wild type and mutants unable to make msDNA are not dramatically different at the times we chose to collect samples for our analysis . Furthermore , the differences in protein abundance between the msDNA mutant and the wild type during anaerobic growth that we re-tested appear to be functionally significant . We show that the growth of mutants that cannot make msDNA , and that have reduced abundance of several alternate electron acceptor reductases needed during anaerobic growth , cannot be rescued by addition of the cognate alternate electrons . Furthermore , the msDNA mutant overproduces periplasmic nitrate reductase ( NapA ) and a wild type level of a second nitrate reductase ( NarG ) . We show that these proteins and thus this pathway are functional , as the exogenous addition of the terminal electron acceptor nitrate rescues the anaerobic growth of mutants unable to make msDNA . Prior reports suggest that the DNA portion of msDNA can be engineered to act as a regulatory molecule by creating an antisense sequence in the DNA loop [45] . Our data suggest that the natural function of msDNA may be to act as a regulatory molecule although we have not yet identified specific regulatory targets . There are two known master regulators of anaerobic metabolism in facultative anaerobes: fnr and arcA [48] . The transcriptional and protein profiles of anaerobically-grown Salmonella mutants deficient in fnr and arcA are established [49 , 50] . With few exceptions , the proteins of altered abundance in our proteomic data align poorly with genes regulated by either fnr or arcA . However , it is difficult to draw meaningful comparisons across our proteomic data and published transcriptional profiles of mutants grown in the absence of oxygen , because of protein profiles with transcript abundance are not directly comparable . Our data raise the possibility that regulation by msDNA may represent an additional pathway to regulate the abundance of proteins necessary for anaerobic metabolism . Further mechanistic study of the anaerobic regulation of gene and protein expression is critical to understanding the behavior of Salmonella in intestinal colonization . Recent evidence suggests that Salmonella exploits the host inflammatory response to gain a competitive advantage in the intestinal lumen [10–13] . Reactive oxygen species produced by neutrophils oxidize thiosulfate to tetrathionate , a compound that Salmonella , but not resident microflora , uses as a terminal electron acceptor [10] . Epithelial-derived nitrate also contributes to the growth of Salmonella in the anaerobic conditions of the intestine by acting as a preferred electron acceptor in these conditions [31] . Some nutrients , such as ethanolamine , are used only during the neutrophilic inflammatory response [13] . We have shown that some of these processes in STm are altered in mutants unable to produce msDNA , along with many other proteins with less clearly defined roles in pathogenesis . We also observed a defect in intestinal persistence of an EPEC mutant lacking its retron RT . Salmonella enterica and E . coli are close phylogenetic relatives and both cause diarrheal illness in susceptible hosts , but there are critical differences in the retron between organisms . The RT of EPEC O127:H6 is located in a different genomic context than the retron of STm and has a GC content of 51 . 8% , similar to the average GC content of 50 . 6% [51] suggesting that this gene was not acquired recently . This GC content in the EPEC retron is in contrast to the GC content of the retron of STm , 30 . 6% compared with the average GC content of 52 . 4% [4 , 52] . Unlike STm , the retron of EPEC O127:H6 lacks an additional ORF . The predicted secondary structures of msDNA from EPEC and STm are similar , however EPEC msDNA is predicted to have mismatched base pairs in the DNA stem [4] . Despite these differences , we report that EPEC mutants lacking the retron RT also have a phenotype during colonization of the intestine . Critical differences also exist between the pathogenesis of EPEC and STm diarrheal diseases . In the intestine , Salmonella lives both in the lumen and invades the epithelium , replicating intracellularly and inducing a profound neutrophilic inflammatory diarrhea [53] . In contrast , EPEC attaches to the intestinal epithelium below the intestinal mucus in these regions of the gastrointestinal tract and remains extracellular [54] . Although our understanding of the molecular mechanism of the development of diarrhea during EPEC infection is incomplete , this infection causes a secretory diarrhea [32] . Thus , the mechanism of EPEC-induced diarrhea is substantially different than the inflammatory diarrhea caused by non-typhoidal salmonellae . Our data suggest that retron RTs are critical for colonization of the intestine by both of these pathogens , yet the phenotypes of these mutants in Salmonella versus EPEC during infection are different . While the role of retron reverse transcriptases and msDNA in intestinal colonization by enteric pathogens is likely to be ubiquitous , we hypothesize based both on our data and on the differences in diseases between these two organisms , that the processes regulated and the regulatory targets themselves are likely to be different . We show that a natural function of msDNA is to regulate protein abundance , the first reported natural function of any msDNA molecule . STm mutants unable to make msDNA poorly colonize the murine intestine . This colonization defect is due to altered abundance of numerous proteins , including those necessary for central anaerobic metabolism , a process known to be necessary for the ability of STm to colonize the intestine of mammals . We observed that an EPEC mutant lacking its retron reverse transcriptase has a reduced ability to persist in the murine intestine , suggesting that the presence and function of msDNA may be broadly applicable to other enteric pathogens . Retrons are also widespread in non-pathogenic eubacteria ( Reviewed in [3] ) including most isolates of the environmental bacterium Myxococcus xanthus [35] . msDNA is present in high copy per cell [1] , suggesting that the regulatory function of this molecule is critical for the lifestyle of the host bacterium . It is puzzling that this molecule appears to have a function under only certain conditions despite the fact that it is produced in abundance . One possible explanation for this phenomenon is that msDNA may sense environmental changes in order to regulate gene expression . This hybrid RNA-DNA molecule represents an exciting new class of bacterial regulatory molecules with broad application to the understanding of the lifestyles of pathogens and non-pathogens alike .
All bacterial strains , plasmids , and primers used for mutant construction are listed in S2 Table , S3 Table ) . All Salmonella strains are derivatives of ATCC 14028s . Enteropathogenic E . coli O127:H6 strain E2348/69 [55] , a generous gift of M . Donnenberg , is the genetic background for all EPEC mutants described here . Mutants were constructed using a modification of the lambda-red recombination technique and antibiotic resistance cassettes removed as previously described [56 , 57] [58] . All Salmonella mutations were moved into a clean genetic background by P22 transduction [59] . Standard cloning protocols were used to generate complementing plasmids [60] . All bacterial cultures were grown at 37°C aerobically with vigorous agitation or standing in an anaerobic chamber with internal atmosphere of 5% H2 , 5% CO2 , and 90% N2 ( Bactron I , ShelLab ) . For anaerobic growth experiments , bacteria were grown overnight aerobically then transferred into the anaerobic chamber and diluted 1:100 into media pre-equilibrated for at least 18 hours . Alternate electron acceptors ( Sigma-Aldrich ) sodium nitrate , sodium fumarate , sodium thiosulfate , and sodium tetrathionate were added to LB to a final concentration of 40mM . Sodium chloride ( Sigma-Aldrich ) at a final concentration of 40 mM served as a negative control . DMSO ( Sigma-Aldrich ) was added to LB to a final concentration of 0 . 1% ( v/v ) . Bacteria were grown in Luria-Bertani ( LB ) broth or LB or MacConkey ( Difco ) agar supplemented with the following antibiotics as appropriate: kanamycin ( 50 mg/L ) , nalidixic acid ( 50 mg/L ) , carbenicillin ( 100 mg/L ) , streptomycin ( 100 mg/L ) , and chloramphenicol ( 20 mg/L ) . All experiments were performed on at least three separate occasions . Bacterial generation number was calculated using the following equation: [log10 ( CFU final ) —log10 ( CFU start ) ]/log10 ( 2 ) . Ethics Statement: 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 Institutional Animal Care and Use Committees of Texas A&M University and North Carolina State University approved all animal experiments ( protocol numbers 2012–084 and 2011–167 ( TAMU ) and 14–132-B ( NCSU ) ) . All experiments that utilized mice were performed using 8–12 week old female C57BL/6J mice ( Jackson Laboratories ) . For competitive infection experiments , mice were infected by gavage with an equivalent ratio of WT and mutant bacteria . The competitive index was determined by dividing the ratio of WT to mutant bacteria in the selected organ by that ratio in the inoculum . For single infections , mice were infected with either WT or mutant bacteria . The harvested tissue was weighed , homogenized , and CFU was determined per gram of tissue collected . Salmonella infections were performed as previously described [15] . For the murine colitis model , mice were administered 20 mg streptomycin in 75 μL sterile water by gavage . Twenty-four hours after treatment , mice were infected with approximately 108 CFU of Salmonella in 100 μL volume by gavage . Feces were collected 24 hours after infection . Mice were euthanized by carbon dioxide asphyxiation at 96 hours post-infection and organs harvested , homogenized , serially diluted , and plated on LB agar with appropriate antibiotics for enumeration of CFU . For the murine typhoid model , mice were treated with 75 μL sterile water by gavage . Mice were then infected and euthanized as above . EPEC mouse infections were performed essentially as previously described [61] . Mice were infected with approximately 108 CFU in 100 μL volume by gavage . Feces were collected every other day for 9 days . Mice were euthanized 10 days post-infection . The aboral 5 cm of small intestine , the entire cecum , and the entire colon were collected . Intestinal contents were exposed through a longitudinal incision . The intestinal segment was placed into sterile PBS and vigorously agitated to remove intestinal contents . Intestinal tissue was washed in sterile PBS to remove remaining ingesta . Intestinal contents and tissue were homogenized separately , serially diluted , and plated on MacConkey agar and LB agar with appropriate antibiotics to enumerate CFU . Samples from mouse ileum , cecum , and transverse colon were collected 96 hours post-infection and fixed in formalin . All tissues were routinely processed and stained with hematoxylin and eosin . All histologic analyses were performed by a veterinary pathologist blinded as to infection group . Tissues were scored ( 0–4 ) for each of the following parameters: polymorphonuclear cell ( PMN ) infiltration , mononuclear leukocyte infiltration , crypt abscess , submucosal edema , villus blunting , and epithelial damage as described [13 , 15 , 62 , 63] . msDNA was isolated from aerobic late log phase cultures normalized by OD600 . Bacteria were lysed as for plasmid isolation ( Qiagen Mini-prep ) and msDNA isolated from the filtered fraction with subsequent ethanol precipitation . msDNA was visualized using a native polyacrylamide gel with in-gel ethidium bromide staining . Cell lines were purchased from American Type Culture Collection ( ATCC ) and used within 15 passages . HeLa cells ( human cervical adenocarcinoma epithelial , ATCC CCL-2 ) were grown as recommended by ATCC . HeLa cells were seeded in 24-well plates at 5 x 104 cells/well approximately 24 h prior to infection . Late-log phase cultures were prepared by inoculating 10 ml LB broth with 0 . 3 ml overnight shaking culture . Flasks were grown at 37°C with agitation for 3 hours . Bacteria were collected by centrifugation at 8000 x g for 90 seconds , resuspended in an equal volume of Hanks’ buffered saline solution ( HBSS , Mediatech ) and added directly to mammalian cells seeded in 24-well plates for 10 minutes . The multiplicity of infection was approximately 50 . Non-internalized bacteria were removed by aspiration . Monolayers were washed three times in HBSS and were then incubated in growth media until 30 min post-infection . Thereafter , gentamicin was added at 50 μg/ml from 30–90 min p . i . to kill extracellular bacteria and reduced to 10 μg/ml from 90 min post-infection For enumeration of intracellular bacteria , monolayers were washed once in phosphate-buffered saline , and then solubilized in 0 . 2% sodium deoxycholate and serial dilutions were plated on LB agar . Swimming motility was performed as previously described [64] . Swimming was assayed on plates containing 0 . 3% Difco Bacto Agar ( LB agar base 25g/L ) . Plates were incubated either in open air or in the anaerobic chamber overnight prior to use for swimming assays . Overnight cultures of bacterial strains were grown at 37°C with agitation and cell numbers normalized by optical density . An aliquot of each normalized culture was transferred into the anaerobic chamber . The WT , ΔSTM3846 , and Δmsd mutants ( 3 μl each ) were spotted onto the same swimming agar plate and incubated at 37°C aerobically or anaerobically for 5 hours . The diameter of the cell spread was measured and compared with that of the WT on the same plate . Each assay was performed in triplicate on three independent occasions ( anaerobic ) or in four replicates on two independent occasions ( aerobic ) . Statistical analysis was performed using GraphPad Prism 6 . All data were log transformed prior to analysis . Statistical significance was set at P < 0 . 05 and was determined using a t-test or ANOVA where indicated . Aerobic overnight cultures of the wild type and the ΔSTM3846 and Δmsd mutants were diluted 1:100 and incubated either aerobically or in an anaerobic chamber ( Coy ) for 4 hours on three independent occasions . Bacteria were pelleted and supernatants discarded . Cell pellets were resuspended in 100 mM NH4HCO3 , pH 8 . 0 and lysed by vigorous vortexing in the presence of 0 . 1 mm silica/zirconia beads . Proteins were denatured and reduced with 8M urea and 5 mM dithiothrietol , respectively , for 30 minutes at 60°C . The proteins underwent enzymatic digestion for 3 hours at 37°C with 1/50 enzyme/protein ( w/w ) ratio of sequencing-grade trypsin . The resultant peptides were desalted for mass spectrometric ( MS ) analysis using C18 solid phase extraction cartridges ( 50 mg , 1 mL , Discovery , Supelco ) . The cartridges were activated with methanol , followed by equilibration with 0 . 1% TFA before loading the samples . The cartridges were then washed with 5% acetonitrile ( ACN ) /0 . 1% TFA and eluted with 80% ACN/0 . 1% TFA . Eluted peptides were concentrated in the vacuum centrifuge and diluted to a concentration of 0 . 5 mg/mL with water for the MS analysis . Digested peptides were loaded into capillary columns ( 75 μm x 35 cm , Polymicro ) packed with C18 beads ( 3 μm particles , Phenomenex ) connected to a custom-made 4-column LC system [65] . The elution was performed using the following gradient: equilibration in 5% B solvent , 5–8% B over 2 min , 8–12% B over 18 min , 12–35% B over 50 min , 35–60% min over 27 min and 60–95% B over 3 min . ( solvent A: 0 . 1% FA; solvent B: 90% ACN/0 . 1% FA ) and flow rate of 300 nL/min . Eluting peptides were directly analyzed either on an Orbitrap ( LTQ Orbitrap Velos , Thermo Scientific , San Jose , CA ) mass spectrometer using chemically etched nanospray emitters [66] . Full scan mass spectra were collected at 400–2000 m/z range and the ten most intense ions were submitted to low-resolution CID fragmentation once ( 35% normalized collision energy ) , before being dynamically excluded for 60 seconds . Tandem mass spectra were searched with MSFG+ against Salmonella enterica serovar Typhimurium 14028s and common contaminant sequences ( downloaded from NCBI , all in forward and reversed orientations ) , using the following parameters: ( i ) partial tryptic digestion , ( ii ) 50 ppm parent mass tolerance , ( iii ) methionine oxidation as a variable modification . The peptides were filtered with a MSGF probability score [67] ≤ 1x10–9 . Peak areas for each peptide were retrieved using the MultiAlign tool [68] , and to ensure the quality of peptide-to-peak matching , the data was filtered with a Statistical Tools for AMT tag Confidence ( STAC ) score ≥ 0 . 7 and uniqueness probability ≥ 0 . 5 [69] . Additionally , proteins were required to have at least 2 peptides and at least one peptide with STAC ≥ 0 . 9 . Peptide abundance values were log transformed and rolled-up into proteins using Qrollup tool , available in DAnTE [70] . Abundance values for each protein across all 32 conditions ( WT , mutants , anaerobic , aerobic conditions , biological replicates , and technical replicates ) were used to calculate a Z-score for each measurement where missing values were filled with 19 . 5 . The Z-score transformation enables comparisons of trends across conditions and proteins to identify relevant abundance changes . | Multicopy single-stranded DNA ( msDNA ) is a unique molecule consisting of both an RNA and DNA portion . This molecule is produced by a reverse transcriptase and has no known natural function despite more than 30 years of study . We report that msDNA is important for both Salmonella Typhimurium and an enteropathogenic E . coli , two pathogens that cause diarrhea in susceptible hosts , to survive in the intestine . Using mutant strains incapable of producing msDNA , we show that msDNA is needed for Salmonella to grow in the absence of oxygen . Mutants grown in oxygen-deficient conditions have substantial changes in overall protein composition , including numerous proteins known to be important for anaerobic metabolism and growth in the intestine . Our findings link msDNA to the ability of Salmonella to thrive in an oxygen-deficient environment similar to the conditions inside the gut . We report that msDNA regulates the quantity of proteins , the first natural function attributed to this molecule . msDNA may represent a new class of regulatory molecules . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [] | 2015 | Multicopy Single-Stranded DNA Directs Intestinal Colonization of Enteric Pathogens |
Despite the growing number of immune repertoire sequencing studies , the field still lacks software for analysis and comprehension of this high-dimensional data . Here we report VDJtools , a complementary software suite that solves a wide range of T cell receptor ( TCR ) repertoires post-analysis tasks , provides a detailed tabular output and publication-ready graphics , and is built on top of a flexible API . Using TCR datasets for a large cohort of unrelated healthy donors , twins , and multiple sclerosis patients we demonstrate that VDJtools greatly facilitates the analysis and leads to sound biological conclusions . VDJtools software and documentation are available at https://github . com/mikessh/vdjtools .
The advent of high throughput sequencing ( HTS ) has opened a new venue for the studies of genomics of adaptive immunity that involve deep profiling of T-cell receptor ( TCR ) and B-cell receptor ( BCR ) gene repertoires encoding a myriad of antigen specificities . Huge volumes of complex data produced by the immune repertoire profiling have led to the development of a diverse set of software tools , which often complement each other . We [1–3] and others [4–7] have recently contributed several tools that handle large amounts of raw HTS data to process it into a human-readable list of clonotypes characterized by Variable ( V ) , Diversity ( D ) , Joining ( J ) segments and V- ( D ) -J junction sequences of receptor genes . While such processed data carry nearly exhaustive information on the sampled immune repertoire , this information yet needs to be convolved , scaled and compared across various samples to result in sound biological conclusions . Post-analysis of immune repertoire data is a challenging task owing to extreme diversity of TCR and BCR sequences . For example , in technically similar microbiome profiling by 16S rRNA sequencing one deals with thousands of operational taxonomic units that represent various species [8] , while typical TCR repertoire samples may contain hundreds of thousands [9 , 10] of clonotypes . Moreover , the species phylogeny and annotation is well developed in the field of microbiology [11] , while immune repertoires remain poorly annotated . To illustrate this , a simple query with “16S rRNA” currently yields more than 8 million records in GenBank , while there are only 37 thousand records annotated as “T-cell receptor” . However , unsupervised methods of studying repertoires , for example based on sample overlap , could turn out very promising , as there exists a relatively limited diversity of overlapping clonotypes [12–15] . In the light of recent advances in storage and processing of immunological big data [16] , community-driven initiatives for immune repertoire data sharing and analysis are likely to emerge , for example VDJserver portal [17] which is currently under development . There are several commonly used ways to survey immune repertoire information obtained from HTS , such as tracking individual clonotypes [18 , 19] , comparing immune receptor segment usage [20 , 21] and comparing repertoire diversity [10] . Still those are overwhelmingly performed using in-house scripts or even manually . This is becoming a major obstacle , as comparison and annotation of samples based on data generated in other studies is critical for comprehensive analysis of immune repertoire sequencing data . In contrast , similar fields , such as metagenomics , have a plethora of such instruments [22] . The VDJtools software package presented here aims at filling this gap by incorporating a comprehensive set of routines for analysis of TCR repertoire sequencing data ( Fig 1 ) . The variety of implemented algorithms range from basic statistics calculation and clonotype table filtering to advanced routines such as repertoire clustering and computationally intensive routines such as clonotype table joining . VDJtools can calculate basic immune repertoire statistics that were commonly used in pre-HTS era repertoire analysis . Those include in silico spectratype ( the distribution of lengths of CDR3 nucleotide sequences ) that was first introduced with corresponding molecular biology assay [23] , and various Variable/Joining segment usage statistics that root in flow cytometry analysis of T- and B-cell populations . The framework provides means for analyzing the diversity of immune repertoires , such as normalized unique clonotype counts ( with an option to account for convergent recombination ) , clonotype frequency distribution , as well as rarefaction curves and lower bound estimates of total repertoire diversity widely applied in ecology field [24] . The concept of repertoire diversity is of great importance , as it reflects the ability of our immune system to effectively withstand a multitude of encountered pathogens [25] . By applying computational methods one could estimate how the diversity is influenced by various processes , such as aging [10] , vaccination , and infection [26] . Diversity measures could also be used to compare the structure of T- and B-cell repertoires in samples derived from a variety of tissues and subjects [27] . Advanced set of VDJtools methods includes cluster analysis of repertoire samples and clonotype tracking which have a wide range of applications . Machine learning methods such as hierarchical clusterization and multi-dimensional scaling can aid in learning T-cell antigen specificities and disease biomarker patterns from high-dimensional TCR data [28] . Clonotype tracking is useful in studying immune repertoire dynamics associated with vaccination [29] , autologous hematopoietic stem cell transplantation ( HSCT ) [19 , 30 , 31] , checkpoint inhibitors [32] , etc . , as well as in detection of minimal residual disease in lymphoid malignancies [33–37] . An overview of 20 recently published immune repertoire studies ( S1 Table ) demonstrates that VDJtools can perform most of emerging post-analysis tasks therefore greatly facilitating the analysis process and removing the need to develop multiple custom scripts . Currently there are few software tools capable to perform post-analysis of immune repertoire data [7 , 38 , 39] , all of which provide less functionality when compared to VDJtools ( S2 Table ) . Moreover , in contrast to VDJtools which can handle output generated by various pre-processing software , these tools only support datasets in their internal formats .
The study was approved by ethics committee of the Russian Children's Hospital from January 20 , 2011 . The core API of the software is implemented in Java/Groovy languages and automatically resolves all dependencies during compilation using Maven . The API includes generalized entities , such as Clonotype , Sample and SampleCollection classes , and allows storing sample metadata using MetadataTable class . The API also contains a comprehensive set of routines for computing sample-specific and cross-sample statistics , which are optimized for parallel computation . VDJtools API can be easily integrated in any software written in Java or related programming languages ( e . g . Groovy , Scala and Clojure ) . VDJtools is an open-source software , the source code can be accessed at GitHub [40] . Comprehensive software documentation is hosted at ReadTheDocs [41] and contains basic usage guidelines ( including the description of common pitfalls ) , a summary of implemented algorithms , as well as examples that cover some typical VDJtools usage cases . The documentation also contains step-by-step instructions for reproducing the analysis described in present paper . VDJtools has a command line interface that allows executing analysis routines that produce tabular and publication-ready graphical output . Tabular output can be used for post-hoc analysis in R or explored in spreadsheet software such as Excel . Plotting parameters are optimized to provide the most intuitive and comprehensive graphical representation for most usage cases while users can specify their own sample groups and factors to be visualized . VDJtools accepts tabular output of commonly used pre-processing software: MIGEC [2] , MiTCR [1] , ImmunoSEQ [38] , IMGT/HighV-QUEST [4] , and MiXCR [3] . VDJtools also supports IgBlast [5] software format . Of note , using IgBlast requires a considerable amount of parsing and post processing , as it only reports Variable segment alignment and doesn’t provide the CDR3 sequence . Moreover , vanilla IgBlast doesn’t accept FASTQ format input , does not provide clonotype assembling ( grouping of sequencing reads with identical Variable segment , Joining segment and CDR3 sequence ) and is not optimized for parallel computations . We have implemented all those features in our wrapper for IgBlast software , MIGMAP , that could be downloaded from [42] . VDJtools converts all input datasets to its own internal format , which is a tab-delimited table containing abundance , CDR3 sequence , V , D and J segment names and markup of CDR3 sequence germline regions . An immune repertoire browser VDJviz which serves as a lightweight GUI for VDJtools was built using Play framework and VDJtools API and could be accessed at [43] . Raw data for multiple sclerosis patients is deposited at SRA ( PRJNA280417 ) . Pre-processed clonotype tables can be found in a separate GitHub repository [44] , which also contains shell scripts that can be used to reproduce the analysis .
We have started our analysis by comparing the repertoire diversity of MS and control samples . To support the diversity measure choice and check for possible biases we have performed a benchmark on previously published T cell immunity aging data [10] and additional ANOVA analysis to identify factors that bias diversity estimates ( S1 Text , S1 Fig , S4 Table ) . We have used common diversity measures: the observed diversity ( number of unique clonotypes ) , Chao [47] and Efron [48] estimates for lower bound on total species diversity , Shannon [49] and Simpson [50] indices , as well as extrapolated Chao estimate [51] . The benchmark , in which correlation with a physiological ( age ) and immune status ( naïve T-cell count ) factors was compared for various diversity estimates , has shown that best correlation can be achieved when samples are normalized to the same size ( TRBM count ) . Correspondingly , ANOVA analysis suggests a strong sampling-related bias . Accounting for this bias is especially important in present case as the rarefaction curves are far from saturation ( Fig 2A ) . Notably , lower bound estimates of total repertoire diversity that are especially affected by sampling bias were applied in some recent studies for the comparison of TCR repertoire diversity under uneven sample sizes [9 , 52] . Using Chao1 estimate [47] for normalized datasets that has shown the best performance together with Efron estimate ( yet is far simpler to compute ) in the aforementioned benchmark , one can discover that MS samples have a significantly lower diversity than the controls ( Fig 2B ) . This suggests a substantial expansion of T-cell clones in peripheral blood of MS patients , an observation previously supported only by local measurements such as Sanger sequencing of individual T-cells and spectratyping assays [53] . As control population is slightly older than MS group one can expect even more profound difference in case exact age matching is achieved for the control group [10] . Still , there is no significant difference for the directly observed sample diversity ( S2 Fig ) , which is likely due to the fact that this estimate doesn’t account for the clonotype frequency distribution in sample and thus is less sensitive . As there is currently no study describing an application of cluster analysis to a large set of immune repertoire datasets coming from different individuals , we have performed a benchmark of various clustering strategies using a recently published twins TCR repertoires study [54] . We have tested the ability to distinguish TCR repertoires of identical twins from those of unrelated individuals for several commonly used similarity measures , correlation of overlapping clonotype frequencies ( R ) , geometric mean of total frequencies of overlapping clonotypes ( F ) , normalized number of overlapping clonotypes ( D , [14] ) , Jaccard [55] and Morisita-Horn indices [56] . Only the F similarity measure showed significant difference for both TCR alpha and beta chain datasets ( S1 Text , S5 Table and S3 Fig ) . At the same time , it should be noted that R and D measures also proved to be useful in other experimental setups . For example , R measure accurately separated TCR alpha repertoires for the T cell subsets and tissues , as well as mutant and control mice Treg repertoires [57] . We have next used cluster analysis to explore whether TCR beta repertoires of MS patients can be distinguished from healthy controls . As some samples were prepared in parallel with single-end sample barcoding , joined and then co-amplified after Illumina adapter ligation , we first checked for the possibility of cross-sample contamination ( S4 and S5 Figs ) . It turned out that direct clustering of samples with F measure resulted in a strong co-clustering of samples prepared in the same batch . To correct for batch effect , we have selected “amino acid NOT nucleotide” clonotype intersection matching rule , i . e . matching of CDR3 amino acid , but not the nucleotide sequences . Hierarchical clustering with F similarity measure and “amino acid NOT nucleotide” clonotype matching rule showed some co-clustering for control but not MS datasets ( Fig 3A ) . Further exploration with multidimensional scaling ( MDS ) method showed that control repertoires of healthy children are more similar to each other according to F similarity measure , while MS repertoires are all different ( Fig 3B and 3C ) . This result is quite similar to our observations of age-related changes in TCR repertoires ( our unpublished data ) . With aging , expansion of antigen-specific clones moves away native repertoires that are initially more close to each other due to the public clonotypes that are frequently produced in recombination [58] . This is in line with observation of early clonal T-cell expansions in MS children ( see “Estimating repertoire diversity” section above ) . Since those expanding T-cell clones , including potentially autoreactive ones , are predominantly private to an MS-affected person [59–61] this leads to the decrease of the overlap between MS repertoires according to the clonotype size-weighted F similarity measure . Keeping in mind that MS was shown to have a Type I-II TCR repertoire bias [62] , i . e . the same prominent Variable segment is used , yet only limited homology between CDR3 region is present in disease specific T-cells , we have performed hierarchical clustering of Variable segment usage profiles ( Fig 3D , note that profiles are weighted by TRBM count ) . The resulting dendrogram distinguishes MS patients and healthy donors with 91% sensitivity and 77% specificity ( P = 0 . 013 , Fisher’s exact test for cluster—group association ) . A post-hoc testing was then performed to find out which Variable segments were more abundant in MS donors than in healthy controls ( S6 Table ) . We have determined that 5 Variable segments had a statistically significant increase in frequency , including TRBV5-6 ( 1 . 6-fold , P = 2x10−5 ) and TRBV5-1 ( 1 . 5-fold , P = 5x10−4 ) , which were previously reported to have a genetic association with MS [61 , 63] . Of note , TRBV20-1 ( 1 . 3-fold , P = 2x10−3 ) which has also emerged in our results was recently shown to have no genetic association with MS in a Sicilian population carrying null allele [64] . This suggests that the observed TRBV20-1—MS association could be either specific for Russian population or represent an indirect biomarker . Further we have compared TCR repertoires of blood samples taken from a single MS patient ( MS8 ) before and after HSCT ( see Fig 4 ) . We have first tracked the clonotypes present before HSCT procedure to the post-transplantation repertoire ( Fig 4A ) . The resulting plot clearly shows that pre-transplantation clones greatly expand ( from ~25% of TRBMs to 75% ) and occupy most of homeostatic space in post-HSCT repertoire . The magnitude of this effect resembles the one we previously observed in an ankylosing spondylitis patient HSCT case [19] and in adult MS autologous HSCT study [31] . Another peculiar finding is that a strong shift in Variable segment usage is observed , while no such change is present for J segment usage ( Fig 4B ) . TRBV15 and TRBV7-8 ranking 10 and 5 replaced the top two Variable segments TRBV20-1 and TRBV29-1 , while top two Joining segments TRBJ2-7 and TRBJ2-1 remained the same . This could not be attributed to CD4/CD8 balance alone , as there is strong differential Joining segment usage between those two populations [65] . Interestingly , a significant HSCT-induced decrease was observed for TRBV5-6 , TRBV5-1 , TRBV5-8 , TRBV7-6 and TRBV20-1 ( P = 0 . 008 , two-tailed paired T-test for log TRBM frequencies ) segments that were enriched in MS patients compared to healthy controls ( see previous section ) . The total frequency of those segments dropped from 20% of TRBMs to 14% . Finally , we have compared CDR3 regions of MS patients to healthy donors using a set of basic features: the length of Variable and Joining segment germline parts remaining within CDR3 region , and VJ junction ( NDN ) size . The length of CDR3 segment itself is a potent marker of antigen receptor reactivity . For example , longer CDR3 sequences may be more characteristic for potentially cross- and self-reactive immune receptors [66] , while CDR3 variants with low number of randomly added “N” nucleotides are characteristic for public clonotypes , including variants specific to common pathogens such as EBV and CMV [67] . As our analysis shows , MS patients are characterized by longer VJ junction region ( Fig 5A ) . To check whether it is due to specific segment usage profile we have compared VJ junctions from all clonotypes of normal donors to the ones coming from clonotypes that have one of Variable segments previously shown to be over-expressed in MS patients ( Fig 5B ) . We have found that aforementioned TRBV5-6 , TRBV5-1 , TRBV5-8 , TRBV7-6 and TRBV20-1 are intrinsically characterized by longer VJ inserts . However , there is still a significant difference in VJ junction size between MS patients and controls for this subset of TRBV segments ( Fig 5C ) . These results may indicate that clonal expansions in MS patients are characterized by more self-reactive T-cell clonotypes than in healthy donors . Alternatively , this could be a more general hallmark of chronic inflammation associated with MS .
A cross-platform binary version of software in a form of executable JAR file is available from [68] . VDJtools software is free for scientific and non-profit use . The source code is available at GitHub repositories [40] and [69] . One important aspect of VDJtools usage not mentioned in the results section is the benchmark of pre-processing software ( S6 Fig ) and library preparation protocols . For this purposes we plan to constantly update VDJtools so it is able to handle the output of newly developed pre-processing software . In future we plan extending VDJtools software to address another highly important problem in the field , the analysis of antibody repertoire [70] . While being applicable to the analysis of BCR clonotypes , VDJtools currently doesn’t account for somatic hypermutations and therefore yet cannot offer a comprehensive analysis for the antibody repertoires . This task requires us to implement algorithms for computing statistics of hypermutation transition patterns and reconstruction of B-cell clonal lineages and visualization of hypermutation graphs . We are also looking forward for the feedback from the community to meet the demand for some exciting novel features that will surely arise in this rapidly growing field . | High-throughput profiling of T- and B-cell antigen receptor repertoires promises great advances in our understanding of the mechanisms underlying adaptive immune system function , treatment of autoimmune and infectious diseases , and development of novel approaches in cancer immunotherapy . A number of recently developed software tools aim at processing immune repertoire data by mapping Variable ( V ) , Diversity ( D ) and Joining ( J ) antigen receptor segments to sequencing reads and assembling T- and B-cell clonotypes . Nevertheless , there still exists a major gap in common methods of data post-analysis in the field: there is no standardized data format so far , and most of data comparative analysis is carried out using a variety of in-house scripts . Here we present VDJtools , a software framework that can analyze output of most commonly used TCR repertoire processing tools and allows applying a diverse set of post-analysis strategies . The main aims of our framework are: To ensure consistency of post-analysis methods and reproducibility of obtained results; to save the time of bioinformaticians analyzing TCR repertoire data by providing comprehensive tabular output and open-source API; and to provide a simple enough command line tool so that immunologists and biologists with little computational background could use it to generate publication-ready results . | [
"Abstract",
"Introduction",
"Design",
"and",
"Implementation",
"Results",
"Availability",
"and",
"Future",
"Directions"
] | [] | 2015 | VDJtools: Unifying Post-analysis of T Cell Receptor Repertoires |
How cells regulate their size from one generation to the next has remained an enigma for decades . Recently , a molecular mechanism that links cell size and cell cycle was proposed in fission yeast . This mechanism involves changes in the spatial cellular distribution of two proteins , Pom1 and Cdr2 , as the cell grows . Pom1 inhibits Cdr2 while Cdr2 promotes the G2 → M transition . Cdr2 is localized in the middle cell region ( midcell ) whereas the concentration of Pom1 is highest at the cell tips and declines towards the midcell . In short cells , Pom1 efficiently inhibits Cdr2 . However , as cells grow , the Pom1 concentration at midcell decreases such that Cdr2 becomes activated at some critical size . In this study , the chemistry of Pom1 and Cdr2 was modeled using a deterministic reaction-diffusion-convection system interacting with a deterministic model describing microtubule dynamics . Simulations mimicked experimental data from wild-type ( WT ) fission yeast growing at normal and reduced rates; they also mimicked the behavior of a Pom1 overexpression mutant and WT yeast exposed to a microtubule depolymerizing drug . A mechanism linking cell size and cell cycle , involving the downstream action of Cdr2 on Wee1 phosphorylation , is proposed .
Dividing cells maintain a stable size from one generation to the next . This suggests that they contain homeostatic mechanisms in which the division cycle is triggered when a particular size is attained . However , the biochemical mechanisms for this have remained unknown puzzles for decades . Sensing mechanisms appear restricted to monitoring concentration changes , so how can such changes reflect cell volume ? Volume and concentration are different types of quantities; the former is sensitive to changes in scale while the latter is not . This issue has been discussed [1] and possibilities have been proposed . Most of these involve measuring the time required for a cellular component to reach a critical concentration beyond which mitosis is triggered [1] , [2] . Because their cell-length phenotypes are directly linked to the time spent in specific cell cycle stages , fission yeast Schizosaccharomyces pombe are especially useful in understanding the relationship between cell size and cell cycle [3] . These 7 µm long rod-shaped newborn cells grow lengthwise to ∼14 µm at which point they divide . A mechanistic model of how these cells might sense size , involving Pom1 and Cdr2 proteins as major players , was recently proposed [4] , [5] . Pom1 is a kinase involved in cell polarization and in establishing the cell division plane [6] , [7] , [8] . Cdr2 is a serine-threonine protein kinase that promotes the G2/M transition by inactivating Wee1 , an inhibitor of Cdc2 [3] , [9] , [10] . In the proposed mechanism , Pom1 inhibits Cdr2 . The size-dependent relief of this inhibition indirectly activates Cdc2 , which promotes entry into mitosis ( Figure 1A ) [4] , [5] . The cell-size-dependence of Pom1 and Cdr2 are proposed to originate from the relative spatial distributions of the two proteins . Pom1 forms a spatial gradient that peaks at the cell tips and decreases towards the middle of the cell ( midcell ) ( Figure 1B ) . This gradient arises from an indirect interaction with microtubules ( MTs ) , mediated through the Tea1 protein [6] , [7] , [11] . During interphase , Tea1 is transported from the nuclear region of the cell to the tips by both “walking” along microtubules and by “riding” on microtubules' growing ends [11] , [12] , [13] . Microtubules occasionally undergo catastrophic collapse , releasing Tea1 in the process . Catastrophe occurs with higher frequency at the tips , causing Tea1 to be delivered preferentially to these regions [14] , [15] . Tea1 anchors to the membrane in a complex positive-feedback process [16] , [17] . Anchored Tea1 recruits Pom1 from the cytosol , sequestering it to the membrane and giving rise to the Pom1 spatial gradient . Conversely , Cdr2 is found in cortical node-like structures on the cell membrane in the midcell region during interphase . Midcell localization appears to be Pom1-dependent , because in cells lacking Pom1 , Cdr2 spreads broadly from midcell to the non-growing end [4] , [5] . The Pom1 gradient is present throughout interphase , but the concentration of Pom1 at midcell is length- ( and thus size- ) sensitive . During early interphase , the Pom1 concentration at midcell is sufficiently high to inhibit Cdr2 from advancing the cell from G2- to M-phase . When the cell reaches a particular length , the Pom1 midcell concentration declines enough for this inhibition to be relieved . This allows Cdr2 to trigger a cascade ( Figure 1A ) that ultimately advances the cell to the M-phase of mitosis . Previous mathematical models have described the control of the G2/M transition [18] , [19] , [20] , [21] . Although some describe the main cell cycle proteins in detail , none includes a specific mechanism for measuring cell size . Here we propose a simple 1D reaction-diffusion-convection mathematical model for a cell-size checkpoint based on the recently proposed Pom1:Cdr2 mechanism . We have attempted to make the model minimal in terms of the assumptions , reactions and components required to exhibit checkpoint behavior as arising from the spatial cellular dynamics of Pom1 and Cdr2 during interphase . The framework combines known chemical features of Pom1 and Cdr2 with the known dynamics of microtubules . The model reproduces phenotypes of a mutant fission yeast strain as well as the effects of two drugs . Our simulations demonstrate that the proposed checkpoint mechanism is feasible from a quantitative perspective .
The Pom1:Cdr2 reaction-diffusion-convection model assumes diffusion along a 1D mesh ( Figure 2A ) . This subsystem involves Pom1 and Cdr2 in cytosolic and membrane-bound forms ( Pc , Pm , Cc , Cm ) . Membrane diffusion is significantly slower than cytosolic diffusion . Pom1 can partition between cytosolic and membrane-bound forms through an uncatalyzed reversible reaction . Cc inserts into the membrane where it is multiphosphorylated by Pom1 . Once fully phosphorylated , Cdr2 is expelled from the membrane and simultaneously dephosphorylated . This mechanism assumes an ordered distributive chain of enzymatic reactions [22] . For simplicity , dephosphorylations are catalyzed by an unspecified and implicit phosphatase whose concentration is assumed to be constant throughout the cell cycle . To model cell growth , equations were derived within a growing domain framework [23] , [24] . Here , we fixed the 1D mesh length by normalizing the x-axis coordinate to cell length ( L ( t ) ) ( 1 ) Thus , one can ensure that a system described in this new coordinate , is bounded within the interval given that its domain is expressed in the old coordinate x by the function L ( t ) . In this fixed domain , the number of mesh points does not change with time such that standard numerical methods can be applied [23] . The interpretation is that the real cellular region represented by a given discretization point is growing . However , because of this fixed domain , the discretized interval and the number of intervals used in the numerical calculations , namely 100 , are invariant with time . This method reduces resolution but not precision ( Figure 2B ) . Given this fixed domain strategy and corresponding rates assigned for the reactions of the chemical model , the system can be described as ( 2 ) Where ( 3 ) subject to no-flux boundary conditions for all components and non-negative initial data . Subscript n refers to the total number of phosphorylation sites on Cdr2 and j refers to those that are phosphorylated . The denominator for each diffusion and convection term of system ( 2 ) arises from the spatial normalization of the growing domain . We assume exponential uniform growth at a time-invariant rate α ( 3 ) . Because of the unidimensional mode of growth in fission yeast , only cell length L ( t ) was required to be modeled . Interfacial reactions ( between cytosol and membrane ) were normalized by the ratio of the surface area to cytosol volume . But because cell volume was approximated by a cylinder , this ratio remained constant , with growth exclusively along the long axis . This allowed us to embed this interfacial normalization ratio into reaction rate constants . The spatial system was numerically solved using the Crank-Nicolson method implemented in Fortran . This subsystem involves microtubule reaction dynamics , which are required for generating the Pom1 and Crd2 spatial gradients . All assumed reactions are given in Table 1 . Although microtubules actually consist of two tubulin isoforms ( α and β ) , only one “lumped” isoform ( T ) was used in the model . Two forms of T were specified , including a GDP-bound form ( TD ) that exchanges irreversibly with GTP to generate a GTP-bound form ( TT ) . An implicit nucleation site reversibly transforms non-growing TT monomers into a growing form ( ) to which units can add . Modeling the addition of each subunit would be impractical , as the real number of subunits per µm , Ns = 1625 [25] , and microtubules in newborn cells can be as long as 3 . 5 µm . We reduced this complexity by discretizing microtubules into N increments each Δd = 0 . 07 µm long . One polymerized MT subunit corresponded to Ns⋅Δd number of monomers , with the reaction designated as ( 4 ) In the rate expression for this reaction , the concentration dependence of TT was not raised to the power to avoid numerical instability . This simplification is reasonable because polymers with different lengths have the same velocity of elongation [15] . Growing polymers convert into shrinking ones ( ) through an irreversible uncatalyzed reaction . The concentration of a microtubule of length i is the sum of growing and shrinking forms , ( 5 ) The rate of this conversion reaction depends on the length of the microtubule relative to cell length , with faster rates occurring with longer polymers [26] . This length-dependence was included in rate constant kcat ( i ) ( Table 1 ) . The reaction for the depolymerization of shrinking polymers was treated analogously . Given the assumptions and reactions described above , the microtubule subsystem is represented by the set of ODEs ( 6 ) subject to non-negative initial data . In the dTD/dt and dTT/dt equations , the factor weights the depolymerization and elongation reaction rates according to the number of TD subunits released or TT subunits consumed per reaction event , respectively . The α-dependent terms in ( 6 ) represent dilution due to cell growth . The microtubule and Pom1:Cdr2 subsystems interact through reactions involving Pom1 and microtubules . In fission yeast , the spatial distribution of Pom1 depends on the cellular movement of Tea1 [6] , [11] , [12] . In our model , Tea1 was not modeled explicitly; rather , it was lumped with Pom1 . Tea1 is transported to the cell tips by “riding” on the microtubules' “plus” ends [27] . It also “walks” along microtubules , as cargo of the motor protein Tea2 [11] , [12] , [28] . Both processes impose a directional velocity to cytosolic Tea1 . In our model , the physical transport of Tea1 is described by a Pom1 convection velocity which depends on microtubule concentrations , as calculated from the MT subsystem . This term serves to transport Pom1 to the cell tips to create the spatial gradient . Without this term , there would be no spatial gradients . The convection velocity was composed of two terms , and , corresponding to riding and walking transport modes , respectively . These terms have the form ( 7 ) In equation ( 7 ) , depends on the number of growing microtubules at a specific spatial point x , while depends on the total number of microtubules passing through that point . Both are calculated from the microtubule concentrations given by system ( 6 ) . and represent experimentally estimated [29] average velocities of elongating microtubules and of Tea2 moving along a MT polymer , respectively . The symbol represents the ceiling function and it is used to calculate the number of normalized discrete MT subunits ni corresponding to a continuous length of x µm . Subunit normalization is described below . From the average velocity of polymer elongation and number of tubulin subunits per micrometer ( NS ) , it is possible to calculate the average number of tubulin subunits added to a given growing polymer per unit of time . An average polymer elongates at the rate ( 8 ) where NA is Avogadro's number and Vc is cell volume . Because the velocity of polymer elongation was estimated from different polymers at different cell volumes , we used an average fission yeast volume ( 74 µm3 assuming a cell radius of 1 . 5 µm ) to calculate the reaction rate . This rate was equated to the rate-law expression for the elongation of a given polymer in our model . The rate was also normalized by the number of tubulin monomers included in one polymerized MT subunit . Using known values for tubulin allowed the rate-constant kel to be calculated as ( 9 ) The depolymerization rate constant was calculated analogously , using the average velocity of shrinkage ( υdp ) . In this case , the rate of depolymerization ( Rdep ) was equated to ( 9 ) which allowed kdep to be calculated as ( 10 ) Tischer et al . determined the frequency of catastrophe for MTs of different lengths by analyzing GFP-tubulin dynamics obtained from fluorescence experiments in fission yeast [15] , [30] . They measured the number of catastrophe events and the MT growth time within defined cellular regions of a statistically significant number of cells of different lengths . We used these data to calculate the catastrophe rate-constant associated with the reaction used in the MT subsystem ( Figure 3 ) . An empirical exponential function ( solid line in Figure 3 ) was fitted to the number of microtubule catastrophes per unit of normalized cell length per unit time . Because catastrophe reactions are first-order ( Table 1 ) , the catastrophe frequency estimated from the exponential regression was defined to be . Nucleation rate-constants knuc1 and knuc2 were set such that simulations yielded an average of 3 . 6 “full length” ( touching the cell tips ) microtubules per cell , based on the observed number of MTs in 73 cells ( see supplementary material of [31] ) . The rate constant for GTP exchange , kex , was set such that an apparent steady-state was reached within 10 min , as reported [32] . The microtubule subsystem was created to provide “tracks” for Pom1 transport within the context of a 1D discretization of the growing cell [33] . The mathematical MT subsystem developed above is not spatially dependent even though it includes components that possess a spatial dimension . To use the time-dependent MT model in the growing cell framework , we assumed that all MTs are nucleated at the 2 central nodes ( nodes i = 50 and i = 51 in Figure 4C ) and that they grow in an antiparallel manner along the axis towards the mesh ends ( i = 1 and i = 100 ) . Cell growth was incorporated into the MT model in the following manner . As cellular volume increased , microtubules were allowed to increase their length until they reached the cell tips . Cell growth ( allowing MTs to grow longer ) was included in the model by increasing the number of microtubules subunits i for the longest polymer accordingly to the incremental growth of the cell length . During simulations , a new ODE associated with a newly added polymer was added each time the cell length increased by Δd µm . The concentration of the new polymer was assumed to be zero at the moment it was introduced . Therefore , the time point representing the longest cells ( where volume had doubled ) contained ∼ twice as many equations as at the beginning of a simulation ( Figure 4 ) . MT lengths were normalized before being positioned into the spatial framework . Microtubule MTi was normalized into a fixed domain , MTni , where the new number of polymerized subunits ni in this domain was given as ( 11 ) In equation 11 , is the length of the real cellular region represented by a discretized point in the fixed domain used to solve the spatial model ( Figure 2B ) . In this way , the results of the MT subsystem were included in the fixed mesh where the Pom1:Cdr2 subsystem was solved . In fission yeast , Cdr2 is part of a complex regulatory signaling cascade that ultimately triggers mitosis [4] . In our model , this cascade was simplified to a zeroth-order ultrasensitive switch modeled as a standard Goldbeter-Koshland function [34] . The switch is only a function of Cdr2 in the midcell region , which we presume correlates to Cdr2 in cortical nodes . The switch is indirectly affected by Pom1 . Since the Pom1 concentration at the midcell region decreases with cell growth , the total Cdr2 membrane form at midcell increases , triggering mitosis when the cell reaches a specific length . This presumption is supported by the co-localization of Wee1 and Cdr2 in medial cortical nodes [35] ( diffusely located in the midcell region ) and by the localization of Wee1 to the nuclear envelope which is also in the midcell region [4] . Wee1 inhibits mitotic entry when it is not phosphorylated , which may be controlled by cellular localization of the two proteins . Cdr2 may effectively inhibit Wee1 only when both proteins are localized to the cortical nodes .
We matched the spatial dimension of the MT subsystem to the spatial discretization interval of the Pom1:Cdr2 subsystem by making . We set the maximum length of microtubules to 3 . 5 µm ( maximum number of subunits N = 50 ) at the beginning of a simulation , representing a newborn cell . At the end of a simulation ( with cell volume doubled ) , N = 100 . Figure 5 shows the change in concentration of MTs as the cell grows . As the volume increased , the overall MT concentration declined exponentially with time . However , the average number of polymers touching the cell tips ( 3 . 6 ) was kept constant between 10 min and the end of the simulation [31] . The experimental data used to build the MT subsystem were associated only with the microtubule bundle tip - the longest MT in the bundle [15] . Thus , the associated parameters used in simulations may slightly overestimate the concentration of the longest MTs and underestimate the concentrations of shorter polymers . The growing domain framework was assigned a doubling-volume time of 100 min . The cellular concentrations of Pom1 and Cdr2 were assumed to be directly proportional to the published fluorescence intensities of their spatial distributions [5] . The overall Pom1 fluorescence was normalized to 2000 copies of Pom1 at the beginning of the simulation [7] . This normalization factor was also used to estimate the number of Cdr2 copies to be 1855 in short cells . In long cells , the areas under the Pom1 and Cdr2 fluorescence curves were greater by ∼62% and 20% , respectively . These copy-number changes were used to calibrate the constant feed terms for Cc and Pc in system ( 2 ) . Diffusion coefficients were estimated from literature values [7] . All other rate constants for the reaction-diffusion-convection system ( 2 ) were empirically adjusted to fit simulations to the normalized fluorescence data with greatest fidelity . Numerical simulations began with a short cell ( 7 µm ) ; initial Pom1 concentrations declined from the poles toward midcell while Cdr2 concentrations were maximal at midcell ( Figure 6B , t = 0 ) . Final concentrations in long cells ( at t = 100 min ) reproduced the fluorescence data with reasonable fidelity ( Figure 6A ) . The convection term reflecting Pom1 transport , as defined by the microtubule dynamics , was more influential than diffusion , affording higher Pom1 concentrations at the cell tips . As the cell lengthened , the limited amount of Pom1 in the cell was predominantly delivered to the tips , leaving a deficiency of Pom1 at midcell . This allowed Cdr2 to accumulate at the cell cortex . Simulations with different initial Pom1 and Cdr2 distributions afforded nearly identical final spatial profiles , demonstrating model robustness ( see Figure 7 for one case ) . These simulations also show the ability of the system to focus Cdr2 to the midcell region as part of the positioning mechanism of the eventual actomyosin ring [36] , [37] . The model was sensitive to parameters related to Pom1 behavior such that formation of the Pom1 gradient dictated the general model behavior . However , different combinations of Pom1 diffusion , Pom1 transport velocities and rate constant for Pom1 detachment from the membrane produced similar overall dynamics . Thus , it is unlikely that the set of parameters used here are unique in their ability to elicit the desired dynamical behavior . Simulations reflecting different experimental conditions were performed to assess the degree to which the model reproduced the cell-size checkpoint behavior of fission yeast . Reducing the growth rate α such that the time of volume-doubling was slowed from 100 to 120 min mimicked the effect of latrunculin-A on a wild-type cell culture . This drug disrupts actin patches and delays entry into mitosis because it increases the time required to reach the size threshold [38] . As required for size checkpoint behavior , mitotic entry in our simulations was delayed at the reduced growth rate but it was triggered at exactly the same volume ( Figure 8A ) . Next we examined the effect of reducing the microtubule concentration; this mimics the effect of methyl benzimidazol-2-yl carbamate ( MBC ) , a microtubule-depolymerizing drug that delays the entry of WT cells into mitosis [39] . Our simulations showed a similar delay ( Figure 8B ) . They suggest that the G2 arrest caused by microtubule depolymerization is , at least in part , a consequence of Pom1 mislocalization rather than sensing microtubule damage , as has been proposed [39] . Cells lacking microtubule interphase bundles are unable to transport Tea1 to the cell tips and consequently fail to retain Pom1 to this region . As a result , the higher Pom1 level at midcell inhibits Cdr2 more effectively , which prevents Wee1 phosphorylation and the cell remains in G2 . Unlike experiments where most of the microtubules were disrupted [35] , our simulations used a reduced MT concentration , which should have a similar effect . In our model , total disruption of microtubules would result in permanent Cdc2 inactivation ( in contrast to the observed delay in Cdc2 activation ) . This difference in behavior probably arises because fission yeast contain other regulators of Cdc2 activation [18] . Finally , we increased the Pom1 concentration in simulations 2-fold , representing Pom1 overexpression mutants which exhibit a dose-dependent cell cycle delay ( 3-4 ) . Mitotic entry was again delayed ( Figure 8C ) . Further increases in the Pom1 concentration delayed the triggering time further .
An earlier version of the model did not include the Pm-catalyzed multiphosphylation of Cm; rather , Pm was modeled to catalyze the expulsion of unphosphorylated Cm from the membrane in a single step . This mechanism was unable to restrict Cdr2 to the midcell region such that the Cdr2 peak was broadened relative to the data ( Figure 9A , dashed green line ) . Including the multisite phosphorylation reactions sharpened this peak ( solid green line ) , with 10 such reactions required to match the data . Each additional assumed phosphorylation reaction sharpened the peak incrementally . We suspect that there might be other factors controlling the Cdr2 spatial linewidth in yeast cells , and do not regard the absolute number of phosphorylation sites required here as being quantitatively accurate . This chain of reactions sets a threshold ratio of kinase/phosphatase below which the fully phosphorylated form of Cdr2 is almost absent . In our case , it sets the ratio of as a function of the spatial Pm concentration . Reactions rates were set such that the Cm forms dominated at Pm concentrations observed at midcell ( Figure 9B , solid purple line ) . As the Pom1 concentration increased towards the cell tips , Cc became the dominant form in these regions . Cdr2 mostly resides on the membrane at the middle cell region because it is constantly ejected from the membrane at the tip regions by Pom1 . Although a single-step reaction can achieve similar Cdr2 ratios at the cell tips , it cannot afford a sharp Cdr2 midcell peak . When different parameter values were used with the single-step reaction to afford midcell ratios similar to those of the multisite mechanism , Cdr2 was not expelled efficiently from the membrane at the cell tips , again yielding a broad Cdr2 peak . It is clear that Pom1 phosphorylates Cdr2 in vitro [4] , [5] but the number of phosphorylation events involved is uncertain . Ten phosphorylation events were required to sufficiently sharpen the Cdr2 peak at midcell , but other processes may contribute to the sharpness of the Cdr2 gradient in real cells such that the actual number of phosphorylation reactions may be fewer than this . An additional unidentified Cdr2 inhibitor may be involved in Cdr2 localization , the effect of which is observed in the phenotype of Pom1 mutants [4] , [5] . Sterol membrane domains may also be involved in Cdr2 cellular localization [35] . In our model , Cdr2 is active as a kinase only when membrane-bound , which we interpret as being when it resides in cortical nodes . Our model also assumes that Pom1 inhibits Cdr2 , not by inhibiting its kinase activity , but by detaching it from these nodes ( Figure 10 ) . This mode of activation/deactivation has some experimental support . First , Cdr2 is essential in forming cortical nodes [4] . Wee1 and Cdr1 ( a direct inhibitor of Wee1 ) localize to these nodes only in the presence of Cdr2 [4] . Cdr1 might only efficiently phosphorylate Wee1 once both proteins are in the nodes , as their local concentrations would be far greater than when they are in the cytosol [42] . Rate enhancement could be as high as the ratio of the cytosol volume to the cortical node volume [42] , assuming first-order dependences . Importantly , Pom1 phosphorylates the non-catalytic terminus of Cdr2 which is responsible for attaching Cdr2 to the membrane [4] , [5] , [35]; this suggests that Pom1 is not affecting the kinase activity of Cdr2 . Also , cortical nodes are disrupted and entry into mitosis is delayed when a mutant Pom1 binds to the cortex at the midcell region [4] , [5] . Cells with defective Cdr2 membrane localization but intact kinase activity respond differently to nutrient starvation relative to wild-type cells [35] , suggesting that Cdr2 localization is closely related to its function . Cdr2 midcell fluorescence increases modestly by the end of G2 phase [5] , but this has not been considered previously as being associated with the mechanism that links cell size to cell cycle events . If there was no increase of the membrane-bound form of Cdr2 at midcell , the Cdc2 triggering mechanism used here could not function appropriately . Whether the mitotic trigger would be sufficiently robust with only the modest observed increase of Cdr2 fluorescence at the midcell region [5] is uncertain . If a switch-like mechanism is responsive to such small concentration changes , one would expect a large variability of cell lengths due to genetic noise [43] , which is not observed . However , other mechanisms may enhance stability and buffer the system against such noise . Positive and double-negative feedback loops can add bistability robustness to the cell size checkpoint [44] , and there are other unidentified components involved in cell-size sensing [4] , [5] , [45] . In any event , our reaction-diffusion-convection system provides a reliable framework for Pom1 and Cdr2 spatial localization where different hypotheses for the link between cell size and cell cycle can be explored . Finally , we have considered whether the Pom1-dependent cell size checkpoint mechanism could be more generally used in eukaryotic cells . Pom1 is a member of the DYRK ( dual-specificity tyrosine-regulated kinase ) family . These proteins are involved in cell cycle regulation and control of cell proliferation and differentiation [46] . Although it is unclear that other Pom1 homologs are used in cell-size checkpoint mechanisms , this possibility is intriguing . Such mechanisms would likely involve size-dependent shifts in protein spatial gradients . Cell size and shape are reportedly involved in controlling the phosphorylation states of cellular components [47] which may be involved in size sensing . Efforts should be made to identify new size-related proteins that are connected to the cell cycle machinery and that exhibit spatial concentration gradients . Such proteins may play key roles in cell-size checkpoint mechanisms . | Cells delay division into two daughter cells until they reach a particular size . However , the molecular-level mechanisms by which they do this have remained unknown until recently . A cell-size checkpoint mechanism in rod-shaped fission yeast cells has recently been shown to involve two proteins , Pom1 and Cdr2 . The concentrations of these proteins in the middle of the cell differ from that at the poles . The changing nature of these spatial gradients as the cell grows is size-sensitive . Pom1 inhibits Cdr2 while Cdr2 stimulates the cell to enter into mitosis . In short cells , the Pom1 concentration in the middle of the cell is so great that Cdr2 is inhibited . As cells grow , the Pom1 concentration in the middle of the cell declines; at some particular size , Cdr2 activates . In this study , we developed a mathematical model that mimics this checkpoint behavior . | [
"Abstract",
"Introduction",
"Model",
"Results",
"Discussion"
] | [
"mathematics",
"computational",
"biology/metabolic",
"networks",
"computational",
"biology/systems",
"biology"
] | 2010 | Mathematical Model of a Cell Size Checkpoint |
Over the past few decades , land-use and climate change have led to substantial range contractions and species extinctions . Even more dramatic changes to global land cover are projected for this century . We used the Millennium Ecosystem Assessment scenarios to evaluate the exposure of all 8 , 750 land bird species to projected land-cover changes due to climate and land-use change . For this first baseline assessment , we assumed stationary geographic ranges that may overestimate actual losses in geographic range . Even under environmentally benign scenarios , at least 400 species are projected to suffer >50% range reductions by the year 2050 ( over 900 by the year 2100 ) . Although expected climate change effects at high latitudes are significant , species most at risk are predominantly narrow-ranged and endemic to the tropics , where projected range contractions are driven by anthropogenic land conversions . Most of these species are currently not recognized as imperiled . The causes , magnitude and geographic patterns of potential range loss vary across socioeconomic scenarios , but all scenarios ( even the most environmentally benign ones ) result in large declines of many species . Whereas climate change will severely affect biodiversity , in the near future , land-use change in tropical countries may lead to yet greater species loss . A vastly expanded reserve network in the tropics , coupled with more ambitious goals to reduce climate change , will be needed to minimize global extinctions .
Accelerated climate change and the destruction of natural habitats through direct human activities are two of the greatest threats to terrestrial biodiversity . In recent decades , they have led to substantial range contractions and species extinctions [1–5] . Even more dramatic environmental change is projected for this century [6–8] . Substantial evidence emphasizes the importance of human land-use changes as a cause of species declines and extinctions [5 , 6 , 9–12] . Recent studies have highlighted existing and future impacts of human-induced climate change on species persistence [2–4 , 13 , 14] and have stressed climate change as a primary concern for the setting of conservation priorities [15 , 16] . Most of these studies have been based on data collected in the temperate zone , where climate change is predicted to be more pronounced . To date , there have been no global forecasts of the relative and synergistic effects of future climate change and habitat loss on vertebrate distributions . Moreover , our conceptual understanding of what makes some regions and species vulnerable to one threat or the other is still limited . We integrated the exposure of species to climate and land-use change through the combined effects of these drivers on global land cover and explored the resulting reductions in range size and possible extinctions within the world's 8 , 750 terrestrial bird species . For this first global assessment , we used the simplifying yet transparent assumption of stationary geographic ranges , which allows us to quantify risk in terms of the projected vegetation changes across a species' current range . Although this assumption yields worst-case projections and a number of factors could modify the local details and timeline of our projections , we think the general picture that emerges is robust: a clear and striking geographic disjunction between the relative impacts of future habitat loss and climate change on global avian diversity . We used the Millennium Ecosystem Assessment ( MA ) global scenarios to provide examples of possible environmental futures [8 , 17 , 18] . The four MA scenarios use plausible ranges of future greenhouse gas emissions and human population and economic growth to estimate how much of a region will be affected by anthropogenic climate change and agricultural expansion . They are characterized by their different approaches to development and ecosystem management . With respect to development , two scenarios ( Global Orchestration and TechnoGarden ) assume the world becomes increasingly globalized; the other two ( Order from Strength and Adapting Mosaic ) assume it becomes increasingly regionalized . With respect to ecosystem management , two scenarios ( Global Orchestration and Order from Strength ) are reactive; they assume that environmental problems causing the breakdown of ecosystem processes are addressed only after they occur . The other two ( TechnoGarden and Adapting Mosaic ) assume such problems are managed more proactively . The modeling framework that is part of the MA integrates the interacting effects of future climate and land-use changes and forecasts expected changes to the geographic occurrence of 18 natural and human-made land-cover types [17 , 19 , 20] . These expected changes are separated into those due to climate change ( change from one natural to another natural land-cover category ) and those due to direct human land-use change ( change from natural to human-caused land-cover type ) . We overlaid the geographic occurrence of these land-cover changes with bird distribution data to estimate the areas transformed to a different habitat and thus presumably lost across each species' global range . In the absence of extensive worldwide surveys , we used refined species extent-of-occurrence maps that minimize range overestimation . We recognize that range maps are a scale-dependent abstraction of species' actual occurrence [21] that limit interpretation at fine geographic scales . However , assuming there are no dramatic geographic or ecological trends in range overestimation , this approach yields reliable and urgently needed insights into the impact and interplay of the two major threats to biodiversity at the global scale .
Range loss varies dramatically across species in all four MA scenarios ( Figures 1 and S1 ) . The mean expected range contraction across all scenarios by the year 2050 is 21–26% ( depending on scenario ) and rises to 29–35% by 2100 ( arithmetic means ) . In the less environmentally conscious scenarios , ~400–900 bird species are projected to have over 50% of their current range transformed to a different habitat by 2050 ( Tables 1 and S1 ) ; this number roughly doubles by 2100 . The species that show minimal loss of range are wide-ranging species , confirming that large ranges provide a buffer against environmental change ( Figures 1 , S1 , and S2 ) . In contrast , the largest potential loss of range size occurs among species that have restricted ranges ( Figure S2 ) ; this fact highlights the double jeopardy for species that already have small population sizes , specialized habitat requirements , and that are exposed to a high risk of extinction from stochastic demographic processes [22] . Small population or range size and rapid loss of habitat are among the characteristics that formally characterize Red List species under grave threat of extinction [23 , 24] . Under all four MA scenarios , roughly 170–260 species are projected to experience substantial ( i . e . , a greater than 50% ) range declines that lead to range sizes of less than 20 , 000 km2 by 2050 ( an additional 83–195 species are projected to experience this by 2100 ) . Should this occur , then under the “restricted distribution” criterion of The World Conservation Union ( IUCN ) ( Criterion B ) , they would likely be classified at least as “vulnerable” in the future ( and as “near threatened” now ) , due to their small range sizes combined with continued decline and range-wide threat ( see [25] for further discussion ) . Fewer than half of the species identified in this way are currently listed by IUCN . Under these criteria the total number of threatened species in the analysis would increase by 19–30% by 2050 and 29–52% by 2100 . Moreover , of the 886 species in this analysis that are already listed as threatened , 418–475 are expected to have further range losses of at least 20% by 2050 under all scenarios . The risk of extinction to these species is thus likely to grow significantly . We initially use the “Adapting Mosaic” scenario to illustrate the geography of environmental change [18] . This relatively optimistic scenario represents a world that deals proactively with environmental issues; nonetheless , between now and 2100 it projects that approximately 25% of areas currently classified as natural will be transformed—16% due to climate change and 9% due to land-use conversions . Although changes to the land cover are projected to occur globally , there are pronounced regional and latitudinal patterns ( Figures 2A and 3A ) . Changes driven by climate change are strongest in the high latitudes ( >30° ) of Siberia and North America , a reflection of the greater temperature increases projected for these regions [26] . In contrast , human land-use change dominates at lower latitudes , specifically in Central and South America , central Africa , and portions of India and China; this reflects the importance of forecasted high levels of economic and population growth in these regions . The alternate “Order from Strength” scenario represents a world with only reactive management of environmental issues and projects a transformation of 28% of land , half of which is due to human land-use change; much larger parts of regions such as central Africa are converted to agriculture ( Figure 2B ) . This scenario would result in direct habitat loss in the tropics and subtropics that is approximately double that of the “Adapting Mosaic” scenario ( compare Figures 3A and 3B ) . In both scenarios , range reductions in species that suffer small-to-intermediate proportional range losses are driven more or less equally by climate and land-use change ( Figure 1 and Table 1 ) . However , the situation is very different for species that are projected to suffer extensive range losses; these are largely caused by direct human land-use change . Each projected outcome reflects the covariance between the spatial distribution of the different impacts and the biogeography of bird distributions . Climate change–induced land-cover changes are consistently projected to have the greatest potential impact on species that live far from the equator , in particular in the large northern landmasses , where individual species tend to exhibit very broad distributions and communities as a whole are low in richness ( Figures 3 and S3 ) . Bird species that live between 0° and 20° latitude have less than half the geographic range size of birds occurring between 40° and 60° latitude ( 5 . 4 × 106 km2 versus 11 . 1 × 106 km2 , arithmetic means ) . Average range sizes across bands of absolute latitude increase steadily toward the poles ( Spearman rank correlation: rs = 0 . 96 , p < 0 . 001 , n = 75; all 8 , 750 species ) , a pattern that is predominantly driven by species in the northern hemisphere ( Figure 3A and 3B ) . The higher latitudes are much poorer in species: only 1 , 186 species occur above 40° North or below 40° South , but 7 , 485 ( ~86% of total ) are located between 20° North and South; species counts per 1° band consistently decrease with increasing absolute latitude ( rs = 0 . 98 , p < 0 . 001 , n = 75 ) . It follows that the high proportional range loss caused by climate-driven land-cover changes in the high latitudes affects a smaller number of species ( Figure 3C and 3D ) . Conversely , even under the environmentally more benign “Adapting Mosaic” scenario , range loss due to land-use change in tropical and subtropical regions will have potentially devastating consequences for the many , more narrowly distributed species found there ( Figure 3A and 3C ) . Land-use change is responsible for more than half of the range contractions in this scenario , and by itself causes twice as many species to lose over half their range ( Figure 3E and Table 1 ) . In the case of the “Order from Strength” scenario , land-use changes below 20° latitude are almost double the magnitude observed in the “Adapting Mosaic” scenario and coincide with a dip in range size and peak in species richness; this concomitance of tropical land-use change and many species with small ranges predicts the dramatic numbers of species potentially experiencing ≥50% range loss ( Figure 3B , 3D , and 3F ) . The projected impact of environmental change differs markedly across the four socioeconomic scenarios ( Table 1 , Figures 1 and S1 ) . These differences are mostly driven by the variation in the magnitude of land-use change between scenarios , which is greater than the variation in projected climate change ( Table 1 ) . Projected exposure roughly corresponds to the economic and ethical values attributed to biodiversity and ecosystem services in each scenario . Scenarios that represent reactive environmental management lead to greater range losses , largely due to direct human land transformation . Conversely , scenarios that focus on environmental protection or technological solutions to environmental problems result in fewer species suffering major range contractions; of these , between one quarter and a half are primarily affected by climate change driven land-cover changes , depending on whether average area loss or an exclusive 50% range loss threshold is used as an indicator ( Table 1 ) . Consistently across all scenarios , the regions with the highest number of species suffering dramatic range contractions are Central America , southeastern Brazil , eastern Madagascar , and the Himalayan highlands ( Figure 2C and 2D ) . All of these have been identified as key biodiversity hotspots in analyses using a diversity of taxa and contemporary rates of habitat loss [27] . Our projections suggest that the Andes and central Africa will also deserve increased attention from conservationists due to high projected levels of habitat loss .
The evaluation we have made of species' exposure to climate change is based on changes in land cover and relies on the well established dependence of land cover on climatic conditions . Our evaluation is transparent and avoids many of the potential conceptual and methodological pitfalls inherent in more complex approaches . But the approach presented here makes some important assumptions: we assume that birds exhibit persistent habitat associations and are limited in their dispersal . In some cases , habitat specialists may be affected by habitat changes finer than those registered by the available land-cover categorization , which would cause us to underestimate climate change impacts for these species . Conversely , range shifts may alleviate the projected impact of climate change [2 , 28 , 29] ( and thus increase the relative importance of other threats ) . This would cause us to overestimate the impact of climate change except for high-altitude species , which face limited area available for dispersal [30] . Similar responses might also mitigate the impact of human land-use change ( but given the geographic separation of impacts , this effect is likely to be small ) . Unfortunately , identifying more highly resolved climatic niche boundaries and estimating range shifts from spatial data are inherently difficult problems . Similarly , more detailed modeling of extinction risk would require us to make critical assumptions about ecological interactions between species , crucial niche components , and changes in potential habitat barriers . Even though notable progress has been made , there is still a lack of general consensus on which of the available modeling approaches provides the best insights given the data limitations for most tropical species [31 , 32] . Furthermore , more detailed models of interactions between climate and land-use change should ideally consider other threats such as infectious diseases , species invasions , and increased persecution which are likely to additionally impact the loss of populations . On a more optimistic note , species currently recognized as specialists may adapt to new habitats including those created under some forms of human encroachment [33 , 34] . Habitats such as regrowth forests , which our analysis counted as lost to primary forest species , may in fact be able to support at least some of the original species pool [35] . Similarly , while refined range maps were used in this analysis , not all parts of the current and projected range will be fully occupied; this will inevitably result in an underestimation of the impact of environmental change for a significant proportion of species , particularly those with specialized niches and heterogeneous distributions across their current geographic range . We acknowledge that further understanding and modeling of these issues is crucially important for accurate predictions at a fine scale . There clearly is need for further broad-scale studies that develop individual species models while exploring the sensitivity of results to assumptions and methods . Complementary progress will come from detailed studies limited to focal regions and few taxa that carefully estimate as many factors as possible potentially driving range shifts , contractions , and adaptations . However , given the detailed information required for such analyses , these studies are unlikely to provide timely advice to decision-makers who must grapple with the issues of climate change and anthropogenic habitat loss now . Further broad-scale work is needed to explicitly model loss of habitat along elevational gradients , assess threats to long-distance migrants [36] , and to additionally take into account the global reserve network that may successfully protect against land-use but not climate change impacts . This study is the first one to our knowledge to investigate exposure to climate change for a full , species-rich clade across the whole world and the first one to concomitantly evaluate the effects of direct land-use change . Our results show notable differences to previous studies [e . g . 15]; this may be due to their assessment of only climate change , methodological differences , and the restriction of the majority of these studies to mostly temperate species ( which are projected to experience highest temperature changes ) . Our results suggest that the impact of climate-change induced land-cover changes on range sizes in birds will likely be considerable . However , habitat loss in economically emerging tropical countries will continue to pose an even more direct and immediate threat to a greater number of bird species . Although the geography , magnitude , and type of impact will depend critically on the socioeconomic pathways different nations choose to follow , even the most optimistic scenarios lead to substantial range contractions of species , especially of those already vulnerable to extinction because of their current restricted ranges . Only by rapidly expanding the network of protected areas in the tropics can we hope to prevent hundreds of species from becoming imperiled or even extinct . The scenarios that proactively acknowledge that the natural environment provides crucial services to the human economy seem likely to conserve both a higher quality of life for the human population and a higher diversity of species .
We evaluated the effect of projected land-cover changes on the breeding distributions of 8 , 750 species of land birds ( out of 9 , 713 total ) , excluding water birds and endemics of small oceanic islands that were to small to be included in the MA projections ( see Tables S2 and S3 for lists of included and excluded species , respectively ) . The classification of species follows Sibley & Ahlquist [37] for nonpasserines and Barker et al . [38] for passerines and was updated for newly described species and recent splits and lumps . Distributions were compiled from the most accurate sources giving expert opinion range ( extent of occurrence ) maps for a given broad geographic region or taxonomic group ( see Figure S4 and Table S2 for details ) . Essentially the same sources were used by [39] . Originally in polygon format , the maps were re-sampled to 0 . 01° resolution in geographic projection for further analysis . Extent of occurrence maps are the only type of distributional information available at global scales . By definition , they overestimate the area of occupancy [40] potentially resulting in a dramatic underestimation of proportional losses in geographic range . To address this issue , the extent of occurrence maps for this analysis were refined by clipping from the rasterized species' range maps those habitat types that are definitely unsuitable for the species . We subjected range maps ( 0 . 01° resolution , geographic projection ) to two clipping steps with finer resolution environmental datasets ( in geographic projection ) . We set the analysis resolution to that of the range maps: whenever the environmental data layers indicated the majority of a 0 . 01° range map grid cell as unsuitable , it was deleted from a species' range . In the first step , we clipped off elevations outside the maximum or minimum observed for the species ( data available for 4 , 726 species [41] ) using the GTOPO30 digital elevation model at 0 . 0083° resolution [42] . For the second step we first compiled potential habitats listed for species from the literature ( [41]; 3 , 472 different habitat descriptions available across all 8 , 750 species ) . We then linked the recorded habitats to one or more of the most representative 22 habitat categories used in the Global Land Cover 2000 database land-cover classification [43] . This resulted in a list of potential land-cover categories occupied by each species . Finally , each species' range map was overlaid with the GLC2000 land-cover map ( in geographic projection at 0 . 0089° resolution , i . e . , ~1 km2 at the equator ) , and all land-cover categories not listed for a species were clipped from the range . Together , these steps caused a mean reduction of 21 . 5% ( standard error 3 . 5% ) of range area compared to the original extent of occurrence maps , while incurring only minimal false absences ( generally below 1% , [Jetz , unpublished data] ) . Qualitatively similar results were gained with unrefined extent of occurrence maps , but they reveal smaller proportional range losses and thus tend to underestimate counts of potentially threatened species ( Table S1 ) . To calculate a species' geographic range size , we first projected a map of the world in geographic projection and 0 . 01° resolution to equal area projection and calculated the true area ( in km2 ) of each 0 . 01° grid cell . Geographic range size was then given by the summed area of all 0 . 01° grid cells occupied by a species . The MA land-cover projections are only available at a coarser resolution ( 0 . 5° , geographic projection ) . Therefore , we separately recorded the summed area ( in km2 ) of all 0 . 01° grid cells occupied by a species in each MA 0 . 5° grid cell . This data then formed the basis for the calculation of proportional range transformations outlined below . All overlays and map calculations were performed using the ESRI Arc and Grid software ( V . 9 . 0; ESRI 2004 , http://www . esri . com ) . The MA developed four scenarios that could be used to examine and compare changes in land use and global climate under a variety of deliberately diverse and different social and political futures . They were developed to compare four possible extreme conditions in the year 2050 and also provide extrapolations to 2100 [8 , 18 , see 44 for discussion of the importance of region and scale] . The scenarios are not predictions; their principal utility is to delineate the range of possible futures . The four scenarios can be briefly described as follows [8 , 18]: ( i ) Adapting Mosaic . In this scenario , regional political responses and economic activity are focused within each major watershed . Local institutions are strengthened and local ecosystems managed proactively . Economic growth is initially low but increases with time . Human population levels approach those estimated for the scenario with the highest rate of human population growth . ( ii ) Order from Strength . This scenario represents a regionalized and fragmented world that is concerned with security and protection; it pays little attention to public goods and takes a reactive approach to environmental problems . It has the lowest economic growth rates of the four scenarios ( they even decrease with time ) , but these are combined with the highest human population growth rates . ( iii ) TechnoGarden . This scenario depicts a globally connected world that relies strongly on environmentally sound technology . Ecosystems are increasingly dependent upon technological fixes . Economic growth is relatively high and accelerates , while human population settles into the midrange of projections . ( iv ) Global Orchestration . Under this scenario , a globally connected society focuses on global trade and economic liberalization but takes a reactive approach to ecosystem problems . It also takes some strong steps to reduce poverty and inequality by investment in infrastructure and education . This scenario has the highest rate of global economic growth and the lowest human population size by 2050 . The MA scenario evaluations are based on the IMAGE 2 . 2 model [19] , a dynamic Earth-system model that estimates future changes to Earth's land-cover in terms of chains of driving forces , pressures , state , and response variables , covering both the natural environment and the socioeconomic system . The model explicitly integrates the forecasts of direct human encroachment with projections of climate change effects on vegetation physiognomy , based on the BIOME model [45] , and considers resulting interactions ( for detailed assumptions on vegetation shifts and adaptation speed see [19] ) . We note that these sort of integrated models will likely experience substantial improvement over the coming decade , and projected hotspots of change may well shift geographically as the field ( and knowledge about regional drivers ) progresses . The model provides information on current and future distributions of 18 different land-cover types at 0 . 5° resolution ( 66 , 661 terrestrial grid cells ) , three of which indicate direct human impact from agriculture or urbanization , cropland , permanent pasture , regrowth forest [18 , 20] . We evaluated changes in land cover between 1985 ( the approximate median of time period over which range maps were compiled in the sources used ) and 2050 and 2100 , respectively . Changes from one of the 15 natural to one of the three human-caused land-cover types were considered as transformations due to land-use change for all but the 822 bird species tolerant of at least minor human encroachment ( following the habitat preference analysis presented above ) . Changes from one natural to another natural land-cover type without direct human impact were considered as transformation due to climate change . In some cases ( <10% of total grid cell transformations ) , the new habitat was among those registered suitable for a species . Given the significant perturbation that most habitat transformations create , we counted this grid cell as lost for the species . Although this approach technically overestimates the potential impact of climate change , qualitatively it does not affect the results , because almost all of these habitat generalist species are widespread and therefore have small proportional range loss . During the 21st century , some areas are forecast to be exposed to climate change before land-use change and vice versa . Therefore , if a grid cell already experienced land-cover change by 2050 , the respective type of change was the one also registered for 2100 . The projected land-cover changes were overlaid with species breeding distributions , and for every transformed 0 . 5° grid cell , the occupied range area registered for this cell and species was subtracted from the original range size . The remaining untransformed range area was calculated to 1-km2 resolution and compared to the original range size . Our approach allows a global and integrated perspective of the effects of land-cover changes driven by climate change and the effects of land-use change on biodiversity . It has the advantage of being transparent and useful for a large number of species . Additionally , it is not fraught by the data limitations and possible methodological pitfalls that are associated with the attempt to quantify species' exact environmental niche and potential for adaptation . Some recent predictions of climate-related extinctions have relied on developing climatic correlates of current species distributions and evaluating potential shifts in these “bioclimatic envelopes” to estimate range loss [15 , 16] . Our method assumes core habitat associations and evaluates the proportion of a species' range that will be transformed to unsuitable habitat by climate change . It uses well substantiated characterizations of species' persistent habitat requirements . Species range shifts in response to changing climate will likely lower the estimated proportional range loss attributed to it . The alleviating effect regarding the impact of land-use change is more equivocal and likely smaller , given that the direction and magnitude of shifts are uncertain and land-use change is mostly subtropical and tropical ( where on the whole , effects of climate change are weaker ) . Environmental change occurring below the spatial resolution of this analysis ( majority area of a 0 . 5° grid cell , i . e . , ~1 , 540 km2 near the equator , ~990 km2 at 50° latitude ) may lead to further habitat and range loss and thus to threats not evaluated here . Conversely , not all parts of a grid cell may experience the land-cover change projected for its majority , potentially overestimating losses . Nonuniform distribution of abundances within species ranges mean that range contraction does not straightforwardly translate into population loss [46] . Finally , more subtle changes in habitat type too fine for the categorization offered in the MA scenarios may lead to additional range losses , while some versatile species may be unaffected by changes of land-cover types . We acknowledge that these issues require further detailed analyses at regional and single species levels . However , there is no reason to assume that they would introduce a systematic bias into our analyses . | Land conversion and climate change have already had significant impacts on biodiversity and associated ecosystem services . Using future land-cover projections from the recently completed Millennium Ecosystem Assessment , we found that 950–1 , 800 of the world's 8 , 750 species of land birds could be imperiled by climate change and land conversion by the year 2100 . These projections are based on the assumption that birds will not dramatically shift their ranges in response to a changing climate , a process that would lessen the range contractions we predict . While climate change will be the principal driver of range contractions at higher latitudes , our projections reveal that land conversion ( e . g . , deforestation , conversion of grasslands to croplands , etc . ) will have a much larger effect on species that inhabit the tropics . This is because birds in the tropics are especially diverse and tend to have small ranges , making them particularly vulnerable to extinction; in contrast , birds at higher latitudes are less diverse and tend to have large ranges . A vastly expanded reserve network in the tropics , coupled with more ambitious goals to reduce greenhouse gas emissions and monitor biodiversity impacts , will be needed to minimize global extinctions . | [
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] | [
"birds",
"ecology"
] | 2007 | Projected Impacts of Climate and Land-Use Change on the Global Diversity of Birds |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.