Search is not available for this dataset
article
stringlengths 4.36k
149k
| summary
stringlengths 32
3.35k
| section_headings
listlengths 1
91
| keywords
listlengths 0
141
| year
stringclasses 13
values | title
stringlengths 20
281
|
|---|---|---|---|---|---|
Hepadnaviridae are double-stranded DNA viruses that infect some species of birds and mammals . This includes humans , where hepatitis B viruses ( HBVs ) are prevalent pathogens in considerable parts of the global population . Recently , endogenized sequences of HBVs ( eHBVs ) have been discovered in bird genomes where they constitute direct evidence for the coexistence of these viruses and their hosts from the late Mesozoic until present . Nevertheless , virtually nothing is known about the ancient host range of this virus family in other animals . Here we report the first eHBVs from crocodilian , snake , and turtle genomes , including a turtle eHBV that endogenized >207 million years ago . This genomic “fossil” is >125 million years older than the oldest avian eHBV and provides the first direct evidence that Hepadnaviridae already existed during the Early Mesozoic . This implies that the Mesozoic fossil record of HBV infection spans three of the five major groups of land vertebrates , namely birds , crocodilians , and turtles . We show that the deep phylogenetic relationships of HBVs are largely congruent with the deep phylogeny of their amniote hosts , which suggests an ancient amniote–HBV coexistence and codivergence , at least since the Early Mesozoic . Notably , the organization of overlapping genes as well as the structure of elements involved in viral replication has remained highly conserved among HBVs along that time span , except for the presence of the X gene . We provide multiple lines of evidence that the tumor-promoting X protein of mammalian HBVs lacks a homolog in all other hepadnaviruses and propose a novel scenario for the emergence of X via segmental duplication and overprinting of pre-existing reading frames in the ancestor of mammalian HBVs . Our study reveals an unforeseen host range of prehistoric HBVs and provides novel insights into the genome evolution of hepadnaviruses throughout their long-lasting association with amniote hosts .
Viruses and their hosts share a rich coevolutionary past that is evidenced by a plethora of viral relics buried within host genomes . A striking example for this is the human genome where genomic relics of ancient , endogenized viruses constitute ∼8% of its total sequence [1] . These “fossils” of viruses have been collectively termed endogenous viral elements ( EVEs ) [2] and originate from host germline integration , followed by vertical transmission and subsequent fixation of virus-derived DNA in the genome of the host population [3] , [4] , [5] . The recent and ongoing availability of numerous genome sequences from non-model organisms [6] , [7] has given rise to the field of paleovirology [8] , the study of anciently integrated viruses , and has yielded the first direct insights into the long-term evolution of certain virus families [2] , [9] , [10] . The vast majority of EVE copies belongs to the Retroviridae family [1] , [5] of viruses which rely on reverse transcription and obligate host genome integration , however , paleovirology has unearthed genomic fossils of all other major groups of eukaryotic viruses [2] , [5] , [11] . Whenever an EVE is present at a unique genomic location , it is possible to date the upper and lower age boundary of viral endogenization events by comparison of orthologous EVE insertions among different host species [5] , providing direct evidence for host-virus coexistence . The Hepadnaviridae are a family of reverse-transcribing dsDNA viruses infecting various species of birds [12] and mammals , including bats [13] , rodents [14] , and primates [15] . In humans , hepatitis B virus ( HBV ) poses one of the most widespread global health problems that affects more than 2 billion people and leads to >500 , 000 deaths per year [16] . Despite the availability of a number of primate genome sequences [7] , HBV EVEs are absent or undetectable in these and other mammalian genomes [10] . In contrast , many bird genomes contain HBV fossils , such as the zebra finch and other songbirds [2] , [10] , [17] , the budgerigar [18] , [19] , and other representatives of Neoaves [10] . Direct evidence from paleovirology suggests a coexistence of birds and HBVs that spans ∼70 million years ( MY ) of the Mesozoic and Cenozoic Eras , with HBV endogenizations dating from >82 million years ago ( MYA ) to <12 . 1 MYA [10] . Based on this fossil record , Hepadnavirus evolution might have either been characterized by an ancient coexistence with amniotes [10] or by a more recent bird-mammal host switch [10] , the latter being in line with the paucity of extant host species and lack of mammalian HBV EVEs . The validity of either hypothesis is largely dependent on the genomic fossil record of HBVs [10] . The same is also the case for the enigmatic origin of the X gene of mammalian HBVs , as X appears to be absent in ancient avian HBV EVEs [10] , while some extant avian HBVs exhibit an X-like gene [20] . The X gene is known to be involved in the generation of liver tumors in chronic HBV infection in humans and woodchucks [21] , [22] , [23] , [24] , [25] , [26] , so the elucidation of the evolutionary emergence of the X gene is of broad relevance to biological and medical research on hepatitis B viruses . Here we report endogenous hepadnaviruses from recently sequenced turtle [27] , [28] , snake [29] , and crocodilian genomes [30] , [31] . Among these EVEs is a near-complete crocodilian HBV genome from the Late Mesozoic and an Early Mesozoic turtle HBV , providing us with the unprecedented opportunity to study the host range , genome evolution and deep phylogeny of Hepadnaviridae . We show that genome organization and replication is highly conserved among HBVs with the exception of the presence of the oncogenic X gene , for which we infer an evolutionary scenario of de-novo emergence in the ancestor of mammalian HBVs . Finally , our hepadnaviral fossil record reveals Mesozoic coexistence of Hepadnaviridae with three of their five major host taxa and supports a scenario of ancient amniote–HBV cospeciation .
We searched the recent saltwater crocodile , gharial , and American alligator draft genome assemblies [30] , [32] using whole viral genomes of the duck HBV ( DHBV; AY494851 ) and the Mesozoic avian eHBV ( eZHBV_C [10] ) , and identified two endogenous crocodilian HBVs ( eCRHBVs; Fig . 1A ) . Likewise , we screened the genomes of turtles ( painted turtle , softshell turtles , and sea turtle [27] , [28] ) , squamate lepidosaurs ( cobra , boa , python , and anole lizard [29] , [33] , [34] , [35] ) , and mammals ( human , opossum , and platypus [36] , [37] , [38] ) for the presence of eHBVs . We detected a single locus in turtle genomes , hereafter referred to as endogenous turtle HBV ( eTHBV; Fig . 1A ) , two endogenous snake HBVs in the cobra genome ( eSNHBVs; Fig . 1A ) , but no EVEs in the remaining squamate and mammalian genomes . Our presence/absence analyses show that all four available cryptodiran turtle genomes plus the sampled pleurodiran ( side-necked ) turtles ( Mesoclemmys and Podocnemis ) exhibit the eTHBV insertion , while it is absent in the orthologous position in crocodilian genomes ( Fig . 1B ) . This suggests that it is of Triassic origin and was endogenized in the ancestor of Testudines that lived 207 . 0–230 . 7 MYA [39] , [40] . eCRHBV1 ( Fig . 1C ) is present in all crocodilians except alligators ( i . e . , Longirostres [41] ) and is 63 . 8–102 . 6 MY [42] old , i . e . , of Cretaceous origin . The second crocodilian EVE ( eCRHBV2 ) is exclusively shared between saltwater and dwarf crocodile; its endogenization thus occurred during the Paleogene in the ancestor of Crocodylidae ( 30 . 7–63 . 8 MYA [42] , ) . Unfortunately , the snake EVEs remain undated , as none of the cobra eSNHBV loci could be aligned to other squamate genomes for ascertainment of EVE presence/absence ( Fig . 1A ) . Given the dense fossil record of crocodilians and turtles that provides multiple calibrations for molecular dating of species divergences [39] , [42] , we suggest that the aforementioned dates are robust age estimates of eCRHBV1 , eCRHBV2 , and eTHBV endogenizations . Furthermore , molecular dating studies using mitochondrial genomes [44] , [45] or nuclear loci [42] , [43] yielded similar results on crocodilian divergence times , and the basal turtle divergence time of 207 MYA [39] , [46] ( i . e . , the Cryptodira–Pleurodira split ) is a nuclear estimate that is well compatible with mitochondrial estimates [47] , [48] and the fossil record [49] . Annotation assigns these five eHBV insertion sequences no extant protein-coding function in their hosts' genomes ( see GenomeBrowser [50]; the two crocodilian eHBVs are located within very large introns and the snake eHBV loci are undetectable in the lizard genome , while the turtle eHBV appears to constitute intergenic sequence ) . In line with this , we identified several frameshifting indels and premature stop codons in all five eHBVs ( S1 Table ) . Most of these were lineage-specific and found at different locations , indicating that they were not present in the common ancestor where the viral integration occurred . To determine whether any of the eHBV fragments may still show any sign of having been functional in the past before incurring stops and frameshifts , we performed likelihood ratio tests of the ratio of the rate of non-synonymous to the rate of synonymous substitutions ( ω ) either fixed to 1 or being allowed to vary freely . As none of these tests provided statistical support for deviation from ω = 1 ( S2 Table ) , there was thus no evidence for non-neutral evolution of these loci in the sampled genomes . Similar observations were previously made in selection tests on avian eHBVs where neutrality could not be rejected [17] , which may suggest that none of the currently known HBV EVEs possess an obvious protein-coding function in their host genomes . The crocodilian and turtle eHBVs' GC content is similar to the GC level of the adjacent flanking sequence of the host ( S1 Figure ) , which suggests that they have resided in the host genome for long enough to show a host-like base composition . Given that we detected no sign of non-neutral evolution of the crocodilian and turtle eHBV loci since their respective endogenization events , another line of evidence for the antiquity of their integration is the level of sequence divergence between orthologous eHBVs . We therefore calculated distances per eHBV locus ( see Materials and Methods ) and applied neutral substitution rates for crocodilians ( 3 . 9×10−10 substitutions/site/year [30] ) and turtles ( 8 . 43×10−10 substitutions/site/year for Pelodiscus sp . and 4 . 77×10−10 substitutions/site/year for Chelonia mydas [30] ) to determine locus-specific estimates of respective endogenization times . Consequently , we inferred integration events to have happened 70 . 3 MYA in eCRHBV1 , 20 . 5 MYA in eCRHBV2 , and 179 . 0 or 316 . 3 MYA in eTHBV . While these dates are compatible with our lower age boundaries of endogenization events derived from eHBV presence/absence patterns ( Fig . 1A ) , we suggest that the distance-based values are less robust estimates , as they rely on a limited number of nucleotides from a single genomic locus and are thus easily prone to biases caused by , for example , variation in substitution rates among lineages ( e . g . , Pelodiscus vs . Chelonia ) or among genomic regions . Extant avihepadnaviruses ( avian HBVs ) and orthohepadnaviruses ( mammalian HBVs ) have a circular genome organization with overlapping open reading frames ( ORFs ) and a streamlined genome size of about 3 . 0 kb and 3 . 2 kb , respectively [14] . The crocodilian , snake , and turtle eHBV fragments comprise up to 81% of an Avihepadnavirus genome ( Fig . 2A ) , permitting us to reconstruct large portions of their genome organization . We detected overlapping regions of the precore/core ( preC/C ) ORF with the polymerase ( pol ) ORF ( eCRHBVs and eTHBV; Fig . 2A ) and of the presurface/surface ( preS/S ) ORF with the pol ORF ( eCRHBVs and eSNHBV1; Fig . 2A ) , which suggests that all known extant and fossil avian , crocodilian , and mammalian HBVs exhibit a highly similar genome organization . This probably also applies to snake and turtle HBVs , because , while the eSNHBV1 and eTHBV fragments only span ∼14 and ∼21% of an HBV genome , they contain a region of overlapping ORFs ( Fig . 2A ) . Finally , we used approaches based on similarity searches and alignments , and did not detect any evidence for an X ORF in our non-avian eHBVs ( Fig . 2A ) . In addition to protein-coding sequences , we detected genomic features related to viral replication ( Fig . 2B–D ) , as the near-complete eCRHBV1 genome comprises the region where avihepadnaviruses and orthohepadnaviruses contain direct repeats ( DR ) and the RNA encapsidation signal ( ε ) . This region lies within the end of the pol ORF and the start of the preC/C ORF [51] ( Fig . 2A ) , but eCRHBV1 exhibited no significant nucleotide sequence similarity against DR+ε sequences of avian and mammalian HBVs . Yet , our structural analyses identified a DR motif of 14 nt that is present in identical copies within pol ( DR2 ) and preC/C ( DR1 ) . We further detected a 54-nt RNA hairpin motif with a priming bulge ( 5′–UUAC–3′ ) identical to the first four RNA nucleotides of the DR motif and reverse complementary to the ( – ) -DNA primer in avian HBVs [51] , suggesting that this is a structure that functionally corresponds to ε of extant HBVs ( Fig . 2B ) . In avian and mammalian HBVs , ε interacts with the ( – ) -DNA primer that is covalently linked to the conserved tyrosine residue of the terminal protein ( TP ) domain of the Pol protein [51] and establishes encapsidation of viral pregenomic RNA [52] , [53] as well as reverse transcription into viral ( – ) -DNA [53] , [54] . Despite the lack of sequence similarity between avian and mammalian HBV ε [51] , as well as the putative crocodilian HBV ε ( see S2 Figure ) , hepadnaviral replication appears to require strong structural constraint on ε with regards to stable base-pairing , as well as the presence of a bulge region and an apical loop ( Fig . 2B–D and refs . [19] , [51] , [55] ) . Only the 4-nt binding sites for the ( – ) -DNA primer within DR and ε exhibit sequence conservation among Hepadnaviridae ( Fig . 2B–D and S2 Figure ) . Recent paleovirological studies on avian eHBVs suggest that extant avihepadnaviruses and orthohepadnaviruses exhibit relatively shallow branches within phylogenetic trees compared to the deep divergences among eHBVs [5] , [10] ( see also S3C Figure ) . This suggests a recent divergence of circulating viruses among each of these two HBV lineages , whereas their endogenous avian counterparts appear to be relics of several distantly related , ancient lineages [10] , [18] with avihepadnaviruses being sister clade to one of them [10] . We reevaluated this by inferring the phylogenetic relationships based on Pol and PreC/C protein sequences of the non-avian eHBVs among Hepadnaviridae . In addition to full-length avian eHBVs and a dense sampling of extant HBVs , we included reverse-transcribing outgroups such as retroviruses , caulimoviruses , and retrotransposons . In phylogenetic trees of both Pol and PreC/C ( Fig . 3B–C , S4A–C Figure ) , the avian eHBVs form ancient , unrelated lineages , but with an eZHBV_C+avihepadnaviruses clade in the Pol tree and an eBHBVs+avihepadnaviruses clade in the PreC/C tree . This reversal in branching order could be explained by interspecific viral recombination events , as have been observed in some extant HBV lineages [56] , [57] , but is more likely due to the very limited amount of phylogenetically informative characters in the short PreC/C protein . Irrespective of the branching order of avian eHBVs , the two crocodilian eHBVs ( eCRHBV1 and eCRHBV2 ) consistently group together as a third major hepadnaviral lineage , and form the sister group of all avian HBVs and eHBVs , which is supported with high bootstrap values in the Pol tree ( Fig . 3B ) . This grouping , of course , is largely dependent on the position of the root of the Hepadnaviridae phylogeny . Our dense ingroup and outgroup sampling yields a Pol tree topology that strongly suggests Orthohepadnaviridae as the first branch among HBVs with respect to the remaining hepadnaviral lineages . Thus , in relation to avian and mammalian HBVs , the phylogenetic position of crocodilian HBVs reflects the host phylogenetic relationships between birds , crocodilians , and mammals [27] , [30] , [40] ( Fig . 3A ) . Unfortunately , it is not possible to include eTHBV in this well-resolved Pol tree , as the turtle EVE spans only a small part ( 16 aa ) of the Pol sequence . Consequently , the phylogenetic affinities of eTHBV are solely inferred from the PreC/C tree , which exhibits a lack of bootstrap support on its backbone , presumably as a consequence of too few phylogenetically informative characters within the PreC/C protein ( 342 aligned aa sites ) . However , the PreC/C tree does recover eTHBV as sister lineage of crocodilian+bird HBVs , which supports the above-mentioned similarity of the deep phylogenetic relationships among HBVs , as well as among their amniote hosts [27] , [30] , [40] ( Fig . 3A ) . Finally , with regards to snake eHBV affinities , the short sequences of eSNHBV1 ( 141 aa Pol ) and eSHBV2 ( 57 aa PreC and 123 aa Pol ) hamper a well-supported resolution of the tree backbones , yet there is topological indication for a grouping of eSNHBV1 with avian HBVs+eHBVs ( S4A Figure ) and eSNHBV2 with crocodilian eHBVs ( S4B–C Figure ) . Annotation suggests that an X or X-like ORF is absent in non-avian eHBVs , while the genomes of orthohepadnaviruses and avihepadnaviruses appear to contain an X and X-like gene , respectively ( Fig . 4A ) . Even when aligning the translated sequences of eHBVs in the region homologous to the putative X-like ORF of avihepadnaviruses [20] , all eHBVs and even several extant avian HBVs exhibit several internal stop codons at conserved positions ( Fig . 4B ) , suggesting that an X-like ORF never existed in these unrelated HBV lineages . While it remains unclear whether the putative X-like gene in DHBV has a function [58] , it is interesting to note that the ribonuclease H ( RNH ) domain ( partially overlapping with the X/X-like ORF region ) has a moderate GC content in eHBVs and avihepadnaviruses ( S3 Figure ) , while mammalian HBV genomes exhibit a conserved X gene and a highly elevated GC content of the RNH domain . Despite both overlapping with the RNH domain of the pol ORF , X and X-like ORFs are found in different reading frames ( Fig . 4A ) . Considering that the Pol protein sequence is homologous among all HBVs and is encoded in the +1 frame , the fact that X resides in the +2 frame and X-like in the +3 frame counters homologization of the codon and protein sequence encoded in the X and X-like ORFs . This provides further evidence that the ancestor of Hepadnaviridae lacked an X or X-like gene and that the X protein arose de novo in the Orthohepadnavirus lineage [10] . The partially overlapping nature of X suggests that it emerged by using an unoccupied reading frame within a pre-existing ORF , a process termed overprinting [59] . We therefore conducted overprinting analyses ( S3 Table ) using the method described by Pavesi et al . [60] for detecting de-novo ORFs based on their codon usage . Although the X codon usage shows an expected weaker correlation with the rest of the viral genome than is the case with the other , older overlapping ORFs ( S3 Table ) , subsampling analyses suggest that the overlapping part of the X ORF is too short to derive a statistically significant conclusion ( S5 Figure ) . In contrast to non-mammalian HBVs where the RNH domain of pol and the start of preC/C overlap , these two ORFs are instead disjoined from each other in mammalian HBVs and together encompass the non-overlapping part of the X ORF ( Fig . 4A ) . It has been proposed that an ORF overlap can easily be eliminated in connection with a duplication of the particular region [61] , which could well have been the case in the ancestor of orthohepadnaviruses and led to the present genome organization . This would also explain why , apart from the aforementioned differences , the locations of all other genomic features of this region have remained unchanged throughout HBV evolution , such as the exact location of DR1 , DR2 , and ε within the pol and preC/C ORFs . To test whether there are still detectable sequence remnants ( i . e . , duplicated amino acid motifs ) of such an ancient segmental duplication , we screened the genomes of all orthohepadnaviruses against themselves as well as each other via translated nucleotide similarity searches and considered only hits that were in the same orientation in the HBV genome . Only one amino acid motif of considerable length ( i . e . , >9 translated aa on the same strand ) appears to be duplicated ( Fig . 4C ) in the entire Orthohepadnavirus genome with up to 50% sequence similarity between the two copies . Both potential duplicates reside within the preC/C ORF , one of them at its very beginning and the other near the 5′ end of the pol ORF . We therefore propose a novel scenario for de-novo emergence of the X ORF in orthohepadnaviruses ( Fig . 4A ) . This builds on the suggestion by Pavesi et al . [60] that the overlapping part of X emerged de novo via overprinting of the pol RNH domain and is completed by our inference of the origin of the non-overlapping part of the X ORF . We hypothesize that the non-overlapping part of X arose by duplication of the first two thirds of the preC/C ORF that extended from the preC/C start to the above mentioned amino acid motif ( Fig . 4D ) . A subsequent deletion of the first half of the first duplicate ( Fig . 4D ) purged the surplus in DR and ε motifs , potentially because it interfered with correct viral replication . If this coincided with the induction of a frameshift mutation ( Fig . 4D ) within the partial duplicate of preC/C , this shifted the intact downstream preC/C ORF ( +2 frame ) by one nucleotide ( +3 frame ) relative to pol that resides in the +1 frame . This would have thus prepositioned the +2 frame of the partial preC/C duplicate for overprinting ( Fig . 4D ) , while keeping the intact preC/C ORF unaffected , as it resides in a different reading frame .
Our study , together with a previous study on a Mesozoic eHBV in birds [10] , provides direct evidence for the coexistence of Hepadnaviridae and three of the five major clades of amniotes during the Mesozoic Era , two of which ( i . e . , crocodilians and turtles ) were previously not thought to be candidate hosts of extant HBVs [14] . The latter is also the case for snakes . While the cobra eHBVs remain undated , the three datable non-avian eHBVs described herein are ≥30 . 7 MY old , so we assume that these non-avian EVEs constitute snapshots of an ancient but now extinct host-virus association . This is in line with the paucity of HBV endogenization events in crocodilian , snake , and turtle genomes , in contrast to birds where dozens of these occurred during their long-lasting and ongoing coexistence [2] , . Furthermore , our non-avian HBV fossils suggest that the minimum age of definite existence of Hepadnaviridae is not >82 MY as suggested in ref . [10] , but >207 MY and thus reaches far into the Mesozoic Era . When considering indirect paleovirological evidence such as our phylogenetic analysis grouping mammalian HBVs as sister to crocodilian+avian HBVs ( but in disagreement with the apparent lack of mammalian HBV fossils [10] ) , Hepadnaviridae could be considered as a considerably older family of viruses with the root of all known HBVs at least in the Early Mesozoic or even in the common ancestor of Amniota . The fact that we identified eHBVs in crocodilian , snake , and turtle genomes implies that Mammalia is the only major lineage of land vertebrates that lacks evidence for the existence of endogenous hepadnaviruses . Unfortunately , it was not possible to determine whether the cobra eSNHBVs or their flanking sequences are present or absent in other squamate lepidosaurs ( anole lizard [35] , python [34] , and boa [33] ) , which can potentially be explained by the accelerated neutral substitution rate characteristic to this clade [27] , [34] that , together with a very high rate of DNA loss [62] , hampers the detection and comparison of orthologous non-functional genomic loci across this level of species divergence . Likewise , fast molecular evolution must have led to the scarcity of ancient transposable element ( TE ) insertions and retroviral EVEs in these genomes [62] . This is not expected in the case of the very slowly evolving genomes of turtles [27] , [28] and crocodilians [30] , [31] , all of which are littered with ancient TEs [63] , and readily explains why Mesozoic eHBVs are still detectable as such in their genomes , even after >200 MY of sequence decay and lack of selective constraint . The absence of endogenous hepadnaviruses in mammals [10] despite dozens of available genome sequences [7] and a rich diversity of extant , exogenous HBV infections [13] remains puzzling . Under the scenario of an ancient coexistence/codivergence of amniotes and Hepadnaviridae , mammalian HBVs would have had equal time for recurring , stochastic germline endogenization of viral fragments as avian HBVs had since the speciation of their amniote ancestor . Also , relative to squamates , mammalian genomes appear to have a much slower rate of DNA loss [62] and a lower substitution rate [27] , suggesting that a fixed HBV endogenization in the germline would have potentially been detectable even after many millions of years . Although the rate of mammalian sequence evolution is somewhat higher than that for birds [27] , [64] , it is less than that of squamates and therefore less likely to erase the evidence for a fixed HBV endogenization unless it were truly ancient . We conclude that , while the so far sequenced representatives of major mammalian lineages generally seem to lack eHBVs , it cannot be excluded that the foreseeable sequencing of thousands of additional mammalian genomes [6] might lead to the unearthing of recent , lineage-specific endogenizations of mammalian eHBVs . Given the evidence that hepadnaviruses coexisted with their amniote hosts at least since the Early Mesozoic , it is striking that the genome organization of HBVs have remained relatively stable over the course of >200 MY , including the patterns of overlapping protein-coding sequences and structures involved in viral replication . The only major difference among HBV genomes appears to be the presence or absence of an X gene . Our analyses provide multiple and independent lines of evidence that the common ancestor of Hepadnaviridae did not exhibit a fourth ORF and that the X gene therefore is an evolutionary novelty that arose in the Orthohepadnavirus lineage [10] . If the expressed X-like protein in duck HBVs is indeed functional [20] , [65] ( note that its function was questioned in ref . [58] ) , then this gene must have emerged de novo within avihepadnaviruses , as its putative ORF region is heavily disrupted by internal in-frame stop codons in all endogenous HBV lineages discovered so far . For example , in eCRHBV1 and eZHBV_C there are more premature stop codons in the ∼120 codons of the X-like ORF than in the total of >1100 codons of the three remaining ORFs together ( compare Fig . 4B with S1 Table and ref . [10] ) . Most importantly , it has previously been overlooked that the X and X-like ORFs cannot represent a single , homologous origin of a gene by overprinting because they lie within different frames of the homologous region of the pol ORF that they overlap with . Any structural [66] or functional [20] similarities between the encoded proteins must have thus evolved independently . A scenario of X emergence via segmental duplication of preC/C and subsequent overprinting of parts of pol and preC/C ORFs provides a simple explanation to why the DR1 ( nested at the 5′ end of preC/C ) and DR2 ( nested at the 3′ end of pol ) sequences are separated by a few hundred bp of non-overlapping , X-specific sequence in orthohepadnaviruses , while they are only a few dozens of nucleotides apart from each other in non-mammalian HBVs where the X gene is missing and pol+preC/C are overlapping instead . It is worth noting that Liu et al . [19] recently reported an avian eHBV genome that was endogenized with partially duplicated pol and preC/C ORFs , suggesting that segmental duplications do occur during replication in the virus particle and also seem to be present in the viral DNA genome that resides in the host nucleus . Finally , the restriction of the presence of X to mammalian HBVs coincides with the notion that chronic HBV infection is associated with HCC development in mammals only , while avian HBVs do not seem to cause HCC in birds [20] . This further adds to the substantial evidence for an oncogenic effect of the X gene of orthohepadnaviruses [21] , [22] , [23] , [24] , [25] , [26] . Although the X protein is known to have several indispensable functions in regulation of protein interactions [26] , [67] , [68] , the initial selective advantage during its de-novo emergence remains enigmatic in the light of the otherwise highly stable , streamlined genomes of Hepadnaviridae .
Subsequent to our initial tBLASTx searches [69] ( cutoff e-value 1e–10 ) for sequence similarity between DHBV/eZHBV_C and non-avian amniote genomes , we extracted all resultant BLAST hits ( including >5 kb of sequence per eHBV flank ) for eTHBV , eCRHBV1 , eCRHBV2 , eSNHBV1 , and eSNHBV2 from turtle , snake , and crocodilian genomes . In the case of genomes that did not yield a tBLASTx hit , we obtained orthologous sequences via BLASTn searches using the aforementioned eHBV flanks . The sequences of each eHBV locus were aligned using MAFFT ( E-INS-i , version 6 , http://mafft . cbrc . jp/alignment/server/index . html ) [70] , followed by manual realignment ( see S1 Dataset for full sequence alignments ) . Presence/absence states were ascertained by standards similar to those used for presence/absence of retroposon insertions [71] . Consequently , the shared presence ( orthology ) of an eHBV is indicated by identity regarding its truncation , orientation , and genomic target site . eHBV absence corresponds to orthologous sequences flanking an empty eHBV target site . To complete our turtle and crocodilian sampling , we sequenced orthologous fragments of the eTHBV locus in pleurodiran turtles ( Mesoclemmys dahli , Podocnemis expansa ) and the eCRHBV2 locus in the dwarf crocodile ( Osteolaemus sp . ) using standard methods [71] . Briefly , PCR reactions ( 5 min at 94°C followed by 35–40 cycles of 30 s at 94°C , 30 s at 45–53°C and 45–60 s at 72°C; final elongation of 10 min at 72°C ) were performed using specific oligonucleotide primers ( see S4 Table ) , followed by direct sequencing . The sequences were deposited in the European Nucleotide Archive ( accession numbers LK391754-LK391756 ) . We tested for evidence of non-neutral evolution in eHBV sequences by comparing nested codon models where ω was fixed to 1 or allowed to vary freely in codeml using model 0 on each pair of closely related host species with codon frequency F3X4 [72] . Model fit was assessed using the likelihood ratio test and evidence for non-neutral evolution was defined as rejection of the null model ( ω = 1 ) . After removal of premature stop codons and frameshifting indels , we analyzed the non-overlapping and overlapping parts of each ORF separately as coding sites are synonymous in one frame but non-synonymous in others in overlapping ORFs . This in principle allows us to interpret the results of the codon model for the non-overlapping sequences . As three of the five non-avian eHBVs are present in orthologous positions in two or more host species , respectively , we estimated nucleotide distances between orthologous sets of sequences . The best-fit model of nucleotide substitution was chosen using jModeltest 2 [73] under the Akaike Information Criterion ( HKY model: -lnL 2610 . 28570 ) and sequences were analyzed in BASEML [72] using the HKY model under a global clock and considering the respective species tree topologies of Fig . 1A . The calculated node ages ( i . e . , half of the distance between a given pair of sequences that diverged since the root of the species tree ) were 0 . 027 for eCRHBV1 ( 2 , 501 bp ) , 0 . 008 for eCRHBV2 ( 1 , 650 bp ) , and 0 . 151 for eTHBV ( 910 bp ) . In order to subsequently date eHBV divergences using these distances , we used neutral substitution rates reported by Green et al . [30] . For crocodilians , they estimated a neutral rate of 3 . 9×10−10 substitutions/site/year based on a whole-genome alignment between saltwater crocodile and American alligator . In the case of turtles , we used neutral substitution rates based on conserved 4-fold degenerate sites [30] , namely 8 . 43×10−10 substitutions/site/year for Pelodiscus sp . and 4 . 77×10−10 substitutions/site/year for Chelonia mydas . We aligned nucleotide sequences of eTHBV , eSNHBV1 , eSNHBV2 , eCRHBV1 , and eCRHBV2 to the whole genomes of DHBV and eZHBV_C [10] . The resulting alignment was used to localize putative start and stop codon positions for hepadnaviral ORFs , as well as to identify frameshifts . Nucleotide and amino acid sequences of hepadnaviral protein-coding genes were reconstructed after replacement of premature stop codons with “NNN” in the nucleotide sequences and removal of frameshift mutations ( see S2 Dataset for the near-complete genome of the crocodilian eCRHBV1 ) . Nucleotides of frameshifting insertions were omitted and frameshifting deletions were accounted for by insertion of “N” residues . DR sequences were identified in the near-complete eCRHBV1 genome by screening the region around the pol ORF end and the preC/C ORF start for identical direct repeat sequences . Furthermore , we analyzed the sequence of the aforementioned region in mfold [74] to locate and reconstruct the putative ε hairpin structure . We aligned polymerase protein sequences from 47 orthohepadnaviruses , 84 avihepadnaviruses , 3 full-length avian eHBVs ( eZHBV_C [10] , eBHBV1+eBHBV2 [19] ) , as well as the crocodilian eCRHBVs ( and , for S4A–B Figure , also the snake eSNHBVs ) using MAFFT and then manually realigned these . Some N-terminal sites of the alignments were problematic ( i . e . , the spacer region of the Pol protein ) and were thus excluded from further analyses . Note that concerning avian eHBVs , we only considered full-length EVEs ( eBHBV1+eBHBV2 from budgerigar [18] , [19] and eZHBV_C from Neoaves [10] ) to minimize missing data in our analyses . Non-hepadnaviral outgroups comprise reverse transcriptase sequences from representatives of caulimoviruses , retroviruses , and retrotransposons , all of which were manually aligned to the aforementioned HBV alignment . C-terminal and N-terminal residues were removed from outgroup sequences if they could not be aligned to the HBV Pol protein . For generating the precore/core protein sequence alignment , the same ingroup sampling was used as for the Pol protein , in addition to the turtle eTHBV fragment ( and , for S4C Figure , also the snake eSNHBV2 fragment ) that spans most of the preC/C ORF . After processing with MAFFT and manual realignment , capsid protein sequences from representatives of retroviruses and retrotransposons were added and manually aligned while strictly following the helix structure-based alignment of ref . [75] . Maximum likelihood phylogenetic analyses of the final Pol and PreC/C alignments ( see S3 Dataset and S4 Dataset , respectively ) were carried out using RAxML [76] ( version 7 . 4 . 7 ) . Amino acid substitution models were chosen based on model testing in MEGA5 [77] using default parameters with either all alignment sites or after partial deletion of missing data ( 95% cutoff for site coverage ) . The respective best-fit models were chosen for Pol ( JTT+G and rtREV+G ) and PreC/C ( JTT+G and WAG+G ) and , as they resulted in the same topologies with similar bootstrap support , we included only the results using the model tested with all alignment sites ( JTT+G for both Pol and PreC/C ) in Fig . 3 and S4 Figure . GC content for windows of X nucleotides length was determined by tallying up all A , C , T and G nucleotides . Only windows where the number of gaps or ambiguous nucleotides was smaller than half of the length of the window were considered . To compare similarity in codon usage between the putatively overprinted and the non-overlapping regions of the genome and older overlapping reading frames , the number of occurrences for each codon in the sequence was tallied . Spearman's rho was then used to obtain correlation coefficients between sequences as a measure of similarity . Codons for which the number of occurrences of the amino acid did not exceed its degeneracy were filtered out . This approach is similar to the method of Pavesi et al . [60] , and assumes that in the case of overlapping reading frames a more newly arisen overprinted sequence will initially be less similar in terms of codon usage to the remainder of the genome because its reading frame is shifted . To assess whether our X sequence was sufficiently long to infer overprinting from this test , we performed a subsampling analysis . Here , we asked how often a randomly selected fragment from the old overlapping reading frame of the same length as the X gene gave a correlation with the non-overlapping region that was as weak as or weaker than the value obtained for X ( 1000 bootstraps ) . Note that this codon similarity analysis assumes that usage ought to be fairly uniform across all open reading frames in the viral genome .
|
Viruses are not known to leave physical fossil traces , which makes our understanding of their evolutionary prehistory crucially dependent on the detection of endogenous viruses . Ancient endogenous viruses , also known as paleoviruses , are relics of viral genomes or fragments thereof that once infiltrated their host's germline and then remained as molecular “fossils” within the host genome . The massive genome sequencing of recent years has unearthed vast numbers of paleoviruses from various animal genomes , including the first endogenous hepatitis B viruses ( eHBVs ) in bird genomes . We screened genomes of land vertebrates ( amniotes ) for the presence of paleoviruses and identified ancient eHBVs in the recently sequenced genomes of crocodilians , snakes , and turtles . We report an eHBV that is >207 million years old , making it the oldest endogenous virus currently known . Furthermore , our results provide direct evidence that the Hepadnaviridae virus family infected birds , crocodilians and turtles during the Mesozoic Era , and suggest a long-lasting coexistence of these viruses and their amniote hosts at least since the Early Mesozoic . We challenge previous views on the origin of the oncogenic X gene and provide an evolutionary explanation as to why only mammalian hepatitis B infection leads to hepatocellular carcinoma .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"cancer",
"genetics",
"genome",
"evolution",
"evolutionary",
"biology",
"microbiology",
"oncogenes",
"hepatitis",
"b",
"virus",
"liver",
"diseases",
"infectious",
"hepatitis",
"hepatitis",
"gastroenterology",
"and",
"hepatology",
"genome",
"analysis",
"genetic",
"elements",
"genome",
"annotation",
"molecular",
"genetics",
"infectious",
"diseases",
"medical",
"microbiology",
"parasitism",
"microbial",
"pathogens",
"molecular",
"evolution",
"hepatitis",
"viruses",
"molecular",
"biology",
"community",
"ecology",
"hepatitis",
"b",
"ecology",
"trophic",
"interactions",
"viral",
"pathogens",
"genetics",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"genomics",
"mobile",
"genetic",
"elements",
"viral",
"diseases",
"computational",
"biology"
] |
2014
|
Early Mesozoic Coexistence of Amniotes and Hepadnaviridae
|
The murine parasite Heligmosomoides polygyrus is a convenient experimental model to study immune responses and pathology associated with gastrointestinal nematode infections . The excretory-secretory products ( ESP ) produced by this parasite have potent immunomodulatory activity , but the protein ( s ) responsible has not been defined . Identification of the protein composition of ESP derived from H . polygyrus and other relevant nematode species has been hampered by the lack of genomic sequence information required for proteomic analysis based on database searches . To overcome this , a transcriptome next generation sequencing ( RNA-seq ) de novo assembly containing 33 , 641 transcripts was generated , annotated , and used to interrogate mass spectrometry ( MS ) data derived from 1D-SDS PAGE and LC-MS/MS analysis of ESP . Using the database generated from the 6 open reading frames deduced from the RNA-seq assembly and conventional identification programs , 209 proteins were identified in ESP including homologues of vitellogenins , retinol- and fatty acid-binding proteins , globins , and the allergen V5/Tpx-1-related family of proteins . Several potential immunomodulators , such as macrophage migration inhibitory factor , cysteine protease inhibitors , galectins , C-type lectins , peroxiredoxin , and glutathione S-transferase , were also identified . Comparative analysis of protein annotations based on the RNA-seq assembly and proteomics revealed processes and proteins that may contribute to the functional specialization of ESP , including proteins involved in signalling pathways and in nutrient transport and/or uptake . Together , these findings provide important information that will help to illuminate molecular , biochemical , and in particular immunomodulatory aspects of host-H . polygyrus biology . In addition , the methods and analyses presented here are applicable to study biochemical and molecular aspects of the host-parasite relationship in species for which sequence information is not available .
Gastrointestinal ( GI ) nematode infections are major causes of disease in both humans and animals . Infections with Ascaris lumbricoides , hookworms ( Necator americanus and Ancylostoma duodenalis ) , Trichuris trichiura , and Strongyloides stercoralis are highly prevalent in developing countries , affecting ∼1 billion people and posing a burden estimated at ∼2 M DALYs ( Disability-adjusted life years ) ( http://apps . who . int/ghodata ) [1] . GI nematodes usually establish chronic infections , surviving in the host for considerable periods of time . This characteristic reflects the ability of these parasites to evade and modulate the host immune response from the early stages of infection while optimizing both feeding and reproduction [2] , [3] . As a result , in addition to their commonly associated effects on host physiology including malnutrition , growth stunting , and anaemia , infection with GI nematodes influences the development and/or severity of co-occurring infections and immune-mediated diseases such as malaria or type 1 diabetes , respectively [4] , 5 . Infection with the nematode Heligmosomoides polygyrus , a natural GI pathogen of mice , has provided a convenient experimental model to understand the biology of GI nematodes and the pathology associated with chronic infections with this class of helminth parasites [6] . Primary infection with H . polygyrus induces a highly polarized Th2 immune response in mice; despite induction of this response , the parasite survives and establishes a chronic infection with the differentiation and activation of host cell types that mediate potent immunoregulatory mechanisms , such as regulatory T cells and alternatively activated macrophages ( AAMΦs ) [7] , [8] . Recent studies indicate that these regulatory responses , especially regulatory T cells , can be stimulated by treatment with H . polygyrus excretory-secretory products ( ESP ) [9]–[12] . These observations suggest that this fraction of the proteome contains many of the immunomodulatory factors responsible for evasion of the host immune response , but the proteins in ESP that mediate these effects remain largely unknown . The use of mass-spectrometry based proteomics has overcome many limitations in the analysis and identification of helminth-derived proteins in ESP [13] . In general , these analyses achieve a remarkable sensitivity in protein identification if either genome , transcriptome , or proteome sequence information is available to support the interrogation of experimentally obtained mass spectra with peptide matching algorithms in database search programs [14] . However , most of this sensitivity is lost when assignation is based on homology with proteins identified in other species , as is the case for H . polygyrus and almost all other relevant parasitic nematode species for which sequence information is not available [15]–[17] . To better understand the molecular mechanisms that lead to the activation and modulation of the host immune response by GI nematodes , we used transcriptome next generation sequencing ( RNA-seq ) technologies and several bioinformatic tools to overcome the limitations in the proteomic analysis of ESP from H . polygyrus . Illumina sequencing ( www . illumina . com ) was employed to generate transcriptomic sequence data in a rapid and cost-efficient way [18] . The transcriptome assembly was used to identify proteins in the ESP using an experimental proteomic approach .
Animal procedures were conducted in accordance with the guidelines and policies of the Canadian Council on Animal Care and the principles set forth in the Guide for the Care and Use of Laboratory Animals , Animal Resources Centre , McGill University . The protocol was approved by the McGill University Animal Care Committee ( Permit Number: 4543 ) . All efforts were made to minimize discomfort and suffering to the animals during handling and manipulation . H . polygyrus was maintained and propagated in male BALB/c mice ( Charles River Laboratories , St . Constant , Canada ) by oral gavage inoculation of 400–450 third-stage larvae ( L3 ) as described [19] . Adult parasites were collected from the small intestine on day 21 post infection under a dissection microscope . Worms were washed extensively with sterile endotoxin-free PBS ( Invitrogen , Burlington , ON , Canada ) containing 80 µg/ml gentamicin ( Schering , Montreal , QC , Canada ) , 100 U/ml penicillin G , 100 µg/ml streptomycin ( Invitrogen ) , and 20 µg/ml polymyxin B ( Sigma , St . Louis , MO ) . Mice were housed in the Animal Care Facility at the Research Institute of the McGill University Health Centre . For RNA extraction , viable worms were harvested from 6 infected mice on day 21 post-infection , and ∼1000 adult female and male worms free of host tissue were selected and extensively washed . After resuspension in 0 . 5 ml PBS , 3 . 0 ml Trizol ( Invitrogen ) were added to the worm suspension . Worms were disrupted with a Polytron homogenizer at maximum speed for 3 min with the tube positioned on ice . Following centrifugation at 12 , 000× g for 10 min at 4°C , the clear upper phase was collected and extracted with chloroform . After centrifugation at 12 , 000× g for 10 min at 4°C , the upper aqueous phase was collected , and RNA was precipitated with isopropanol . RNA was centrifuged at 12 , 000× g for 10 min at 4°C . The RNA pellet was washed with 75% ethanol , followed by centrifugation at 7 , 500× g for 5 min , and the RNA pellet was dissolved in water . The 260/280 ratio of the sample was >1 . 6 . The RNA samples were stored at −70°C and until sequencing at the McGill University and Génome Québec Innovation Centre . Total RNA quality was verified on an RNA chip using an Agilent 2100 Bioanalyzer and quantified using a NanoDrop ND-1000 UV-VIS spectrophotometer ( Thermo Fisher ) . A cDNA library was prepared from 5 µg total RNA using the mRNA-Seq Sample Preparation Kit ( Illumina ) , according to the manufacturer's recommendations . Quality of the library was verified on a DNA 1000 chip using the Agilent 2100 Bioanalyzer and quantified by PicoGreen fluorimetry . The library was subjected to 108 single-read cycles of sequencing on an Illumina Genome Analyzer IIx as per the manufacturer's protocol . Cluster generation was performed on a c-Bot ( Illumina ) with a single read cluster generation kit . Sequencing was performed once using a 36 cycle sequencing kit v4 . ESP were prepared using a modification of previously described methods [9] . Briefly , adult worms were collected as described above , and viable worms were selected , washed , and cultured at a density of ∼1000 worms per ml of serum-free RPMI 1640 medium ( Invitrogen ) supplemented with 2% glucose ( Sigma ) and antibiotics for 36 h at 37°C . The supernatant was harvested , centrifuged at 8 , 000× g for 10 min to remove eggs and debris , and concentrated using an Amicon centrifugal filter device with a 3 kDa cut-off ( Millipore , Billerica , MA ) . The protein concentration in ESP preparations was determined with a Bradford Reagent kit ( Bio-Rad , Hercules , CA ) according to the manufacturer's instructions . For proteomic analysis , a pooled sample of ESP prepared from 4 harvests of adult worms from a total of 40 mice was used . The 4 ESP preparations were pooled after their migration patterns on 4–20% acrylamide SDS-PAGE were confirmed to be similar . Pooled ESP was stored at −80°C until analysis at the McGill University and Génome Québec Innovation Centre . ESP were resuspended in loading buffer containing 2-mercaptoethanol , and ∼100 µg protein were separated by SDS-PAGE through a 3 cm gradient gel ( 7–15% acrylamide ) as described [20] . Following gel staining with Coomassie Brilliant Blue G , the entire lane was subjected to automated band excision using the Picking Workstation ProXCISION ( Perkin Elmer ) to generate 15 bands per lane ( 5–7 pieces/line ) . Proteins from gel bands were subjected to reduction , cysteine-alkylation , and in-gel tryptic digestion in a MassPrep Workstation ( Micromass , Manchester , UK ) as previously described [20] . Twenty µl of the tryptic digest solution were injected on a Zorbax 300SB-C18 pre-column ( 5×0 . 3 mm , 5 µm ) previously equilibrated with water containing acetonitrile ( 5% ) and formic acid ( 0 . 1% ) using the Micro Well-plate sampler and the IsoPump modules of an Agilent 1100 Series Nanoflow HPLC . Following washing for 5 min at 15 µl/min , the pre-column was back-flushed to a 75 µm i . d . PicoFrit column ( New Objective , Woburn , MA ) filled with 10 cm of BioBasic C18 packing ( 5 µm , 300 Å ) by the acetonitrile gradient supplied by the Agilent series 1100 Nanopump to allow elution of the peptides towards the mass spectrometer at a flow rate of 200 ηl/min as described [20] . Eluted peptides were analyzed in a Q-TOF micro ( Waters Micromass , Manchester , UK ) equipped with a Nanosource modified with a nanospray adapter ( New Objective , Woburn , MA ) . The MS survey scan was set to 1 s ( 0 . 1 s interscan ) and recorded from 350 to 1 , 600 m/z . MS/MS scans were acquired from 50 to 1 , 990 m/z , scan time was 1 . 35 s , and the interscan interval was 0 . 15 s . Doubly and triply charged ions were selected for fragmentation with collision energies calculated using a linear curve from reference collision energies . MS raw data from a single run were acquired on the Data Directed Analysis feature in the MassLynx ( Micromass ) software with a 1 , 2 , 4 duty cycle ( 1 sec in MS mode 2 peptides selected for fragmentation , maximum of 4 sec in MS/MS acquisition mode ) . MS/MS raw data were transferred from the Q-TOF Micro computer to a 50 terabyte server and automatically manipulated for generation of peaklists by employing Distiller version 2 . 3 . 2 . 0 ( http://www . matrixscience . com/distiller . htmls ) with peak picking parameters set at 5 for Signal Noise Ration ( SNR ) and at 0 . 4 for Correlation Threshold ( CT ) . The peaklisted data were then searched by employing Mascot version 2 . 3 . 01 ( http://www . matrixscience . com ) and X ! Tandem version 2007 . 01 . 01 . 1 ( http://www . thegpm . org ) against the 6 open reading frames ( ORF ) translation of the transcriptomic assembly ( see below ) . Searches were restricted to up to 1 missed ( trypsin ) cleavage , fixed carbamidomethyl alkylation of cysteines , variable oxidation of methionine , 0 . 5 mass unit tolerance on parent and fragment ions , and monoisotopic . Scaffold ( version Scaffold_2_05_02 , Proteome Software Inc . , Portland , OR ) was used to validate MS/MS-based peptide and protein identifications . Peptide identifications were accepted if they could be established at greater than 95 . 0% probability as specified by the Peptide Prophet algorithm [21] . Protein identifications were accepted if they could be established at greater than 95 . 0% probability and contained at least 2 identified peptides . Protein probabilities were assigned by the Protein Prophet algorithm [22] . Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony . Reads from Illumina sequencing were trimmed in a process that consisted of search and clipping for adapter sequences , elimination of the first 16 bases of the reads to remove random hexamers , and quality trimming using a Q20 threshold on the 3′ end . The assembly was done with Velvet 1 . 0 . 13 with a kmer value set at 43 [23] . Oases 0 . 1 . 6 ( http://www . ebi . ac . uk/~zerbino/oases/ ) was then used for final transcriptome assembly . Loci generated from the Oases assembler were subjected to analysis by BLASTx and BLASTn to identify putative homologues in C . elegans , other parasitic nematodes , and organisms other than nematodes ( e-value of ≤1e-05 ) . Full assembly will be available at Nembase4 ( http://www . nematodes . org/nembase4/ ) ( Submission date: May 6th , 2011 ) [24] . Gene Ontology ( GO ) annotations were performed using BLAST2GO [25] . Mapping of GO terms was performed on the hits retrieved from the initial search with BLASTx for protein homologues against the NCBI non-redundant database with a minimum expected value of 1×10−3 and a high scoring segment pair cut-off of 33 . The annotation algorithm was set with default parameters; pre-eValue-Hit-Filter of 1×10−6 , annotation cut-off of 55 , and GO weight of 5 . Identification of enriched GO terms in the secretome dataset compared to the transcriptome was done by assessing P values from Fisher's exact tests applying robust false discovery rate ( FDR ) using the integrated framework Gossip [26] . InterProScan [27] , [28] searches were performed using the built-in feature of BLAST2GO using the conceptual translation from the longest ORF of each locus . Enrichment analysis of exported InterPro terms in the ESP vs . transcriptome datasets was also performed by assessing adjusted P values to control for FDR from Fisher's exact tests run using FatiGO [29] on the integrative online platform Babelomics ( http://babelomics . bioinfo . cipf . es ) [30] .
To identify proteins in H . polygyrus ESP , ∼100 µg ESP were separated by SDS-PAGE . The entire lane was excised in 15 pieces , digested with trypsin , and analyzed by LC-MS/MS . A preliminary protein identification attempt was performed on the complete MS data set ( 10 , 227 spectra ) using the protein sequences from nematodes in the UniProt database ( taxonomy ID:6231 , September 17 , 2010 ) as a search source for the peptide matching algorithms . After validation with Scaffold ( v . 2_05_02 ) , 20 proteins were assigned ( 95% probability ) with a total number of assigned spectra ranging between 2 to 12 and between 2 to 7 unique peptides assigned per sequence . Nineteen of 20 identified proteins were homologues of proteins from nematodes other than H . polygyrus ( Table S1 ) . To provide a more suitable information source for the peptide assignation software and to increase the number of proteins identified in the H . polygyrus ESP , an RNA-seq analysis of this organism was carried out . Using the GAIIx platform from Illumina , ∼24 . 7 million reads of raw data , amounting to >2 . 7 Gbp , were obtained from H . polygyrus poly-A selected mRNA . Initial assembly was performed after the removal of adapters , random hexamer primer sequences , and quality control trimming using Velvet 1 . 0 . 13 [23] , generating 76 , 616 contigs . Final assembly with Oases 0 . 1 . 6 resulted in 33 , 641 total transcripts ( isoforms ) in 29 , 918 loci ( 3723 alternative splice events ) ( Table 1 ) . These values do not include sequences <100 bp , which were removed for downstream analysis . Searching for protein homologues in the H . polygyrus assembly with BLASTx identified 18 , 816 ( 55 . 9% ) transcripts sharing homology with proteins from C . elegans ( E cut-off 1×10−5 ) and 15 , 338 ( 45 . 6% ) with proteins from Brugia malayi ( Table 1 ) . Only 4 sequences were found to return mouse proteins as the first BLASTx output ( E cut-off 1×10−15 ) , indicating a low degree of host RNA contamination in the preparations . Functional annotation using BLAST2GO allowed us to assign GO terms to 14 , 094 ( 41 . 9% ) sequences; 764 different cellular component terms were assigned to 8 , 694 sequences , 1 , 805 molecular function terms to 11 , 671 sequences , and 4 , 078 biological process terms to 11 , 296 sequences ( Table 1 ) . The most frequently annotated GO terms within these three categories were “integral to membrane” ( GO:0016021 , 1 , 513 sequences ) , “protein binding” ( GO:0005515; 3 , 782 sequences ) , and “embryonic development ending in birth or egg hatching” ( GO:0009792 , 2 , 239 sequences ) ( Tables 2 and S2 ) . Distribution of GO terms at level two indicated that “binding” ( GO:0005488 , 49% of annotated sequences ) and “catalytic activity” ( GO:0003824 , 32% ) were the two major molecular function categories ( Figure 1 , left ) . In the case of biological process , the most represented categories at level two were “cellular process” ( GO:0009987 , 17% ) , “metabolic process” ( GO:0008152 , 13% ) , “multicellular organismal process” ( GO:0032501 , 10% ) , “developmental process” ( GO:0032502 , 10% ) , and “biological regulation” ( GO:0065007 , 10% ) . Functional domains and protein families were assigned to H . polygyrus transcripts using InterProScan [27] , [28]; 17 , 342 ( 51 . 5% ) sequences retrieved at least one protein signature , 2 , 204 different functional domains were predicted in 15 , 435 ( 45 . 9% ) sequences , and 1 , 896 protein families in 5 , 760 sequences ( 17 . 1% ) ( Table 1 ) . A protein kinase-like domain ( IPR011009 ) was found in 360 sequences ( 1 . 1% ) , which represents the most frequently found predicted domain in the annotated dataset . In the case of protein families , the P-type , K/Mg/Cd/Cu/Zn/Na/Ca/Na/H-transporter family ( IPR001757 ) was the most abundant , found on 80 sequences ( 0 . 2% ) ( Table 3 ) . The translation of the 6 ORFs of each transcript from the RNA-seq assembly was used as input for the matching algorithms in the protein identification software . Using this strategy , 209 proteins were identified with a total number of assigned spectra between 132 and 2 with 2 to 19 unique peptides assigned per sequence . It should be noted that one sequence appears twice as it was assigned to 2 different ORFs ( Locus_541_Transcript_1/4_Confidence_0 . 692 , frames 4 and 5 ) ( Tables 4 and S3 ) . Manual verification of peptide assignments showed that all the identified peptides group in a single ORF . Annotations from the non-redundant list of ESP hits ( 208 proteins ) were extracted from the full transcriptome data set for further analysis . 642 GO terms could be annotated to sequences from the ESP subset ( 54 . 8% of the identified sequences ) , identifying 52 different cellular component terms in 47 ( 22 . 6% ) sequences , 87 molecular functions in 107 ( 51 . 4% ) sequences , and 167 biological processes in 89 ( 42 . 8% ) sequences ( Table 1 ) . At Level 2 , within the molecular function category , 8 of the 14 terms initially found in the full transcriptome dataset were also found in the ESP subset . The GO terms “binding” ( GO:0005488 , 49% of annotated sequences ) and “catalytic activity” ( GO:0003824 , 36% ) were the most abundant terms in this category ( Figure 1 , right ) . In the biological process category , 19 of 25 terms were found in the ESP subset . Although the proportion of annotated terms in the ESP subset was slightly different than in the whole transcriptome dataset , the terms “multicellular organismal process” ( GO:0032501 , 14% ) , “developmental process” ( GO:0032502 , 13% ) , “biological regulation” ( GO:0065007 , 12% ) , “cellular process” ( GO:0009987 , 12% ) , and “metabolic process” ( GO:0008152 , 10% ) were the most represented terms in both ( Figure 1 , left panels ) . InterProScan hits assigned to the ESP subset predicted at least one protein signature for 158 ( 76 . 0% ) sequences , identifying 70 functional domains in 104 ( 50 . 0% ) sequences and 41 protein families for 70 ( 33 . 7% ) sequences . The cysteine-rich secretory protein , antigen 5 , and pathogenesis-related 1 protein ( CAP ) domain ( IPR014044 ) with 25 ( 12 . 0% ) sequences identified , was the most abundant domain in the ESP subset . The allergen V5/Tpx-1-related family ( IPR001283 ) , associated with the CAP domain , was the most prevalent found in the ESP subset ( Table 3 ) . The proteins were organized according to the number of assigned spectra , indicative of protein abundance ( Table 4 ) [31] . The most abundant hits organized in this manner were categorized into 3 main groups according to their annotated features . The first group is the proteins predicted to contain the CAP domain belonging to the allergen V5/Tpx-1-related family . This group of proteins is described in the annotation tables as homologues of venom allergen-like proteins ( VAL ) , A . caninum secreted proteins , or activation-associated secreted proteins ( ASP ) . The second group is composed of globin homologues . Proteins found within this group were annotated with the biological process GO term “oxygen transport” ( GO:0015671 ) and the molecular function terms “heme binding” ( GO:0020037 ) , “oxygen transporter activity” ( GO:0005344 ) , and “oxygen binding” ( GO:0019825 ) . Although not predicted from the InterproScan in all these sequences , the globin-like domain ( IPR009050 ) and globin family ( IPR000971 ) were also annotated to some of these hits . The third group of most abundant proteins contains vitellogenin ( Vtg ) homologues . Most of these proteins are predicted to contain the characteristic Vtg open β-sheet ( IPR15255 , IPR 15817 ) domain as well as domains associated with lipid transport ( IPR015819 , IPR001747 , and IPR015816 ) and GO terms associated with the molecular function of “protein binding” ( GO:0005515 ) and the biological processes “embryonic development ending in birth or egg hatching” ( GO:0009792 ) , “determination of adult lifespan” ( GO:0008340 ) , and “positive regulation of growth rate” ( GO:0040010 ) ( Figure 1 , right panels ) . GO terms enrichment analysis using GOSSIP [26] identified terms that were over-represented in the ESP subset compared to the total transcriptome dataset ( Table S4 ) . Using adjusted P-values to control FDR ( significance set at p<0 . 05 ) as criterion for statistical significance , 14 terms within the biological process category and 8 within the molecular function category were enriched in the ESP subset . In accordance with the number of globin homologues found in the ESP subset , the biological process term “oxygen transport” ( GO:0015671 ) and its parent “gas transport” ( GO:0015669 ) were enriched in the ESP subset . Consistent with the globins and their putative role in oxygen transport via heme prosthetic groups , the molecular function terms “oxygen binding” ( GO:0019825 ) , “oxygen transporter activity” ( GO:0005344 ) , and “heme binding” ( GO:00200037 ) , together with their parent terms , “iron ion binding” ( GO:0005506 ) and “tetrapyrrole binding” ( GO:0046906 ) , were also enriched in this subset . Two other groups of hierarchically-related enriched biological process terms were delineated for their association with Vtg homologues in the ESP subset . The first group comprises the term “determination of adult lifespan” ( GO:0008340 ) and its parent “multicellular organismal process” ( GO:0032501 ) . The second group consists of “positive regulation of growth rate” ( GO:0040010 ) and parent terms “regulation of growth rate” ( GO:0040009 ) , “positive regulation of growth” ( GO:0045927 ) , and “regulation of growth” ( GO:0040008 ) . The identification of 3 homologues of glutathione-S-transferase also accounts for the enrichment of these terms . In the molecular function category , two groups of enriched terms were associated with proteins of lower relative abundance . One group includes homologues of retinol and/or fatty acid binding protein as well as repetitive ladder antigens and Vtg homologues , which have the putative ability to bind and transport vitamin A and/or lipids . These proteins were annotated under the terms “retinol binding” ( GO:0019841 ) and their parents , “retinoid binding” ( GO:0005501 ) , “isoprenoid binding” ( GO:0019840 ) , and “lipid binding” ( GO:0008289 ) . The other group is composed of certain proteases in the ESP subset , particularly several zinc metallopeptidase homologues . GO annotations in this group included the term “metallopeptidase activity” ( GO:0008237 ) and the parent terms “peptidase activity acting on L-aminoacid peptides” ( GO:0070011 ) and “peptidase activity” ( GO:0008233 ) . In addition , the molecular function terms “intramolecular oxidoreductase activity” ( GO:0016860 ) and “nucleoside diphosphate metabolic process” ( GO:0009132 ) were also enriched in the ESP dataset . The first was associated with homologues of protein disulfide isomerase , triosephosphate isomerase ( TPI ) , and macrophage migration inhibitory factor ( MIF ) . The latter included homologues of nucleoside diphosphate kinases ( NDPK ) , calcium activated nucleosidases , and ribonucleotide reductases ( RNR ) . Furthermore , InterPro domain enrichment analysis was performed using FatiGO [29] ( Table S5 ) . Likewise , adjusted P-values to control FDR were used as criteria for statistical significance ( p<0 . 05 ) ; 23 domains and families were enriched in the ESP subset compared to the transcriptome dataset . Consistent with what was found in the enrichment analysis of GO terms , there was an enrichment of predicted families and domains associated with homologues of peptidases , globins , nucleosidases , glutathione-S-transferases , Vtg and retinol and/or fatty acid binding proteins . In addition , CAP domain ( IPR014044 ) and its related allergen V5/Tpx-1-related family ( IPR001283 ) and Ves allergen ( IPR002413 ) , along with the transthyretin-like family ( IPR0001534 ) , were enriched in the ESP dataset .
The ESP fraction of the proteome from parasitic nematodes is thought to contain many of the effector molecules that contribute in a direct or indirect way to establishment and survival within the host [32] . The H . polygyrus-mouse model is a convenient system for the study of human chronic gastrointestinal parasitism; potent immunomodulatory effects of ESP preparations from this parasite have been documented [9]–[12] . Specification of the protein composition of ESP is an important step toward compiling a comprehensive list of the proteins responsible for these effects . In addition , the transcriptomic analysis-based protein identification presented here highlights other aspects related to the biology of GI nematode infections that may illuminate new therapeutic strategies . Proteomic approaches involving mass spectrometry have been applied for the characterization of ESP in several helminth species [33] . Protein identification in this manner has typically been empowered by the availability of information resulting from genome sequencing projects . Our preliminary results exemplify how the lack of this type of information and the reliance on sequences of protein homologues from different nematode species severely limit protein identification of H . polygyrus ESP; these factors would similarly limit such analyses from other unsequenced species . To overcome this limitation , we sequenced the transcriptome of H . polygyrus using Illumina technology to provide the peptide matching software with the resulting RNA-seq de novo assembly . Next-generation sequencing technologies applied to the study of parasitic nematode transcriptomes offer an efficient way to understand how these organisms orchestrate their biochemical and molecular processes within the host [34]–[36] . However , we show here that its potential includes the use of this information to study specific aspects of the proteome . In particular , the H . polygyrus RNA-seq assembly was used as a reference for the identification of proteins present in the ESP . Mass spectrometry-based proteomics has started to be exploited for the validation and/or correction of sequence datasets and associated annotations [37] . To a certain extent , this is the case for the present analysis . On the other hand , the overall output of the protein identification process is dependent on the searching space explored , in this case the 6 ORFs of the RNA-seq assembly . In addition to the sequence coverage , factors that may affect the quality of the de novo RNA-seq assembly include the performance of the assembly program as well as errors in individual reads during sequencing and genetic variation in the transcribed sequences , which complicates the recognition of sequence overlap during assembly [38] . How these factors and others ( e . g . , instrumental aspects of mass spectra acquisition ) alter the final output has not been studied extensively . In practical terms , this imposes the need for further validation when using such a dataset for downstream analysis . Comparison of frequencies and distribution of annotations provides a way to describe the degree of functional specialization of proteins in the ESP relative to the total transcriptome . GO terms enrichment analysis revealed how some of the components of the H . polygyrus ESP may be involved in processes associated with the transport and/or uptake of nutrients from the host as well as possible involvement in signalling pathways . Globin homologues in the ESP were enriched in functional annotation categories related to oxygen and heme binding . Nematode globins are distantly related to those in vertebrates and are known or predicted to play a role in several processes , given their expression in different anatomical patterns and diversity in gene structure and amino acid sequence [39] , [40] . Although a more precise understanding of the multiple functions of nematode globins is needed , it can be expected that their role in oxygen transport and supply must be critical in the low oxygen conditions of the host microenvironment , where the adult H . polygyrus attaches to and coils around the duodenal villi [41] . In this context , globin functions can vary from transport and delivery to oxygen sink depending on the affinity of oxygen binding . For example , the high oxygen affinity globin from A . suum has been proposed to prevent toxic effects of oxygen for this parasite [40] , [42] . In addition , parasitic as well as free living nematodes are heme auxotrophs [43] , and thus secreted globins may also participate as heme carriers for the supply of this prosthetic group required for many other biological processes . Another group of enriched functions found in the ESP are related to binding of lipids and retinoids . Proteins associated with these functions are involved in the transport of these hydrophobic molecules as substrates for energy metabolism , membrane biosynthesis , and signalling [44] . Identified proteins in this group include homologues of nematode polyprotein allergens/antigens ( NAR ) , fatty acid and retinol binding ( FAR ) proteins , and Vtg proteins . NAR and FAR proteins comprise classes of small ( ∼14 kDa and ∼20 kDa , respectively ) lipid binding proteins from nematodes . NAR proteins bind both retinol and fatty acids; they are synthesized as repetitive polypeptides in tandem and are subsequently cleaved into multiple functionally similar proteins [45] , [46] . FAR proteins exhibit higher affinity for retinol than for fatty acids [47] . In addition to a role in the acquisition of small lipids from the host or the microbiota , their role as parasite secreted proteins has been proposed to be the sequestration or delivery of signalling lipids to host cells [44] . Their possible role in sequestering vitamin A from the host has been associated with the pathology of parasitic nematode infections . Among these are visual impairment caused by infections with Onchocerca volvulus [47] and vitamin A deficiency in patients infected with A . lumbricoides , possibly due to malabsorption [48] . Sequestration of vitamin A may also contribute to immunomodulation as it is required for host adaptive immunity and is involved in the differentiation of CD4+ T helper ( Th ) cells and B cells . In particular , vitamin A deficiency leads to impaired intestinal immune responses , including antibody-mediated responses directed by Th2 cells [49] , [50] . Vtg proteins form a highly diverse family in the large lipid transfer protein ( LLTP ) superfamily . In addition to the ESP from H . polygyrus , these proteins have also been identified in ESP from other parasitic GI nematodes [51] , [52] . In C . elegans , Vtgs are implicated in the delivery of nutrients to support embryonic development , hence the enrichment of biological process terms associated with growth regulation . They are secreted from the intestine to the pseudocoelomic space where they transit through the gonadal basal lamina and then through the sheath pores for receptor-mediated oocyte endocytosis [53] , [54] . Therefore , it is likely that their presence in ESP from parasitic GI nematodes is the result of egg release . However , the involvement of Vtg-like proteins in modulation of insect host immune responses [55]–[57] suggests a possible additional role in negotiation of the host-parasite interface . Peptidase activity was another GO function enriched in the ESP protein set . Helminth proteases participate in the establishment , development , and maintenance of infection [58] . In H . polygyrus , developmental regulation of ESP-proteases suggest possible roles in exsheathment , invasion of the mucosa , and immune regulation during the larval stages , and feeding and migration during the adult stage [59] . Nothing is known about the substrate specificities of the H . polygyrus ESP-proteases . However , by analogy to the proteolytic cascade required for haemoglobin degradation by hookworms [60] , several components of which were also identified in A . caninum ESP [52] , the identified aspartyl , cysteine , and metalloproteinases from H . polygyrus are predicted to participate in degradation of host proteins acquired during tissue feeding . The identification of enzymes involved in nucleotide metabolism suggests a possible role of ESP in modulation of host signalling pathways . Regulation of local levels of extracellular nucleotides could affect the activity of host purinergic receptors , which mediate a variety of cellular responses , including elements of the immune system [61] . Enzymes involved in nucleotide metabolism have previously been identified in ESP from parasitic nematodes [20] , [62]–[64] . These include nucleoside diphosphate kinases , nucleosidases , and adenosine deaminases that participate in the formation of activators of purinergic receptors from ATP or UTP , such as AMP , UMP , adenosine , or inosine [61] . In addition , the homologue of ribonucleotide reductases in H . polygyrus ESP may contribute precursors for this pathway through the generation of deoxynucleotides from ribonucleotides . In addition to proteins of interest based on comparison of GO annotation between datasets , homologues of ASP or VAL proteins were also highlighted for their abundance and number of isoforms identified . These proteins are characterized by the presence of the CAP domain ( also known as SCP-like domain ) and belong to the allergen V5/Tpx-1-related family of proteins , a group of evolutionarily related eukaryotic extracellular proteins whose function remains largely unknown [32] , [65] , [66] . InterPro terms associated with this domain and families were found to be enriched in the ESP dataset . Members of this family include cysteine-rich sperm proteins ( CRISPs ) , insect venom allergens , and plant pathogenesis family-1 ( PR-1 ) proteins . Reasons to suspect a role for these proteins at the nematode-host interface ( including pathogenesis ) include the rapid and specific release of N . americanus ASP-2 during the transition from larval to parasitic stages as well to their neutrophil chemoattractant activity [67] , [68] , and the angiogenic effects of several O . volvulus ASPs [69] . In addition to proteins highlighted on the basis of enrichment of functional annotation , other relevant proteins in H . polygyrus ESP include homologues of glycolytic and metabolic enzymes . Of particular interest are triosephosphate isomerase ( TPI ) , fructose bisphosphate aldolase A ( FBPA ) , and enolase ( ENO ) , which have consistently been reported in nematode ESP , a pattern suggesting that their release cannot be simply due to worm death or damage during culture [32] . While the function of these proteins remains obscure in the context of host-nematode relationships , there is evidence of the association of these enzymes with host cell surface components and their involvement in functions unrelated to glycolysis , including microbial pathogenesis and autoimmune disorders [70]–[73] . Possible immunomodulators also include a homologue of MIF , a parasite protein that mimics a mammalian cytokine , which has been reported in many nematode ESPs . MIFs are usually associated with pro-inflammatory responses . However , in contrast to the mammalian cytokine , nematode MIF acts in a Th2 environment to induce AAMΦs [32] , [74] , [75] . In addition , the cysteine protease inhibitor ( CPI ) homologue identified in H . polygyrus ESP may modulate immune responses to unrelated antigens by inhibition of antigen processing and presentation by antigen presenting cells or by inhibition of T-cell proliferation , which may contribute to the state of cellular hypo-responsiveness characteristic of chronic parasitic nematode infections [76]–[78] . Also of interest are the previously characterized C-type lectins ( CTL ) from H . polygyrus and galectin homologues identified in the ESP in the present study [79] . Their role as immunomodulators is suggested by the involvement of these carbohydrate-binding proteins in a variety of immune functions [80]–[83] as well as the eosinophil attracting activity that has been reported for a galectin from Haemonchus contortus [84] . Finally , the presence of homologues of peroxiredoxin ( PRX ) and glutathione S-transferase ( GST ) in H . polygyrus ESP suggests a role for enzymes involved in detoxification of reactive oxygen species ( ROS ) released from the host [85] , [86] . Other roles for these enzymes may include the induction of AAMΦs , as shown for a helminth PRX , promotion of Th2 immune responses , and the involvement of GST in heme transport and detoxification [87]–[89] . In conclusion , we employed the next-generation sequencing and proteomic approaches to gain insights into the transcriptome of adult H . polygyrus and used the dataset to identify protein components of the ESP . Comparison of functional annotation categories of the total transcriptome , which provides a picture of the total proteome , with those of the ESP subset allowed us to identify functions and associated proteins that may play a role at the host-parasite interface , where many events critical for success of the infection occur . The data presented here contribute to the identification of individual components that may be responsible for the immunomodulatory activity that has been reported for H . polygyrus ESP . Moreover , methods and analyses presented here are useful for the study of biochemical and molecular aspects of nematode biology in other species for which sequence information is not available .
|
Gastrointestinal ( GI ) nematode infections are major causes of human and animal disease . Much of their morbidity is associated with establishment of chronic infections in the host , reflecting the deployment of mechanisms to evade and modulate the immune response . The molecules responsible for these activities are poorly known . The proteins released from nematode species as excretory-secretory products ( ESP ) have potent immunomodulatory effects . The murine parasite Heligmosomoides bakeri ( polygyrus ) has served as a model to understand several aspects related to GI nematode infections . Here , we aimed to identify the protein components of H . polygyrus ESP through a proteomic approach , but the lack of genomic sequence information for this organism limited our ability to identify proteins by relying on comparisons between experimental and database-predicted mass spectra . To overcome these difficulties , we used transcriptome next-generation sequencing and several bioinformatic tools to generate and annotate a sequence assembly for this parasite . We used this information to support the protein identification process . Among the 209 proteins identified , we delineated particular processes and proteins that define the functional specialization of ESP . This work provides valuable data to establish a path to identify and understand particular parasite proteins involved in the orchestration of immune evasion events .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"biotechnology",
"medicine",
"biochemistry",
"infectious",
"diseases",
"model",
"organisms",
"global",
"health",
"immunology",
"biology",
"proteomics"
] |
2011
|
Proteomic Analysis of Excretory-Secretory Products of Heligmosomoides polygyrus Assessed with Next-Generation Sequencing Transcriptomic Information
|
The proteins responsible for the key molecular events leading to the structural changes between the developmental stages of Echinococcus granulosus remain unknown . In this work , azidohomoalanine ( AHA ) -specific labeling was used to identify proteins expressed by E . granulosus protoscoleces ( PSCs ) upon the induction of strobilar development . The in vitro incorporation of AHA with different tags into newly synthesized proteins ( NSPs ) by PSCs was analyzed using SDS-PAGE and confocal microscopy . The LC-MS/MS analysis of AHA-labeled NSPs by PSCs undergoing strobilation allowed for the identification of 365 proteins , of which 75 were differentially expressed in comparison between the presence or absence of strobilation stimuli and 51 were expressed exclusively in either condition . These proteins were mainly involved in metabolic , regulatory and signaling processes . After the controlled-labeling of proteins during the induction of strobilar development , we identified modifications in protein expression . The changes in the metabolism and the activation of control and signaling pathways may be important for the correct parasite development and be target for further studies .
The parasite Echinococcus granulosus is a cestode tapeworm that acts as the causative agent of cystic echinococcosis ( cystic hydatid disease ) , one of the 17 neglected tropical diseases to be recently prioritized by the World Health Organization [1] . During its life cycle , the E . granulosus adult worm resides in the intestine of the definitive host ( e . g . , dogs ) , releasing their eggs with the host feces . Following ingestion by the intermediate host ( e . g . , domestic ungulates ) , the eggs release oncospheres that penetrate the intestinal wall and then migrate to various organs of the host . At the organ site , the oncosphere develops in the larval stage of the parasite , the hydatid cyst ( metacestode ) . The pre-adult forms ( protoscolex , PSC ) are asexually formed in the cyst germinal cellular layer and liberated into the lumen of hydatid cysts [2–7] . In the cyst cavity , PSCs may remain in an inactive state for years until the structural integrity of the cyst is lost and they exhibit a dual developmental capacity . When ingested by a definitive host , PSCs sexually differentiate into fully developed , segmented adult worms in a process called strobilation . Alternately , upon hydatid cyst rupture and the release of its contents into the peritoneal cavity of an intermediate host , PSCs can dedifferentiate into secondary hydatid cysts [8] . This dual developmental capacity of the parasite and its requirement for more than one host to complete its life cycle are associated with its ability to readily respond to host environmental changes and regulate its gene expression and protein synthesis [9–11] . Transcriptional and proteomic studies have identified differentially expressed genes and proteins between the different life stages and cyst components of E . granulosus [6 , 12–14] . However , the identities of the proteins responsible for key molecular events that lead to structural changes of the parasite and its transition between different developmental stages remain essentially unknown . One possible reason for this is the difficulty of indirectly associating changes in gene expression to the response from a particular stimulus . Consequently , the direct visualization and identification of newly synthesized proteins ( NSPs ) is useful for revealing the spatiotemporal characteristics of proteomes during development [15] . Recently , the application of bioorthogonal non-canonical amino acid tagging ( BONCAT ) and fluorescent non-canonical amino acid tagging ( FUNCAT ) have been described for the non-radioactive labeling , visualization , purification and identification of NSPs [11 , 16 , 17] . In BONCAT , newly synthesized proteins containing non-canonical amino acids containing either azide or alkyne moieties , such as the methionine ( Met ) analogue azidohomoalanine ( AHA ) , are chemically combined with affinity tags . The alkyne or azide functional groups used in BONCAT require further purification steps , whereas FUNCAT uses fluorescent tags for in situ visualization . BONCAT has been used for labeling NSPs in response to different stimulus in mammalian [16 , 17] and bacterial [18] cells . Moreover , BONCAT has been used in combination with FUNCAT to show NSPs in zebrafish [19] . Further adaptations have allowed the application of these methods to identifying NSPs in model organisms such as Caenorhabditis elegans , fruit fly and mouse [20] . Here , we report the application of FUNCAT and BONCAT followed by confocal , SDS-PAGE and MS analyses to study E . granulosus NSPs and to identify 365 AHA-labeled NSPs during the E . granulosus strobilar development in vitro . Some of the identified proteins have important functions in key processes for the survival and development of the parasite , such as metabolic reactions and processes involved in host/parasite relationships . The further applicability of these methods for developmental studies in parasitic flatworms is also discussed .
Hydatid cysts from E . granulosus ( G1 genotype ) were obtained from the naturally infected livers and lungs of cattle routinely slaughtered in a local abattoir ( São Leopoldo , RS , Brazil ) . Finding hydatid cysts during mandatory animal inspection renders contaminated viscera unfit for human consumption , and E . granulosus contaminated livers and lungs were donated by the abattoir for use in this work . PSCs were collected by aspiration , decanted by gravity and washed several times with PBS , pH 7 . 4 [13] . The viability of PSCs was determined using the trypan blue exclusion test [21] and the motility was evaluated visually using an inverted light microscope . Only PSCs with viability greater than 90% were used for further analysis . The PSCs were genotyped by one-step PCR [22] . PSCs were cultured in vitro as previously described [23] , with minor modifications . Briefly , PSCs were incubated at 37°C and 5% CO2 for 15 min with pepsin ( 2 mg/mL , Sigma , St . Louis , MO , USA ) in Hanks’ Balanced Salt Solution ( HBSS ) , pH 2 . 0 . The PSCs were then washed three times with PBS containing antibiotics ( 100 IU/mL penicillin and 100 mg/mL streptomycin , Sigma , St . Louis , MO , USA ) before being incubated with the appropriate medium . To stimulate the PSCs to undergo strobilar development ( SSD ) , the PSC suspension was transferred to a biphasic medium . Approximately 500 PSCs were used per mL of liquid medium over the solid base . The biphasic medium contained coagulated newborn calf serum ( Gibco , Auckland , NZ ) as the solid phase , which was obtained by heating the serum at 76°C in a water bath for 10 to 30 min . Each 100 mL of the liquid phase consisted of 83 . 5 mL RPMI without Met ( Gibco , Grand Island , NY , USA ) , 15 mL fetal bovine serum ( Vitrocell , Campinas , SP , BR ) , 1 . 15 mL glucose ( Merck , Loughborough , UK ) 30% in 18 megaOhm water , 0 . 35 mL taurocholate 0 . 2% ( Sigma , St . Louis , MO , USA ) in HBSS , 100 IU/mL penicillin/streptomycin and AHA ( Invitrogen , Eugene , OR , USA ) in a final concentration of 50 μM . The control without stimuli for strobilar development ( NSD ) consisted of PSCs maintained in a monophasic medium containing RPMI without Met , 15% of fetal bovine serum and AHA ( 50 μM ) . The negative control ( NC ) for AHA incorporation consisted of the NSD condition without AHA . The cultures were maintained at 37°C and 5% CO2 for 24 h for the proteome analysis or 72 h for the other experiments .
The proteins synthesized by E . granulosus PSCs after 72 h of cultivation in the presence of AHA ( SSD ) were labeled with TAMRA and visualized under UV to verify their electrophoretic profiles . The pattern of bands in the total extract of PSCs was assessed by inspection of the gels stained with coomassie blue ( Fig 1A ) . The NSPs of the PSCs could be visualized by UV in the same gel . The analysis of the NSPs revealed a complex but lower number of distinct banding patterns that ranged from 10 to 225 kDa ( Fig 1B ) . In the negative control condition without AHA , the total proteins were visualized with coomassie blue ( Fig 1C ) , but no band corresponding to NSPs was visualized under UV ( Fig 1D ) except for a non-incorporated TAMRA residual band , indicating the high specificity of the reaction . To evaluate whether the AHA-labeling methodology could be applied to identifying NSPs from in toto PSCs and to verify whether the labeled proteins were associated with a particular site or structure in PSCs undergoing strobilation ( SSD ) , Alexa Fluor 488-labeled AHA-containing NSPs were analyzed in whole mount PSCs by confocal microscopy after 72 h of in vitro culture ( Fig 2 ) . The fluorescence signal corresponding to the labeled NSPs was widely distributed and could be detected all over the PSCs ( Fig 2A ) , though there was a possible correlation between the NSPs and the suckers . In contrast , no NSP-associated fluorescence could be detected in the control experiments . The levels of DAPI fluorescence remained stable under the different conditions ( Fig 2B–2D ) . The estimated fluorescence for the NSP-Alexa Fluor and DAPI are shown in Fig 1E . We next explored the identities of the NSPs expressed in PSCs undergoing strobilation over a 24-h time window . As we searched for proteins involved in the strobilation process , we chose a shorter culture time because longer times may dilute the importance of these molecules . Scaffold was used to validate 365 non-redundant proteins in the AHA samples ( Fig 3 ) . In the non-AHA samples , we found 14 proteins that were excluded from subsequent analyzes . The purified NSPs from PSCs either with strobilation stimuli ( SSD ) or without strobilation stimuli ( NSD ) were identified by LC-MS/MS , which allowed for the identification of 248 and 275 proteins , respectively . A list of the newly synthesized proteins identified in both samples is provided in S1 Table . We found 51 proteins ( 23 SSD and 28 NSD ) in only one of the conditions studied in at least two replicates . We normalized the other 233 identified proteins by NSAF and applied Student’s t-test to find the differentially expressed proteins . 75 proteins were considered to be differentially expressed , with 34 considered to be more highly expressed in SSD and 41 more highly expressed in NSD ( Table 1 ) . In SSD , several proteins are involved in metabolic reactions ( Fig 4 ) and may indicate how the parasite obtains energy during its development into the adult worm . A feasible metabolic pathway for E . granulosus involves phosphoenolpyruvate , which is carboxylated to give oxaloacetate . Oxaloacetate is reduced to malate , which is then either oxidatively decarboxylated to pyruvate or reduced to succinate . Moreover , pyruvate can be converted into acetyl-CoA , which can participate in the tricarboxylic acid cycle or be converted into acetate . A functional annotation of the NSPs is presented in Table 1 ( data summarized in Fig 5A ) . The most representative terms in the SSD were related to post-translational modification , protein turnover , and chaperones ( O , 30% ) , energy production and conversion ( C , 19% ) , cytoskeleton ( Z , 11% ) , intracellular trafficking , secretion , and vesicular transport ( U , 9% ) and transcription ( K , 7% ) . In the NSD condition , the most representative functions were cytoskeleton ( Z , 20% ) , signal transduction mechanisms ( T , 17% ) , post-translational modification , protein turnover , and chaperones ( O , 15% ) and RNA processing and modification ( A , 10% ) . A hierarchical clustering of the up-regulated proteins listed in Table 1 was performed using the Z-score calculation on NSAF , and the results were represented as a heat map ( Fig 5B ) . Two main clusters of the NSPs during the strobilar stimuli showing high ( red ) or low ( green ) expression .
Techniques that allow for the selective labeling of molecular targets are powerful tools for understanding the molecular pathways involved in the strobilation process and the identification of key proteins that are activated in response to specific stimuli . The high specificity of AHA-labeling has been proven to be non-toxic as it does not alter the global protein synthesis rates or cause significant protein misfolding or degradation [16 , 19 , 20 , 31] . Furthermore , AHA is incorporated exclusively into NSPs without interfering with preexisting proteins . Ethical and practical difficulties in undertaking in vivo studies bring up the necessity to develop in vitro systems . Thus , we have successfully obtained an in vitro model of E . granulosus PSCs undergoing strobilar development under low potential sources of Met and conditions that facilitate the efficient incorporation of AHA . The genomic data for E . granulosus were also analyzed and confirm the feasibility of AHA labeling because proteins with non-terminal Met residues are required for this purpose ( S1 Text ) . Another factor contributing to the success of AHA-labeling is that the genome of this parasite showed no machinery for the endogenous synthesis of Met [6] . SDS-PAGE and confocal microscopy were useful in detecting the incorporation of AHA . The confocal microscopy showed a possible correlation between the NSPs and PSCs suckers . In Mesocestoides corti , the apical massif is a polynucleated cell mass that differentiates into several cell types [32] . This structure is a part of the tegmental syncytium and is located at the top of the scolex , next to the suckers . Although studies using DNA labeling in E . granulosus have not demonstrated the existence of a proliferation site in this region [33] , the possible correlation between the NSPs and the suckers may correspond to an increased protein synthesis site . The expression pattern identified by proteomic analysis of NSPs from PSC without stimuli for strobilar development revealed proteins involved in basic cellular functions , such as metabolic processes , regulation of biological processes and cellular component organization . In this sample , we have identified proteins related to transcription , translation and cytoskeletal . In contrast , proteomic analysis of NSPs from SSD samples indicated changes in parasite metabolism and this has been reported with development [34–36] In more advanced stages of worm development , there is a shift from cytosolic to mitochondrial metabolism , which tends to produce more acetate and succinate , two end products with a higher energy yield than lactate [36–38] . Thereby , changes in energy-producing pathways associated with maturation may be essential for both the correct progression of parasite life cycle as well as survival . We found a SNW domain-containing protein that is a member of the SNW gene family . Human SNW encodes a transcription coactivator that can interact with vitamin D receptor ( VDR ) and retinoid X receptor ( RXR ) [39] . It is believed that VDR and RXR may play key roles in stimulating PSCs to develop into adult worms [6 , 40] . The binding and activation of these receptors by bile acid salts regulates the expression of genes involved in differentiation , development , homeostasis and metabolism . Therefore , finding an SNW protein may indicate the presence of an active state of VDR/RXR in SSD , which is plausible given these findings . We also found three charged multivesicular body proteins that are components of the endosomal sorting complex required for transport III ( ESCRT-III ) [41–43] . ESCRT-III participates in the degradation of the surface receptor proteins , the formation of endocytic multivesicular bodies and the down-regulation of several signaling pathways . We also identified a clathrin light chain , a subunit of clathrin that participates in several membrane traffic pathways [44–46] . Extracellular vesicles are derived from the multivesicular body and act in host/parasite relationships and cell–cell signaling [47 , 48] . Although this is a preliminary result for E . granulosus , it is encouraging to find proteins related to these functions . This cell-cell communication via exosome-like vesicles has been related to sexual differentiation , survival and population density [49] . Interestingly , a comparison of the SSD up-regulated proteins showed no apparent correlation with the previously published RNAseq data [6] . However , RNAseq data includes only expression profiles from either pepsin-activated PSCs or adult worms collected from dogs , with no data available for the transition between these two stages . Therefore , we believe that this only reinforces the importance of our experimental approach in the attempt to identify early molecular events that are triggered by a developmental stimulus . This is the first report of an efficient labeling and identification of NSPs with AHA in flatworms , which provides an interesting tool for use in the search for regulatory molecules in E . granulosus and other parasitic organisms . The temporally controlled and context-dependent labelling of synthesized proteins allow the association between molecular changes and the processes occurring during induction of strobilation . Whereas the steps of induction between different stages play a central role to the correct development of the parasite , the knowledge of such processes can have great value to the improvement of new disease control strategies . Although there is vaccine for the intermediate host [50] , the WHO recommends that other stages should also be targeted for intervention [1] , which would make more efficient control . Still , considering that regulatory processes may be conserved among different helminths , the results obtained here can serve as a starting point for control studies of other parasites .
|
In the life cycle of the parasite Echinococcus granulosus , hydatid cysts produce the pre-adult form , which has the ability to either differentiate into an adult worm ( strobilation ) or dedifferentiate into a secondary hydatid cyst . We used different protein tags that allowed for the visualization and purification of proteins produced specifically after the induction of strobilar development to identify proteins that might be involved in this process ( temporally controlled and context-dependent ) . As a result , we found proteins that are involved in important processes during development , such as energy metabolism , control pathways and cellular communication . We believe that these results will be useful for the development of scientific approaches to controlling and preventing cystic hydatid disease .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Identification of Newly Synthesized Proteins by Echinococcus granulosus Protoscoleces upon Induction of Strobilation
|
Recent studies have demonstrated that the topography of thalamocortical ( TC ) axon projections is initiated before they reach the cortex , in the ventral telencephalon ( VTel ) . However , at this point , the molecular mechanisms patterning the topography of TC projections in the VTel remains poorly understood . Here , we show that a long-range , high-rostral to low-caudal gradient of Netrin-1 in the VTel is required in vivo for the topographic sorting of TC axons to distinct cortical domains . We demonstrate that Netrin-1 is a chemoattractant for rostral thalamic axons but functions as a chemorepulsive cue for caudal thalamic axons . In accordance with this model , DCC is expressed in a high-rostromedial to low-caudolateral gradient in the dorsal thalamus ( DTh ) , whereas three Unc5 receptors ( Unc5A–C ) show graded expression in the reverse orientation . Finally , we show that DCC is required for the attraction of rostromedial thalamic axons to the Netrin-1–rich , anterior part of the VTel , whereas DCC and Unc5A/C receptors are required for the repulsion of caudolateral TC axons from the same Netrin-1–rich region of the VTel . Our results demonstrate that a long-range gradient of Netrin-1 acts as a counteracting force from ephrin-A5 to control the topography of TC projections before they enter the cortex .
In the central nervous system , the vast majority of axonal projections are organized topographically . The dorsal thalamus ( DTh ) is a pivotal forebrain structure , receiving sensory inputs from the periphery and communicating with the cerebral cortex via thalamocortical ( TC ) axons . Each thalamic nucleus projects topographically to a unique set of cortical areas ( interareal , first-order level of topography ) , and subsequently , axons emerging within a given thalamic nucleus establish a topographic map of a given sensory modality within each cortical area ( intra-areal , second-order level of topography ) . Numerous anatomical studies have demonstrated that the interareal topography of TC projections is organized so that rostromedial thalamic neurons project to more-rostral cortical areas than caudolateral nuclei , which tend to project to more-caudal cortical areas [1–3] . The developmental mechanisms leading to the initial guidance and topographic sorting of TC axons , first in the ventral telencephalon ( VTel ) and , ultimately , in the dorsal telencephalon ( or cortex ) , are still poorly understood at the molecular level ( reviewed in [4 , 5] ) . Previous results suggested that the precise topography characterizing TC projections arises from the progressive sorting of axons by a series of cues present along their pathway rather than by those exclusively present in their final target , the cortex [4 , 5] . First , analysis of mouse knockouts for genes patterning the ventral telencephalon , including Ebf1 and Dlx1/2 , revealed a severe disruption of the topography of TC axon projections [6] . Second , genetic manipulation of rostral patterning molecules such as FGF8 affects the relative positioning of cortical areas without initially changing the topography of TC projections to the appropriate “cortical domain” [7 , 8] . These findings suggested a model in which TC axons are guided to their appropriate cortical domain by extracortical cues , i . e . , before reaching the cortex . Interestingly , at later stages , unidentified cortical cues are able to redirect thalamic axon outgrowth to the appropriate cortical area inside the cortex proper [8] , a result also found in heterotopic cortical grafting experiments [9] . Using a novel in vitro assay ( the “whole-mount telencephalic” assay ) , we have demonstrated that axons originating from different rostrocaudal domains of the DTh respond differentially to topographic cues present in the VTel that guide these axons to specific cortical domains [10] . Currently , the only axon guidance cue identified to play a role in the topographic sorting of TC axons is ephrin-A5 , which is expressed in a high-caudal to low-rostral gradient in the VTel [11] . In a complementary fashion , several EphA receptors , including EphA4 , EphA3 , and EphA7 , are expressed in high-rostromedial to low-caudolateral gradients in the DTh [11] . Using the whole-mount telencephalic assay developed by Seibt et al . [10] , a parallel study demonstrated that the graded expression of ephrin-A5 in the VTel and some of its receptors such as EphA7 and EphA4 in the DTh play a role in the topographic sorting of TC axons in the VTel . Interestingly , ephrin-A5–EphA4 double-knockout ( dKO ) mice show a significant and fully penetrant topographic shift of TC projections at the level of the VTel , leading to the misprojection of some thalamic motor axons to aberrantly more-caudal areas such as the primary somatosensory cortex . However , TC axon projections still display a significant level of topography in the ephrin-A5–EphA4 dKO mouse [11] , suggesting the existence of other axon guidance cues involved in the topographic sorting of TC axons in the VTel [5] . This idea is reinforced by the fact that the caudal outgrowth of TC axons in the VTel is not easily explained simply by the lack of responsiveness of EphA-deficient caudal thalamic axons to caudally enriched ephrin-A5 [11] . In the present study , we demonstrate that Netrin-1 is expressed in a high-rostral to low-caudal long-range gradient within the VTel . Using a novel quantitative axon tracing technique with high spatial resolution , we show that Netrin-1–deficient embryos show a severe disruption of the topography of TC projections at the level of VTel before they enter the cortex . Interestingly , both rostral and caudal thalamic axons are affected in the Netrin-1 knockout mouse , and we further demonstrate that both the attractive and repulsive functions of Netrin-1 are required for proper topographic projections of TC axons along the anteroposterior axis of the VTel . These results ( 1 ) provide new insights into the molecular and cellular mechanisms specifying the topography of TC axons and ( 2 ) demonstrate that the secreted ligand Netrin-1 can specify the topography of projection of large ensembles of axons , a function almost exclusively attributed to the membrane-bound ephrin/Eph signaling system [12] and more recently to another set of secreted cues ( Wnt ) and their receptor Ryk [13] .
The introduction of fluorescent carbocyanine dyes as axonal tracers represented a technical breakthrough in our ability to map the development of neuronal connectivity , especially at embryonic stages in rodents [14] . However , fluorescent carbocyanine dyes such as DiI present several important limitations , including long diffusion time , imcompatibility with immunofluorescent techniques , and most importantly , significant diffusion at the site of injection . In order to circumvent most of these problems , we adapted a well-established anterograde axon tracing technique using lysine-fixable , low molecular weight biotinylated dextran amine ( BDA , or biocytin ) microinjection in the DTh of embryonic mouse [15 , 16] . Using microinjections of BDA in the DTh of mouse embryos ( Figure S1A ) enables high spatial resolution ( few hundreds neurons labeled; Figure 1B′–1D ) and is fully compatible with immunofluorescence ( Figure S1B–1D ) . Complete anterograde filling of the axon is achieved over long distances ( 1–2 mm ) within only 4–6 h following BDA injection , as shown by the presence of large growth cones at the tip of the majority of axons ( Figure S1E ) . In order to normalize , register , and quantify the topography of TC axon projections in the VTel from microinjections performed in multiple mouse embryos , we developed a series of quantitative tools using axon tracing and image analysis allowing ( 1 ) reconstruction of the size and position of the BDA injection site in the DTh ( Figure S1G , S1I , S1K , and S1M ) , as well as ( 2 ) tracing of thalamic axon projections throughout the telencephalon ( Figure S1F , S1H , S1J , and S1L ) , and ( 3 ) quantitative analysis of the mapping of TC projections resulting from multiple injections in a large number of individuals . In order to best represent the degree of topographic sorting achieved by thalamic axons in the VTel ( i . e . , just before entering the cortex ) , we chose an anatomical landmark lying at the interface between the ventral and the dorsal telencephalon: the corticostriatal boundary ( CSB; also known as the pallial–subpallial boundary; Figures 1E , S1P , S1R , and S2 ) . The axon density maps used throughout this study ( as shown in Figure S1S ) is a flat , 2-D representation of the CSB as viewed by a virtual observer looking at the telencephalon from a lateral perspective ( see Figure 1E ) . Using this approach , we precisely and quantitatively mapped the organization of the axonal projections originating from different regions of the DTh at the level of the VTel . To do this , we performed a series of random microinjections of BDA in the DTh of E15 . 5 ( Figure S3 ) , when TC axons are still pioneering the VTel en route to the cortex , and E18 . 5 mouse embryos , a stage when all thalamic axons have reached the cortex [17] . Only injections representing less than 5% of the total volume of the DTh were analyzed in order to ensure that small groups of thalamic neurons are labeled , thus maintaining high spatial resolution . Our results show that at E18 . 5 , thalamic axons are highly segregated at the CSB according to their origin along two main axes of the DTh: the rostrocaudal axis ( Figure 1A–1A3 ) and the mediolateral axis ( Figure 1B–1B3 and 1C–1C3 ) . Axons originating from the rostral third of the DTh cross the CSB ( and therefore enter the cortex ) at a more-rostral level ( Figure 1A1 ) than axons originating from progressively more-caudolateral levels of the DTh ( Figure 1A2 and 1A3 ) . The same segregation is found for axons originating at different levels of the mediolateral axis of the DTh: axons originating from the medial third of the DTh reach the CSB at more-rostrodorsal levels ( Figure 1C , 1C1 , and 1B–1B3 ) than axons originating from progressively more-lateral thalamic domains ( Figure 1C2 , 1C3 , and 1B–1B3 ) , which cross the CSB at progressively more-caudal levels . A converse way to represent the topography of thalamic axon projections in the VTel is to categorize thalamic axon populations based on where they cross the CSB and ask where they originate within the DTh ( Figure 1D–1D3 ) . This “reverse anatomy” approach reveals that axons crossing the CSB at rostral levels originate from more-rostromedial levels of the DTh ( Figure 1D1 ) than axons crossing the caudal CSB , which originate from progressively more-caudolateral levels of the DTh ( Figure 1D2 and 1D3 ) . Interestingly , the general topography of TC projections is already present at E15 . 5 , as demonstrated using the same analysis ( see Figure S3 ) , confirming that the topographic sorting of TC axons is controlled by axon guidance cues present in the VTel when TC axons pioneer this intermediate target while forming the internal capsule ( E14–E15; [10] ) . These results ( 1 ) confirm previous studies that have primarily explored the organization of thalamic projections along the mediolateral axis of the DTh [6 , 7 , 10 , 18 , 19] , ( 2 ) reveal that TC projections are also organized along the rostrocaudal axis as proposed previously [3 , 10 , 11 , 20] , and therefore ( 3 ) that , most importantly , the overall axis of topography of TC projections is rostromedial to caudolateral [5]; ( 4 ) this new quantitative tool provides for the first time , to the best of our knowledge , a framework for the quantitative analysis of the function of axon guidance cues in the specification of the topography of TC projections in vivo . These results reveal that TC axons are organized in a precise “canvas” at the CSB as a consequence of axon guidance mechanisms specifying the topography of TC projections in the VTel , i . e . , before they enter the dorsal telencephalon [4 , 5] . In order to identify some of the axon guidance cues patterning the topography of TC axon projections in the VTel , we used a whole-mount telencephalic assay recapitulating in vitro some of the key aspects of TC pathfinding observed in vivo , including the rostrocaudal axis of TC projections [10] . Using this in vitro assay , we tested whether the mantle region of the rostral part of the VTel contains a chemoattractive cue for axons originating from the rostral thalamus by performing grafts of the mantle ( postmitotic ) region isolated from the rostral VTel ( heterotopic graft ) or the caudal VTel ( homotopic graft ) into the caudal VTel of a whole-mount telencephalon ( see Figure 2A ) . In control experiments , homotopic grafts ( caudal VTel into the caudal VTel ) axons originating from the rostral DTh ( DTR; see Figure S11 for isolation ) specifically invade the rostral domain of the VTel ( arrow in Figure 2B ) as observed in control nongrafted whole-mount telecenphalic cocultures ( see also [10 , 11] ) . Therefore , grafting itself does not perturb the topography of DTR axon projections in the VTel . However , when a small explant of rostral VTel is grafted heterotopically into the caudal VTel , rostral thalamic axons are significantly attracted towards the caudal VTel ( arrows in Figure 2C; see also quantification in Figure 2D ) , overall randomizing the outgrowth of rostral DTh axons in the VTel . This result strongly suggests the presence of a chemoattractive cue for rostral thalamic axons in the rostral part of the VTel . Earlier studies have shown that Netrin-1 is expressed in the mantle ( postmitotic ) region of the VTel of mouse embryos where the internal capsule forms [21–24] . We carefully examined the spatial pattern of Netrin-1 expression using two independent approaches at E14 . 5 and E15 . 5 , when the vast majority of TC axons pioneer the VTel to form the internal capsule en route to the cortex in the mouse embryo [17 , 24] . First , using in situ hybridization performed on horizontal and coronal sections of E14 . 5 ( Figure 3A and 3B ) or E15 . 5 mouse embryos ( Figure 3C–3E and 3I ) , we found that Netrin-1 mRNA is expressed in a high-rostral to low-caudal gradient in the mantle region of the VTel . We took advantage of a gene trap mouse line in which a LacZ expression cassette was inserted into the first intron of the Netrin-1 coding sequence ( Ntn1Gt ( pGT1 . 8TM ) 629Wcs allele , abbreviated Ntn1LacZ; [21] ) . As shown in Figure 3F and 3G , anti-β-galactosidase immunofluorescence in E15 . 5 Ntn1LacZ/+ mouse embryos recapitulates faithfully the graded expression of Netrin-1 mRNA at the same age ( Figure 3C and 3D ) . In order to examine the spatial relationship between this gradient of Netrin-1 expression and TC axons in the internal capsule , we performed L1 immunofluorescent staining in combination with anti-β-galactosidase immunofluorescence in Ntn1Lacz/+ embryos at E15 . 5 ( Figure 3H ) . A quantitative analysis of both Netrin-1 mRNA expression and anti-β-galactosidase immunofluorescence along the rostrocaudal axis of the VTel reveals an almost linear high-rostral to low-caudal gradient ( Figure 3I ) . Therefore , this gradient of Netrin-1 expression represents a good candidate to exert a function in the control of the topography of TC axons along the rostrocaudal axis of the VTel . Inspection of the internal capsule of wild-type or Netrin-1 knockout embryos at E18 . 5 using anti-L1 staining ( which labels both TC and callosal , but not corticothalamic , axons [25] ) failed to reveal any major axon outgrowth defect ( see Figure S4 ) : horizontal sections of E17 . 5 Netrin-1 knockout embryos revealed no obvious decrease in the number of thalamic axons compared to wild-type littermates at the level of ( 1 ) the thalamic peduncle ( axon bundle crossing the diencephalic to telencephalic boundary ) , ( 2 ) the internal capsule , or ( 3 ) the corticostriatal boundary compared to wild-type embryos ( Figure S4 ) . This qualitative analysis suggested that Netrin-1 is not simply required in vivo for proper outgrowth of thalamic axons into the internal capsule as suggested previously [22] . As shown later using whole-mount telencephalic cocultures , wild-type DTR or caudal DTh ( DTC ) axons growing in the VTel of Netrin-1–deficient embryos confirms quantitatively the absence of axon outgrowth defect compared to control wild-type telencephalon . Therefore , we conclude that Netrin-1 expression is not required for extension of thalamic axons in the VTel . In order to test whether Netrin-1 controls the guidance of TC projections in the VTel , we performed BDA microinjections in the DTh of both wild-type ( n = 34 ) and Netrin-1LacZ/LacZ ( Netrin-1−/−; n = 17 ) E18 . 5 embryos . A qualitative illustration of the type of topographic projection defect observed in the Netrin-1 knockout is shown in Figure 4 A–4G following a relatively large injection of BDA ( more than 5% of DTh volume; injection not used for our quantitative analysis ) in the rostral part of the DTh of a control ( Netrin-1+/−; Figure 4B–4D ) or Netrin-1 knockout embryo ( Figure 4E–4G ) . Using an oblique plane of section revealing the entire tract of TC projections from the DTh to the cortex ( see Figure 4A; [26] ) , we show that thalamic axons originating from the rostral DTh invade more-caudal territories of the VTel of Netrin-1 knockout embryos ( arrows in Figure 4E–4G ) than in control embryos ( Figure 4B–4D ) . Our quantitative analysis of a large number of BDA injections in E18 . 5 embryos reveals a profound disruption of the topography of TC projections in the Netrin-1 knockout mouse ( see Figures 4 and S5 ) . The significance of the differences between each axon density map of TC projections is tested statistically using a two-way analysis of variance ( ANOVA ) test comparing Netrin-1+/+ and Netrin-1−/− embryos ( Figure S6 ) . First , when clustered along the rostrocaudal axis of the DTh , the most significant differences in the pattern of TC projections concerns axons originating from the rostral thalamus , which reach the dorsal telencephalon at significantly more-caudoventral levels of the CSB . Thalamic axons originating from both the medial and caudal third of the DTh reach the CSB at a significantly more-rostral level in the Netrin-1−/− embryos than in wild-type control ( Figures 4J , 4K , and S6B–S6D ) Similar disruption of the topography of TC projections in the VTel is visible when examining the mediolateral organization of thalamic projections ( Figure 4L–4O ) : axons originating from the medial and central part of the DTh reach the CSB at more-caudoventral levels in Netrin-1 knockout compared to wild-type mouse embryos . Additionally , axons originating from the lateral-most third of the DTh in Netrin-1 knockout embryos are significantly shifted rostrodorsally at the level of the CSB compared to wild-type control ( Figure 4O and S6H ) . Using a reciprocal analysis , we confirmed these results by categorizing neuron position within the DTh based on the rostrocaudal level at which their axons cross the CSB ( Figure 4P ) . This analysis confirms the severe disruption of the topography of thalamic projections characterizing the Netrin-1 knockout embryos . Axons crossing the CSB at its rostral-most third originate from the rostromedial part of the DTh in wild-type embryos ( green in Figures 4Q and S6J ) , but in contrast , originate from a more widespread area of the DTh in the Netrin-1 knockout , including the extreme caudolateral territories of the DTh ( Figures 4Q and S6J ) . Axons crossing the CSB along its medial third originate from a more-caudolateral domain of the DTh in Netrin-1 knockout compared to wild-type littermates ( Figures 4R and S6K ) . Strikingly , the reverse is found for thalamic axons crossing the CSB along its caudal third , where cell bodies are found in a more-rostromedial position of the DTh in Netrin-1 knockout than in wild-type littermates ( Figures 4S and S6L ) . We also examined whether Netrin-1 is controlling the segregation of thalamic axons along the dorsoventral axis of the VTel . Using the reverse anatomy approach , we found that in wild-type mice , thalamic axons crossing the dorsal half of the CSB tend to originate from more-rostral domains of the DTh , whereas thalamic axons crossing the ventral half of the CSB tend to originate from the caudal DTh ( green in Figure S7 ) . The segregation along this axis is less marked than along the rostromedial to caudolateral axis ( see Figure 1 ) . We found small but significant differences in thalamic axon segregation along the dorsoventral axis of the CSB between control and Netrin-1–deficient embryos ( Figure S7 ) suggesting that Netrin-1 might play a role in segregating thalamic axons along the dorso-ventral axis of the internal capsule . These results are surprising because they suggest that Netrin-1 gradient in the VTel is not only attracting rostromedial thalamic axons in the rostral part of the VTel , but might also act as a repulsive cue for caudolateral thalamic axons . In other words , in the absence of Netrin-1 in vivo , rostromedial thalamic axons are shifted caudally according to their responsiveness to Netrin-1 ( compatible with the removal of a rostral attractant in the VTel ) , but at the same time , caudolateral thalamic axons are shifted rostrally according to their responsiveness to Netrin-1 ( compatible with the removal of a rostral repulsive cue in the VTel ) . However , there is a potential caveat with this interpretation: Netrin-1 is not only expressed in the VTel , it is also expressed in the DTh itself ( see arrowheads in Figure 3A and 3C ) . Therefore , at this point , we could not exclude that some of the topographic defects of TC axon outgrowth observed in vivo in the Netrin-1 knockout embryos could be due to Netrin-1 expression in the DTh itself . In order to test whether the graded expression of Netrin-1 in the VTel is required for the establishment of the topography of TC projections , we took advantage of our whole-mount telencephalic coculture assay in order to uncouple the genotype of the DTh and the telencephalon ( see [10 , 11] ) . As shown in Figure 5A and 5E , we performed whole-mount cocultures between wild-type E14 . 5 EGFP-expressing dorsal thalamic explants ( rostral DTh , Figure 5A–5D; or caudal DTh , Figure 5E–5H ) with telencephalic vesicles isolated from isochronic wild-type ( Figure 5B and 5F ) or Netrin-1−/− embryos ( Figure 5C and 5G ) . Our results show that in the absence of Netrin-1 in the VTel , a significant proportion of axons originating from the rostral part of the DT are shifted caudally ( red arrow in Figure 5C ) compared to control cocultures ( arrow in Figure 5B ) . However , a contingent of rostral thalamic axons is still projecting to the rostral third of the VTel ( arrowheads in Figure 5C ) . The quantification of these cocultures ( Figure 5D ) demonstrates that axons originating from the rostral DTh and growing in Netrin-1–deficient telencephalon can be subdivided into two subpopulations that are both significantly shifted caudally ( two peaks in Figure 5D ) compared to control cocultures ( green arrow in Figure 5D ) . Next , we performed whole-mount telencephalic cocultures using axons originating from the caudal part of the DTh ( DTC ) . As shown previously [10] , DTC axons diffusely invade caudal territories of the VTel ( Figure 5F and 5H ) . However , caudal DTh axons growing in a Netrin-1–deficient telencephalon do not show a preferential caudal outgrowth ( arrowhead in Figure 5G ) but instead display a more random distribution in the VTel ( Figure 5H; see also Figure S8 ) . Taken together , these results demonstrate that the high-rostral to low-caudal gradient of Netrin-1 expression in the VTel is required for the differential topographic mapping of thalamic axons before they reach the cortex . These results also suggest that the function of Netrin-1 in the topographic sorting of TC axons in the VTel requires both its attractive and repulsive properties . We first tested whether Netrin-1 differentially affects DTh axon populations through either attractive or repulsive activity , by testing directly whether Netrin-1 has a differential effect on different TC axons using collagen cocultures between E14 . 5 DTR or DTC explants and aggregates of HEK 293 that are stably expressing Netrin-1 [27] ( Figure S9 ) . These results show that DTR axons are significantly attracted by Netrin-1 ( in accordance with [22 , 28] ) , but at the same time , that DTC axons are not attracted , but rather moderately repulsed , by Netrin-1 in vitro . One of the drawbacks of this collagen coculture assay is that TC axons are not growing through their “natural” environment , and therefore , axon responsiveness to specific axon guidance cue could be biased because axons do not express the right complement of axon guidance receptors . A precedent for this has been well documented in the developing spinal cord where commissural axons only up-regulate surface expression of Robo receptors after crossing the midline and therefore are not responding to the midline repellent Slits before they reach the midline [29] . In order to better test the differential effects of Netrin-1 on the response of thalamic axons originating from the rostral and caudal thalamus in a contextual environment , we performed whole-mount telencephalic assays in which a source of Netrin-1–expressing HEK293 cells ( [23] ) is ectopically grafted in the caudal VTel ( see Figure 6A and 6E ) . Axons originating from the rostral thalamus now invade more-caudal territories of the VTel , overriding the repulsive effect of ephrin-A5 and other putative caudal repulsive activity [11] , suggesting that they are attracted towards a caudal source of Netrin-1 ( arrows in Figure 6C ) , whereas the control 293 cells graft has no effect ( Figure 6B ) . Interestingly , the reverse is found for caudal thalamic axons , which are significantly shifted rostrally when confronted with a caudal source of Netrin-1 ( Figure 6G and 6H ) as compared to control grafts ( Figure 6F and 6H ) . Note that this caudal source of Netrin-1 imposed experimentally is likely to disrupt the endogenous gradient of Netrin-1 still present at high levels in the rostral part of the VTel ( see Figure 3 ) . These results strongly suggest that in the VTel , Netrin-1 functions as a chemorepulsive cue for caudal thalamic axons and a chemoattractive cue for rostral thalamic axons . So far , our results imply that dorsal thalamic neurons express different Netrin-1 receptors conferring attractive ( for DTR axons ) or repulsive ( for DTC axons ) responses to Netrin-1 . In order to substantiate this hypothesis , we examined the pattern of expression of several transmembrane receptors known to mediate Netrin-1 responsiveness: Deleted in Colorectal Cancer ( DCC ) , known to mediate chemoattraction to Netrin-1 [27] , and homologs of the Caenorhabditis elegans Unc5 receptor called Unc5A–C ( also called Unc5H1–3 ) , known to mediate chemorepulsion to Netrin-1 upon heterodimerization with DCC [30 , 31] . A fourth mammalian ortholog of Unc5 has been recently identified , but its affinity for Netrin-1 has not been assessed yet [32] . We performed in situ hybridization for DCC and Unc5A , Unc5B and Unc5C ( Figure 7 ) on serial horizontal sections of E14 . 5 mouse embryos in order to best visualize differences of Netrin-1 receptor expression along the rostromedial to caudolateral axis of the DTh . First , we wanted to define accurately the caudal extent of the DTh on horizontal sections from E14 . 5 mouse embryos . To do this , we used two markers: first the transcription factor Gbx2 , which is a reliable marker of the DTh at E14 . 5 [33 , 34] , and the transcription factor bHLHB4 , which has been recently identified as a marker of the pretectum , which is immediately caudal to the DTh during embryogenesis [35] . Our results show that these two markers reliably identify the caudal limit of the DTh on horizontal sections of E14 . 5 mouse embryos along the dorsoventral axis of the diencephalic–mesencephalic boundary ( Figure 7C–7K ) . Therefore , in the rest of our analysis , we used the caudal limit of Gbx2 expression as a marker of the caudal limit of the DTh ( see lines in Figure 7N , 7Q , 7T , and 7W ) . We found that DCC mRNA is expressed at high levels in the rostromedial part of the DTh ( Figures 7L–7Q and S10A and S10B ) but is also expressed at lower levels in more-caudolateral territories of the DTh ( arrows in Figures 7U and S10C ) . In contrast , Unc5A , Unc5B , and Unc5C are expressed in nonoverlapping caudolateral domains of the DTh ( Figure 7M , 7N , 7P , 7Q , 7S , and 7T , respectively , and Figure S10D–S10L ) . The star in Figure 7N , 7Q , 7T , and 7W marks the peak of Unc5A–C expression . These complementary patterns of expression are compatible with our model , suggesting that axons originating from the rostromedial DTh ( which are attracted by Netrin-1 in the rostral part of the VTel ) express DCC only , whereas thalamic axons originating from the caudolateral domain of the DTh are repulsed by Netrin-1 in the rostral VTel and express Unc5A–C , as well as low levels of DCC . Interestingly , DCC and Unc5A–C are highly expressed in other parts of the diencephalon , including the epithalamus ( Figure S10B , S10C , S10E , and S10H ) , the ventral thalamus ( Figure S10B , S10E , and S10K ) , as well as in the pretectum ( Figure 7L–7W ) , suggesting other functions during diencephalic/mesencephalic development . We tested whether DCC is required in the topographic projection of thalamic axons by using a well-characterized function-blocking anti-DCC antibody ( clone AF5 [27] ) in the whole-mount telencephalic coculture assay ( Figure 8A and 8E ) . Our results show that DTR axons specifically invade the rostral domain of the VTel when cultured in the presence of isotype-control mouse IgG ( Figure 8B ) , but in the presence of function-blocking anti-DCC antibodies , DTR axon outgrowth is significantly randomized ( Figure 8C ) and grows significantly more caudally than in control cocultures ( Figure 8D ) . Similarly , blocking DCC function tends to randomize the outgrowth of DTC axons , which invade significantly more-rostral domains of the VTel ( arrow in Figure 8G and 8H ) compared to DTC axons in control cocultures ( Figure 8F and 8H ) . Overall , these results strongly suggest that DCC receptor function is required both for the attraction of DTR axons to rostral Netrin-1–rich territories of the VTel and for the repulsion of DTC axons away from the same domain . We next tested whether Unc5 receptor function is required for the topographic projections of DT axons in the VTel . We used a commercially available polyclonal antibody initially raised against the extracellular domain of Unc5H1 ( anti-rat Unc5H1 , R&D Systems ) and reported to act as a function-blocking reagent against both Unc5A and Unc5C ( Unc5H1 and Unc5H3 , respectively [36] ) . We verified the cross-reactivity of this anti-rat Unc5H1 antibody with mouse Unc5A , 5B , and 5C proteins using a biochemical approach ( see Figure S12 ) . Our results show that anti-rat Unc5H1 binds to mouse Unc5A and Unc5C , but not Unc5B , and that its relative affinity for Unc5C when standardized to anti-myc immunoreactivity is about a third of its affinity for Unc5A ( Figure S12 ) . We used this reagent to block Unc5A/C receptor function in the whole-mount telencephalic assay using both DTR ( Figure 9A–9D ) and DTC ( Figure 9E–9H ) . Our results show that blocking Unc5A/C receptor function does not have any significant effect on the guided outgrowth of DTR axons in the rostral domain of the VTel ( Figure 9C and 9D ) compared to control ( Figure 9B and 9D ) . In contrast , blocking Unc5A/C receptor function had a highly significant effect on DTC outgrowth , inducing a significant shift of DTC axon outgrowth into the rostral Netrin-1–rich domain of the VTel ( Figure 9G and 9H ) compared to control ( Figure 9F–9H ) . These results suggest that Unc5A/C receptors are required for the repulsion of DTC axons away from the rostral Netrin-1–rich domain of the VTel but do not play any role in the attraction of DTR axons towards the same region . We tested whether Unc5 receptor expression is the critical determinant of the difference between DTR and DTC axons towards Netrin-1 in the VTel . To do this , we overexpressed the Unc5C receptor in rostral thalamic neurons where it is normally expressed at low levels . We implemented an ex vivo slice electroporation technique developed recently by Cobos et al . [37] . Following focal microinjection of plasmid expressing myristoylated- ( m ) Venus or Unc5C-IRES-mVenus in the DTh and slice electroporation , explants corresponding to DTR or DTC were cocultured for 4 d in vitro with isochronic whole-mount telencephalon ( Figure 10A ) . This technique results in clear visualization of single thalamic axons or small axon fascicles that were traced individually in ImageJ and plotted on a common reference for quantification ( Figure 10A ) . Our results show that overexpression of Unc5C ( but also Unc5A or B; unpublished data ) is sufficient to convert the preferential outgrowth of DTR axons in the rostral domain of the VTel ( Figure 10B ) into outgrowth in the caudal domain of the VTel ( Figure 10C ) as observed with DTC axons ( Figure 10D ) . The quantification ( Figure 10E ) demonstrates that the topography of DTR axon outgrowth overexpressing Unc5C does not differ from DTC axons but is significantly different from control DTR axons in the VTel . These results show that differential Unc5 receptor expression is a critical determinant in the topographic outgrowth of thalamic axons originating from the rostromedial compared to the caudolateral part of the thalamus in response to Netrin-1 in the VTel .
Recent studies provided evidence showing that TC axons are topographically organized in response to axon guidance cues located in the VTel [6 , 10 , 11] . However , the exact 3-D organization of TC axons in VTel , where they form the internal capsule with descending corticofugal axons , has remained elusive because of the lack of quantitative analysis . Qualitative analysis based on carbocyanine injections in single brains suggested that TC axons are segregated according to their origin in the DTh along the mediolateral axis [18 , 19 , 39] as well as the rostrocaudal axis [10] . Our quantitative analysis demonstrates that both axes are equally important , and we show that at the level of the CSB , i . e . , before invading the cortex , thalamic projections are highly organized along a rostromedial to caudolateral axis ( Figure 1 ) . Therefore , there is a precise “blueprint” of the topography of TC projections generated before entering the cortex as suggested previously [5] . Where exactly is this topography initiated within the VTel ? Axons entering the VTel show a loose degree of organization when pioneering the internal capsule , and axons originating from different regions of the thalamus have to redistribute or “fan out” over a large area: at E14/15 , thalamic axons pioneer the internal capsule as a bundle referred to as the thalamic peduncle , roughly 100–200-μm wide along its rostrocaudal axis . These axons will redistribute over approximately 2–3 mm when they reach the CSB and enter the cortex . Based on previous and present results , we proposed that TC axon sorting occurs progressively as the axons grow along the mediolateral axis of the VTel [5] ( Figure 11 ) . Interestingly , the only two axon guidance molecules ( ephrin-A5; [11] and Netrin-1; present study ) identified so far as playing a significant role in this topographic sorting of TC axons in the VTel are both expressed in the most lateral part of the mantle region of the VTel and are therefore likely expressed by postmitotic neurons forming the striatum . Future studies will address how opposing gradients of ephrin-A5 and Netrin-1 are generated . Two interesting possibilities come to mind: first , this gradient is the result of patterning cues such as Shh or fibroblast growth factors ( FGFs ) specifying the rostrocaudal identity of ventral telencephalic regions , and/or second , this graded expression of Netrin-1 is the result of a graded density of cells migrating rostrocaudally within the VTel from a point source . Recent evidence suggests that thalamic axons' responsiveness to Netrin-1 expressed in the VTel is modulated by extracellular serotonin levels [39] . Bonnin et al . [39] reports that in vitro , in the absence of serotonin , Netrin-1 is repulsive to “anterior” thalamic axons and attractive to “posterior” thalamic axons , which seems at odds with our present results . However , the authors provide evidence that in the presence of high concentrations of serotonin ( 30 μM ) , these responses are reversed , and now Netrin-1 is attractive to anterior thalamic axons and repulsive to posterior thalamic axons . Several technical differences might account for the potential discrepancies between these and our results . First , Bonnin et al . are microdissecting very small parts of the DTh “by hand , ” whereas , as previously published by our group ( see also Figure S11 ) , we can only isolate thalamic explants reproducibly from serial 250-μm–thick vibratome slices [10] . Second , as previously shown , serotonin levels are high in rodent embryonic tissue; since serotonin is present at high levels in the maternal circulation and undergoes specific uptake in the placenta [40 , 41] , one would expect axons throughout the mouse embryo to be exposed to high levels of extracellular serotonin . We did test the responsiveness of DTR or DTC explants to a source of Netrin-1 in a collagen assay in the presence of horse serum-containing medium ( presumably containing levels of serotonin comparable to blood levels found in vivo ) and confirmed that DTR axons are attracted by Netrin-1 source but that DTC are either not responsive or slightly repulsed by a source of Netrin-1 in these in vitro conditions ( Figure S9 ) . More importantly , knockout mice for the main vesicular monoamine transporter ( VMAT2 ) expressed in the embryonic brain have been generated and show undetectable levels of all three major monoamines ( serotonin , dopamine , and noradrenalin ) in the developing brain [42] . These mice show a delayed , but otherwise unaffected , TC development , and certainly no defect in the topography of ventrobasal ( VB ) thalamic axons targeting to the primary somatosensory cortex ( S1 ) [42] . These and other data strongly suggest that serotonin and other monoamines are not required for the initial establishment of the topography of TC projections in vivo [42] . Regardless of the intracellular signaling pathways mediating Netrin-1 signaling in vivo , our quantitative analysis of axon tracing in wild-type and Netrin-1 knockout embryos ( Figures 4 and S5–S7 ) provides unequivocal genetic evidence for the requirement of Netrin-1 in vivo for ( 1 ) the preferential growth of rostromedial thalamic axons into the rostrodorsal part of the VTel and ( 2 ) for the preferential growth of caudolateral thalamic axons into the caudoventral part of the VTel . The fact that in the Netrin-1 knockout embryos , axons originating from the rostromedial domains of the DTh are significantly shifted caudoventrally compared to controls and axons originating from the caudolateral domains of the DT are significantly shifted rostrodorsally ( Figures 4 and S6 ) compared to control embryos ( Figure 1 ) represents the strongest evidence in favor of our model suggesting that Netrin-1–rich rostral domain of the VTel normally acts as an attractant for rostromedial thalamic axons and a repulsive cue for caudolateral thalamic axons . This model is corroborated by our DCC and Unc5A–C expression data and more significantly by our function-blocking experiments demonstrating that a high level of DCC receptor expression by rostromedial thalamic axons mediates attraction towards the Netrin-1–rich domain of the VTel ( Figure 8 ) and that both DCC and Unc5A–C expression are required for repulsion of axons originating from caudolateral domain of the DTh away from the Netrin-1–rich rostral domain of the VTel ( Figure 9 ) . Interestingly , Netrin-1 is also required for the organization of thalamic axons along the dorsoventral axis of the developing DTh ( see Figure S6 ) , an axis that is not tightly associated with differences of expression of DCC and/or Unc5A–C . Future experiments will explore whether other Netrin-1 receptors such as Neogenin or the recently identified Unc5D ( Unc5H4 ) are differentially expressed along the dorsoventral axis of the DTh and mediate segregation of thalamic axons in the VTel . In the retinotectal projection , axons located along the nasotemporal axis of the retina are topographically mapped along the anteroposterior axis of the optic tectum through the action of EphA–ephrinA signaling system [12] , whereas axons originating along the dorsoventral axis of the retina project topographically along the mediolateral axis of the tectum . Recent evidence demonstrates that the topography established along the mediolateral axis of the optic tectum is regulated by EphB–ephrinB signaling ( reviewed in [12 , 43] ) . However , theoretical modelization suggested that the graded expression of a single axon guidance cue is not sufficient for specifying a continuous topographic map along both axis of the tectum; at least one other gradient of an additional cue is necessary for proper topographic map formation in the retinotectal system [44–47] . Indeed , recent evidence shows that a gradient of Wnt3 along the mediolateral axis of the tectum counterbalances the attractive function of ephrinB1 [13] . Our results also suggest that a Netrin-1 gradient counterbalances the function of the ephrin-A5 gradient identified in the VTel [11] and that the combined expression of these two cues ( possibly along with other graded cues ) is required for the establishment of TC topography in the VTel ( Figure 11 ) . The main difference between the retinotectal and the TC projections is that TC axons are first sorted in the VTel , their main intermediate target , before they reach their final target , the cortex . This reflects the “nested” nature of TC projections: at embryonic stages , axons from distinct parts of the mouse thalamus ( ultimately corresponding to different thalamic nuclei ) are first sorted to different cortical domains in the intermediate target , the VTel ( interareal topography ) , but at early postnatal stages , neurons in each sensory thalamic nuclei project topographically within each cortical area ( intra-areal mapping; i . e . ; sensory map formation ) [48] . Interestingly , the sorting of axons along the rostromedial to caudolateral axis of the thalamus at the level of the internal capsule ( i . e . , before they reach the cortex ) is perfectly conserved in humans as shown recently by tract-tracing studies using diffusion tensor imaging [49] , which suggests that the establishment of TC topography in the VTel is the result of an evolutionary conserved developmental mechanisms in mammals . Importantly , the basic topography of TC projections is specified in the VTel before individual thalamic nuclei can be identified cytoarchitecturally [10 , 11 , 34] . Careful examination of the expression pattern of several transcription factors , including Ngn2 , Lhx2 , Lhx9 , and Gbx2 , from E12 to postpartum day 0 ( P0 ) demonstrated that their expression is regionalized between E12 and E14 , well before the appearance of distinct thalamic nuclei ( E15–E16 ) [34] . These results suggested that phenotypic traits of thalamic neuron identity , such as their patterns of axon projections , are intrinsically specified by the combinatorial expression of transcription factors , a model based on specification of motor neuron identity in the developing spinal cord [50 , 51] . Interestingly , an experimental validation of this model was provided recently by the analysis of the function of the bHLH transcription factor Ngn2 , which specifies the topography of TC projections to the frontal cortex by controlling the responsiveness of thalamic axons to so-far unidentified intermediate axon guidance cues present in the VTel [10] . Based on the present results , we can hypothesize that the caudal shift displayed by rostral thalamic axons of the Ngn2 knockout embryos in the VTel could be due to down-regulation of Netrin-1 , or ephrin-A5 responsiveness . Further experiments will determine whether Netrin-1 and ephrin-A5 receptors examined in this and previous studies ( DCC , Unc5A–C , and EphA4 ) have altered expression profiles in the DTh of Ngn2 knockout embryos . Several studies have implicated Netrin-1 as an intermediate target cue for both corticofugal [23] and TC axons [22 , 39] . Netrin-1 was first shown to stimulate the outgrowth of descending corticofugal axons in vitro and attract these axons towards the internal capsule [23] . Interestingly , Netrin-1 expression in the VTel has also been proposed to stimulate thalamic axon outgrowth [22] . This study provided evidence for a decreased number of thalamic axons invading the VTel as well as a disorganized internal capsule , using DiI tracing and L1 staining . Despite careful examination , we did not find evidence of a significant decrease in the number of thalamic axons in the VTel of Netrin-1−/− embryos compared to control ( see Figure S4 ) , and the longest DTh axons' length was not significantly altered when axons were growing in Netrin-1−/− or control VTel ( Figure S8 ) . Furthermore , in our whole-mount telencephalic assay , wild-type DTR axons grew equally well in a wild-type or a Netrin-1–deficient VTel ( Figure 5 ) , suggesting that Netrin-1 does not play a critical role in the stimulation of thalamic axon outgrowth in vivo . The potential discrepancies between our results and the study by Braisted et al . ( 2000 ) [22] could be due to methodological differences or to differences in the genetic background of the Netrin-1 knockout mice between the two studies . As a precedent , mice presenting a null mutation in the Netrin-1 receptor Unc5C on the inbred C57BL/6J ( B6 ) genetic background display abnormal projections of both trochlear nerve and motor neuron axons , but these defects are greatly attenuated on a hybrid B6 × SJL background [52] . The authors have provided evidence for a locus representing a genetic suppressor of Unc5C function on mouse chromosome 17 [52] . In mammals , there are at least five genes encoding transmembrane receptors for Netrin-1: DCC ( Deleted in Colorectal Cancer ) and Neogenin receptors mediate the attractive response elicited by Netrin-1 , whereas Unc5A–C family members mediate repulsion elicited by Netrin-1 either as homodimers or heterodimers with DCC [53 , 54] . Our current results show an interesting regionalization of Netrin-1 receptor expression along the rostromedial to caudolateral axis of the DTh . At E14 . 5 , when the topography of TC axons is initiated in the VTel , but before individual thalamic nuclei are formed , DCC is expressed at high levels in a rostromedial domain of the DTh , whereas Unc5A , Unc5B , and Unc5C are expressed in largely nonoverlapping caudolateral domains of the DTh ( see Figure 7 ) . Interestingly , Unc5C expression pattern is more widespread than Unc5A and B , and seems to overlap at least partially with the rostromedial domain of DCC expression ( see Figure 7 ) . Our function-blocking experiments demonstrate that DCC is required for the guidance of rostral thalamic axons to the Netrin-1–rich rostral domain of the VTel , whereas DCC and Unc5C are required for the proper repulsion of caudal thalamic axons to the same Netrin-1–rich region . A recent study has implicated ephrin-A5–EphA4 signaling in the initiation of the topography of TC axon projection in the VTel [11] . Three EphA receptors ( EphA4 , A3 , and A7 ) were shown to be expressed in high-rostromedial to low-caudolateral gradients in the E14 . 5 DTh , whereas the ephrin-A5 ligand was found to be expressed in a high-caudal to low-rostral gradient in the VTel ( Figure 11A ) . This study also provided in vivo and in vitro functional evidence demonstrating that both ephrin-A5 expression in the VTel and EphA4 expression in the DTh were required for the proper topographic projection of thalamic axons [11] . Taken together , our results and those of Dufour et al . ( 2003 ) [11] suggest that rostromedial thalamic neurons express high levels of DCC and EphA receptors conferring to their axons both attractive responsiveness to rostral Netrin-1 and repulsive responsiveness to caudal ephrin-A5 , respectively , resulting in repulsion from the caudal domain and attraction to the rostral domain of the VTel . Our DCC function-blocking experiments demonstrate that DCC is required for the attraction of rostral thalamic axons to the rostral domain of the VTel . In contrast , progressively more-caudolateral thalamic neurons express lower levels of EphA receptors and higher levels of Unc5A–C receptors , and we demonstrate that this decreased sensitivity to the repulsive effect of ephrin-A5 is accompanied by an increasing sensitivity to the repulsive action of rostral Netrin-1 . Future experiments will determine whether Netrin-1 is playing this function in large-scale mapping of ensembles of axons in other projection systems .
Mice were used according to a protocol approved by the Institutional Animal Care and Use Committee at the University of North Carolina-Chapel Hill , and in accordance with National Institutes of Health guidelines . Time-pregnant females were maintained in a 12-h light/dark cycle and obtained by overnight breeding with males of the same strain . Noon following breeding is considered as E0 . 5 . Netrin-1 knockout mice ( Ntn1Gt ( pGT1 . 8TM ) 629Wcs , abbreviated Ntn1LacZ ) were generated by crossing between heterozygous mice [21 , 55] . The initial line was on a mixed C57Bl6 and Sv129 background , and was backcrossed for more than ten generations on BALB/c background ( Jackson Laboratories ) . Genotyping of Netrin-1LacZ mice was performed by the University of North Carolina genotyping core facility using quantitative PCR detecting the presence of zero , one , or two copies of the lacZ transgene . The genotype of embryos heterozygote or homozygote for the Netrin-1 transgene was confirmed by anatomical defects described previously ( absence of callosal and anterior commissure projections ) [21] . Transgenic mice expressing EGFP under the control of CMV enhancer/chicken β-actin promoter were maintained by heterozygous crossing on a Balb/C background for more than ten generations [56] . Briefly , isolated hemispheres from E14 . 5 to E18 . 5 mouse embryos were microinjected ( PicoSpritzer III; General Valve Corp . ) using a medial approach with a 10% solution of lysine-fixable BDA ( 3 , 000 MW ) . Following incubation in oxygenated artificial cerebrospinal fluid ( aCSF ) for 5 h at 37 °C , the hemispheres were immersion fixed in 4% PFA . Injected hemispheres were sectioned coronally using a vibratome ( LEICA VT1000S ) at 100-μm thickness , permeabilized , and then incubated with Alexa546-conjugated streptavidin ( 1:1 , 000 in PBS + 0 . 1% Triton X-100 + 0 . 3% BSA ) to reveal BDA . ( See Figure S1 . ) Sense and antisense probes for mouse Netrin-1 , DCC , Unc5A , Unc5B [27 , 57] , and Unc5C [31 , 58] were generated as described previously . In situ hybridizations were performed as previously described using DIG-labeled probes [11] . Whole-mount telencephalon/dorsal thalamic cocultures were maintained on organotypic slice culture inserts , fixed , and stained for immunofluorescence as previously described [10 , 59] . The following primary antibodies were used: polyclonal rabbit anti-β-galactosidase ( 1:1 , 000; Molecular Probes ) , monoclonal antineurofilament 165kD ( clone 2H3; 1:2 , 000; Developmental Hybridoma Bank ) , polyclonal rat anti-L1 cell adhesion molecule ( 1:1 , 000; Chemicon ) as well as polyclonal chicken and rabbit anti-GFP ( 1:2 , 000; Molecular Probes ) . The following secondary antibodies were used: Alexa-488 , −546 , and −647 conjugated goat anti-chicken , anti-rabbit , or anti-mouse IgG ( 1:2 , 000; Molecular Probes ) . Full-length cDNA clones of mouse Unc5A , Unc5B , and Unc5C ( Image ID: Unc5A , 6813463; Unc5B , 6417563; and Unc5C , 40085998 ) were purchased from Open Biosystems . The open reading frame of each clone was amplified by PCR using LA Taq polymerase ( TAKARA BIO ) and cloned with a Myc-tag at the carboxy-terminus into a modified pCIG2 vector , which drives expression of cloned cDNA from chicken ß-actin promoter and CMV enhancer . All constructs were confirmed by DNA sequencing . COS7 cells were transfected with green fluorescent protein ( GFP ) and Unc5A-Myc , Unc5B-Myc , or Unc5C-Myc expression vectors using Lipofectamine 2000 ( Invitrogen ) . Cells were lysed in RIPA buffer ( 50 mM Tris-Hcl [pH 7 . 5] , 150 mM NaCl , 1% Triton X-100 , 0 . 1% SDS ) supplemented with protease inhibitors ( Complete Protease Inhibitors Cocktail Tablets; Roche ) 48 h after transfection . Protein samples ( 20 μg each ) were run on 4%–12% gradient SDS-PAGE . gels ( Invitrogen ) and transferred to Hybond-P membranes ( Amersham ) . Membranes were preincubated in 5% nonfat dry milk and 0 . 1% Tween-20 in Tris-buffered saline and incubated with the primary antibodies in the same solution . The primary antibodies used were mouse anti-Myc antibody ( clone 9B11 , 1:2 , 000; Cell Signaling Technology ) , goat anti-Unc5A antibody ( anti-rat Unc5h1 , 1:200; R&D Systems ) , and rabbit anti-GFP antibody ( IgG fraction , 1:2 , 000; Molecular Probes ) . The fluorescent signals were generated using corresponding HRP-conjugated secondary antibodies and ECL-Plus Western Blotting Detection Reagents ( Amersham ) , and were detected using a Typhoon 9400 image scanner ( Amersham ) . To visualize axons of dorsal thalamic neurons , we transfected myristoylated-Venus ( mVenus ) plasmid ( pCX-myrVenus , kindly provided by Anna-Katerina Hadjantonakis , 1 μg/μl ) . For Unc5 overexpression , the mixture ( 0 . 5 μg/μl each ) of Unc5A-Myc , Unc5B-Myc , and Unc5C-Myc ( unpublished data ) or Unc5C-Myc alone ( 1 μg/μl ) was transfected together with mVenus plasmid ( 1 μg/μl ) . Control experiments included transfection of mVenus plasmid alone or the mixture of mVenus and Myc-tag empty plasmids . These two conditions gave similar results , and therefore , we combined them as control . Electroporation into slices was performed essentially as described previously [37] . Briefly , coronal slices ( 250 μm ) of E14 . 5 wild-type brains were prepared using a Leica VTel 1000S vibratome , and plasmid solution was pressure injected through a glass pipette into the rostral or caudal DTh ( DTR or DTC; see Figure 10 ) using a Picrospritzer III ( General Valve ) microinjector . Electroporations were performed with gold-coated electrodes ( GenePads 5 × 7 mm; BTX ) using an ECM 830 electroporator ( BTX ) and the following parameters: five 5-ms–long pulses separated by 500-ms intervals at 100 V . After electroporation , the DTh was dissected and cocultured with E14 . 5 wild-type whole-mount telencephalon for 4 days , followed by fixation and immunostaining for GFP as described previously [10] . To identify GFP-positive axons , we used either of the two methods based on the axon density in each sample . When the number of axons in the sample was small , as shown in Figure 10D , we identified axons by single-axon tracing using the NeuronJ plug-in for ImageJ . Traced axons were superimposed on a reference framework with common origin , and the percentage of pixels in caudal , medial , or rostral 60° radial bin was quantified . When axon density in the sample was too high to identify individual axons , we quantified the percentage of EGFP-pixel distribution in the same way as in single-axon tracing . Thus , the two methods utilize essentially identical quantification analyses , and we combined the results obtained by using these two methods . Fluorescent immunostaining was observed using a LEICA TCS-SL laser scanning confocal microscope equipped with an Argon laser ( 488 nm ) , green helium-neon laser ( 546 nm ) , and red helium-neon laser line ( 633 nm ) mounted on an inverted DM-IRE2 microscope equipped with a Marzhauzer X-Y motorized stage allowing large-scale tiling of whole-mount telencephalic cocultures obtained by scanning multiple fields using a long working distance 10× objective followed by an automatic tiling function available from the LEICA confocal software .
|
The functional properties of each structure in the central nervous system are critically dependent on the precision of neuronal connectivity . The cerebral cortex in particular is a highly organized structure divided into many distinct cortical areas underlying important sensory , motor , and cognitive functions in the brain . Each primary cortical area receives its synaptic inputs from the periphery via the dorsal thalamus . The main relay station for sensory information to the cortex , the thalamus , can be divided into specific nuclei projecting topographically to individual cortical areas . How is the complex topography of thalamic axon projection to individual cortical areas specified during development ? Recent evidence demonstrated that thalamic axons are routed to different cortical domains before they enter the cortex , by putative axon guidance cues present in the ventral forebrain . In the present study , we provide evidence that a secreted axon guidance cue , Netrin-1 , expressed in a long-range gradient in the ventral forebrain , plays a critical role in the establishment of the topography of thalamic projections by directing different subsets of axons to specific cortical domains . These results provide important insights into the molecular mechanisms responsible for shaping the topographical patterns of thalamocortical axon projections in mammals .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"neuroscience"
] |
2008
|
Topography of Thalamic Projections Requires Attractive and Repulsive Functions of Netrin-1 in the Ventral Telencephalon
|
Over the last decades , researchers have characterized a set of “clock genes” that drive daily rhythms in physiology and behavior . This arduous work has yielded results with far-reaching consequences in metabolic , psychiatric , and neoplastic disorders . Recent attempts to expand our understanding of circadian regulation have moved beyond the mutagenesis screens that identified the first clock components , employing higher throughput genomic and proteomic techniques . In order to further accelerate clock gene discovery , we utilized a computer-assisted approach to identify and prioritize candidate clock components . We used a simple form of probabilistic machine learning to integrate biologically relevant , genome-scale data and ranked genes on their similarity to known clock components . We then used a secondary experimental screen to characterize the top candidates . We found that several physically interact with known clock components in a mammalian two-hybrid screen and modulate in vitro cellular rhythms in an immortalized mouse fibroblast line ( NIH 3T3 ) . One candidate , Gene Model 129 , interacts with BMAL1 and functionally represses the key driver of molecular rhythms , the BMAL1/CLOCK transcriptional complex . Given these results , we have renamed the gene CHRONO ( computationally highlighted repressor of the network oscillator ) . Bi-molecular fluorescence complementation and co-immunoprecipitation demonstrate that CHRONO represses by abrogating the binding of BMAL1 to its transcriptional co-activator CBP . Most importantly , CHRONO knockout mice display a prolonged free-running circadian period similar to , or more drastic than , six other clock components . We conclude that CHRONO is a functional clock component providing a new layer of control on circadian molecular dynamics .
Circadian rhythms are ubiquitous in daily life , coordinating the sleep–wake cycle along with oscillations in hormone secretion , blood pressure , and cognitive function [1] , [2] . While a central master-pacemaker is located in the suprachiasmatic nuclei ( SCN ) of the hypothalamus , cell autonomous rhythms are generated throughout the body . The CLOCK/BMAL1 transcriptional complex lies at the core of the molecular clock . These proteins bind E-box elements in the promoters of target genes [3] . The Period and Cryptochrome gene families are prominent among these targets , and their products ultimately repress CLOCK/BMAL1 activity and their own transcription [4] , [5] . A second loop regulates Bmal1 expression through the opposing actions of the REV–ERB and ROR nuclear receptor protein families [6] , [7] . Circadian oscillations are in turn subject to multiple layers of control . The casein kinase I proteins ( CSNK1D and CSNK1E ) and the F-box and leucine-rich repeat proteins ( FBXL3 , FBXL21 ) [8]–[10] regulate the nuclear accumulation and/or stability of clock components , respectively . Moreover , recent evidence highlights the importance of metabolic cofactors and histone modifiers ( e . g . , HDAC3 , P300 , CBP , SIRT1 , and NAMPT ) in modulating these feedback loops . The understanding of circadian timekeeping has demonstrated far-reaching importance . Allelic variation in clock components has been associated with circadian , sleep , and mood disorders [8] , [11]–[13] . Mutational and epidemiologic studies have linked clock genes with neoplastic and metabolic phenotypes [2] , [14] . However , the current model of the circadian pacemaker is likely incomplete . Indeed , quantitative circadian trait analysis maps most loci to regions unassociated with known clock genes [15] . In an attempt to identify these missing regulatory components , researchers have moved beyond the costly and laborious mutagenesis screens that identified the first clock components [16] , [17] . Recent studies have turned to higher throughput genomic and proteomic approaches . A screen for activators of BMAL1 transcription [7] , a screen for proteins that bind CLOCK [18] , and proteomic analysis of the BMAL1 [19] and PERIOD [20] , [21] protein complexes have all identified proteins that function in circadian control . Here we present an alternative , computer-assisted approach aimed at accelerating clock gene discovery . We used probabilistic machine learning to integrate heterogeneous , genome-scale datasets [22]–[24] and identify candidate clock genes that functionally resemble known clock components . We screened the top candidates for physical interactions with a subset of clock components using a mammalian two-hybrid assay . Candidates were further screened for circadian function in an in vitro system . We focused our attention on three promising initial candidates . Here we demonstrate the utility of this approach with data from the first of these candidates , Gene Model 129 ( Gm129 ) , to have its circadian function characterized in both cells and knockout mice . We confirmed that Gm129 physically interacts with core clock genes and regulates the molecular oscillator . In addition , Gm129 oscillates in multiple tissues , functionally represses the activity of the CLOCK/BMAL1 transcriptional complex , and most importantly , influences the free-running circadian period of locomotor activity in mice . In view of its role as a computationally highlighted repressor of the network oscillator , we have renamed the gene Chrono .
Using the above empirical distributions and a modified version of the Naïve Bayes learning algorithm , we quantified the evidence provided by each feature that a given gene is a member of the core circadian network [22] , [32] . We relied on the prior assumption , informed by experimenter judgment , that increasing possession of each of these features lends increasing evidence of a role in the circadian clock . We used the empirical cumulative distribution function ( ECDF ) describing exemplar “clock genes” or “nonclock genes” to estimate the probabilities that a randomly selected “clock gene” or “nonclock gene” would possess a metric value at least as extreme as the one observed . We term the ratio of these probabilities a “circadian evidence factor” ( Materials and Methods , Eq . 4 ) . The evidence factors arising from particular features and metric value are shown in the right panels of Figure 1B–E ) . The evidence amassed from all five features is encapsulated by a “combined evidence factor . ” Computation of combined evidence factors requires knowledge of the joint cumulative probability distributions for these features among both “clock genes” and “nonclock genes . ” These joint cumulative distribution functions are “learned” from the examples under the “Naïve” assumption of conditional independence . Evidence factors from each individual feature are multiplied to calculate the combined evidence factor ( Materials and Methods , Eq . 6 ) . This approach differs from the standard Naïve Bayes statistical learning approach only in that cumulative distribution functions are used rather than probability density functions . We ranked genes based on this combined evidence . The top 20 candidates ( Figure 1F ) include 10 of the exemplar clock components along with Tef [33] and Nfil3 [34] , two genes with established circadian functions . Moreover , Wee1 , a canonical cell cycle gene , is known to be regulated by the circadian clock [35] . Although , to our knowledge , the hypothesis that Wee1 directly regulates clock function has not been tested . Inspecting the top 50 ranked genes , several other genes known to be involved in the circadian clockworks appear . These include Dbp [36] , Insig2 [37] , and Nampt [38] . In order to evaluate the utility of this ranking in the discovery of novel clock genes , we applied 10-fold cross-validation . We sequentially removed all possible pairs of clock components from the exemplar distribution , ignoring our prior knowledge of their role in orchestrating circadian rhythms . In each case , we then recomputed the combined evidence factors based on this reduced knowledgebase and tested our ability to “rediscover” these clock genes using different ranking cutoffs . Based on this analysis , we estimate that ∼50% of true clock components would be recovered by screening the top 50 genes ( Figure S1A ) . We also compared the use of evidence factors with two prepackaged machine learning algorithms . Using the same features , we ranked genes using a Gaussian Naïve Bayes classifier and a Flexible Naïve Bayes classifier [39] . The three methods all yield comparable performances using cutoffs less than ∼1 , 000 , but the evidence factor method outperforms the other two beyond this point . Importantly , the top candidates from all three methods show a very high degree of overlap ( Figure S1B ) . Only rankings from the evidence factor approach were used in selecting genes for further screening . However , results from all three probabilistic learning methods are presented in the Supporting Information section . The cycling feature makes the largest single contribution to the combined evidence factors , but it does not completely dominate this ranking . Hundreds of genes demonstrate strong cycling in the tissues analyzed and other features determine the relative ranking among these . Moreover , some candidates , like Hdac11 , are largely prioritized based on the combined strength of other features . Given the rarity of bona fide clock genes , any method that is not 100% specific will result in a number of false positives . As the ranking cutoff is increased , the number of nonclock genes incorrectly identified will also increase . As in other screening applications where one is searching for a “needle in a haystack , ” a secondary validation of candidates is needed . Assuming different numbers for the total number of core clock components , we estimated the false positive rate for different screening cutoffs ( Figure S1C ) . The ultimate value of this approach will be determined by its ability to identify previously unrecognized clock components . We tested the top 25 novel candidates for physical interactions with a subset of proteins from the negative arm of the molecular clock ( BMAL1 , BMAL2 , CLOCK , NPAS2 , CRY1 , CRY2 . PER1 , PER2 , and PER3 ) . Three of these candidates ( Gm129 , Ifitm1 , and Cbs ) demonstrated both physical binding with at least one of the included clock components and a statistically significant change in circadian reporter period after knockdown in the NIH 3T3 model system ( Figure S2 ) . Of note , although Gm129 might have been identified simply by its strong cycling , Cbs and Ifitm1 are identified by virtue of a combination of features . Bellow we present a more detailed investigation of the previously uncharacterized candidate , Gm129 , here renamed Chrono . These data show that Chrono meets the formal definition of a mammalian circadian clock gene . Our previous microarray data suggested that Chrono expression cycles with a 24-h period in liver , pituitary , and NIH 3T3 cells [25] , [40] . We used quantitative PCR ( qPCR ) to confirm cycling in the liver and further evaluated transcript cycling in skeletal muscle and white fat ( Figure 2 ) . The circadian oscillations in Chrono expression are of a similar magnitude to those observed for known clock factors Nr1d1 and Per2 . Consistent with our results , temporal profiling in rat skeletal muscle [41] and lung [42] , as well as mouse SCN [43] , also revealed daily oscillations in Chrono expression . Several genome-wide , ChIP-seq studies in mouse liver [43]–[45] have identified the E-boxes in the Chrono gene promoter among those genomic regions most tightly bound by BMAL1 protein . Time course microarray studies from SCN and liver demonstrate that Chrono expression is reduced in Clock mutant animals and loses circadian rhythmicity ( Figure S3A ) [46] , [47] . Moreover , Chrono expression becomes arrhythmic in the livers of Cry1/Cry2 double knockout animals ( Figure S3B ) [48] . In total , Chrono demonstrates robust circadian expression in multiple tissues and appears to be directly regulated by the molecular clock . We employed a mammalian two-hybrid screen to identify physical interactions between CHRONO and a subset of known clock components . As expected , many core clock proteins physically interacted , as indicated by specific activation of a UAS:Luc reporter in transfected Human Embryonic Kidney 293 cells containing the SV40 T-Antigen ( HEK 293T ) ( Figure 3A , Table S1 ) . Interactions between CHRONO and both BMAL1 and PER2 were also observed , with >20-fold induction of luciferase activity . BMAL1–CHRONO and PER2–CHRONO complex formation were confirmed through co-immunoprecipitation ( co-IP ) ( Figure 3B and C ) . Bi-molecular Fluorescence Complementation ( BiFC ) using Venus , an enhanced yellow fluorescent protein ( YFP ) , was then used to map BMAL1/CHRONO interactions to cell nuclei ( Figure 3D ) . Notably , when S-tagged CHRONO was overexpressed with both BMAL1 and CLOCK BiFC fusion proteins , CHRONO appeared to colocalize with the CLOCK/BMAL1 heterodimer in nuclear bodies , suggesting that CHRONO continues to interact with BMAL1 while part of this functional circadian complex . To evaluate the functional consequences of these physical interactions , we monitored Per1:luciferase activity in unsynchronized HEK 293T cells transiently transfected with Clock/Bmal1 . Per1:luc reporter activity is enhanced by Clock/Bmal1 transfection but repressed by the overexpression of either Cry1 or Chrono ( Figure 3E ) . As has been previously demonstrated , CLOCKH360Y and BMAL1G612E missense mutants are resistant to CRY-mediated repression [49] . In contrast , CHRONO-mediated repression is unaffected by these point mutations ( Figure 3E ) . The same pattern was observed in the expression of Nr1d1 , an endogenous CLOCK/BMAL1 target ( Figure S4 ) . Alternatively , CHRONO knockdown augments Per1:luc reporter activity ( Figure 3F ) . These data suggest that CHRONO and CRY1 have distinct binding sites and/or functional mechanisms . Small interfering RNA ( siRNA ) mediated knockdown of C1orf51 , the human homologue of Chrono , markedly dampened circadian oscillations in a genome-wide circadian screen [28] . Using NIH 3T3 cells expressing a Bmal1:dLuc reporter as a second model system , we tested the effects of four different short hairpin RNA ( shRNA ) constructs that reduced Chrono transcript expression and protein abundance ( Figure S5 ) . Comparing the pooled results to control demonstrates that Chrono knockdown reduces amplitude and increases circadian period ( Figure 4A–F ) . To definitively establish the role of CHRONO in modulating circadian behavior , we obtained transgenic mice from the Knockout Mouse Project [50] . These mice incorporate a transgenic construct ( Figure S6A ) whereby the Chrono encoding region is flanked by Lox-P sites ( Chronoflx/flx ) and utilizes a “knockout-first” cassette [51] . The transgenic allele is a knockout at the level of RNA processing . We mated heterozygous transgenic mice to obtain homozygous Chrono knockout mice ( Chronoflx/flx ) , wild-type littermate controls ( Chrono+/+ ) , and heterozygotes ( Chronoflx/+ ) . qPCR confirmed that , when compared to wild-type littermate controls , mRNA expression was halved in heterozygotes ( Chronoflx/+ ) and abolished to basal levels in homozygote knockouts ( Chronoflx/flx ) ( Figure S6B and C ) . As shown in Figure 4G and H , wild-type , heterozygous , and homozygous knockouts were all well entrained to the 12∶12 light∶dark ( L∶D ) cycle and maintained a 24-h period . Under free-running conditions , homozygous Chrono knockouts exhibited a statistically significant ( p<0 . 05 ) ∼25-min increase in circadian period as compared to wild-type controls ( Figure 4I ) . Heterozygous knockouts display an intermediate period . The magnitude of this period change is similar to that observed in Clock ( ∼20 min ) [52] , Per1 ( ∼40 min ) [53] , Per3 ( ∼30 min ) [54] , Nr1d1 ( ∼20 min ) [6] , Rorb ( ∼25 min ) [55] , and Npas2 ( ∼12 min ) [56] knockout animals . These data strongly suggest that endogenous Chrono expression plays an important regulatory role in the mammalian circadian clock . Light , however , does not appear to directly influence CHRONO expression in the SCN . The SCN microarray data of Jagannath et al . does not reveal a significant change in Chrono expression following a nocturnal light pulse [57] . Moreover , in our own experiments , the phase shifting response of Chrono knockout mice to light pulses at ZT16 or ZT22 are not significantly different from control . Thus the primary role of CHRONO in the circadian clock appears to be in modulating core oscillator function and output timing rather than oscillator entrainment . In a recent report , BMAL2 was shown to function as a tissue-specific paralogue of BMAL1 [58] . However , CHRONO specifically binds BMAL1 and not BMAL2 ( Figure 3A ) . Moreover , CHRONO functionally represses the transcriptional activity of the BMAL1/CLOCK complex but not the activity of the BMAL2/CLOCK complex ( Figure 5A ) . In order to identify the region of BMAL1 required for CHRONO binding , we generated mutant BMAL1 proteins with truncated N- or C-terminal regions ( BMAL178–626 , BMAL11–445 ) ( Figure 5B ) and tested their interaction with CHRONO using the mammalian two-hybrid assay . Deletion of the C-terminal domain of BMAL1 ( BMAL11–445 ) completely abolished CHRONO binding , whereas deletion of the N-terminal domain ( BMAL178–626 ) had no effect ( Figure 5C ) . We next exploited the strong sequence homology between BMAL1 and BMAL2 to localize the CHRONO binding site within the BMAL1 C-terminal region . We swapped corresponding sections of the BMAL1 and BMAL2 C-terminal domains . As expected , the construct containing the N-terminal of BMAL1 and the full C-terminal of BMAL2 ( BMAL1–BMAL2 ) did not interact with CHRONO in the two-hybrid assay and was relatively immune to CHRONO-mediated repression ( Figures 5C–E ) . A chimeric protein including the N-terminal region of BMAL2 with the longer BMAL1 C-terminus ( BMAL2–BMAL1#1 ) interacted with CHRONO and phenocopied wild-type BMAL1 with regard to CHRONO-mediated repression ( Figure 5D–F ) . Sequence alignment between C-terminal domains of BMAL1 and BMAL2 reveals a region of poor alignment ( 514–594 ) . Insertion of this unique region of the BMAL1 protein ( 514–594 ) into BMAL2 C-terminus rendered the chimeric protein ( Bmal2–Bmal1#2 ) responsive to CHRONO-induced repression ( Figure 5E–F ) . This CHRONO binding region is adjacent to , but distinct from , the CRY1 interacting terminus [59] . Thus , CHRONO functions as a specific transcriptional co-repressor of BMAL1 through interaction with a unique C-terminal domain adjacent to the CRY1 binding region . This domain is both necessary and sufficient for physical and functional interactions with CHRONO . Previous studies suggested that CBP also binds to the BMAL1 C-terminus [59] , [60] . Thus , we hypothesized that CHRONO might interfere with BMAL1–CBP binding . We generated plasmids encoding BMAL1 and CBP fused to the C- and N-terminal regions of the Venus YFP . We then utilized BiFC to visualize BMAL1–CBP interactions in HEK 293T cell nuclei . BMAL1–CBP complex formation induced a yellow BiFC signal ( Figure 6A ) . Co-expression of native or S-tagged CHRONO severely dampened BMAL1–CBP complementation . Western blotting ( Figure S7A ) confirmed stable abundance of BMAL1 and CBP proteins , implicating altered binding as the source of the reduced BiFC signal . Lastly , the ability of CHRONO to interfere with BMAL1–CBP binding was verified by co-IP analysis showing that overexpression of intact CHRONO reduced BMAL1–CBP complex formation ( Figure 6B ) . A functional impairment in the ability of BMAL1 to recruit CBP is expected to reduce histone acetylation of CLOCK/BMAL1 target regions . To assess the influence of CHRONO on the histone acetyl-transferase activity of the BMAL1/CLOCK complex , we performed a ChIP study using an antibody targeting acetylated histone H3 lysine 9 ( H3–K9 ) ( Figure 6C ) . PCR was used to specifically evaluate H3–K9 acetylation near the Per1 promoter E-box . Control samples obtained from immortalized human osteosarcoma ( U2OS ) cells 24 and 36 h after dexamethasone synchronization demonstrated a temporal variation in target acetylation . U2OS cells overexpressing CHRONO demonstrated a blunted temporal profile in Per1 promoter H3–K9 acetylation , with loss of the increased acetylation normally observed 24 h after synchronization [61] , [62] . In order to confirm that abrogated CBP/BMAL1 binding contributes to the CHRONO-mediated modulation of circadian dynamics , we constructed several CHRONO truncation mutants . All constructs that retained the 108–212 region reduced BMAL1–CBP binding as assessed by BiFC ( Figure 6D and E ) . As has been previously demonstrated , overexpression of CBP , along with BMAL1 and CLOCK , enhances Per1:luc expression in unsynchronized cells ( Figure 6F ) . Those same CHRONO constructs that abrogated BMAL1–CBP complex formation also repressed CBP-enhanced Per1:luc reporter activity ( Figures 6E and F and S7B ) . This pattern of activity among CHRONO truncation mutants was further mirrored in their ability to colocalize with BMAL1 ( Figure S7C ) . Stable expression of the constructs in synchronized cells reveals the same pattern in their ability to modulate circadian reporter expression ( Figure S7D and E ) . Thus , the abrogation of the BMAL1–CBP binding provides a plausible mechanism whereby CHRONO might influence circadian dynamics . In summary , our data demonstrate that Chrono ( i ) oscillates with a circadian frequency in multiple tissues , ( ii ) physically interacts with BMAL1 and PER2 , ( iii ) specifically reduces BMAL1/CLOCK-mediated transcription independently of CRY1 , ( iv ) affects the free-running circadian period of mice , and ( v ) interferes with BMAL1–CBP binding , functionally repressing the CLOCK/BMAL1 complex and modulating the circadian acetylation of target genes . Most importantly , CHRONO knockout mice display a long free-running circadian period similar to or more drastic than six other clock components . These data establish a role for Chrono in the mammalian circadian oscillator . Like CIPC [18] , CHRONO appears unique to the vertebrate genome . Given the repressive function of CRY proteins , the evolutionary development of additional CLOCK/BMAL1 repressors in vertebrates highlights the importance of fine control of circadian rhythms . Transcriptional oscillations can differ in their amplitude , frequency , phase , basal expression , and waveform shape . The ability to independently control these characteristics likely requires multiple , tunable genetic parameters . The specificity of CHRONO-mediated repression for BMAL1 over BMAL2 , along with tissue-specific variation in the expression of BMAL1 and BMAL2 , may thus facilitate local tuning of circadian oscillations . Of course , there remain important , unanswered questions with regard to the function of CHRONO in modulating circadian dynamics . Although the abrogation of BMAL1–CBP is a plausible mechanism for CHRONO-mediated repression , it may reflect only part of its circadian function . Moreover a nuanced understanding of how this repression leads to a period-lengthening phenotype in the knockout animal will likely require a greater understanding of kinetics and network compensation . Our BiFC data demonstrate that overexpressed CHRONO co-localizes with the CLOCK/BMAL1 complex in nuclear bodies . It was previously shown that BMAL1 recruits CBP primarily when localized to promyelocytic leukemia ( PML ) nuclear bodies [63] . Thus the interruption of the CBP–BMAL1 binding within these nuclear structures is consistent with the potent repression induced by CHRONO overexpression ( Figure 3E ) . Indeed , while this work was in revision , Annayev et al . [64] also reported that CLOCK/BMAL1 transcription is efficiently repressed by CHRONO ( GM129 ) . Our experimental work adds both a description of the circadian locomotor phenotype of the Chrono knockout mouse and an understanding of the mechanism by which this repression is mediated . Although our work focused on the interaction between CHRONO and BMAL1 , CHRONO might also influence circadian physiology through its interaction with PER2 . It was recently reported that PER2 also localizes to PML nuclear bodies [65] . The importance of CHRONO/PER2 binding ( Figure 3A , C ) , both within this complex and more generally , remain unexplored . PER2 not only binds with cryptochromes but also interacts with nuclear receptors NR1D1 and RORA [66] . Our preliminary tests ( Figure S8 ) show that overexpression of CHRONO enhances the PER2/NR1D1 complex formation . The recruitment of this established circadian repressor provides another mechanism for CHRONO-enhanced repression of the circadian network . The importance of CHRONO/PER2 binding and a broader analysis of the role of CHRONO in the circadian network will require further study . The extent to which CLOCK can recruit CBP/P300 independently of BMAL1 also remains unclear [67] . Given the highly redundant structure of the circadian oscillator [68] , the ability of CLOCK to recruit a co-activator hints that there may be a functional paralogue of CHRONO acting on the other half of the BMAL1/CLOCK complex . Perhaps most importantly , the knockout and targeted disruption of several other clock factors have been shown to not only influence circadian period but also downstream physiological changes in metabolism [69] and sleep homeostasis [70] , [71] . More detailed phenotyping of CHRONO knockout mice will be required to identify any such deficits . Machine learning has recently been applied to complex biological problems including drug discovery [72] , protein translation [73] , and gene interaction networks in yeast [74] . We used a simple form of probabilistic machine learning to integrate sparse existing data whose joint distribution is hypothesized to yield a more specific ranked list of candidate genes . Although follow-up experimentation is an important part of this process , the identification of Chrono reflects the ability of this approach to find genes regulating circadian behavior . To our knowledge , this is the first application of these methods to identify genes responsible for complex neurological behaviors . We anticipate that the investigation of other candidates will advance the understanding of circadian rhythms . Indeed , in addition to CHRONO , our initial screening of the top 25 novel candidates identified two other proteins that both bind clock components and modulate in vitro circadian oscillations . To facilitate the experimental characterization of these and other candidates , a more exhaustive candidate ranking is provided in Table S2 . As bona fide clock components are discovered and high-quality datasets become available , exemplar distributions can be re-evaluated and feature metrics can be improved . Thus , this integrated computational and experimental approach presents a path for leveraging genome scale data to develop insight into circadian biology .
All animal experiments were performed with the approval of the Institutional Animal Care and Use Committee ( IACUC Protocol Numbers 801906 and 803945 ) . Unless otherwise specified , all computations were done in the R programming environment [75] . The derivation of circadian evidence factors closely follows that for Bayes factors [32] , and our strategy follows the Naïve Bayes Classifier approach of “learning” the feature distributions from the training data . We considered an individual feature described by metric , and a single arbitrary gene with observed metric value . The event space is divided in two disjoint events: and . These events correspond to a randomly selected gene having a metric value at least as extreme as or the randomly selected gene having a metric value less than . The events are labeled and , respectively . The use of an interval rather than a point allows us to regularize the sparse empirical data for the estimation . Each gene is assumed to belong to either the set of clock genes ( Cgene ) or the set of nonclock genes ( NCgene ) . By Bayes' Theorem: ( E1 ) and ( E2 ) Dividing ( E1 ) by ( E2 ) yields: ( E3 ) Substituting the definition of , the middle term of ( E3 ) becomes: ( E4 ) The left-hand side of ( E3 ) is the posterior odds of a gene being a core clock component conditional on observing a metric value greater than or equal to . The last term represents the general odds of clock gene membership without additional feature information . Thus , the posterior odds of a gene belonging to the set of clock genes ( given a metric value greater than or equal to ) is equal to the product of and the a priori odds . Our analysis included n = 5 clock gene features . For each metric , the event space is divided into two disjoint events— and for some —and these events are labeled and , respectively . Following the steps above: ( E5 ) The middle term in equation ( E5 ) is the factor by which the a priori odds of clock gene membership must be adjusted to recover the posterior odds after all of the observed data . It represents the combined evidence factor ( ) given all five features . Given the number of features , the training set of circadian clock components is too sparse to approximate the required joint distribution without some regularizing assumption . We follow the typical Naïve Bayes approach and show that , given conditional independence of the included features , is simply the product of the individual evidence factors . By definition , random variables with probability density functions and joint probability density function are conditionally independent given a random variable if and only if: Using this definition and the definition of the events , the denominator of can be simplified: Similarly , the numerator of becomes , and the cumulative evidence is equal to: ( E6 ) Given either the distribution of metric values among exemplar clock genes or the distribution among the genome at large , the probabilities of obtaining a metric greater than , or equal to , that observed was approximated with the ecdf ( ) function in R . For any given gene and feature , the ratio of these probabilities was computed to obtain the value of . Metric values greater than the maximum value observed among exemplar clock components were assigned the same evidence factor as that maximum value . Combined evidence factors are the product of the feature-specific factors . If no data were available for a given gene and feature , this feature was ignored by setting the corresponding evidence factor to be 1 . The ubiquity and homology metrics were both Boolean variables , and the standard Bayes factor formula was used for these features . The use of combined evidence factors was compared with two prepackaged , supervised machine learning algorithms in the R programming environment: a Gaussian/Normal Naïve Bayes classifier within the “e1071” package [80] and a Flexible Naïve Bayes classifier [39] within the “klaR” package [81] . Probabilistic learning algorithms were preferred as they do not require a prior weighting of the importance of the various features [24] . For training , genes not in the exemplar clock group were labeled as nonclock genes , and the classifier was trained on the entire dataset . Genes were rank ordered on the posterior probability of clock gene membership after the model was applied to the data . For the Flexible Naïve Bayes implementation , kernel density estimation was performed with the default value for the “window parameter . ” This default uses a heuristic formula to adjust the window of kernel density estimate based on the number of data points . We sequentially removed all possible pairs of clock components from the exemplar distribution and retrained the various learning algorithms on the reduced exemplar sets , testing our ability to theoretically recover these known clock genes using different ranking cutoffs ( Figure S2A ) . The three methods all had comparable performance using cutoffs less than ∼1 , 000 , but the evidence factor method outperformed the other two beyond this point . The top candidates from all three methods show a very high degree of overlap ( Figure S2B ) . We estimated the false discovery rate ( FDR ) of the Evidence Factor approach by combining the sensitivity analysis with an assumed total number of clock components to generate an expected number of true and false positives at different ranking thresholds ( Figure S2C ) . Preprocessed microarray data obtained from WT and Clock mutant animals as reported by Miller et al was downloaded from the Circa database [47] and replotted . A single apparent outlier from the SCN data ( Mutant , original time point 46 ) is excluded from the plot as this value was greater than any other SCN expression value from WT or mutant animals , and ∼3× the replicate measure . Cel files from the Cry1/Cry2 double mutant were obtained from NIH GEO and normalized via GCRMA [76] . Exon-array cel files describing the transcriptional response of WT and melanopsin knockout animals to sham control and following a light pulse [57] were downloaded from NIH GEO . Data were extracted , annotated , quantile normalized , and log transformed at the gene level using the Affymetrix Expression Console package ( v1 . 1 ) . The probeset corresponding to Gm129 was then separately analyzed . In both WT and knockout animals , when compared to sham control , Chrono expression did not significantly change 30 , 60 , or 120 min after light pulse . C- and N-terminal regions of an enhanced variant YFP called Venus were fused with identified constructs . Expression vectors of S-tagged full-length CHRONO ( 1–385 ) , and its various deletion mutants were cotransfected with GFP–BMAL1 expression vector or BiFC fusion plasmids encoding VC–BMAL1 , CLOCK–VN , or CBP–VN . At 16 h post-transfection , the cells were fixed with 4% paraformaldehyde in PBS and incubated with anti–S-tag ( Bethyl Laboratories , Inc . ) and anti-hC1orf51 ( Santa Cruz Biotechnology ) antibodies , followed by secondary antibodies conjugated to Alexa Fluor 568 ( Invitrogen ) . Cells were visualized using fluorescein isothiocyanate and tetramethylrhodamine isothiocyanate filters in fluorescence microscopy . A modified Aschoff type II procedure was used , facilitating the exposure of animals to light pulses before their free-running rhythms had drifted apart significantly [84] , [85] . Animals were entrained to a 12∶12 L∶D cycle and then placed in constant darkness ( D∶D ) prior to a 30-min light pulse . The light pulses were initiated at zeitgeber times ( ZTs ) 16 or 22 on the second day of D∶D . Animals remained in DD for 7 d following the light pulse . Daily activity onset times were determined using ClockLab Data Collection software ( Actimetrics ) and were exported for further analysis . The phase response was calculated as the difference between activity onset predictions as determined by prepulse and postpulse regression lines computed in R . The prepulse regression line was fit from activity onset data for 5 d prior to the light pulse . The postpulse regression line was determined from the first through seventh days in D∶D following the pulse [85] . We used 1 µg total RNA to generate cDNA with the High Capacity cDNA Archive Kit using the manufacturer's protocol ( Applied Biosystems ) . qPCR reactions were performed using iTaq PCR mastermix ( BioRad ) in combination with gene expression assays ( Applied Biosystems ) on a 7800HT Taqman machine ( Applied Biosystems ) . Importin 8 was used as an endogenous control for all experiments . Mammalian two-hybrid constructs were generated by PCR with primers containing the flanking restriction sites that allow for in-frame cloning of the full-length ORF ( not including the start ATG codon ) into pACT or pBIND plasmids ( Promega ) . The two-hybrid reporter plasmid pGL4P–4XUAS was generated by inserting 4× repeats of the Gal4 UAS binding sites into the pGL4P vector ( Promega ) . Epitope-tagged cDNAs were generated by PCR with primers containing the flanking restriction sites that allow for in-frame cloning of the full-length ORF ( not including the start-ATG codon ) into pFlag [49] or pTag3C plasmids ( Stratagene ) . For plasmids expressing S-tagged CHRONO ( both wild-type and truncation mutants ) , full-length and truncated DNA fragments of the gene were amplified with upstream and downstream primers containing S-tag-encoding sequence ( KETAAAKFERQHMDS ) and were subcloned into pCMV Sport6 or pcDNA3 . 1 expression vectors ( Invitrogen ) using NotI and XhoI restriction enzymes . The shRNA construct sequences were as follows: NS shRNA , CAACAAGATGAAGAGCACC; Sh234 , GACTGGAGTTGCATCCTAT; Sh235 , GAGCCAGCATTGGTGTCAT; Sh236 , GACTTGGTTTCCTCACATA; Sh237 , GGAGAACGTTATCTAGGAA; Sh238 , GGAGCCTCGTTGCCACAGT; Sh239 , GAACCTTGCTGCAGGTGGA; and Sh240 , GTGTCATCCTTGTCCTCCA . Sh 238 , 239 , and 240 were ultimately found to be ineffective by Western and/or PCR . The Taqman probe identifiers were as follows: For Mus musculus: Arntl , Mm00500226_m1; Arntl2 , Mm00549497_m1; Per1 , Mm00501813_m1; Per2 , Mm00478113_m1; Per3 , Mm00478120_m1; Nr1d1 , Mm00520708_m1; Chrono ( Gm129 ) , Mm01255906_g1; Importin 8 , Mm01255158_m1 . For Homo sapiens: Arntl , Hs00154147_m1; Arntl2 , Hs 00368068_m1; Clock , Hs00231857_m1; Per1 , Hs00242988_m1; Per2 , Hs00256144_m1; Nr1d1 , Hs00253876_m1; Chrono ( C1orf51 ) , Hs00328968_m1; Gapdh , Hs99999905_m1 .
|
Daily rhythms are ever-present in the living world , driving the sleep–wake cycle and many other physiological changes . In the last two decades , several labs have identified “clock genes” that interact to generate underlying molecular oscillations . However , many aspects of circadian molecular physiology remain unexplained . Here , we used a simple “machine learning” approach to identify new clock genes by searching the genome for candidate genes that share clock-like features such as cycling , broad-based tissue RNA expression , in vitro circadian activity , genetic interactions , and homology across species . Genes were ranked by their similarity to known clock components and the candidates were screened and validated for evidence of clock function in vitro . One candidate , which we renamed CHRONO ( Gm129 ) , interacted with the master regulator of the clock , BMAL1 , disrupting its transcriptional activity . We found that Chrono knockout mice had prolonged locomotor activity rhythms , getting up progressively later each day . Our experiments demonstrated that CHRONO interferes with the ability of BMAL1 to recruit CBP , a bona fide histone acetylase and key transcriptional coactivator of the circadian clock .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"systems",
"biology",
"molecular",
"neuroscience",
"behavioral",
"neuroscience",
"genomics",
"gene",
"expression",
"genetics",
"gene",
"regulation",
"biology",
"and",
"life",
"sciences",
"molecular",
"genetics",
"computational",
"biology",
"neuroscience"
] |
2014
|
Machine Learning Helps Identify CHRONO as a Circadian Clock Component
|
Light is an important environmental cue that affects physiology and development of Neurospora crassa . The light-sensing transcription factor ( TF ) WCC , which consists of the GATA-family TFs WC1 and WC2 , is required for light-dependent transcription . SUB1 , another GATA-family TF , is not a photoreceptor but has also been implicated in light-inducible gene expression . To assess regulation and organization of the network of light-inducible genes , we analyzed the roles of WCC and SUB1 in light-induced transcription and nucleosome remodeling . We show that SUB1 co-regulates a fraction of light-inducible genes together with the WCC . WCC induces nucleosome eviction at its binding sites . Chromatin remodeling is facilitated by SUB1 but SUB1 cannot activate light-inducible genes in the absence of WCC . We identified FF7 , a TF with a putative O-acetyl transferase domain , as an interaction partner of SUB1 and show their cooperation in regulation of a fraction of light-inducible and a much larger number of non light-inducible genes . Our data suggest that WCC acts as a general switch for light-induced chromatin remodeling and gene expression . SUB1 and FF7 synergistically determine the extent of light-induction of target genes in common with WCC but have in addition a role in transcription regulation beyond light-induced gene expression .
Organisms synchronize their behavior and physiology with the geophysical day-night cycle by acute signal transduction of rhythmically reoccurring cues and via anticipatory processes controlled by circadian clocks . Circadian clocks are biological timing systems that operate from the cellular to the organismal level . They are crucially dependent on interconnected transcriptional and posttranscriptional feedback loops that are intimately connected with metabolism [1 , 2 , 3 , 4] . Although dispensable for clock function per se , light is generally a strong cue for the synchronization of endogenous circadian oscillations with the environmental day-night cycle [5 , 6 , 7] . Light also induces acute transcriptional responses , particularly in photosynthetic organisms and in fungi [8 , 9] . White Collar 1 ( WC1 ) is the major blue-light photoreceptor of Neurospora crassa . WC1 and its partner WC2 are GATA-family DNA binding proteins that assemble into the hetero-dimeric transcription factor ( TF ) White Collar Complex ( WCC ) [10 , 11] . WCC is essential for light-induced gene expression . It regulates carotenoid biosynthesis , asexual spore formation ( conidiation ) and sexual reproduction [12] . WC1 contains a flavin-binding light-oxygen-voltage ( LOV ) blue-light photoreceptor domain [13 , 14 , 15] . Light exposure of such LOV domains induces a covalent flavin-cysteinyl photo-adduct and formation of LOV domain dimmers [16 , 17 , 18] . Light-activation of WCC results in dynamic homo-dimerization of WCC protomers , which then bind to specific light-responsive DNA elements ( LREs ) to activate transcription of target genes [15 , 19 , 20 , 21 , 22] . WCC is also the core TF of the circadian clock of Neurospora . It supports self-sustained circadian gene expression rhythms in the dark and synchronizes the circadian oscillator with rhythmic exogenous light cues [2 , 23 , 24 , 25 , 26] . The dark form of WCC supports a low amplitude circadian nucleosome occupancy rhythm at the so-called clock-box in the frq promoter [27 , 28] while light-activated WCC supports nucleosome remodeling at the light-responsive element ( LRE ) close to the transcriptional start site [29] . Light-activated WCC enhances transcription of the sub1 gene , which encodes a GATA-family TF [19 , 30] . It has been proposed that light-induced accumulation of SUB1 drives in a hierarchical fashion expression of a subset of so-called late light-responsive genes on a second tier [30] . However , substantial levels of sub1 are already expressed in the dark [19 , 30 , 31] and SUB1 target genes are still light-inducible to a lower extent in the absence of SUB1 [30] suggesting a more complex regulation of SUB1-dependent light-inducible genes . We show here that SUB1 cannot activate transcription of light-inducible genes in the absence of WCC . Rather , such genes are activated by WCC and SUB1 supports the activity of WCC . Light-activation of WCC is associated with nucleosome eviction at its binding sites and target promoters . Efficient remodeling of some nucleosomes is dependent on SUB1 . SUB1 interacts with the TF Female Fertility 7 ( FF7 ) . SUB1 and FF7 contribute to transcription of a subset of light-inducible genes but are also required for efficient expression of a large number of non light-inducible genes .
We analyzed expression of SUB1 in the dark and light to confirm that sub1 is a direct target of WCC as shown previously [19 , 30] . When Neurospora was grown in constant darkness SUB1 was rhythmically expressed ( S1A Fig . ) . Basal expression levels of SUB1 were essentially independent of WCC but upon light exposure expression of SUB1 was rapidly induced in a WCC-dependent manner ( S1B Fig . ) . The data demonstrate that sub1 is a light-induced and clock-controlled gene . To investigate a possible crosstalk of the GATA-family TFs SUB1 and WCC we analyzed by RNA-seq the transcriptomes of wt and Δsub1 strains grown in the dark and after light-exposure . We identified 519 light-inducible genes in wt ( S1C Fig . , S1 Table ) . 319 of these genes were also identified recently in a similar analysis [31] . More than 80% of the light-inducible genes ( 417 ) identified in our study responded rapidly to the light cue ( >2x induction within 30 min ) , strongly suggesting that these are immediate early genes directly controlled by WCC . About 50% of the transcripts accumulated to maximal levels after 30 min and the RNA levels decreased subsequently ( early genes ) . The remainder of the transcripts reached peak expression levels somewhat later ( late genes ) . Light-induction of a substantial subset of genes ( 189 ) was severely impaired in Δsub1 ( Fig . 1A ) . Both early and late accumulating light-induced transcripts were affected in Δsub1 . The temporal transcription dynamics of SUB1-affected genes corresponds to the transient activity profile of light-activated WCC suggesting that SUB1 may cooperate with WCC rather than acting independently and downstream of WCC . In addition , deletion of sub1 affected expression of 593 genes that were not induced by light , indicating a major role of SUB1 beyond regulation of light-induced transcription ( S1D Fig . ) . We then analyzed by qPCR the SUB1-dependent regulation of the light-inducible genes rds1 and hyr1 , which harbor a WCC binding site in their promoters [19] . Expression of both genes was rapidly induced by light and the light-induction was attenuated in a Δsub1 strain ( Fig . 1B ) . In contrast , light-induction of the WCC-dependent vvd gene was not significantly affected by SUB1 ( S1E Fig . ) . To assess the SUB1-dependence of light-induced transcription initiation at the rds1 and hyr1 promoters we analyzed the recruitment kinetics of RNA polymerase II ( RNAPII ) in wt and Δsub1 strains . SUB1-dependent recruitment of Ser5 phosphorylated RNAPII , which is indicative of transcription initiation , was already detected 5 min after light-induction ( S1F Fig . ) , i . e . prior to the light-induced accumulation of newly synthesized SUB1 ( S1G Fig . ) . The data indicate that the previously synthesized , old SUB1 cooperates with light-activated WCC to activate transcription at the rds1 and hyr1 promoters . Moreover , we analyzed recently published ChIP-seq data of recruitment kinetics of RNAPII in response to a single 1 min light-pulse [32] . The data revealed that 40% of the SUB1-dependent light-inducible genes and 43% of the SUB1-independent genes showed a rapid ( 5–10 min ) and transient increase in RNAPII occupancy ( S1H Fig ) , suggesting that these light-inducible target genes of SUB1 are directly activated by WCC . The rest of SUB1-dependent ( 60% ) and independent ( 57% ) light-inducible genes did not show significant RNAPII recruitment under these conditions ( 1 min light-pulse ) , suggesting that longer periods of light exposure are required for their maximal activation . Since SUB1 and WCC contain GATA-family DNA-binding domains we asked if these TFs bind to the same sites in the hyr1 and rds1 promoters . We therefore constructed by gene replacement a strain expressing a FLAG-HIS tagged version of SUB1 and performed tandem chromatin immunoprecipitation ( ChIP ) [33] . In dark-grown mycelia SUB1FLAG-HIS was detected at a low level at the WCC binding sites in the rds1 and hyr1 promoters ( Fig . 1C ) . Recruitment of SUB1FLAG-HIS was enhanced after light exposure of mycelia , suggesting that light-activated WCC facilitates binding of SUB1 directly and/or indirectly by supporting light-induced expression of new SUB1 . To assess if SUB1 binding also affects recruitment of WCC we analyzed binding of WCC in wt and Δsub1 strains . Light-induced binding of WCC to the rds1 and hyr1 promoters was attenuated in a Δsub1 strain ( Fig . 1D ) , indicating that SUB1 supports recruitment of light-activated WCC to these promoters . Hence , SUB1 and WCC mutually facilitated their recruitment to overlapping or nearby binding sites in the rds1 and hyr1 promoters . We next determined by ChIP-seq the binding sites of WCC on a genome-wide scale . We , in collaboration with others , had previously identified binding sites of light-activated WCC on the basis of a ChIP-seq analysis with rather low sequence coverage [19] . Here , we determined WCC binding sites using two further independent ChIP-seq approaches . In the first approach we fragmented chromatin by sonification and identified 466 light-inducible putative WCC binding sites ( S2A Fig . , S2 Table ) by tandem affinity ChIP of a TAP-tagged WC2 [34] . In the second approach , chromatin was gently fragmented by MNase digestion and we identified 218 putative binding sites of WCC by ChIP with WC2 antibodies ( S2B Fig . , S2 Table ) . The combined analyses revealed 582 light-dependent putative WCC binding sites with 92 highly confident binding sites that were identified by both approaches ( Fig . 2A , S2 Table ) . Analysis of binding motifs by MEME [35] revealed a GATC-containing consensus motif ( S2C Fig . ) , similar to previously identified WCC binding motifs [10 , 15 , 19] . Further analysis revealed that WCC sites were enriched in tandem GATC motifs with a preferential pairwise spacing of 10–30 bp ( Fig . 2B , C and D ) . This arrangement of GATC motifs may reflect that light-activated WCC is a dimer of two WC1/WC2 protomers [21] and thus can potentially bind up to four GATC motifs . The 92 highly confident WCC binding sites were associated with 91 genes and 51 of these genes were light-inducible . This enrichment of light-inducible genes suggests that these binding sites are functionally relevant . 41 WCC binding sites contain a consensus tandem GATC motif . Binding of WCC to 30 of these sites correlated with light-inducible expression of the associated genes . However , 11 sites with a consensus tandem GATC motif were not associated with light-inducible expression of neighboring genes under the conditions analyzed here ( S5 Table ) . Although the vast majority of light-inducible genes responded rapidly to light-cues ( see above ) we have not detected highly confident WCC binding sites ( n = 92 ) in all light-inducible genes ( n = 519 ) , raising the question of whether they are directly controlled by the WCC . A visual inspection of the WC2 ChIP-seq coverage at promoters of light-inducible genes without significant binding sites revealed putative light-dependent WCC sites that were not detected by our peak-calling algorithm ( S2D Fig . ) . A subsequent ChIP-PCR analysis revealed a light-dependent enrichment of WCC at such regions ( S2D Fig . ) indicating that they are true WCC binding sites and suggesting that the associated light-inducible genes are directly controlled by the WCC . Hence , not all of the WCC binding sites were detected by our ChIP-seq analysis . In order to assess the possible effect of light on SUB1 binding on a genome-wide scale we performed tandem ChIP-seq [33] of SUB1FLAG-HIS using dark grown and light-exposed mycelial cultures . We identified 617 binding sites that were associated with 562 genes ( Fig . 2E , S2 Table ) . 63 genes were light-inducible suggesting a moderate but significant enrichment of SUB1 at light-inducible genes ( p < e-07 ) . However , binding of SUB1 to the majority of sites ( 527 ) was independent of light . Hence , the elevated expression of SUB1 in light-exposed mycelia does not support increased SUB1 binding on a genome-wide scale . Recruitment of SUB1 to 50 sites ( associated with 50 genes ) was , however , enhanced by light ( S2E Fig . ) and 27 of the associated genes were light-inducible . Interestingly , 23 of these 27 genes harbor overlapping ( S2F Fig . ) or close by WCC binding sites . The data indicate that light-induced SUB1 binding occurs mainly at light-inducible WCC target genes . Analysis of SUB1 binding sites by MEME revealed the bipartite motif a/cGATc/g-x6-a/cTGc/t ( Fig . 2F , upper panel ) . This motif was highly enriched ( 223 / 617 ) and located in the center of the SUB1 binding sites ( Fig . 2G ) . The MEME analysis revealed in addition two GTA-rich motifs , which were present in 82 and 63 sites , respectively ( Fig . 2F , lower panels ) . In order to assess the difference between SUB1-dependent and independent light-activated genes we next analyzed recruitment of SUB1 and WCC to light-inducible genes . Sequence coverage of the SUB1 ChIP was higher at promoters of SUB1-dependent light-inducible genes while WCC occupancy was higher at promoters of SUB1-independent light-inducible genes ( S2G and S2I Fig . ) . The data suggest that SUB1 may preferentially support light-induced transcription at promoters with lower affinity for WCC . To assess whether WCC and/or SUB1 support light-induced chromatin remodeling we performed nucleosome mapping of dark-grown and light-exposed wt , Δsub1 and Δwc2 strains by MNase digestion of DNA followed by paired-end sequencing . Two independent replicates were analyzed . The average length of protected fragments was about 140–150 bp and fragments longer than 100 bp were considered nucleosomal DNA , while protected smaller fragments were considered footprints of TFs or other DNA-binding proteins . The nucleosome occupancy profiles at WCC binding sites were bimodal . A rather broad region ( about ± 2kb ) with moderately reduced nucleosome occupancy likely reflects that WCC binding sites are enriched in promoters , which are generally rather nucleosome free , while the confined nucleosome-free region in the center may reflect the actual binding site of WCC ( Fig . 3A , dashed lines ) . In the dark nucleosome occupancy at WCC sites was similar in all the stains ( S3A Fig . , left panel ) . Upon light-exposure , the nucleosome occupancy decreased in wt and Δsub1 but not in Δwc2 , demonstrating that the activated WCC reduces nucleosome occupancy at its binding sites ( Fig . 3A , solid lines ) . Maximal depletion of nucleosomes was observed in wt , i . e . when SUB1 and light-activated WCC were present ( Figs . 3A and S3A , right panel ) . Analysis of an independent nucleosome analysis supported these results ( S3B Fig . ) . The light-induced nucleosome loss at 20 highly confident WCC binding sites was impaired in Δsub1 while nucleosome eviction was independent of SUB1 at the remaining 72 WCC sites ( Fig . 3B ) . Nucleosome occupancy at SUB1 binding sites was rather low in wt , Δwc2 and even in Δsub1 and essentially independent of light ( Figs . 3C , S3A ) . The data suggest that SUB1 binding sites are either intrinsically free of nucleosomes or that other chromatin remodelers keep these sites open . To obtain potential footprints of the WCC and SUB1 , we analyzed MNase-resistant DNA fragments that were shorter than typical fragments protected by nucleosomes . Protected DNA fragments < 100 bp accumulated in light-dependent fashion at WCC binding sites ( S3C Fig . , left panel ) , suggesting that they correspond to a footprint of the light-activated WCC . In contrast , a potential footprint of SUB1 was not affected by light ( S3C Fig . , right panel ) . Together these observations indicate that binding of the light-activated WCC triggers depletion of nucleosomes from its binding sites . SUB1 contributes to the light-induced nucleosome removal at WCC binding sites . Binding sites of SUB1 are also rather devoid of nucleosomes , even in the absence of SUB1 , and the nucleosome occupancy of SUB1 sites was independent of light and WCC . To identify on a genome-wide level light-induced nucleosome remodeling in promoters and genes , we aligned the +1 nucleosomes of all annotated transcription start sites ( TSSs ) . In transcribed regions the nucleosomes were regularly spaced by 176 bp and nucleosome occupancy was rather high ( Fig . 4A ) . In contrast , nucleosome spacing was irregular and occupancy was lower in promoters , similar to corresponding observations in other species [36 , 37 , 38] . Light , WCC and SUB1 did not affect nucleosome occupancy of genes and promoters on a genome-wide scale ( Fig . 4A ) . However , light triggered nucleosome remodeling at the promoter of the SUB1-dependent rds1 gene . A light-induced loss of nucleosomes was detected at the overlapping WCC and SUB1 binding sites ( Fig . 4B and C ) . The light-induced nucleosome loss was attenuated in Δsub1 and absent in a Δwc2 strain , indicating that removal was strictly dependent on the activated WCC and supported by SUB1 . A light-induced loss a nucleosome was also observed at the WCC binding site of the hyr1 promoter ( S4A Fig . , left panel ) , which was , however , not dependent of SUB1 . Light triggered a substantial loss and repositioning of nucleosomes at the vvd promoter and gene ( S4A Fig . , right panel ) . The light-induced nucleosome dynamics were similar in wt and Δsub1 but absent in Δwc2 , indicating that chromatin remodeling of the vvd gene by the light-activated WCC was independent of SUB1 . The pronounced depletion of nucleosomes in the transcribed region of vvd is likely due to the synchronous activation of the rather strong vvd promoter in the entire ensemble of nuclei . Similar losses of nucleosomes were observed in the transcribed region of other highly expressed light-inducible genes such as al-1 , cry and con-10 but was less pronounced or not detectable in less active genes such as hyr1 , ncu00309 and frq ( S1 Table ) . With the exception of highly transcribed genes the light-induced eviction of nucleosomes was generally confined to one or two nucleosomes overlapping the WCC binding sites . We observed eviction of individual nucleosomes at several high affinity binding sites of the WCC that were not associated with transcription initiation of a neighboring or close-by gene ( Figs . 4D and S4B ) . We confirmed the light- and WCC- dependent nucleosome eviction at these sites by independent histone H2A ChIP-PCR ( Fig . 4E ) . These observations suggest that the light-activated WCC supports eviction of nucleosomes at its binding sites independent of transcription . To address whether SUB1 can activate transcription of light-inducible genes without the WCC we expressed in wc1-deficient ( wc1mut ) [39] and wc1-proficient ( wc1+ ) strains a FLAG-tagged SUB1 under control of the inducible quinic acid 2 ( qa2 ) promoter . SUB1FLAG was expressed at low level in the absence of QA and expression levels were elevated in the presence of QA ( Figs . 5A , S5A ) . Expression levels of hyr1 , which is a SUB1-affected light-inducible gene , were generally higher in a wc1+ background than in the corresponding wc1mut strains ( Fig . 5B ) . In light , QA-induced SUB1FLAG supported expression of hyr1 at a high level in the presence of WCC ( wc1+ ) but not in the absence of WCC ( wc1mut ) ( Figs . 5B and S5B ) . Together the data indicate that SUB1 cannot activate expression of hyr1 independently of the WCC . Maximal expression of hyr1 requires the presence of SUB1 and light-activated WCC . To analyze binding of SUB1FLAG to the hyr1 promoter we performed ChIP-PCR . In a wc1+ background binding of QA-induced SUB1FLAG was more efficient in light than in dark while binding of SUB1 was independent of light in a wc1-deficient strain ( Fig . 5C ) . The data suggest that the light-activated WCC supports recruitment of SUB1 to the hyr1 promoter . Corresponding results were obtained for the rds1 promoter ( S5C Fig . ) . Since the WCC facilitates recruitment of SUB1 , expression of SUB1-dependent light-inducible genes could be limited by SUB1 abundance . To test this hypothesis , we generated WCC-proficient strains expressing SUB1 under control of the tubulin ( tub ) and the ccg1 promoter , respectively . Dark-grown mycelial cultures of wt , Δsub1 , tub::sub1 and ccg1::sub1 were exposed to light and SUB1 levels and the kinetics of hyr1 expression were measured ( Fig . 5D and E ) . In wt , SUB1 was expressed at low level in the dark and accumulated , as expected , to high levels after light exposure . In contrast , SUB1 levels were constitutively low in tub::sub1 and constitutively high in ccg1::sub1 ( Fig . 5D ) . In the dark hyr1 expression correlated well with the SUB1 levels in the respective strain . In response to light , hyr1 levels increased with similar kinetics ( 5–7 fold ) in all strains , reaching the highest expression in ccg1::sub1 and the lowest level in Δsub1 . Interestingly , hyr1 levels were similar in wt and tub::sub1 despite light-induced accumulation of substantial amounts of SUB1 in wt ( Fig . 5D upper panel ) . Hence , the SUB1 that was newly synthesized under control of the WCC did not independently activate hyr1 on a second hierarchical tier . Corresponding results were obtained when light-induced expression of the SUB1-dependent genes rds1 and ncu00309 was analyzed ( S5D , S5E Fig . ) . Over-expression of SUB1 in a WCC-deficient background ( S5F Fig . ) did neither support elevated expression of the SUB1 target genes ( rds1 and ncu00309 ) in the dark nor in light ( S5G and S5H Fig . ) . Together the data indicate that SUB1 functionally cooperates with the dark-form and the light-activated WCC but cannot activate transcription of light-inducible genes in the absence of WCC . Thus , SUB1 supports the activity of WCC in synergistic manner but is not a bona-fide transcription activator of light-inducible genes . We noted that light-induction of rds1 was strongly attenuated by deletion of sub1 but was not affected by SUB1 overexpression . In contrast , light-induction of hyr1 was only moderately affected by deletion of sub1 but was strongly enhanced by SUB1 overexpression , suggesting that the SUB1 level in wt was limiting for hyr1 expression under the conditions analyzed . To detect hyr1-type genes that respond only to high levels of SUB1 we analyzed the light-inducible transcriptome of ccg1::sub1 in comparison to a wt strain . In wt ( replicate 2 ) we identified 657 light inducible genes ( S5I Fig . , S3 Table ) . Most of these genes were also identified by the independent analysis ( replicate 1 ) described above ( S5J Fig . ) . Light-induction of 121 genes was significantly enhanced in ccg1::sub1 ( S5K and S5L Fig . ) . Together with the group of genes down-regulated in Δsub1 ( see Fig . 1 ) the data indicates that about 40% ( 264 ) of the light-inducible genes are co-regulated by SUB1 . To identify interaction partners of SUB1 we performed tandem affinity purification of SUB1FLAG-HIS . By subsequent mass-spectrometry we identified Female Fertility-7 ( FF7 ) as a potential interaction partner of SUB1 . FF7 has a Gal4-type Zn ( 2 ) -Cys ( 6 ) binuclear cluster domain and a putative acetyl transferase domain with similarity to maltose acetyl transferase . To confirm the interaction we constructed a strain expressing FLAG-HIS tagged FF7 and performed reciprocal anti-FLAG and anti-SUB1 immunoprecipitations . SUB1 co-immunoprecipitated with FF7FLAG-HIS ( Fig . 6A ) and vice versa , FF7FLAG-HIS was pulled down with SUB1 antibodies ( Fig . 6B ) . The pull-down efficiency was quite low , suggesting that the interaction is rather unstable . To identify the binding sites of FF7 and to investigate whether SUB1 and FF7 co-localize on the genome , we performed tandem ChIP-seq of FF7FLAG-HIS from light-exposed mycelial cultures . We identified 2756 putative FF7 binding sites that were associated with 2315 genes ( S4 Table ) , suggesting a rather ubiquitous role of FF7 . Analysis of FF7 binding sites by MEME revealed a AACCGC motif ( Fig . 6C , upper panel ) that was highly enriched in the center of the FF7 binding sites ( Fig . 6D ) . A “GTA” rich motif , similar to the one found in the SUB1 ChIP-seq , was also found in ChIPed FF7 sites ( Fig . 6C , lower panel ) . This motif might be a more general element associated with promoters since it was not enriched at the center of the binding sites . To assess the potential relationship of FF7 , SUB1 and WCC we analyzed the occupancies of the transcription factors at their own and the respective binding sites of the other TFs ( S6A Fig . ) . FF7 binding was enriched at WCC and at SUB1 sites . Similarly , SUB1 binding was enriched at WCC and at FF7 sites . WCC sites were , possibly due to their low number ( n = 92 ) , neither enriched at SUB1 sites ( n = 617 ) nor at FF7 sites ( n = 2756 ) . On a genome wide scale , about 70% ( 422 / 617 ) of the SUB1 binding sites were also occupied by FF7 indicating a highly significant ( p < e-10 ) co-occurrence of these factors on the DNA ( Fig . 6E ) . Similarly , 72% ( 68 / 92 ) of the WCC binding sites overlap with FF7 . Interestingly , essentially all SUB1 and WCC overlapping sites ( 28 / 29 ) harbor also FF7 binding sites ( Fig . 6E ) . Examples are shown in Figs . 6F and S6B . The 28 overlapping WCC , SUB1 and FF7 binding sites were associated with 25 expressed genes . 10 of these genes were light-inducible and dependent on SUB1 and/or FF7 , 5 genes were light-inducible but independent of SUB1/FF7 and 10 genes were not light-inducible ( S1 and S6 Tables ) . Thus , the majority of functional WCC sites with overlapping SUB1 and FF7 sites are also co-regulated by SUB1 and FF7 . To analyze the functional cooperation of FF7 with SUB1 and WCC , we determined by RNA-seq the transcriptome of a Δff7 strain grown in the dark and after light-exposure . RNA levels of 192 of 519 light-inducible genes were reduced in Δff7 ( S6C Fig . , S1 Table ) . 109 of these genes show impaired light-induction in Δff7 and in Δsub1 ( Fig . 6G ) . We confirmed the impaired light-induction of the rds1 gene by independent RNA measurements ( S6D Fig . ) . Furthermore , expression of 440 genes was reduced in Δff7 in a light-independent manner ( S1 Table ) and expression of 278 of these genes was also reduced in Δsub1 ( S6E Fig . ) . Finally , we performed nucleosome mapping of Δff7 after light-induction to assess whether FF7 contributes to light-induced nucleosome eviction at WCC binding sites . The light-induced nucleosome loss at the rds1 promoter was independent of FF7 despite impaired light-induction of the rds1 gene in Δff7 ( Fig . 6H ) . Similarly , we did not detect impaired light-induced nucleosome removal at other WCC sites in Δff7 ( Fig . 6I ) suggesting that FF7 is required for transactivation rather than eviction of nucleosomes . Together the data suggest that WCC , SUB1 and FF7 have distinct functions and cooperate to regulate subsets of genes in a combinatorial fashion .
WC1 and WC2 , which constitute the WCC , and SUB1 are GATA-family transcription factors of Neurospora crassa . Here we analyzed their roles in regulation of light-induced nucleosome remodeling and gene expression . We identified about 500 light-inducible genes . SUB1 regulates a substantial subset of light-inducible genes ( 264 ) in cooperation with the WCC and a larger number of non light-inducible genes . However , SUB1 , even when overexpressed , cannot induce transcription of its light-inducible target genes in the absence of WCC . Hence , SUB1 does not independently activate a subset of late light-inducible genes on a second hierarchical tier . The immediate light-induced recruitment of RNAPII , even in response to a 1 min light-pulse , suggests that more than 40% of the light-induced SUB1 target genes are directly activated by the WCC photoreceptor rather than indirectly via light-induced synthesis and accumulation of a TF acting on a second hierarchical tier . How does SUB1 cooperate with the WCC to regulate light-inducible gene expression ? Transcription activation in eukaryotes is based on regulation of DNA accessibility to RNAPII . This is often achieved by cooperation of several transcription regulators and co-regulators facilitating in combinatorial fashion modification and eviction of histones and subsequent recruitment of general transcription machinery and RNAPII [40 , 41 , 42 , 43 , 44 , 45] . Conceptually the light-activated WCC and SUB1 could exert identical functions . In this case the TFs would contribute independently and in approximately additive manner to the transcriptional output of common target genes . This is obviously not the case since SUB1 cannot activate its light-inducible target genes in the absence of WCC . Rather , it seems likely that WCC and SUB1 contribute distinct functions to activate common target promoters . We show that nucleosome occupancy profiles at binding sites of WCC and SUB1 are rather low even in the absence of the cognate TF , suggesting that they are either intrinsically nucleosome-free or that sequence specific machinery keeps these sites open . Additional light-induced eviction of nucleosomes at WCC binding sites is strictly dependent on the WCC . SUB1 contributes to the WCC-dependent light-induced nucleosome loss at a subset of promoters , suggesting that WCC and SUB1 act in combinatorial rather than additive fashion . Light-induced eviction of nucleosomes at WCC binding sites is independent of transcription . The remodeling activity associated with the WCC might be similar to the pioneer-like activity of circadian transcription factor CLOCK/BMAL1 in mice and prepare promoters for activation for other TFs [46] . The light-activated WCC recruits NGF1 , a H3K14 acetyl transferase homologous to yeast GCN5 [47] and may , similar to its less active dark form , also recruit the rather ubiquitous ATP-dependent chromatin remodeler SWI/SNF [28] to evict nucleosomes as reported for the promoter of the frq [29] . How are the differentially regulated subsets of SUB1-dependent and independent light-inducible genes defined ? The highly homologous Zn-fingers of WC1 , WC2 and SUB1 bind GATC-related core sequence motifs and flanking sequences seem to distinguish their specificity . Although binding motifs of the dark form of the WCC have not been determined it seems to interact with GATG motifs in the clock box of the frq promoter [48] and presumably also with GATC motifs present in the vvd LRE [49] . Deletion or mutation of the Zn-fingers of WC1 or WC2 abolishes WCC activity in the dark [50 , 51] . Hence both Zn-fingers seem to have the capacity to interact with DNA and contribute to WCC activity . When activated by light protomers of WCC dimerize dynamically [21] . The increased ChIP efficiency of light-activated WCC may reflect tighter DNA binding of the WCC dimer , which could potentially interact with up to four GATC-related motifs via the Zn-fingers of two WC1 and two WC2 subunits . Analysis of the 92 highly confident binding sites of light-activated WCC revealed tandem GATC motifs spaced by ≤ 30 bp that are enriched in the center of the binding sites . Strong LREs might thus be determined by number and spacing of GATC motifs . Indeed , highly occupied LREs contain tandem or more GATC motifs ( Fig . 2D ) . Hence , the molecular mechanism underlying light-induced recruitment of WCC seems to be based on a gain in binding avidity . Weak and highly dynamic protein-DNA interactions of WCC protomers with individual GATC related motifs and weak protein-protein interactions between WCC protomers are mutually stabilized at LREs containing properly spaced GATC motifs . The majority of light-inducible genes were not associated with significant WCC binding sites , i . e . sites detectable by two independent ChIP-seq replicates . The median expression level of light-inducible genes without significant WCC binding site was rather low and many promoters of these genes contain putative low affinity binding sites ( S2D Fig . ) . ChIP-PCR analysis revealed that WCC is in fact recruited in light-dependent fashion to such low affinity sites . Furthermore , promoters of such genes are significantly enriched in tandem GATC repeat motifs . In addition , a large fraction of light-inducible genes without detectable WCC binding sites responded immediately ( within a few minutes ) even to a short light pulse . These observations suggest that most light-inducible genes are directly activated by the WCC rather than via induction and accumulation of sufficient SUB1 that would then indirectly induce transcription of genes on a second hierarchical tier . However , since the WCC controls expression of several TFs in addition to SUB1 [19 , 30] , accumulation of some of these TFs could induce genes on a second hierarchical tier . Detectable accumulation of the corresponding transcripts may , however , require longer time periods than analyzed in this or previous studies [30] . Interestingly , our combined analyses of the light-inducible transcriptome and the WCC cistrome revealed 13 genes ( S5 Table ) that were not light-inducible despite the presence of highly confident WCC binding sites with tandem GATC motifs in their promoters . To support expression of these genes the WCC may cooperate with unknown TFs , which were not active under the experimental conditions analyzed . A bipartite sequence motif is highly enriched in binding sites of SUB1 ( Fig . 2F , upper panel ) . The first half of this motif ( a/cGATc/g ) is related to binding motifs of GATA-family proteins . The second part of the motif ( a/cTGc/t ) is located a full helical turn of the DNA away ( center to center ) and could reflect an additional sequence-specific contact of SUB1 or correspond to the binding site of an interaction partner of SUB1 . We identified FF7 as dynamic interaction partner of SUB1 . FF7 contains a putative O-acetyl transferase domain and a Zn2Cys6 binuclear cluster DNA binding domain that binds to AAGCGC motifs and not to a/cTGc/t . An interaction partner other than FF7 was not detected in our affinity purified SUB1 preparation and SUB1 is not predicted to harbor a second DNA binding domain in addition to its GATA-type Zn-finger . Hence , the functional role of the a/cTGc/t remains elusive . The large fraction of SUB1 binding sites overlapping with FF7 binding sites ( ∼70% ) and the large fraction of SUB1-affected genes co-regulated by FF7 ( ∼ 60% of the light-inducible and ∼ 47% of non light-inducible genes ) suggests that SUB1 might generally cooperate with FF7 . However , FF7 has ∼4-fold more genomic binding sites than SUB1 . Hence , FF7 seems to have a broader role and it may cooperate with other TFs in addition to SUB1 , consistent with the more severe phenotype of Δff7 ( reduced conidiation and female fertility ) in comparison to a Δsub1 strain . When SUB1 and FF7 binding sites are sufficiently close , individually weak interactions of SUB1 and FF7 with their cognate binding sites as well as weak interactions of SUB1 and FF7 ( as detected by pull-downs ) might be mutually stabilized to specify the subset of genes regulated by these two TFs . Pathways and cues regulating SUB1 activity are not known . The expression levels of SUB1-affected light-inducible target genes were roughly proportional to SUB1 abundance , suggesting that WCC and SUB1 act synergistically . At light-inducible genes that are independent of SUB1 the corresponding activity might not be rate limiting or provided by other , unidentified TFs with or without the help of FF7 . Combinatorial cooperation of TFs with the WCC would allow differential regulation of subsets of the large group of light-inducible genes . Thus , WCC could regulate the fold induction of genes , i . e . light versus dark ratio , while cooperating TFs such as SUB1 might synergistically support gene expression in light and in dark . Hence , the apparent set of genes that respond in significant manner to light may crucially depend on the activity of TFs cooperating with the WCC .
Neurospora strains; wt ( FGSC #2489 ) , Δsub1 ( FGSC #11127 ) , Δwc2 ( FGSC #11124 ) , Δff7 ( FGSC #11073 ) , wc1mut ( FGSC #4398 ) , bd ( FGSC #1859 ) , his-3 ( FGSC #6103 ) used in this study were acquired from FGSC . wc1+ ( FGSC #9718 ) was used to generate ff7FlagHis , sub1FlagHis and wc1+qa2::sub1Flag strains . wc1mut was used to generate wc1mut qa2::sub1Flag . bd Δwcc , his-3 [52] was used for integration of ccg1::sub1 into the his-3 locus to obtain bd ΔWCC ccg1::sub1 . his-3 strain was crossed to Δsub1 to create Δsub1 , his-3 strain that was used to generate tub::sub1 and ccg1::sub1strains . Δsub1 , his-3 strain transformed with empty vector was used as a sub1 KO strain in RNA measurements together with ccg1::sub1 and tub::sub1 strains . Standard growth medium contained 2% glucose , 0 . 5% L-arginine , 1× Vogel's medium , and 10 ng / mL biotin . For light-induction experiments , indicated strains were grown in petri plates until mycelial mats formed . Mycelial pads ( 1 cm ) were cut out and grown for 1 day in light at 25°C and transferred to darkness for 24 h before cultures were exposed to light ( 100 μE ) and harvested after the indicated time periods . For replicate 1 of the light-induction experiment wt , Δsub1 and Δff7 strains were analyzed whereas wt and ccg1:sub1 strains were analyzed for the replicate 2 . For the QA induction experiments 0 . 3% QA ( final ) was added to 24 h dark grown and constant light grown cultures . Samples were harvested before and 4 hours after the addition of QA . For nucleosome mapping conidia of wt , Δsub1 , Δwc2 and Δff7 ( strains were inoculated in 200 ml media , grown in light for 2 days and transferred to darkness for 24 h . Dark grown and 20 min light exposed cultures were crosslinked with 0 . 5% paraformaldehyde ( FA ) for 10 min . FA was quenched with 125 mM glycine for 5 min . Δff7 was not included into the replicate 2 nucleosome mapping . The yeast in vivo recombination system [53] was used to generate sub1FlagHis , ff7FlagHis and qa2::sub1Flag strains . Transformation of Neurospora was performed as described [53] . Primers are listed in S6 Table . Extraction of proteins and subcellular fractionation were performed as described [54] . SUB1 rabbit antibody was generated against the peptide “RKRQLEQRSIRPKPTDDRN” . H2A antibody was generated against the peptide “CHQNLLPKKTGKTGKNASQEL” Western blotting was performed as described [55] . Protein concentration was estimated by measuring absorption at 280 nm ( NanoDrop , PeqLab ) . Enhanced chemiluminescence signals were detected with X-ray films ( Fuji Film Tokyo , Japan ) . RNA was prepared with peqGOLD TriFAST ( peqLab , Erlangen , Germany ) and reverse transcribed with the QuantiTect Reverse Transcription Kit ( QIAGEN , Hilden , Germany ) . Transcript levels were analyzed by quantitative real-time PCR in 96-well plates with the StepOnePlus Real-Time PCR System ( Applied Biosystems ) . TaqMan Gene Expression Master Mix ( Applied Biosystems ) and UPL probes ( Roche ) were used . Primers and probes are listed in S6 Table . The previously published MNase digestion protocols [56 , 57] were optimized for Neurospora . 400 mg ground mycelial powder from each culture was resuspended in 3 . 75 ml MNase digestion buffer ( 250 mM sucrose , 60 mM KCl , 15 mM NaCl , 15 mM Tris-HCl pH 7 . 4 , 3 mM MgCl2 , 1 mM CaCl2 , 0 . 2% NP-40 ) with freshly added 0 . 5 mM DTT , 0 . 5 mM Spermidine and protease inhibitors ( EDTA-free , Roche ) . The suspension was mixed by vortexing and incubated on ice for 5 minutes . Next 750 μl aliquots were distributed to five 1 . 5 ml Eppendorf tubes . MNase powder ( Sigma-N3755-500 ) was resuspended in 850 μl MNase resuspension solution ( 10 mM HEPES-KOH pH 7 . 6 , 50 mM KCl , 1 . 5 mM MgCl2 , 0 . 5 mM EGTA , 10% glycerol ) . Aliquots were digested with different amounts ( 0 [control] , 0 . 75 , 1 . 5 , 3 , 6U ) of MNase to cover both , sensitive and resistant sites . All samples were incubated at 25°C for 1 hour with shaking at 400 rpm in a Themo-mixer . The reaction was stopped by adding stop buffer ( final 0 . 2% SDS , 10 mM EDTA pH 8 . 0 ) . Samples are centrifuged at 20000 g for 20 min to pellet the cell debris . Supernatants were transferred to new tubes and 2 μl RNAse Cocktail Enzyme mix ( Life Technologies , AM2286 ) was added and incubated at 37°C for 45 minutes to degrade RNA . To degrade proteins , 15 μl proteinase K solution ( Life Techonolgies , AM2548 ) was added and samples were incubated at 65°C for 2 hours . DNA was precipitated with EtOH in the presence of 40 μg glycogen ( Thermo ) and further cleaned by using PCR clean-up kit ( Promega ) . Aliquots were analyzed by electrophoresis in a 1 . 7% Agarose gel ( 80 volts for 1 hour ) to visualize nucleosomal DNA ladders . Paired-end libraries for sequencing were prepared as described below . Neurospora cultures were crosslinked with 1% FA for 10 min . FA was quenched 5 min with 125 mM glycine . ∼ 400 μl ground mycelial powder was resuspended and digested in 600 μl MNase digestion buffer for 1 h at 25°C with 15 U of MNase ( Sigma-N3755-500 ) ( In addition an independent sample from a 1 min light-induced culture was digested with 5 U MNase ) . The reactions were stopped by adding final 5 mM EDTA pH 8 . 0 . Then 400 μl ChIP lysis buffer ( 50 mM HEPES , 150 mM NaCl , 1 mM EDTA , 1% Triton-X , 0 . 1% SDS and 0 . 1% NaDOC ) was added and samples were centrifuged at 4°C , 15000 g . The rest of the ChIP protocol was performed as described [19] . DNA for ChIP-seq was pooled from two-independent experiments . Light-induction was performed for the indicated time periods and cultures were cross-linked in constant light with 1% FA for 15 min . FA was quenched with 125 mM glycine for 5 min . 6 aliquots of 600 μl ground mycelial powder were used for each time point for TAP-WC2-ChIP-seq . Mycelia were dissolved in 1 ml ChIP lysis buffer ( 50 mM HEPES pH 7 . 4 , 150 mM NaCl , 1 mM EDTA , 1% Triton-X , 0 . 1% SDS and 0 . 1% NaDOC ) with freshly added 0 . 5 mM DTT and protease inhibitors ( Roche cOmplete Protease Inhibitor Cocktail-EDTA free ) . Sonification was performed with SonoLabTM Covaris Version 7 . 0 . 20 . 0 ( average incident power 36 watt , peak incident power 180 watt , duty factor 20 percent , cycles / burst 200 count and duration 160 seconds ) . Samples were centrifuged at 15000 g for 20 min at 4°C . Supernatants of the six aliquots per time points were combined and incubated with 100 μl IgG Sepherose/time point ( GE Healthcare ) for 3 h at 4°C . Beads were washed 2x with ChIP lysis buffer , 1x Lindet ( 250 mM LiCl , 1% NP-40 , 1% NaDOC , 1 mM EDTA , 10 mM Tris / HCl; pH 8 ) and 2x with TAP buffer ( 50 mM Tris–HCl , pH 7 . 5 , 150 mM NaCl , 1 . 5 mM MgCl2 , 0 . 1% NP-40 , 0 . 5 mM DTT ) . Elution was performed with two consecutive digestions with 50 U TEV protease ( Invitrogen ) at 16 C° 2 h and ON at 4 C° . Final 3 mM CaCl2 was added to the combined elution and incubated with 130 μl Calmodulin Sepharose 4B ( GE Healthcare ) for 3 . 5 h . Beads were washed 5 times with TAP buffer with 3 mM CaCl2 . The elution and DNA extraction was performed as described [19] . DNA for ChIP-seq was collected from three-independent experiments . Tandem ChIP with Ni-NTA enrichment , followed by anti-Flag immunoprecipitation was performed as described [33] . Primers and probes used for the ChIP-PCR are listed in S6 Table . NEBNext Ultra RNA Prep kit with NEBNext Multiplex oligos was used for cDNA preparation at Bioquant Deep Sequencing Core Facility . PolyA selection was performed at the beginning of the protocol . The size and the quality of the libraries were checked with a 2100 Bioanalyzer . Un-paired sequencing with 50 bp read length was performed with a HiSeq 2000 at GeneCore EMBL Heidelberg . Raw reads can be accessed at SRA database under the accession numbers listed in S7 Table . 50 bp long raw reads were mapped to Neurospora crassa genome NC10 by using Bowtie [58] . Maximum 3 mismatches were allowed for the mapping . Reads mapping to more than one site were discarded . Gene expression was quantified by the number of reads falling into annotated exons . Normalization between samples was performed by using the size factor formula as described [59] . Genes with low RNA levels ( lower 20% of all annotated genes ) were excluded from further analysis . In order to identify significant light-induced genes , differential gene expression analysis was performed . Read counts of the genes were assumed to follow negative binomial ( NB ) distribution , Gi≈NB ( μi , σi ) , where μi is mean and σi is variance . Since no replicates information were available , σi was estimated based on the mean and variance correlation . A “locfit” R package was used to fit the relationship between the mean and variance . By adapting the Robinson and Smyth Exact Test [60] , the two side p-value can be computed as follow , p=∑f ( a , b ) ≤f ( Gtreat , Gcontrol ) f ( a , b ) ∑f ( a . b ) where a+b=Gtreati+Gcontroli , a , b∈0 . . ( Gtreati+Gcontroli ) . Gtreati is the read counts of ith gene in treatment condition , Gcontroli is the read counts of ith gene in control condition , f ( a , b ) is the product of f ( a ) and f ( b ) , which can be computed using dnbinom of R package . After finding the genes that have a significant ( p < 0 . 05 ) increase in the reads in either 30 , 60 or 120 minutes compared to DD reads , we set another cut off ( 2 fold ) to further filter the candidates . In order to identify genes that show impaired light-induced in Δsub1 light-induced time points were compared and p values were calculated as described above . The time points that have significantly lower reads ( p < 0 . 05 ) in the knock-out strains were identified . Similar analysis and p values were used to identify up-regulated genes in ccg1:sub1 compared to wt ( replicate 2 ) . We analyzed ChIP-seq data from Cesbron et al . to assess the kinetics of RNAPII recruitment to 519 light-inducible genes ( RNA-seq replicate 1 ) . Significantly increased levels ( p < 0 . 05 ) of transcribing RNAP ( RNAPII-S2P ) were detected at 221 genes already 1 , 5 and 10 min after a 1 min light-pulse . Libraries from purified MNase-digested DNA were prepared without size selection to detect nucleosomal DNA and putative footprints of transcription factors . NEBNext Ultra™ DNA Library Prep Kit for Illumina ( E7370L ) was used for library preparation at the Bioquant Deep Sequencing Core Facility by using manufacturer`s instructions . The size and the quality of the libraries were analyzed with a 2100 Bioanalyzer . Paired-end sequencing with 100 bp read length was performed with a HiSeq 2000 at BGI , Hong Kong . Raw reads can be accessed at SRA database under the accession numbers listed in S7 Table . For nucleosome mapping the fragment length was set to be between 100 bp and 1000 bp with forward and reverse conformation . For mismatches and multiple alignments the same settings were used as in the single end mapping . The non-mapped reads were further processed to remove the adapters , mapped to the genome again and analyzed independently . The middle 50 bp of the pair end position was used to generate the wiggle file format genome nucleosome coverage . Wig files were visualized with IGV genome browser [61] . The normalization factor was computed by using 90th percentile of each experiment . Smoothing was carried out by using Kernel Regression Smoother package of R . 1500 bp upstream and downstream of annotated TSS position were used for genome wide nucleosome coverage analysis . The center of the +1 nucleosome of genes was defined by the maximum read coverage in a window of 200 bp around the annotated TSS . Each nucleosome was estimated to cover 176 bp , and the nucleosome coverage was estimated by the area under the curve . To analyze the nucleosome coverage at the transcription factor binding sites ( BS ) the nucleosome mean coverage at the summit of the BS was calculated . As the position of a BS relative to a TSS follows a NB , a random NB using the same parameter was generated . The background nucleosome mean coverage was then computed by using the random relative position of BS to TSS . The nucleosome mean coverage was determined by using the normalized nucleosome mean coverage and plotted as ratio to remove the possible bias related to the promoters . ChIP DNA libraries were prepared with NEBNext ChIP-Seq Library Prep Reagent Set for Illumina with NEBNext Multiplex oligos at Bioquant Deep Sequencing Core Facility . A 2100 Bioanalyzer was used to check the quality and size of the libraries . Un-paired sequencing with 50 bp read length was performed with a HiSeq 2000 at BGI , Hong Kong . Mapping was performed as in RNA-seq analysis . In order to identify peaks a sliding window of 150 bp with a step of 50 bp was used to scan the genome and quantify the read intensity . The read intensity was normalized by using the total mapped reads . Wig files were visualized with IGV genome browser [61] . The significantly enriched windows were computed by fitting the read intensity to a Poisson distribution . A ChIP binding site was called if 4 continuous windows were statistically higher ( p value < 1e-05 ) compared to the control ChIP-seq . FGSC #9718 strain was used as a background to identify the significant windows for FLAG-HIS ChIP-seq . Dark ChIP of each experiment was used as a background for TAP-WC2 and MNase-WC2 ChIP . Another cut off was determined based on the coverage of the identified ChIP peak . We excluded the binding sites with low enrichment ( < 1 . 5 fold , < 90th percentile of the coverage ) by comparing the peak coverage to general coverage of the ChIP-seq . Raw reads can be accessed at SRA database under the accession numbers listed in S7 Table . Upstream and downstream genes were analyzed for the annotation of the peaks . The gene was annotated to a peak if there was a peak detected within 1000 bp upstream of the TSS and 500 bp downstream of TSS . If the peak was not close to a promoter of any gene , the binding site was annotated to the nearest genes with the expected orientation ( i . e . binding sites are upstream of the annotated TSS ) with no distance limitation . If there were no genes in the upstream/downstream of the peak with the correct orientation , the peak was not annotated to any gene . The SUB1 binding sites of DD and 30 min light-induction samples were merged based on the coordinate location to find light-induced SUB1 binding sites . Using linear regression between the DD and 30 min light-induction read intensity of the peaks , light-regulated SUB1 binding sites were identified . The linear regression was fitted using R , and the clustering was based on the 80% confidence prediction interval . Motif analysis was done by using MEME motif search software [35] . 300 bp DNA regions surrounding the summit of the ChIP-seq peaks were used . For the WCC ChIP-seq “GATC” motif the pair wise distribution was calculated within these 300 bp regions .
|
In this study we have investigated the roles of the Neurospora transcription factors ( TFs ) WCC and SUB1 in light-activation of transcription . In principle TFs could exert identical functions for transcriptional activation and the extent of transcription will be determined by the sum of activity of the TFs . In this case however , we found that the activity of the main blue-light photoreceptor WCC is essential for the activation of light-inducible genes . SUB1 cooperates synergistically with the WCC to enhance expression of a subset of genes controlled directly by the light-activated WCC but cannot activate its light-inducible target genes in the absence of WCC . WCC evicts nucleosomes at its binding sites . This process is supported by SUB1 at a subset of common target genes . Light-dependent nucleosome loss generally correlates with but is not dependent on induction of transcription . Light-induced nucleosome eviction by the WCC/SUB1 could sensitize promoters for activation via endogenous and exogenous cues other than light , which may modulate the plasticity of the light-responsive transcriptome .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Combinatorial Control of Light Induced Chromatin Remodeling and Gene Activation in Neurospora
|
Mitochondria form a dynamic tubular reticulum within eukaryotic cells . Currently , quantitative understanding of its morphological characteristics is largely absent , despite major progress in deciphering the molecular fission and fusion machineries shaping its structure . Here we address the principles of formation and the large-scale organization of the cell-wide network of mitochondria . On the basis of experimentally determined structural features we establish the tip-to-tip and tip-to-side fission and fusion events as dominant reactions in the motility of this organelle . Subsequently , we introduce a graph-based model of the chondriome able to encompass its inherent variability in a single framework . Using both mean-field deterministic and explicit stochastic mathematical methods we establish a relationship between the chondriome structural network characteristics and underlying kinetic rate parameters . The computational analysis indicates that mitochondrial networks exhibit a percolation threshold . Intrinsic morphological instability of the mitochondrial reticulum resulting from its vicinity to the percolation transition is proposed as a novel mechanism that can be utilized by cells for optimizing their functional competence via dynamic remodeling of the chondriome . The detailed size distribution of the network components predicted by the dynamic graph representation introduces a relationship between chondriome characteristics and cell function . It forms a basis for understanding the architecture of mitochondria as a cell-wide but inhomogeneous organelle . Analysis of the reticulum adaptive configuration offers a direct clarification for its impact on numerous physiological processes strongly dependent on mitochondrial dynamics and organization , such as efficiency of cellular metabolism , tissue differentiation and aging .
Whereas experimental biology provides important insights into numerous characteristics and protein complexes responsible for cellular physiology , understanding the functional properties of many organelles can best be achieved when cell-wide structural complexity is included . This is provided by mathematical models able to explore a wide range of spatial and temporal scales . In the present study we describe the spontaneous emergence of a cell-wide network of mitochondria as an effect of a small set of dynamical rules well characterized in biological studies . Mitochondria are elongated intracellular organelles present in eukaryotic organisms ranging from yeasts to mammals . They are well known to produce the majority of cellular ATP , the universal form of energy required for most of cellular reactions and to be a critical checkpoint for the initiation of apoptosis . In the past few years mitochondria came into focus by ongoing discoveries of their central role in aging [1]–[3] , ischemia [4] , [5] , development of cancer and common neurological and metabolic diseases [6]–[9] - processes dependent on complex interactions between mitochondrial subunits . In the cell they are present in variable numbers ranging from few to hundreds of entities , forming a dynamic partially interconnected reticular network which spreads over the whole cytosolic volume excluding the nucleus . The network architecture is rather diverse and flexible , is able to adjust itself on a time scale of minutes depending on the actual physiological condition , and is highly variable among different cell types . In the fully interconnected state , the network edges are approximately of cylindrical shape with a typical diameter a few hundred nm and highly varying lengths up to more than 10 µm [10] . The opposite extreme is the fully fragmented condition , where the mitochondria are roughly spherical vesicles with diameter similar to that of the aforementioned cylinders . In the majority of situations , a cell contains many networked clusters of varying sizes along with numerous individual mitochondria , thus representing an intermediate state between the two extremes ( Fig . 1 A ) [11] , [12] . Quantitative description of this complex structure is currently not available , but it is thought to result from mitochondrial dynamics , governed by their constant intracellular motion along the cytoskeleton filaments and the ability to fuse and divide at varying positions at a time scale of minutes to hours [13]–[15] . The past years of experimental effort greatly enriched biochemical understanding of fusion and fission processes , which are accomplished via assembling the reaction-specific macromolecular complexes in mitochondrial membranes . Fission requires assembly of circular oligomers involving dynamin-like protein Drp1 attached to the outer membrane anchor Mff ( in mammals ) followed by GTP-dependent scission of the mitochondrial body perpendicular to the cylinder long axis . The fusion is performed by proteins ( in mammals these include Mfn1 , Mfn2 and Opa1 ) mediating tethering and subsequent connection of membranes surrounding the two input organelles and thereby generating a continuous body [16]–[19] . Despite variations in regulation and homology levels of the protein components , the above blueprint of the fusion/fission progression was found to be largely universal among different species [17] . Much less clear is what influence these elementary events have on the structural properties of the network as a whole and what their implications are regarding the functional efficiency of the cellular chondriome . Studies of other complex systems found that collective dynamics is often associated with phenomena not directly deducible from behavior of their isolated constituents [20] , [21] . In relation to mitochondria , this distinction is potentiated by experimentally established interconnection between the organelle’s functionality and its reticulum configuration [22] , [23 and Refs . herein] . Here , we report a whole-cell dynamical model of the chondriome able to capture both the recognizable variability of tissue-specific mitochondrial architectures and the network intrinsic flexibility in response to metabolic requirements , as observed experimentally [12] , [24] . First , we examine the mitochondrial reticulum by image analysis of fluorescence microscopy in order to identify the key types of fission and fusion processes responsible for the reticulum connectedness . Then , the network structure and dynamics are recreated numerically using both mean-field ordinary differential equations and explicit stochastic agent methods . The evolutionary graph-based representation introduces well-definable concepts making possible accurate characterization of mitochondria in quantitative terms . Its in-depth analysis shows that ( a ) the two types of events - fusion and fission of mitochondrial bodies - are sufficient to explain the observed diverse organization of the mitochondrial reticulum , ( b ) the geometrical properties of the network are directly related and can be calculated from the frequencies of these two processes ( and vice versa ) based on a few assumptions well supported by experimental data , ( c ) the cellular mitochondrial reticulum should exhibit a percolation transition and ( d ) its plasticity can be well explained by the vicinity of its functional regime to the critical point .
The major characteristics determining collective properties of mitochondria are the ability of these elongated organelles to form tubular threads here referred to as segments and to further interconnect them into branched net-like structures extending over the volume of the cytosol within a cell . This continually changing reticular arrangement is clearly observable with fluorescence microscopy ( Fig . 1 A ) . A comprehensive model of cellular chondriome must be based on a correct representation of essential dynamics of its components . Despite the overall agreement that the mitochondrial morphology is shaped by a delicate balance between fusion and fission [17] , [25] , different kinds of these processes are conceivable for the formation of three-dimensional tubular objects such as mitochondria . Which processes are actually active in living cells requires experimental elucidation . Upon noting that the underlying dynamic behavior of mitochondria leaves a distinctive and stable footprint on the resulting network structure , the details of fission and fusion processes can be deduced from the morphological still image analysis . This is because different types of fission and fusion mechanisms ( e . g . tip-to-tip in contrast to side-to-side fusion , etc . ) do involve process-specific kinds of network nodes . Segmentation processing of confocal microscopy images was employed here to experimentally establish the types of fission and fusion events forming the mitochondrial reticulum based on its structural properties . Configuration of a spatial network is specified by the connectedness of its branches at node points , expressed as node degree k ( number k of edges connecting the node to the rest of the network , Fig . 1 C ) and does not explicitly involve branch lengths or other geometrical attributes [26] , [27] . As a consequence , the mere presence or absence of nodes of degree k , rather than their relative abundance is sufficient in order to deduce a structure of the underlying dynamics and thus formulate a general model of the mitochondrial network and its evolution . The main advantage of this procedure is circumvention of the problems related to explicit 4-dimensional reconstruction of the network , not sufficiently reliable yet ( due to the limited resolution of the current microscopes in z-dimension , required to be less than the tubule radius , potentiated by the augmented noise resulting from fast scanning ) [28] . With the structural aspects kept fixed by the experimental assessment , the consecutive mathematical model will explore all the possible geometrical configurations by varying the reaction rates . This standpoint mirrors the available evidence that organisms and tissues share fundamentally similar kinds of mitochondrial fission and fusion events , despite the differences in protein structure or compositional details of the corresponding molecular machineries [16] . The image processing algorithm identifies the network segments and their connection points ( Materials and Methods , Fig . 1 B ) . Although some of the free mitochondrial ends ( i . e . nodes of degree k = 1 ) in the pictures are artifacts arising from the apparent cutting a reticulum branch by confocal slicing , false higher degree nodes cannot be produced that way , leaving the branching points ( k>2 ) unaffected . This enables the determination of the branching point types using single confocal sections per cell , therefore selecting the most advantageous focus position near the cover glass , where the reticulum structure is best resolvable ( Fig . 1 A ) . With a proper resolution , image voxel size has an effect on visible lengths of the mitochondrial tubules ( and hence on apparent fraction of the bulk nodes k = 2 ) but not on the reticulum branching organization . Because the network structure is determined solely by the branching node types , quantitative image analysis is restricted to these only , without experimental evaluation of the bulk nodes contribution ( inset at the bottom part of Fig . 1 B ) . On the other hand , in view of the importance of the nodes of degree two for the segment sizes and kinetics , these are included in the detailed model ( see the next Section ) . There , relative contribution of all the node types will emerge dynamically from the fission/fusion activity . Computational image analysis of mtGFP-harboring mitochondria in HeLa cells revealed an exclusive presence of apparent node degrees 1≤k≤4 ( n = 7 , see Materials and Methods for details ) . Complete lack of higher degree nodes indicates a very limited number of fission and fusion mechanisms involved in the network dynamics , in accordance with the small number of protein complex types known to perform these reactions [16] , [24] . Relative abundance of nodes of different degrees is important , because it reveals geometrical constraints favoring the node formation of particular degree . So , the mechanisms capable of generating the branching node types k = 3 and 4 are dissimilar and can be enumerated explicitly . Assuming single node formation/disruption times much shorter than the characteristic time of fusion/fission kinetics , the degree 3 nodes result exclusively from interactions of a mitochondrial tip ( k = 1 ) with a side surface ( k = 2 ) ( a ) , while the k = 4 nodes can only be created either by ( b ) fusion of a mitochondrial tip to an already existing branching site node ( k = 3 ) , or ( c ) by fusion of two bulk sites ( k = 2 ) in organelles touching each other side-to-side . Thus , comparative abundance of nodes of degrees 3 and 4 reflects the relative impact of the above reaction types on mitochondrial structure . For example , the detection of a high fraction of crossing tubules ( k = 4 ) would evidence for the significance of ( b ) or ( c ) for the network dynamics . Without the above assumption of fast elementary events , also interactions of three and four nodes would be conceivable , although experimental observations of such processes were not reported in the literature . Because of the expected small frequency of these higher-order interactions , the mathematical model utilizes events involving two nodes only: For example , generation of a k = 3 node from a triple of free tips would then involve a pair of tip-to-tip and tip-to-side fusions quickly following each other . Because the thickness of confocal slices cannot be made less than approximately two times the typical mitochondrial diameter , some of the detected branching nodes represent false positive connections resulting from the overlay of two unrelated organelles along the microscope optical path . However , this artifact can be corrected for by estimating its probability from the known volume fraction of the cytosol occupied by mitochondrial bodies and their diameter . In the examined micrographs , the occurrence of all apparent k = 4 nodes is 17 . 0±6 . 9% of the total branching points ( Supplementary Material Table S1 ) . The correction due to the aforementioned optical artifact reveals that at least 80% of those nodes with k = 4 result from the overlay of unrelated organelles . In contrast , for k = 3 the overlay contributes to 7% of the detected nodes , and for k<3 its effect is negligible . Hence , ∼96% of the actual branching points are of degree 3 , while the fraction of k = 4 nodes is statistically insignificant . Although measurements performed here are focused at peripheral cellular regions with lower density of microtubules , the extreme dominance of the simplest branching type strongly indicates that the general ability of mitochondria to perform complex fusions of type ( b ) and ( c ) is extremely low . We conclude that the spatial network of mitochondria is essentially connected with branching points of degree 3 resulting from the tip-to-side fusion activity ( a ) , consistent with reports using alternative methods [29] . Accordingly , the model discussed in the following is designed to exclusively exhibit nodes with 1≤k≤3 . We represent the mitochondrial reticulum with a graph ( nodes linked by edges , Fig . 1 C ) consisting of the following node types: free ends of mitochondrial segments ( k = 1 ) , bulk sites ( k = 2 ) and branching points ( k = 3 ) . Network edges interconnecting the nodes define minimal ( indivisible ) constituents of the organelle and their length l introduces a spatial scale to the network . The graph formalism does not imply actual physical existence of discrete subunits of uniform size l inside the mitochondrial bodies , but incorporates in a formal way their divisibility potential . Physically , l corresponds to the average distance between the membrane-bound fusion or fission complexes projected on the mitochondrial body axis . Finite value of l reflects the fact that at any moment the chondriome contains finite number of the molecular machines potentially capable of the reactions . When fluorescently stained , such complexes are directly observable as discrete punctuate structures on the surface of mitochondrial membranes [30] , [31] . Thus , the network edges reflect the maximal achievable fragmentation state of the chondriome , which disintegration into numerous vesicle-like tiny mitochondria is routinely observed experimentally and can be induced by strong downregulation of fusion or upregulation of fission [30] , [32] . In the model , this potential for fission is provided by bulk nodes interconnecting the edges into linear threads and by branching nodes . On the cellular scale ( >>l ) , the overall dynamics then leads to a structure ( Fig . 1 C ) similar to the real reticulum network as seen in Fig . 1 A . With the chondriome internally discretized by the edges and nodes , the graph-based formulation does not require any externally imposed coordinate system or lattice . As a whole , the network has L edges amounting to the total mitochondrial length in the cell . Because fission potential of the mitochondria is restricted by the finite number of the available molecular machineries , the discrete model reflects the biological situation better than continuous representation would do . In this organelle , the discretization parameter L representing the divisibility limit should be viewed as an important observable . For example , in the specific case of HeLa cell line , the total mitochondrial length as estimated by visual inspection corresponds roughly to 5000 µm . Because under the microscope mitochondria ( diameter ≈0 . 2 µm ) in the maximally fragmented configuration resemble spherical vesicles [30] , [32] , their diameter can be used as typical network edge length l = 0 . 2 µm , thus giving the value of the discretization parameter L = 3·104 ( ≈5000/0 . 2 ) . Keeping L as a free model parameter provides simple means for exploring the chondriome sizes relevant for different cell types . Applying the principle of minimal complexity sufficient for the creation of the above network structure , in general we postulate two fusion and two fission types [33] represented as reaction processes on nodes Xk ( Fig . 1 D ) : Thus , the biological fusion or fission processes correspond to the network node transformations ( Fig . 1 D ) governed by specific evolution rules ( Eq . 1 ) . The graph-based , non-spatial formulation of the reticulum has the advantage that it allows studying structural development imposed by fusion/fission dynamics while omitting the detailed consideration of the underlying regulatory pathways , which may be organism or tissue-specific . By not discriminating biochemical protein species responsible for the corresponding reactions , the processes of Eq . 1 are set to account for their cumulative phenomenological effect by reproducing the observed spatiotemporal dynamics of the reticular structure , rather than mechanochemical description of inter-protein interactions . Still , taking into consideration that in mitochondria only one type of fission molecular apparatus is found experimentally , the description can be further simplified by assigning equal probability of a fission event per network edge: b2 = ( 3/2 ) b1 ≡ ( 3/2 ) b . This connection of fission rates for bulk and branching nodes via the number of participating edges reflects the fact that in both cases the fission event occurs by scission across a tubular body of the organelle . Both the network dynamics and its equilibrium configurations result from intensities ( a[] , b ) of a few well-defined reactions , Eq . 1 . The current study considers in detail a wide range of constant a[] and b , which is sufficient for the characterization and explanation of the experimentally observed reticulum variability [11] , [12] , [16] , [18] , [30] , [32] . Notably , interpretation of the reaction propensities may be extended by turning them into explicit functions of time- and concentration-dependent protein-specific kinetic processes . This would allow examination of the upstream regulatory mechanisms . However , such an expansion would involve inclusion of extramitochondrial biochemical pathways , exceeding the scope of this report . The modeling framework proposed here ( Eq . 1 ) can be utilized as long as the assumptions of ( a ) the network topology ( 1≤k≤3 ) , and ( b ) absence of correlations between reaction events are fulfilled . This makes the model well suited for examination of mitochondrial networks , but would require major modifications in order to be applied to other cellular spatial networks such as endoplasmic reticulum . The condition ( b ) implies spatial isotropy and homogeneity inside the cell volume . Some anisotropy could arise in the very periphery of widely spread cells where microtubules may become arranged in bundles or preferentially oriented towards the cellular distal edge . So , care should be taken when applying the model to such cells . On very short spatial and temporal scale comparable with the kinetics of single motor proteins , mitochondrial fission and fusion may become also temporally correlated due to the influence of cytoskeleton . For example , two tips of mitochondria created by a recent division event but still attached to the same microtubule would have higher immediate propensity for fusion . However , such a deviation can occur only if the density of the cytoskeleton is sufficiently low , e . g . in the very periphery of the cell . In the bulk of the cytosol , frequent transitions between the fibers resulting from the high density of the cytoskeletal mesh common for the eukaryotes are expected to average the effect of single filaments out , leading to a fast decay of such correlations . This supposition was checked and confirmed by control simulations using an extended model where the mitochondrial network elements were assigned spatial positions by connecting them to explicit mesh of cytoskeletal fibers ( data not shown ) . For configurations and densities of microtubules typical for central cellular regions , the prevailing majority of mitochondrial segments were found in the immediate proximity to several differentially oriented fibers simultaneously , leading to their fast reorientation . Consequently , the well-mixed environment is sufficient for the present investigation , in which the extremely short time scales are not considered . Accounting for limited diffusivity in crowded cytosolic environment was shown to be important for accurate representation of molecular chemical systems due to their impact on effective kinetics relative to dilute conditions [34] . Mitochondria are much larger objects and are driven actively by motor proteins , but their mobility can still be affected by hindrances and other distortions commonly modeled by variations in diffusion coefficients . This kind of influence is accounted for in the current graph-based representation too , although a different approach is utilized . The dynamics of mitochondria is governed here by node transformation rates , and thus the diffusion in physical space is not explicitly implemented . Instead , the actual effective values for fusion/fission rates are taken from experimental measurements performed in living cells [15] , which overall account for all known and unknown factors affecting the dynamics , without discrimination between those internal and external to mitochondria . This corresponds to a well-mixed approximation and allows the model to reproduce the architectural build-up of the reticulum while avoiding ambivalences related to still scarce experimental data on complex motility patterns of mitochondria subject to multiple regulatory mechanisms . Importantly , the framework of dynamic quantitative graph introduced here can be utilized for an upgraded model where coordinates , velocities , and/or external forces are assigned explicitly to the network constituents . In addition to fusion/fission dynamics , the mitochondrial reticulum is subject to ongoing biological renovation in the course of import/export of material accompanied by organelle degradation due to selective mitochondrial autophagy ( mitophagy ) . Carried out by protein molecules quite uniformly distributed over the mitochondrial body , the import/export is not known to influence or to be sensitive to the network structure . The total mitochondrial mass varies synchronously with the cell mitotic cycle or as a response to signals changing gene expression patterns – both processes much slower than the fusion or fission [15] , [18] , [25] . Hence , value of L , the parameter accounting for the organelle size is assumed time-independent here . The mitophagy potentially could influence the size distribution by preferentially depleting very small clusters because it is limited to small ( ∼1 µm ) organelles . However , its impact on the mitochondrial architecture is negligible under physiological conditions because of the differences between time scales characteristic for the autophagy and the fission/fusion: while the newly imported or synthesized mitochondrial material was experimentally found to persist there for up to a few weeks , the mitochondrial motility is sufficiently fast to adapt the network conformation on the time scale of minutes to hours [13]–[15] , [35]–[37] . Hence , the reticulum will be considered here well-equilibrated from the material turnover point of view . Notably , the model formulation allows for an extension explicitly incorporating autophagy into alternative implementations specifically targeted for investigation of longer durations . This can be useful if properties exceeding the mitochondrial structural organization , such as quality control mechanisms , were to be examined in detail [2] , [38] . For the following , a mitochondrial segment is defined as one or more network edges connected only through bulk sites ( nodes of degree 2 ) , and a cluster as a detached set of interconnected segments possibly containing also branching nodes . When expressed in the numbers of nodes per cell xk , Eqs . 1 translate into a system of differential-algebraic equations governing the network dynamics: ( 2 ) The terms on the right-hand side of the first two of Eq . 1 represent the node kinetics due to each of the fission and fusion processes . For example , consumption of mitochondrial tips ( k = 1 nodes ) in the course of tip-to-tip fusion -a1x1 ( x1-1 ) is a product of the total number of node pairs potentially capable of fusion x1 ( x1-1 ) /2 and the fusion propensity for a pair 2a1 . The last of Eq . 2 reflects conservation of the mitochondrial mass in the cell expressed as total number of edges L . Technically , it restricts the system phase space x ≡ ( x1 , x2 , x3 ) to a plane which position and orientation are uniquely specified by L and the network structure k ≡ ( 1 , 2 , 3 ) respectively . While changing the applied discretization L of the model merely shifts the phase plane parallel to itself , relative proportions of the nodes and thus the state of the reticulum as a whole are well defined by the parameterization parameters ( a[] , b ) . Alteration of the reticulum discretization L corresponds to a different expression level of fission/fusion complexes without changing their relative abundance , while the values of a[] and b reflect independent activities of each of the network transformation processes . The steady state of the system is a function of the ratio of fusion and fission rates c1 ≡ a1/b and c2 ≡ a2/b rather than the rates themselves: ( 3 ) A unique set of real non-negative solutions x describing the network state corresponds to each triple of parameters ( c1 , c2 , L ) . Fig . 2 exemplifies the steady-state solutions ( Eq . 3 ) for a particular chondriome size corresponding to HeLa cell line ( L = 3·104 , see previous Section ) and a wide range of ( c1 , c2 ) . In the extreme of infinitely strong tip-to-side fusions c2→+∞ , the network tends to a fully connected “crystal” x → ( 0 , 0 , ( 2/3 ) L ) consisting of branching sites only . Because in this regime the reticulum has no other node types , its geometry is insensitive to c1 . The opposite case ( c2 → 0 ) allows for many different configurations specified by the ratio of tip-to-tip fusion and fission rates c1: the set of possible states is flanked here by a fully fragmented x → ( 2L , 0 , 0 ) , c1→0 and a single loop x → ( 0 , L , 0 ) , c1→+∞ . Reticulum states corresponding to some of these extremes were experimentally induced by artificial manipulation of mitochondrial fission or fusion activities [30] , [32] . However , physiologically the most relevant parameter range is fairly narrow ( see below ) , with all xk being far from saturation ( Fig . 2 ) . The deterministic description of the mitochondrial geometry ( Eq . 2 ) establishes a well-defined analytical connection between the molecular biochemical parameters ( a[·] , b , L ) and network-wide structural variables like the average numbers of nodes xk , mean segment length for non-loop segments and total number of segments ( x1+3x3 ) /2 comprising the reticulum . These represent the simplest quantitative measures of the chondriome architecture . They can be determined experimentally from the analysis of high-resolution cell reconstructions recorded with sub-diffraction microscopy , which is currently being extensively developed and shall become available in biological laboratories [39]–[41] . This type of recordings could also go beyond the mean values and measure among others the probability distributions of mitochondrial segment lengths and cluster sizes ( discussed below ) . In addition to a more complete characterization , the latter variables may prove critical for clarification of key features of this organelle by revealing potential effects of stochastic fluctuations on mitochondrial operation and homeostasis ( see the Section “Percolation phase transition in the mitochondrial reticulum” below ) . In order to obtain a more detailed insight into the expected mitochondrial reticulum architecture , an agent-based stochastic simulation [42] of the same system was developed . As was verified by investigation of the potential fusion points in a setting with explicit representation of the cytoskeleton ( data not shown ) , for the long-term evolution a non-spatial approximation is justified inside the majority of the cytosolic volume . Hence , a stochastic model was developed , where a set of L reactant objects corresponding to the network edges is submitted to the processes of fission and fusion as above ( Eq . 1 ) , i . e . between random nodes without explicit positioning in space . Reaction events and timings are put under the operation of the Gillespie algorithm [43] where nodes participating in a particular event are chosen randomly with equal probability among nodes of the same type . In the absence of detailed biochemical data on fission and fusion rates , the simulation parameters were adjusted to reproduce the experimentally observed average frequencies of fusion and fission events ≈0 . 25 ( cell·sec . ) −1 [15] . Explicit representation of the cellular mitochondrial system within the agent model allows for a comprehensive insight into the network characteristics surpassing the mean-field approximation of the deterministic description above . Hence , the following discussion is focused on the stochastic properties of the reticulum expressed in terms of statistical distributions . Alternatively to the node-based description , the discussed network can also be viewed as a set of four types of ( partially ) connected segments , discriminated by degrees of the two end nodes ( Fig . 1 E ) . The network state is then characterized by segment numbers of each of the four types: separate open-end segments , separate loops , as well as surface and internal segments of branched clusters . Index i denotes the segment length measured in edges . Using the agent-based model , we find that with a good accuracy the steady-state distributions of mitochondrial segment lengths ( Fig . 3 A ) can be expressed as a superposition of two qualitatively different , fast and slow decaying , terms with their relative strength being strongly dependent on c1 and less on c2 . The two components can be examined analytically by considering a simplified network where tip-to-side fusions are switched off ( i . e . c2 = 0 ) . As the degree of nodes is restricted here to k = 1 and 2 , such a network consists of ( 1-1 ) and ( 2-2 ) segment types only , and equations governing their dynamics can be formulated explicitly . The segment numbers of length i and are derived by taking into account all possible transitions between segment populations: ( 4 ) Here the prime under the sum denotes omission of the term j = i . The consecutive terms on the right-hand side of the Eq . for correspond to ( 1 ) creation of the segment resulting from fission of longer segments , ( 2 , 3 ) disruption and creation due to fission of same-size open-end segments and loops respectively , ( 4 ) formation upon fusion of shorter segments , and ( 5 , 6 ) disruption resulting from fusion to different and same size segments , respectively , and ( 7 ) formation of loops . Expressions for the steady-state segment length distributions in this network reveal unambiguous distinction in length distributions between loops and open-ended segments ( Fig . 3 B ) : ( 5 ) with being the total number of open-ended segments . While Eq . 5 takes into account finite-size effects , as anticipated in real cells , for an idealized infinite system one would expect a geometrical distribution of because of the uniform probability density for a segment production or disruption over bulk nodes . On the other hand , the sizes of loops are governed by power-law . As a result , among the small segments the number of loops strongly exceeds the amount of the open-ended segments , while for the long segments the proportions are reversed . In the general network ( c2>0 ) , the tip-to-side fusion and fission generates branched clusters consisting of variable segment numbers . The possibility of transitions between them induces a mild deviation from Eq . 5 . Still , for all segment types other than disconnected loops ( 2-2 ) , the geometric distribution , where p is the probability of a segment end provides a good approximation to the agent-based result . The mean length of segments can be estimated from the number of nodes introduced in the deterministic model ( Fig . 3 C , solid line ) and is therefore related to fusion and fission rates through Eq . 3 . In the equilibrated system , is essentially equal among all non-loop segment types or cluster sizes ( Fig . 3 C ) . Importantly , due to the strong decrease of at lower i ( cf . Eq . 5 , Fig . 3 B ) , loop segments amount to only a few percent of the total mitochondrial mass ( Fig . 3 C , crosses ) , hence the mean length can be viewed as a good network-wide measure of mitochondrial reticulum structure ( Fig . 3 C , solid line ) . The geometric law of segment lengths is a direct consequence of the balance between opposing actions of fusion and fission . The pattern results from the positional independence of fission sites along the body of a mitochondrial tube , accompanied by the equal chances of network nodes to become involved in a fusion event ( i . e . with probability being independent from the lengths of the segments ) . This corresponds to the dynamic behavior of a well-mixed system , as expected upon a cell-wide equilibration of the chondriome resulting from the permanent motion of individual mitochondria along the cytoskeleton filaments [13] , [14] . For general values of c1 and c2 the system contains numerous segment clusters of variable sizes . Because the lengths of segments in a cluster are geometrically distributed independent random variables , the sizes j of clusters ( i . e . the total number of network edges within a cluster ) with r segments ( Fig . 4 B ) in the agent-based system conform ( Fig . 4 A , colored markers ) to the negative binomial distribution ( Fig . 4 A , dashed lines ) ( 6 ) where p is the probability of a segment end . The integral size distribution of all clusters in the system ( 7 ) involves weighting by the number of r-segment clusters g ( r ) , which has no known closed form expression for this network geometry . Yet , upon using the agent-simulated g ( r ) together with Eq . 6 , one recovers n ( j ) numerically ( Fig . 4 A , black solid line , grey stars ) . Variations in cluster sizes n ( j ) expected in a single cell are illustrated in the scatter plot of cluster sizes vs . number of segments in the cluster , for a set of different tip-to-side fusion rates c2 ( Fig . 4 C ) . The dependences can be viewed as linear , with their coefficients ( slopes of the scatter ) corresponding to well-defined mean segment lengths discussed above . Increase in c2 leads to the conversion of bulk sites into branching sites and the resultant gradual reduction of ( see also Fig . 3 C ) accompanied by a higher segment number . Much less intuitive are c2-dependent nonlinear changes in variance ( spread ) of the cluster sizes ( Fig . 4 C ) discussed in the following . For the simplified dynamics considered above ( c2 = 0 , Eqs . 4 , 5 ) , n ( j , r ) would be reduced to a mixture of geometric and power laws ( see previous Section ) , because such a network consists from single-segment clusters only . The presented reticulum model ( Eq . 1 ) exhibits a percolation phase transition in parameter c2 ( tip-to-side fusion/fission ratio ) . In the thermodynamic limit of infinitely large network , the transition over a percolation threshold would correspond to an abrupt formation of a giant ( percolating ) cluster . At critical value a qualitatively new structure arises - a global cluster created from otherwise unconnected chondriome components spreading over the remote intracellular regions ( Fig . 5 , blue stars and schemes ) . In the finite network of a real cell the abrupt change of the network order is smoothed in the vicinity of the critical point , but rapid emergence and then dominance of the giant cluster for c2> is manifested when the fractional size of the largest mitochondrial cluster is plotted as a function of c2 while keeping c1 constant . Continuous ( second order ) phase transitions are marked by a peak in size fluctuations or other susceptibility measures at the point of criticality [44] . In the modeled reticulum the peak is visible as a sharp change in variability of non-percolating cluster sizes , e . g . expressed as the average number of segments per cluster ( Fig . 5 , pink circles ) . In a scatter plot , the rapid rise in cluster size fluctuations near the transition point is undoubtedly noticeable as elevated spread in sizes of individual clusters comprising the cellular chondriome ( Fig . 4 C , yellow and green versus red and blue ) . In contrast , no phase transition is found in tip-to-tip rate c1 , where the critical point should be expected only at c1 →+∞ , similar to a one-dimensional percolation problem on the lattice [44] . This result can be intuitively understood: While increase in tip-to-tip fusion rate c1 merely raises the average segment length , the tip-to-side fusions ( c2 ) interconnect the segments among themselves and thus promote the formation of a global networked structure . Although from a mathematical perspective the predicted criticality is not surprising given the network topology assumed in the model , its relevance for the structure of mitochondria may have long-standing implications for cellular organization and functionality . The predicted phase transition implies two distinct classes of mitochondrial structures: a subcritical network consisting of a relatively uniform set of multiple disconnected mitochondria , as opposed to a supercritical one characterized by the presence of a dominant giant cluster accumulating the majority of the mitochondrial mass , accompanied by a few much smaller satellite mitochondria ( Fig . 5 bottom part ) . On one hand , from the experimental perspective , because only a relatively narrow range of c2 values corresponds to the transitional regime , the great majority of cells under unselective conditions could be expected to be found far from the transition point . On the other hand , frequent observation of intermediate chondriomes would indicate an underlying regulatory or self-tuning mechanism selectively promoting quasi-critical configurations . Both highly fragmented and highly interconnected structures were induced experimentally in several cell types with pharmacological treatment severely shifting the fission/fusion balance [30] , [32] . However , quantitative experimental investigations of the chondriome's undisturbed geometrical organization are , to our knowledge , not yet available . Here , we preliminarily examined the relevance of the predicted percolation transition for chondriome structure by measuring relative sizes of disconnected mitochondrial clusters in confocal images analyzed in the first Section ( see also Materials and Methods ) . We find that in a standard mammalian cell line ( HeLa , Supplementary Material Table S1 ) , the largest mitochondrial cluster comprises 35% of the total visible reticulum ( Fig . 5 , dashed line ) , a fraction far from both the high-end and low-end saturation echelons expected for single-phase configurations ( Fig . 5 , blue stars ) . This value is accompanied by high ( st . dev . 20% ) diversity among cells , pointing to an elevated level of fluctuations in the cluster sizes . This result is further supported by the examination of relative sizes of mitochondrial clusters belonging to the same cell ( data not shown ) . Both parameters indicate that under the normal physiological conditions the mitochondrial network may operate near the percolation transition point . However , due to small sampling of this preliminary measurement , currently only a compatibility with this interpretation can be stated . A more rigorous validation of the percolation transition predicted here will be achievable , when precise tuning of fission and fusion frequencies along with three-dimensional reconstruction of mitochondrial networks in living cells becomes biologically possible .
Growing experimental evidence indicates the fundamental interdependence between mitochondrial metabolic activity and its network structure [22] , [23 and Refs . herein] . Yet , until now , the theoretical analysis of this vital organelle was either focused on its biochemical and electrophysiological aspects [45]–[47] , or was reducing its architecture to a set of linear objects [48] . The examples of other complex systems indicate , that the manifestation of collective properties of the mitochondrial net should be critically dependent on proper representation of its structure and dimensionality [20] , [21] , [44] , [49] , requiring a cell-wide multiscale formulation . The current work examined the chondriome organization experimentally and introduced its network-based mathematical representation . This led to a detailed insight into the organelle , emerging as a mesh of tubular segments interconnected into larger flexible clusters able to reach distant cellular regions . We find that fusion and fission dynamics should lead to a branched reticulum of tubules which lengths are well approximated by a geometric law and which mean size in equilibrium is determined by relative rates of these processes ( Eq . 3 ) . The whole network consists of disconnected clusters of such tubules . The cluster sizes are well approximated with a superposition of negative binomial distributions ( Eqs . 5 , 6 ) . Notably , the distribution shape is distinctly convex ( Fig . 4 A ) , featuring numerous tiny clusters coexisting along with a few relatively large entities . This property is expected to promote the experimentally observed disposal of damaged mitochondria by cellular autophagosomes [2] , [50]–[52] . One reason for this is that mitochondrial dysfunction was shown to inactivate the mitochondrial fusion machinery , consequently leading to smaller mitochondrial entities [53] , [54] . In addition , prevalence of small clusters present in the network supports the formation of autophagosomal bodies , which are not able to engulf objects larger than a few µm in mammals [37] , [55] . Moreover , smaller clusters are expected to exhibit high statistical variance in functional efficacy facilitating the determination of removal candidates based on the inner membrane potential gradient or similar markers . In this way , the viability of an organism could be optimized by maintaining homeostasis of high-quality mitochondrial material on the whole-cell level [38] . Gradual disruption of this quality control e . g . as a result of natural aging is known to coincide with simultaneous rearrangement of mitochondrial reticulum structure due to alteration of fission and fusion rates [2] , [15] , [51] . By experimentally manipulating mitochondrial dynamics , the network reorganization was found to be sufficient for the induction of a substantial slowing down of the aging process [1] . The cluster size distribution predicted here offers a quantitative explanation for these and similar observations related to the strong dependence of mitochondrial structure and function [23] . To what extent the ongoing network dynamics is able to smooth out the differences constantly arising from diverse functional activity in distant parts of the chondriome ? On a shorter time scale ( ∼minutes ) , the ongoing fission prevents complete homogeneity of the mitochondria over the cell body . Inside the mitochondrial clusters , additional equilibration results from molecular diffusion along the organelle tubules [56] . The geometric ( exponential ) distribution of the segment lengths predicted here is characterized by a high variance , with very long tubules connected to multiple shorter ones . It would be interesting to check what biological implications this diversity has for the organelle performance . For example , taking into account that key mitochondrial proteins in the inner membrane tend to compartmentalize while their diffusion is very slow [56]–[59] , the presence of very long segments could significantly hinder equilibration of compositional differences between reticulum branches . The balanced fusion and fission dynamics as in Eq . 1 leads to a network capable of a phase transition , i . e . that possessing two qualitatively dissimilar organizational modes . The critical transitional region lays in a narrow range of tip-to-side fusion/fission rates ( c2 , Fig . 5 ) where the reticulum structure is able to rapidly change its configuration , i . e . is very susceptible to the proper balance of these opposing processes . An experimental examination of the chondriome structure in HeLa cells reveals that its geometry corresponds to the transitional regime , characterized by the maximal heterogeneity in sizes of the network subcomponents . If this result is confirmed for other cell types , this would imply that under normal physiological conditions the cell has to maintain the fission and fusion rates quite precisely matched , despite the necessary flexibility in other parameters . The tight positioning inside the narrow transitional region rather than in one of the numerous configurations away from the critical point would induce questions about factors responsible for such a specific arrangement . Their assessment exceeds the scope of the current study , whilst the details of the network dynamics and architecture examined here could serve as an important counterpart . Similar combinations of high susceptibility and phenotypic robustness were found in other complex adaptive systems [60]–[62] . Independently from the underlying factors , it would be , indeed , advantageous for a cell to operate in the vicinity of the critical point because here the mitochondrial reticulum can be reconfigured with minimal energetic and temporal cost . Examples for such transformations include the quick and radical mitochondrial fragmentation upon activation of the apoptotic cascade or in the course of mitosis , where it is essential for a proper partitioning of the organelle between the daughter cells [63] , [64] . Furthermore , high susceptibility of the mitochondrial network in the critical regime to small changes of the branching parameter naturally generates a high diversity between cells . Well known experimentally , it was often discarded as inessential intercellular noise amid a homogeneous population . However , this view is now changing due to recent observations connecting cellular structural aspects to its functional and gene expression patterns [65] . For example , in the course of organogenesis , the narrowly positioned regime of elevated flexibility can facilitate the tissue-specific reticulum alteration . With such a mechanism in place , during the differentiation phase cells can attain chondriomes best suitable for specialized energetic needs , inducing the variability of mitochondrial geometries found in different tissues and cell types [11] . In conclusion , the proposed model explains the self-organization of the chondriome into a dynamic network and its operation as a cell-wide adjustable construct , which large-scale characteristics provide the necessary connection between microscopical biochemical parameters and qualitative features central for the functionality of living cells . The predicted size distributions of segments and clusters are easily interpreted in physiological terms and verifiable by corresponding experiments .
HeLa cells were grown in Dulbecco Minimum Essential Medium ( Sigma ) , supplied with 10% FCS ( PAA ) and 1% Penicillin Streptomycin ( Sigma ) . Effectene ( Qiagen ) was used for transient transfection in 3 . 5 cm Petri dishes ( IBIDI ) according to the manufacturer's instructions . mtGFP experiments were performed 24 h after transfection . 4- ( 2-hydroxyethyl ) -1-piperazineethanesulfonic acid ( HEPES , 10 mM final concentration ) was added to the medium 1 h before measurements . Live cell imaging was done at 37°C using Nikon Eclipse TE2000-E microscope . A small pinhole ensured that the thickness of confocal slices did not exceed 800 nm . In order to reduce the effects of perinuclear region irregularities , the system was focused to a fraction of the intracellular volume adjacent to the Petri dish bottom . Raw confocal images of HeLa cells ( Fig . 1 A ) with fluorescently visualized mitochondria were subjected to digital analysis designed to determine the reticulum network structure . As the first step , the images were thresholded and skeletonized after contrast optimization . The threshold position was chosen for each image such that no mitochondrial signal was lost while the background was cut off . Subsequently , the binary maps representing spatial graphs of the reticulum resulting from the skeletonization operation ( Fig . 1 B , main field ) were treated with a segmentation algorithm designed to identify and statistically analyze mitochondrial linear segments and branching points interconnecting them into clusters . An input skeletonized image consists of white pixels ( 1 ) on the background ( 0 ) ( Fig . 1 B ) , the former classified as nodes of the mitochondrial network . Two white pixels were considered forming a connected graph if they were found adjacent to each other by applying an 8-connectivity criterion [66] . Degrees k of the resulting network nodes were calculated as the number of neighbors adjacent to the corresponding white pixel and , if appropriate , corrected to account for oversampling . The segmentation algorithm scanned the map starting from one of its edges and proceeded line by line . Upon encountering a white pixel , the graph to which it belongs was followed using a depth-first search method [67] until the whole cluster of adjacent pixels was traversed ( upper box in Fig . 1 B shows the segmented structure ) . These pixels were then excluded from further search and the procedure advanced until the whole image matrix was processed . The image processing algorithm requires no initial assumptions regarding possible distributions of the reticulum segment lengths , their clustering or extent of connectedness . Image contrast adjustment , thresholding and skeletonization were done using ImageJ ( US National Institutes of Health , Bethesda , MD ) . Segmentation analysis algorithms , statistical and visualization procedures , as well as ordinary differential equation numerical solutions were implemented in MATLAB ( The MathWorks , Natick , MA ) . Stochastic agent-based model of the mitochondrial network was designed using Intel Corp . ( Santa Clara , CA ) C++ compiler and run under Linux v2 . 6 kernel . Random numbers were generated using VSL routines , part of Intel Corp . ( Santa Clara , CA ) Math Kernel Library . For non-commercial use , the computer files comprising the agent-based model can be obtained free of charge upon contacting one of the corresponding authors ( the e-mail addresses are given on the first page ) .
|
Mitochondria control energy production , initiation of cell death and several other critical cellular processes . Most often , they form a constantly reshaping tubular reticulum spread over the cytosol . Despite extensive knowledge of mitochondrial physiology , accurate description of their large-scale architecture is missing , partly due to substantial variability of reticulum geometries found in different cell types , and a capability for fast radical changes . We address this shortcoming with a mathematical model representing the organelle as a cell-wide dynamical network subjected to opposing actions of fission and internal fusion – processes known experimentally but not yet accurately specified . This opens a way for the quantitative characterization of the large-scale organization by showing how particular types of the internal dynamics can shape the reticulum into the whole multitude of configurations observed in biological studies . Further analysis reveals that for a specific value of tip-to-side fission/fusion rates the network should undergo a radical reorganization . Because of the high morphological sensitivity to minute changes in fusion or fission rates close to the critical point , cells can quickly adapt the mitochondrial operation and structure to their actual needs at a low expenditure of energy .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"death",
"cellular",
"structures",
"subcellular",
"organelles",
"cellular",
"stress",
"responses",
"synthetic",
"biology",
"biophysics",
"simulations",
"theoretical",
"biology",
"biology",
"biophysics",
"systems",
"biology",
"cell",
"biology",
"biophysic",
"al",
"simulations",
"computational",
"biology",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] |
2012
|
Emergence of the Mitochondrial Reticulum from Fission and Fusion Dynamics
|
Soil-transmitted helminths ( STH ) – a class of parasites that affect billions of people – can be mitigated using mass drug administration , though reinfection following treatment occurs within a few months . Improvements to water , sanitation and hygiene ( WASH ) likely provide sustained benefit , but few rigorous studies have evaluated the specific WASH components most influential in reducing infection . There is a need for alternative analytic approaches to help identify , characterize and further refine the WASH components that are most important to STH reinfection . Traditional epidemiological approaches are not well-suited for assessing the complex and highly correlated relationships commonly seen in WASH . We introduce two recursive partitioning approaches: classification and regression trees ( C&RT ) and conditional inference trees ( CIT ) , which can be used to identify complex interactions between WASH indicators and identify sub-populations that may be susceptible to STH reinfection . We illustrate the advantages and disadvantages of these approaches utilizing school- and household-level WASH indicators gathered as part of a school-based randomized control trial in Kenya that measured STH reinfection of pupils 10 months following deworming treatment . C&RT and CIT analyses resulted in strikingly different decision trees . C&RT may be the preferred approach if interest lies in using WASH indicators to classify individuals or communities as STH infected or uninfected , whereas CIT is most appropriate for identifying WASH indicators that may be causally associated with STH infection . Both tools are well-suited for identifying complex interactions among WASH indicators . C&RT and CIT are two analytic approaches that may offer valuable insight regarding the identification , selection and refinement of WASH indicators and their interactions with regards to STH control programs; however , they represent solutions to two distinct research questions and careful consideration should be made before deciding which approach is most appropriate .
Infection with soil-transmitted helminths ( STH ) , intestinal nematodes , is classified by the World Health Organization ( WHO ) as a neglected tropical disease ( NTD ) . More than 1 billion people are infected and up to 5 . 3 billion are at risk of infection with at least one species of STH , including roundworm ( Ascaris lumbricoides ) , whipworm ( Trichuris trichiura ) , or hookworm ( Necator americanus or Ancylostoma duodenale ) [1]–[3] . STH infection occurs through fecal exposure , either through the skin in contaminated soil ( in the case of hookworm ) or ingestion of fecal material , typically in soil , on food , or on fingers [4] . Morbidity is most acute in school-age children , though high levels of hookworm infection can persist into adulthood [4] . It is estimated that between 5 and 39 million disability adjusted life years are lost due to STH infection [5] , [6] . Though a recent review found limited evidence [7] , STH infections have been found to impact on growth and nutrition of children [8] and reduce pupil absence in some studies [9] , [10] . Control of STH is a priority for the WHO [11] and several countries , including Kenya , are scaling up mass drug administration in school-age children to reduce STH-related morbidity [12] , [13] . These infections can be treated safely and effectively with the anthelminthic drugs albendazole or mebendazole [4] , [14] . However , in the absence of improved access to water , sanitation , and hygiene ( WASH ) , reinfection occurs and the prevalence and intensity of infection can reach pre-treatment levels in as few as six months , with 94% reinfection after 12 months [15] . Access to WASH includes hardware – such as toilet facilities that separate human feces , protected water supply , and soap – as well as behaviors , such as hand washing at key times and toilet use . The UNICEF and WHO Joint Monitoring Program ( JMP ) is the most widely cited source of data on what is considered “improved” water supply and sanitation [16] , but the JMP does not provide guidance on hand washing , nor are its definitions specific to STH control . Even in countries with moderate access to improved water and sanitation in sub-Saharan Africa there is considerable geographic inequity [17]; these same marginalized populations without access are the ones with high risk of STH [18] . WASH components thought to be most critical for control of STH are the use of a clean toilet facility and the presence of water and soap for hand washing; however , few randomized trials have been conducted to assess the relationship between WASH and STH infection . Three randomized controlled trials have found evidence that improved hand washing with soap can lead to lower STH infection [19]–[21] . Nonetheless , in a study by Dumba et al . , researchers did not find any impact of a participatory hygiene and sanitation transformation ( PHAST ) intervention compared with a control group that received deworming alone [22] . A recent meta-analysis of 36 , mostly observational , studies suggested that access to and use of sanitation facilities is associated with significant reductions in the prevalence of STH infection , with an odds ratio [OR] of 0 . 54 ( 95% CI: 0 . 43–0 . 69 ) for A . lumbricoides , 0 . 58 ( 95% CI: 0 . 45–0 . 75 ) for T . trichiura , and 0 . 60 ( 95% CI: 0 . 48–0 . 75 ) for hookworm [23] . In a separate meta-analysis , soap use ( OR: 0 . 53 , 95% CI: 0 . 29–0 . 98 ) , wearing shoes ( OR: 0 . 38 , 95% CI: 0 . 18–0 . 81 ) and drinking treated water ( OR: 0 . 45 , 95% CI: 0 . 36–0 . 58 ) were associated with lower STH infection [24] . Access to piped water was associated with lower infection with A . lumbricoides ( OR:0 . 39 , 95% CI: 0 . 39–0 . 41 ) and T . trichiura ( OR: 0 . 57 , 95% CI: 0 . 45–0 . 72 ) . However , because nearly all studies in these meta-analyses were observational , it was not possible to disentangle the impacts of individual WASH components or the relationship between WASH and socio-economic status , potentially biasing many of these results . WHO has set the goal of elimination of STH as a public health problem by 2020 , which is provisionally defined as a prevalence of moderate- and high-intensity STH infection of <1% ( WHO , 2012 ) . To achieve this goal , and to sustain the gains made possible through mass drug administration , WASH improvements and intersectoral collaboration will be critical [11] , [25] . However , identifying and characterizing those WASH components that are most effective at reducing or preventing STH infection is non-trivial , in part because of the ethical challenges of conducting randomized control trials which are necessary for establishing causal relationships [24] , and yet will be essential for developing evidence on the success of STH control programs [18] . One challenge is that access to the different components of WASH in both the public and private sphere is highly interrelated , and little is known about the relative contributions of each independent WASH component in mitigating infection with STH . Furthermore , readily measurable WASH components relevant for STH control have not been identified or validated . Indeed , current WHO guidelines for STH control refer to WASH in general terms [11] , [26] . The vast majority of studies examining the association between WASH components and STH infection have considered the main effects; however , because of the inherent connectedness of WASH components – e . g . water must be present for hand washing to occur – it is also critical to consider interactions . The number of potentially measurable WASH components is quite large , and when one also considers all the potential first , second- , and higher-order interaction terms , most datasets would not have sufficient power to detect all important associations using standard analytic approaches . A need exists to identify alternative analytic approaches to help identify , characterize , and further refine those WASH components that are most important to STH infection . The goal of this analysis is to introduce two analytic approaches that are relatively new to the NTD and WASH communities: classification and regression trees ( C&RT ) and conditional inference trees ( CIT ) . Both C&RT and CIT are a type of recursive partitioning , a nonparametric analytic approach well-suited for handling datasets with large numbers of predictor variables , identifying complex interactions , and selecting independent variables that are most predictive of or associated with the outcome [27] . These approaches are particularly useful for hypothesis generation and as a precursor to other model building approaches . We demonstrate how both methods can be applied to a dataset measuring household- and school-level WASH components and STH infection in Kenyan school children . This is a secondary analysis of the data; the primary results from this study have been reported elsewhere [9] . We discuss the relative merits and weaknesses of each approach and make recommendations for their uses .
Data collection for this study was approved by the Institutional Review Board at Emory University and the Ethics Committee at Great Lakes University of Kisumu ( Kenya ) . We obtained a loco parentis from the head teacher at each school . Children provided oral consent to participate in this study , which was documented on the electronic data collection form . The ethics committees approved both a waiver of parental consent and the use of oral consent for study participants . This study utilized data from a cluster-randomized trial to assess the impact of improved school and household WASH access on STH infection in Nyanza Province , Kenya from 2007–2009 [28] . Data for this analysis were collected in February 2009 – the final survey round of the trial – from 1 , 106 students in 39 public primary schools ( Checklist S1: STROBE Checklist ) . Twenty of these schools had been randomly selected to receive a school-based WASH intervention that included construction of ventilated-improved pit latrine facilities at the school , hand washing and drinking water storage containers , teacher training on hygiene behavior change , and a one-year supply of dilute sodium-hypochlorite used for treatment of drinking water at the point of use . Pupils in all schools – both intervention and control – were dewormed at baseline ( May , 2007 ) and midterm ( April , 2008 ) using 400 mg of albendazole . Pupils from grades 3 to 5 who were between the ages of 7 and 13 and had been dewormed during the previous round of data collection were randomly selected and enrolled into the original trial . The mean number of pupils was 302 and 275 in the intervention and control schools , respectively . This age group was selected because they experience the greatest burden of A . lumbricoides and T . trichiura , though peak morbidity for hookworm occurs later [29]–[31] . Systematic random selection of 30 pupils was conducted using a list of pupils from the school records , though some pupils were absent the day of the study . Only one child per household was enrolled to avoid the need to adjust for intra-household correlations . Of the 1106 students included in the study , 1095 provided a single analyzable stool and had valid Kato-Katz results . The original sample included pupils from 40 schools ( 20 intervention and 20 control ) . However , one control school was dropped from the analysis after children were treated with an additional round of deworming drugs . Stool samples were collected and transported to the laboratory in cool boxes and examined microscopically within one hour of preparation using the Kato-Katz method [32] . Each stool sample was processed on two separate slides and read by different laboratory technicians to ascertain the eggs per gram of each STH species . Presence of infection was defined as detection of one or more eggs on either slide . Because all individuals were dewormed 10 months prior to the study , any infection observed was interpreted as incident infection . This analysis includes data on pupils from both intervention and control arms . Data on individual demographics , household WASH conditions , and school WASH conditions were collected using structured observations and questionnaires . Pupils were interviewed to determine their age , sex , shoe wearing , comfort using the latrine at home and school , knowledge on hand washing and water supply treatment , opinion about latrine conditions , access to hand washing and drinking water at school , and their soil eating behavior ( known as pica or geophagy ) , a common practice in western Kenya [33] . As a complement to the direct observations made at each school , pupil responses regarding school-based access to drinking water , hand washing water , soap , and latrines were aggregated at the school-level as an estimate of school WASH access . One caregiver – typically the maternal head of household – for each pupil enrolled in the study was interviewed in his or her home to determine if one or both parents was alive and , if alive , the highest level of education achieved , socio-economic status through an asset index; access to an improved drinking water source , as defined by UNICEF and WHO [34]; if treated water is used; and the presence and condition of a household latrine . School head teachers were interviewed about the school's access to an improved water source during the dry season , pupil to latrine ratio , and latrine conditions . Enrollment data were taken from official school records . In order to calculate socio-economic status , we used a principal component analysis ( PCA ) using assets observed at the household [35] . These assets included household construction materials , ownership of goods such as a TV and radio , and connection to electricity [36] . PCA was also used to construct an index of sanitation conditions at both the household and school , which included odor , presence of flies , presence of feces , wall material , condition of the slab , and presence of a functioning door . These components were put on a scale from 1–4 and the resulting value was a relativistic score of the average conditions for all latrines at the school or for the latrine at home . Two scores – latrine cleanliness and latrine structure – were derived based on factor loading . We did not include the score for latrine conditions at the home in our tree analysis , since that would limit the analysis to children with latrines at home . Acceptable latrines were classified as those for which no parameter scored in the lowest two values for each of the five sanitation categories . This study compares two different recursive partitioning approaches: C&RT and CIT . Recursive partitioning is a nonparametric regression approach; it is a form of hierarchical clustering in which the data are sequentially split into dichotomous groups such that each resulting group contains increasingly similar responses for the outcome [37] , [38] . Recursive partitioning has several advantages over traditional logistic regression . C&RT and CIT are supervised clustering approaches; they create partitions based on an outcome variable , as opposed to other clustering approaches such as k-means and PCA , which do not involve the outcome [39] , . As nonparametric approaches , C&RT and CIT make no assumption of a monotonic or parametric relationship with the outcome , can be used to identify complex interactions among the independent variables without a priori specification of interaction terms , and can handle datasets where the number of independent variables is high relative to the number of observations . This final feature is particularly attractive to studies such as this , where a goal is to identify a few best predictors from many . Both C&RT and CIT result in the formation of a decision tree with three levels consisting of a root node , internal nodes , and terminal nodes . Every tree starts with a “root node” that contains the sample of data from which the tree will be grown ( e . g . the study population ) . The data are then partitioned into two “child nodes” based on the value the independent variable ( IV ) that best meets some partitioning criterion . The resulting child nodes each contain a subset of the original data . Each child node may be further partitioned , again based on the value of an IV . This process continues until no further partitions remain or some set of partitioning criteria are no longer met , resulting in terminal nodes . Terminal nodes , by definition , cannot have offspring . C&RT and CIT differ in the partitioning criteria used to select the IVs . Under C&RT the data are partitioned according to the IV that results in the greatest improvement in the distribution homogeneity of the outcome [41] , also referred to as reducing node impurity . Put another way , the data are split according to the IV that best improves predictive accuracy in the child nodes . The predictive accuracy of each potential binary split is considered independently and the split offering the greatest improvement is chosen to partition the data . The initial tree generated by the recursive partitioning process of C&RT tends to be large ( i . e . contain many splits of the data ) and runs the risk of over-fitting the data . This motivates a second stage of tree construction called “pruning” , which can be viewed as analogous to backwards selection in linear regression . Through pruning , partitions of the data that are deemed to be the most superfluous are removed from the bottom-up . Cross-validation is then used to select the optimal sub-tree from the initial tree . In this study , C&RT analysis was performed using the ‘rpart’ package in R , version 2 . 13 . 2 , available at http://cran . r-project . org/web/packages/rpart/index . html . For a more detailed description of this method see Therneau et al [42] . With CIT , the partitioning criterion is based on statistical significance and , unlike C&RT , accounts for conditional relationships between IVs . In the first step of the algorithm , the global null hypothesis of independence between all the IVs and the outcome is tested; if the null cannot be rejected , partitioning stops . If the global null hypothesis is rejected , then the IV that is the most significant in the model , conditional on the other covariates , is selected . When the selected IV is dichotomous , the choice of the best binary split is trivial; for non-dichotomous variables , the algorithm identifies the best binary split from all possible splits . Because CITs are based on statistical inference , pruning is not necessary . In this study , CIT analysis was performed using the ctree function in the ‘party’ package in R , version 2 . 13 . 2 , available at http://cran . r-project . org/web/packages/party/index . html . See Hothorn et al for more information on this method [43] . All of the demographic and WASH indicators listed in Tables 1–3 were included in the analysis as IVs to be selected to partition the trees . The outcome of interest was any STH infection , coded dichotomously , with a “1” indicating the presence of at least one infection by A . lumbricoides , T . trichiura , or hookworm . Both C&RT and CIT trees were grown with the restriction that each node must have a minimum of 20 observations . The C&RT results were validated using 10-fold cross-validation and the optimal tree was selected by pruning to the smallest tree within one standard error of the minimum cross-validated error tree [44] . Two different CITs were generated . In the primary analysis a minimum p-value of 0 . 05 was used for the partitioning criterion; in a secondary analysis , p-values were adjusted for multiple comparisons , using the Bonferroni correction .
Table 1 contains the prevalence of key demographic and WASH variables measured at the pupil level . Over half ( 575; 52% ) of the pupils surveyed were boys with a mean age of 10 . 4 years . Approximately a third ( 383; 35% ) of pupils were observed without shoes at school and 138 ( 13% ) reported some form of soil eating , known as geophagy . Pupils had a mixed impression of their school latrines , with 55% and 64% reporting the latrines to be dirty and have a strong odor , respectively; while 66% reported the latrines to be “comfortable . ” Table 2 contains information regarding the household demographic and WASH characteristics . The prevalence of orphanhood was high in the households surveyed , with 12% of mothers and 33% of fathers deceased . Of the households with living mothers , only 67 ( 6% ) had completed at least secondary education and 423 ( 39% ) had no formal education . Nearly half of the households had an improved water source in the dry season ( 537; 49% ) . The presence of a latrine was high ( 63% ) , though few hand washing stations were observed ( 39% ) . Information regarding school-level WASH characteristics is in Table 3 . Available water for drinking and hand washing was observed at the time of interview in nearly 60% of schools . Only half of the schools had an improved water source in the dry season , while approximately 70% had an improved source in the rainy season . The frequency with which pupils reported constant availability of water for drinking and hand washing at school varied widely between schools . Most students reported that soap was not always available at school . Of the 1095 pupils tested by Kato Katz , 18% tested positive for at least one worm infection , with 81 pupils testing positive for A . lumbricoides , 75 for hookworm and 74 for T . trichiura ( Table 4 ) . Thirty-three children ( 3% ) tested positive for two STH species . No samples were positive for all three worm types . More detailed information on the worm burden and main effects in this data have been previously published in Freeman et al [9] . Initial attempts to generate a classification tree failed to result in anything other than the root node ( the starting dataset with no partitions ) after pruning . This means that after cross-validation the tree generated showed no significant improvement in predictive accuracy over the starting dataset . A second C&RT analysis was performed , this time with sensitivity weighted more heavily than specificity under the assumption that in a situation of STH control , the identification of true STH infections is likely to be prioritized over true no infections . This was achieved by setting a misclassification cost of 2∶1 for STH positive vs . STH negative infections . Typically the C&RT algorithm tries to minimize the proportion of misclassified cases , where misclassification costs are taken to be equal for every case ( e . g . those individuals with both positive and negative stool examinations for STH ) . With the 2∶1 weighting employed , misclassified positive individuals count twice as much as misclassified negative individuals . The 2∶1 misclassification weighting resulted in a pruned classification tree with five partitions and six terminal nodes for predicting the incidence of infection by any of the three helminths . The following independent variables appeared in the final pruned tree: “latrine structure” , “latrine cleanliness” , “% pupils reporting drinking water always available at school” , and “father status” ( Figure 1 ) . The first split of the tree was the PCA factor for school latrine cleanliness , indicating that this variable was best at classifying STH infection status in the data . For both PCA-derived school sanitation variables appearing in the final tree ( latrine cleanliness and latrine structure ) , the C&RT algorithm found an optimal partition of the PCA scores; however , because the numeric values of these scores have no external generalizability we relabeled the dichotomous groups as having “better” or “worse” sanitation conditions . There were six terminal nodes in the final C&RT tree . Terminal node T5 had the greatest proportion of positive cases with 19 of the 28 children positive for one or more species of helminth; this terminal node corresponds to schools with good latrine cleanliness and structure but low reported drinking water availability . The C&RT algorithm labeled terminal nodes T4 and T5 as predictive of a “positive” STH infection , while the remaining four terminal nodes were predictive of no STH infection . Note that terminal node T4 is predictive of “positive” infection despite only 41% of observations being positive because of the 2∶1 weighting favoring sensitivity over specificity . Table 5 shows the distribution of STH infection at each pair of child nodes emanating from the internal nodes in the C&RT analysis . Based on the classification tree , pupils with “worse” latrine cleanliness scores were twice as likely to be infected with STH ( 30% ) , compared to those with “better” latrine cleanliness scores ( 15% ) . Among pupils in schools with “better” latrine cleanliness scores , those with “better” latrine structure had twice the rate of infection ( 24% ) , compared to those with “worse” latrine structural conditions ( 12% ) . For schools with better latrine structure , greater values for “% pupils reporting drinking water always available at school” was predictive of lower levels of infection . Among those schools with “worse” latrine cleanliness and greater drinking water availability at school , the pupil-level variable for father's education status was identified as the best classifier of STH infection . The optimal partition of this pupil-level ordinal variable , which distinguished deceased fathers from living fathers with various levels of education ( see Table 2 ) , occurred between deceased fathers and living fathers regardless of education status . The primary CIT tree , generated without adjustment for multiple comparisons , had 12 terminal nodes ( Figure 2 ) . The IV most significantly associated with STH infection , conditional on all other variables in the model , was “District” . Of the 11 IVs appearing in the tree , 5 were measured at the schoollevel , 5 at the household level and 1 ( “age” ) at the pupil level . Five of the IVs in the final tree were demographic measures while the remaining six were WASH indicators . Terminal node T7 had the greatest proportion of positive cases with 44% ( n = 12 ) of individuals testing STH positive . This node corresponds to pupils from Kisumu East or Rachuonyo Districts with low family SES who attend a school with high enrollment rates and little to no soap reportedly available . Table 6 shows the distribution of STH infection for each pair of child nodes emanating from the internal nodes in the CIT analysis . In Nyando District , greater reported availability of water for hand washing in the schools was associated with a greater incidence of STH infection ( 34% vs . 16% , from terminal nodes T1 and T2 ) . In the Kisumu East and Rachuonyo Districts , among those schools with high enrollment rates ( >80th percentile ) , pupils whose household was in the lowest SES quintile had twice the incidence of STH infection ( 32% vs . 16%; Table 6 ) . The IV “% pupils reporting soap always available at school” appeared twice in the tree in Figure 2 ( internal nodes 8 & 10 ) but with opposite directions of association; at internal node 8 , little to no soap availability was associated with increased STH infection ( 44% vs . 24% ) , whereas at internal node 10 lack of soap was associated with decreased STH infection ( 11% vs . 27% ) . The second conditional inference tree , grown with p-values adjusted for multiple comparisons , is shown in Figure 3 . This tree is a sub-tree of the tree in Figure 2 and represents a more conservative approach . The tree in Figure 3 has four terminal nodes , with the greatest STH incidence seen in the terminal node for pupils in Nyando District attending schools where more than 63% of the students reported hand washing water available ( 34% of pupils in this node were positive for STH ) . The branching of the IVs in both the classification and conditional inference trees can be used to identify potential interactions between WASH indicators that may be important predictors of STH infection . The classification tree illustrates that when latrine cleanliness is better , latrine structure is an important determinant of STH infection . In this instance , worse latrine structure predicts lower STH infection ( Figure 1 ) . When both latrine cleanliness and structure are good , the pupil-reported presence of water at school is important , with poor water availability ( <24% of students reporting constant water availability ) associated with higher STH infection ( T5 , Figure 1 ) and more consistent water availability predicting less STH infection ( T6 , Figure 1 ) . The conditional inference tree suggests that the interaction of living in the Nyando District and the reported availability of hand washing water at school are together associated with STH infection ( Figure 2 ) . Similarly , for those living in the Kisumu East and Rachuonyo Districts the CIT analysis identified an interaction between pupil enrollment , low SES and age ( among other interactions present in the tree ) .
C&RT and CIT are two analytic tools that may be of use to the NTD and WASH communities , depending on the research objective . When prediction of the outcome is the goal , C&RT is likely to be the most favorable tool , whereas CIT is good for identifying the IVs most significantly associated with the outcome . Both methods can be used to identify complex interactions in the data; however , these interactions should be interpreted in the context of the tool ( e . g . prediction vs . association ) . These interactions can then be incorporated into subsequent parametric analyses or used to generate hypotheses for future research . This study supports the WHO's goal for STH elimination by contributing to the research on the impacts of WASH in mitigating STH infection . With the help of this and future research it will hopefully be possible to identify the WASH indicators of greatest relevance for STH control .
|
Soil-transmitted helminths ( STH ) are pervasive enteric parasites that lead to cognitive , nutritional and educational sequelae . Mass drug administration is employed to reduce morbidity , but reinfection occurs rapidly in the absence of changes to other environmental conditions , such as improvements to water , sanitation and hygiene ( WASH ) . Since WASH behaviors and conditions are highly interrelated , typical epidemiological methods are limited . Few rigorous studies have assessed the impact of WASH components as they complement deworming and even fewer have sought to prioritize among the available indicators or identify complex interactions . In this paper we introduce two recursive partitioning approaches: classification and regression trees ( C&RT ) and conditional inference trees ( CIT ) . We demonstrate these two tools using data from a school-based cluster-randomized trial conducted in Kenya . We discuss the advantages and disadvantages of each tool and give examples of how they may be used to improve STH control programs .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"helminth",
"infections",
"environmental",
"health",
"medicine",
"and",
"health",
"sciences",
"environmental",
"epidemiology",
"epidemiology",
"global",
"health",
"neglected",
"tropical",
"diseases",
"hookworm",
"diseases",
"tropical",
"diseases",
"soil-transmitted",
"helminthiases",
"parasitic",
"diseases",
"health",
"care"
] |
2014
|
Exploring the Relationship between Access to Water, Sanitation and Hygiene and Soil-Transmitted Helminth Infection: A Demonstration of Two Recursive Partitioning Tools
|
Group B Streptococcus ( GBS ) is a common agent of bacterial sepsis and meningitis in newborns . The GBS surface capsule contains sialic acids ( Sia ) that engage Sia-binding immunoglobulin-like lectins ( Siglecs ) on leukocytes . Here we use mice lacking Siglec-E , an inhibitory Siglec of myelomonocytic cells , to study the significance of GBS Siglec engagement during in vivo infection . We found GBS bound to Siglec-E in a Sia-specific fashion to blunt NF-κB and MAPK activation . As a consequence , Siglec-E-deficient macrophages had enhanced pro-inflammatory cytokine secretion , phagocytosis and bactericidal activity against the pathogen . Following pulmonary or low-dose intravenous GBS challenge , Siglec-E KO mice produced more pro-inflammatory cytokines and exhibited reduced GBS invasion of the central nervous system . In contrast , upon high dose lethal challenges , cytokine storm in Siglec-E KO mice was associated with accelerated mortality . We conclude that GBS Sia mimicry influences host innate immune and inflammatory responses in vivo through engagement of an inhibitory Siglec , with the ultimate outcome of the host response varying depending upon the site , stage and magnitude of infection .
Group B Streptococcus ( GBS , S . agalactiae ) is a Gram-positive encapsulated bacterium colonizing the vagina of 15–30% of healthy women . GBS is a leading cause of neonatal pneumonia , septicemia , and meningitis [1] , [2] , [3] , and GBS colonization during pregnancy increases the incidence of preterm rupture of membranes and premature delivery [4] , [5] . A decrease in the incidence of early-onset GBS disease in many developed countries has occurred following implementation of universal antenatal culture screening and use of intrapartum antibiotic prophylaxis ( IAP ) [6] . In contrast , the use of IAP has not had a similar impact on the incidence of late-onset disease , which now represents approximately one-third of total cases [6] . Up to 50% of late-onset GBS cases develop meningitis , which carries a very high incidence of neurocognitive sequelae – among survivors 13% had severe , 17% moderate and 18% mild disability at 5 years [7] . A systematic review of incidence data since 2000 estimated an overall incidence of GBS infection in infants less than 3 months of age of 0 . 53 cases per 1 , 000 births in the European region and 0 . 67 cases per 1 , 000 births in the Americas [8] . Less developed countries reporting no IAP use have an overall a 2 . 2-fold higher incidence of early-onset infection compared with those reporting any use . Overall case fatality rates are approximately 7–10% in this recent meta-analysis [8] . Of concern , the emergence of antibiotic-resistant GBS strains has been recently reported , likely a consequence of widespread administration of IAP [1] , [9] , [10] . Moreover , invasive GBS infections occurring in non-pregnant adult populations , especially the elderly and immune-compromised , have been documented with increasing frequency [11] , [12] , [13] . GBS expresses a capsular polysaccharide ( CPS ) that is a major virulence factor contributing to evasion of host immune defense mechanisms [14] . GBS CPS can be divided into ten serotypes ( Ia , Ib , II–IX ) , each with unique structural and antigenic features . Nevertheless , all the GBS CPS repeating units share a critical conserved element: a terminally capped sialic acid ( Sia ) . Sia is a nine-carbon backbone sugar present abundantly in the surface glycocalyx of all mammalian cells , and in this manner GBS decorates its surface with a common host epitope – a form of molecular mimicry . The sialylated CPS is recognized as a critical factor for GBS survival in vivo [15] , interfering with host complement system functions to block C3b deposition and limit C5a production [16] , [17] . An important facet of Sia biology is the function of Sia-binding immunoglobulin-like lectins ( Siglecs ) , receptors that are largely expressed across the major leukocyte lineages to mediate important innate and adaptive immune functions [18] , [19] , [20] , [21] . A major subset is the CD33-related Siglec family ( CD33rSiglecs ) , most of which possess a cytoplasmic domain containing both a membrane-proximal immunoreceptor tyrosine-based inhibitory motif ( ITIM ) and a membrane-distal ITIM-like motif [18] , [22] . Studies have shown that ITIMs can be tyrosine phosphorylated to recruit Src homology 2 domain-containing tyrosine phosphatases ( SHP ) -1 and SHP-2 , and to trigger inhibitory signaling when the corresponding receptors are cross-linked with agonist monoclonal antibodies [23] , [24] , [25] . This suggests that the natural function of the inhibitory CD33rSiglecs is the engagement of ubiquitous endogenous Sia motifs acting as “self-associated molecular patterns” in the host to limit the baseline activation of leukocytes [26] . Through Sia mimicry , GBS has the potential to engage inhibitory CD33rSiglecs and dampen leukocyte innate immune responses , which may promote the survival of the pathogen [27] . In earlier work , we took advantage of Sia-blocking and Sia-nonblocking monoclonal antibodies against human Siglec-9 to provide in vitro evidence that GBS Sia-dependent engagement of Siglec-9 could blunt neutrophil oxidative burst and extracellular trap formation [28] . The significance of this phenomenon in the context of in vivo infection remained unproven , in large part due to rapid evolution of CD33rSiglecs in primates , such that the composition of the CD33rSiglec family varies greatly between primates and rodents . For example , while humans have eleven CD33rSiglecs , mice only have five . Murine Siglec-E has been broadly detected on the surface of innate immune cells [25] , [29] , [30] , [31] and has been proposed to represent the functional paralog of human Siglec-9 . Very recently , Siglec-E knockout ( KO ) mice have been generated and confirmed to have an inhibitory role in leukocyte activation [31] . The availability of these animals thus allows us to test for the first time in vivo the pathogenic consequence of Sia molecular mimicry on bacterial interactions with host inhibitory Siglecs . We report that Siglec-E indeed interacts with GBS in a Sia-dependent manner , triggering SHP-1 recruitment to its intracellular domain and diminishing GBS-triggered myeloid cell innate immune and inflammatory responses . Exaggerated responses to GBS challenge in Siglec-E KO mice represented a “double-edged sword” , limiting GBS dissemination in sublethal systemic challenge , but promoting accelerated septicemia and increased mortality upon high-dose systemic challenge .
Siglec-E has been shown to bind a wide range of sialylated lipid probes that have α2 , 3 , α2 , 6 , or α2 , 8 sialyl linkages in glycan arrays [32] . We first ascertained whether Siglec-E can specifically engage GBS that expresses an α2 , 3-linked sialyllactosamine CPS . Fc chimeras for Siglec-E and Siglec-9 were prebound to protein A-coated plates in a high-avidity format , and FITC-labeled wild-type ( WT ) GBS ( serotype III strain COH-1 ) or its isogenic Sia-deficient mutant ( ΔSia ) were added to the wells . The GBS WT strain but not the GBS ΔSia mutant bound to Siglec-E , the proposed functional paralog of Siglec-9 , indicating that the GBS-Siglec-E interaction is indeed Sia-dependent ( Figure 1A ) . To confirm that Siglec-E interaction with GBS is specific for Sia , we first examined the surface charge of WT GBS and ΔSia mutant . As shown in Figure S1A , the two strains bound positively charged poly-L-lysine equivalently , which indicates that GBS WT and ΔSia mutant have similar overall surface charge . In addition , we found GBS lost its Siglec-E binding ability when the side chain of CPS Sia was truncated by mild periodate treatment ( Figure S1B ) , further supporting that the GBS-Siglec-E interaction is Sia-dependent . WT GBS engaged Siglec-9 to dampen human neutrophil activation [28] . The first question we sought to address is whether Siglec-E KO leukocytes lacking expression of this major ITIM-containing Siglec demonstrate increased bactericidal activity . WT GBS was inoculated into freshly isolated whole blood of WT or Siglec-E KO mice and bacteria colony-forming units ( CFU ) were enumerated at the indicated time points . As shown in Figure 1B , growth of GBS was better controlled in Siglec-E deficient mouse blood than in WT mouse blood . Because Siglec-E has been detected in bone marrow cells , neutrophils , peritoneal macrophages and monocyte/macrophage cell lines [25] , [31] , we prepared bone marrow-derived macrophages ( MBDMs ) to study the impact of Siglec-E on the innate immune response to GBS infection . In accordance with greater restriction of GBS growth in whole blood from Siglec-E KO mice , absence of Siglec-E expression on macrophages enhanced their phagocytic ( Figure 1C and Figure S2 ) and bactericidal ( Figure 1D ) activity . In addition , WT GBS stimulated greater secretion of the pro-inflammatory cytokines TNF-α ( Figure 1E ) and IL-6 ( Figure 1F ) from Siglec-E deficient macrophages both at 6 h and 24 h post infection . This phenomenon was Sia-dependent , as the isogenic GBS ΔSia mutant and LPS triggered similar amounts of pro-inflammatory cytokines from the WT compared to the Siglec-E KO cells ( Figure 1 E and F and Figure S3 ) . The ITIM signaling motif can antagonize kinase-dependent activation cascades by recruiting tyrosine phosphatases SHP-1 and SHP-2 [33] , [34] , [35] . Siglec-E is constitutively associated with a low level of SHP-1 in neutrophils , even in the absence of Siglec-E tyrosine phosphorylation [31] . We asked whether the recruitment of SHP-1 to Siglec-E can be further enhanced upon encountering the Sia-expressing pathogen GBS . WT MBDMs were incubated with or without WT GBS or the ΔSia mutant for 30 min , and Siglec-E was immunoprecipitated and probed for the co-immunoprecipitation of SHP-1 . A baseline level of endogenous SHP-1 association with Siglec-E in the unstimulated MBDMs was observed , consistent with a previous report on mouse bone marrow cells [31] . The SHP-1 recruitment was slightly reduced in the ΔSia mutant-treated cells compared to that of unstimulated cells , which suggests the activation of the ΔSia mutant reduces endogenous SHP-1 association to Siglec-E in the absence of Sia-Siglec-E engagement . On other hand , enhanced SHP-1 recruitment was observed on macrophages exposed to WT GBS where Sia engagement occurs ( Figure 1G ) . Bacterial infections characteristically activate pattern recognition receptors such as Toll-like receptors ( TLRs ) to initiate MAP kinase signaling cascades and NF-κB activation . We hypothesized that activation of MAP kinase and NF-κB in response to GBS may be increased in macrophages lacking Siglec-E-mediated inhibitory signaling . Indeed , WT GBS triggered prolonged ERK phosphorylation and IκB degradation in Siglec-E deficient macrophages than WT macrophages ( Figure 1H ) but no difference was observed in p38 phosphorylation . Prolonged ERK phosphorylation and IκB degradation triggered in response to GBS challenge in the Siglec-E KO macrophages was consistent with the enhanced secretion of inflammatory cytokines ( Figure 1 E and F ) , whereas the GBS ΔSia mutant triggered similar levels of ERK phosphorylation and IκB degradation ( Figure S4 ) . We also examined the expression of suppressor of cytokine signaling ( SOCS ) -1 and SOCS-3 on the WT and Siglec-E KO macrophages after GBS or LPS stimulation to exclude a potential general signaling termination defect in the Siglec-E KO macrophages . Similar SOCS-1 and SOCS-3 expression was observed on WT and Siglec-E KO macrophages after LPS and GBS stimulation , indicating that Siglec-E deficiency does not result in a pervasive defect in signal initiation and termination ( Figure S5 ) . Our data suggest that Sia present in the GBS surface CPS engages Siglec-E and induces SHP-1 recruitment to diminish MAP kinases and NF-κB activation . This pattern of pathogen subversion of host responses by engagement of the inhibitory Siglec was not observed in the Siglec-E KO macrophages nor in WT macrophages stimulated with the GBS ΔSia mutant . In an earlier study , we showed that removal of the tonic inhibitory signals from Siglecs by cleaving their endogenous cis-ligands with pneumococcal sialidase provokes increased leukocyte inflammatory responses [36] . Since the loss of the principal inhibitory Siglec ( Siglec-E ) on murine macrophages increased their bacterial killing , ERK and NF-κB signaling , and inflammatory cytokine release in vitro ( Figure 1 ) , we next examined whether intranasal GBS challenge triggered exaggerated cytokine responses in vivo . At an early time point ( 6 h post-infection ) , no significant differences in bacteria load in lung tissues ( Figure 2A ) nor infiltrated immune cells from the bronchoalveolar lavage ( BAL ) ( Figure 2B ) were observed . However , significantly higher levels of IL-1β ( Figure 2C ) and IL-6 ( Figure 2D ) were found in the BAL fluid from GBS-infected Siglec-E KO mice . Moreover , a significant decrease in secretion of the anti-inflammatory cytokine IL-10 was observed in the lung tissue of Siglec-E KO mice after GBS challenge ( Figure 2E ) , consistent with a previous report that presence of Siglec-9 on cultured THP-1 macrophages resulted in reduced TNF-α mRNA expression accompanied with increased IL-10 mRNA levels [37] . The elevated inflammatory cytokine production and the decreased secretion of IL-10 in the Siglec-E deficient mice in response to acute localized GBS infection suggested that loss of the immunoregulatory ITIM-containing Siglec-E in leukocytes skews the host immune response toward a more inflammation prone set point . Mice lacking Siglec-G were more susceptible to intestinal perforation-induced sepsis due to disruption of a Siglec-G/CD24 interaction , which leads to a failure in repressing host danger signaling-mediated inflammation [38] . Reduced IL-10 secretion together with excessive inflammatory cytokine production in Siglec-E KO mice upon local GBS infection let us speculate that , lacking inhibitory signals from the major Siglec expressed on macrophages , may lead to exaggerated inflammation and septicemia during high-dose systemic GBS infection . Sialylated CPS is recognized as a critical factor for GBS survival in vivo [15] , [39] , and the LD50 values of CPS deficient strains are up to 105-fold greater than the WT strains [15] , [40]; thus we are only able to model sustained infection with WT GBS and not the ΔSia mutant . When WT and Siglec-E KO mice were challenged with 3×108 CFU WT GBS intravenously , 80% of WT mice survived the observation period whereas the majority of mice lacking Siglec-E died within 48 h ( Figure 3A ) . In parallel with reduced survival rates in Siglec-E KO animals , serum levels of IL-6 ( Figure 3B ) and acute phase protein , serum amyloid A ( SAA ) ( Figure 3C ) 18 h after infection were substantially higher in Siglec-E KO animals relative to WT controls . The increased mortality of Siglec-E KO mice was not attributable to differences in bacterial burden in the animals , since similar GBS CFU were recovered from the blood ( Figure 3D ) and brains ( Figure 3E ) of WT and Siglec-E KO mice . These results suggested that the accelerated fatality in Siglec-E KO mice upon high dose GBS challenge involved impaired regulation of inflammatory cytokine production . Loss of Siglec-E expression was harmful for animals upon high-dose lethal GBS challenge due to excessive inflammation . However , as GBS Sia CPS is known to dampen neutrophil bactericidal activity by targeting Siglec-9 [28] , we postulated that under lower-dose challenge conditions Siglec-E KO animals could benefit from loss of negative feedback on innate immunity response pathways . Consequently , WT and Siglec-E KO mice were infected with sublethal dose of GBS intravenously . GBS counts detected in the blood 4 h post-infection in each group were essentially identical , suggesting consistent establishment of systemic infection ( Figure S6A ) . None of the mice died within 48 h under sublethal GBS challenge , and infected mice were euthanized 48 h postinfection to assess bacterial dissemination to the brain and other organs . Whereas no significant overall difference in the blood GBS CFU was observed between WT and Siglec-E deficient mice , a few Siglec-E KO mice showed complete clearance of GBS in the blood ( Figure S6B ) . When compared to the WT mice , GBS-infected Siglec-E KO mice had a modest reduction in bacterial counts in the kidney ( P = 0 . 046 , Figure 4A ) . Bacterial brain GBS CFU counts were also significantly lower in Siglec-E than in WT mice ( P = 0 . 0055 ) ( Figure 4B ) , as was the ratio of CFU recovered from the brain over the bloodstream ( P = 0 . 0178 ) ( Figure 4C ) . We further measured the cytokine production in the mice after nonlethal GBS infection . Elevated serum IL-6 and SAA in the GBS-infected Siglec-E deficient mice was observed 18 h postinfection ( Figure S7A and B ) , although there was no difference in blood and brain bacterial loads in WT and Siglec-E deficient mice at the earlier infection stage ( Figure S7C and D ) . No detectable inflammatory cytokines ( TNF-α , IL-1β and IL-12 ) were present in the serum collected at the time mice were sacrificed ( 48 h ) . The failure of cytokine detection may be due to the quick drop of systemic serum inflammatory cytokines after infection . However , quantative RT-PCR analysis revealed that the GBS-infected Siglec-E deficient mice had increased transcript levels of IL-1β and IL-12 ( Figure 4 D and E ) compared to WT mice , consistent with the findings that loss of Siglec-E expression augmented inflammatory cytokine production in vitro and in vivo . Corroborating what we found in the localized intranasal pneumonia model , reduced IL-10 transcript production in Siglec-E KO mice was observed after low-dose GBS infection ( Figure 2E and 4F ) . Our data show for the first time that an ITIM-containing Siglec can promote expression of the anti-inflammatory IL-10 in response to a sialylated bacterial pathogen . We observed reduced GBS brain CFU counts in Siglec-E KO mice after intravenous infection despite similar blood CFU counts in WT and Siglec-E KO animals . Microglial cells are the major resident macrophage population in the central nervous system parenchyma and are also part of the blood-brain barrier composition , which collectively poise them to function as a first line of defense against invading pathogens [41] , [42] , [43] . Since macrophages lacking Siglec-E expression exhibited greater bactericidal activity and inflammatory cytokine production , important for the control the growth and dissemination of bacteria , we tested whether microglia cells in the Siglec-E KO mice might contribute to reduced spread of GBS into the central nervous systems upon a systemic GBS infection . Microglial cells derived from WT animals showed low level expression of Siglec-E ( data not shown ) , and loss of Siglec-E expression on microglia cells enhanced their GBS bactericidal activity ( Figure 4G ) . In accordance with greater inflammatory cytokine secretion in the Siglec-E deficient macrophages following WT GBS stimulation ( Figure 1 E and F ) , WT GBS also stimulated greater TNF-α secretion the Siglec-E KO microglia cells 24 h post infection . Once again , the GBS ΔSia mutant triggered similar amounts of TNF-α secretion in WT vs . Siglec-E KO cells ( Figure 4H ) .
In the present study , we report that GBS can engage mouse Siglec-E , a functional paralog of human Siglec-9 , and directly address the consequence of host responses in vivo in the context of presence or absence of this major inhibitory Siglec expressed on the innate immune cells during live infection with a sialylated bacterial pathogen . Loss of Siglec-E expression significantly enhanced macrophage inflammatory and bactericidal activity against GBS . The outcome and host survival was associated with the magnitude of infection and infection-induced inflammatory responses . Animals lacking Siglec-E expression appeared to benefit by the elevated inflammatory and bactericidal responses upon a low dose GBS infection . In contrast , the detrimental overwhelming inflammation accelerated mortality in the Siglec-E-deficient animals followed by a lethal dose challenge . Leukocyte activation is controlled by a sophisticated balance between stimulatory and inhibitory signals through activating and inhibitory receptors , respectively . Bacterial infections activate pattern recognition receptors , such as TLRs , which to recognize highly conserved microbial motifs to initiate MAP kinase and NF-κB activation required for cell activation . However , the termination of such activation signals is equally critical to fine-tune the magnitude of inflammation to mitigate host cell damage , such that the overall response is dependent on all downstream signals delivered by the engaged receptors . Many inhibitory receptors contain one or more characteristic sequences ( V/IXYXXL/V ) in the cytoplasmic domain classified as an ITIM . Phosphorylation of the tyrosine residue within the ITIM allows the binding of protein tyrosine phosphatases SHP-1 and/or SHP-2 or the inositol phosphatase SHIP through their SH2 domains [44] , [45] , [46] . When recruited to the complex , these phosphatases act to block signal transduction by dephosphorylating key proteins or lipids of a signaling cascade . Thus , regardless of the structural type of an inhibitory receptor , the inhibitory mechanism is similar . The molecular features and signaling properties of the ITIM-containing CD33rSiglecs point to a potentially important role in the limitation of excess inflammatory responses upon cell activation [21] . Reduced CD33 surface expression on human monocytes by RNA interference silencing or antibody blockade resulted in the increased secretion of IL-1β , IL-8 , and TNF-α [47] . On the other hand , overexpression of WT Siglec-9 , but not Siglec-9 with a mutated tyrosine into phenylalanine on the cytoplasmic ITIM motif , led to reduced TNF-α and IL-6 production upon TLR ligation [37] . In addition , ligation of Siglec-9 on monocyte-derived dendritic cells by tumor-derived Sia-containing mucins or anti-Siglec-9 antibodies suppressed IL-12 production upon LPS stimulation [48] . Similar findings can be extended to the murine Siglecs in that crosslinking of Siglec-E on LPS-treated macrophages also impaired their inflammatory cytokine production [30] . In accordance with these in vitro cell experiments , we found that GBS engages Siglec-E in a Sia-dependent manner to recruit SHP-1 and to influence ERK and NF-κB signaling , thereby reducing the associated inflammatory cytokine secretion and bactericidal activity ( Figure 1 ) . We further demonstrated that the singular loss of the major inhibitory Siglec , Siglec-E , on innate immune cells was sufficient to augment the inflammatory cytokine secretion accompanied with reduced anti-inflammatory IL-10 production during a local GBS intranasal challenge ( Figure 2 ) . This provides the first evidence to demonstrate that a sialylated pathogen can exploit inhibitory CD33rSiglec to negatively modulate the host inflammatory status during an in vivo infection . A recent report demonstrated accelerated neutrophil and macrophage recruitment in the Siglec-E deficient lungs upon intranasal LPS administration [31] . The authors found this phenomenon was mediated through Siglec-E by suppression of CD11b outside-in signaling when Siglec-E was engaged by the sialylated CD11b ligand , fibrinogen . This Siglec-E-mediated suppression was Sia-dependent , since asiolo-fibrinogen , LPS and immuno-complex exposure did not alter the same downstream signaling observed with sialylated fibrinogen . These data examining endogenous Sias on the host fibrinogen , together with our observations regarding exogenous Sia on pathogens , emphasize the broad significance of Sia-dependent immune cell regulation through Siglec engagement . Deficiency in inhibitory pathways results in profound immune defects characterized both by decreased activation thresholds and hyperresponsiveness phenotypes , which often lead to autoimmunity and chronic inflammation [49] , [50] , [51] , [52] . Mice deficient in Siglec-F displayed enhanced eosinophilic inflammation [53] , while animals lacking Siglec-G expression were more susceptible to intestinal perforation-induced sepsis due to failure in repressing host danger signaling-mediated inflammation [38] . In general , a fine balance must be achieved when encountering pathogens: sufficient innate immune responses must be generated in order to eliminate pathogens , but the inflammatory activation must not be too large to cause widespread host tissue damage . Upregulation of CD33rSiglecs and/or their cognate Sia ligands may represent a means for the host to attenuate and control inflammatory exacerbation by enhancing inhibitory signaling after infection . Siglec-E expression can be induced by pathogen associated molecular patterns such as ligands for TLR2 , 4 , 7 , and 9 in a MyD88-dependent manner [30] . Reduced IL-10 secretion together with excessive inflammatory cytokine production in Siglec-E KO animals upon local GBS infection ( Figure 2 ) let us evaluate whether absence of Siglec-E expression may lead to uncontrolled inflammation and tissue damages during a systemic lethal challenge . Indeed , exaggerated mortality was observed on the Siglec-E KO with higher serum IL-6 and SAA in the setting of similar bacterial burdens ( Figure 3 ) . Given the role of inhibitory CD33rSiglec in controlling exaggerated inflammation , sialylated pathogens may take advantage by molecular mimicry to blunt the bactericidal functions of the immune cells in which inhibitory CD33rSiglecs are expressed through their higher-density trans Sia ligands . We previously reported that GBS CPS Sia mimicy allows engagement of Siglec-9 to dampen neutrophil bactericidal functions [27] , [28] . Here we further demonstrate that in sublethal GBS infection , Siglec-E deficient animals exhibited reduced bacteria dissemination and may have benefited from the enhanced inflammatory responses and reduced anti-inflammatory IL-10 production . Thus GBS can trigger inhibitory signals through engaging Siglec-E to reduce overall innate immune responses in the WT animals ( Figure 4 ) . The established virulence function of the GBS Sia CPS is twofold , involving interference with the complement pathway [16] , [17] but also downregulation of myeloid cell innate immune activities via engagement of inhibitory CD33rSiglecs . CD33rSiglecs are expressed primarily on leukocyte subsets , and whereas some are highly restricted to a certain cell type , other CD33rSiglecs have relatively broad expression patterns . In addition , several CD33rSiglecs can be present on the same cell type in which they may perform a specific function or have functional redundancy . Since the potential contribution of each CD33rSiglec is determined primarily by its cellular expression profile , detailed comparative analyses of the specific cell types that express CD33rSiglecs are essential to clarify their unique roles [18] , [54] . Here , we surprisingly found that in addition to the previously known cells that express Siglec-E , such as neutrophils and macrophages , its expression was also present on the brain microglia cells . Microglia cells are the major resident macrophage population in the CNS parenchyma and part of the blood-brain barrier ( BBB ) composition . The parenchymal CNS is an anti-inflammatory environment with high local concentrations of inflammation-suppressive cytokines such as IL-10 and TGF-β [55] , [56] . Furthermore , the BBB is able to restrain CNS inflammation by excluding plasma proteins as well as peripheral immune cells and their associated inflammatory molecules [57] . Expression of inhibitory Siglec-11 on human microglia cells has been shown to alleviate neurocytotoxicity [58] . Therefore , Siglec-E on microglia may play a physiological role when encountering its endogenous cis ligands . Although microglia are believed to function as first line of defense against invading pathogens [41] , [42] , [43] , the sialylated GBS may suppress microglial responses by engaging Siglec-E on microglia to facilitate the establishment of meningitis upon systemic GBS infection . To support this hypothesis , we found that Siglec-E KO microglia cells had greater microbicidal and cytokine responses against GBS ( Fig . 4 G–H ) compared to WT microglia cells . In conclusion , our data show that GBS Sia mimicry can attenuate host innate immune responses through engagement of an inhibitory Siglec on leukocytes , with the potential outcome on the host response likely to vary dependent upon on the site , stage and magnitude of infection . In localized and early-stage infection , we propose this process can serve to diminish host macrophage innate immune functions , promoting GBS survival and virulence . In overwhelming infection , which could include fulminant neonatal sepsis after ascending infection of the placental membranes , engagement of GBS Sia by inhibitory CD33rSiglecs such as Siglec-9 could ultimately act to mitigate the dysregulated inflammatory response and cytokine storm . Similarly , microglial Siglec-E expression could reduce inflammatory responses in the privileged CNS compartment . A number of other invasive human bacterial pathogens associated with meningitis , including Haemophilus influenzae , Neisseria meningitidis and Escherichiae coli serotype K1 , display Sia in their CPS or cell wall lipooligosaccharides with the potential to engage members of the CD33rSiglec family . Further analysis of Sia-Siglec interactions during host-pathogen encounters could provide novel targets for therapeutic intervention to modify infectious disease outcome .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health under a protocol approved by the Institutional Animal Care and Use Committee at the University of California , San Diego ( Animal Welfare Assurance Number: A3033-01 ) . All efforts were made to minimize suffering of animals employed in this study . A well-characterized human serotype III GBS isolate ( COH-1 ) from a case of human neonatal sepsis was used in this study . Generation of the isogenic ΔSia mutant , containing an in-frame allelic replacement of the neuA gene encoding CMP-N-acetylneuraminic acid synthetase with a chloramphenicol acetyltransferase cassette , was previously described [59] . GBS were propagated in Todd-Hewitt broth , THB , ( Difco , BD Diagnostics ) at 37°C without shaking . For infection studies , bacteria were cultivated at 37°C to mid-exponential phase and resuspended to an OD600 of 0 . 4 , followed by serial dilution and enumeration of CFU ( colony-forming units ) to verify each experimental inoculum . Siglec-Fc and FITC-labeled GBS interactions were performed with slight modifications from our previously described method [27] . To prepare FITC-labeled GBS , early exponential phase grown GBS was pelleted , washed and then labeled with 0 . 1% fluorescein isothiocyanate ( FITC ) /PBS for 1 h at 37°C with rotation . Bacteria were extensively washed with PBS to remove trace amounts of free FITC then resuspended in PBS . 96-well plates were coated with 1 µg/well protein A in 50 mM carbonate buffer ( pH 9 . 5 ) overnight at 4°C . Wells were washed and blocked with 1%BSA/PBS for 1 h at room temperature . Human Siglec 9-Fc , produced as previously described [60] , or murine Siglec E-Fc [29] was immobilized to individual wells for at least 3 h at room temperature . Wells were washed three times , and 1×107 FITC labeled-GBS added to each well . Plates were centrifuged at 2000 rpm for 10 min and incubated for 30 min at 37°C . The initial fluorescence intensity was verified , wells washed to remove unbound bacteria , and the residual fluorescent intensity ( excitation , 488 nm; emission , 530 nm ) measured using a CytoFluorII fluorescent plate reader . Murine bone marrow-derived macrophages were derived from bone marrow cells collected from femur and tibia and cultured with conditional media containing macrophage colony-stimulating factors ( M-CSF ) for 7 d . For phagocytosis assays , GBS ( 2 . 5×106 CFU ) were added to 5×105 macrophages ( multiplicity of infection or MOI = 5 ) , followed by incubation for 30 min or 1 h . Cells were washed with PBS three times followed by addition of medium containing penicillin ( 5 µg/ml ) and gentamicin ( 100 µg/ml ) for 1 h to kill extracellular bacteria . Cells were then washed , detached , and lysed with 0 . 025% Triton X-100 to release intracellular bacteria . Bacterial CFU were enumerated by serial plating on the THA plates . For bacterial killing assays , 1×105 GBS was added to macrophages ( 5×105 ) at MOI = 0 . 2 for 1 or 2 h , followed by addition of 50 µl of 0 . 6% Triton X-100 to lyse cells . Recovered GBS were plated on THA plates for CFU enumeration . Assays were performed in triplicate and were repeated three times . To detect cytokine secretion after GBS infection , 1×105 macrophages were stimulated with 106 GBS for 30 min , and then penicillin and gentamicin added as above to prevent the bacterium overgrowth . The culture supernatant was collected 6 h or 24 h after GBS infection for cytokine ELISA analysis . Isolation of primary microglia cells from adult animals was performed as previously described [61] . In brief , PBS-perfused brains were collected and digested in an enzymatic solution contained 116 mM NaCl , 5 . 4 mM KCl , 26 mM NaHCO3 , 1 mM NaH2PO4 , 1 . 5 mM CaCl2 , 1 mM MgSO4 , 0 . 5 mM EDTA , 25 mM glucose , 1 mM cysteine and 20 units/ml papain ( all from Sigma ) for 90 min at 37°C , followed by 0 . 5 mg/ml of DNase I ( Roche ) treatment for 5 min at room temperature . The resulted brain mixture was resuspended in 20 ml of 20% isotonic Percoll/HBSS , carefully overlaid with 20 ml HBSS and centrifuged at 200 g for 20 min with slow acceleration and no brake . The pellet containing mixed glial cells was collected , washed and cultured in the Dulbecco's modified Eagle's/F12 medium with GlutaMAX supplemented with 10% FBS , 100 units/ml penicillin , 100 µg/ml streptomycin ( all from Invitrogen ) and 5 ng/ml of carrier-free murine recombinant granulocyte and macrophage colony-stimulating factor ( GM-CSF ) ( R&D Systems ) . The medium was changed twice a week until the cells became confluent . A nearly pure microglial cell population was collected from the mixed glial culture supernatant without any prior shaking and plated in the 48-well plates ( 2×105 cells/well ) or 96-well plates ( 1×105 cells/well ) for 3 days in medium without GM-CSF . Bacteria killing assays and cytokine secretion after GBS infection were performed as described above , using MOI of 0 . 2 for 2 h and MOI of 10 for 24 h , respectively . Concentrations of cytokines in macrophage supernatants collected over time post-infection , or cytokines in mouse BAL fluids , lung homogenates and serums from infected animals were detected by corresponding commercial ELISA kits ( TNF-α and IL-6 from R'D; IL-1β , IL-10 and IL-12 from BD Biosciences; SAA ( serum amyloid A ) from Life Diagnostics ) Macrophages were lysed in lysis buffer ( 50 mM Tris , pH 8 , 150 mM NaCl , 1% NP40 ) containing protease inhibitor cocktail ( Roche ) and phosphatase inhibitor cocktail ( Santa Cruz Biotechnology ) . Cell lysates were then separated on a 10% SDS-PAGE and transferred to a PVDF membrane . The membrane was probed with the anti-phospho p44/42 MAPK ( T202/Y204 ) , anti-phospho p38 , anti-p44/42 , or anti-IκB antibodies ( all from Cell Signaling Technology ) , and then followed by appropriate HRP-conjugated secondary Abs ( Bio-Rad ) and ECL reagent ( Thermo Scientific ) . Macrophages were stimulated with COH1 WT or ΔSia mutant for 30 min at MOI = 10 . Cells were then lysed in RIPA buffer with protease and phosphatase inhibitor cocktail . Siglec-E was immunoprecipitated by mouse anti-Siglec-E mAb ( generated in the Crocker lab ) . Immunoblots were probed with Ab to SHP-1 ( Santa Cruz Biotechnology ) and Siglec-E ( R&D systems ) , respectively , and then followed by appropriate HRP-conjugated secondary Abs and ECL reagent . All animal experiments were approved by the Committee on the Use and Care of Animals , UCSD and performed using accepted veterinary standards . For the murine intranasal challenge model , mice ( Siglec-E knockout mice or C57B/6 controls , 10–12 weeks ) were lightly anesthetized by intraperitoneal injection of ketamine and xylazine , and 50 µl of PBS containing 108 GBS was then administered into the nostrils of the mice . The inoculum dose was confirmed by viable count after plating on THA plates . Infected animals were sacrificed 6 h post infection . Blood was collected via terminal cardiac puncture . For bronchoalveolar lavage ( BAL ) fluid collection , the trachea was exposed and 0 . 8 ml PBS ( without calcium and anticoagulant ) was injected twice by using an 18 guage needle connected to 1 ml syringe . Cells from BAL were counted and stained with APC-conjugated rat anti-mouse F4/80 mAb ( AbD Serotec ) and FITC-conjugated rat anti-mouse Gr-1 mAb ( Caltag Medsystems ) to analyze the infiltrated cell population after GBS infection . The left lung lobe was collected was collected for bacterium CFU enumeration and cytokine detection . For monitoring mouse survival after systemic intravenous challenge , mice ( 10–12 weeks ) were intravenously infected with 3×108 GBS and observed for 10 d . To measure serum cytokine secretion after GBS infection , blood was collected 18 h post infection for cytokine ELISA analysis , along with measurement of bacterial load in the blood and brain . For the GBS meningitis model , mice ( 10–12 weeks ) were intravenously infected with 108 GBS and , bacteria CFU in the blood was examined 4 h later to ensure mice received similar challenge doses . Then 48 h after injection , samples of blood , brain/meninges , and kidney were collected aseptically from mice after euthanasia . Bacterial counts in blood and tissue homogenates were determined by plating serial dilutions . Bacterial counts in brain and kidney were corrected for differences in organ weight . Brain bacterial counts were also corrected for blood contamination using the blood concentration and a conservative estimate of the mouse cerebral blood volume [62] . RNAs were isolated from various tissues using Trizol ( Life Technologies ) and used as a template for cDNA preparation by iScript kit ( Bio-Rad ) . Quantitative PCR was performed using iQ SYBR Green Supermix ( Bio-Rad ) according to standard protocols . Primers used were for IL-1β 5′-ACTACAGGCTCCGAGATGAAC-3′ and 5′-CCCAAGGCCACAGGTATTTT-3′; for IL-12 , 5′-CGTGCTCATGGCTGGTGCAAAG-3′ and 5′-CTTCATCTGCAAGTTCTTGGGC-3′; and for GAPDH , 5′-AACTTTGGCATTGTGGAAGGGC-3′ and 5′-GGTAGGAACACGGAAGGCCATG-3′ . Primers for IL-10 were obtained from QuantiTect Primer Assay ( Qiagen ) Graphpad Prism version 5 was used for statistical analysis . Statistical significance was accepted at P<0 . 05 . Data shown were either pooled from two independent experiments or representative data from two independent experiments conducted with biological replicates . The significance of differences between different groups for the animal experiments was determined using the Mann-Whitney test . Unpaired t test , one-way ANOVA and two-way ANOVA were used for two groups , multiple groups or two parameters comparisons , respectively .
|
The bacterium Group B Streptococcus ( GBS ) causes serious infections such as meningitis in human newborn babies . The surface of GBS is coated with a capsule made of sugar molecules . Prominent among these is sialic acid ( Sia ) , a human-like sugar that interacts with protein receptors called Siglecs on the surface of our white blood cells . In a test tube , GBS Sia binding to human Siglecs can suppress white blood cell activation , reducing their bacterial killing abilities; however , the significance of this during actual infection was unknown . To answer this question , we studied mice for which a key white blood cell Siglec has been genetically deleted . When infected with GBS , white blood cells from the mutant mice are not shut off by the pathogen's Sia-containing sugar capsule . The white blood cells from the Siglec-deficient mice are better at killing GBS and are able to clear infection more quickly than a normal mouse . However , if the mice are given an overwhelming dose of GBS bacteria , exaggerated white blood activation can trigger shock and more rapid death . These studies show how “molecular mimicry” of sugar molecules in the host can influence a bacterial pathogen's interaction with the immune system and the outcome of infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2014
|
Group B Streptococcus Engages an Inhibitory Siglec through Sialic Acid Mimicry to Blunt Innate Immune and Inflammatory Responses In Vivo
|
Regeneration in adult chordates is confined to a few model cases and terminates in restoration of restricted tissues and organs . Here , we study the unique phenomenon of whole body regeneration ( WBR ) in the colonial urochordate Botrylloides leachi in which an entire adult zooid is restored from a miniscule blood vessel fragment . In contrast to all other documented cases , regeneration is induced systemically in blood vessels . Multiple buds appear simultaneously in newly established regeneration niches within vasculature fragments , stemming from composites of pluripotent blood cells and terminating in one functional zooid . We found that retinoic acid ( RA ) regulates diverse developmental aspects in WBR . The homologue of the RA receptor and a retinaldehyde dehydrogenase-related gene were expressed specifically in blood cells within regeneration niches and throughout bud development . The addition of RA inhibitors as well as RNA interference knockdown experiments resulted in WBR arrest and bud malformations . The administration of all-trans RA to blood vessel fragments resulted in doubly accelerated regeneration and multibud formation , leading to restored colonies with multiple zooids . The Botrylloides system differs from known regeneration model systems by several fundamental criteria , including epimorphosis without the formation of blastema and the induction of a “multifocal regeneration niche” system . This is also to our knowledge the first documented case of WBR from circulating blood cells that restores not only the soma , but also the germ line . This unique Botrylloides WBR process could serve as a new in vivo model system for regeneration , suggesting that RA signaling may have had ancestral roles in body restoration events .
Some of the most fundamental issues in developmental biology concern the ability of metazoans to regenerate . In most multicellular organisms , adult stem cells maintain organs' homeostasis throughout life and facilitate tissue repair after injury or disease [1] . Several organisms are capable of regrowing amputated organs and body parts , for example , amphibian limbs , lens , and retina [2–4] . Phylogenetic perspectives reveal a decrease in regenerative abilities concomitant with an increase in animal body complexity [5] . Thus , only a few adult animals manifest massive regeneration events , and these events occur mostly in less complex multicellular organisms such as sponges [6] , cnidarians [7] , and flat worms [8] . An exception is the phenomenon of whole body regeneration ( WBR ) in the highly evolved urochordates subfamily , Botryllinae . In this group of sedentary colonial organisms , which are the closest living relatives of vertebrates [9] , a fully functional adult can regenerate within 10–14 d from any isolated minute fragment ( 0 . 1 mm ) of a blood vessel containing a limited number ( n = 100–300 ) of blood cells [10 , 11] . The filter-feeding Atlantic urochordate B . leachi ( Figure 1A ) is a common encrusting colonial sea squirt , probably a Mediterranean species that has spread ubiquitously [12] . Animals are found in very shallow waters , under stones , on algae , pilings , floats , and other substrata . Each colony is composed of several to thousands of genetically identical modules ( zooids ) ( Figure 1A , arrowheads ) , each 2–3 mm long , embedded within a gelatinous matrix called the tunic . Zooids are arranged in systems of two parallel elongated and often serpentine rows . A network of blood vessels connects all zooids within a colony , from which pear-shaped vascular termini ( ampullae ) extend toward the colony margins ( Figure 1A , arrows ) . Once the planktonic larva metamorphoses into the founder individual ( called the oozooid ) , a colony of zooids develops by a weekly budding process . This process is comprised of four successive phases ( A–D ) [13] , during which new zooids bud from the thoracic body wall of the oozooid and the subsequent zooids , a phenomenon called palleal budding [14] or blastogenesis [15–17] . Each blastogenic cycle ends in a massive apoptotic event ( Phase IV—the takeover phase ) in which parental zooids are resorbed concurrently with the maturation of buds into adult , functional zooids . A unique WBR event , in which buds develop at the bases of vascular ampullae , was observed in congener botryllid species under adverse environmental conditions [18 , 19] , a regeneration phenomenon that had also been induced experimentally [10 , 11] , providing a unique model system for WBR . Retinoic acid ( RA ) is a low molecular weight biologically active metabolite of Vitamin A , which plays diverse functional roles in a wide variety of tissues and organs [20–22] . It plays a conserved role in controlling positional information and cellular differentiation during embryonic development . RA action is mediated at the metabolism and signaling levels . RA-synthesizing enzymes , such as retinaldehyde dehydrogenase ( Raldh ) , together with degrading enzymes ( Cyp26 ) create the spatiotemporal distribution of RA during development and refine it [23] . Signaling is mediated through RA binding to two nuclear receptors , retinoic acid receptor ( RAR ) and retinoic X receptor , through which it acts as a ligand-dependent transcription factor [24 , 25] , subsequently inducing gene transcription . RA has also been recorded as being involved in restricted regeneration of tissues and organs in adult vertebrates [26] . In urodele amphibians , administering RA into amputated limbs resulted in regeneration of extra limb tissue [27–29] . In adult rat lung , RA induces alveolar regeneration , and systemic RA treatment can reverse pathological features of experimental emphysema [30] . In adult rat spinal cord , RAR isoform RARβ2 is able to promote functional regeneration of sensory axons [31] . Despite the identification of RA receptors exclusively in chordates , and the consideration of RA signaling as a chordate innovation [32] , remarkably little evidence exists connecting the activation of RA signaling to conserved chordate biological traits of regeneration . A single RA receptor [33] in lower chordates ( such as urochordates and cephalochordates ) offers an advantage when searching for direct RA-signaling functions . This is not the case in vertebrates where , due to redundancy in RAR genes , RAR localization fails to provide sufficient information on RA-signaling functions . Here , we endeavor to elucidate cellular processes and RA signaling during WBR in the urochordate B . leachi in order to reveal traits and evolutionary routes of regeneration similar to those recorded in its vertebrate counterparts [33] .
Experiments ( n = 95 ) were conducted on small B . leachi fragments ( each containing 1–92 isolated ampullae , cut from different colonies ) that were well attached to the substrate . Whole body restoration was observed in 80 fragments , irrespective of the number of blood vessels , the blastogenic stage ( A–D ) of the colony on the day of operation , the original colony size ( 1–6 systems , 3–41 zooids ) , or the amputated vasculature location ( peripheral , marginal , or from vessels located centrally ) ( Figure 1 ) . Following surgery and isolation of vessel fragments ( Figure 1B ) , we observed an immediate vasculature contraction and restriction of blood flow . The next day , vascular connections grew between ampullae within each fragment , creating a new circulatory system ( Figure 1C and 1D , arrows ) . Then , blood flow was restored and blood cells were observed traveling ( as in untreated colonies ) in opposite directions , briefly in one direction and then back . At the same time , the lumen of some blood vessels became opaque and shrank progressively . During the first seven days , isolated ampullae were highly dynamic , changing their orientations and shapes within the tunic matrix , coalescing into each other , and moving within the tunic ( Figure 1E and 1F ) . These directional movements and new vascular connections created a dense network of anastomosed vessels at selected locations within the tunic embedment . Vasculature movements were more conspicuous when blood vessels were spaced out , creating a dense mass of vessels on one side of the fragment and a semitransparent gelatinous tunic matrix deprived of blood vessels on the other side ( Figure 1F ) . Blood propulsion in the newly formed vascular network slowed down progressively . Concomitantly , an opaque mass of blood vessels was formed , followed by formation of an internal small transparent vesicle ( Figure 1G , arrow ) that grew and developed two openings in opposite corners ( Figure 1H ) . By days 10–14 , a fully operating filter-feeding zooid , equipped with functional atrial and peribranchial siphons ( both facing upwards , as in normal zooids ) ( Figure 1H , arrows ) , developed from the opaque mass of blood vessels . The first generation of zooids was larger ( 1 . 18 mm2 ) than regular zooids ( 0 . 78 mm2 ) and exhibited a spherical shape , unlike the elongated pear-like shape of regular colony zooids . Irrespective of the number of blood vessels ( 1–92 ) presented in each fragment and of the number of buds ( 1–24 ) developing throughout the amputated fragment , only one zooid per fragment ( n = 80 ) was fully formed . Yet , successful regeneration , indicated by the opening of siphons , varied between fragments , being faster ( within 10 d ) in fragments characterized by spacious blood circulation or fragments exhibiting early onset of a new vascular system . When a new vascular system failed to form properly between blood vessels , buds developed in between ampullae at vessel's joints . WBR was not documented in cases ( n = 15 ) where fragments failed to separate completely from colonies , leaving a single zooid fragment or an intact blastogenic bud . Regenerating vasculature fragments ( n = 116 ) were sacrificed at sequential daily intervals and were processed for histological observations . Three distinctive morphological phases were identified at the cellular level . Regular cellular vasculature morphology was revealed by using control intact ampullae ( n = 10 ) . A homologue of RAR ( Bl-RAR ) was cloned from the cDNA of regenerating blood vessels . An 808-bp fragment was sequenced , revealing 87% identity to RAR of the budding tunicate Polyandrocarpa misakiensis . A domain search revealed a conserved 157-amino acid fragment , corresponding to the ligand-binding domain of hormone receptors present in all RARs [34] , ( Figure 3A , red line ) . Furthermore , Bl-RAR is phylogenetically clustered with other urochordate RAR family members in a general glade of other vertebrate RARs and is significantly distinguished from other RA receptor families ( Figure 3B ) . To gain insight into the temporal expression pattern of RAR during regeneration , total RNA was extracted from fragments of regenerating blood vessels at different developmental stages , and RT-PCR was performed using specific primers . Intact blood vessels ( controls ) did not express the RAR transcript ( Figure 3C , lane 1 ) . Similar outcomes were obtained when PCR product from control ampullae was amplified 40 cycles , revealing no traces of the transcript . In contrast , RAR transcript was detected in cDNA of regenerating ampullae as early as 19 h after isolation from the colony ( Figure 3C , lane 2 ) . A PCR transcript was expressed continuously in PCR products throughout subsequent regeneration stages for up to 11 d ( Figure 3C , lanes 3–12 ) . The spatial expression of RAR was assessed during WBR . Blood vessels were separated from B . leachi colonies and left to regenerate in 1-l plastic tanks containing seawater . Intact and regenerating vasculature fragments were fixated at temporal intervals , and in situ hybridization was employed on 5-μm thick paraffin sections . Intact blood vessels showed no detectable hybridization with the probe ( unpublished data ) , while regenerating blood vessels showed a specific staining pattern that was visible from day 2 ( Figure 3D , arrow ) . The early expression pattern of the RAR probe specifically demarcated regeneration sites in aggregates of haemocytes at various niches within the blood vessels . In cases where several regeneration niches were established and buds developed simultaneously within the ampullae , RAR expression pattern was localized specifically in all buds ( Figure 3E , arrows ) . Later on , RAR continued to stain , specifically , regenerating buds through the subsequent developmental stages of spheres , invaginations , and organogenesis ( Figure 3F ) . No detectable staining had ever been observed in blood vessel epithelium or circulating blood cells that did not participate actively in regeneration . Specific sense probes were used as controls revealing no detectable staining pattern ( unpublished data ) . In order to identify cellular sources of RA synthesis and metabolism , a 337-bp fragment was amplified and subsequently cloned from cDNA of regenerating blood vessels . The deduced amino acid was most similar to Aldehyde dehydrogenase ( Aldedh ) from the budding ascidian P . misakiensis ( E-value = 4e−55 ) and revealed high sequence similarity to Aldedh2 family members from chick and mouse ( 66% identity , E-value = 2e−42 ) ( Figure 4 ) . A domain search revealed an Aldedh domain ( Figure 4A , red line ) conserved in all Aldedh family members [35] . B . leachi Raldh ( Bl-Raldh ) expression was analyzed by in situ hybridization on sections of naïve and regenerating blood vessels and revealed a distinct expression pattern during WBR . In naïve ampullae , Bl-Raldh is expressed exclusively in a population of circulating macrophage cells scattered throughout the vasculature ( Figure 4B , arrows ) . Following amputation , RA synthesis was documented in macrophage cells ( Figure 4C , arrows ) in regeneration areas adjacent to nonstained , small cell morula-like aggregates ( Figure 4C , arrowheads , rectangle and Figure 4E enlargement ) . During the regeneration process , Bl-Raldh was not expressed in lymphocyte-like cells , morula-like cells and structures , other common blood cell types , or in the surrounding vessel epithelium . Concomitantly to its strong expression levels in macrophage cells , starting from the blastula-like stage , Bl-Raldh displayed expression patterns overlapping RAR expression throughout bud development ( Figure 4D , arrow ) . A similar correlate expression pattern of RAR and Raldh was observed throughout colony astogeny ( unpublished data ) . Specific sense probes were used as controls revealing no detectable staining pattern ( unpublished data ) . Therefore , the sequence similarity between Bl-Raldh and other vertebrate Raldh molecules combined with the specific parallel expression patterns of Bl-Raldh and Bl-RAR during WBR and colony astogeny classifies this gene product as a Raldh family member . Following the documented complexity of the Aldedh gene family in urochordates [36] , we cannot exclude the presence and/or interplay of additional Raldh genes during WBR . To assess the functional roles of RA synthesis during WBR , peripheral ampullae were dissected from colonies and left to regenerate in 1-l plastic tanks containing the RA synthesis inhibitor 4-Diethylaminobenzaldehyde ( DEAB ) . For control purposes , peripheral ampullae were separated from B . leachi colonies and left to regenerate in 1-l tanks containing DMSO alone . We conducted two sets of regeneration groups employing different DEAB concentrations ( 10 μM and 100 μM ) ( Table 2 ) . Both experimental groups produced similar morphological outcomes . Characteristics of early-stage regeneration were similar between experimental cases ( n = 22 ) and controls ( n = 12 ) and also showed similarity to the previously characterized cases ( n = 95 ) . DEAB-treated regenerating ampullae shrank . While blood vessels in these fragments started directional movements inside the tunic matrix ( Table 2 ) , complete ampullar aggregations in the form of opaque masses of ampullae were never documented ( n = 22 ) . Instead , vessel coalescence had gradually slowed down and stopped completely after day 2 , leaving blood vessels in an intermediate state ( Figure 5A; compare to Figure 1F ) . In control fragments , as in the former studied cases , zooids appeared 10–14 days postsurgery . In contrast , no single zooid had ever developed in any of the DEAB-treated fragments ( Table 2 ) , even after a follow-up observational protocol of 30 d ( n = 11 ) in which all DEAB-treated blood vessel fragments eventually degenerated . Similar results were obtained with the RA synthesis inhibitor Geranial , 3 , 7-Dimethyl-2 , 6-Octadienal ( Citral ) . Two experimental groups were conducted using different concentrations of Citral , showing a dose-dependent phenotype , ranging from complete attenuation of vessel movements ( at 60 μM ) to complete aggregations ( at 20 μM ) ( Figure 5B and 5C , respectively; Table 2 ) . Similar to DEAB treatments , no regeneration was documented even after a follow-up observational protocol of 30 d ( n = 23 ) ( Table 2 ) . We sacrificed 10-μM and 100-μM DEAB-treated and control fragments for histological observations at temporal regeneration intervals . Controls showed the regular regeneration morphology as specified earlier . In DEAB-treated fragments , internal lumen was compartmentalized and taken over by partially formed regeneration niches . Yet , at both DEAB concentrations , there was an unusual increase in cell density inside the vessels' lumen , with more connections between cells ( unpublished data ) . At 100-μM DEAB concentration ( n = 9 ) , buds did not develop and multinucleated giant cells appeared , scattered throughout the vessel lumen ( Figure 5D , arrows ) . At 10-μM DEAB concentration ( n = 13 ) , morphologically abnormal buds formed that failed to develop organ structures and retained a simple epithelial morphology ( Figure 5E ) . In severe cases , masses of undifferentiated aggregated cells occupied the interior of the vessel lumen ( Figure 5F ) . All malformed buds subsequently degenerated . PCNA immunostaining revealed distinctive proliferation at aggregated masses , while no staining was detected in blood vessels ( Figure 5G ) . We disrupted RA receptor function by way of RNA interference ( RNAi ) as well as the RAR pan-antagonist BMS-493 separately . Ampullae , surgically separated from B . leachi colonies , were soaked for 6 d in seawater containing control small interfering RNAs ( siRNAs ) and siRNAs generated against Bl-RAR , as previously shown [37] . Outcomes were morphologically monitored daily followed by histological analyses . Knockdown of Bl-RAR expression was verified in intact colonies after 6 d of incubation by comparing the abundance of RAR to actin transcripts ( Figure 6A ) . As clearly seen in siRNAs-treated colonies , RAR expression was nearly eliminated ( Figure 6A , lane 2 ) while no significant change was observed in the RNAi control experiment ( Figure 6A , lane 1 ) compared to naïve colonies ( Figure 6A , lane 3 ) . Differences between control RNAi and RNAi experiment fragments were visible as of day 2 , when blood vessels moved actively in control fragments , compared to arrest movement in siRNAs-treated fragments ( unpublished data ) . After 6 d , control experiments using control siRNAs completed vasculature movements , forming opaque vessel masses as previously specified ( Figure 6B ) . In contrast , no single opaque vessel mass appeared in RNAi affected experiments ( n = 14 ) . Instead , blood vessels remained apart , situated haphazardly within the tunic matrix as in the premature stages ( Figure 6C ) . In several experiments , vessel movement halted in an intermediate state , similarly to the chemically inhibited experiments ( Figure 6D ) . A similar phenotypic morphology was observed with the RAR pan-antagonist BMS-493 ( Figure 6E; Table 2 ) . Both controls , RNAi and BMS-493-treated experiments , were histologically examined in hematoxylin-eosin stained sections ( 10 μm ) . The regenerating buds in RNAi treatments were malformed ( Figure 6F , arrow; Table 2 ) , and the regeneration niches were mostly taken over by disorganized masses of undifferentiated cells . In mild cases , buds reached progressive developmental stages but failed to invaginate properly ( Figure 6G , arrow ) . A similar histological phenotype is observed in BMS-493-treated fragments exhibiting malformed epithelial spheres ( Figure 6H , arrows; Table 2 ) . Next , we examined the gain-of-function effects of RA on WBR by the systemic administration of 10 μM or 50 μM all-trans RA to regenerating blood vessel fragments ( n = 28 ) ( Table 2 ) . Control blood vessel fragments ( n = 11 ) were immersed in seawater containing DMSO alone . On day 6 , control fragments were still in the midregeneration stage , characterized by movements within the tunic matrix . However , in RA-treated fragments , vessels had already completed orientation within the tunic matrix ( day 1 ) , completed vasculature aggregations ( days 1–4 ) , and developed into opaque vessel masses . Furthermore , at the early day-6 stage , 11 of the RA-treated fragments ( 39% ) had already developed fully functional adult zooids from the vessel masses with active atrial and peribranchial siphons . On day 6 , all RA-treated fragments were sacrificed for histological examinations . Results showed further accelerated regeneration throughout the entire vasculature . Haemocytes aggregated systemically along blood vessels and not only near the vessel epithelium . Buds were present in many blood vessels and revealed extremely high variation in developmental states , from cell aggregates to well-developed zooids . The regenerating zooids manifested fully differentiated organ systems ( Figure 7; Table 1 ) , including stigmata with cilia ( Figure 7A , arrows ) , a completed digestive system , and a pair of contracting siphons carrying normally developed tentacles . Different from normal blastogenic buds , blood cells colonized extensively these regenerating buds between atrial folds and throughout internal cavities ( Figure 7B , arrows ) . The consecutive generation of palleal buds , at progressive developmental stages , had already started to grow from both sides of the regenerating zooids . In one case , the regenerating bud progressed to the stage where secondary buds formed on the primary palleal buds ( Figure 7C , arrow and arrowhead , respectively ) . In marked contrast to control regenerating fragments where only a single bud developed , numerous buds at different regeneration niches had simultaneously reached the final stages of organogenesis ( 2–5 functional zooids ) ( Figure 7D , arrows ) , exhibiting fully differentiated organ systems that reproduced asexually through palleal budding . At high concentrations of RA ( 50 μM ) , buds developed abnormally with an enlarged digestive tract , pyloric gland , and swelling of the atrial folds ( unpublished data ) .
Whereas most animals share the ability to repair damaged tissue , the phenomenon of massive body restoration is a rare event in highly complex animals [5] . Here , we studied the unique whole body regeneration phenomenon in a colonial urochordate , revealing the capability of rebuilding the entire zooid body from minute vasculature fragments . We defined three consecutive developmental phases , shedding light on a unique regeneration process that differs from known regeneration model systems in several fundamental criteria ( Table 3 ) . In other epimorphic regenerations , restoration starts with the formation of a transient structure termed blastema [38] . However , in the Botrylloides WBR , instead of blastema , regeneration niches are structured de novo inside vasculature . Furthermore , while a single restoration center is induced in all other model systems studied [5] , a “multifocal regeneration niche” system is generated in the Botrylloides WBR . A possible explanation for this regeneration strategy in Botrylloides is the presence of a systemic inductive regeneration cue . Regeneration in Animalia is regularly induced by inductive signals [39 , 40] . In model systems , such as regeneration of limb in salamander or fin in zebrafish [41 , 42] it is regulated by locally induced signals . We suggest that in B . leachi , WBR , in contrast to other cases , signals might be propelled throughout circulation ( cellular and humoral elements ) , leading to many regeneration foci in any single regeneration entity . The Botrylloides system is also the first documented case , to our knowledge , of WBR originating from circulating blood cells ( Table 3 ) and reveals the capability of restoring not only the soma , but also the germ line [10 , 11] . As previously reported [10 , 11 , 19] , Botrylloides WBR initiates through aggregates of haemocytes within newly formed vasculature niches; each housing up to 100 small round cells . Any haphazardly isolated peripheral ampulla ( out of hundreds of ampullae fringing in a single colony ) , containing only 100–300 circulating cells [10 , 11] , is capable of successful WBR . What is the cellular source of the regenerating bud ? Their small cell size , cellular morphology , and strong ubiquitous hematoxylin staining suggest primitive stem cell origin [19]; however , direct proof should be addressed in further controlled studies . In Botrylloides WBR , as in embryonic development [43] , the transformation of the initial state of aggregated cells into complete animal structures reveals a fundamental attribute of animal development . Both processes have very different starting points ( a fertilized egg versus pluripotent blood cells ) ; yet both converge at a common endpoint of adult morphology through the establishment of body axes , self-assembly , and differentiation of tissues and organs [44 , 45] . We can draw close morphological similarities between the initial developmental stages in Botrylloides WBR ( cell aggregates , development of blastula-like structures , and invaginations of epithelial layers ) and fundamental embryonic developmental stages ( morula , blastula , and gastrula ) . A similar resemblance transpires when comparing Hydra WBR from dissociated cells [46] to embryonic development , implying a basic evolutionary root underlying whole body restoration events [47 , 48] . The conserved signaling pathways governing the development of tissues and organs are one of the major objects in developmental biology . Redeployment of these embryonic signaling pathways at different biological contexts enables quicker responses by multicellular organisms to adverse environmental conditions , such as RA signaling in the regeneration of specific tissues and organs in various vertebrate model systems [26 , 49 , 50] . A functional example of redeployment of a developmental mechanism comes from the mammalian central nervous system . Embryonic central nervous system undergoes regeneration through RARβ2-stimulated RA signaling , unlike adult central nervous system , which does not express RARβ2 . Redeployment of RA signaling through overexpression of RARβ2 indeed promotes functional regeneration of adult central nervous system [31] . It had already been shown that RAR is expressed in mesenchymal cells of developing buds in the colonial tunicate P . misakiensis , where it regulates developmental aspects in normal budding processes , including the induction of a secondary axis in developing buds [51–53] . Following the evaluation of the biological redeployment strategy , this study reveals , for the first time to our knowledge , the important roles of RA in botryllid ascidian WBR . We show that RAR is expressed during WBR , specifically in haemocytes within the vasculature . In contrast to tissue- and organ-specific expression patterns of RA receptors in vertebrates , RAR expression in botryllid regeneration does not follow developmental processes expressed in normal blastogenesis of tissue and organ structuring , but is expressed ubiquitously throughout the entire body . This observation suggests that in addition to its role in early stages of regeneration , RAR expression also represents other RA activities , such as maintenance or other yet unknown proceedings . We further show that the entire body of developing zooids is susceptible to RA signaling . The use of both DEAB , Citral , BMS-493 , and RNAi-mediated knockdown of Bl-RAR to disrupt RA synthesis and RA receptor function , collectively result in WBR arrest and bud malformation . The administration of all-trans RA into fragments of blood vessels results in accelerated regeneration and the unusual development of multibuds , leading to restored colonies with multiple functional zooids . This observation could indicate that in regular Botrylloides WBR , the systemic developmental inhibition of all buds but one , could be achieved by controlling the levels of RA transcriptional cascade . We found , during early stages of WBR , that highly Aldedh-positive macrophages form foci of expression that correspond to foci of multiple initiations of regeneration , strengthening the notion that RA is required for the early stages of this process [54 , 55] . Regeneration as a central discipline in biology holds great promise not only for the understanding of species-specific developmental issues , but also for deduction of its evolutionary roots and for medical applications [56] . In the past few years , ascidians have become model organisms for the study of a wide variety of biological phenomena , including developmental processes , embryogenesis , stem cell biology , and immunology [57–60] . This group of organisms demonstrates basic mechanisms of biological phenomena , similar to those observed in vertebrates . In light of the unique regenerative power of botryllid ascidians , WBR in B . leachi may serve as a model system for studying the evolutionary roots of organ regeneration , lost in most vertebrate taxa .
Colonies of B . leachi were collected from underneath stones in shallow waters along the Mediterranean coast of Israel . The colonies , with thin layers of stony material attached to them , were carefully pealed off from the substratum using industrial razor blades and individually tied with fine threads onto 5 × 7 . 5-cm glass slides . Colonies were cultured on slides placed in 17-l tanks of standing seawater system [33] . Within several days of culture , ampullar contractions and expansions resulted in complete or partial sliding of colonies from their natural calcareous substrate onto the glass slides . Colonies were cleaned weekly by carefully removing debris ( empty substrates and other settled organisms ) from the glass slides with industrial razor blades and fine brushes . Isolation of marginal ampullae and blood vessel fragments was performed under a dissecting microscope using an industrial razor blade and a fine tungsten needle . Next , the dissected colonies were removed from the glass slides onto other slides , and the colonial fragments were cut further into smaller fragments using a fine tungsten needle . Blood vessel fragments were left to regenerate in 17-l tanks and were monitored daily by observing slides under a dissecting microscope . Whole fragment pictures were taken with a Supercam camera ( Applitec , http://www . applitec . co . il ) . Immunohistochemistry and general histology were performed as described by Lapidot [61] . Dissected blood vessels were fixed in Bouin's solution for 1 h , dehydrated in 70% ethanol , and embedded in paraffin wax . For general histological observations , we used the regular hematoxylin-eosin staining protocol . For immunohistochemistry , serial 5-μm thick sections were prepared , attached to SuperFrost Plus microscope slides ( Menzel-Glaser , http://www . menzel . de ) , dewaxed , and antigen retrieval was performed by microwaving ( 480 W ) the sections for 30 min in 10-mM citrate buffer ( 600 ml , [pH 6] , 30 min ) . After a 5-min cooling period , up to 1 l in volume distilled water was added , and the slides were incubated for additional 10 min at room temperature , followed by several washes with TBS . Endogenous peroxidase activity in the sections was blocked by incubation in Dako EnVision+ System Peroxidase block ( AEC , catalogue number K4005 , http://www . dako . com , ) for 6 min , and the slides were washed with water . Nonspecific binding sites were blocked by incubation in 1% BSA ( Sigma , http://www . sigmaaldrich . com ) in 50 mM TBS−0 . 1% Tween 20−0 . 01% Triton × 100 for 1 h at room temperature . The slides were washed with TBS for 10 min , stained with a mouse anti-PCNA monoclonal antibody ( Dakocytomation , 1:100 in TBS , 100 μl per slide ) for 1 h at room temperature , and then incubated for 12 h at 4 °C . After washing with TBS for 5 min × 3 , peroxidase-labeled polymer-conjugated goat anti-mouse immunoglobulin ( Dako EnVision+ ) was added; the sections were incubated for 1 h and washed with TBS for 5 min × 3 . Then , 2–3 drops of AEC substrate , chromogen ( 3-amino-9-ethylcarbazole hydroperoxide , Dako , EnVision+ ) was added for 20–60 min ( upon judgment ) , and the reaction was stopped by rinsing with distilled water . Control and experimental sections were mounted with Hydromount ( National Diagnostics , http://www . nationaldiagnostics . com ) , covered with 24 × 50-mm cover glasses ( Medite , http://www . medite . ch ) , observed with the Leica DMIRE2 inverted microscope , and photographed with a Leica FX300 camera ( http://www . leica . com ) . Total RNA was isolated from fragments of regenerating blood vessels with RNeasy Mini or Midi kits ( Qiagen , http://www . qiagen . com ) . A cDNA fragment of 808 bp was amplified using degenerate oligonucleotide primers ( forward: GGVTGYAAGGGITTCTT; reverse: TTCATVAKCATYTTIGGGAA ) directed against two conserved regions of the ligand-binding domain of hormone receptors present in all RARs . Amplification was obtained using Bl-RAR sequence specific primers ( forward: TCGACGCTTTCGGGCATAC , reverse: AAGACGGCAAAGCGGGAGAG ) . A B . leachi Raldh cDNA fragment of 337 bp was cloned from cDNA of regenerating blood vessels and amplified using sequence specific primers ( forward: AGAATTTCCTTGGAGCTTGG; reverse: ACCCTGTTCAATGTCGCTG ) . A B . leachi cytoplasmic actin cDNA fragment of 338 bp was cloned from cDNA of B . leachi colonies and amplified using the following primers ( forward: GAAATCGTGCGTGACATCAAAG; reverse: GCGGTGATTCCCTTCTGCATAC ) . BLAST results confirmed it as a cytoplasmic actin . PCR products of B . leachi Actin , RAR , and Raldh were cloned using pGEM-T easy vector ( Promega , http://www . promega . com and Qiagen ) PCR cloning kit , respectively . Sequences related to Bl-RAR and Bl-Raldh were identified with BLAST , and ClustalW was used to obtain the multiple alignment of both sequences . All sequences of RAR- and Raldh-related proteins used for the multiple alignments were obtained from the EMBL Nucleotide Sequence Database ( http://www . ebi . ac . uk/embl/ ) . RT-PCR analysis was performed by extracting total RNA from regenerating blood vessels at temporal stages using RNeasy Mini or Midi kits ( Qiagen ) as a template . After cDNA production by reverse transcriptase , PCR reactions were performed ( 40 cycles at 94 °C for 30 s , 30 s at 50 °C , and 60 s at 72 °C ) using specific primers corresponding to Bl-RAR and Bl-cytoactin . B . leachi colonies , colonial fragments containing blood vessels , and regenerating fragments were fixed overnight in 4% paraformaldehyde , dehydrated in 70% methanol , embedded in paraffin , and cut into 5-μm sections . Both Bl-RAR and Bl-Raldh clones were used to obtain sense and antisense DIG-labeled RNA probes that were synthesized using the DIG RNA labeling kit ( SP6/T7 , Roche Molecular Biochemicals , http://www . roche . com ) . Hybridization of probes to tissue sections was performed according to Breitschopf [62] for paraffin-embedded tissue . DIG-labeled RNAs on samples were revealed using anti-DIG antibody ( Roche ) . Samples were observed with the Leica DMIRE2 inverted microscope and photographed with a Leica FX300 camera . Stock solutions were made in the dark . All experiments were carried out in aluminum foil covered plastic tanks . Solutions were changed every other day . All-trans RA was purchased from Sigma , and a 0 . 1-mM stock solution was made by adding DMSO . Stock solutions were frozen in −20 °C as aliquots of 100 μl . Immediately prior to experiment , an aliquot of stock solution was thawed and further diluted when added in different concentrations to the experimental 2-l seawater tanks . Citral ( cis + trans , Fluka [Sigma] ) 60-mM stock solution was prepared in 100% ethanol and was further diluted to 20 μM and 60 μM by administrating into 1-l seawater tanks . We prepared DEAB ( Fluka ) stock solution by dissolving 0 . 177 g DEAB in 1 ml DMSO within an aluminum covered Ependorf vial . The stock solution was further diluted to 10 μM and 100 μM by adding to 1-l seawater tanks . We prepared 25-μM BMS-493 by diluting 20 μl of 25 mM stock solution in 20 ml seawater . Glass slides with B . leachi fragments of different regeneration phases were placed in the water tanks , left to regenerate , and examined daily under a dissecting microscope . Whole fragments' pictures were taken with a Supercam camera ( Applitec ) . Three different siRNAs to Bl-RAR were generated using the Silencer siRNA Construction kit ( Ambion , http://www . ambion . com ) . The specific primers used were as follows: As1 5′-AAATGGTATGCCCGAAAGCGTCCTGTCTC-3′ and S1 5′-AAACGCTTTCGGGCATACCATCCTGTCTC-3′; As2 5′-AAACTACAAGAACCTCTCGTCCCTGTCTC-3′ and S2 5′-AAGACGAGAGGTTCTTGTAGTCCTGTCTC-3′; As3 5′-AACGCAACTTCAAAGTTGCGCCCTGTCTC-3′ and S3 5′-AAGCGCAACTTTGAAGTTGCGCCTGTCTC-3′ . Three additional sets of primers and an unrelated control siRNA [37] were generated by Ambion according to the Bl-RAR sequence provided . siRNA was delivered by the submersion of animal colonies for 2 d , followed by separating the blood vessels from the colonies and immersing the regenerating blood vessels in seawater containing 25 nM siRNA for an additional 6 d , during which the medium was changed every other day . Control siRNAs and siRNA-treated regenerating blood vessels were monitored daily and collected for further analyses . Apoptotic nuclei were stained using a Klenow fragEL DNA fragmentation detection kit ( TUNEL ) ( QIA21 , Calbiochem , http://www . emdbiosciences . com/CBC ) according to the manufacturer's protocol of paraffin-embedded tissue . Negative controls were generated by substituting the Klenow enzyme in the labeling reaction mix with dH2O .
The EMBL Nucleotide Sequence Database ( http://www . ebi . ac . uk/embl ) accession numbers cDNA fragments discussed in this paper are DQ523226 , EF125176 , and EF125177 .
|
Whole body regeneration ( WBR ) in Animalia is rare , confined to morphologically less complex taxa such as sponges , cnidarians , and flatworms . In the chordate phylum , only colonial ascidians ( invertebrate chordates also known as sea squirts ) have the documented ability to wholly regenerate . Once separated from the colony , any minute fragment of peripheral blood vessel ( about 1 mm in length , containing 100–300 blood cells ) of the colonial ascidian Botrylloides leachi regenerates an entire functional adult within one to three weeks . By following cellular and molecular events in Botrylloides WBR , we revealed that this system proceeds differently from regeneration events in other model organisms by several fundamental criteria . This is , for example , to our knowledge the first documented case of WBR initiating from circulating blood cells that restore not only the body tissue , but also the germ line . We found that retinoic acid ( RA ) signaling , previously reported in the regeneration of specific vertebrate tissues and organs , plays a major role in WBR via RA receptor expression throughout the entire regenerating animal . This suggests that RA signaling may have had ancestral roles in body restoration events . Elucidating the processes involved in this WBR system will improve our understanding of the nature of regeneration and the reduced regeneration capabilities represented in so many vertebrates .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods",
"Supporting",
"Information"
] |
[
"developmental",
"biology",
"cell",
"biology",
"vertebrates",
"evolutionary",
"biology"
] |
2007
|
Systemic Bud Induction and Retinoic Acid Signaling Underlie Whole Body Regeneration in the Urochordate Botrylloides leachi
|
Clincialtrials . gov , NCT01119521
Almost a quarter of the global population is infected with Mycobacterium tuberculosis ( M . tb ) [1] , placing millions at risk for tuberculosis ( TB ) disease . Five to 15% of infected individuals progress to active TB disease within their lifetimes . A spectrum of clinical disease severity may occur , associated with poorly regulated immune responses , including infection site and systemic inflammation and in some cases excessive inflammation [2 , 3] . TB disease-driven inflammation in the lung is characterized by presence of activated neutrophils , macrophages and lymphoid tissues in infected foci , as well as the presence of soluble inflammatory mediators at sites of disease [4–7] . Disease-driven systemic inflammation manifests with high levels of cytokines , chemokines , acute phase proteins , and other inflammatory mediators detectable in peripheral blood [8–10] . Peripheral blood transcriptomic profiling of patients with active TB disease demonstrated elevation of inflammatory gene expression pathways , including interferon ( IFN ) stimulated genes ( ISG ) , myeloid inflammatory genes , and FC receptor/complement pathway genes [11–14] . There is limited knowledge about blood transcriptomic changes prior to development of clinical TB disease . We recently identified and validated a whole blood transcriptional signature of risk of TB disease , detectable long before disease manifests [15] . This 16-gene signature likely represents a small fraction of the host immunological changes that characterize progression to active TB disease [11–14] . Here , we proposed that a more extensive interrogation of immunological changes prior to clinical disease manifestations would enhance understanding of TB disease progression . We hypothesized that the spectrum of outcomes of M . tb infection–from asymptomatic quiescent infection , to subclinical disease ( detectable only by special investigation ) , to symptomatic clinical TB disease [16]–is reflected by specific peripheral blood immunological and inflammatory profiles . We therefore applied a new in-depth transcriptomic , proteomic , and cellular analysis of blood collected from the adolescent cohort that was used to discover the above-mentioned whole blood signature of risk for TB disease . We also examined blood from an independent cohort of adults revaccinated with Bacille Calmette-Guérin ( BCG ) to validate functional adaptive immune response profiles that we found were associated with the inflammatory profiles observed during disease progression .
Whole blood RNA isolated from PAXgene tubes collected from ACS progressors and controls was analyzed by RNA Sequencing ( RNA-Seq ) . Nonlinear kinetic analysis of gene expression differences between the two groups was assessed over time , beginning 2 years before TB diagnosis of progressors . During progression to TB 1 , 494 genes were upregulated and 1 , 646 genes were downregulated in progressors , compared with controls . A wide range of kinetic changes in longitudinal expression of different genes was discernible ( Fig 1B and S2 Table ) . Transcriptional module enrichment analysis revealed that a hierarchy of biological processes drove the development of progression-associated changes in gene expression ( Fig 2A and S3 Table ) . Up-regulation of IFN response modules ( containing ISGs such as STAT1 , STAT2 , IFITs , GBPs , MX1 , OAS1 and IRF1 ) , preceded up-regulation of myeloid inflammation and monocyte modules ( containing MyD88 , ICAM1 , IL-6 amongst many others ) ( Fig 2A and 2B , S1A Fig ) . These changes preceded down-regulation of modules associated with specific lymphocyte cell populations ( Fig 2A , S3 Table ) . Further , differential expression of the 16 genes that comprise the previously described whole blood signature of risk for TB disease [15] preceded all other modules [18 , 19] , including the IFN response modules themselves , and their induction magnitude was among the highest detected ( Fig 2C ) . This result suggests that genes contained in the previously-described whole blood signature of risk comprise a novel subset of ISGs that were robustly differentially expressed at the earliest stages of TB progression . Since monocytes were strongly implicated in the module enrichment analysis above , suggesting either changes in peripheral monocyte numbers or changes in monocyte gene expression , we determined whether genes in the signature of risk were up-regulated in isolated monocytes from progressors compared to controls . Expression of FCGR1A/B and SERPING1 , the inflammasome adaptor ASC/PYCARD and NCF1 ( p47phox ) , which mediates inflammatory signaling in TB [20] , were increased in monocytes from progressors who expressed the whole blood signature of risk for TB , compared to controls who did not express the signature of risk ( S8 Table , S2A Fig ) . These results indicate that expression of the signature of risk of TB in whole blood may , in part , derive from enhanced expression of these transcripts in the monocyte population , suggesting that systemic monocyte activation , in addition to increased monocyte abundance ( see below ) , may comprise a hallmark of TB progression that precedes the neutrophil activation associated with disease [11] . We complemented blood transcriptional analysis with proteomic analysis of plasma collected longitudinally at the same time points prior to disease manifestation in progressors from the adolescent cohort . Relative concentrations of >3 , 000 proteins were quantified with multiplexed slow off-rate modified DNA aptamers ( SOMAmer reagents ) [21 , 22] . During progression to TB , levels of 179 plasma proteins increased while 251 decreased , compared with controls ( S1B Fig , S4 Table ) . Module enrichment analysis of plasma proteins revealed coordinated kinetic changes during progression , comparable to what was observed by transcriptomic analysis ( S5 Table ) . Complement cascade modules ( containing Complement factor I , H and B , C1s , C2 , C3b , C5 , C9 ) , were up-regulated earliest during progression , compared with controls ( Fig 3A ) . These changes were present at the same time as upregulation of IFN response genes shown by transcriptomic analysis . This was followed by changes in blood coagulation modules ( containing coagulation factor X , fibrinogen , D-dimer , fibrinogen gamma chain , thrombospondin 1 , SERPIN A1 and D1 , and platelet factor 4 ) and by myeloid inflammation modules ( containing CXCL9 , CCL1 , CD163 , IL-6 and RANTES ) , which emerged after complement activation , around 200 days before TB diagnosis . Finally , several modules associated with tissue remodeling ( containing MMP1 , MMP9 , MMP12 and tissue inhibitor of MMPs 2 and 3 ) , hemostasis , and platelet activation emerged within 200 days before TB ( Fig 3A , 3B and 3C ) . Despite the detection of many individual proteins down-regulated during TB progression ( S4 Table ) no significant enrichment for down-regulated protein modules was detected ( S5 Table ) . Many proteins that significantly changed in abundance during progression could not be mapped to protein modules . Those with demonstrated importance for host defense against M . tb included granulocyte peptides neutrophil defensin 1 ( HNP-1 , DEFA1 ) , cathelicidin ( CAMP ) , beta-defensin-110 ( DEFB110 ) and -131 ( DEFB113 ) and neutrophil-activating peptide 2 ( NAP2 ) –all of which were upregulated . Leukotriene A4 hydrolase ( LTA4H ) , which has been implicated as an important regulator of the balance between protective and pathogenic inflammation [3 , 23] , and NK cell products ( killer cell immunoglobulin-like receptor 2DL4 ( KIR2DL4 ) and granzyme K ) , were downregulated . Lastly , levels of total IgG and IgA were elevated in plasma from progressors ( S4 Table ) . To identify links between temporal changes in blood mRNA and plasma proteome data , we tested whether specific modules were over-represented within sets of proteins that exhibited progression-associated changes in abundance with kinetics that were consistent with the kinetics of key transcriptional modules . Specifically , we determined which proteins ( and protein modules ) showed differential abundance at deviation days that coincided with the interquartile range ( IQR ) of deviation days for differentially regulated genes from the ACS signature of risk of TB and the IFN response and inflammation modules ( S9 Table ) . No proteins with significant enrichment ( p<0 . 05 ) within defined protein modules ( S5 Table ) had deviation days that coincided with the deviation day IQR of the 16 genes within the whole blood signature of risk of TB . However , proteins that were temporally associated with the IFN response modules included CXCL10 ( IP-10 ) , STAT1 and Tryptophanyl-tRNA ligase ( WARS , WRS , SYWG ) ; and those proteins mapping to the inflammation pathway included Calgranulin C ( S100A12 or EN-RAGE ) , alpha-1-antitrypsin ( SERPINA1 ) and Myeloblastin or proteinase 3 ( PR3 ) , MMP9 and Ficolin-1 ( FCN1 ) ( S9 Table ) . These data provide protein-level confirmation for the finding from gene expression analyses that the IFN response precedes myeloid inflammation . Our transcriptomic and proteomic data highlight profound inflammatory processes during progression that are detectable more than a year before TB diagnosis . Since inflammation is known to regulate both myelopoiesis and lymphopoiesis [24] , we sought to investigate changes in peripheral blood cell subsets during progression . First , we investigated kinetic changes in whole blood transcripts associated with granulocytes , monocytes , T cells and B cells . mRNA expression of FFAR2 ( a representative granulocyte gene ) and CD14 ( a representative monocyte gene ) were significantly upregulated while CD28 ( a representative T cell gene ) and CD79A ( a representative B cell gene ) were downregulated in progressors relative to controls ( Fig 4A ) . Analysis of gene modules representing these four blood cell subsets supported these changes ( Fig 4B ) and suggested that modulation of myeloid and lymphoid compartments during TB progression was secondary to the up-regulation of the genes in the whole blood signature of risk for TB and the induction of IFN response genes in general , which markedly preceded changes in peripheral blood cellularity . To confirm these data , we enumerated proportions of major blood cell subsets by flow cytometry ( S3A Fig ) . Within 200 days of TB diagnosis relative proportions of CD14+ monocytes were significantly increased while CD3+ T cells were depleted in progressors , relative to controls ( Fig 4C ) . These changes were accompanied by T cell activation in progressors , indicated by elevated expression of HLA-DR on CD4 T cells , with concomitant decreases in relative proportions of CD45RA-CCR7+ central memory CD4 and CD8 T cells ( Fig 4C ) . T cells and specifically antigen-specific IFNγ-expressing CD4 T cells are necessary for successful control of M . tb infection [25 , 26] . In vitro studies have shown that type I IFNs can inhibit the macrophage antimycobacterial response mediated by IFNγ [27] . To determine if expression of the whole blood signature of risk for TB and IFN response gene module is associated with concomitant functional changes to T cells , we performed RNA-Seq transcriptome profiling of T cells sorted from adolescent progressor and control PBMCs . Comparing transcriptomes of T cells obtained from TB progressors that expressed the whole blood signature of risk to those from controls that did not express the whole blood signature of risk revealed 277 genes that were significantly differentially expressed between the populations ( S6 Table ) . Modular analysis showed that genes associated with hypoxia response and cell cycle were prominent amongst genes that were expressed at lower and higher levels , respectively , in progressor T cells ( S2B and S2C Fig and S7 Table ) . Another striking result was that the Th17-associated genes IL-17F , IL-23R , RORC and CCR2 were expressed at lower levels in T cells from progressors expressing the whole blood signature of risk for TB ( Fig 5A ) . This suggests that induction of Th17 responses may be inhibited in progressors with high ISG expression in the blood—a result that may have implications for vaccination in persons exposed to M . tb . To test the hypothesis that expression of the ISGs comprising the signature of risk for TB in whole blood is associated with a concomitant suppression of Th17 function , we analyzed T cell responses to Bacille Calmette-Guerin BCG revaccination in an independent cohort of South African adults with latent M . tb infection [17] . This cohort exhibited a broad range of scores for the whole blood signature of risk for TB , spanning low to high IFN response magnitudes before BCG administration . Three weeks after BCG administration BCG-specific CD4 T cells expressing Th1 cytokines and IL-17 were enumerated by flow cytometry ( S3B Fig and Fig 5B ) . Frequencies of BCG-specific CD4 T cells that co-expressed IFNγ and IL-17 as well as relative proportions of BCG-specific IFNγ+ CD4 T cells that co-expressed IL-17 were inversely correlated with whole blood expression of the signature of risk for TB ( Fig 5C ) . By contrast , frequencies of BCG-specific CD4 T cells expressing any cytokine , co-expressing IFNγ and TNF , or relative proportions of BCG-specific IFNγ+ CD4 T cells that co-expressed TNF , were not associated with the whole blood expression of the signature of risk for TB , and neither were frequencies of total IL-17+ CD4 T cells ( S4 Fig ) . These data suggest that underlying systemic inflammatory perturbations that are associated with risk of TB progression , as indicated by the signature of TB risk score , may interfere with induction or maintenance of antigen-specific Th17 cells after vaccination .
We report orchestrated , sequential changes in blood mRNA , soluble protein and cellular responses during the transition from asymptomatic M . tb infection to active pulmonary TB disease in a prospective , longitudinal cohort of adolescents . Particularly striking was that these changes exhibited a spectrum of kinetics , with a minority of responses exhibiting detectable differences 1–2 years before diagnosis , and the largest suite of differences between progressors and controls being observed most proximal to TB disease . These data suggest that TB progression is a slow but steady transition from an immunologically quiescent state , via nondiscrete progressive stages of inflammatory perturbation to the highly inflammatory , clinical manifestations ( fever , cough , hemoptysis and weight loss ) of microbiologically confirmed , active TB disease . Our results suggest that an intermediate M . tb infection state that appears consistent with incipient or subclinical TB in individuals with no other signs of TB disease , can be revealed with blood biomarkers , such as the whole blood signature of risk for TB progression [15] , and specific elevation of IFN response gene modules and activation of the complement cascade . Our recent report of infants with QFT conversion values which exceeded > 4IU/mL of IFN-γ and who were at exceptionally high risk of TB disease within 6 months of QFT conversion [28] , support this finding . This is also consistent with the description of subclinical TB disease recently reported in a proportion of asymptomatic , antiretroviral therapy naïve , HIV-infected individuals with latent M . tb infection , who presented with pulmonary abnormalities on combined positron emission and computed tomography ( PET-CT ) [29] . Of note , four of the ten individuals with evidence of subclinical disease developed symptomatic active TB within 6 months , suggesting that such individuals are progressing towards clinical disease . Our study lends a timeline to the different immunological stages of TB progression . The first of these , detected up to 18 months before TB diagnosis , included elevated expression of the signature of risk genes themselves and expression of IFN responses genes and complement activation more broadly . Although we did not detect elevated levels of soluble IFNα , β or IFNγ proteins themselves in plasma from progressors , as was also observed in a proteomic study of TB disease [22] , it is well-established that expression of type I IFNs by M . tb-infected macrophages can be activated by bacterial DNA via STING signaling following binding to cGAS [30–32] . Further , M . tb-induced mitochondrial stress and abundance of mitochondrial DNA in the cytosol of infected macrophages were also recently shown to drive IFNβ expression [33] . These papers suggest that the IFN response signature may be directly activated by M . tb bacilli . However , our data do not reveal whether the IFN response cascade is induced by type I or II IFNs . Other well-known inducers of type I IFNs , such as viral infections , may also underlie the IFN response and even contribute to a higher risk of progression to active TB [25 , 34] . For example , influenza infection has been shown to reduce host resistance to M . tb in mice [35] . Plasma protein levels of the established interferon ( type I and/or type II ) -induced proteins CXCL10 , STAT1 and Tryptophanyl-tRNA ligase were up-regulated with kinetics that were consistent with the up-regulation kinetics for the transcriptomic IFN response . Further , increases in protein levels of the inflammatory proteins Calgranulin C , alpha-1-antitrypsin , Myeloblastin , MMP9 and Ficolin-1 were temporally associated with upregulation of the inflammation transcriptomic module . Calgranulin C , and the serine proteases alpha-1-antitrypsin and Myeloblastin are involved in the neutrophil response and have been previously implicated in mycobacteria-induced inflammation [36] . Interestingly , the M . tb secreted proteins Ag85 and ESAT-6 have been shown to induce expression of Ficolin-1 [37] and MMP-9 [38] , respectively . Ficolin-1 is a pattern recognition molecule that can activate complement via the lectin pathway [39] and MMP-9 plays an important role in macrophage recruitment and granuloma establishment after M . tb infection [38] . These data further support the interpretation that the inflammatory responses observed during progression may be directly triggered by increased bacterial replication . We also observed a strong complement activation signal during the very early stages of TB progression , which coincided with elevation of the IFN response . In light of the well-described roles of complement components , such as C1 , C3 and C4 , in M . tb recognition and phagocytosis [40 , 41] , these data are also consistent with host innate sensing of M . tb in progressors , either via increased pathogen load or greater access to the bacterium . We were not able to establish from our data whether complement was activated through a particular pathway . Mycobacteria are able to activate complement through the antibody-dependent and antibody-independent classical pathways , alternative pathway activation and the lectin pathway [39 , 42] . We observed upregulation of components consistent with activation of the classical pathway via C1q , while levels of total IgG and IgA were also upregulated in progressors . Differential abundance of mannose-binding lectin ( MBL ) and ficolin-1 proteins in plasma also implicate activation of the lectin pathway . Finally , while a number of proteins involved in activation of the alternative pathway were elevated in progressors , these can also indicate complement activation by any pathway [39] . Complement may also be activated by pro-inflammatory stimuli and , in turn , components of the complement cascade are known regulators of inflammation [39] . Our data thus highlight that the complex interplay between M . tb , inflammation , antibody responses and complement activation needs greater exploration . Secondary to these IFN responses and complement activation during progression , increased myeloid cell inflammation , platelet activation and blood coagulation , with concurrent enrichment of peripheral blood monocytes and other myeloid cells were observed 12 to 6 months before disease diagnosis . Finally , within the most proximal 6 months before TB disease , changes in lymphocytes , including suppressed T and B cells and enrichment in neutrophils , were detected . The latter coincided with activation of tissue remodeling pathways that included elevated expression of several MMPs . High MMP concentrations correlate with lung immunopathology in TB disease , demonstrating the role of MMPs as effectors of matrix destruction in TB [43 , 44] . We speculate that our data suggest that breakdown of the extracellular lung matrix may occur months before clinical TB manifestation . If so , intervention during early stages of progression may allow prevention of pulmonary caseation , necrosis and cavitation , which are associated with poor treatment outcome [45] . Our findings are consistent with those of a recent study of blood transcriptional signatures in the cynomolgus macaque model of TB [46] . This study revealed elevated IFN responses , myeloid inflammation , complement activation and coagulation/platelet and myeloid lineage pathways , and decreased T cell , B cell and cytotoxicity pathways , in M . tb-infected macaques prior to clinical manifestation and divergence into active and latent TB . Further , macaques that ultimately developed active TB had elevated expression of IFN response signatures and lower expression of lymphoid cell gene modules by 30 days post-M . tb infection , compared with animals that maintained latent infection [46] . Our findings also complement those of numerous investigators who described transcriptomic signatures of active TB disease , characterized by highly elevated expression of ISGs and upregulated myeloid inflammation , neutrophil and FC receptor/complement pathways [11–14] . Such inflammatory signatures were previously also reported in a small proportion of apparently healthy individuals and led to the recognition that asymptomatic infection with M . tb , traditionally referred to as latent TB , exists as a spectrum that ranges from quiescence to subclinical TB disease [16] . Our results support this interpretation and add a timeline to the transition through apparent stages within the spectrum . It should be noted that transition from quiescent infection , through incipient and subclinical TB to active pulmonary disease in different individuals was highly heterogeneous . A limitation of our study is that the time of exposure and/or M . tb infection in most progressors and controls was unknown , precluding interpretation of the events that precede establishment of M . tb infection . Finally , our T cell transcriptomic results demonstrate that progression was associated with modulation of the functional states of T cells , particularly suppressed expression of genes associated with the Th17 compartment [47 , 48] in progressors that expressed the whole blood signature of risk for TB . Systemic expression of IFN response genes that comprise the signature of risk occurred concomitantly with Th17 inhibition . A negative correlation between type I IFN responses and Th17 responses has been reported in other systems [49–51] . This link between high expression of ISG as measured by the signature of risk of TB , and an alteration in T cell functional capacity was confirmed by analysis of an independent cohort . In South African adults who expressed the signature for risk of TB in whole blood , BCG revaccination induced significantly lower frequencies of IFNγ+IL-17+ and decreased proportions of IL-17-expressing IFNγ+ CD4 T cells . A major implication of this result is that immune responsiveness to vaccination may be modulated by the inflammatory milieu associated with progression to active TB , and even other immune modulations that result in systemic persistent expression of IFN response genes , such as viral infections . Further research is required to dissect the mechanistic link between inflammatory and cellular events that may underlie this observation , and to understand the true implications of this finding . Our study shows that sequential inflammatory dynamics precede TB disease manifestation characterized by specific alterations in blood transcriptomic , proteomic and cellular signatures . The detectable immunological and tissue remodeling perturbations observed in progressors suggest that new vaccination and drug treatment strategies and/or host-directed therapies may be required to control M . tb in persons with subclinical TB disease , while identifying potential targets ( and potential targets to avoid ) for successful interventional approaches to prevent progression to active TB . Careful investigation of this phenomenon is warranted .
We analyzed samples from M . tb-infected participants of the South African Adolescent Cohort Study ( ACS ) , previously evaluated to identify and validate the signature for risk of TB [15] . Briefly , 6 , 363 healthy adolescents , aged 12–18 years , were enrolled between July 2005 and April 2007 and follow-up was completed by February 2009 . Approximately half of the adolescents were evaluated at enrollment and every 6 months during 2 years follow-up; the other half was evaluated at baseline and at 2 years . At enrollment and at each visit , clinical data were collected , 2 . 5mL blood was collected directly into PAXgene blood RNA tubes ( PreAnalytiX ) and blood was collected in Cell Preparation Tubes ( BD Biosciences ) and peripheral blood mononuclear cells and plasma were isolated using density gradient centrifugation . Only adolescents with M . tb infection at enrollment , or those who developed active TB disease more than 6 months after M . tb infection was first detected were included in our analyses , diagnosed by a positive QFT ( Qiagen; >0 . 35 IU/mL ) and/or a positive TST ( 0 . 1mL dose of Purified Protein Derivative RT-23 , 2-TU , Staten Serum Institute; >10mm ) . According to South African policy , QFT and/or TST positive adolescents were not given therapy to prevent tuberculosis disease . Progressors were adolescents who developed active TB disease during follow-up , defined as intrathoracic disease , with either two sputum smears positive for acid-fast bacilli or one positive sputum culture confirmed as M . tb complex ( mycobacterial growth indicator tube , BD BioSciences ) . For each progressor , two matched controls who remained healthy during follow-up were selected and matched by age at enrolment , gender , ethnicity , school of attendance , and presence or absence of prior episodes of tuberculosis disease ( Table 1 and Fig 1 ) . Participants were excluded if they developed tuberculosis disease within 6 months of enrollment or QFT and/or TST conversion , to exclude early asymptomatic disease that could have been present at the time of evaluation , or if they were HIV infected . Participants with diagnosed or suspected tuberculosis disease were referred to a study-independent public health physician for treatment according to national tuberculosis control programs of South Africa . Effects of IFN responses on T cell responses after BCG revaccination were assessed in M . tb-infected adults who participated in a previous trial of BCG revaccination [17 , 52] . Briefly , we recruited healthy 18 to 40 year old South African adults , who were strongly TST positive ( ≥ 15mm induration when tested with PPD RT-23 ) ; HIV-seronegative; received BCG at birth and had a visible BCG scar . In this phase I trial , participants , recruited from the population of Worcester in the Western Cape , South Africa were randomized in parallel into two groups in a 1:1 ratio as previously described [17 , 52] . Participants in the first group were observed for 7 months , then vaccinated with BCG , and subsequently treated with isoniazid ( INH ) 6 months later ( Observation-BCG-INH ) . Participants in the second group received a course of 6 months of INH within a maximum period of 7 months , followed by BCG vaccination ( INH-BCG-Observation ) . Danish strain 1331 BCG Vaccine SSI ( Statens Serum Institut , Copenhagen , Denmark ) , the BCG vaccine used in the South African national immunization program and one of the most widely administered BCG vaccines worldwide , was administered intradermally at an adult dose of 2 to 8 x 105 CFUs . INH ( Westward Pharmaceutical Corporation , Eatontown , NJ , USA ) was administered daily at 5mg/kg rounded up to the nearest 100mg ( maximum dose 300 mg/day ) , and INH adherence was monitored by pill counts at clinic visits and random urine INH metabolite testing ( 18 ) . All participants provided written , informed consent . Whole blood was collected in PAXGene tubes and in Sodium-Heparin tubes from participants and processed within 45 minutes of phlebotomy , as previously described ( 19 ) , at enrollment , 1 month after isozianid preventive therapy initiation , at BCG vaccination , at 3 and 5 weeks , and 1 year post-vaccination . RNA was isolated from PAXGene tubes as described above . Heparinized blood was stimulated and processed for measurement of T cell responses by whole blood intracellular cytokine staining ( WB-ICS ) assay , as previously described [17] . The signature of risk of TB was measured in samples collected before BCG re-vaccination and the functionality of the T cell response to BCG revaccination was measured 3 weeks after vaccination . ACS study protocol , including sample collection , utilization and analyses , were approved by the Human Research Ethics Committee of the Faculty of Health Sciences , University of Cape Town . Written informed consent was obtained from parents or legal guardians , and written informed assent from each adolescent . The BCG revaccination trial protocol , including sample collection , utilization and analyses , were approved by the Medicines Control Council ( MCC ) of South Africa , Human Research Ethics Committee ( HREC ) of the University of Cape Town and the University Hospitals Case Medical Center institutional review board . The trial was registered on ClinicalTrials . gov ( NCT01119521 ) . Written informed consent was obtained from all participants . Generation of the whole blood RNA-Seq data was previously described [15] . RNA was extracted from PAXgene tubes , globin transcripts were depleted and ( GlobinClear , Life Technologies ) cDNA libraries were prepared using Illumina mRNA-Seq Sample Prep Kit . RNA-Seq was performed by Expression Analysis Inc . , at 30 million 50bp paired-end reads , on Illumina HiSeq-2000 sequencers . For monocytes and T cells , RNA was extracted from cells left unstimulated , stimulated with M . tb antigens ( ESAT-6/CFP-10 or Ag85A/B; T cells ) , or infected with M . tb ( monocytes and T cells ) . RNA-Seq was performed by Expression Analysis Inc . as described [15] or ( Unstimulated and M . tb antigen-stimulated T cells ) Beijing Genomics Institute ( Shenzen , China ) after performing amplification ( Clontech SMARTer Universal Low Input RNA Kit ) . RNA-Seq alignment , QC , and gene-level summarization for whole blood , monocytes , and T cells were also performed as described [15] . Whole blood , monocyte and CD4 T cell RNA-Seq data was aligned to the hg19 human genome using gsnap [53] as in the original study [15] . Normalized gene-level expression estimates were derived from mapped read pairs following the procedure implemented previously [54] . Briefly , mapped read pairs were assigned to genes by collapsing all transcripts into a single gene model and counting the number of reads that fully overlap the resulting exons using htseq ( v . 0 . 6 . 0 ) [55] , with strict intersection and including strand information . Gene models for protein-coding genes were downloaded from Ensembl ( GRCh37 . 74 ) . Reads that mapped to multiple locations were only counted once and those mapping to ambiguous regions were excluded . Log2-transformed values of counts normalized by adjusted library counts were computed using the cpm function of the edgeR package [56] . For monocyte and CD4 T cell transcriptomic analyses , both RNA-Seq and qRT-PCR measurements of the ACS signature of risk of TB score ( S1 Table ) were used to classify samples as positive ( > 0 . 6 in both RNA-Seq and qRT-PCR ) or negative ( < 0 . 4 in both RNA-Seq and qRT-PCR ) , to ensure robust classification . Cryopreserved plasma samples collected from BD Vacutainer Cell Preparation Tubes with Sodium Heparin ( BD Biosciences ) were analysed by using SOMAscan Version 3+ 3000plex assay , a multiplexed modified DNA aptamer array that quantifies 3000 proteins at 3 different plasma dilutions , as reported previously [21 , 22] . Data from all samples were log2 transformed , normalized and calibrated using standard hybridization and calibration procedures . Prospective RNA-Seq data of progressors were realigned to the time point at which active tuberculosis was diagnosed ( TimeToDiagnosis ) , as described in [15] , thereby synchronizing the cohort with respect to outcome . Differences in gene-level mRNA expression or protein concentrations between each progressor sample and the average of demographically matched control samples were computed using the published ACS metadata ( S1 Table [15] ) . TimeToDiagnosis values were assigned to each sample according to the original definitions . The log2 fold change values between progressor and control biomarkers were modeled as a nonlinear function of TimeToDiagnosis for the entire population using the smooth . spline function in R with three degrees of freedom . Ninety-nine percent confidence intervals for the temporal trends were computed by performing 2000 iterations of spline fitting after bootstrap resampling from the full dataset . Cyropreserved PBMC from progressors and controls were thawed and used for M . tb antigen stimulation experiments to sort monocytes and T cells for transcriptomic analyses . Specifically , CD14+ monocytes were sorted by positive selection using Miltenyi CD14 microbeads on an AutoMACS Pro to a purity of >90% ( verified by flow cytometry ) . Two x 105 sorted monocytes were subsequently stimulated with 2x106 CFU/ml live H37Rv M . tb in 0 . 5mL final volume , or left unstimulated for 6 h at 37°C . Similarly , for analyses of T cells , thawed PBMC were rested at 37°C for 4–6 hours , and 1x106 live PBMC were stimulated with 1x106 CFU/ml live H37Rv in 0 . 5 mL final volume , at 37°C for 12 hr , or with pools of 15mer peptides overlapping by 10 amino acids ( 1μg/ml/peptide ) , of ESAT-6 and CFP-10 , or Ag85A and Ag85B . Stimulation in media with 0 . 27% DMSO served as the negative control . Anti-CD28 and anti-CD49d co-stimulatory antibodies ( 1μg/ml , BD Biosciences ) were added to the peptide pool stimulated and negative control conditions . After stimulation at 37°C for 12 hours , T cells were purified from PBMC by negative selection using Miltenyi Pan-T cell isolation kit on an AutoMACS Pro ( for peptide and negative controls ) , or manually using MACS columns under BSL-3 conditions for M . tb-stimulated samples to a purity of >99% ( verified by flow cytometry ) . Purified T cells were lysed in RNeasy RLT buffer ( QIAgen ) while purified monocytes and M . tb-stimulated T cells were lysed in PrimeStore MTM buffer ( Longhorn Vaccines and Diagnostics ) . RNA was extracted from sorted cell subsets using RNeasy Plus Micro kit ( QIAgen ) and subjected to RNA sequencing as described above . To test for coordinated changes in functionally-associated genes and proteins enrichment analysis was performed using predefined module definitions [18 , 19 , 57] . Module enrichments were performed by treating the timepoint of deviation between progressors and controls , defined as the day before TB diagnosis on which the 99% CI deviated from a log2 fold change of 0 , for each gene/protein as a predictor . Only modules with enrichment ( adjusted p-value < 0 . 05 ) and more than 9 genes or proteins with kinetic response were considered to have a significant kinetic response during progression . For the progressor and controls analyses , PBMC were stained anti-CCR7-PE ( 150503 ) at 37°C for 20 min and thereafter at 4°C for 30 mins with a cocktail of the following antibodies: CD45-FITC ( 2D1 ) , CD11c-PerCP-Cy5 . 5 ( Bu15 ) , CD19-PE-Cy7 ( SJ25C1 ) , CD45RA-APC ( 550855 ) , HLA-DR-AF700 ( L243 , BioLegend ) , CD14-Qdot585/605 ( TuK4 ) , CD3-V450 ( UCHT1 ) , CD8-Qdor655 ( 3B5 , Invitrogen ) , CD4-V500 ( RPA-T4 ) and the live/dead dye , 7AAD ( ViaProbe ) . All antibodies and dyes were from BD Biosciences , unless otherwise indicated . Cells were acquired on a BD LSRFortessa flow cytometer and analysed using FlowJo ( v9 . 2 ) . Dead cells and doublets were excluded before cell subset proportions were computed . The gating strategy on a representative donor sample is shown in S1A Fig . For the BCG revaccination analyses , cryopreserved , fixed whole blood samples were thawed and stained at 4°C in BD perm/wash buffer ( BD ) with CD3-ECD ( UCHT1 , Beckman-Coulter ) , CD4-Qdot605 ( S3 . 5 , Invitrogen ) , CD8-APC-H7 ( SK1 ) , TCR-γδ-BV421 ( B1 ) , CD56-BV711 ( HCD56 , Biolegend ) , IFNγ-Alexa700 ( B27 ) , TNFα-PECy7 ( MAb11 , eBioscience ) , IL-2-APC ( MQ1-17H2 ) , IL-17-Alexa488 ( N49-653 ) , IL-22-PE ( IC7821P , R&D Systems ) . All antibodies and dyes were from BD Biosciences , unless otherwise indicated . Cells were acquired on a BD LSRFortessa flow cytometer and analysed using FlowJo ( v9 . 2 ) . The gating strategy on a representative donor sample is shown in S1B Fig . Datasets are available in the online appendix , or for the RNA-Seq data , as follows: Whole blood transcriptomes: GSE79362 T cell and monocyte transcriptomes: GSE103147
|
To define biological mechanisms that underlie progression of Mycobacterium tuberculosis infection to active tuberculosis , we followed M . tuberculosis-infected adolescents longitudinally . Those who ultimately developed tuberculosis disease ( “progressors” ) were compared with matched controls , who remained healthy . Whole blood transcriptomic and plasma proteome analyses showed sequential modulation of immunological processes . Type I/II interferon signalling and complement cascade were elevated 18 months before tuberculosis diagnosis , while changes in myeloid inflammation , lymphoid , monocyte and neutrophil responses occurred more proximally to tuberculosis disease . Analysis of gene expression in purified T cells revealed early suppression of Th17 responses in progressors . This was confirmed in an adult BCG re-vaccination cohort , where expression of interferon response genes in blood was associated with suppression of IL-17 expression by BCG-specific CD4 T cells . We concluded that sequential inflammatory dynamics and immune alteration precede tuberculosis disease manifestations , with important implications for developing diagnostics , vaccines and host-directed therapies for tuberculosis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"blood",
"cells",
"inflammatory",
"diseases",
"medicine",
"and",
"health",
"sciences",
"body",
"fluids",
"immune",
"cells",
"immunology",
"tropical",
"diseases",
"bacterial",
"diseases",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"tuberculosis",
"proteins",
"gene",
"expression",
"t",
"cells",
"biochemistry",
"diagnostic",
"medicine",
"blood",
"anatomy",
"cell",
"biology",
"tuberculosis",
"diagnosis",
"and",
"management",
"monocytes",
"physiology",
"interferons",
"genetics",
"biology",
"and",
"life",
"sciences",
"cellular",
"types"
] |
2017
|
Sequential inflammatory processes define human progression from M. tuberculosis infection to tuberculosis disease
|
Insecticide resistance ( IR ) can undermine efforts to control vectors of public health importance . Aedes aegypti is the main vector of resurging diseases in the Americas such as yellow fever and dengue , and recently emerging chikungunya and Zika fever , which have caused unprecedented epidemics in the region . Vector control remains the primary intervention to prevent outbreaks of Aedes-transmitted diseases . In many high-risk regions , like southern Ecuador , we have limited information on IR . In this study , Ae . aegypti IR was measured across four cities in southern Ecuador using phenotypic assays and genetic screening for alleles associated with pyrethroid IR . Bottle bioassays showed significant inter-seasonal variation in resistance to deltamethrin , a pyrethroid commonly used by the Ministry of Health , and alpha-cypermethrin , as well as between-city differences in deltamethrin resistance . There was also a significant difference in phenotypic response to the organophosphate , Malathion , between two cities during the second sampling season . Frequencies of the resistant V1016I genotype ranged from 0 . 13 to 0 . 68 . Frequencies of the resistant F1534C genotype ranged from 0 . 63 to 1 . 0 , with sampled populations in Machala and Huaquillas at fixation for the resistant genotype in all sampled seasons . In Machala and Portovelo , there were statistically significant inter-seasonal variation in genotype frequencies for V1016I . Resistance levels were highest in Machala , a city with hyperendemic dengue transmission and historically intense insecticide use . Despite evidence that resistance alleles conferred phenotypic resistance to pyrethroids , there was not a precise correspondence between these indicators . For the F1534C gene , 17 . 6% of homozygous mutant mosquitoes and 70 . 8% of heterozygotes were susceptible , while for the V1016I gene , 45 . 6% homozygous mutants and 55 . 6% of heterozygotes were susceptible . This study shows spatiotemporal variability in IR in Ae . aegypti populations in southern coastal Ecuador , and provides an initial examination of IR in this region , helping to guide vector control efforts for Ae . aegypti .
In Ecuador , dengue , chikungunya , and Zika viruses are present and transmitted to people by the Aedes aegypti mosquito , causing a high burden of febrile illness in susceptible populations . This species is a particularly effective vector of these aboviruses because it has evolved to live in urban environments , lay its eggs in small containers of water in and around human dwellings , and feed preferentially on humans [1] . In Ecuador in 2016 , there were 13 , 612 reported cases of dengue fever [2] , 2 , 025 cases of chikungunya , [3] 3 , 531 cases of Zika fever [4] . For all three of these diseases , actual incidence in Ecuador is likely much higher than the reported figures indicate , because many cases are asymptomatic or mild , and access to laboratory diagnostics can be limited [5] . Furthermore , the co-circulation of these viruses can lead to more complex disease outcomes , and nonspecific febrile symptoms make clinical differentiation of the three diseases difficult [6] . To reduce the burden of disease , public health organizations rely heavily on vector control methods , particularly insecticide applications [7] . Additionally , individuals purchase products such as aerosols , repellents , and nets to reduce mosquito populations and prevent disease transmission at the household level , with low-income households in Machala , Ecuador , spending as much as 10% of their discretionary income on these items [5] . While vector control is considered to be the only tool available for the control of these arthropod-borne viruses , the extent to which these interventions produce significant reductions in disease burden has been difficult to ascertain [8 , 9] . There are four main categories of chemical insecticides regularly used for disease vector control: organophosphates , pyrethroids , carbamates , and organochlorines , with organophosphates and pyrethroids being the most widely used in Ecuador for Ae . aegypti control [10] . For all four of these insecticide classes , regular deployment in locations around the world has been associated with the development of insecticide resistance ( IR ) in targeted vector populations , resulting in resurgences of mosquito-borne diseases when vector control fails [11–13] . Although public health organizations recommend monitoring and managing IR [14] , these practices are often resource-intensive , meaning that many areas do not have the capacity to conduct regular IR testing . Additionally , the largely unregulated application of insecticide treatments at the household level can influence local-scale IR . In an experimental study , exposure to commercially available aerosolized insecticides , applied in a simulation of typical household use , resulted in significant increases in genotypic and phenotypic IR in Ae . aegypti [15] . These results , along with field observations [16–18] , indicate that fine-scale selection pressures can contribute to IR development . When resistance remains undetected , public health organizations may spend substantial time and resources applying insecticides that are ineffective [19] and may have negative environmental and human health impacts [20] . A better understanding of the spatial and temporal variability in IR would greatly enhance the ability of vector control to predict and mitigate IR in vector populations . Research has shown that IR status can differ dramatically across cities within a country [21–23] , but further work needs to be conducted to understand the spatial scale of this variability , particularly across cities with differing vector control needs and strategies . Furthermore , most available reports from within-country spatial scales determine IR status at a single point in time , but IR exhibits temporal fluctuations that should be taken into consideration [24 , 25] . This study addresses a substantial knowledge gap regarding IR in Ae . aegypti by investigating differences in IR across both space ( with four cities included in the study area ) and time ( with sampling occurring in three seasons ) , while considering both genetic and phenotypic lines of evidence to determine IR status . Furthermore , assays were conducted with multiple commonly-used insecticides , allowing for comparisons of effectiveness that can inform public health decision makers about local-scale vector-control . In Ecuador , vector control is conducted by field workers of the Ministry of Health ( MoH ) in arbovirus endemic areas , as well as focal control in and around homes with suspected arboviral infections . Interventions to control adult mosquitoes include indoor residual spraying ( IRS ) with deltamethrin ( pyrethroid ) and ultra-low volume ( ULV ) fumigation with malathion ( organophosphate ) . Interventions to control immature mosquitoes include application of an organophosphate larvicide ( temephos/Abate ) to containers with standing water , as well as community mobilization to eliminate larval habitats . The objectives of this study were to evaluate overall IR status and to describe seasonal and inter-city variability in IR of Ecuadorian populations of Ae . aegypti , specifically testing for susceptibility to insecticides commonly used in mosquito control campaigns by the MoH . In recent work in El Oro province , researchers demonstrated resistance to deltamethrin in phenotypic assays and detected genetic mutations associated with resistance to pyrethroids in Ae . aegypti [26] . Beyond this small-scale study , however , IR monitoring is not regularly performed and research in this area has been limited , meaning that the prevalence of IR in these Ae . aegypti populations is largely unknown . Due to the high morbidity associated with the viruses transmitted by Ae . aegypti , as well as concerns regarding the cost and efficacy of vector control efforts , MoH leaders have indicated that there is a critical need for operational research on IR in Ecuador .
This study was conducted as part of a longitudinal cohort study examining social-ecological correlates of arboviral risk in southern Ecuador . The study protocol was reviewed and approval by Institutional Review Boards ( IRBs ) at SUNY Upstate Medical University ( IRBNET ID 4177710–25 ) , the Luis Vernaza Hospital in Guayaquil , Ecuador , and the Ecuadorean Ministry of Health . Prior to the start of the study , all adult participants ( 18 years of age or older ) engaged in a written informed consent process . Data collection for this project was also conducted with input from local vector control and public health organizations . Sampling coordinates were stored in a secured database and all resulting data were pooled to evaluate citywide trends in the analysis , meaning identifiable individual sites were not shared . Field samples of Ae . aegypti were collected in four cities in El Oro , a dengue endemic province in southern coastal Ecuador ( Fig 1 ) . The cities included in this study range from low to high dengue case burden ( e . g . in 2017 incidence ranged from 0 . 78–16 . 8 per 10 , 000 people ) . In Machala ( 3°15’09”S , 79°57’20”W; 6m elevation , 279 , 887 people ) , a port city and important center of agribusiness , 136 unique sites were visited . In Huaquillas ( 3°28’33”S , 80°13’33”W; 15m elevation , 57 , 366 people ) , a town on the border with Peru , 142 unique sites were visited . Both of these cities are located at sea level along the coast and have endemic transmission of dengue , typically seeing higher numbers of annual reported cases ( Machala 148 , Huaquillas 33 , reported in 2017 ) . In Portovelo ( 3°42’58”S , 79°37’08”W; 645 m elevation , 13 , 673 people ) , 52 unique sites were visited and in Zaruma ( 3°41’31”S , 79°36’47”W; 1 , 155m elevation , 25 , 615 people ) , 42 unique sites were visited . These are both mining towns located further inland and at higher elevations , with Zaruma highest , with limited autochthonous transmission ( Portovelo had 20 cases on average per year from 2014–2017 , and Zaruma 4 ) . The discrepancies in case burden between the cities translate into differential mosquito control demands at the municipal level , making our selection of study sites a diverse backdrop for investigating IR . Collection of field samples was conducted in 2017 over three sampling periods: Season 1 ( February 1 –April 30 ) , Season 2 ( May 1 –June 30 ) , and Season 3 ( July 1 –August 31 ) . Seasons were explicitly chosen to collect mosquito eggs at different phases of annual arbovirus transmission in Ecuador , sampling during the peak ( Season 1 ) , decline ( Season 2 ) , and low transmission ( Season 3 ) . These collection seasons correspond with historical trends in both dengue transmission and mosquito densities in El Oro Province [27 , 28] , and matched the observed dengue transmission pattern during the study ( see S1 Fig , weekly dengue cases for each city in 2017 ) . A longitudinal cohort study across the four cities included household surveys with heads of households , regarding the purchase of insecticides . Surveys were conducted from May to July 2017 . We additionally collected information from the MoH in each city regarding the type , timing , and method of insecticide application over the duration of the present study . Ae . aegypti eggs were collected from households in the four cities using ovitraps lined with oviposition paper [29] . Two or three ovitraps were placed at each trapping site; details on the number of traps by season are shown in S2 Fig . Households were selected from ongoing surveillance study sites where the MoH designated areas of the cities as having high historic risk of dengue . These houses are distributed across each of the four cities to capture geographic variability , and are part of a larger study looking at arboviral and vector dynamics in a cohort of homes over 3 years . Ovitrapped houses for this study were purposely targeted to collect the greatest number of eggs , so ovitraps were strategically placed in areas known to have a greater abundance of eggs based on previous surveillance efforts , representing a subset of the larger household cohort study as well as additional sites nearby . Ovitraps are a sensitive means of identifying the presence of Ae . aegypti , especially in areas with low vector densities [30] . Papers with eggs were collected in the field and transported to the Center for Research on Health in Latin America ( CISeAL ) in Quito , Ecuador , where all insect rearing and handling was performed under standard insectary conditions ( 28 ± 1°C; 80% +/- 10% relative humidity; 12h light , 12h dark photoperiod ) . Egg hatching was achieved by placing the papers in a plastic tray containing distilled water . Larvae were fed finely ground fish flakes . Upon pupation , mosquito specimens were placed in cages for adult emergence . To increase the number of adult specimens available for experimentation , F0 adults were allowed to mate and adult female mosquitoes were blood fed on mice and allowed to oviposit . Collected eggs ( F1 generation ) were hatched and maintained under the aforementioned conditions . Upon pupation , F1 specimens were sorted by sex and females were used for further experimentation . Males were killed by freezing and discarded . We monitored phenotypic resistance to pesticides using the bottle bioassay method as described by Brogdon and Chan [31] . Briefly , we coated glass bottles with ethanolic-insecticide solutions at established diagnostic doses: 10 μg / bottle for both deltamethrin and alpha-cypermethrin ( both pyrethroids ) , and 50 μg / bottle malathion ( an organophosphate ) . Each bioassay replicate consisted of four pesticide-coated bottles and one control bottle , coated only with the ethanol diluent . After the ethanol had evaporated from the bottles , groups of 15–25 female mosquitoes were introduced into each of the four treatment bottles , resulting in a total combined count of mosquitoes in the treatment bottles for each replicate ranging from 62 to 100 . Mortality was recorded at 15-minute intervals . Mosquitoes were considered knocked down when they were incapable of flying or maintaining an upright posture on the surface of the bottle . Mortality counts were recorded after 30 minutes of exposure , following the protocol for diagnostic time for Ae . aegypti given in [31] . This period was chosen to be consistent with CDC protocols used at other sites , and following the specific diagnostic time recommended for South American Ae . aegypti . We did not have a reference susceptible colony to establish specific diagnostic time; we did not record final mortality times . The number of F1 females procured from field-collected specimens allowed us to perform one to three bioassay replicates for each city . The selection of pesticides ( deltamethrin , alpha-cypermethrin and malathion ) corresponds to pesticides typically used for vector control in Ecuador . In addition to the monitoring of phenotypic resistance , we also performed genetic screening for two mutations of the voltage-gated sodium channel gene ( V1016I and F1534C ) that are associated with IR to pyrethroids in Ae . aegypti . Both mutations have been previously described and are reported to alter the transmembrane domain of the voltage-gated sodium channel , preventing pyrethroids from binding to these sites [28 , 29] . Previous genotyping work has shown that Ae . aegypti that are homozygous for the isoleucine variant at position 1016 ( I1016 ) are resistant to pyrethroid treatments , while those that are homozygous for the valine variant ( V1016 ) allele are susceptible , and those that are heterozygous show intermediate resistance [32] . At position 1534 , the cysteine variant ( C1534 ) confers resistance , and the phenylalanine variant ( F1534 ) is susceptible , while the heterozygote typically shows intermediate resistance , though there is some disagreement regarding the extent to which the heterozygous genotype confers resistance , independent of additional resistance mechanisms [32 , 33] . Following the aforementioned bioassays , DNA was extracted from specimens of known phenotype ( susceptible/resistant to each of the pesticides used for testing ) . Genomic DNA was obtained using the Wizard Genomic DNA Purification Kit ( Promega Corporation , Madison , WI , USA ) , following the protocol established by the manufacturer . Quantification of the concentration ( ng/μL ) and purity ( absorbance index at 260nm / 280 nm ) of the obtained DNA was completed using a Nano-Drop 1000 V3 . 7 spectrophotometer ( Thermo Fisher Scientific Inc . , Wilmington , DE , USA ) . Screening for the V1016I and F1534C mutations was performed as previously described [28 , 29] , using a Bio-Rad real-time thermocycler CFX96 ( Bio-Rad , Hercules , CA , USA ) . Results were visualized using CFX Manager Software ( Version 3 . 1 , Bio-Rad , Hercules , CA , USA ) . Determination of the genotype was possible through the visualization of PCR product melting curves . In the case of the V1016I mutation , a melting peak at 79°C corresponded to isoleucine ( I/I–resistant mutant ) and a melting peak at 85°C corresponded to a valine ( V/V–susceptible wild type ) ( S3 Fig ) . In the case of the F1534C mutation , a melting peak at 85°C corresponded to cysteine ( C/C , resistant mutant ) and a melting peak at 80°C corresponded to a phenylalanine ( F/F–susceptible wild type ) ( S4 Fig ) . Genotypic and allelic frequencies were calculated for both mutations . Genotypic frequencies for all four cities were mapped using ArcMap© version 10 . 6 ( Environmental Systems Research Institute , Redlands , California , USA ) . Inter-city and inter-seasonal differences in genotypic frequencies were tested for statistical significance ( α = 0 . 05 ) using Fisher’s exact test of independence in the R software for statistical computing , version 3 . 4 . 3 ( The Foundation for Statistical Computing , Vienna , Austria ) . When significant differences were detected , post-hoc pairwise Fisher’s tests with Bonferroni corrections were then performed to identify which cities and seasons differed significantly . The same statistical methods were used to examine phenotypic frequencies of resistance , as determined by bottle bioassays for deltamethrin , alpha-cypermethrin , and malathion , between cities for each collection season and in relation to corresponding genotypes . Additionally , populations for each city and season were assessed for Hardy-Weinberg equilibrium using the exact test in the HardyWeinberg R package [34] .
In Machala , 27 . 5% ( 14/51 ) surveyed homes reported purchasing pyrethroid insecticides to use at home , while 18 . 9% ( 10/53 ) of surveyed homes in Huaquillas reported these purchases . Of households surveyed in Portovelo and Zaruma , 45 . 6% ( 22/48 ) and 36 . 5% ( 19/52 ) respectively , reported purchase of pyrethroid insecticides for home use . Ministry of Health insecticide applications in Machala utilized a combination of deltamethrin , alpha-cypermethrin , and malathion over the duration of the study . Both Portovelo and Zaruma applied deltamethrin and malathion during the peak transmission season , followed by deltamethrin only for the following months . Huaquillas included malathion and an additional ( unidentified ) product during peak transmission season , followed by deltamethrin and an additional ( unidentified ) product in the following months ( S5 Fig ) . High levels of IR to deltamethrin , alpha-cypermethrin , and malathion were detected in all four cities , as determined by bottle bioassays . Machala had the lowest combined mortality averaged across the three seasons ( 18 . 29% , SE±5 . 75 ) , indicating the highest level of resistance , followed by Portovelo ( 31 . 08% , SE±6 . 23 ) and Huaquillas ( 44 . 79% , SE±10 . 87 ) . The mean mortality rate for Zaruma was 23 . 67% ( SE±17 . 93 ) in Season 1 , which was the only time period in which sufficient numbers of Ae . aegypti were collected for bioassays at this location . The proportion of mosquitoes that were resistant to deltamethrin differed significantly across cities in all three collection seasons ( Fisher’s exact test P-value <0 . 001 ) ( Fig 2A ) . This pattern was also seen in the post-hoc pairwise tests ( S1 Table ) . Mean mosquito mortality after 30 minutes of exposure to deltamethrin in bottle bioassays ranged from 0 . 70% ( SE±0 . 6 ) in Machala to 18 . 99% ( SE±0 ) in Huaquillas in Season 1 , from 7 . 09% ( SE±4 . 19 ) in Portovelo to 16 . 88% ( SE±7 . 51 ) in Huaquillas in Season 2 , and from 0 . 37% ( SE±0 . 37 ) in Machala to 56% ( SE±2 . 49 ) in Huaquillas in Season 3 , with no mortality observed in control groups ( Fig 2A ) . Mean mosquito mortality after 30 minutes of exposure to alpha-cypermethrin across bottle bioassay replicates ranged from 59 . 24% ( SE±13 . 86 ) in Zaruma to 59 . 41 ( SE±1 . 49 ) in Machala in Season 1 , from 12 . 1% ( SE±0 . 42 ) in Machala to 17 . 27 ( SE±0 . 33 ) in Portovelo in Season 2 , and from 0 . 38% ( SE±0 . 38 ) in Machala to 11 . 39 ( SE±1 . 03 ) in Portovelo in Season 3 , with no mortality observed in control groups ( Fig 2B ) . Significant differences in mean mortality rates between cities were detected in seasons two and three ( Fisher’s exact test P value <0 . 001; Fig 2B ) , with Machala having the highest proportion of phenotypically resistant specimens in these two seasons ( S2 Table ) . Collection counts were insufficient for alpha-cypermethrin bioassays in Portovelo for Season 1 , Huaquillas for Seasons 1 and 3 , and Zaruma for Seasons 2 and 3 . Mean mosquito mortality after 30 minutes of exposure to malathion in the bottle bioassays ranged from 5 . 07% ( SE±0 . 35 ) in Machala to 3 . 03% ( SE±1 . 52 ) in Zaruma in Season 1 , from no mortality in Machala to 1 . 45% ( SE±1 . 45 ) in Portovelo in Season 2 , and from 0 . 78% ( SE±0 . 78 ) in Machala to 2 . 86% ( SE±0 . 08 ) in Portovelo in Season 3 , with no mortality observed in control groups . Statistically significant differences in Malathion resistance were only detected in season two between Machala and Portovelo ( Fisher’s exact test P value <0 . 001 ) . Due to low sample sizes , bioassays could not be conducted in Portovelo for Season 1 , Zaruma for Seasons 2 and 3 , and Huaquillas for all three seasons , and therefore further comparisons were not feasible . Inter-seasonal variations in phenotypic resistance for each city were also assessed for statistical significance . In the deltamethrin bottle bioassays , mean mortality in Machala increased from Season 1 to Season 2 ( Post-hoc Fisher’s exact test P value <0 . 001 ) , but decreased from Season 2 to Season 3 ( Post-hoc Fisher’s exact test P value <0 . 001 ) . Mean mortality increased from Season 1 to Season 2 ( Post-hoc Fisher’s exact test P value = 0 . 003 ) and Season 2 to Season 3 ( Post-hoc Fisher’s exact test P value <0 . 001 ) in Huaquillas . Mean mortality increased in Portovelo from Season 1 to Season 2 ( Fisher’s exact test P value = 0 . 005 ) and from Season 2 to Season 3 ( Fisher’s exact test P value = 0 . 02; S3 Table ) . In the alpha-cypermethrin bottle bioassays , significant inter-seasonal differences were only detected in Machala , where the percent mortality decreased from Season 1 to Season 2 ( Post-hoc Fisher’s exact test P value < 0 . 001 ) and from Season 2 to Season 3 ( Post-hoc Fisher’s exact test P value = 0 . 05; S4 Table ) . Similarly , significant inter-seasonal differences in mean mortality in the malathion treatment were only detected in Machala , where mortality decreased from Season 1 to Season 2 ( Post-hoc Fisher’s exact test P value = 0 . 05; S5 Table ) . Observed genotype frequencies varied significantly between cities only in Season 1 for both V1016I ( Fisher’s exact test P value <0 . 001 ) and F1534C alleles ( Fisher’s exact test P value <0 . 001; Fig 3 ) . Pairwise post-hoc analysis revealed significant differences in the frequencies of genotypes I/I ( mutant ) and V/I ( heterozygous ) for the V1061I gene , with Huaquillas ( n = 34 ) having a significantly higher frequency of heterozygotes compared to Machala ( n = 22; Post-hoc Fisher’s exact test P value <0 . 001 ) and Portovelo ( n = 22; Post-hoc Fisher’s exact test P value = 0 . 03; S6 Table ) . Portovelo and Zaruma ( n = 40 ) differed in the frequency of I/I ( mutant ) and V/V ( wild type ) genotypes , with Zaruma having a significantly higher frequency of wild type mosquitoes during Season 1 ( Post-hoc Fisher’s exact test P value = 0 . 05; S6 Table ) . Genotypic frequencies of C/C ( mutant ) and F/C ( heterozygous ) genotypes of the F1534C resistance gene differed across cities in Season 1 , with Zaruma having a significantly higher proportion of the heterozygous genotype than Huaquillas and Machala ( Post-hoc Fisher’s exact test P values = 0 . 02 and <0 . 001 , respectively; S7 Table ) . Although the frequencies of F1534C genotypes also varied significantly in the Season 3 ( Fisher’s exact test P value = 0 . 03 ) , conservative post-hoc analysis did not reveal any significant pairwise relationships ( S7 Table ) . Significant inter-seasonal variation in V1016I genotype frequencies was detected in Huaquillas ( Fisher’s exact test P value = 0 . 04 ) , Machala ( P value <0 . 001 ) , and Portovelo ( P value <0 . 001 ) , though conservative post-hoc analysis did not identify significant pairwise differences in Huaquillas . In Machala , the proportion of the I/I ( mutant ) and V/I ( heterozgote ) genotypes differed significantly from Season 1 to Season 3 ( Post-hoc Fisher’s exact test P value <0 . 001 ) , with the I/V genotype increasing while the I/I genotype decreased ( S8 Table ) . In Portovelo , there were significant differences in the frequencies of the I/I and V/I genotype between Seasons 1 and Season 2 and between Season 1 and Season 3 ( Post-hoc Fisher’s exact test P values = 0 . 05 and 0 . 001 , respectively ) , with the frequency of the I/I genotype decreasing while V/I increased . Similarly , the frequency of the V/V ( wild type ) genotype increased significantly from Season 1 to Season 3 ( Post-hoc Fisher’s exact test P value = 0 . 005 ) while the I/I genotype decreased ( S9 Table ) . No significant inter-seasonal differences in F1534C genotype frequency were detected . There were statistically significant associations between the resistance genotypes and phenotypic resistance results for 632 Ae . aegypti that were genotyped and subjected to the pyrethroid ( deltamethrin or alpha-cypermethrin ) bottle bioassays . Due to low sample sizes , we pooled pyrethroid assay results ( Tables 1 & 2 ) , and present the separated analyses in supplemental information ( S10 and S11 Tables ) . For mutation V1016I , while the majority of the resistant individuals ( 96 . 4% ) presented the mutant or heterozygous genotype , 12 ( 3 . 6% ) phenotypically resistant individuals presented the wild type ( V/V ) genotype that is typically associated with susceptibility to pyrethroids , and 40 ( 13 . 5% ) susceptible individuals presented the genotype that typically confers resistance . Similarly , at the F1534C locus , 276 ( 93 . 2% ) individuals with the mutant ( C/C ) genotype , typically associated with resistance , were susceptible to pyrethroid treatments . Due to low sample sizes of the wild type ( F/F ) for both pyrethroids , separate Fisher’s exact tests were not conducted ( S6 Table ) . Because the two loci studied here contribute additively to resistance , the proportions of resistant and susceptible individuals for each combined V1016I and F1534C genotypes were also considered . Of the 227 mosquitoes with the mutant I/I and the mutant C/C genotype , 187 ( 82 . 4% ) were resistant in the pyrethroid assay . There were twelve genotyped mosquitoes that were heterozygous at both loci , and of these , six individuals were resistant; likewise , only three genotyped specimen were homozygous wild type at both loci , and all three were susceptible ( Tables 1 & 2 ) . The V1016I genotype frequencies indicate that populations from Huaquillas ( Fisher’s exact test P value = 0 . 03 ) and Machala during Season 1 ( Fisher’s exact test P value = 0 . 03 ) and Portovelo during Season 3 ( Fisher’s exact test P value = 0 . 02 ) were not in Hardy-Weinberg equilibrium ( S12 Table ) .
In this study , exposure to diagnostic doses of deltamethrin , alpha-cypermethrin and malathion resulted in mortality rates below 80% after 30 minutes of insecticide exposure in all populations tested , regardless of collection season . Based on the World Health Organization´s ( WHO ) recommendations for assessing the significance of detected resistance [31] , our results suggest that these Ae . aegypti populations should be considered as resistant to all the insecticides considered in our study . Furthermore , the results of the bioassays for malathion susceptibility are of particular interest , as they indicate that these populations are extremely resistant to malathion , with no population showing more than 5 . 07% mortality , and populations from Machala reaching values as low as 0% mortality during Season 2 . The CDC bottle bioassay used in this study is one of several methods used to detect resistance in mosquito populations . The other most commonly used method , the WHO susceptibility test , is a response-to-exposure analysis that uses insecticide-impregnated papers obtained directly from a distribution center [34] . While this pre-manufactured quality means there is greater consistency and control in the administration of the WHO susceptibility test , the CDC bottle bioassay can be conducted without specialized equipment , which often makes it the assay of choice in resource-limited settings [35] . In addition to these methods , IR can be assessed based on calculation of the lethal concentration needed to kill half of a sample of mosquitoes ( LC50 ) after topical application of the insecticide of interest [36] . This method allows for the calculation of resistance ratios to quantify and compare resistance across populations; however , this test is more time-consuming and requires specialized equipment , so it is not as commonly conducted [37] . In this study , molecular characterization showed that the resistance-associated mutant alleles V1016I and F1534C are present at all the locations studied . In particular , allele F1534C was present at very high frequencies ( close to 1 ) in all the locations studied and across all three seasons , similar to results from recent work in Mexico [35] . This suggests that this gene has been subjected to selective pressures in the past and is approaching or has reached fixation in these populations . By contrast , the results from the test for Hardy-Weinberg equilibrium for the V1016I gene and the significant inter-seasonal differences in genotype frequencies indicate that the populations are still responding to varying selective pressures and are not in a state of equilibrium . Both of these mutations typically confer resistance to dichlorodiphenyltrichloroethane ( DDT ) as well as pyrethroids [36 , 37] , meaning selection for these alleles likely began with earlier widespread usage of DDT ( which was used by the MoH until 1996 [38] ) and has persisted as pyrethroids became more commonly used [39] . Significant distinctions in genotypic and phenotypic frequencies were not detected for many cities in this study during subsequent seasons . It is worth noting that for some locations and seasons , periods of low vector densities in the field resulted in the collection of low numbers of mosquito eggs . Overall , collection counts were consistently low in Zaruma , a small city that has the highest elevation of the four study cities . This observation is consistent with other work that has documented a negative relationship between elevation and the probability of Ae . aegypti presence [40] . Similarly , collection counts during Season 2 and Season 3 were not high enough for bioassays with each of the three insecticides of interest . These seasons have historically corresponded with periods of low mosquito activity and dengue transmission in Ecuador , as observed in 2017 [27 , 28] This scarcity of F0 individuals translated into missing data for some cities and/or very low frequency counts in some categories , thus providing a limited basis for making quantitative comparisons . The detection of resistance to organophosphate and pyrethroid insecticides in Ecuador is consistent with broader regional trends identified in recent years . In neighboring Peru , Ae . aegypti strains have been shown to be resistant to multiple organophosphates and pyrethroids in WHO susceptibility tests [41] . Similarly , pyrethroid resistance has been detected in Colombia in both CDC and WHO bioassays , though the levels of resistance vary throughout the country [22] . However , Colombian Ae . aegypti populations tested with both WHO and CDC bioassays were broadly susceptible to malathion , in contrast to the widespread resistance seen in the Ecuadorian Ae . aegypti in this study [42] . Throughout the rest of the Americas , pyrethroid resistance , as measured in studies using LC50 or percent mortality , appears to be broadly distributed , particularly in Brazil and French Guiana , though some populations in Costa Rica , Panama , and northern Colombia are still susceptible [10] . The temporal and spatial variability in the results from the bioassays highlight the importance of regularly conducting IR monitoring across multiple locations to understand the true extent of IR and make appropriate vector control decisions . For example , in Machala , the mortality rate of Ae . aegypti treated with alpha-cypermethrin decreased significantly from Season 1 to Season 3 , indicating this population was becoming less susceptible throughout the course of the study . By contrast , the mortality rate associated with the deltamethrin assays on populations from Huaquillas increased from Season 1 to Season 3 , meaning this population was becoming more susceptible over time . The impact of city-level insecticide application on IR is difficult to infer with the information we obtained from the MoH of each municipality . Portovelo did not use alpha-cypermethrin and we saw no differences in alpha-cypermethrin mortality rates across seasons . Machala used some form of deltamethrin , alpha-cypermethrin , and malathion throughout the study duration , but mortality rates differed for these insecticides by season , with mortality rates decreasing for both alpha-cypermethrin and malathion . There was some seasonal variation in method of application , strength of insecticide solution , and neighborhood coverage , which could impact the IR of our sampled mosquito populations . Continued work in this area could determine if our observed trends are due to seasonal fluctuations , differential insecticide application parameters , local-scale movements of Ae . aegypti populations with varying levels of resistance , or long-term , inter-annual trends . There were also statistically significant differences in mortality rates across the cities for the deltamethrin bioassays in all three seasons and in two of the three seasons for the alpha-cypermethrin bioassays . Considering this variability in the resistance phenotypes found within a single province of Ecuador , organizations involved in decision-making about insecticide applications should be cautioned against inferring the IR status of one Ae . aegypti population based on the status of populations in neighboring municipalities . To better contextualize this work for appropriate vector-control decision-making , the relationships between genotypes , IR bioassay results , and actual IR status in the field should be considered . While the V1016I and F1534C mutations are known markers of pyrethroid resistance in Ae . aegypti , the results of this study showed that the genotypes were not perfectly predictive of resistance phenotypes , even when both the V1016I and F1534C genotypes were considered . In the recorded frequencies of genotypes versus bioassay outcomes , resistant and susceptible phenotypes were observed for each genotype , although the resistant phenotype was still statistically associated with the mutant genotypes for both genes . This is likely due to other IR mechanisms , such as metabolic detoxification processes [19] that could influence IR status in these populations; however , these mechanisms were not considered in the current study . Additionally , factors such as temperature , larval nutrition , larval density and age have been shown to influence insecticide susceptibility in Aedes mosquitoes , leading to discrepancies between bioassay results and the actual outcomes of insecticide treatments in the field [43] . Further work on IR in Aedes populations could identify and possibly reconcile differences between results from the laboratory and the field . To comprehensively evaluate IR status , future studies should also investigate the effectiveness of insecticide application methods , intensity , timing , and coverage by households and the MoH , as well as the impact of larvicides , such as temephos , which is commonly used in this study area , as well as Bacillus thuringensis israelensis ( Bti ) , which is a common control method in other parts of the world . Geographic methods , particularly spatial statistics and modelling , are well suited for understanding the patterns and drivers of IR at meso- and local scales . Employing these approaches can lead to more targeted , efficient , and sustainable vector control efforts . Future research in this area should continue to explore the spatial and temporal variability in IR among Ae . aegypti populations . Recent work in Yucatan , Mexico , demonstrated that IR levels could vary significantly across neighborhoods within the same city [24] . Additionally , work on the temporal dynamics of resistance could be beneficial for vector control decision-making . For example , in a study on pyrethroid-resistant , field-derived Ae . aegypti , researchers demonstrated that susceptibility to pyrethroids could be restored within ten generations when the selective pressure of regular insecticide treatments was removed [25] . While this experiment was conducted in a controlled environment , similar work within the context of urban environments , such as the four cities included in this study , could help calibrate timing of insecticide class rotations , allowing for better long-term management of susceptibility . In conclusion , the Ae . aegypti collected in these four cities in Ecuador showed varying levels of resistance to the insecticides tested , and these measures typically changed over the course of the three seasons during which sampling took place . Regular IR monitoring should be conducted as long as insecticide applications remain an integral component of vector control activities , particularly in areas where these operations are deployed to control arbovirus transmission . Beyond this monitoring process , appropriate alternative management strategies should be deployed when IR is detected . These strategies can include biological control and community mobilization to reduce Ae . aegypti breeding sites .
|
Mosquito control is the primary method of managing the spread of many diseases transmitted by the yellow fever mosquito ( Aedes aegypti ) . Throughout much of Latin America the transmission of diseases like dengue fever and Zika fever pose a serious risk to public health . The rise of insecticide resistance ( IR ) is a major threat to established vector control programs , which may fail if commonly used insecticides are rendered ineffective . Public health authorities in southern coastal Ecuador , a high-risk region for diseases vectored by Ae . aegypti , previously had limited information on the status of IR in local populations of mosquitoes . Here , we present the first assessment of IR in adult Ae . aegypti to insecticides ( deltamethrin , Malathion , and alpha-cypermethrin ) routinely used in public health vector control in four cities along Ecuador’s southern coast . Observed patterns of IR differed between cities and seasons of mosquito sampling , suggesting that IR status may fluctuate in space and time . The highest overall resistance was detected in Machala , a city with hyperendemic dengue transmission and a long history of intense insecticide use . Monitoring for IR is an integral component of vector control services , where alternative management strategies are deployed when IR is detected .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"death",
"rates",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"ecuador",
"chemical",
"compounds",
"variant",
"genotypes",
"geographical",
"locations",
"malathion",
"organic",
"compounds",
"animals",
"genetic",
"mapping",
"organophosphates",
"population",
"biology",
"infectious",
"disease",
"control",
"insect",
"vectors",
"public",
"and",
"occupational",
"health",
"agrochemicals",
"infectious",
"diseases",
"south",
"america",
"chemistry",
"disease",
"vectors",
"insects",
"agriculture",
"arthropoda",
"insecticides",
"people",
"and",
"places",
"population",
"metrics",
"mosquitoes",
"eukaryota",
"organic",
"chemistry",
"ethers",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"species",
"interactions",
"organisms"
] |
2019
|
Seasonal and geographic variation in insecticide resistance in Aedes aegypti in southern Ecuador
|
Axons require a constant supply of the labile axon survival factor Nmnat2 from their cell bodies to avoid spontaneous axon degeneration . Here we investigate the mechanism of fast axonal transport of Nmnat2 and its site of action for axon maintenance . Using dual-colour live-cell imaging of axonal transport in SCG primary culture neurons , we find that Nmnat2 is bidirectionally trafficked in axons together with markers of the trans-Golgi network and synaptic vesicles . In contrast , there is little co-migration with mitochondria , lysosomes , and active zone precursor vesicles . Residues encoded by the small , centrally located exon 6 are necessary and sufficient for stable membrane association and vesicular axonal transport of Nmnat2 . Within this sequence , a double cysteine palmitoylation motif shared with GAP43 and surrounding basic residues are all required for efficient palmitoylation and stable association with axonal transport vesicles . Interestingly , however , disrupting this membrane association increases the ability of axonally localized Nmnat2 to preserve transected neurites in primary culture , while re-targeting the strongly protective cytosolic mutants back to membranes abolishes this increase . Larger deletions within the central domain including exon 6 further enhance Nmnat2 axon protective capacity to levels that exceed that of the slow Wallerian degeneration protein , WldS . The mechanism underlying the increase in axon protection appears to involve an increased half-life of the cytosolic forms , suggesting a role for palmitoylation and membrane attachment in Nmnat2 turnover . We conclude that Nmnat2 activity supports axon survival through a site of action distinct from Nmnat2 transport vesicles and that protein stability , a key determinant of axon protection , is enhanced by mutations that disrupt palmitoylation and dissociate Nmnat2 from these vesicles .
The chimeric fusion protein WldS ( Entrez Gene ID 22406 ) affords robust protection of injured axons in vitro and in vivo [1] , [2] and extends axon survival in several disease models [3]–[9] . The WldS protein incorporates full-length Nmnat1 ( Entrez Gene ID 66454 ) , an NAD synthesising enzyme whose enzymatic activity is necessary for the protective effect of WldS [10] , [11] . Additionally WldS harbours an N-terminal region that causes its partial re-distribution from the nucleus to an axoplasmic localization that is necessary for axon protection in vivo [1] , [10]–[12] . We recently identified the related NAD-synthetic enzyme Nmnat2 ( Entrez Gene ID 226518 ) as a labile , endogenous axon survival factor whose constant supply from cell bodies into axons is required for axon survival in primary culture . Specific depletion of Nmnat2 causes neurite degeneration without injury . After injury , endogenous Nmnat2 is depleted rapidly in the distal stump of neurites , initiating the process of Wallerian degeneration . If the more long-lived WldS protein is present in the axon , Nmnat2 is still depleted at the same rate; however , the NAD-synthetic enzyme activity of the stable WldS protein substitutes for that of Nmnat2 , resulting in a significant delay of Wallerian degeneration [13] . Supporting a role of Nmnat2 in axon survival , strong overexpression of Nmnat2 delays Wallerian degeneration in vitro , and this protective effect is dependent on its enzymatic activity [13] , [14] . Furthermore , Nmnat2 overexpression delays injury-induced axon degeneration in zebrafish [15] and alleviates neurodegeneration in the P301L mouse model of tauopathy [16] . Spontaneous synapse and axon loss in Drosophila lacking dNmnat , which can be partially rescued by murine Nmnat2 , also supports a key role for endogenous axonal Nmnat activity in axon survival [17] . Neurons are extremely polarized cells with processes extending up to centimetres or even meters beyond the cell body . Moreover , in some neurons the axon constitutes over 99% of total cytoplasmic volume [18] . Despite growing evidence for the local , axonal synthesis and regulation of some proteins [19]–[21] , many others appear to be synthesized only in the cell body and rely on axonal transport to reach their site of action in the axon or synapse . This supply process is a huge logistical challenge , and not surprisingly , any impairment affects axonal function or survival . Indeed , recent work has illustrated that there is a significant , early impairment in axonal transport in many neurodegenerative conditions and , for at least some of these , impaired axonal transport appears to cause the degenerative process [22] . Given that Nmnat2 is essential for axon maintenance [13] and its mRNA has not been found in axons [23]–[26] , the delivery of this short-lived protein into axons is likely to limit axon survival when axonal transport is impaired through injury , aging , or disease . Although the WldS protein can overcome the need for Nmnat2 , its clinical application faces the problem that this gain-of-function chimeric fusion protein is expressed only in WldS mice and a few strains of transgenic organisms . Instead , understanding and manipulating the delivery , turnover , or intra-axonal targeting of the endogenous survival factor , Nmnat2 , is a more promising route to influence axon survival . Thus , it is important to understand how healthy neurons ensure a steady supply of Nmnat2 and the mechanisms that regulate its delivery , turnover , and activity in axons . Nmnat2 localizes to vesicular structures and undergoes fast axonal transport in the neurites of primary culture neurons [13] , [27] . Previous work showed that association of Nmnat2 with Golgi membranes in HeLa cells requires palmitoylation of the two adjacent cysteine residues C164/165 . In the absence of these residues , Nmnat2 adopts a more diffuse , cytosolic localization [27] , [28] . This palmitoylation site is located within the isoform-specific targeting and interaction domain ( ISTID ) of Nmnat2 . ISTID regions are found in all three mammalian Nmnat isoforms and , in contrast to the more conserved core catalytic domains that make up the remainder of the proteins , are highly divergent between isoforms and are thus thought to account for the differential subcellular localizations of the Nmnats [28] . Here we report that the small , centrally located exon 6 is both necessary and sufficient for palmitoylation , stable membrane association , and vesicle-mediated delivery of Nmnat2 into axons . By manipulating its localization , we then test the hypotheses that these transport vesicles are the sites of Nmnat2 axon-protective action and that vesicular Nmnat2 is somehow protected from rapid turnover , enabling it to reach the ends of long axons before being degraded . Surprisingly , we find that a diffuse , nonvesicular localization enhances Nmnat2 axon protection through increased protein stability . Our results support a model in which Nmnat2 subcellular localization regulates its turnover and protective capacity and suggest a site of action distinct from its transport vesicles .
Previously , we reported that , in primary culture neurons , Nmnat2-EGFP is transported in particulate structures with an anterograde bias and at velocities in the range of fast axonal transport [13] . To shed light on the identity of this trafficking organelle , we utilized dual-colour , live-cell imaging of primary culture superior cervical ganglia ( SCG ) neurons to visualize the co-transport of Nmnat2 with established axonal transport cargos and organelle markers . Co-migration of two fluorescent markers was quantified using kymographs ( see Materials and Methods ) , counting only moving particles . We detected significant co-migration of Nmnat2 with several Golgi network markers ( Golga2 , Syntaxin6 , TGN38; Figure 1D–F , L ) . Similarly , high levels of co-migration were observed for synaptic vesicle markers ( SNAP25 , Synaptophysin , SynaptotagminI; Figure 1I–K , L ) . Interestingly , however , no significant co-migration was found for mitochondria ( Figure 1A , L ) , lysosomes ( lamp1; Figure 1B , L ) , or active zone precursor vesicles ( Bassoon; Figure 1G , L ) , and only partial co-migration was detected for a marker of the ER-Golgi intermediate compartment ( ERGIC ) ( ERGIC53/LmanI; Figure 1C , L ) . Thus , Nmnat2 undergoes fast axonal transport in a Golgi-derived vesicle population that overlaps with synaptic vesicle precursors and is distinct from mitochondria , lysosomes , active zone precursor vesicles , and the ERGIC . Next , we tested whether the reported mechanism of Golgi-targeting in HeLa cells [27] , [28] applies to neuronal cell bodies and vesicle targeting in axons . In order to define the requirements for membrane association in neurons , we used a photoactivation assay in SCG cell bodies . Photoactivatable GFP ( PA_GFP [29] ) was fused to variant Nmnat2 sequences and microinjected along with an mCherry marker to identify injected cells . PA_GFP was activated in a small region of the cell body and the pool of activated PA_GFP followed over time . For quantification , we compared the fraction of fluorescence intensity that remained in the activated area with a non activated area elsewhere in the cell body ( excluding the nucleus ) . Activated PA_GFP alone diffused very rapidly throughout the cell body , resulting in an even distribution of fluorescence after about 5–10 s ( Figure 2 ) . In contrast , Nmnat2-PA_GFP was retained within the originally activated area , suggesting strong membrane association that was stable over the course of the experiment ( Figure 2; Table 1 ) . This indicates that the majority of Nmnat2 is stably membrane-bound and most of the spread of fluorescence that did occur was slow and resulted from transport of vesicle-bound Nmnat2-PA_GFP out of the activated region ( see Figure 2A ) . The C164/165 palmitoylation site is located at the centre of the 27 amino acids encoded by exon 6 of Nmnat2 ( see Figure S1A for Nmnat2 primary structure with relevant regions highlighted ) . To test whether this exon is sufficient for membrane targeting , we created an exon6-PA_GFP construct and subjected it to the membrane association assay described above . Interestingly , we observed a very strong membrane association that was indistinguishable from that seen with full-length Nmnat2-PA_GFP ( Figure 2; Table 1 ) . This same sequence also targeted EGFP to transport vesicles in neurites that co-migrated with full-length Nmnat2-mCherry . The degree of co-migration was similar to that between Nmnat2-EGFP and Nmnat2-mCherry ( Figure 3B , C , and H ) . Together , these results indicate that exon 6 encodes residues sufficient for efficient , stable membrane association and the resulting vesicular fast axonal transport of Nmnat2 . We then confirmed the requirement for the C164/165 palmitoylation site for membrane association in neurons , using a C164S/C165S construct ( Nmnat2ΔPS-PA_GFP; see Figure S1B for an overview of all mutant constructs used in this study ) . Photoactivation in SCG cell bodies resulted in a rapid spread of fluorescence , similar to PA_GFP alone ( Figure 2 ) , although quantification revealed that the spread was slightly slower than for PA_GFP , even when accounting for differences in molecular size ( see k values in Table 1 ) [30] , suggesting that other residues contribute weakly to membrane association . GAP43 ( Entrez Gene ID 14432 ) , which associates with membranes through palmitoylation of a similar double-cysteine motif , also requires an adjacent group of basic residues for efficient and stable membrane association [31]–[33] and shows partial co-migration with Nmnat2 ( Figure 1H , L ) . We tested whether a similar mechanism applies to Nmnat2 by mutating the five basic residues encoded by exon 6 ( K151A , K155A , R162A , R167A , and R172A ) . This construct , Nmnat2ΔBR-PA_GFP , showed intermediate membrane association . A pool of diffusible material spread quickly throughout the cell ( as seen with PA_GFP ) , but a significant portion of the signal remained in the originally activated area for the duration of the experiment , suggesting the presence of a pool of strongly membrane-bound material ( as seen with Nmnat2-PA_GFP and exon6-PA_GFP ) ( Figure 2 ) . These results suggest that , in addition to the palmitoylated cysteines themselves , basic residues that surround C164/165 are also required for efficient palmitoylation and membrane association . However , the mobility of the diffusible portion of Nmnat2ΔBR-PA_GFP was not significantly different from that of Nmnat2ΔPS-PA_GFP ( Figure 2; Table 1 ) . Accordingly , a double mutant , Nmnat2ΔPSΔBR-PA_GFP , also showed the same mobility as either of the single mutants , thus still exhibiting slower diffusion than PA_GFP alone . To confirm that these changes in membrane association reflect the degree of palmitoylation of Nmnat2 , we used radiolabelling to measure palmitate incorporation into wild-type and mutant Nmnat2 ( Figure S2 ) . In agreement with previous findings [27] , [28] we found that FLAG-Nmnat2ΔPS loses all detectable palmitoylation . Interestingly , however , a small but significant portion of palmitate incorporation was maintained in FLAG-Nmnat2ΔBR as predicted by the membrane association assay , further supporting the idea that exon 6 basic residues are necessary to enable efficient palmitoylation and only a small amount of palmitoylation can occur in their absence . To further substantiate the role of palmitoylation in Nmnat2 membrane association , we treated SCG neurons with 2-Bromopalmitate ( 2-BP ) , a lipid-based inhibitor of palmitoylation . As predicted , treatment with 2-BP substantially reduced membrane association of wild-type Nmnat2-PA_GFP in the photoactivation assay ( Figure S6A , B ) . Next , we sought to test the effect of exon 6 mutations on axonal transport of Nmnat2 . As expected , mutation of the palmitoylation site in Nmnat2ΔPS-EGFP led to a diffuse , nonvesicular distribution in neurites . We detected little co-migration with Nmnat2-mCherry ( Figure 3E ) , which was not significantly different from EGFP alone ( Figure 3A , H ) . Like other cytosolic proteins , nonspecific or transient membrane association may help deliver this protein to neurites in this system [34] . For Nmnat2ΔBR-EGFP , the amount of diffuse , nonvesicular fluorescence signal was also greatly increased , but we still observed significant co-migration with Nmnat2-mCherry ( Figure 3D , H ) , consistent with the residual palmitoylated , membrane-bound component inferred from the photoactivation and palmitate labelling experiments . Nmnat2ΔPSΔBR-EGFP and Nmnat2Δex6-EGFP ( lacking all of exon 6 ) were both similar to Nmnat2ΔPS-EGFP with no further reduction in co-migration with Nmnat2-mCherry ( Figure 3F–H ) . Taken together , these results suggest that , within exon 6 , both the palmitoylation site and surrounding basic residues are necessary for efficient , stable membrane association , and vesicular axonal transport of Nmnat2 . We next tested the hypothesis that vesicle targeting is necessary for Nmnat2-mediated axon protection . We injected wild-type or variant Nmnat2-EGFP together with a DsRed2 fluorescent marker into SCG cell bodies and transected their neurites 48 h later . Due to its short half-life , Nmnat2-EGFP protects neurites for 24 h after transection only if strongly overexpressed . Surprisingly , however , both Nmnat2ΔPS-EGFP and Nmnat2ΔBR-EGFP preserved transected neurites significantly more strongly when only 0 . 001 µg/µl of DNA was injected ( Figure 4A , C ) . Interestingly , the protective effects of these two mutations were additive . At 0 . 002 µg/µl , Nmnat2ΔPSΔBR-EGFP showed strongly preserved neurites up to 72 h , significantly more than either single mutant ( Figure 4B , D ) . To rule out any influence of the EGFP tag on these results , we also found that untagged Nmnat2ΔPSΔBR protects neurites significantly better than untagged wild-type Nmnat2 ( 0 . 01 µg/µl; Figure S3 ) . Thus , vesicle association is dispensable for Nmnat2-mediated neurite protection in primary culture , and missense mutations disrupting vesicle association boost Nmnat2 axon protective capacity to a modest but significant degree . These surprising findings prompted us to investigate the effects on axon protection of deletions within and around exon 6 . In particular , three deletion mutants were recently reported to retain enzyme activity ( Nmnat2Δ32-EGFP , Nmnat2Δ43-EGFP , Nmnat2Δ69-EGFP [35] ) . Remarkably , all these deletion mutants and a mutant lacking exon 6 only ( Nmnat2Δex6-EGFP ) protected neurites far more strongly than any of the missense constructs above . Even microinjection of very low DNA concentrations ( 0 . 0005 µg/µl ) preserved around 80% of neurites for 72 h after transection ( Figure 5 ) , compared with less than 10% for Nmnat2-EGFP or Nmnat2ΔPSΔBR-EGFP at this concentration . Surprisingly , this even exceeds the level of protection achieved by WldS-EGFP at this concentration ( around 30% intact neurites at 72 h ) , illustrating the very strong enhancement of axon protective capacity in these mutants ( Figure 5 ) . The presence of the EGFP tag was not necessary for the observed increase in protection ( Figure S4 ) . To rule out the possibility that deletion of exon 6 induces a novel gain-of-function in Nmnat2 that is independent of its NAD-synthesis activity , we introduced an enzyme-dead mutation into Nmnat2Δex6 . His24 is conserved in all three mammalian Nmnats and is critical for Nmnat2 NAD-synthesis activity as well as its ability to protect axons after cut [14] , [36] , [37] . Convincingly , a Nmnat2Δex6H24D enzyme-dead mutant did not protect neurites after injury ( Figure S5 ) , although we found that this mutation also reduces the stability of Nmnat2Δex6H24D ( unpublished data ) . Together , these findings suggest that , in addition to the palmitoylation site and surrounding basic residues , other exon 6 residues influence the axon protective capacity of Nmnat2 , while the additional sequences deleted outside of exon 6 in Nmnat2Δ32 , Nmnat2Δ43 , and Nmnat2Δ69 appear to have little further effect . Next , we sought to identify the mechanism by which these mutations increase axon protective capacity . As the short half-life of Nmnat2 limits survival of injured axons [13] , we decided to investigate protein stability using an emetine chase assay . HEK293 cells transiently expressing FLAG-Nmnat2 or one of its mutant forms were treated with 10 µM emetine to inhibit protein synthesis . Nmnat2 turnover was then measured as the rate of decline of the FLAG-Nmnat2 signal over time relative to a more stable control ( FLAG-WldS ) [13] . We had envisaged that vesicular Nmnat2 may be relatively stable , allowing it to reach the ends of long axons , and in dynamic equilibrium with a less stable , but perhaps more active cytosolic form . However , all mutations disrupting membrane targeting were found to increase protein stability ( Figure 6A , B; Table 2 ) . Moreover , the FLAG-Nmnat2ΔPSΔBR double mutant showed significantly higher protein stability than either FLAG-Nmnat2ΔPS or FLAG-Nmnat2ΔBR , supporting a model in which an increase in protein stability contributes to the increase in protective capacity observed for these missense mutations . Intriguingly , however , the very strongly protective Nmnat2Δex6 construct was no more stable than Nmnat2ΔPSΔBR ( Figure 6A , B; Table 2 ) , suggesting that factors other than protein stability underlie the further increase in its protective capacity . As Nmnat2 degradation is blocked by proteasome inhibitor MG132 [13] , we then asked whether Nmnat2 becomes ubiquitinated and whether these stabilizing mutations reduce ubiquitination . Wild-type and mutant FLAG-Nmnat2 were overexpressed in HEK293 cells , and the K48-specific ubiquitin binding domain of Dsk2 was bound to Glutathione-Sepharose beads and used to immunoprecipitate ubiquitinated proteins . An inactive mutant form of the Dsk2 UBA was used as a control [38] , [39] , and 20 µM MG132 was added 6 h prior to cell lysis to increase the abundance of ubiquitinated proteins . FLAG-Nmnat2 produced a clear ladder of ubiquitinated products ( Figure 6C ) , and all missense mutants and Nmnat2Δex6 showed significantly less ubiquitination ( Figure 6C , D ) . Thus , reduced ubiquitination is likely to contribute to the increased stability and protective capacity of these mutants . Based on these results , we hypothesized that palmitoylation and vesicle association cause wild-type Nmnat2 to become destabilised through increased levels of ubiquitination . In contrast , the nonvesicular , cytosolic location of the Nmnat2 mutants reduces ubiquitination and increases protein stability and protective capacity . The Nmnat2ΔPS data , and strongly enhanced protective capacity of Nmnat2Δex6 over Nmnat2ΔPSΔBR , indicate that the observed effects do not just reflect removal of lysine residues . In agreement with this model , inhibiting palmitoylation directly with 2-BP resulted in reduced levels of ubiquitination on FLAG-Nmnat2 ( Figure S6C , D ) . Furthermore , 2-BP treatment enhanced the half-life of FLAG-Nmnat2 in the emetine chase assay ( Figure S6E , F ) . Based on this , we predicted that a cytosolic , stable Nmnat2 mutant with increased protective capacity ( such as Nmnat2ΔPSΔBR ) would revert to a wild-type behaviour when re-targeted to vesicles . To test this , we first attached sequences to the N-terminus of Nmnat2ΔPSΔBR-PA_GFP that would re-target it to the same Golgi-derived vesicle population as wild-type Nmnat2 . For this , we used exon 6 of Nmnat2 ( Nmnat2ΔPSΔBR-Nterex6-PA_GFP ) or the signal peptide and transmembrane domain of TGN38 ( Nmnat2ΔPSΔBR-NterTGN-PA_GFP ) . These constructs were stably targeted to membranes as assessed by the photoactivation assay ( Figure S7A , B ) , and as predicted , re-targeting significantly reduced the ability to protect injured neurites ( Figure 7A , B ) . We also confirmed that Nmnat2ΔPSΔBR-NterTGN showed substantial levels of ubiquitination that were not detectable in Nmnat2ΔPSΔBR ( Figure 7E ) and that the protein stability of Nmnat2ΔPSΔBR-NterTGN was significantly reduced as expected ( Figure 7C , D ) . While these results suggest that re-targeting cytosolic Nmnat2 to membranes reverts its stability and protective capacity to lower levels as expected , we cannot rule out the possibility that the N-terminal sequences have a direct effect on Nmnat2 stability . To address this issue , we used a commercially available heterodimerisation system ( iDimerize , Clontech ) , in which two proteins tagged with DmrC and DmrA domains , respectively , undergo heterodimerisation after addition of a soluble “A/C heterodimeriser” compound [40] . We used TGN38-DmrC-HA to provide the membrane anchor for re-targeting of cytosolic DmrA-Nmnat2ΔPSΔBR-PA_GFP based on the strong co-migration of TGN38 with Nmnat2 ( see Figure 1 ) . Thus , this system overcomes the abovementioned limitation of the N-terminal targeting sequence as all that is required to induce membrane re-targeting is addition of the small molecule heterodimeriser compound . The photoactivation assay confirmed successful re-targeting , as mobility of DmrA-Nmnat2ΔPSΔBR-PA_GFP was significantly reduced after addition of heterodimeriser , albeit less strongly than using the N-terminal targeting sequence above ( Figure S8 ) . Correspondingly , addition of heterodimeriser also significantly reduced the axon protective capacity of DmrA-Nmnat2ΔPSΔBR-PA_GFP , but only in the presence of TGN38-DmrC-HA ( Figure 8A , B ) . Even though the reduction in protective capacity observed in response to N-terminal re-targeting was stronger than that achieved by heterodimerisation , this was reflected in a lower degree of membrane re-targeting in the heterodimerisation system ( compare Figure S7A , B and Figure S8 ) . Additionally , we confirmed that re-targeting of cytosolic Nmnat2ΔPSΔBR to membranes through heterodimerisation resulted in increased levels of ubiquitination ( Figure 8F , G ) and a decreased protein half-life ( Figure 8C–E ) , confirming the results obtained with N-terminally re-targeted mutants above . At this point , it is interesting to ask whether the observed changes after Nmnat2 re-targeting arise from a special property of the vesicle membranes that Nmnat2 exon 6 and TGN38 target to , or whether they reflect a more general effect of Nmnat2 membrane association . To test this , we attached the mitochondrial outer membrane anchor ( a . a . 1–37 ) of TOM20 to the N-terminus of Nmnat2ΔPSΔBR ( Nmnat2ΔPSΔBR-NterMOM ) . Note that , as with Nmnat2ΔPSΔBR-NterTGN , the Nmnat2 portion of this construct faces the cytosol . The NterMOM tag led to stable membrane association in the photoactivation assay ( Figure S7A , B ) . To confirm targeting to mitochondria , we co-stained neurons expressing Nmnat2ΔPSΔBR-NterMOM-PA_GFP with MitoTracker dye and observed largely overlapping staining patterns ( Figure S7C ) . We then subjected Nmnat2ΔPSΔBR-NterMOM to the ubiquitination assay and found no evidence for increased levels of ubiquitination relative to Nmnat2ΔPSΔBR ( Figure 7E ) . This suggests that the induction of ubiquitination after membrane attachment is indeed specific for transport vesicle membranes . Despite the absence of detectable ubiquitination , however , we found Nmnat2ΔPSΔBR-NterMOM to be destabilized relative to Nmnat2ΔPSΔBR ( Figure 7C , D ) , which resulted in a loss of protective capacity ( Figure 7A , B ) . These findings suggest that targeting to the mitochondrial outer membrane destabilizes Nmnat2 through a mechanism distinct from that operating on transport vesicle membranes . As described above , deletion of exon 6 led to a very strong increase in Nmnat2 protective capacity without any further changes in stability with respect to Nmnat2ΔPSΔBR . To further explore this dissociation between protective capacity and protein stability , we re-targeted Nmnat2Δex6 to vesicle membranes using the N-terminal TGN38 tag . Nmnat2Δex6-NterTGN was stably targeted to membranes in the photoactivation assay ( Figure S9 ) and showed increased ubiquitination ( Figure 9C ) and reduced protein stability ( Figure 9D , E ) with respect to Nmnat2Δex6 . Interestingly , however , this did not affect its protective capacity , which remained indistinguishable from Nmnat2Δex6 up to 72 h after cut ( Figure 9A , B ) . This finding suggests that the increase in protective capacity resulting from loss of exon 6 is sufficient to strongly delay degeneration even when only a low , residual level of protein remains .
Nmnat2 is required for axon survival and is the only confirmed endogenous Nmnat isoform in axons . However , its ability to promote axon survival is limited by its short half-life . We have identified a series of mutations that extend Nmnat2 half-life without disrupting enzyme activity and which significantly increase axon protection . For deletion mutants lacking exon 6 the efficacy even surpasses that of WldS . Surprisingly , these changes arise when Nmnat2 targeting to a population of post-Golgi axonal transport vesicles is disrupted and are reversed when vesicle targeting is restored , indicating a nonvesicular site for the axon survival function of Nmnat2 . The more stable and protective variants are less prone to ubiquitination through a mechanism likely to involve subcellular targeting and not just lysine availability . We identify cysteine-linked palmitoylation as the vesicle targeting mechanism and propose modulation of this targeting as a promising , novel therapeutic strategy for axonopathies . We show that Nmnat2 axonal transport vesicles carry Golgi markers as well as synaptic vesicle markers . In contrast , we find no evidence of Nmnat2 undergoing co-transport with mitochondria . This suggests that Nmnat2 is involved in the regulation of cytosolic and not mitochondrial NAD metabolism in the axon , especially since mitochondria are not thought to take up cytosolic NAD under normal conditions [41] . Thus , any potential influence of Nmnat2 on mitochondrial function is likely to be mediated through indirect mechanisms . Furthermore , given that removal of Nmnat2 from its vesicles does not impair its protective capacity , it seems unlikely that Nmnat2 exists in close proximity to important downstream targets on its own transport vesicles . If this were the case , removal of Nmnat2 from this microenvironment would be expected to result in reduced levels of neurite protection . Instead , it appears that the regulation of overall cytosolic NAD metabolism by Nmnat2 is critical for axon survival . Given the importance of axonal transport , surprisingly little is known about sequences targeting proteins to axons . The mechanism of Nmnat2 membrane association appears very similar to that reported for another axonally transported protein , GAP43 . Both proteins lack a transmembrane domain or alternative membrane targeting structures and depend fully on palmitoylation of a double-cysteine motif for membrane association [27] , [28] , [42] , [43] . For GAP43 , it was reported that , in addition to the palmitoylated cysteine residues themselves , several adjacent basic amino acid residues are also necessary for efficient membrane association [31]–[33] . We found a similar mechanism to operate for Nmnat2 . Our results suggest that basic amino acid residues surrounding the palmitoylation site in exon 6 are involved in mediating initial membrane contact and allow palmitoylation to establish stable membrane anchoring . In their absence , the level of palmitoylation is strongly reduced and membrane targeting of Nmnat2 is less efficient , resulting in a higher level of diffuse , soluble protein . However , once palmitoylation has occurred , membrane association seems to be as stable as for wild-type Nmnat2 over the time-scale of our analysis . This view is also supported by our finding that Nmnat2ΔBR , despite its increased level of diffuse fluorescence in neurites , is still associated with the correct population of transport vesicles . These similarities between GAP43 and Nmnat2 suggest a common underlying axonal targeting motif based on a dual-cysteine palmitoylation site and adjacent or surrounding lysine and arginine residues . Palmitoylation regulates the axonal transport and subcellular sorting of several axonally delivered proteins [44] and is unique among the fatty-acid modifications in that it is readily reversible . It is now well established that palmitate cycling , the recurrent addition and removal of palmitate groups to a target protein , is an important regulatory mechanism in many cases [45] . Thus it is possible that endogenous Nmnat2 undergoes similar cycles of palmitoylation and depalmitoylation . If this is the case , a model could be envisaged in which Nmnat2 is palmitoylated and vesicle bound for the purpose of delivery over long distances into axons . Once in the axon , depalmitoylation could cause Nmnat2 to detach from vesicles and instead assume a diffuse , cytosolic localization in order to carry out its function in regulating axoplasmic NAD metabolism . Thus , palmitate cycling would effectively regulate the ratio of vesicle-bound to diffuse Nmnat2 , and hence ultimately the axon protective capacity of Nmnat2 . Modulation of the Nmnat2 palmitoylation-depalmitoylation cycle by targeting the relevant palmitoyltransferase and thioesterase enzyme ( s ) might hence present a useful tool to alter the course of axon degeneration . We also found support for the hypothesis that increased protein stability underlies the mechanism by which diffuse , cytosolic Nmnat2 becomes more highly protective . Wild-type Nmnat2 , which is mainly vesicle-bound , is very short-lived both in cell lines and in the neurites of primary culture neurons [13] . Here we found it to have a protein half-life of around 40 min , whereas soluble , cytosolic Nmnat2 mutants have a longer half-life . This means that while wild-type Nmnat2 is very rapidly depleted when its supply stops after neurite transection , cytosolic mutants with increased protein stability have an increased potency to protect neurites against degeneration , due to a combination of higher steady-state levels in the neurites before transection and a longer protein half-life after transection . Moreover , the reduced level of ubiquitination in these cytosolic mutants suggests that increased protein stability is a direct result of reduced turnover of Nmnat2 by the ubiquitin-proteasome system . This fits with our previous findings that inhibition of the ubiquitin proteasome system , which was shown to delay Wallerian degeneration [46] , stabilizes endogenous Nmnat2 [13] . Thus our data support a model in which vesicle-bound Nmnat2 is unstable due to its high levels of ubiquitination , which in turn results in its rapid turnover and short protein half-life . Releasing Nmnat2 from its vesicles reduces ubiquitination and leads to a more stable protein with a higher axon protective capacity . Our results indicate that Nmnat2 can undergo ubiquitination on lysine residues outside of exon 6 as Nmnat2 mutants lacking exon 6 lysines ( Nmnat2ΔPSΔBR and Nmnat2Δex6 ) can still be ubiquitinated when re-targeted to membranes . Interestingly , this destabilising effect of palmitoylation-mediated membrane attachment contrasts with findings for several palmitoylated transmembrane domain proteins , including cell-surface receptors [47]–[50] and SNARE proteins [51] , for which it was found that palmitoylation increases protein half-life through reduction of ubiquitin-proteasome mediated degradation . Our results indicate that the effect of palmitoylation on ubiquitination and protein stability might differ for proteins lacking transmembrane regions ( such as Nmnat2 ) . The mechanism by which palmitoylation-mediated vesicle-association causes high levels of ubiquitination in Nmnat2 is as yet unclear and will be the object of future studies . One interesting possibility is the localization of a relevant ubiquitin ligase to the surface of Nmnat2 transport vesicles . Such a vesicle-specific mechanism is supported by our finding that re-targeting cytosolic Nmnat2 mutants to mitochondrial outer membranes does not induce detectable ubiquitination . Alternatively , vesicular axonal transport may deliver Nmnat2 to parts of the axon with higher levels or activity of relevant elements of the ubiquitin proteasome system . A recently published study reported one such element regulating the turnover of Drosophila Nmnat ( dNmnat ) in axons . The Drosophila E3 ubiquitin ligase Highwire was found to be necessary and sufficient for rapid turnover of dNmnat , and of ectopically expressed mammalian Nmnat2 , in the distal stump of injured axons . In its absence , dNmnat persists and degeneration is delayed [52] . Together with our findings , this suggests that ubiquitination regulates the course of axon degeneration both in mammals and Drosophila and that Nmnat2 axonal transport vesicles play an important role in bringing together dNmnat or Nmnat2 with their respective ubiquitin ligases . Furthermore , our results indicate that subcellular localization and protein stability are not the only determinants of Nmnat2 axon protective capacity . Deletion of exon 6 dramatically increases Nmnat2-mediated neurite protection without any further increase in protein stability . Our findings with the enzyme-dead exon 6 deletion mutant suggest that this strong increase in protective capacity depends on Nmnat2 enzymatic activity . However , the reduced stability of this mutant means we cannot completely rule out the possibility that other , nonenzymatic mechanisms contribute to the rise in protective capacity . Interestingly , deletion of exon 6 overcomes the reduction in Nmnat2 protective capacity upon membrane re-targeting that was observed for point mutants . This rescue occurred despite the destabilising effects of membrane attachment , which were unchanged by the removal of exon 6 . This suggests that exon 6 regulates Nmnat2 axon protective function through various mechanisms , which could include protein-protein interactions or additional posttranslational modifications . In summary , we show that Nmnat2 , normally the least axon protective of the three endogenous Nmnat isoforms due to its short half-life , can be converted to a highly protective molecule by disrupting its targeting to axonal transport vesicles . While the importance of these vesicles for long-range axonal trafficking is clear , we suggest that Nmnat2 must dissociate to carry out its axon survival function optimally . We also propose that cytosolic NAD metabolism is central to the axon survival mechanism . Our data establish the principle that Nmnat2 can be modified to promote axon survival and highlight modulation of its palmitoylation state as a route to achieve this . Unlike WldS or other Nmnats , this approach utilizes a protein already identified in wild-type axons , raising the attractive prospect of converting an endogenous axonal protein into one with a protective capacity that matches or even exceeds that of WldS .
Nmnat2-EGFP , FLAG-Nmnat2 , and FLAG-WldS constructs were described previously [13] . WldS-EGFP was created by insertion of the WldS coding sequence into the MCS of pEGFP-NI vector ( Clontech ) . Nmnat2-mCherry was created by replacing the EGFP coding sequence of Nmnat2-EGFP with the mCherry coding sequence from pmCherry-NI ( Clontech ) . Nmnat2ΔPS-EGFP , Nmnat2ΔBR-EGFP , and Nmnat2ΔPSΔBR-EGFP were created from Nmnat2-EGFP using the QuikChange II Site Directed Mutagenesis Kit ( Stratagene ) according to the manufacturer's instructions . Nmnat2Δex6-EGFP was created by PCR amplification of the Nmnat2-EGFP vector excluding exon 6 and introduction of a SacII restriction site to allow vector re-ligation . FLAG-tagged , untagged , and PA_GFP tagged Nmnat2 wild-type or mutants were created by insertion of the appropriate Nmnat2 mutant into the MCS of pCMV-Tag2A ( Stratagene ) , pCMV-Tag4A ( Stratagene ) , and pPAGFP-NI [29] ( Addgene plasmid 11909 ) vectors , respectively . Exon6-EGFP and Exon6-PA_GFP constructs were created by PCR of exon6 from Nmnat2 and insertion into the MCS of pEGFP-NI and pPAGFP-NI vectors . DmrA-Nmnat2ΔPSΔBR-PA_GFP was created by PCR amplification of the DmrA coding sequence from pHet-NucI vector ( Clontech ) and insertion at the N-terminus of Nmnat2ΔPSΔBR-PA_GFP . For the TGN38-DmrC-HA construct , the DmrC-HA coding sequence was PCR amplified from the pHET-1 vector ( Clontech ) and inserted in place of the GFP coding sequence of TGN38-EGFP . Nmnat2ΔPSΔBR-Nterex6-PA_GFP was created by PCR amplification of Nmnat2 exon 6 and insertion at the N-terminus of Nmnat2ΔPSΔBR-PA_GFP . The N-terminal TGN38 targeting sequence was made by fusing the TGN38 signal peptide sequence ( a . a . 1–20 ) to its transmembrane domain surrounded by linker sequences ( a . a . 281–330 ) . This sequence was then inserted at the N-termini of Nmnat2ΔPSΔBR and Nmnat2Δex6 to create Nmnat2ΔPSΔBR-NterTGN and Nmnat2Δex6-NterTGN , respectively . Nmnat2ΔPSΔBR-NterMOM was created by addition of amino acids 1–37 of Mus musculus TOM20 to the N-terminus of Nmnat2ΔPSΔBR . Nmnat2Δex6H24D was created by site-directed mutagenesis of Nmnat2Δex6 . For organelle markers , the following accession numbers were used for PCR primer design . Constructs were amplified from mouse brain cDNA and inserted into the MCS of pEGFP-NI ( Clontech ) or ptagRFP-NI ( Evrogen ) vectors . TGN38-EGFP , NM_009443; Syntaxin6-EGFP , NM_021433; Synaptophysin-EGFP , NM_009305; mito-tagRFP , AK003116 ( bp 1–72 ) ; LmanI-EGFP , AK011495; Golga2-EGFP , NM_133852; GAP43-EGFP , BC080758; SynaptotagminI-EGFP , NM_001252341 . All constructs were verified by DNA sequencing ( Beckman Coulter Genomics ) . Nmnat2 deletion mutants ( Δ32 , Δ43 , and Δ69 ) were a gift from Prof . Giulio Magni ( Ancona , Italy ) . GST-Dsk2 UBA was kindly provided by Dr . Simon Cook ( Cambridge , UK ) . The SNAP25-EGFP construct was a gift from Dr . Luke Chamberlain ( Glasgow , UK ) . GFP-Bassoon was a gift from Prof . Eckart Gundelfinger ( Magdeburg , Germany ) . Lamp1-RFP [53] was from Addgene ( plasmid 1817 ) . All animal work was carried out in accordance with the Animals ( Scientific Procedures ) Act , 1986 , under Project License 80/2254 . C57BL/6JOlaHsd mice were obtained from Harlan UK ( Bicester , UK ) . Dissociated superior cervical ganglia cultures were prepared and maintained in culture as described previously [13] . For live-cell imaging of axonal transport , cells were transferred ( on the day of imaging ) into imaging medium in order to improve performance and detectability of fluorescent proteins [54] . Cells were viable and appeared morphologically normal in imaging medium for at least 3 days . Imaging medium consisted of 1 . 80 mM CaCl2 ( Sigma ) , 0 . 25 µM Fe ( NO3 ) 3 ( Sigma ) , 0 . 81 mM MgSO4 ( AnalaR ) , 5 . 33 mM KCl ( AnalaR ) , 44 . 05 mM NaHCO3 ( Sigma ) , 110 . 34 mM NaCl ( AnalaR ) , 0 . 92 mM NaH2PO4 ( Sigma ) , 4 , 500 mg/l glucose ( AnalaR ) , 110 mg/l sodium pyruvate ( PAA ) , 2 mM glutamine ( Invitrogen ) , 100 ng/ml 7S NGF ( Invitrogen ) , 1% penicillin/streptomycin ( Invitrogen ) , 4 µM aphidicolin ( Calbiochem ) , 1× MEM amino acids ( PAA ) , 30 mg/l glycine ( AnalaR ) , and 42 mg/l serine ( Sigma ) in sterile distilled water . For live-cell imaging of photoactivatable GFP ( PA_GFP ) , cells were transferred ( on the day of imaging ) into Hibernate-E medium ( Invitrogen ) with added 2 mM glutamine ( Invitrogen ) , 100 ng/ml 7S NGF ( Invitrogen ) , 1% penicillin/streptomycin ( Invitrogen ) , 2% B27 supplement ( Invitrogen ) , and 4 µM aphidicolin ( Calbiochem ) . For heterodimerisation , 500 nM A/C heterodimeriser ( Clontech ) was added 8 h before imaging . Where indicated , 40 µM 2-BP ( Sigma ) was added immediately after microinjection . DNA microinjections into the nuclei of primary culture SCG neurons were performed as described [13] . For dual-labelling live cell imaging , both DNA constructs were used at a concentration of 0 . 03 µg/µl in the injection mix . For single-color axonal transport imaging , constructs were used at 0 . 05 µg/µl in the injection mix . For photoactivation experiments , 0 . 05 µg/µl of the photoactivation construct was co-injected with 0 . 01 µg/µl mCherry expression construct except for experiments involving heterodimerisation where 0 . 03 µg/µl each of DmrA- and DmrC-tagged constructs were co-injected with 0 . 01 µg/µl mCherry expression construct . Seventy-five cells were injected per dish , and imaging was performed 24 h after microinjection . For experiments on neurite degeneration after cut , 0 . 01 µg/µl DsRed2 expression vector was co-injected with the relevant Nmnat2 constructs at varying concentrations ( see text ) . We injected 75–100 cells in each dish and neurites were cut 48 h after microinjection . MitoTracker Red CMXRos ( Invitrogen ) was used according to the manufacturer's instructions . Time-lapse imaging of axonal transport was performed on an Olympus CellR imaging system ( IX81 microscope , Hamamatsu ORCA ER camera , 100×1 . 45 NA apochromat objective , 485 and 561 nm laser excitation ) . During imaging , cell cultures were maintained at 37°C in an environment chamber ( Solent Scientific ) . Images were captured at 4 ( single-color ) or 2 . 5 ( dual-colour ) frames per second for 1–2 min . The extent of axonal co-migration of two fluorescent protein markers was analysed in time-lapse recordings of individual neurites . Using ImageJ software version 1 . 44 ( Rasband , W . S . , ImageJ , NIH , Bethesda , Maryland , USA , http://imagej . nih . gov/ij/ , 1997–2011 ) , the same neurite was straightened , contrast adjusted , and projected as a kymograph for both image stacks . The kymographs were then merged to create an overlay . Co-migration was scored for each particle trace according to whether it was detectable in only one or both of the kymograph channels . Co-migration was determined only for moving particles ( i . e . , traces that were not exclusively vertical in kymographs ) . Parameters of axonal transport of fluorescently labelled proteins were determined from straightened time-lapse recordings of individual neurites using the Difference Tracker ImageJ software plugin [55] . Additionally , the percentage of moving and stationary particles was scored manually on kymographs . Photoactivation imaging was carried out on an Olympus FV1000 point scanning confocal microscope system ( IX81 microscope , 60×1 . 35 NA plan super apochromat objective , 488 and 561 nm laser excitation ) . Microinjected cell bodies were identified based on their mCherry fluorescence . Imaging settings were adjusted to standard settings ( 5× zoom , scan rate 8 µm/s , frame rate 3 s/frame ) . After taking a pre-activation image , PA_GFP was activated by a 100 ms pulse of a 405 nm laser at 50% intensity in a 100 pixel region of interest in the cell body . Images were then taken every 3 s for a total of 5 min . For analysis , two circular regions of interest of identical size ( 50 pixel diameter ) were selected in the cell body . One was placed in the originally activated area , while the other one was placed 10–20 µm away in an area that was not activated by the original laser pulse . For quantification , the percentage of combined fluorescence in these two areas that remained in the originally activated area was determined for each time point . Data were fitted for exponential decay , and decay constant ( k ) was calculated using GraphPad Prism 5 . 04 . Degeneration of ds-Red2 labelled neurites was determined for the same field of view at indicated time points after neurite transection . The percentage of neurites remaining continuous and morphologically normal compared to the initial time point was scored for each field . For experiments involving heterodimerisation , A/C heterodimeriser ( Clontech ) was added to the relevant dishes 8 h before cut at a final concentration of 500 nM . Fresh medium ( with heterodimeriser where appropriate ) was added 24 and 48 h after cut . HEK 293 cells were maintained in culture as described [13] . For transfection , cells were grown to 80% confluency in 10 cm or 24-well dishes and transfected using lipofectamine 2000 ( Invitrogen ) according to the manufacturer's instructions . For emetine chase experiments , HEK293 cells in 24-well dishes were co-transfected with WldS-FLAG and the appropriate FLAG-Nmnat2 construct ( turnover of Nmnat2 mutants ) or with Flag-Nmnat1 , TGN38-DmrC-HA , and DmrA-Nmnat2ΔPSΔBR-FLAG ( turnover of re-targeted mutants ) . For relevant experiments , 100 µM 2-BP was added 6 h post-transfection . Twenty-four hours after transfection , cells were treated with 10 µM emetine ( Sigma ) and samples were taken at indicated time points . For ubiquitination experiments , HEK293 cells in 10 cm dishes were transfected with the appropriate FLAG-Nmnat2 construct and , for re-targeting experiments , with empty pCMV-Tag4A or TGN38-DmrC-HA constructs . Twenty-four hours after transfection , 20 µM MG132 ( Sigma ) was added to the medium . Six hours later , cells were lysed and subjected to GST-Dsk2 UBA ( wild-type or mutant ) pulldown assay as described [38] , [39] . Where indicated , 100 µM 2-BP was added 6 h post transfection . SDS-PAGE analysis and Western blotting analysis were performed as described [13] , [56] . Mouse monoclonal anti-FLAG ( Sigma , M2 ) was used at 1∶3 , 000 . AlexaFluor680-conjugated anti-mouse secondary antibody ( Molecular Probes , Eugene , OR , USA ) was used at 1∶5 , 000 . Blots were scanned and quantified using the Odyssey imaging system ( LI-COR Biosciences , Lincoln , NC , USA ) . HEK293 cells in six-well dishes were transfected with the appropriate FLAG-Nmnat2 construct . Twenty-four hours after transfection , 0 . 5 mCi/ml [9 , 103H]-palmitate ( Perkin Elmer ) was added to the medium . After 6 h , cells were washed in PBS , lysed in 500 µl lysis buffer ( 20 mM Tris pH 7 . 5 , 137 mM NaCl , 1 mM EGTA , 1% TritonX-100 , 10% glycerol , 1 . 5 mM MgCl2 , 50 mM NaF , 1 mM Na3VO4 , and protease inhibitor mix ( Roche ) ; all chemicals AnalaR unless stated otherwise ) . The lysate was centrifuged for 10 min , 13 , 000 rpm . Following overnight incubation of the lysate with 5 µg of anti-FLAG antibody ( Sigma ) , 50 µl of washed Sepharose beads ( GE Healthcare ) was added and mixed for another 3 h at 4°C . Beads were washed thrice in lysis buffer and twice in wash buffer ( 50 mM Tris , pH 8 . 0 ) . Bound protein was eluted with Laemmli sample buffer ( BioRad ) and boiling for 5 min and processed for SDS-PAGE . After transfer to PVDF membrane , blots were dried and radiolabel was detected by exposure on Tritium phosphor screen ( Fuji ) for 14 d . Statistical analyses and graph fitting were performed using GraphPad Prism 5 . 04 ( GraphPad Software Inc . ) and SPSS Statistics 19 ( IBM ) .
|
Neurons are polarized cells that rely on bidirectional transport to deliver thousands of cargos between the cell body and the most distal ends of their axons . One cargo that is of particular importance is the NAD-synthesising enzyme Nmnat2 . This surprisingly unstable protein is produced in the cell body and its constant supply into axons is required to keep them alive . If this supply is interrupted , Nmnat2 levels in the distal axon drop below a critical threshold , leading to axon degeneration . The rapid turnover of Nmnat2 contributes critically to the time course of axon degeneration . If its half-life could be extended , axons may be able to survive transient interruptions of its supply . In this study , we find that disruption of Nmnat2 localization to axonal transport vesicles increases both its half-life and its capacity to protect injured neurites . Specifically , association of Nmnat2 with transport vesicles reduces it stability by making it vulnerable to ubiquitination and proteasome-mediated degradation . These findings suggest that modulation of the subcellular localization of Nmnat2 on transport vesicles could serve as a potential avenue for therapeutic treatment of axon degeneration .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"molecular",
"neuroscience",
"neurobiology",
"of",
"disease",
"and",
"regeneration",
"cell",
"biology",
"membranes",
"and",
"sorting",
"biology",
"molecular",
"cell",
"biology",
"neuroscience"
] |
2013
|
Subcellular Localization Determines the Stability and Axon Protective Capacity of Axon Survival Factor Nmnat2
|
Ocular herpes simplex virus infection can cause a blinding CD4+ T cell orchestrated immuno-inflammatory lesion in the cornea called Stromal Keratitis ( SK ) . A key to controlling the severity of SK lesions is to suppress the activity of T cells that orchestrate lesions and enhance the representation of regulatory cells that inhibit effector cell function . In this report we show that a single administration of TCDD ( 2 , 3 , 7 , 8- Tetrachlorodibenzo-p-dioxin ) , a non-physiological ligand for the AhR receptor , was an effective means of reducing the severity of SK lesions . It acted by causing apoptosis of Foxp3- CD4+ T cells but had no effect on Foxp3+ CD4+ Tregs . TCDD also decreased the proliferation of Foxp3- CD4+ T cells . The consequence was an increase in the ratio of Tregs to T effectors which likely accounted for the reduced inflammatory responses . In addition , in vitro studies revealed that TCDD addition to anti-CD3/CD28 stimulated naïve CD4+ T cells caused a significant induction of Tregs , but inhibited the differentiation of Th1 and Th17 cells . Since a single TCDD administration given after the disease process had been initiated generated long lasting anti-inflammatory effects , the approach holds promise as a therapeutic means of controlling virus induced inflammatory lesions .
Ocular infection with herpes simplex virus ( HSV ) can result in a chronic immuno-inflammatory reaction in the cornea which represents a common cause of human blindness [1] , [2] . The pathogenesis of stromal keratitis ( SK ) involves numerous events , but studies in murine SK models indicate that lesions are mainly orchestrated by CD4+ T cells that recognize virus derived peptides , or perhaps altered self proteins unmasked in the damaged cornea [1]–[4] . The severity of SK can be influenced by the balance of CD4+ effector T cells and Foxp3+ regulatory T cells ( Treg ) [5] , [6] . Procedures that change this balance represent a promising approach for therapy . This has been achieved either by adoptive transfer with Treg populations [6] or the repeated administration of reagents that can cause naïve CD4+ T cells to convert to become Treg [7] , [8] . From a therapeutic angle , procedures that could shift the balance of T effectors and Treg after a single drug administration would represent a convenient maneuver . Recent evidence from studies to control autoimmunity and graft-versus-host disease indicate that the objective might be achieved by the administration of stable agonists of the aryl hydrocarbon receptor ( AhR ) [9]–[11] . The AhR is a cytosolic transcription factor that can be activated by different ligands . These include the physiological ligand tryptophan photoproduct 6-formylindolo ( 2 , 3-b ) carbazole ( FICZ ) , and synthetic molecules such as 2 , 3 , 7 , 8- tetrachlorodibenzo-p-dioxin ( TCDD ) [10] , [12] . Signaling through the AhR has consequences that include changes in innate cell function , as well as some modulatory effects on several aspects of T cell immunity [13] , [14] . For example , Weiner and colleagues showed that TCDD administration could suppress the induction of experimental autoimmune encephalomyelitis ( EAE ) , an effect attributed to a reduction of proinflammatory T cells along with the expansion of Treg [9] . By a similar mechanism , TCDD had suppressive effects in an autoimmune diabetes model [15] . Similarly , the administration of TCDD prior to the induction of colitis led to reduced lesions along with an increase in the Treg population [16] . In graft versus host disease ( GVHD ) too , the reduced lesions in TCDD treated animals was attributed to the expansion of adaptive Tregs that suppressed allospecific cytotoxic T cell generation [11] , [17] . Modulating AhR by TCDD has also been shown to control the differentiation of Type 1 regulatory T cells ( Tr1 ) in vitro , which produce IL-10 and are instrumental in the prevention of tissue inflammation , autoimmunity as well as GVHD [18] . Although AhR ligation can result in reduced inflammatory lesions , in some circumstances lesions may be exacerbated . This was noted in the Weiner studies when the physiological ligand FICZ , rather than TCDD , was used for treatment [9] . In this study administration of FICZ boosted Th17 differentiation and increased the severity of EAE . Proinflammatory effects of AhR activation were also noted in a model of rheumatoid arthritis [19] , where synoviocytes were exposed to different concentrations of TCDD and shown to produce inflammatory cytokines . Additional proinflammatory effects of AhR ligation were also associated with pulmonary neutrophilia [20] , [21] , as well as with the induction and expansion of IL-17+ secreting CD4+ T cells ( Th17 ) that expressed high levels of AhR receptors [22] , [23] . Currently , it is not clear why AhR activation causes either an increased , or a reduced effect on inflammatory reactions , but the stability of the ligand used for AhR stimulation is one suspected explanation [24] . Accordingly , TCDD is a non-degradable high affinity ligand for AhR and most studies using this ligand report inhibitory effects on inflammatory reactions [24] , [25] . The effects of AhR agonists have not been evaluated in microbe induced inflammatory lesions . In this report , we show that a single administration of the stable AhR ligand TCDD was highly effective at suppressing the severity of ocular immuno-inflammatory lesions caused by HSV . The outcome was attributed to inhibitory effects on inflammatory IFN-γ+ secreting CD4+ T cells ( Th1 ) and Th17 cells . However , since Foxp3+ regulatory T cell numbers remained unchanged by the treatment , the balance between T effectors and Tregs favored the latter population . TCDD was also shown to cause apoptosis ex vivo of Foxp3- CD4+ T cells and could cause some naïve T cells to convert to Foxp3+ CD4+ T cells . Since a single TCDD administration given after the disease process had been initiated generated long lasting anti-inflammatory effects , the approach holds promise as a therapeutic means of controlling virus induced inflammatory lesions .
To evaluate the role of AhR engagement on the outcome of ocular HSV infection , mice were given a single intraperitoneal ( IP ) administration of TCDD on day 1 post-infection ( pi ) , and the effect on the severity of ocular lesions was compared to untreated controls . All treated animals developed significantly reduced lesions compared to controls , but around 40% of the animals developed clinical signs typical of herpes encephalitis before the end of the 15 day observation period and had to be terminated ( Figure 1A–D ) . Ocular viral loads were also increased in the TCDD treated group ( Figure 1E ) . Accordingly , the drug was judged to be effective but would not be recommended for use when virus is present and actively replicating in the cornea . In other experiments , the physiological AhR ligand FICZ was administered daily starting at day 1 pi . This drug was without significant effects on lesion severity ( Figure 1F ) , and none of the treated animals developed herpetic encephalitis ( data not shown ) . In additional experiments , TCDD administration was begun on day 5 pi , a time when levels of replicating virus in the cornea were barely detectable and inflammatory lesions start to become evident [3] . This treatment procedure resulted in significantly reduced lesion severity , as well as the extent of corneal neovascularization , compared to untreated infected controls ( Figure 2A–D ) , and none of the treated animals developed encephalitis . The treatment procedure delayed the time of lesion appearance and average severity scores were significantly less at most time points over a 15 day observation period . For example , on day 12 pi , whereas 10 of 12 eyes from untreated animals had lesion scores of 3 or above , only 2 of 14 eyes in the treated group had lesions of such severity ( Figure 2B ) . An example of comparative severity of control and treated animals is shown in the histological sections in Figure 2E . In additional experiments terminated on day 28 pi , the pattern of results was similar with treated animals showing significantly diminished lesions compared to untreated controls ( Figure 2F ) . In conclusion , ligation of the AhR with a single administration of TCDD given 5 days after virus infection significantly diminished HSV induced immunopathology . To measure the effect of TCDD treatment on the cellular composition of SK lesions collagen digested corneas were analyzed by FACS and compared to controls at day 15 pi . The combination of three independent experiments is shown in Figure 3A–D . As shown in Figure 3D , the average number per cornea of neutrophils and CD4+ T cells was reduced in the treated group by 2 . 03 fold and 4 . 7 fold respectively when compared to untreated controls . In separate experiments of the same design , pools of corneas were processed to quantify mRNA of selected cytokines ( IL-1β , TNF-α , IL-6 , IFN-γ , and IL-17 ) and chemokines ( CCL20 , CXCL9 , CXCL10 , and CXCL11 ) by quantitative real time PCR ( Q-RTPCR ) . As shown in Figure 3C , the consequence of TCDD treatment was a reduction in the levels of several proinflammatory cytokines and chemokines . However , levels of the cytokine IL-10 was increased to 1 . 4 fold in samples from treated compared to controls . Taken together , our results show AhR ligation by TCDD significantly reduced the total cellular infiltration of CD4+ T cells and neutrophils , as well as the amount of proinflammatory cytokines and chemokines . To measure the consequences of TCDD treatment on the T cell subset composition of SK lesions at day 15 pi , pools of corneas from treated and control animals were collagen digested to recover the T cell population . Part of the pool was stimulated in vitro for 4 hours with PMA and ionomycin to enumerate cells that were either IFN-γ or IL-17 producers . The other fraction was used to enumerate Foxp3+ CD4+ T cells . In the experiment shown , there was an average 12 . 3 fold reduction of Th1 cells and a 9 . 4 fold reduction of Th17 cells in treated compared to control corneas . The numbers of Foxp3+ T cells were almost identical in corneal pools from treated and control animals ( Figure 4C ) . Two additional experiments provided a similar pattern of results . Taken together , our results show that a consequence of TCDD treatment was to increase the ratio of total numbers of Foxp3+ CD4+T cells to both , Th1 and Th17 cells ( Figure 4D ) . Parallel studies of a similar design were performed with T cells isolated from the draining lymph nodes ( DLN ) and spleen collected from the same animals used for the corneal studies . The results shown in Figure 5A–D demonstrate that Th1 and Th17 cell frequencies and total numbers per organ were significantly reduced in TCDD recipients when compared to controls . However the frequencies of Foxp3+ Tregs , compared as a fraction of total CD4+ T cells , were increased in treated animals when compared to controls . Additionally , when the ratio of total numbers of Treg per T effectors was compared to controls , a significant increase in the number of Treg per Th1 or Th17 cells in the TCDD treated mice was evident ( Figure 5E ) . To compare levels of IFN-γ and IL-17 produced by CD4+ T cells from infected and treated or untreated mice , sorted CD4+ T cells were isolated from DLN on day 10 pi and stimulated in vitro with PMA and ionomycin . When comparing the number of IFN-γ and IL-17 secreting CD4+ T cells by ICCS , averages were reduced for both in TCDD treated animals ( Figure 5G ) . Similarly IFN-γ secreting levels measured by ELISA were reduced 2 . 9 fold as a consequence of TCDD treatment ( Figure 5H ) . Results from the previous section indicated that there was a shift in the balance between Tregs and T effectors towards Tregs , as well as a reduction in the production of proinflammatory cytokines . To further determine how TCDD could change the balance of Treg to T effectors , naïve splenocytes from DO11 . 10RAG2-/- ( 98% naïve CD4+ T cells ) animals were stimulated in vitro with plate bound anti-CD3 and anti-CD28 , in the presence of IL-2 . Cultures were either untreated or treated with graded amounts of TCDD ( from 0 . 1 µM to 0 . 25 µM ) . Cultures with 0 . 25 µM of TCDD significantly triggered the conversion of approximately 6 . 2% of CD4+ T cells into Foxp3+ CD4+ T cells , as compared to 0 . 3% in the untreated controls ( Figure 6B ) . Other cultures were TCR stimulated in a cytokine cocktail reported to induce either Th1 or Th17 cells in the additional presence of different doses of TCDD . The outcome was a significant decrease in both Th1 and Th17 cell induction ( Figure 6C–D ) with the highest TCDD dose studied ( 0 . 6 µM ) causing a disappearance of the majority CD4+ T cells from the cultures ( data not shown ) . Taken together , our results indicate that activation of AhR signaling by TCDD can induce some CD4+ T cells to become Foxp3+ , but it is inhibitory to the generation of IFN-γ+ CD4+ and IL-17+ CD4+ T cells . To determine if TCDD had differential effects on Foxp3+ and Foxp3- CD4+ T cell proliferation , Foxp3-GFP mice were infected and some treated with TCDD on day 5 pi . After an injection of 5-Bromo-2-deoxyuridine ( BrdU ) on day 8 pi , experiments were terminated on day 9 pi and proliferation of both Foxp3-CD4+ and Foxp3+ cells was detected by BrdU incorporation . Our results show that TCDD treatment significantly reduced the proliferative response of the Foxp3- CD4+ T cell population in both corneas and lymphoid tissue , but was without significant inhibitory effects on the Foxp3+ CD4+ T cell population . Instead , the effect of TCDD on Foxp3+ cells in the cornea was to cause a modest increase in proliferation ( Figure 7A–B ) . These effects could explain in part the balance between Tregs and T effectors in corneal lesions . Prior studies had shown that TCDD administration in vivo causes thymocytes to undergo apoptosis [26] . We determined if apoptosis of Foxp3- CD4+ T cells could account for the reduced numbers of T effectors . We performed experiments with CD4+ T cells isolated from DLN or spleen on day 8 pi from HSV infected Foxp3-GFP mice . Cells were cultured ex vivo in the presence of TCDD for 5 hours and apoptosis of Foxp3- and Foxp3+ CD4+ T cells was measured using Annexin-V staining . The result showed a dose dependent increase in the apoptosis of Foxp3-CD4+ T cells , but no significant apoptosis of Foxp3+CD4+ T cells ( Figure 8A and C ) . Notably , there was no difference in the frequencies of Tregs with the addition of different concentrations of TCDD as compared to media ( Figure 8B–C ) . Taken together , these results indicate that AhR signaling by TCDD , can promote the apoptosis of Foxp3- CD4+ T cells in vitro , but did not cause the same effect in Foxp3+ Treg .
SK is a blinding immuno-pathological lesion induced by ocular infection with HSV [1]-[3] . Novel treatment procedures are needed to replace the current long term use of antivirals and corticosteroids which have unwanted side effects [27]–[29] . A key to controlling the severity of SK lesions is to suppress the activity of T cells that orchestrate lesions and enhance the representation of cellular and humoral events that inhibit effector cell function . In this report , we have evaluated the use of a novel approach to achieve lesion control in a murine model of SK . We demonstrate that modulation of AhR signaling with a single dose of a synthetic stable molecule ( TCDD ) causes cellular changes in the cornea after HSV infection that account for significantly reduced SK lesion severity . The outcome of therapy was reduced effector Th1 and Th17 cells that orchestrate lesions , a reduction of neutrophils that are mainly responsible for damage to the cornea , as well as an increase in the representation of Foxp3+ Treg . Accordingly , when the ratio of Treg per T effectors was compared to controls , a significant increase in the number of Treg per Th1 as well as Th17 cells in the TCDD treated mice was evident . Foxp3+CD4+ T cells are assumed to function by inhibiting the inflammatory effects of T effectors either directly , or by the generation of counter inflammatory molecules [30] . Since a single administration of TCDD provided effective treatment that lasted for as long as one month , this approach could represent an effective novel therapy for a lesion that is a common cause of human blindness . Aryl hydrocarbon receptors are found in animals in many levels of the evolutionary scale . They can recognize numerous low molecular weight synthetic chemicals as well as a list of endogenous ligands , some of which are photoproducts of tryptophan breakdown [24] , [31] . Several cell types express AhR that includes some , but not all cells , involved in innate and adaptive immunity [32] . Our own interest in AhR ligands stemmed from recent reports that synthetic AhR agonists had anti-inflammatory activity [9] , [15] . Moreover , the dioxin TCDD can provide long term activation of the AhR since it is resistant to metabolic degradation [25] . As a consequence , a single administration can result in long term effects on immune mediated diseases . A recent report using the animal model of multiple sclerosis , EAE , showed disease suppression when animals were pretreated with TCDD [9] . The diminished lesions in treated animals were correlated with expansion of the Foxp3+ CD4+ T cell population and the levels of some cytokines produced by effector cells were reduced . The expansion of the Treg population was explained in part by conversion of naïve T cells to become Treg as could be demonstrated in vitro . In our studies too , we observed that a single TCDD administration was an effective means of reducing HSK ocular lesions , but with our model the outcome appeared to be more the consequence of suppressed numbers of Th1 and Th17 T cells that orchestrate SK , than any notable effect on the expansion of Tregs . Accordingly , the cytokine producing cells in lesions were reduced several fold in treated animals , whereas Treg numbers remained approximately the same in treated and controls . We did confirm the Weiner group [9] observations that TCDD can cause some naïve T cells to convert and become Foxp3+ in vitro , but in our hands this was a modest effect . This notwithstanding , it could be that the relative increase in Treg in the SK lesions of treated animals was the explanation for the reduced lesions , the Treg acting by inhibiting the functions of effectors as well as producing anti-inflammatory cytokines such as IL-10 . Equally possible , however , was that the reduced lesions were the direct consequence of the fewer numbers and less functional effectors in the corneas of treated animals . Such effectors would be less able to recruit inflammatory cells such as neutrophils that are considered responsible for much of the tissue damage of SK [33] , [34] . The reduced numbers of effectors would likely arise either , or both , from an inhibitory effect of TCDD on effector cell proliferation and differentiation , or be explained by the drug causing apoptosis of effectors . The latter effect could readily be demonstrated by in vitro studies with TCR stimulated CD4+ T cells cultured with TCDD . In addition , Foxp3- CD4+ T cells from treated animals proliferated less in vivo than did cells from control animals . In some reports TCDD was shown to prematurely terminate the proliferation and decrease the survival of CD4+ T cell , although differential effects on T cell subsets were not investigated [35] . Nevertheless , since regulatory T cells may be more resistant to apoptosis than conventional T cells [32] , [36] frequencies of Tregs would be expected to increase when other T cell populations are depleted . The reduced number of effector cells present in drug treated animals in our studies were also functionally impaired in their ability to mediate inflammatory reactions . Accordingly , ex vivo stimulation of DLN cells from drug-treated mice produced lower levels of some proinflammatory cytokines as well as chemokines responsible for neutrophil recruitment than cells from control animals . During in vitro studies , AhR ligation was shown to affect the differentiation of T helper subsets , behaving differently under identical culture conditions depending on the ligand used . TCDD for example , was shown to trigger the conversion of Foxp3-CD4+ into Foxp3+CD4+ without the need for TGF-β addition [9] , [37] , to dampen Th17 differentiation [37] and to increase the frequency of IL-17 secreting cells induced by TGF-β plus IL-6 [38] . On the other hand , FICZ under identical culture conditions promoted Th17 differentiation , but not Treg differentiation [37] , [38] . Other ligands too , such as kynurenine ( the first tryptophan metabolite of the IDO pathway ) , was shown to optimally generate Tregs in the presence of TGF-β [39] . In our in vitro experiments not only did we find a conversion of naïve T cells into Tregs , but also provided support for the notion that the TCDD interfered with the primary induction of both Th1 and Th17 cells . Thus , in the presence of TCDD , TCR stimulated naïve CD4+ T cells cultured in conditions to cause their differentiation into either Th1 or Th17 cells , resulted in significant suppression . As it currently stands , our mechanistic experiments cannot establish which is the major explanation for the in vivo anti-inflammatory effects of TCDD against SK . Further investigations are needed and are underway . The use of TCDD represents a potentially valuable approach to control SK since a single injection provided an excellent level of lesion control for at least a month pi . So far our results can only be considered as quasi therapeutic since treatment was begun 5 days after infection , a time when most infectious virus has been eliminated but clinical lesions are yet to become evident [1]–[3] . Moreover , we elected to study only the dose shown to be effective in an autoimmune disease model [9] . Since in some studies the outcome of treatment has been shown to be dose dependent , [35] similar dose response studies are warranted in the SK system and these are planned . Nevertheless , our approach does stand in contrast to most other investigations where treatment was begun prior to disease induction , or before natural disease is expected to occur . With ocular HSV infection in mice , such an early treatment approach would not be recommended because when TCDD was given one day after infection up to half of the animals succumbed to lethal infection of the CNS . Others too have observed that TCDD administration in viral infections can result in increased mortality [40]–[42] . For example , with influenza A virus infection AhR activation by the administration of TCDD decreased the survival time to lethal infection and resulted in mortality with a non lethal dose of the virus [42] , [43] . The cause of lethality was unclear since TCDD treated animals cleared the virus from the lungs as well as a non treated mice [44] . More than likely animals succumbed to lung pathology associated with increased neutrophilia found in the lungs of TCDD treated mice [21] , [45] . Curiously however , in our model TCDD administration reduced the numbers of neutrophils in the infected corneas . Aesthetically , the use of TCDD for therapy , a molecule often castigated as an environmental pollutant , may have minimal appeal . However , the use of a natural ligand for AhR such as FICZ that can be metabolized by the body may not represent a good option . Several studies using FICZ have observed that inflammatory lesions can be exacerbated by such a treatment [9] , [23] . For example , in the studies on EAE by Weiner and colleagues [9] , FICZ treatment resulted in more severe lesions . A similar outcome was reported too by Stockinger and colleagues [23] . In our own studies , we observed no beneficial , or in fact harmful , effects when we treated HSV infected mice with FICZ . One reason AhR ligation with certain ligands can cause enhanced inflammatory lesions is that Th17 T cells , mainly responsible for mediating some inflammatory diseases , express high levels of AhR [22]17 , 18] . Consequently , ligation of AhR on Th17 cells can cause cell expansion and the production of cytokines that contribute to tissue damage [9] , [23] . In the SK model , Th17 cells appear to play only a minor role in SK pathogenesis [46] which may explain our failure to observe adverse effects of FICZ therapy . It could be however that using non-toxic ligands such as 2- ( 1′H-indole-3′-carbonyl ) -thiazole-4-carboxylic acid methyl ester which induces Foxp3+ T CD4+ T cells and suppresses EAE [47] could lead to a more acceptable therapeutic approach for SK . In conclusion , our results are consistent with the observation that modulation of AhR signaling through the use of TCDD plays a role in influencing the expression of SK lesions . The mechanisms involved to explain the outcome were multiple , and involve a change in the balance between effector and regulatory T cells . We anticipate that manipulating AhR signaling , preferable with non-toxic ligands , could represent a useful approach to control an important cause of human blindness .
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Research Council . All animals were housed in Association for Assessment and Accreditation of Laboratory Animal Care ( AAALAC ) -approved animal facilities . The protocol was approved by the Institutional Animal Care and Use Committee of the University of Tennessee ( PHS Assurance number A3668-01 ) . HSV-1 eye infection was performed under anesthesia ( avertin ) , and all efforts were made to minimize animal suffering . Female 6 to 8 weeks old C57BL/6 mice were purchased from Harlan Sprague Dawley ( Indianapolis , IN ) . BALB/c DO11 . 10 RAG2 -/- mice were purchased from Taconic and kept in our pathogen free facility where food , water , bedding , and instruments were autoclaved . All manipulations were done in a laminar flow hood . All experiment procedures were in complete agreement with the Association for Research in Vision and Ophthalmology resolution on the use of animals in research . HSV-1 RE Tumpey and HSV-RE Hendricks were propagated and titrated on Vero cells ( American Type Culture Collecting no . CCL81 ) using standard protocols . The virus was stored in aliquots at −80°C until use . CD4-allophycocyanin ( RM4 . 5 ) , CD4-FITC ( RM4 . 5 ) , Foxp3-PE ( FJK-16s ) , anti-IFN-γ-FITC ( XMG1 . 2 ) , anti-IL17-PE ( TC11-18H10 ) , CD45-allophycocyanin ( 30-F11 ) , CD11b-PerCP ( M1/79 ) , Ly6G-PE ( 1A8 ) . Corneal infections of C57BL/6 mice were done under deep anesthesia induced by IP injection in tribromoethanol ( avertin ) as previously described [48] . Mice's corneas were scarified with a 27-gauge needle , and a 3 µl drop containing the specific viral dose was applied to the eye . Eyes were examined on different days pi ( dpi ) with a silt-lamp biomicroscope ( Kowa Company , Nagoya , Japan ) measuring the progression of SK lesion severity and angiogenesis of individual mice . The scoring system was as follows: 0 , normal cornea; +1 , mild corneal haze; +2 , moderate corneal opacity or scarring; +3 , severe corneal opacity but iris visible; +4 , opaque cornea and corneal ulcer; +5 , corneal rupture and necrotizing keratitis [49] . The severity of angiogenesis was recorded as described previously [50] . According to this system , a grade of 4 for a given quadrant of the circle represents a centripetal growth of 1 . 5 mm toward the corneal center . The score of the four quadrants of the eye were then summed to derive the neovessel index ( range 0–16 ) for each eye at a given time point . TCDD ( Sigma Aldrich ) diluent was evaporated with nitrogen and reconstituted with DMSO . Female 6 to 8 weeks old C57BL/C mice were ocularly infected under deep anesthesia with 1×104 PFU of HSV-1 RE Tumpey and divided randomly into groups . Animals in the treated groups were either treated with TCDD on day 1 pi or day 5 pi IP , being the dose administered of 1 µg/mice . Animals in the control groups were treated the same days ( either day 1 or day 5 pi ) with DMSO IP . Mice were observed for SK and angiogenesis progression from day 5 until day 15 or 28 as described elsewhere [49] . Most of the experiments were repeated at least three times . FICZ ( Biomol International , L . P . , Plymouth Meeting , PA ) was dissolved in DMSO . Female 6 to 8 weeks old C57BL/C mice were ocularly infected under deep anesthesia with 1×104 PFU of HSV-1 RE Tumpey and divided randomly into groups . Animals in the treated groups were either treated daily with FICZ from day 1 pi to day 11 pi ( IP ) , being the dose administered of 1 µg/mice . Animals in the control groups were treated the same days with DMSO IP . Mice were observed for HSK and angiogenesis progression from day 5 until day 15 as described elsewhere [49] . Eye swabs were taken from infected corneas using sterile swabs at the indicated time points . Infected corneas were extracted on day 6 pi and placed on ice sterile 2 . 0-mL straight-wall ground-glass tissue homogenizers ( Wheaton ) with media and homogenized . Homogenates were centrifuged ( 2 , 250 g at 4°C ) for 5 min , place on ice , and immediately plated . Titrations were performed by a standard plaque assay as described previously [51] . Titers were calculated as log10 pfu/ml per a standard protocol [52] . Eyes from control and TCDD treated mice were extirpated on day 15 pi and snap frozen in OCT compound ( Miles , Elkart , IN ) . Six micron thick sections were cut , air dried in a desiccation box . Staining was performed with hematoxylin and eosin ( Richard Allen Scientific , Kalamazoo , MI ) . RNA was extracted from cells and tissue using TRIzol LS reagent ( Invitrogen ) . Total cDNA was made with 500ng of RNA using oligo ( dT ) primer . Quantitative PCR ( Q-RTPCR ) was performed using SYBR Green PCR Master Mix ( Applied Biosystem , Foster City , CA ) with iQ5 real-time PCR detection system ( Bio Rad , Hercules , CA ) using 5 µl of cDNA for 40 cycles . The expression levels of different molecules were normalized to β-actin using Δ threshold cycle method calculation . Relative expression between mock infected samples and control or day 5 TCDD treated samples from day 15 pi were calculated using the 2-ΔΔCt formula: ΔΔCt = ΔCt , sample - ΔCt , reference . Here , ΔCt is the change in cycling threshold between the gene of interest and the ‘housekeeping’ gene β-actin , where ΔCt , sample was the Ct value for any day 5 TCDD treated or control samples from day 15 pi normalized to the β-actin gene and ΔCt , reference was the Ct value for the mock infected samples ( scratched and infected only with PBS ) also normalized to β-actin . Each of the samples was run in duplicates to determine sample reproducibility , and a mean Ct value for each duplicate measurement was calculated . The PCR primers used were the following: βactin F 5′-CCTTCTTGGGTATGGAATCCTG-3′ and R 5′-GGCATAGAGGTCTTTACGGATG-3′ , IL-6 F 5′-CGTGGAAATGAGAAAAGAGTTGTGC-3′ and R 5′- ATGCTTAGGCATAACGCACTAGGT-3′ , TNF-α F 5′-CAGCCTCTTCTCATTCCTGCTTGTG-3′ and R 5′- CTGGAAGACTCCTCCCAGGTATAT-3′ , IL-1β F 5′-GAAATGCCACCTTTTGACAG-3′ and R 5′- CAAGGCCACAGGTATTTTGT-3′ , IFN-γ F 5′-GGATGCATTCATGAGTATTGC-3′ and R 5′- GCTTCCTGAGGCTGGATTC-3′ , IL-17A F 5′-GCTCCAGAAGGCCCTCAG-3′ and R 5′- CTTTCCCTCCGCATTGACA-3′ , IL-10 F 5′- CCTTTGACAAGCGGACTCTC-3′ and R 5′- GCCAGCATAAAAACCCTTCA-3′ , CXCL-9 F 5′-CAAGCCCCAATTGCAACAAA-3′ and R 5′- TCC GGA TCT AGG CAG GTT TGA-3′ , CXCL-10 F 5′-TGC TGG GTC TGA AGT GGG ACT-3′ and R 5′- AAG CTT CCC TAT GGC CCT CA-3′ , CXCL-11 F 5′-GGTCACAGCCATAGCCCTG-3′ and R 5′- AGCCTTCATAGTAACAATC-3′ , CCL-20 F 5′-GCCTCTCGTACATACAGACGC-3′ and R 5′- CCAGTTCTGCTTTGGATCAGC-3′ . CD4+ T cells were purified from pooled DLN single cell suspension obtained from HSV-infected mice using a mouse CD4+ T cell isolation kit ( Miltenyi Biotec , Auburn , CA ) . The purity was achieved at the extent of 90% . Purified CD4+ T cells were analyzed by Flow cytometry and ELISA after stimulation for the expression of IFN-γ and IL-17 . DLN single cell suspensions from individual mice were collected at day 15 pi . Cells were stimulated in vitro with anti-CD3 ( 2 µg/ml ) and anti-CD28 ( 1 µg/ml ) for 48 h at 37°C . Additionally DLN single cell suspensions from mice were also collected at day 10 pi and CD4+ T cells were purified using magnetic columns . Cells were then stimulated in vitro with PMA ( 50 ng ) and ionomycin ( 500 ng ) for 4 h at 37°C . The concentrations of IFN-γ and IL-17 were measured by sandwich ELISA kits from eBioscience . Splenocytes isolated from DO11 . 10 RAG2 -/- mice were used as a precursor population for the induction of Foxp3+ in CD4+ T cells as described elsewhere [6] . Briefly , 2×106 splenocytes after RBC lysis and several washings were cultured in 1ml volume previously optimized doses of plate bound anti-CD3 Ab ( 0 . 123 µg/ml in 200 µl total volume ) , rIL-2 ( 25–100 U/ml ) and TGFβ ( 2 . 5–10 ng/ml ) for 5 days at 37°C in a 5% CO2 incubator . Different concentrations of TCDD were also added . After 5 days samples were characterized for Foxp3 intranuclear staining using an eBioscience kit and analyzed by flow cytometry . Naïve CD4+ T cells were stimulated for 4 to 5 days with plate bound antibody to CD3 ( 4 µg/ml ) and anti CD28 ( 2 µg/ml ) . For Th1 differentiation recombinant mouse IL-12 ( 10ng/ml ) and anti IL-4 ( 10 µg/ml ) were used . In the case of Th17 differentiation TGF-β ( 2 . 5ng/ml ) , IL-6 ( 30 ng/ml ) , anti IL-4 ( 10 µg/ml ) and anti IFN-γ ( 10 µg/ml ) were added . Concentrations of TCDD were added into cultures at the beginning of the experiment . After 5 days samples were analyzed by intracellular cytokine staining for the production of IFN-γ and IL-17 using a BD biosciences kit and then flow cytometry . The culture mediums used were IMDM ( Sigma-Aldrich ) for Th17 differentiation or RPMI 1640 ( Sigma-Aldrich ) for Th1 differentiation , both supplemented with 2×10−3 M L-glutamine , 100 U/ml penicillin , 100 µg/ml streptomycin , 5×10−5 M β-mercaptoethanol , and 5% FCS [53] . Foxp3+-GFP knock-in animals were kindly provided by Dr . M . Oukka of Seattle Children's Research Institute . Mice were infected and divided into two groups: non-treated and TCDD treated mice . 8 days after ocular HSV 1 infection mice were injected IP with BrdU ( 1mg/mouse ) and were terminated 12 hours later . 9 dpi , host Foxp3+CD4+ and Foxp3-CD4+ T cells that incorporated BrdU were analyzed by staining with anti BrdU antibody using an APC BrdU flow kit from BD Pharmingen as per the manufacturer's instructions . Samples were acquired with a FACSCalibur ( BD biosciences ) , and the data were analyzed using the FlowJo software . DLN cells and splenocytes isolated from HSV-infected Foxp3-GFP C57BL/6 mice at 8 days pi were incubated for 5h with various concentrations of TCDD in 96 well flat-bottom plate in 5% CO2 incubators . After incubation period was over , cells were stained for annexin V using a kit from BD biosciences . Additionally cells were costained for CD4 . Stained cells were analyzed immediately by flow cytometry . Most of the analyses for determining the level of significance were performed using unpaired two-tailed Student's t test . Values P≤0 . 001 ( *** ) , P≤0 . 01 ( ** ) , P≤0 . 05 ( * ) were considered significant . Results are expressed as means ±SEM . For some experiments , as mentioned in the figure legends , a one-way ANOVA test was applied . CD4 ( MGI:88335 ) , IFN-γ ( MGI:107656 ) , Foxp3 ( MGI:1891436 ) , IL-17 ( MGI:107364 ) , IL-1β ( MGI:96543 ) , TNF-α ( MGI:104798 ) , IL-6 ( MGI:96559 ) , CCL20 ( MGI:1329031 ) , CXCL9 ( MGI:1352449 ) , CXCL10 ( MGI:1352450 ) , CXCL11 ( MGI:1860203 ) , IL-10 ( MGI:96537 ) , β-actin ( MGI:87904 ) , CD45 ( MGI:97810 ) , CD11b ( MGI:96607 ) , Ly6G ( MGI:109440 ) , CD3 ( MGI:88332 ) , CD28 ( MGI:88327 ) , Annexin V ( MGI:106008 ) , IL-12 ( MGI:96540 ) , IL-4 ( MGI:96556 ) , TGF-β ( MGI:98725 ) , IL-6 ( MGI:96559 ) .
|
This report describes a novel approach to control a blinding immuno-inflammatory reaction in the eye caused by herpes simplex virus . We showed that a single administration of TCDD , a stable agonist of the aryl hydrocarbon receptor , significantly reduced the severity of herpes keratitis lesions . The outcome of the therapy was a change in the balance of effector cells responsible for orchestrating lesions , with regulatory cells able to inhibit the inflammatory effects of the effectors . Since a single administration of TCDD provided effective treatment that lasted for as long as one month , this approach could represent a valuable therapy for a lesion that is a common cause of human blindness .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"immunopathology",
"medicine",
"infectious",
"diseases",
"inflammation",
"immunity",
"virology",
"immunology",
"biology",
"microbiology",
"viral",
"diseases",
"immunomodulation",
"immune",
"response"
] |
2011
|
Controlling Viral Immuno-Inflammatory Lesions by Modulating Aryl Hydrocarbon Receptor Signaling
|
The insulin/IGF-1 signaling pathway plays a critical role in stress resistance and longevity , but the mechanisms are not fully characterized . To identify genes that mediate stress resistance , we screened for C . elegans mutants that can tolerate high levels of dietary zinc . We identified natc-1 , which encodes an evolutionarily conserved subunit of the N-terminal acetyltransferase C ( NAT ) complex . N-terminal acetylation is a widespread modification of eukaryotic proteins; however , relatively little is known about the biological functions of NATs . We demonstrated that loss-of-function mutations in natc-1 cause resistance to a broad-spectrum of physiologic stressors , including multiple metals , heat , and oxidation . The C . elegans FOXO transcription factor DAF-16 is a critical target of the insulin/IGF-1 signaling pathway that mediates stress resistance , and DAF-16 is predicted to directly bind the natc-1 promoter . To characterize the regulation of natc-1 by DAF-16 and the function of natc-1 in insulin/IGF-1 signaling , we analyzed molecular and genetic interactions with key components of the insulin/IGF-1 pathway . natc-1 mRNA levels were repressed by DAF-16 activity , indicating natc-1 is a physiological target of DAF-16 . Genetic studies suggested that natc-1 functions downstream of daf-16 to mediate stress resistance and dauer formation . Based on these findings , we hypothesize that natc-1 is directly regulated by the DAF-16 transcription factor , and natc-1 is a physiologically significant effector of the insulin/IGF-1 signaling pathway that mediates stress resistance and dauer formation . These studies identify a novel biological function for natc-1 as a modulator of stress resistance and dauer formation and define a functionally significant downstream effector of the insulin/IGF-1 signaling pathway . Protein N-terminal acetylation mediated by the NatC complex may play an evolutionarily conserved role in regulating stress resistance .
The ability to cope with fluctuating environmental stresses is critical for animal survival . Environmental stresses include a wide range of factors such as extremes in temperature , oxidation , and metal availability . A stress response might promote tolerance against a specific challenge or provide broad-spectrum resistance , and a critical question in this field is how specific stress responses mediate resistance to one or more forms of environmental challenge ? The nematode Caenorhabditis elegans is an important model system for studies of stress resistance . In response to stresses such as high temperature , low nutrient availability , and high population density , developing larvae will adopt an alternative L3 stage called dauer that is stress resistant [1] . Studies of dauer formation led to the discovery of an insulin/IGF-1 signaling pathway as a critical regulator of this stress response [2] . Loss-of-function mutations in daf-2 and age-1 cause dauer constitutive phenotypes , whereas loss-of-function mutations in daf-16 cause dauer defective phenotypes [1] . daf-2 encodes an insulin/IGF-1 receptor homolog , and age-1 encodes a phosphatidylinositol-3-OH kinase catalytic subunit homolog [3] , [4] . In addition to mediating a developmental switch in larvae , this pathway functions throughout the life of the animal to mediate stress resistance , since daf-2 loss-of-function mutations cause increased tolerance to multiple stresses and an extended lifespan [3] , [5] , [6] . These daf-2 mutant phenotypes are suppressed by mutations in daf-16 , indicating that daf-16 is a major downstream effector of the insulin/IGF-1 signaling pathway that is negatively regulated by daf-2 activity . daf-16 encodes a FOXO transcription factor [7] , [8] . Because DAF-16 plays a central role in promoting longevity and stress tolerance , a major goal has been to identify and characterize DAF-16 transcriptional targets [9]–[17] . Although several genes have been demonstrated to be directly regulated by DAF-16 , the understanding of how DAF-16 target genes mediate stress resistance and longevity remains fragmentary . The insulin/IGF-1 signaling pathway regulates stress tolerance , metabolism , and longevity in multiple species including mammals [18]–[22] . Thus , the identification of physiologically significant DAF-16 targets may suggest strategies for promoting longevity and stress tolerance in C . elegans and higher eukaryotes . The metal zinc is a nutrient that is essential for all organisms and plays many roles in biological systems . Zinc functions in signal transduction pathways and contributes to protein structure and activity [23]–[25] . Zinc deficiency and excess both cause a wide spectrum of defects , demonstrating the importance of zinc homeostasis [26]–[31] . Zinc deficiency appears to be deleterious due to the reduced function of many zinc-requiring proteins and signaling events [25] . The mechanisms underlying excess zinc toxicity are not well characterized; excess zinc may displace other physiological metals or bind to low-affinity sites , leading to altered or decreased protein function [32] . In addition to zinc , several other metals are toxic in excess , including cadmium , nickel , and copper [33]–[35] . To characterize mechanisms of metal stress resistance , we performed a forward genetic screen for mutations that caused resistance to high levels of dietary zinc [36] . We reasoned that tolerance to high zinc could be caused by two general mechanisms: ( 1 ) mutations might affect zinc metabolism and reduce the accumulation of toxic zinc , or ( 2 ) mutations might cause alterations that promote growth and survival in the presence of high levels of zinc . We identified haly-1 , which encodes the enzyme histidine ammonia lyase that metabolizes histidine [37] . Loss-of-function mutations in haly-1 do not affect zinc accumulation but rather cause an increase in histidine levels resulting in increased chelation of zinc and nickel , and chelation by histidine reduces the toxicity of these metals [37] . Notably , haly-1 mutations have not been demonstrated to cause resistance to other stressors . By contrast , mutations in daf-2 and age-1 , members of the insulin/IGF-1 pathway , cause resistance to a broad-spectrum of stressors including the metals cadmium and copper , oxidative stress , and heat stress [5] , [35] , [38] . Here we describe natc-1 , a new gene that was discovered in the screen for worms that are resistant to high zinc toxicity , which encodes the C . elegans N-α-acetyltransferase 35 . N-α-acetyltransferases ( NATs ) are highly conserved among eukaryotes and function as protein complexes to transfer the acetyl group of acetyl coenzyme A to the α-amino group of the first amino acid of a target protein . natc-1 encodes an auxiliary subunit of the NatC complex which acetylates proteins that begin with the amino acids Met-Leu , Met-Phe , Met-Ile , and Met-Trp [39] , [40] . natc-1 mutations caused resistance to multiple metals , oxidation , and heat , indicating that natc-1 modulates broad-spectrum stress resistance , similar to mutations in insulin/IGF-1 signaling genes . Interestingly , the natc-1 promoter contains an evolutionarily conserved DAF-16 binding site , and DAF-16 binds natc-1 in vivo [11] , [41] , [42] . We demonstrated that natc-1 is transcriptionally repressed by DAF-16 activity and that natc-1 interacts with genes in the insulin/IGF-1 signaling pathway to mediate stress resistance and dauer formation . These results indicate that natc-1 is directly regulated by DAF-16 and functions as a downstream effector of the insulin/IGF-1 signaling pathway . Supporting this model , mutations in natc-1 that increase stress resistance are epistatic to daf-16 mutations . These results provide novel insights into the transcriptional regulation of natc-1 by the insulin/IGF-1 signaling pathway and the biological function of protein N-terminal acetylation in mediating stress resistance . Furthermore , our data elucidate a new mechanism used by the insulin/IGF-1 signaling pathway to mediate stress tolerance and dauer formation .
We performed a forward genetic screen to identify mutant strains that are resistant to the growth arrest and lethality caused by high levels of dietary zinc [36] . Nineteen mutations were identified and positioned in the genome using a genome-wide map of single nucleotide polymorphism ( SNP ) markers [36] . Here we focus on two of these mutations , am134 and am138 , that caused significant resistance to dietary zinc toxicity ( Figure 1A , B ) . Both mutations displayed tightest linkage to the same SNP , pkP5513 , positioned at +0 . 1 map units on chromosome V ( Figure 1C ) . Three factor mapping experiments indicated that am138 is positioned between dpy-11 and unc-42 , a 325 kilobase pair interval that contains pkP5513 ( Figure 1C ) [36] . To identify the gene affected by these mutations , we performed whole genome sequencing using DNA from the am134 and am138 mutant strains . Candidate mutations in the mapping interval were identified by comparing the am134 and am138 DNA sequence to the wild-type DNA sequence . The am134 and am138 strains both contained candidate mutations in the predicted open reading frame T23B12 . 4 , suggesting that these mutations caused resistance to zinc toxicity . The predicted T23B12 . 4 protein is homologous to human N-α-acetyltransferase 35 , an auxiliary subunit of the NatC complex . Thus , we named this gene natc-1 . The mutation in the am134 mutant strain , which was induced with the mutagen ENU , is a C to T substitution that changes codon 691 from arginine ( CGA ) to stop ( TGA ) . This nonsense mutation is predicted to truncate 110 amino acids from the NATC-1 protein ( Figure 2A , B ) . The mutation in the am138 mutant strain , which was induced by ENU mutagenesis , is a 186 base pair deletion that eliminates portions of exons 1 and 2 and all of intron one ( Figure 2A , B ) . This deletion eliminates the codons for amino acids 13–59 , and the mutated open reading frame is predicted to have a frame shift and encounter a stop at the new codon 15 . The loss of coding sequences and early truncation suggest natc-1 ( am138 ) is likely to be a strong loss-of-function or null allele . To test the hypothesis that mutations in natc-1 cause resistance to zinc toxicity , we analyzed the natc-1 ( ok2062 ) deletion mutation that was generated by the C . elegans knock out consortium [43] . natc-1 ( ok2062 ) is a 1 , 108 base pair deletion that eliminates a portion of exon 3 , intron 3 , exon 4 , intron 4 , and a portion of exon 5 ( Figure 2A , B ) . natc-1 ( ok2062 ) mutant animals displayed significant resistance to zinc toxicity , similar to natc-1 ( am134 ) and natc-1 ( am138 ) mutant animals ( Figure 1B ) . This result supports the hypothesis that mutations in natc-1 cause resistance to zinc toxicity . To independently test the hypothesis that the am134 mutation in natc-1 causes resistance to zinc toxicity , we determined whether a wild-type version of natc-1 could rescue this phenotype . We generated five independently derived transgenic strains containing extrachromosomal arrays with wild-type copies of natc-1 in the background of natc-1 ( am134 ) . All the transgenic strains displayed a significant decrease in zinc resistance when compared to their non-transgenic siblings . This is indicative of a more wild-type phenotype and rescue activity ( Figure 1C , Figure S1A ) . To determine if an intact natc-1 open reading frame is necessary for rescue activity , we generated transgenic animals containing a natc-1 locus that encodes a mutant protein with a 112 amino acid deletion in the background of natc-1 ( am134 ) . These transgenic animals did not display rescue of the mutant phenotype , indicating that the rescue activity of the natc-1 locus requires an intact open reading frame ( Figure 1C , Figure S1B ) . Together , these results demonstrated that natc-1 is the gene affected by am134 and am138 ( reference allele ) , and that mutations in natc-1 caused resistance to zinc toxicity . To characterize the products generated from the natc-1 locus , we analyzed natc-1 mRNA . The C . elegans expressed sequence tag ( EST ) project isolated multiple cDNAs corresponding to natc-1 , and we determined the DNA sequence of three independently derived cDNAs . The cDNA sequences were used to infer the mRNA sequence from exon 3 to the 3′ end , including the position of the polyA tail 330 nucleotides downstream of the TGA stop codon . To characterize the 5′ end of the transcript , we conducted a 5′ RACE experiment that showed the natc-1 mRNA contains a 22 nucleotide splice leader 1 ( SL1 ) sequence that begins 14 base pairs upstream of the start codon . Together , the analysis of cDNAs and 5′ RACE indicated that the natc-1 mRNA contains 8 exons and defined the complete predicted open reading frame ( Figure 2A ) . The predicted NATC-1 protein contains 799 amino acids . To determine the expression pattern and sub-cellular localization of NATC-1 , we generated transgenic natc-1 ( am138 ) animals expressing NATC-1 protein fused to green fluorescent protein ( GFP ) under the control of the native natc-1 promoter . Live animals were imaged with confocal fluorescence microscopy , and NATC-1::GFP was detected in a wide range of cells and tissues in a pattern that suggests cytoplasmic localization ( Figure 3 ) . NATC-1::GFP was detected throughout development from early larval stages through late adulthood . To confirm that the expression pattern of NATC-1::GFP is representative of the expression pattern of endogenous NATC-1 , we demonstrated that the extrachromosomal array expressing NATC-1::GFP rescued the natc-1 ( am138 ) zinc-resistance phenotype ( Figure 1C , Figure S1C ) . Comparison of NATC-1 protein sequence to databases using the method of BLAST revealed that NATC-1 is most similar to N-α-acetyltransferase 35 proteins , which are auxiliary subunits of the NatC complex [44] . Figure 2B shows an alignment of C . elegans NATC-1 with Drosophila melanogaster and human proteins; C . elegans NATC-1 is 24% identical to human NAA35 , suggesting that it may have similar biochemical functions . NATC-1 is an auxiliary subunit of the NatC complex , and C . elegans B0238 . 10 ( NATC-2 ) is the predicted catalytic subunit [45] . The NatC complex catalyzes the acetylation of the N-termini of translating proteins ( Figure 2C ) . The NatC complex specifically acetylates translating proteins that begin with Met-Ile , Met-Leu , Met-Trp , or Met-Phe [39] , [40] . To identify predicted NatC target proteins , we conducted a bioinformatic analysis using the fully sequenced C . elegans genome . Approximately 4 , 300 proteins have Ile , Leu , Trp , or Phe in amino acid position two . These proteins represent ∼17% of the C . elegans proteome and are candidates to be acetylated by the NatC complex . Zinc resistance displayed by natc-1 mutant animals could be explained by two general models: ( 1 ) natc-1 mutant animals have decreased levels of zinc , perhaps as a result of reduced zinc uptake or increased zinc excretion and ( 2 ) natc-1 mutant animals have normal levels of zinc but increased tolerance to high zinc toxicity . To distinguish between these possibilities , we used inductively coupled plasma mass spectrometry ( ICP-MS ) to measure total animal zinc content . Synchronized populations of animals were cultured with NAMM , harvested , and analyzed for zinc content . The total animal zinc content of natc-1 ( am134 ) , natc-1 ( am138 ) , and natc-1 ( ok2062 ) mutant animals was not consistently different from wild-type animals when cultured with or without supplemental zinc ( Figure S2 ) . These results indicate that mutations in natc-1 cause resistance to zinc toxicity by increasing the ability of the animal to tolerate excess zinc that results from a high zinc diet rather then by reducing zinc accumulation . Mutations in haly-1 that cause resistance to high zinc toxicity were identified in the same genetic screen as mutations in natc-1 [37] . The mechanism of action of mutations in haly-1 appears to be accumulation of histidine , which is hypothesized to reduce high zinc toxicity by chelation of the ion . To determine if mutations in natc-1 may cause resistance to high zinc toxicity by a similar mechanism , we analyzed natc-1 ( am138 ) ;haly-1 ( am132 ) double mutant animals . If natc-1 ( am138 ) and haly-1 ( am132 ) cause zinc resistance by affecting the same pathway or process , then the resistance to high zinc toxicity phenotypes might not be additive . Interestingly , natc-1 ( am138 ) ;haly-1 ( am132 ) double mutant animals displayed enhanced resistance to high zinc toxicity compared to natc-1 ( am138 ) or haly-1 ( am132 ) single mutant animals ( Figure S3 ) . This result suggests that resistance to high zinc toxicity caused by natc-1 mutations may be mechanistically distinct from that caused by haly-1 mutations . To determine if natc-1 causes resistance to additional metals , we cultured wild-type animals and natc-1 ( am138 ) animals on NAMM plates supplemented with cadmium , nickel , or copper . Concentrations that caused ∼50% sterility of wild-type animals were chosen for each metal to maximize the sensitivity of the assay . natc-1 mutant animals displayed improved growth and development compared to wild-type animals when cultured with 200 µM zinc , 20 µM cadmium , 50 µM nickel , or 300 µM copper ( Figure 4A–D ) . These data suggest that mutations in natc-1 cause resistance to toxicity induced by both physiological and non-physiological metal ions . To determine if mutations in natc-1 cause resistance to stressors in addition to metal ions , we analyzed heat stress . Wild-type and natc-1 mutant animals were cultured at 35°C and survival times were monitored . natc-1 ( am138 ) animals displayed a significant 19% extension of survival compared to wild-type animals ( Figure 5A , Table 1 ) . These data suggest that mutations in natc-1 cause resistance to heat toxicity . To analyze oxidative stress , we cultured animals with 40 mM paraquat and monitored survival time . natc-1 ( am138 ) animals displayed a significant 40% extension of survival compared to wild-type animals ( Figure 5B , Table 1 ) . Taken together , these data suggest that mutations in natc-1 cause resistance to a broad-spectrum of stressors , including high levels of multiple metal ions , oxidative damage , and high heat . One hypothesis that might explain the stress resistance phenotype is that natc-1 ( lf ) mutations stimulate the unfolded protein response . To investigate this hypothesis , we used the method of qRT-PCR to analyze the mRNA levels of the stress-induced genes hsp-4 , hsp-6 , and hsp-16 . 2 . Wild-type and natc-1 ( am138 ) animals did not display statistically significant differences in mRNA levels for these genes ( p>0 . 05 ) , suggesting that loss of natc-1 activity does not stimulate the unfolded protein response . Furthermore , gst-4 mRNA levels , which are induced by oxidative stress and proteosomal dysfunction [46] were not significantly altered in natc-1 ( am138 ) mutants compared to wild type ( p>0 . 05 ) . To determine how natc-1 activity affects longevity , we analyzed the lifespan of wild-type and natc-1 ( am138 ) mutant animals . natc-1 ( am138 ) animals displayed a significant 31% reduction in mean lifespan compared to wild-type animals ( Figure 5C , Table 1 ) . These data suggest that natc-1 is a lifespan assurance gene when animals are cultured at an optimal temperature with abundant food , conditions that minimize stress . To analyze the regulation of natc-1 , we used qRT-PCR to monitor the level of natc-1 mRNA . Because natc-1 ( lf ) mutations cause zinc resistance , we examined the transcriptional response to high dietary zinc . natc-1 mRNA levels were not affected by 200 µM supplemental dietary zinc ( p>0 . 05 ) , suggesting that natc-1 transcription is not regulated to promote zinc tolerance . To further analyze regulation , we examined the insulin/IGF-1 signaling pathway because it plays a pivotal role in stress resistance in C . elegans [2] . daf-2 encodes the insulin receptor that functions to inhibit dauer formation and stress resistance; daf-2 ( e1370 ) is a partial loss-of-function mutation that causes a temperature-sensitive dauer constitutive ( Daf-c ) phenotype in larvae and an increased stress resistance phenotype in adults [6] , [35] , [38] . daf-16 encodes a FOXO transcription factor that is a crucial downstream target that is negatively regulated by the DAF-2 pathway; daf-16 ( mu86 ) is a null mutation that causes a dauer defective ( Daf-d ) phenotype in larvae and a reduced stress resistance phenotype in adults [7] . Interestingly , Lee et al . ( 2003 ) used bioinformatic techniques to identify a putative DAF-16 binding site ( TTGTTTAC ) positioned 90 base pairs upstream of the predicted start codon of the natc-1 locus [11] ( Figure 6A ) . This predicted DAF-16 binding site is evolutionarily conserved in natc-1 homologues in Caenorhabditis briggsae and Drosophila melanogaster , suggesting it is functionally important [11] . We analyzed the genomic locus of human NAA35 , the homolog of C . elegans NATC-1 , and identified four predicted DAF-16 binding sites , consistent with the model that these binding sites might be conserved during evolution ( Figure S4 ) . Furthermore , Gerstein et al . ( 2010 ) and Riedel et al . ( 2013 ) used the method of chromatin immunoprecipitation followed by massively parallel DNA sequencing to demonstrate that DAF-16 protein interacts with the natc-1 locus in vivo [41] , [42] ( G . Ruvkun and C . Riedel , personal communication ) ( Figure 6A ) . Based on these observations , we hypothesized that natc-1 transcription is directly regulated by binding of the DAF-16 transcription factor . According to this model , altering DAF-16 activity is predicted to alter natc-1 mRNA levels . When cultured in standard laboratory conditions , worms display a low level of DAF-16 activity because the insulin/IGF-1 pathway is strongly activated [47] . Therefore , comparing daf-16 ( lf ) animals to control animals would not be highly informative . To test our hypothesis , we analyzed natc-1 mRNA levels in daf-2 ( e1370 ) mutant animals , since daf-2 mutant animals have been demonstrated to have increased DAF-16 nuclear localization and activity [48] . These daf-2 ( lf ) phenotypes are similar to the consequences of food deprivation or stress , suggesting daf-2 ( lf ) mutant worms have initiated a starvation or stress response [1] . daf-2 mutant animals displayed a ∼2-fold decrease in natc-1 mRNA levels compared to wild-type animals ( Figure 6B ) . To confirm that this change was statistically significant , we analyzed six independent biological replicates of both wild-type and daf-2 RNA . These data are consistent with the model that DAF-16 represses transcription of natc-1 , since DAF-16 activity is increased in daf-2 mutant animals . To directly test the function of daf-16 , we analyzed daf-16;daf-2 double mutant animals; the decrease in natc-1 mRNA levels was abrogated in these animals , demonstrating that daf-16 is necessary for the regulation of natc-1 ( Figure 6B ) . To confirm that daf-16 mutant animals do not contain daf-16 activity , we analyzed daf-16 transcript levels by qRT-PCR; daf-16 transcripts were detected in wild-type animals but were undetectable in the daf-16;daf-2 double mutant animals . These data suggest that DAF-16 is a transcriptional repressor of natc-1 and natc-1 is an effector of the insulin/IGF-1 signaling pathway that functions downstream of DAF-16 . The insulin/IGF-1 signaling pathway mediates entry into an alternative third larval stage called dauer that has distinctive metabolic and developmental features that promote longevity and stress resistance [3] , [8] , [49] . To test the function of natc-1 in insulin/IGF-1 signaling , we analyzed dauer larvae formation in natc-1 ( am138 ) animals . A single mutation in natc-1 did not cause a Daf-c phenotype ( Figure 6C ) . However , natc-1 ( am138 ) strongly enhanced dauer formation in the daf-2 ( e1370 ) background , compared to the daf-2 ( e1370 ) single mutant animals ( Figure 6C ) . To determine if the daf-2 ( e1370 ) ;natc-1 ( am138 ) Daf-c phenotype was daf-16 dependent , we analyzed daf-16 ( mu86 ) ;daf-2 ( e1370 ) ;natc-1 ( am138 ) triple mutant animals for dauer formation at 25°C . None of the daf-16;daf-2;natc-1 triple mutant animals displayed dauer formation ( N = 111 ) , indicating that daf-16 is required for this Daf-c phenotype . Together , these data demonstrated that natc-1 was necessary to inhibit dauer formation , although the effect was only observed in a sensitive genetic background . We hypothesized that the enhancement of the Daf-c phenotype caused by a mutation in the natc-1 auxiliary subunit reflects the reduction or loss of the acetylation activity of the NatC complex . To test this hypothesis , we analyzed the function of the predicted catalytic subunit of the NatC complex , B0238 . 10 , which we named natc-2 . We used the method of feeding RNAi to reduce natc-1 and natc-2 activity in a daf-2 ( e1370 ) mutant background . A significant increase in dauer formation was observed compared to control RNAi ( Figure 6D ) . These data indicate that the catalytic subunit encoded by natc-2 is necessary to inhibit dauer formation , suggesting that the acetylation activity of the NatC complex mediates insulin/IGF-1 signaling . Mutations in the insulin/IGF-1 receptor daf-2 cause increased longevity , while the natc-1 ( am138 ) mutation causes a shortened lifespan [50] . To further characterize the genetic interaction between natc-1 and daf-2 , we analyzed the lifespan of wild-type , natc-1 ( am138 ) , daf-2 ( e1370 ) , and daf-2 ( e1370 ) ;natc-1 ( am138 ) animals . While natc-1 ( am138 ) shortens wild-type lifespan , natc-1 ( am138 ) had no effect on the daf-2 ( e1370 ) longevity phenotype ( Figure S5 , Table 1 ) . This result suggests that daf-2 activity is necessary for the natc-1 ( am138 ) mutation to cause a reduction of lifespan . To characterize the role of natc-1 in stress resistance mediated by the insulin/IGF-1 signaling pathway , we analyzed interactions between natc-1 , daf-16 , and daf-2 in response to heat and high zinc stress . Single mutations of natc-1 and daf-2 cause resistance to heat stress , and daf-2 ( e1370 ) ;natc-1 ( am138 ) double mutant animals displayed enhanced stress resistance compared to either single mutant animal ( Figure 7A ) . One interpretation of this additivity is that natc-1 and daf-2 function in the same pathway , but neither single mutation maximizes the potential of the pathway to increase stress resistance; this is consistent with the fact that the daf-2 ( e1370 ) allele causes a partial loss-of-function . The alternative interpretation is that natc-1 and daf-2 function in parallel to mediate stress resistance . Compared to wild-type animals , daf-16 ( mu86 ) animals displayed a mild sensitivity to heat stress ( Figure 7B , Table 1 ) . daf-16 ( mu86 ) ;natc-1 ( am138 ) double mutant animals displayed heat stress resistance similar to natc-1 single mutant animals . ( Figure 7B , Table 1 ) . These data indicate that natc-1 is epistatic to daf-16 with respect to heat stress resistance , consistent with the model that natc-1 is a downstream effector that is negatively regulated by daf-16 . To further analyze this pathway , we determined if daf-16 was necessary for natc-1 ( am138 ) to cause resistance to zinc toxicity . Attempts to analyze daf-2 and daf-2;natc-1 mutant animals for resistance to high zinc toxicity were not successful , since supplemental zinc caused a high rate of dauer formation in these mutant animals , precluding an analysis of growth rates ( Figure S6 ) . natc-1 ( am138 ) caused similar resistance to high zinc toxicity in both a wild-type and daf-16 ( mu86 ) background ( Figure 7C ) , indicating that daf-16 function is not necessary for the natc-1 ( am138 ) zinc-resistance phenotype . These data support the model that natc-1 functions downstream of daf-16 to mediate zinc resistance , and together these results suggest that natc-1 is acting as a key downstream effector of the C . elegans insulin/IGF-1 signaling pathway ( Figure 8A , B ) .
The DAF-16 FOXO transcription factor is the key downstream target of the insulin/IGF-1 signaling pathway that promotes stress resistance and longevity [5] , [35] , [50]– . A critical goal in this field is to identify the functionally significant targets of DAF-16 , since these genes are hypothesized to mediate stress resistance , nutrient utilization , and aging . A variety of approaches have been used to identify DAF-16 target genes , including bioinformatic , genomic , and reverse genetic techniques [9]–[17] . Lee et al . ( 2003 ) used the biochemically demonstrated DNA binding sequence of DAF-16 to bioinformatically identify predicted binding sites in the C . elegans genome . One of these sites is positioned 90 base pairs upstream of the natc-1 start codon , suggesting that DAF-16 binds the natc-1 promoter . A similar FOXO transcription factor binding site is present in the promoters of genes homologous to natc-1 in Caenorhabditis briggsae , Drosophila melanogaster , and humans , suggesting that this transcription factor binding site has been conserved during evolution and is functionally significant [11] . Gerstein et al . ( 2010 ) and Riedel et al . ( 2013 ) used the technique of chromatin immunoprecipitation followed by DNA sequencing to identify in vivo binding sites for DAF-16 ( G . Ruvkun and C . Riedel , personal communication ) [41] , [42] . These studies independently demonstrated that DAF-16 binding is significantly enriched at the natc-1 locus , consistent with the hypothesis that DAF-16 occupies the predicted binding site . Here we demonstrated that natc-1 is transcriptionally repressed by DAF-16 . Worms cultured in standard laboratory conditions with abundant food have high activity of the DAF-2 insulin/IGF-1 receptor and low activity of DAF-16 . By contrast , daf-2 loss-of-function mutant animals have high activity of DAF-16 and display enhanced stress resistance and extended longevity . daf-2 ( lf ) mutant animals displayed reduced levels of natc-1 transcripts , and natc-1 transcripts were restored to WT levels in daf-16 ( lf ) ;daf-2 ( lf ) double mutant animals . These results suggest that daf-16 activity reduces the level of natc-1 transcripts , and daf-2 activity increases the level of natc-1 transcripts by negatively regulating daf-16 . Lee et al . ( 2003 ) used the technique of Northern blotting to examine natc-1 transcript levels and did not detect a substantial difference between wild-type and daf-2 ( e1370 ) mutant animals [11] . By contrast , Riedel et al . ( 2013 ) used the technique of high-throughput sequencing of mRNA ( mRNA-Seq ) [42] and detected a significant decrease ( ∼2-fold ) in natc-1 mRNA levels in daf-2 ( e1370 ) animals compared to wild-type animals ( G . Ruvkun and C . Riedel , personal communication ) . A possible explanation of these findings is that the techniques qRT-PCR and RNA-Seq , which gave similar results , are more quantitative than Northern blotting and were able to detect a small but statistically significant difference in transcript levels . Based on the presence of an evolutionarily conserved DAF-16 binding site in the natc-1 promoter , the in vivo binding of DAF-16 to the natc-1 locus , and our observed alterations in natc-1 transcript levels caused by mutations in daf-2 and daf-16 , we propose the model that DAF-16 is a direct transcriptional repressor of natc-1 ( Figure 8B ) . Our genetic analysis showed that natc-1 inhibits stress resistance and dauer formation and genetically interacts with the insulin/IGF-1 signaling pathway , suggesting that natc-1 is a physiologically relevant target of DAF-16 . These results do not exclude the possibility that natc-1 might be regulated by an additional pathway ( s ) in parallel to the insulin/IGF-1 pathway; indeed , the natc-1 promoter has been reported to interact with multiple transcription factors such as PQM-1 , SKN-1 , and PHA-4 , suggesting there are additional regulatory control mechanisms [41] , [53] . natc-1 mutant animals were discovered in an unbiased forward genetic screen for resistance to the stress of high dietary zinc . Further analysis revealed that natc-1 ( lf ) mutations cause resistance to a wide range of stressors , such as heat , oxidation , and multiple metals , similar to daf-2 ( lf ) mutations . daf-2 functions upstream of daf-16 in a signaling pathway , and daf-16 ( lf ) mutations are epistatic to daf-2 ( lf ) mutations . By contrast , our model that natc-1 functions downstream of daf-16 predicts that the stress resistance caused by natc-1 ( lf ) mutations is epistatic to daf-16 ( lf ) mutations . Indeed , we observed that the resistance to heat and high levels of dietary zinc caused by natc-1 ( lf ) mutations was epistatic to daf-16 ( lf ) mutations . Thus , both molecular and genetic studies support the model that natc-1 is directly targeted by DAF-16 to promote stress resistance . Consistent with this model , we demonstrated that natc-1 inhibits dauer formation , since natc-1 ( lf ) mutations enhanced the dauer constitutive phenotype of daf-2 ( lf ) mutations . Although a variety of DAF-16 target genes have been demonstrated to interact genetically with the insulin/IGF-1 signaling pathway , these target genes are primarily activated by daf-16 , and loss-of-function of these target genes impairs the stress resistance mediated by daf-16 activity . For example DAF-16 promotes transcription of the superoxide dismutase sod-3 [5] . sod-3 functions to promote resistance to oxidative stress [54] . Thus , sod-3 represents a class of DAF-16 targets that are transcriptionally activated by DAF-16 and promote stress resistance ( Figure 8A ) . To our knowledge , our results with natc-1 are the first demonstration of a direct DAF-16 target gene that is repressed by DAF-16 to promote stress resistance ( Figure 8A ) . Thus , these studies make a novel contribution to the understanding of how the activity of the insulin/IGF-1 signaling pathway mediates stress resistance . In addition to stress resistance , DAF-16 also promotes longevity [55] . Because mutations in natc-1 reduce longevity , DAF-16 likely regulates lifespan through a natc-1-independent mechanism . N-terminal acetyltransferases ( NATs ) are multi-subunit enzymes that catalyze the transfer of the acetyl group of acetyl coenzyme A to the α-amino group of the first amino acid of a target protein . N-terminal acetylation is a widespread modification that affects the majority of eukaryotic proteins; for example , ∼80–85% of human proteins are N-terminally acetylated [56] . Eukaryotes possess multiple NAT complexes ( NatA-NatE ) that vary in subunit composition and substrate specificity [57] . NAT activity is relevant to human health , since mutations in a human NAT gene are associated with Ogden syndrome , an X-linked disorder that is lethal in infancy [58] . Whereas the biochemical activity of NAT enzymes is well characterized , the functional roles of these enzymes are only beginning to be explored . The NatC complex is comprised of the catalytic subunit Naa30 and the auxiliary subunits Naa35 and Naa38 , and it acetylates proteins with the N-terminal sequences Met-Leu , Met-Phe , Met-Ile , and Met-Trp [39] , [40] , [44] , [59] , [60] . Genetic studies in yeast revealed NatC complex subunits are necessary for dsRNA virus particle assembly and WT growth rate in media lacking a fermentable carbon source [61] , [62] . The Arabidopsis thaliana Naa30 homolog was identified in a screen for photosynthesis altered mutants , and Naa30 mutants display decreased chloroplast density and slow growth [63] . The zebrafish Naa35 homolog is the embryonic growth-associated protein ( EGAP ) , and knock down studies with morpholinos suggest that it is necessary for WT growth and development [64] . The rat Naa35 homolog was discovered based on increased expression in healing corneal epithelium , and it is highly expressed in developing rat cornea and skin [65] . In human cells , reducing the levels of NatC complex subunits causes reduced cell viability and p53-dependent apoptosis [66] . These results suggest that the NatC complex functions to promote growth and development in a wide range of organisms . Our studies of C . elegans natc-1 make a unique contribution to understanding the biological function of NatC , since these results are the first molecular and genetic characterization of an animal with a NatC subunit mutation . Genetic studies demonstrated that mutations in natc-1 increased resistance to multiple environmental stressors including excess metal , heat , and oxidation . These results suggest that natc-1 activity reduces stress resistance . Environments with low food , high population density , and high temperatures promote formation of dauer larvae . Dauers are a stress resistant form that allows animals in unfavorable environmental conditions to suspend development , resist environmental stresses , and be prepared to resume reproductive development when conditions improve [1] . Dauer formation is mediated by the insulin/IGF-1 signaling pathway [1] . The capacity to form dauer larvae highlights the importance of maintaining a balance between promoting growth and reproduction and surviving environmental stressors . In a sensitized genetic background , natc-1 activity inhibited dauer formation . Because a developmental switch mediates the decision between dauer larvae and larvae destined for reproductive development , these findings indicate that natc-1 activity promotes a development fate characterized by growth and reproduction . These findings identify novel phenotypes associated with a subunit of the NatC complex . To determine how the NatC complex might execute these newly discovered functionalities , we bioinformatically identified predicted NatC target proteins in C . elegans . NatC might modulate processes such as stress resistance and dauer formation by acetylating groups of proteins with similar functionalities . To investigate this hypothesis , we performed a gene ontology ( GO ) analysis . Predicted NatC target proteins in C . elegans were enriched for 20 functional classes . To determine if NatC has an evolutionarily conserved preference for these functional classes , we used bioinformatics to predict NatC target proteins in humans . GO analyses of human NatC target proteins identified 11 functional classes that were enriched . Electron carrier and oxidoreductase activity were enriched in C . elegans and human GO analyses suggesting that the NatC complex may regulate a subset of proteins associated with these functionalities ( Table S1 ) . We further determined if predicted NatC target proteins were enriched for a specific cellular localization . GO analysis of NatC target proteins identified 4 cellular component terms enriched in C . elegans and 18 enriched in humans . NatC target proteins were enriched for mitochondrial localization in both C . elegans and humans ( Table S2 ) . These conserved functionalities and cellular components may inform future experimental efforts aimed at understanding how the NatC complex regulates physiology such as dauer entry and stress tolerance . To identify protein targets of the NatC complex that might contribute to the mutant phenotype by having altered acetylation in natc-1 ( lf ) mutant animals , we identified predicted protein targets that are implicated in zinc metabolism and stress resistance . Published genes implicated in C . elegans zinc metabolism include the cation diffusion facilitator ( CDF ) zinc transporters cdf-1 , sur-7 , cdf-2 , and ttm-1 [67]–[70] , the metallothioneins mtl-1 and mtl-2 [71] , and haly-1 [37] . Of these seven genes , only cdf-2 is a putative target of the NatC complex . Mutations in cdf-2 affect zinc accumulation [69] , whereas natc-1 ( lf ) mutations do not alter zinc accumulation , suggesting that cdf-2 is not a critical target of the NatC complex in zinc resistance . Rather , we hypothesize that natc-1 mutations cause high zinc resistance by triggering mechanisms that allow animals to ameliorate the toxicity of high levels of zinc . This hypothesis is supported by the observation that natc-1 mutations cause resistance to a wide variety of stressors . Genes implicated in stress resistance include those encoding proteins involved in reactive oxygen species metabolism and dauer formation . The NatC complex is predicted to target several proteins that met these criteria; sod-1 , sod-2 , and sod-3 encode superoxide dismutases that increase oxidative stress resistance [54] , mev-1 encodes cytochrome b , a subunit of the mitochondrial respiratory chain complex II , and frh-1 encodes a frataxin ortholog that promotes the oxidative stress response [72] , [73] . Several predicted protein targets are encoded by genes that influence dauer formation; tph-1 encodes a tryptophan hydroxylase , daf-36 encodes an oxygenase , and cyp-35a3 encodes one of 42 cytochrome P450 proteins predicted to be targeted by the NatC complex [17] , [74] , [75] . Previous studies of stress responses and dauer formation have largely focused on the importance of transcriptional regulation . Our work suggests posttranslational modifications like N-terminal acetylation might also play an important role , and future proteomic analyses may help identify key effectors of stress tolerance and dauer formation . Our findings are relevant to a general principle that organisms must balance growth and reproduction , which is facilitated by nutrient-rich environments , with stress resistance and quiescence , which are adaptive in nutrient poor and/or high stress environments . Plants that evolved in low-resource , stressful environments share a common set of traits , including relatively low rates of growth , photosynthesis , tissue turnover , and nutrient absorption [76]–[78] . This has been named the “stress resistance syndrome” ( SRS ) , since these plants are resistant to a wide spectrum of physiologic stressors [79] . SRS may represent an adaptive strategy for coping with harsh environmental conditions [76] , and this process has interesting analogies to dauer formation in C . elegans . Both dauer formation and SRS represent organisms balancing growth and stress tolerance to promote survival . The NatC complex may have an evolutionarily conserved role in mediating this balance between growth and stress tolerance . Consistent with this hypothesis , loss-of-function of the catalytic subunit of the Arabidopsis NatC complex causes decreased photosynthetic activity [63] , an SRS trait . Additionally , our genetic studies demonstrate that loss-of-function of natc-1 promotes dauer formation and inhibits reproductive development , a trait analogous to SRS . Therefore , we propose that the NatC complex may have evolutionarily conserved functions in maintaining the balance between promoting growth and reproduction and resisting stressful environmental conditions . Given that NatC functions to mediate the balance between growth and stress resistance , which is finely tuned by environmental conditions , it is important to determine how the activity of NatC is regulated . However , little is known about the regulation of these enzymes . In yeast , the natc-1 homolog ( MAK10 ) protein levels are glucose repressible , but it was not established whether regulation occurs at the level of RNA or protein [61] . Here we demonstrated that natc-1 is negatively regulated at the level of transcription by DAF-16 . The insulin/IGF-1 signaling pathway responds to environmental cues , such as nutrient availability , temperature , and dauer pheromone , by regulating the activity of DAF-16 . Therefore , our findings establish a direct link between environmental sensing mediated by the insulin/IGF-1 signaling pathway and protein N-terminal acetylation mediated by NatC . These results make a new contribution to understanding NatC regulation in several ways . First , NAT subunits have not previously been reported to be regulated at the level of transcription . Second , this is a novel demonstration that a NAT complex is regulated by the insulin/IGF-1 signaling pathway , linking an environmental sensing pathway to the regulation of protein N-terminal acetylation ( Figure 8A , B ) . NatC complexes appear to have an evolutionarily conserved role in modulating growth and stress resistance , and our findings suggest that they may also have a conserved role in responding to the insulin/IGF-1 signaling pathway . We have molecularly characterized two genes identified by screening for mutant animals that display increased tolerance to excess dietary zinc [36]: natc-1 and haly-1 [37] . These are the only genes that have been demonstrated to cause resistance to excess dietary zinc in an animal , and mutations in these two genes appear to act by very different mechanisms . First , these genes encode proteins with distinct functions . haly-1 encodes an enzyme that metabolizes histidine , and haly-1 mutant animals display increased levels of histidine . natc-1 encodes a subunit of a protein N-terminal acetylation complex , suggesting that protein N-terminal acetylation of many proteins is altered in these mutant animals , although this prediction has not been biochemically tested . Second , the spectrum of stress resistance caused by mutations in these two genes is distinct . Increased histidine appears to chelate and detoxify excess zinc and nickel , but the effect is quite specific since haly-1 mutant animals are not resistant to the toxicity caused by other metals [37] . By contrast , here we demonstrated that natc-1 mutations cause broad-spectrum stress resistance , including resistance to multiple metals , heat , and oxidation . Consistent with the model that these genes function by distinct mechanisms , the resistance to excess zinc toxicity caused by mutations of natc-1 and haly-1 was additive . These findings raise a general question about stress resistance; how does a mutation in a single gene such as natc-1 promote resistance to a broad-spectrum of stresses ? ( 1 ) One possibility is that the single-gene mutation results in a cascade of events that changes the activity of many proteins in the cell . In this model , each specific change in activity might mediate resistance to only one or a small number of stressors . For example , changes in haly-1 activity only mediate resistance to zinc and nickel . However , the combination of many different changes in activity could mediate broad-spectrum resistance . This is an attractive model for daf-2 mutant animals , which display broad-spectrum stress resistance and are documented to have changes in the expression of many genes as a result of the regulation of the DAF-16 transcription factor . This model is also attractive for natc-1 , since this enzyme is predicted to mediate the N-terminal acetylation of many different proteins . ( 2 ) An alternative model is that diverse environmental stresses converge on a single type of important molecular damage . For example , heat , oxidation , and excess metals may all cause toxicity as a result of similar damage , such as protein unfolding . In this model , changing the activity of a single gene might confer broad-spectrum stress resistance by enhancing tolerance to the major form of cellular damage . These two basic models represent extremes of a continuum , and are not mutually exclusive . Our results document that resistance to high zinc toxicity can be increased by mutations that cause specific or broad-spectrum stress resistance , contributing to a conceptual framework for understanding stress resistance .
C . elegans strains were cultured at 20°C on nematode growth medium ( NGM ) seeded with E . coli OP50 unless otherwise noted [80] . The wild-type C . elegans strain and parent of all mutant strains was Bristol N2 . The following mutations were used: daf-16 ( mu86 ) [7] is a strong loss-of-function or null mutation of the DAF-16 forkhead transcription factor; daf-2 ( e1370 ) [6] is a partial loss-of-function mutation of the DAF-2 insulin/IGF-1 receptor; haly-1 ( am132 ) [37] is a strong loss-of-function or null mutation of the HALY-1 histidine ammonia lyase . natc-1 ( am134 ) and natc-1 ( am138 ) were identified in a genetic screen for resistance to high zinc toxicity [36] , backcrossed four times to wild type , and are described here; natc-1 ( ok2062 ) was obtained from the C . elegans knockout consortium [43] and backcrossed four times to wild type . The back crossing procedure replaced ∼94% of the genome of mutant strains with wild-type DNA that has not been exposed to chemical mutagenesis , minimizing background mutations . Double mutant animals were generated by standard methods , and genotypes were confirmed by PCR or DNA sequencing . Hermaphrodites were cultured on NGM , and one embryo was transferred to a 35×10 mm Petri dish containing NAMM [36] supplemented with zinc sulfate ( ZnSO4 ) , cadmium chloride ( CdCl2 ) , nickel chloride ( NiCl2 ) , or copper chloride ( CuCl2 ) and 5× concentrated E . coli OP50 as a food source . Dishes were analyzed daily until progeny were observed or the animal died , except Figure S3 where dishes were analyzed only until day 6 . Animals that generated one or more live progeny were scored as “fertile adults . ” To determine the metal concentration for these assays , we generated dose response curves of fertility for each metal using wild-type animals and selected the concentration that caused ∼50% of wild-type animals to fail to display fertility ( Figure 4 ) . Large populations of animals were cultured on NAMM supplemented with 0 or 200 µM zinc sulfate ( ZnSO4 ) . The animals were desiccated to determine dry weight , and total zinc content was determined by ICP-MS as described by Murphy et al . ( 2011 ) [37] . Live transgenic animals were immobilized using levamisole in phosphate buffered saline ( PBS ) and mounted onto a thin pad of ∼7 . 5% agarose . More than 100 transgenic animals were analyzed , and representative images are presented . All images were captured on a PerkinElmer spinning disk confocal microscope utilizing Volocity imaging software . Gravid adult hermaphrodites were bleached to obtain embryos . Embryos were allowed to hatch in M9 buffer to synchronize at the L1 stage and cultured at 15°C on NGM . For heat stress assays , animals were shifted to 35°C on day 1 of adulthood . Animals were analyzed hourly for spontaneous or provoked motility and pharyngeal pumping; animals displaying none of these traits were scored as dead . Animals were briefly exposed to room temperature ( 24–25°C ) for scoring . For the experiment shown in Figure 7A , animals were cultured continually at 35°C until hourly scoring began at 12 hours; summary statistics were not calculated in this case because some data points were not collected . For oxidative stress assays , day 3 adults were transferred to NGM dishes supplemented with 40 mM methyl viologen dichloride hydrate ( paraquat ) , fed E . coli OP50 and cultured at 20°C . We analyzed day 3 adults to avoid the high frequency of matricidal hatching in response to oxidative stress displayed by younger adults . Animals were scored every 12 hours for survival . For lifespan assays , L4 animals were cultured on NGM at 20°C ( defined as day 0 ) and fed E . coli OP50 . Adult hermaphrodites were transferred to fresh Petri dishes every day until the cessation of progeny production and analyzed every day for survival . In heat stress , oxidative stress and lifespan assays , animals that displayed matricidal hatching or a vulval-bursting phenotype were omitted from the analysis . To analyze dauer formation , we transferred embryos to NGM with E . coli OP50 at 15–25°C until adult animals began to lay embryos , approximately 3–5 days depending on the temperature . Animals were scored as dauer or non-dauer using a dissecting microscope based on the visible radial constriction phenotype [81] . To determine the effect of zinc , we conducted this assay on NAMM supplemented with zinc sulfate ( ZnSO4 ) . To analyze genetic regulation of dauer formation , we performed feeding RNAi as described by Kamath et al . ( 2001 ) with minor modifications [82] . Briefly , daf-2 ( e1370 ) P0 hermaphrodites and F1 progeny were incubated at 20°C and continuously fed RNAi expressing bacteria . F1 progeny were scored as dauer or non-dauer after approximately 4 days . We used the empty vector control ( L4440 ) and clones encoding dsRNA corresponding to T23B12 . 4 ( natc-1 ) and B0238 . 10 ( natc-2 ) from the Ahringer RNAi Library [83] . The DNA sequence of each clone was confirmed by standard methods . Plasmid pJM5 is pBlueScript SK+ ( Stratagene ) containing a 3 , 356 base pair fragment of C . elegans genomic DNA from fosmid WRM067bF02 that extends 139 base pairs upstream of the predicted natc-1 start codon and 395 base pairs downstream of the predicted stop codon . To generate pJM6 , we modified pJM5 by digestion with BstEII ( New England Biolabs ) and religation resulting in a 382 base pair deletion that removes parts of exons 2 and 3 and all of intron 2 . The resulting pJM6 mutant open reading frame is predicted to truncate at amino acid 33 in a premature stop codon ( TAG ) . To generate the Pnatc-1::NATC-1::GFP::unc-54 3′UTR translational fusion protein construct ( pJM8 ) , we inserted the natc-1 genomic locus ( without the stop codon ) into pBlueScript SK+ with the GFP coding sequence and the unc-54 3′ UTR . The DNA sequence of each plasmid was confirmed by standard methods . Transgenic animals were generated by injecting natc-1 ( am134 ) hermaphrodites with pJM5 or pJM6 , and natc-1 ( am138 ) hermaphrodites with pJM8 . All injections were done with the dominant Rol marker pRF4 [84] . We selected independently derived Rol self progeny that transmitted the Rol phenotype . These transgenes formed extrachromosomal arrays , since the Rol phenotype was transmitted to only a sub-set of the self-progeny . To analyze transgenic rescue of the natc-1 ( am134 ) or natc-1 ( am138 ) resistance to high zinc toxicity phenotype , we calculated the fraction of transgenic animals on baseline and high zinc concentrations . nonRol animals were presumed to lack the extrachromosomal array and were thus non-transgenic . We defined rescue as a significant decrease in percentage of transgenic animals able to survive ( and thus be quantified ) on toxic zinc conditions ( 300 µM supplemental zinc ) compared to the baseline zinc concentration ( 0 µM supplemental zinc ) , as described by Murphy et al . ( 2011 ) [37] . To identify protein targets of the NatC complex , we wrote a custom Perl script that computationally identified predicted NatC targets in C . elegans and human protein databases from the National Center for Biotechnology Information ( NCBI ) . Gene ontology ( GO ) analysis was performed using GOrilla [85] by comparing predicted NatC targets to the entire proteome for each species . We report significant GO functional terms ( p<0 . 001 ) according to Eden et al . ( 2009 ) [85] . Three natc-1 cDNA clones ( EST ) were obtained from the National Institutes of Genetics , Japan ( yk194g4 , yk262c3 , and yk420a1 ) . We determined the complete sequence of these cDNAs using standard techniques . These data were used to infer the mRNA sequence from exon 3 to the polyA tail attached 330 nucleotides downstream of the TGA stop codon . To experimentally define the 5′ end of the natc-1 mRNA , we used 5′ RACE System V2 . 0 ( Invitrogen ) according to the manufacturer's instructions . These data were used to infer the natc-1 mRNA sequence from the 22 nucleotide splice leader 1 ( SL1 ) sequence that begins 14 base pairs upstream of the start codon to exon 3 . To generate synchronous populations of worms for RNA extraction , we treated gravid adult hermaphrodite animals with a mixture of bleach and 4M sodium hydroxide ( NaOH ) and cultured embryos overnight in M9 solution at 20°C , resulting in L1 stage arrest . L1 larvae were transferred to NGM plates at 20°C , fed E . coli OP50 , and allowed to develop to the L4 stage ( approximately 2 days ) . RNA isolation and cDNA synthesis were performed as described by Davis et al . ( 2009 ) [86] . Quantitative , real-time PCR was performed using an Applied Biosystems 7900HT Fast Real-Time PCR system and the Applied Biosystems SYBR Green Master Mix . mRNA fold change was calculated using the comparative CT method [87] . Forward and reverse amplification primers were: rps-23 5′- aaggctcacattggaactcg and 5′- aggctgcttagcttcgacac; mtl-1 5′-ggcttgcaagtgtgactgc and 5′-cctcacagcagtacttctcac; natc-1 5′-tcagctttacgggtccaatg and 5′-ccgaaaatgctctgtggttac; daf-16 5′- gacggaaggcttaaactcaatg and 5′- gagacagattgtgacggatcg . All data were analyzed utilizing the two-tailed students t-test of samples with unequal variance unless otherwise specified . For binary data such as dauer entry and fertility , the Chi-squared test was utilized . P-values less than 0 . 05 were considered statistically significant .
|
What are the mechanisms used by animals to cope with stressful environments that inflict damage or restrict essential processes such as growth , development , and reproduction ? One strategy is changes in physiology that increase stress resistance , and an extreme version of this strategy is diapause , an alternative developmental state that is enduring and stress resistant . In the nematode C . elegans , stress tolerance and entry into a diapause state called dauer larvae are mediated by the conserved insulin/IGF-1 pathway . Specifically , the FOXO transcription factor DAF-16 promotes stress tolerance and dauer larvae development . However , the targets of DAF-16 that mediate these processes remain largely elusive . Using an unbiased forward genetic screen to discover new mediators of stress tolerance , we identified natc-1 , a novel target of DAF-16 and the insulin/IGF-1 pathway . natc-1 encodes a conserved subunit of the N-terminal acetyltransferase C ( NAT ) complex . The NatC complex modifies target proteins by acetylating the N-terminus . We demonstrated that natc-1 mediates diapause entry and stress tolerance . Furthermore , we elucidated regulation of NatC by demonstrating that natc-1 is a direct transcriptional target that is repressed by DAF-16 . These findings may be relevant to other animals because both the insulin/IGF-1 signaling pathway and the NAT system are conserved during evolution .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"invertebrates",
"animal",
"genetics",
"caenorhabditis",
"gene",
"regulation",
"animals",
"dna",
"transcription",
"gene",
"function",
"animal",
"models",
"mutation",
"caenorhabditis",
"elegans",
"model",
"organisms",
"molecular",
"genetics",
"research",
"and",
"analysis",
"methods",
"gene",
"expression",
"genetic",
"screens",
"gene",
"identification",
"and",
"analysis",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"organisms"
] |
2014
|
The DAF-16 FOXO Transcription Factor Regulates natc-1 to Modulate Stress Resistance in Caenorhabditis elegans, Linking Insulin/IGF-1 Signaling to Protein N-Terminal Acetylation
|
Phage-encoded serine integrases mediate directionally regulated site-specific recombination between short attP and attB DNA sites without host factor requirements . These features make them attractive for genome engineering and synthetic genetics , although the basis for DNA site selection is poorly understood . Here we show that attP selection is determined through multiple proofreading steps that reject non-attP substrates , and that discrimination of attP and attB involves two critical site features: the outermost 5–6 base pairs of attP that are required for Int binding and recombination but antagonize attB function , and the “discriminators” at positions −15/+15 that determine attB identity but also antagonize attP function . Thus , although the attachment sites differ in length and sequence , only two base changes are needed to convert attP to attL , and just two more from attL to attB . The opposing effect of site identifiers ensures that site schizophrenia with dual identities does not occur .
Establishment of lysogeny by temperate bacteriophages typically involves site-specific integration of the phage genome into the host chromosome . Integration is catalyzed by a phage-encoded Integrase protein ( Int ) mediating site-specific recombination between phage and bacterial attachment sites ( attP and attB respectively ) , and generates attachment site junctions ( attL and attR ) as products of the reaction ( Figure 1A ) [1] . There are two major classes of phage integrases – corresponding to the tyrosine- and serine-recombinase families – that use distinct mechanisms of strand exchange and have different site and protein requirements [2] . The tyrosine integrases typically utilize a relatively large attP site ( ∼250 bp ) containing multiple binding sites for integrase , a host-encoded integration host factor , and a recombination directionality factor ( RDF ) that binds and bends DNA to confer directionality of recombination [3] , [4] . In contrast , serine-integrases use simple attachment sites ( <50 bp ) , have no host factor requirements , and the RDF does not act through direct binding to DNA [5]–[7] . Because of these features , serine-integrases function well in heterologous systems , making them attractive for genome engineering in human , mouse , drosophila , and malarial cells [8]–[11] , as well as powerful switches for synthetic genetic circuits and microbial data storage systems [12] , [13] . The serine-integrase ( Int ) coded by mycobacteriophage Bxb1 is relatively large ( 500 aa ) and contains an N-terminal catalytic domain ( ∼150 aa ) common to all serine-recombinases , and a C-terminal domain ( CTD; 350 aa ) that binds DNA [14] , [15] . The attP and attB sites are small and have different length requirements , 48 bp for attP and 38 bp for attB . Strand cleavage occurs about an asymmetric central dinucleotide within a protein-mediated synaptic complex , followed by rotation and religation [16] . Recombination is highly selective for the cognate attachment sites , attP and attB for integration , and attL and attR for excision , and is strongly directional , such that excision only occurs in the presence of the recombination directionality factor ( RDF ) , Bxb1 gp47 ( Figure 1A ) [7] . attP and attB are functionally symmetrical such that the central dinucleotide is the sole determinant of integration polarity [15] , and the sequences of both sites are partially symmetric , although outside of an 8 bp common core there is only limited sequence similarity ( Figure 1B ) . Bxb1 Int binds as a dimer to attP and attB with similar affinities ( Kd 70 nM ) , but somewhat tighter to attL and attR ( Kd 15 nM ) ; CTD binds as a monomer to each half site with similar affinities for the B , B′ and P half sites ( ∼120 nM ) , and a somewhat lower affinity for the P′ half site [17] . These general features are shared by other well-studied serine integrase systems [5] , [18]–[24] . Selection of cognate sites that support recombination involves multiple steps in the reaction [25] , [26] . DNA binding is required but is not sufficient , and in the absence of the RDF , synapsis only occurs between Int dimers bound at attP and attB [15] , [27] . Because synapsis is protein-mediated , Int presumably adopts different configurations when bound at different att sites with synapsis requiring compatible configurations [17] , [27] . However , synapsis of attL and attR in the presence of the RDF is orientation dependent , suggesting that an Int protomer bound at a B-type half site ( either B or B′ ) can only productively interact with a P-type half site ( P or P′ ) [26] . Substitutions in the attB site of φC31 show that specific DNA sequences are also important for post-synaptic events [25] . Here , we investigate what specific sequences are required for Bxb1 Int to recognize its attP site and to functionally distinguish attP and attB . We show that there are two critical site components . One is the outermost flanks of attP that are required for Int binding and for recombination , but which also prevent usage as attB . The second is a key discriminator position at positions −15 and +15 where a T∶A/A∶T base pair ( in B and B′ half sites respectively ) is required for both Int binding and recombination as an attB site , but which interferes with attP functionality . The identities of attP and attB are mutually exclusive , but they can be interconverted with mutations in the critical discriminator and flanking motifs .
To determine the sequence contributions of Int binding to attP , we initially constructed a series of altered half-site substrates with transition mutations at each of 23 positions within the P half site , and examined the binding of the Bxb1 Integrase CTD ( Figure 1C , Figure S1 , Table 1 ) . Substitutions at four positions ( −19 , −20 , −21 , and −23 ) are strongly deleterious to CTD binding and no complex is observed even at the highest protein concentrations tested ( Figure 1D , Figure S1A ) . These positions are all at the extreme flank of the P half site and – with the exception of position −19 – are outside of the corresponding minimal site requirements for attB ( Figure 1B ) . They also are all symmetrically conserved between the P and P′ half sites ( Figure 1B ) . Substitutions at most of the other positions in the P site also impair CTD binding but to lesser extents ( Figure 1C , 1D , Figure S1A , Table 1 ) . A summary of all mutant site activities is shown in Table 2 . We extended this analysis to determine how Int CTD binds to full site attP substrates containing P site substitutions ( Figure 1E , 1F ) . CTD binding to wild-type attP has a somewhat unusual pattern in that two complexes are formed ( complex 1 and complex 2; Figure 1E ) whose identities are not clear , and it is presumed that the slower migrating complex ( complex 2 ) contains CTD protomers bound to both half sites , and that complex 1 has only a single CTD protomer . However , higher concentrations of CTD do not drive attP DNA from complex 1 into complex 2 unless there is a nick at the center of the site [17] , suggesting that DNA rigidity contributes to interference between CTD protomers binding to both half sites . In general , the impact of P-substitutions on CTD binding to these substrates reflects those seen with half-site DNAs ( Figure 1E , 1F , Figure S1B ) although the −19 , −20 , −21 and −23 substitutions have a more modest impact suggesting that binding of CTD to the P′ half site can stimulate CTD with mildly cooperative binding to the P half site . Full length Int binds cooperatively as a dimer to attP forming a single complex ( Figure 1G , 1H , Figure S1C ) , and substitutions in the P component generally have only mildly reduced binding , including the flank positions that strongly impair CTD binding . No single base substitution reduces Int binding by more than about 10-fold ( Figure 1G , 1H , Figure S1C , Table 1 ) . A similar series of binding experiments were performed with substitutions in the P′ arm ( Figure 2 ) . The cognate mutations generally have similar effects on binding to a P′ half site as to the P half site , although the binding to the wild-type site is relatively weak and determining affinities is more difficult ( Figure 2A ) . Substitutions at positions +23 and +21 are the most deleterious to binding , with lesser effects by other mutations . In the context of the full attP site , the substitutions primarily influence the formation of complex 2 by CTD ( Figure S2A ) , and none of the mutants tested has a substantial impact on Int binding ( Figure 2B , Figure S2B , Table S1 ) . Finally , we examined the impact of double substitutions at symmetrically related positions in both half sites ( Figure 2C , 2D , Table 1 ) . Double substitutions at positions −23/+23 , −21/+21 , −20/+20 , −19/+19 strongly interfere with CTD binding ( Figure 2C ) , and substantially reduced Int binding ( Figure 2D ) . Some of the double mutants – such as −23/+23 have poor CTD binding – but Int itself binds reasonably well . Overall , these binding data illustrate the important roles of the extreme flanking sequences for recognition of attP by Int , and the important but lesser contributions at a large number of positions in the inner part of the site . We surveyed all of the P-mutants ( in the attP context ) for their ability to support integrative recombination ( Figure 3A ) , and analyzed subsets of these as well as P′ mutants and double mutants in further detail ( Figure 3B , 3C , 3D , Figure S3 ) . In general , most of the single substitutions in the attP flanks ( −23 , −21 , −20 , −19 , +19 , +20 , +21 , +23 ) are deleterious for recombination , even though Int binding to most of these substrates is only mildly affected ( Figure 3B , 3C ) . Similarly , single substitutions at −4 and +4 also impair recombination ( Figure 3B , 3C ) , even though Int binds reasonably well ( Figure 1H , Figure 2B , Table 1 , Table S1 ) . Double mutants that strongly interfere with Int binding ( e . g . −21/+21 , −20/+21 , −19/+19 ) not surprisingly are strongly defective in recombination ( Figure 3D , Figure S3C ) . The poor recombination of the −4/+4 double mutant reflects the behaviors of the single substitutions at position 4 , and Int binds reasonably well to the double mutant ( Figure 3D , Figure 2D ) . These observations show that the attP sequence influences not only Int binding , but is also important for subsequent steps in the reaction , either synapsis or post-synaptic events . Moreover , there are two distinct types of effect: the attP flank sequences that are required for CTD recognition but are also important for recombination ( although we cannot rule out that the recombination defect is largely a consequence of poor Int binding ) , and the −4/+4 positions that have a modest contribution to CTD binding , but are critical for recombination . A summary of mutant site activities is shown in Table 2 . For those substrates to which Int binding is observed but recombination is impaired , the defect could be at the requirement for synaptic complex formation between attP and attB , or a post-synaptic event involving strand cleavage , rotation or rejoining . To examine this , we tested mutant substrates for their ability to form synaptic complexes with an attB suicide substrate ( Figure 4 ) [7] . Substrates with single mutations in either P or P′ that support Int binding at reasonable levels ( Figure 1 , Figure 2 , Table 1 , Table S1 ) generally show good synaptic complex formation , with milder defects in attP flank mutants as well at the −4 and +4 positions ( Figure 4A , 4B , Figure S4 ) . In general , mutants with mild defects in synaptic complex formation ( T-21C , T-20C , G-4A , A+21G , A+20G , C+19T , C+4T ) are also strongly defective in recombination , even though Int binds reasonably well to most of these substrates ( Figure 4A , 4B , 4C , Figure S4 ) . But even the A+21G substrate – to which Int binds normally ( Figure 2B ) – forms good synaptic complexes at high Int concentrations ( Figure 4B ) , even though recombination is strongly impaired ( Figure 3C ) . Among the double mutants , the substitutions at positions 21 , 20 and 19 fail to form synaptic complexes ( Figure 4A , 4B , 4C ) but this reflects the strong defects in Int binding . Extended incubation promotes synapsis for the −/+19 mutant ( Figure 4C ) . In contrast , the strong recombination defect of the −4/+4 mutant appears to result from strong inhibition of synapsis . Cleavage assays show that mutants with single substitutions at the +21 , +20 and +19 positions are strongly defective in cleavage ( Figure 4D ) , even though they can form synaptic complexes – albeit inefficiently . In contrast , other single substitutions – primarily in the P half of attP – appear to undergo cleavage reasonably well . The difference between the cleavage capacity of P and P′ mutants could reflect the asymmetry of the attB suicide substrate ( in which only the top strand contains a gap ) and only cleavage of the bottom strand is required to generate a dsDNA cleaved product ( Figure 4D ) . These data are thus consistent with the interpretation that the Int protomer bound to the P′ half site is specifically responsible for cleavage of the bottom strand . Nonetheless , these observations show that single base substitutions ( such as T-21C ) can inhibit post-cleavage events in the reaction , such as rotation or rejoining ( Figure 4D ) . A summary of all mutant site activities is shown in Table 2 . Taken together , these observations show that there are multiple stages in the integration reaction where the sequence of attP influences recombination . These can be thought of as a series of proofreading events in which the site sequence is interpreted for whether it is permissive for recombination . In the initial binding stage for example , the T-21C/A+21G mutant is strongly defective in binding and recombination does not occur . At the next step of synapsis , a mutant such as G-4A/C+4T is bound reasonably well by Int ( Figure 2D ) , but this mutant is rejected for synapsis ( Figure 4C ) . But even if a mutant such as T-21C is bound by Int , synapses with attB and undergoes cleavage , it is impaired for rotation or religation . This is consistent with a model in which site-selection involves the formation of specific conformations of protein-DNA complexes , and inappropriate conformations prevent not only synapsis but also post-synaptic events . The experiments described above identify the roles of specific base pairs in attP that enable it to recombine with attB . The sequences at the extreme flanks of attP play critical roles in both Int binding and recombination , but it is unclear to what extent these contribute to attP identity . Specifically , mutations at positions −20 , −21 , and −23 strongly interfere with CTD binding to a half site substrate , although these are outside of the minimal length of an attB substrate ( Figure 1B ) . So although CTD binds well to a B half site substrate [17] it does not recognize these P mutants as though they are B-type sites . Furthermore , we note that Int binds quite well to single mutants such as T-21C but is poor at recombination , so a plausible explanation is that the conformation of the Int promoter bound at the mutant half site has adopted the conformation as if it were bound to a B-type site , effectively converting the mutant attP site into attL . Nonetheless , the finding that such single mutant sites can synapse with attB ( Figure 4 ) – which attL is not able to do – argues strongly against that . Closer examination of the similarity of the P and B half site sequences show that 13 of the 18 positions are conserved , with differences at positions −5 , −8 , −11 , −15 , and −18 ( Figure 5A ) . With the exception of −18 , all of these are in symmetrically conserved positions in attB ( Figure 5A ) and are thus candidates for playing roles in determining the identities of attP and attB , perhaps explaining the failure of CTD to bind to the attP flank mutants as though it were a B-type site . To address this , we first determined the impact of single substitutions in these conserved positions of B half site substrates ( Figure 5B ) . The only position with strong inhibition of CTD binding is the position at −15 , showing that this is critical for B-type site recognition . We note that the cognate position in attP is not symmetrically conserved and is a 5′-GC ( top strand-bottom strand ) base pair at both −15 and +15 ( Figure 5A ) . Transition mutations in attP at these positions have little impact on binding of either CTD or Int , or on recombination ( Figure 1 , Figure 2 , Figure 3 ) . To define the elements determining site identity , we constructed two hybrid sites ( Hybrid-1 and Hybrid-2; Figure 5A ) . Both contain the inner part of attB onto which is added differing lengths of the attP flanks; Hybrid-1 and Hybrid-2 have attP sequences from −15/+15 and −18/+18 to the ends , respectively ( Figure 5A ) . Int binds remarkably well to both of these hybrid substrates , with affinities of Kd = 13 nM and 7 nM respectively ( Figure 5C ) , similar to binding of Int to attL and attR , and 4–5 times better than to either attP or attB [17] . Hybrid-2 retains its ability to recombine as an attB substrate – although with somewhat reduced efficiency ( Figure 5D , 5E ) – but fails to act as an attP site . The extreme attP flanking sequences thus appear to impair attB function , but incompletely . In contrast , Hybrid-1 has completely lost its attB identity , but interestingly has gained attP identity , recombining with attB albeit inefficiently ( Figure 5D , 5E ) . Hybrid-1 and Hybrid-2 differ by only four bases ( −15 , +15 , +16 , +17; Figure 5A ) and these must then encompass the critical discriminatory positions . Positions 16 and 17 are not symmetrically-related , but are shared between the B and P half sites ( Figure 5A ) so we constructed two additional substrates; Hybrid-3 adds G+16A/T+17G to Hybrid-1 symmetrizing them with their counterparts in the P and B sites , and Hybrid-4 also symmetrizes the position at +15 ( i . e . G+15C ) . Both hybrids are good Int binding sites ( Figure 5C , Kd = 15 nM and 10 nM respectively ) and both function as attP substrates with Hybrid-4 having near wild-type levels of activity; neither functions as an attB site . These observations suggest that the −15 and +15 positions are discriminator bases playing critical roles in site identity . We therefore tested whether addition of a G-15T substitution ( introducing the B-type base pair ) to a half-site attP substrate containing a T-21C substitution ( to which CTD fails to bind; Figure 1C ) would restore CTD binding ( Figure 5F ) . We do observe CTD binding to this substrate , although weakly , and a substrate with the same two mutations in both attP half sites ( Hybrid-5; T-21C/G-15T/G+15A/A+21G ) behaves similarly ( Figure 5F ) . However , if the two P-site mutations ( T-21C/G-15T ) are in a full attP context ( i . e . with a wild-type P′ site; Hybrid-6 ) , then CTD binds well with efficient formation of complex 2 ( Figure 5F ) . If these two mutations restore a B-type interaction then Hybrid-6 should act as an attL-like substrate . We observe that both CTD and Int ( Int Kd = 10 nM ) bind to Hybrid-6 with similar patterns to attL ( Figure 5F ) , and Hybrid-6 is functionally indistinguishable from attL for recombination ( Figure 5G ) ; it does not function as either attP or attB . The full Int protein binds slightly less well to Hybrid-5 ( Kd = 120 nM ) but Hybrid-5 has acquired the ability to function as an attB site , albeit inefficiently , and lost the ability to function as attP ( Figure 5D , 5E ) . These experiments illustrate the critical roles in the flanking sequences and the −15/+15 base pairs in site identity . Finally , we constructed two sites that are derivatives of attP with G-15A/G+15A and G-15T/G+15A mutations , but with wild-type attP flanking sequences ( Hybrid-7 and Hybrid-8 respectively ) . Int binds well to both substrates ( Kd = ∼10 nM for both ) , but neither function as attB , and both work only poorly as attP , with Hybrid-8 working substantially worse than Hybrid-7 ( Figure 5D , 5E ) . These behaviors are consistent with the interpretation that not only is the T∶A/A∶T ( at B and B′ half site respectively ) base pair required for attB identity , but that it also antagonizes attP identity . Likewise , the inability of Hybrid-8 to act as an attB site suggests that the attP flanking sequence also antagonizes attB identity .
Phage-encoded serine integrases show a remarkable selectivity for suitable recombination partner DNAs . This selectivity is inherently related to the biological requirement that these site-specific recombination systems have strong directional control , such that integration and excision do not occur under undesirable circumstances . One consequence of this is that the system must strongly discriminate , for example , between attP and the attachment junctions attL and attR , each of which differs from attP by one B-type half site . Because synapsis is a requirement for strand cleavage and is protein-mediated , we assume that different conformations of protein-DNA complexes are the ultimate determinants of site selection . The analysis of attP mutants described here provides further support for this model , but also reveals that the attP sequence plays a role in controlling post-synaptic events . Previous analysis showed that the ability to form synaptic complexes is a critical stage in site-selection , although this was based on testing sites to which Int binds but which have substantial sequence differences . The more subtle changes of point mutations show that a block to synapsis can still be observed , such as with the −4/+4 mutant , but that most of the other mutants tested are competent to synapse , even though they may be defective for recombination ( Figure 6A ) . Although we would have predicted that such mutants would be blocked in cleavage , this does not appear to be the case , and at least for single mutations in the P site , cleavage can still occur . The attP sequence thus plays an important role in controlling activity , from Int binding through to post-cleavage events ( Figure 6A ) . This mirrors the role of the attB sequence in φC31 integration , where mutations interfere with Int binding or synapsis , but also block DNA cleavage [25] . In general , the requirement for satisfying multiple different reactions stages is akin to going through multiple security checks at an airport , needing to pass each one of them before being permitted to board the plane . The architectures of the Bxb1 attP and attB sites reflect three types of components ( Figure 6B ) . The first , is the inner part , which we define as encompassing the 28 bp from −14 to +14 , and is present in both attP and attB . Although the sequences of inner-B and inner-P sites differ at a total of nine positions , few appear to play major roles in discrimination between attP and attB , although most make small contributions to binding . For examples , Hybrid-1 , which contains inner-B but with attP flanks attached works quite well as an attP substrate . Within this region , the −4 and +4 positions are curious as they contribute to CTD binding in spite of being relatively close to the crossover site to which the N-terminal domain must interact , and the −4/+4 double mutant is strongly defective in synapsis , even at concentrations at which Int binds well . We note that double substitutions at positions equivalent to Bxb1 −4/+4 [corresponding to −3/+3 in φC31 [25]] have little impact on binding or recombination in φC31 , although changes equivalent to Bxb1 −3/+3 ( −2/+2 in φC31 ) are defective in cleavage [25] . It seems likely that different serine-integrases ‘read’ their sequences in different ways , while sharing in common the process of conformational proof reading at multiple steps in the reaction . The second architecture feature is the key discriminator positions at −15 and +15 ( which we refer to as Discriminator-L and Discriminator-R ) . The T∶A/A∶T ( in B and B′ half sites respectively ) base pair is critical for Int binding to attB , and for identity as an attB site , and when the G-15T mutation is introduced into a half site containing the T-21C , CTD binding is partially restored , presumably with a B-type conformation . This is confirmed by the observation that in the context of the full attP site with a wild-type P′ site , Hybrid-6 works with full activity as an attL site . Thus , although Int discriminates strongly between attP and attL , only two base substitutions are needed to interconvert their identities ( Figure 6B ) . Furthermore , repetition of the same two substitutions in the P′ now produces a site with attB identity ( Hybrid-5 ) albeit with reduced activity , and eliminates attP identity . We note that although inclusion of the T∶A/A∶T base pair ( in B and B′ half sites respectively ) at both −15 and +15 in attP site with proper flanks ( Hybrid-8 ) is not sufficient to switch from attP to attB identity , it severely impairs attP function , and thus antagonizes attP identity . Most other substitutions at the −15/+15 positions in attP that we tested have little impact on binding or recombination . The third architectural feature is the two flanking sequences of attP that have no counterpart in attB . Flank-L and Flank-R ( −18 to −24 , and +18 to +24 , respectively , Figure 5A ) are symmetrically conserved and are required for both efficient binding of Int and recombination . Simply adding these to a site with inner-B and attB discriminators at −15 and +15 ( Hybrid-2 ) does not prevent the site from acting as attB , but considerably impairs it , showing that these not only are required for attP function , but are also somewhat anti-attB . We note that the flanking sequences of φC31 attB are also important for efficient recombination by φC31 Int , although these are all encompassed within the site length requirements for attP [25] . In all large serine-recombinase systems in which the site requirements have been examined , attP is longer than attB [17] , [21]–[24] , [28] , [29] , and we therefore propose that the use of the extreme attP flanking sequences to confer attP identity is a common feature . The use of the −15/+15 discriminator position in other systems is unclear , although we predict that it may be a common site feature , with different systems using different positions for this function . The way in which Int recognizes these features are unclear and no structural information is available . However , we propose that a common DNA binding feature within CTD recognizes the inner parts of both attP and attB , and we predict that this lies within the N-terminal part [CTDa; [17]] of CTD ( Int residues 155–287 ) . Although CTDa alone does not bind DNA efficiently , when connected to the N-terminal catalytic domain ( i . e . to include Int residues 1–287 ) it binds DNA , albeit weakly [17] , but recognizes attP and attB similarly . A zinc-finger motif common to serine integrases – and proposed to be involved in DNA recognition [28] – is located in Bxb1 CTDb at residues 297–354 [17] , and we postulate that this specifically recognizes the attP flanking sequences . A striking conclusion from these studies is the simplicity with which site identities can be changed with only a few mutations ( Figure 6B ) . There are likely to be multiple pathways for inter-conversion , and two are shown in Figure 6B . In one pathway , introduction of the single T-21C substitution generates a substrate that binds Int but fails to undergo recombination , and likely fails to act as any type of attachment site . Adding one more substitution ( G-15T ) converts this into a fully functional attL site ( Hybrid-6 ) , and introducing the same mutations to convert the P′ site into a B′-like site generates attB identity . A second pathway involves addition of the short attP flanking sequences to attB ( Hybrid-2 ) which then retains attB identity but functions poorly . Adding GC base pairs at the −15 and +15 positions then results in a switch to attP function . It is noteworthy that none of the inter-conversion pathways we have described generate substrates that can act as both attP and attB , although this is perhaps not unexpected considering that the key identifiers ( attP flanks and the discriminators ) antagonize one identity while promoting the other . We also recognize that there are clearly additional contributions to site identity and function , as substrates such as Hybrid-7 and Hybrid-8 function as attP , but relatively inefficiently . It seems likely that a combination of activities and integration of several site components will be common to other serine-integrase systems , although because there is so much sequence diversity among the sites , often without substantial symmetry and with few positions shared between attP and attB , understanding site selection and identity in other serine integrase systems will likely require empirical determination . Serine-integrases are attractive systems for genome manipulation in heterologous systems as well as for construction of synthetic genetic circuits [8] , [12] , [13] , [30] . The Bxb1 system has good attributes for these applications and shows strong site specificity even in large genomic contexts including human , Drosophila , and Plasmodium genomes [9] , [11] , [31] . This selectivity derives from multiple proofreading steps in site selection , together with the requirement of key sequences conferring site identity , and understanding these will contribute to the use of serine-integrases for engineering purposes .
Plasmids pMY1 , pMOS-attB , pMOS-attP and pMOS-attR containing 343 bp and 50 bp of attB , 200 bp of attP and 376 bp of attR , respectively , have been described previously [14] , [15] , [26] . DNA fragments ( 50 bp ) containing wild-type and mutant attP sites were prepared by annealing complementary oligonucleotides . Mutant attP DNAs containing a single gpInt binding site were prepared by either mutating a half-site ( attP-mut P half-site or attP- mut P′ half-site ) or by eliminating a half-site ( attP-P half-site or attP-P′ half-site ) . These sites are obtained by annealing the necessary pairs of oligonucleotides ( Table S2 ) . Mutations were all transitions unless otherwise stated . Suicide substrate attB ( 50 bp ) was prepared as described earlier [17] and has a gap 4 nucleotides 5′ of the scissile bond of the top strand ( at P site ) . It is presumed to trap synaptic complexes in which all Int-DNA covalent linkages are formed , but in which religation fails due to loss of the 4-base DNA strand between the gap and the cleavage site on the top strand . Bxb1 integrase , CTD and gp47 were purified as described earlier [14] , [17] . Stocks of gpInt , CTD and gp47 proteins were diluted as appropriate in 10 mM Tris ( pH-7 . 5 ) , 1 mg/ml Bovine serum albumin ( BSA ) and 1 mM Dithiothreitol ( DTT ) . DNA substrates were prepared by 5′ end labeling of one oligonucleotide of each pair and annealing . Approximately 0 . 1 pmol of labeled DNA was incubated with either gpInt and CTD in a buffer containing 20 mM Tris ( pH-7 . 5 ) , 25 mM NaCl , 10 mM EDTA , 10 mM Spermidine , 1 mM DTT , and 1 µg Calf Thymus DNA , in a total volume of 10 µl . Reactions were incubated at 37°C for one hour and the protein-DNA complexes separated on a native 5% ( unless otherwise stated ) polyacrylamide gel at 4°C . Gels were dried , exposed to a phosphorimager screen overnight and scanned ( Fuji Phosphoimager ) . Kd was determined as the Int or CTD concentration in which one half of maximal binding was observed . If multiple complexes were observed the apparent Kd was deduced from the protein concentration at which half of the DNA remained unbound . In vitro integrative recombination assays were performed as described previously [15] in a recombination buffer containing 20 mM Tris ( pH-7 . 5 ) , 25 mM NaCl , 10 mM EDTA , 10 mM Spermidine and 1 mM DTT in final volume of 10 µl . Reactions using supercoiled pattB DNA contained 0 . 03 pmol of pMOS and 50 bp of attP DNA . The integration reactions were incubated at 37°C for up to 1 h and heat inactivated at 75°C for 15 min . The products were separated by electrophoresis in 0 . 8% agarose in 1× TBE running buffer and visualized by ethidium bromide staining . In vitro excision were carried out between 376 bp of attR in pMOS-attR and linear attL ( 50 bp ) in the above recombination buffer , gpInt and gp47 were added as indicated . The reaction were carried out at 25°C for 2 hours and separated on a 0 . 8% agarose gel . For synaptic complex formation and cleavage assays , 5′-end labeled suicide attB ( 50 bp ) DNA was incubated with Int and attP DNA under the same conditions as for DNA-binding . After 1 hour incubation at 37°C reactions were heat inactivated at 75°C for min 15 min . For cleavage assays reactions were treated with 1 mg/ml Proteinase K and 0 . 2% SDS at 55°C for 15 min .
|
Site-specific recombinases catalyze recombination between two specific DNA sites to generate the products of recombination . The Integrase encoded by mycobacteriophage Bxb1 is a member of the serine-recombinase family and catalyzes strand exchange between attP and attB , the attachment sites for the phage and bacterial host , respectively . Although the DNA sites are relatively small ( <50 bp ) , the reaction is highly selective for these sites and is also strongly directional . Here , we address the question of what sequences within attP are required for it to act as an attP site and identify the key sequence features that are required not just for Integrase binding but also for synapsis and post-synapsis events . We also have identified the key determinants of attP and attB identity , and although the sites are different in sequence and length , they can be interconverted with just two base changes in each of the half sites .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"genetic",
"mutation",
"mutation",
"types",
"nucleic",
"acids",
"genetics",
"biology"
] |
2013
|
Attachment Site Selection and Identity in Bxb1 Serine Integrase-Mediated Site-Specific Recombination
|
Genome-scale metabolic models have proven useful for answering fundamental questions about metabolic capabilities of a variety of microorganisms , as well as informing their metabolic engineering . However , only a few models are available for oxygenic photosynthetic microorganisms , particularly in cyanobacteria in which photosynthetic and respiratory electron transport chains ( ETC ) share components . We addressed the complexity of cyanobacterial ETC by developing a genome-scale model for the diazotrophic cyanobacterium , Cyanothece sp . ATCC 51142 . The resulting metabolic reconstruction , iCce806 , consists of 806 genes associated with 667 metabolic reactions and includes a detailed representation of the ETC and a biomass equation based on experimental measurements . Both computational and experimental approaches were used to investigate light-driven metabolism in Cyanothece sp . ATCC 51142 , with a particular focus on reductant production and partitioning within the ETC . The simulation results suggest that growth and metabolic flux distributions are substantially impacted by the relative amounts of light going into the individual photosystems . When growth is limited by the flux through photosystem I , terminal respiratory oxidases are predicted to be an important mechanism for removing excess reductant . Similarly , under photosystem II flux limitation , excess electron carriers must be removed via cyclic electron transport . Furthermore , in silico calculations were in good quantitative agreement with the measured growth rates whereas predictions of reaction usage were qualitatively consistent with protein and mRNA expression data , which we used to further improve the resolution of intracellular flux values .
Cyanothece spp . are unicellular , diazotrophic cyanobacteria that temporally separate light-dependent oxygenic photosynthesis and glycogen accumulation from N2 fixation at night [1] . When grown under nutrient excess , Cyanothece sp . strain ATCC 51142 ( thereafter Cyanothece 51142 ) cells can accumulate significant amounts of storage polymers including glycogen , polyphosphates , and cyanophycin [2] . The inter-thylakoid glycogen granules are significantly larger in size than those found in other cyanobacteria , which points at an unusual branching pattern and packaging of this compound . From a biotechnological perspective , this presents an intriguing theoretical possibility to accumulate substantially higher amounts of polyglucose without any significant increase in the number of granules [3] . Cyanothece 51142 is also of interest for bioenergy applications due to its ability to evolve large quantities of H2 . Remarkably , H2 production in this organism can occur under light conditions in the presence of O2 and is mediated by nitrogenase [4] , [5] Sequencing of the Cyanothece 51142 genome [6] has enabled application of high-throughput genomic approaches to study the unique physiological and morphological features of this organism . Transcriptomic and proteomic studies have been conducted to analyze global gene expression patterns under a variety of environmental conditions and infer regulatory pathways that govern the organism's diurnal growth [7] , [8] . The availability of genomic information also provides means to construct genome-scale constraint-based models of metabolism , which are powerful tools for systems-level analysis and prediction of biological systems response to environmental cues and genetic perturbations [9] , [10] . Such models have been developed for a variety of biological systems [9] but only in a few studies has this approach been applied to photosynthetic microorganisms , including Synechocystis sp . PCC 6803 [11]– , Rhodobacter sphaeroides [14] , and Chlamydomonas reinhardtii [15] , [16] . However , the modeling of metabolism in oxygenic photoautotrophs is an intriguing problem due to the complexity of photosynthetic and respiratory electron transport chains , and the potential effects of two distinct photosystems upon the generation and fate of reductant and energy that drives the remainder of metabolism . In this work , we developed the first genome-scale metabolic model of Cyanothece 51142 and used a combination of computation and experimental approaches to investigate how photosynthetic and respiratory fluxes affect metabolism . Discrete representation of PS II and PS I and their integration with multiple respiratory pathways enabled modeling of photon fluxes and electron flux distributions under conditions of variable light quality and intensity . The predicted changes in growth rates of Cyanothece 51142 in response to changes in light input were experimentally tested using a photobioreactor with controlled sources of monochromatic 630 and 680 nm light . We also carried out computational and experimental analyses of light- and nitrogen-limited chemostat growth of Cyanothece 51142 and used mRNA and protein expression data to constrain model-predicted flux distributions . Both in silico and experimental data suggest that respiratory electron transfer plays a significant role in balancing the reductant ( NADPH ) and ATP pools in the cells during photoautotrophic growth . This study is a first step towards a systems-level analysis of cyanobacterial metabolism , as it integrates information into a genome-scale reconstruction to understand metabolism qualitatively and quantitatively through a constraint-based analysis [9] . We also discuss strategies for improving internal flux distributions through integration of in silico simulations and data .
To build a constraint-based metabolic model of Cyanothece 51142 , a genome-scale metabolic network was reconstructed using the genome annotation and data from NCBI [6] , SEED [17] , KEGG [18]–[20] , and CyanoBase [21] , [22] . The resulting iCce806 network contains 806 genes and 667 metabolic and transport reactions ( see Dataset S1 and Tables S1 , S2 , S3 for network details ) . Most of the 42 reactions without genes associated with them were added to complete metabolic pathways needed for biomass production . The final reconstruction encompasses central metabolic pathways such as the Calvin-Benson cycle , the pentose phosphate pathway ( PPP ) , reactions within the tricarboxylic acid ( TCA ) cycle , as well as , the complete set of anabolic pathways involved in biosynthesis of glycogen , cyanophycin , amino acids , lipids , nucleotides , vitamins , and cofactors . Pathways for glycolate synthesis ( via ribulose-1 , 5-bisphosphate carboxylase/oxygenase , i . e . , photorespiration ) , glycolate conversion to serine , and glycerol catabolism are also included . Photosynthetic electron transfer associated with the thylakoid membrane is represented as a set of four separate reactions , including light capture by photosystem II ( PS II ) and photosystem I ( PS I ) , electron transfer between the two photosystems , and cyclic electron transfer around PS I . Similarly , respiratory electron transfer is represented by reactions catalyzed by terminal cytochrome c oxidase ( COX ) , quinol oxidases ( QOX , both bd- and bo-types ) , NADH dehydrogenases ( NDH , type 1 and 2 ) , and succinate dehydrogenase . In addition , two reactions ( NADP+- and ferredoxin- requiring ) for flavin-dependent reduction of O2 ( i . e . , Mehler reactions ) were included . A simplified scheme of the photosynthetic and respiratory electron transfer reactions in iCce806 is shown in Figure 1 . For initial testing , we examined the ability of the constraint-based model of iCce806 to predict growth under photoautotrophic ( using light and fixing CO2 ) , heterotrophic ( using glycerol in the dark ) , and photoheterotrophic ( using glycerol and light ) conditions with different nitrogen sources . In silico calculated biomass yields , which simulated carbon or light- limited growth ( Figure S1 ) , qualitatively agreed with previously reported growth data for Cyanothece 51142 [1] , [2] , [23] . Other non-growth conditions that were simulated with the model , included nitrogen fixation as occurs during the dark phase of Cyanothece's ciracadian cycle [1] . In this case , the oxidation of glycogen provides reductant and ATP for nitrogenase , and we examined the model's ability to quantitatively predict the amount of nitrogen ( N2 ) that could be fixed and stored in the dark , by maximizing cyanophycin production from glycogen . Although H2 is an obligate co-product of the nitrogenase reaction , no H2 was produced in the initial simulations under dark N2-fixing conditions , contradicting experimental observations . Model examination revealed that all of the nitrogenase-generated H2 was utilized by hydrogenases to reduce NAD ( P ) and ferredoxin , which ultimately increased cyanophycin production . When the three hydrogenase reactions ( HDH_1 , HDH_2 , and UPHYDR ) were eliminated from the model , the predicted ratio of fixed N2 to consumed glycogen depended on the non-growth associated ATP requirement ( NGAR ) , and was estimated to be 0 . 3 ( NGAR = 2 . 8 ) or 0 . 67 ( NGAR = 0 ) mole N2/mole glycogen , which was in accordance with an experimentally measured value of 0 . 51 [2] . Under this condition , the model predicted that H2 production would have same yields as fixed N2 ( 0 . 3 to 0 . 67 mole H2/mole glycogen ) due to the stoichiometry of the nitrogenase reaction . We also evaluated how fluxes through electron transfer reactions are affected by the nitrogenase flux under N2-fixing dark conditions . With glycogen being the sole source of reductant for both ATP-generating oxidative phosphorylation and N2 reduction , a balance between fluxes through respiratory pathways and nitrogenase reaction is needed . In the absence of the hydrogenase reactions , the model predicted that O2 reduction via COX , QOX , or Mehler reactions are required to consume NADH resulting from glycogen catabolism ( Figure S2 ) . The model predicts that the COX reaction is required to achieve the maximum N2 fixation rate since it generates more ATP than the QOX or Mehler pathways ( ∼9 O2 are needed per N2 fixed ) . This is consistent with the results from recent proteomic studies showing the CoxB1 ( cce_1977 ) subunit of COX is more predominant during the dark [24] , [25] . These results suggest terminal oxidases are important under dark N2-fixing conditions not only to generate an intracellular anaerobic environment for nitrogenase , but also to provide ATP for nitrogenase activity . As photosynthesis and respiratory electron transport chains are interconnected in cyanobacteria [26] , these pathways were allowed to interact in the iCce806 model . To perform model robustness analysis , we computationally explored the impact of key photosynthetic and respiratory pathways on growth rate and intracellular flux distributions under varying photon uptake flux for PS I , while the photon uptake flux for PS II was fixed at 20 mmol·g−1 AFDW·h−1 ( Figure 2 ) . First , the model was evaluated assuming only linear photosynthetic electron transfer . In this case , all alternative reductant sinks including the proton and O2 reduction as well as cyclic photosynthetic reactions around PS I were eliminated from the model ( Figure 2A ) . Under this condition , growth only occurred at one value of photon uptake flux for PS I and extracellular organic products ( ethanol , lactate and/or alanine with trace amounts of formate ) would have to be secreted in order to generate enough ATP to support biomass production . Second , when cyclic photosynthetic reactions were added back , the photon uptake flux for PS I could vary with a fixed photon uptake flux for PS II , but significant amounts of extracellular products were still formed until the photon uptake flux for PS I exceeded ∼85 mmol·g−1 AFDW·h−1 ( Figure 2B ) . No growth occurred unless PS I photon uptake flux was greater than or equal to the photon uptake flux for PS II . Only when the model was allowed to use both cyclic photosynthesis and O2 reduction reactions were no extracellular products predicted and the photon uptake flux for PS I could be less than that for PS II ( Figure 2C ) . Since experimental data does not indicate that any by-products including H2 or organic acids are produced by Cyanothece 51142 at a detectable level during photoautotrophic growth with excess ammonium , a plausible mechanism for balancing growth through the generation of additional ATP may involve activity of the cytochrome oxidases . The discrete representation of PS II- and PS I-mediated reactions and their interactions with multiple respiratory reactions in iCce806 enabled further in silico analysis of growth and electron flux distributions under photoautotrophic conditions of variable light quality and intensity . In this case , the complete model was used to explore which reactions would be used to support maximal photoautotrophic growth rates for different levels of PS II and PS I photon uptake fluxes . To predict the corresponding growth rates under light-limited conditions , we constrained the photon uptake fluxes ( ranging from 0 to 60 mmol·g−1 AFDW·h−1 ) through each photosystem . The resulting phenotypic phase plane ( PhPP ) contained three distinct regions ( Figure 3A ) : in two regions growth was limited only by fluxes through PS II ( region 1 ) or PS I ( region 3 ) , while in region 2 growth was limited by both PS II and PS I photon uptake fluxes ( i . e . , increases in either flux would improve growth rate ) . By adding artificial ATP or NADPH generating reactions ( ADP+HPO4+H→ATP+H2O and NADP+H→NADPH ) to the model and analyzing changes in predicted maximal growth rates , we were able to identify that in regions 1 and 3 growth was NADPH/reductant-limited , while in region 2 it was limited by energy supply ( Figure 3A ) . To analyze the effect of photon uptake rates on electron flux distributions , we calculated the flux ranges using flux variance analysis ( FVA ) for all photosynthetic and respiration reactions within each PhPP region ( Figure 3B ) . In this instance , PhPP FVA was run with constraints that restrict the model to a given region and to the maximum growth for each point in the region ( in contrast , standard FVA is used at a single point in a region ) . Using PhPP FVA , we identified active ( both minimum and maximum flux values are positive or negative ) , inactive/blocked ( minimum and maximum fluxes are both zero ) , and optional ( which could have at least one zero and one non-zero flux value somewhere in the region ) reactions leading to optimal solutions in each PhPP region . This new analysis technique allowed classification of reaction usage across entire regions of the PhPP and is not restricted to fixed points within a region . While linear photosynthesis was active and Mehler reactions were blocked across the entire PhPP , there were differences in the usage of photosynthetic and respiratory reactions observed within all three regions ( Figure 3B ) . Surprisingly , while generation of NADPH from reduced ferredoxin via linear photosynthesis is the key source of reductant , ferredoxin-NADP+ oxidoreductase ( FNR ) was predicted to be active in region 2 , but optional in regions 1 and 3 . Closer examination of in silico calculated electron flux distributions revealed that , in addition to FNR , the model utilized a cycle involving glutamine synthetase , glutamate synthase and transhydrogenase , resulting in ATP-driven NADPH production . In regions 1 and 3 , the model predicts there is excess ATP , and so this cycle can be used instead of FNR to move electrons from ferredoxin to NADPH . However , this cycle is unlikely to be of any physiological relevance since there has been no experimental data supporting this route for making NADPH , and FNR is essential for photoautotrophic growth in unicellular cyanobacteria such as Synechococcus 7002 [27] . Differences in the predicted usage of respiratory reactions were also found . In region 1 , where growth is limited by the flux through PS I , at least one of the COX and QOX reactions must be active to oxidize excess electron carriers ( Pc , Cyt c6 , or Pq ) generated from PS II . Similarly , in region 3 under PS II flux limitation , excess electron carriers ( Pq , Fd ) must be reduced via NDH-1 or –2 or ferredoxin-dependent cyclic electron transfer ( FdPq ) . Conversely , due to ATP limitation in region 2 , the model favored reactions with higher proton pumping capacities and so both the QOX and FdPq reactions were inactive . The usage of COX was optional in region 2 and depended on photon uptake rates ( e . g . , COX reaction was inactive at the boundary between regions 2 and 3 ) . The model predictions ( Figure 3A ) were compared to batch growth experiments in the LED-photobioreactor which allowed instantaneous measurements of initial growth and photon uptake rates by Cyanothece 51142 cells exposed to different intensities and ratios of 630 and 680 nm light ( Table 1 ) . When Cyanothece 51142 cultures were illuminated with both 630 nm and 680 nm light , initial growth rates generally correlated with the total photon flux through PS II and PS I , with higher growth rates observed at 80 mmol·g−1 AFDW·h−1 total photon flux and 630 nm∶680 nm light ratio of 2∶1 . When cultures were exposed to only a single wavelength of light ( batch experiments 6–10 ) , i . e . , either 630 or 680 nm , Cyanothece 51142 cells displayed a similar trend with higher growth rates observed at higher photon flux intensities . The predicted growth rates were within 7% of the experimentally measured values , except for the two cases where single 630 nm wavelength irradiances were used ( Table 1 ) . The reasons for this are unclear but may be due to other physiological and/or biochemical phenomena such as state transitions that are not contained within the model but are operating in vivo . Data from these batch experiments ( batch experiments 1–5 , Table 1 ) were also used to estimate the growth ( GAR ) and non-growth ( NGAR ) associated ATP requirements . NGAR is the amount of energy spent to maintain the cell ( i . e . , maintenance energy ) . GAR is defined as energy expenditures used on protein and mRNA turnover or repair , proton leakage , and maintenance of membrane integrity; it does not include ATP spent on polymerization reactions , which are already accounted for in the macromolecular synthesis pathways of the network . The time-averaged growth and photon uptake rates were used to constrain the model and the maximal amount of ATP hydrolysis was calculated ( Figure S3 ) for each batch experiment . A plot of growth rate versus maximum ATP hydrolysis flux was generated and a linear fit used to estimate the GAR and NGAR values [28] . Specifically , the slope of the fitted line is the GAR ( 544 mmol·g−1 AFDW·h−1 ) , and the y-intercept is NGAR ( 2 . 8 mmol·g−1 AFDW·h−1 ) . The estimated GAR value is significantly higher than those reported from other bacteria [29]; however , these model estimates assume that all absorbed photons lead to photosynthetic fluxes ( 100% quantum efficiency ) and that the overall efficiency of ATP production via all electron transfer reactions ( photosynthetic and respiratory ) are accurate . Depending on the growth condition the quantum yields can change , and for Cyanothece 51142 this value was reported to be between ∼70–100% for photoautotrophic growth [23] . Upon further analysis , we found the estimated Cyanothece ATP requirements were most sensitive to reductions in quantum efficiency and the amount of ATP generated by photosynthesis and respiration ( Table S4 ) . Since neither quantum efficiency nor combined photosynthetic and respiratory ATP production were experimentally measured for Cyanothece 51142 , the original estimates , GAR = 544 and NGAR = 2 . 8 were used in all growth simulations . Chemostat cultures grown under light and ammonium limitations were used to calculate metabolic fluxes and further understand reductant partitioning pathways in Cyanothece 51142 . The differences in biomass composition between these growth conditions indicated a major shift in carbon partitioning pathways ( Figure 4; and Table S5 ) . In ammonium limited cultures , carbohydrates comprised almost half of cell biomass; in contrast , under light limitation , Cyanothece 51142 cells contained higher amounts of protein , nucleic acids , and approximately 10% cyanophycin . The quantitative biomass composition measurements were used to generate two separate biomass equations for the metabolic model; experimentally measured growth rate , photon uptake rates , and O2 production rates were included as additional model constraints ( Table S6 , in this case no mRNA or protein expression data is used by the model ) . Using FBA and through minimization of the overall flux magnitude , we calculated representative flux distributions under light and ammonium limitations ( values listed in Table S1 ) . As expected , changes in flux values were consistent with differences in measured biomass compositions used in the simulations: under light limitation , fluxes increased for reactions involved in biosynthesis of amino acids , nucleotides and cyanophycin , while ammonium limitation resulted in flux increases for glycogen biosynthesis . Comparisons of global transcriptome profiles displayed by Cyanothece 51142 during ammonium- and light-limited chemostat growth also reflected the rewiring of cellular metabolism ( Table S7 ) . Under ammonium limitation , significant increase in relative mRNA abundances was observed for genes involved in N2 fixation ( cce_0198 , cce_0545–0579 ) , iron acquisition ( cce_0032–0033 , cce_1951 , cce_2632–2635 ) , respiratory electron transport ( cce_1665 , cce_3410–3411 , cce_4108–4111 , cce_4814–4815 ) as well as peptide transport , synthesis , and protein repair ( cce_0392 , cce_1720 , cce_3033 , cce_3054–3055 , cce_3073–3075 ) . Among the most highly expressed genes in ammonium-limited Cyanothece 51142 cells was the one encoding 6-phosphogluconate dehydrogenase ( cce_3746 ) , a key PPP enzyme . Under light limitation , the major changes in the transcriptome of Cyanothece 51142 included upregulation of genes encoding: components of the photosynthetic apparatus and electron transport chain ( cce_0776 , cce_0989–0990 , cce_1289 , cce_2485 , cce_2959 , cce_3176 , cce_3963 ) ; pigment biosynthesis ( cce_0920 , cce_1954 , cce_2652–2656 , cce_2908 , cce_4532–4534 ) ; CO2 uptake and fixation machinery ( cce_0605 , cce_3164–3166 , cce_4279–4281 ) ; ATP synthase ( cce_2812 , cce_ 4485–4489 ) , and protein synthesis machinery ( cce_ 4016–4030 ) ( Table S7 ) . Global proteome profiles of Cyanothece 51142 corroborated the shifts in gene expression ( Table S8 ) . The abundance of proteins from central metabolism ( glycolysis , TCA , and pentose phosphate pathway ) all had significant differences between cells grown under ammonium- and light-limited conditions . Enzymes of the oxidative PPP branch , namely glucose-6-phosphate dehydrogenase ( cce_2535–2536 ) , 6-phosphogluconolactonase ( cce_4743 ) and 6-phosphogluconate dehydrogenase ( cce_3746 ) , showed increased abundances under ammonium limited conditions . Similarly , two-fold increase in abundance levels was observed for gluconeogenesis proteins , including fructose 1 , 6-bisphosphatase ( cce_4758 ) , glucose-6-phosphate isomerase ( cce_0666 ) , glyceraldehyde-3-phosphate dehydrogenase ( cce_3612 ) , and phosphoglycerate kinase ( cce_4219 ) . In contrast , relative abundances of proteins catalyzing the conversion of glycerate-3P to pyruvate ( cce_1789 and cce_2454 ) were unchanged or up-regulated ( pyruvate kinase cce_3420 ) in light-limited cells . Consistent with the results from global mRNA profiles was the up-regulation of Cyanothece 51142 proteins involved in photosynthesis and carbon fixation under light-limited conditions ( Table S8 ) . Notably , two key components involved in the electron transfer to PS I , namely plastocyanin ( cce_0590 ) and cytochrome b6 ( cce_1383 ) , displayed elevated peptide abundances in light-limited cells . Since there may be more than one flux distribution that is consistent with the experimentally measured rates of growth , photon uptake , and O2 production we used FVA to identify required ( flux must be non-zero ) , optional ( flux may or may not be zero ) , or inactive ( flux must be zero ) reactions under light- and ammonium-limited growth conditions . As our initial simulations ( Table 2 ) produced a large number of optional reactions ( 170 out of 667 for both growth conditions ) , that represent uncertainty regarding usage , we subsequently used the transcriptome and proteome data ( TPD ) to further constrain the model . Using a modification to a previously developed approach [30] , we obtained a flux distribution that was consistent with measured rates and TPD while reducing the overall flux magnitude ( Table S1 ) . In this analysis , flux was favored through reactions for which proteins were detected and disfavored through reactions associated with undetected proteins and transcriptome data less than a given threshold ( e . g . , log2 of mRNA expression level is less than 8 ) . The model constrained by TPD predicted that the majority of reactions in central metabolism would be active under both chemostat conditions ( Figure 5 ) . In addition , we subsequently applied FVA employing additional constraints arising from the TPD . Comparison between FVA results with and without TPD constraints demonstrated a significant decrease in the number of ambiguities ( the optional reaction set ) when TPD is used ( Table 2 ) . While the number of optional reactions was reduced by incorporating TPD into the model , the flux spans ( difference between maximum and minimum values ) of individual fluxes was still large ( >30 mmol·g−1 AFDW·h−1 for some central metabolic reactions , Table S1 ) . These large flux spans could arise from cycles or alternative pathways in the model , and deleting these features from the model could subsequently reduce the flux spans . FVA was repeated using measured growth , photon uptake , and O2 release rates under light-limited conditions as constraints and with optional reactions were deleted ( similar results were found for ammonia limited conditions , data not shown ) . Flux spans for reactions in central metabolism ( Figure 5 ) were then calculated for a series of single or double reaction deletions in silico . The purpose of this analysis was to identify those reactions that exert the greatest impact on the flux span in central metabolism ( Figure 6A ) . Single deletions of glyceraldehyde-3-phosphate dehydrogenase ( GAPD or GAPD_NADP ) or hydrogenase ( HDH_1 ) reduced the average central metabolic flux span the most ( from 74 to 22 mmol·g−1 AFDW·h−1 ) . Other single deletions with significant effects included FNR and NDH-1 , which are involved in photosynthesis and respiration . The reaction deletions shown in Figure 6A all had a larger impact on reducing average central metabolic flux span than did imposition of constraints based on TPD . There were cases where single deletions had large effects on other specific reactions , but only modest effects on overall central metabolic flux spans . For example , a single deletion in phosphogluconate dehydrogenase ( PGDHr ) reduced the span for glucose-6-phosphate isomerase flux ( PGI ) to 0 ( Figure 6B ) , but only reduced the average central metabolic flux span by ∼0 . 7 mmol·g−1 AFDW·h−1 . The in silico analysis of double reaction deletions did not yield any new double deletions that would reduce the average central metabolic flux span significantly . However , some double deletions strategies did reduce flux spans of individual reactions .
Several cyanobacterial metabolic models ( all for Synechocystis PCC 6803 ) have been published , which represented photosynthesis as two lumped reactions [12] , [31] for linear ( PSII , Cyt b6f , PSI , and FNR ) and cyclic ( PS I and Cyt b6f ) pathways . In this study , we modeled photosynthesis as a larger set of separate reactions [13] as this structuring allowed analysis of the effects of different illumination on the production and partitioning of reductant through photosynthetic and respiratory reactions , as well as the contribution of different electron transfer pathways to growth . Our PhPP FVA results showed how different photosynthetic and respiratory electron transport chain components are used to maximize biomass production under different lighting regimes . It was not surprising that linear photosynthesis was active in all three regions because the cell needs photons from both PSI and PSII to grow under photoautotrophic conditions . However , the Mehler reactions were inactive in all three regions when we only consider maximal growth rate solutions . In regions 1 and 3 , reducing equivalents ( e . g . , NADPH ) limit growth and the Mehler reactions would lower the amount of reducing equivalents available for growth . The Mehler reactions are less energetically efficient than NADH dehydrogenase and cytochrome oxidase so the model would not use them in region 2 , where ATP is limiting . So while the Mehler reactions can carry flux in the model , using these reactions lowers the maximum growth rate making them inactive ( blocked reactions ) in our PhPP analysis . A recent study showed that the Mehler reactions are operational in Synechocystis sp . PCC 6803 , serving as a sink for excess electrons [32] . These reactions are also likely to be active in Cyanothece 51142 , since the associated proteins were detected in the proteomic data ( Table S8 ) . As a result the model only predicted non-zero Mehler fluxes when the proteomic data were used to constrain the model ( Table S1 ) . In the absence of cyclic photosynthesis , other products including water ( produced by COX , QOX or Mehler reactions ) , H2 ( via hydrogenase ) , or small organic compounds ( alanine , ethanol , lactate and formate ) were predicted to be necessary in order to balance the electrons and ATP needed to support growth . In the presence of linear and cyclic photosynthesis reactions , these products must also be produced unless significant amounts of cyclic photosynthesis occurs ( >3 times the amount of linear photosynthesis ) . Since H2 and small organic compounds are not generally produced under photoautotrophic conditions with excess ammonium , any additional energy is most likely supplied by cytochrome oxidase activities that reduce photosynthetically produced O2 . Interestingly , in the absence of cytochrome oxidase activities in the model , the PS I fluxes must always be greater than or equal to the PS II fluxes . It was shown that the marine cyanobacteriium Synechococcus has a PS I/PS II protein ratio >1 , which has been explained as a mechanism to protect PS II from photo-damage [33] . Under conditions with high levels of PS II activity , cytochrome oxidase activity may ensure an adequate supply of oxidized plastoquinone ( needed for PS II ) and reduce O2 concentrations to limit photorespiration . Similarly , cyclic electron flow via NADH dehydrogenase- or ferredoxin-dependent routes have also been experimentally demonstrated to play important roles in balancing the amount of NADPH and ATP produced via photosynthesis . Synechocystis 6803 mutants lacking ndhD genes ( encoding subunits of NDH-1 ) had significantly lower cyclic photosynthesis activity [34] . Although the mechanism of electron transfer from ferredoxin to the plastoquinone pool ( without using NDH ) is still unclear , its activity has been demonstrated in green algae [35] and higher plants [36] . Our computational simulations also showed that , under light-limited photoautotrophic conditions , cyclic electron transfer involving NADH dehydrogenase ( NDH-1 ) is needed for maximal growth if ATP ( rather than NADPH ) is limiting . In an environment where PS I photon availability is high relative to PS II , cyclic electron transport is needed ( Figure 2 ) to increase availability of PS I substrates ( reduced PC or Cyt c6 ) and protect against photo-damage . Cyclic electron flow has been experimentally shown to help protect the photosynthetic apparatus from photo-damage [37]–[39] In addition to studying the interactions between components of the photosynthetic and respiratory components computationally , we also experimentally evaluated cells grown under continuous light conditions in light- and ammonia-limited chemostats . The measured 630 nm and 680 nm photon uptake and O2 production rates suggests that reductant was being directed towards O2 via the Mehler , QOX , and/or COX reactions . In both chemostat conditions , the model predicted that steady-state growth rate could have been achieved using lower photon uptake rates by decreasing the amount of reductant generated by PS II that was predicted to reduce O2 . A limitation to flux balance analysis is that a wide range of flux values may be consistent with the constraints in the computational model . An iterative application of computational and experimental methods is an important strategy to improve the comprehensive understanding of cyanobacterial metabolism . We have begun to apply this iterative approach , by including mRNA and protein expression datasets as additional constraints beyond biomass composition and physiological rate measurements . Experimentally-measured TPD were successfully used to further constrain the model , and thereby reduce uncertainty and increase the number of required ( that is , metabolically active ) reactions ( Table 2 ) . However , there remained discrepancies in that the model did not predict flux through all reactions for which proteins were experimentally detected . Such discrepancies can be used to subsequently improve the model with previously developed approaches [40]–[42] . For example , an earlier version of the model did not predict flux through proline oxidase , even though proteome data demonstrated that proline oxidase was synthesized . This prediction arose because the model did not contain a reaction in which FADH2 ( a product of the proline oxidase reaction ) could be reoxidized to FAD . After experimental confirmation that proline can be used as a nitrogen source ( implying activity of proline oxidase ) by Cyanothece 51142 ( data not shown ) , a FADH2 recycling reaction was included in the final iCce806 model . Even with these additional TPD constraints , a wide range of flux values remained feasible ( Figure 6 ) . We should note that we did not take real enzymatic activities into account ( which can be affected by post-translational modifications ) , as we did not have this type of data for the two conditions examined . Such data , if available , could be used as additional factors for determining whether to favor or disfavor fluxes through associated reactions ( See Material and Methods ) . Other constraint-based methods for incorporating gene expression data use similar Boolean on/off type of constraints to restrict fluxes [30] , [43] , [44] and would be expected to yield results similar to those described herein . Thus , novel computational methods which can more quantitatively constrain the metabolic flux values are still needed . The strategy of evaluating fluxes for reaction deletions in silico can be used to identify knockout mutants that can potentially improve the resolution of intracellular flux distributions . A flux that is well resolved would have a small span meaning we can more definitively state its value . If the mutants show no growth defects then the corresponding reactions may not be used under the conditions tested , or alternative pathways not included in the model may occur . Either way , this information could be used to better resolve the intracellular flux distribution or improve the metabolic model . For Cyanothece 51142 , this would require development of a genetic system ( such a system already exists for another Cyanothece strain [45] ) as experiments with mutants would have the most potential to improve resolution of central metabolic fluxes during photoautotrophic growth . Also , as a complement to the in silico reaction knockouts that our simulations predict would reduce the flux spans associated with central metabolic reactions , the photobioreactor employed here provides a system whereby cultivation conditions can be rigorously controlled and some aspects of physiological state monitored continuously . In addition , cells from steady-state or perturbed cultures can be interrogated via physiological or biochemical analyses to experimentally test the predictions of the computational models for wild type or mutants . As the number of available cyanobacterial models continues to grow , cross-species physiological , genomic , and metabolic comparisons will enable the identification of core networks and contribute towards improving our understanding of metabolic processes in cyanobacteria .
Cyanothece 51142 was grown in modified ASP2 medium [46] amended with 0 . 75 mM K2HPO4 , 0 . 03 mM FeCl3•6H2O , and 17 mM NH4Cl which substituted NaNO3 as the nitrogen source . Routinely , the cells were maintained under continuous white light illumination ( 50 µmol photons·m−2·s−1 ) in 1-L Roux bottles sparged with CO2-enriched air ( 0 . 3% vol/vol ) . Culture purity was confirmed by plating on DIFCO Bacto Tryptic Soy Broth and DIFCO Luria-Bertani solid media ( BD Diagnostic Systems , Franklin Lakes , NJ ) as well as by phase contrast or acridine orange fluorescent microscopy . Controlled batch and chemostat cultures of Cyanothece 51142 were grown in a 7 . 5-L borosilicate glass vessel operated at 5 . 5-L working volume under the control of New Brunswick Bioflo 3000 bench-top bioreactor ( New Brunswick Scientific , Edison , NJ ) . The vessel was housed in a custom-made black anodized aluminum enclosure equipped with light-emitting diodes ( LED ) generating 680 nm and 630 nm light for the preferential excitation of chlorophyll a and phycobilin pigments , respectively . Built-in sensors allowed for automatic adjustment of incident and transmitted light intensities using custom-designed control module . Both hardware and software components of the LED enclosure and the control module were developed at Pacific Northwest National Laboratory ( US Patent Application # 20100062483; http://appft1 . uspto . gov ) . All experiments were carried out under continuous illumination in modified ASP2 medium sparged with CO2-enriched argon ( Ar ) ( 0 . 2% vol/vol ) . Agitation , temperature , pH , and gas flow rates were maintained at 250 rpm , 30°C , 7 . 4 , and 2 . 8 L/min , respectively . Incoming and off-gas composition was constantly monitored by an in-line mass spectrometry based gas analyzer MGA iSCAN ( Hamilton Sundstrand , Pomona CA ) . Cell density was monitored spectrophotometrically at 625 , 678 , and 730 nm . To establish a light-limited chemostat culture , the photobioreactor was inoculated with 10 mL of mid-log phase Cyanothece 51142 cells and maintained as a batch culture under 630 nm and 680 nm illumination at 40 and 70 µmol photon·m−2·s−1 , respectively , until the culture reached late logarithmic stage . Chemostat mode was initiated by continuous inflow of medium at a dilution rate of 0 . 05 hr−1 that resulted in a steady-state optical density ( OD730 ) of 0 . 20 . Similarly , a nitrogen-limited continuous culture of Cyanothece 51142 was established using low-nitrogen ASP medium containing 0 . 75 mM NH4+ . The ammonium-limited chemostat was maintained under identical operating conditions in regard to the culture dilution rate and optical density under incident light at 38 . 5 and 73 . 5 µmol photon·m−2·s−1 for 630 nm and 680 nm wave lengths , respectively ) . The light uptake fluxes ( mmol·g−1 AFDW·h−1 ) were determined by multiplying the light consumption rates ( µmol photon m−2s−1 ) by the surface area of cell culture exposed to light ( m2 ) and dividing by the amount of biomass in the reactor ( g AFDW ) . The light consumption rates were determined by subtracting the transmitted light intensity from the values of incident light intensity after corrections were made for the abiotic consumption of light to account for the gas bubbles and probes in the reactor . Cells in the 5 . 5 L working volume were assumed to be equally exposed to the light at all times . Based on the inner diameter and height of the liquid culture at working volume , the surface area was 0 . 1403 m2 . The amount of biomass in the reactor was determined from the working volume and biomass concentrations . Biomass ash-free dry weight ( AFDW ) was measured using centrifuged ( 11 , 000× g , 4°C ) cell pellets as described previously [29] . Total protein , reducing carbohydrates , RNA , and DNA were assayed using standard analytical techniques [47]–[49] . The total lipid fraction was measured gravimetrically after an extraction from a known volume of freeze-dried culture using previously published methodology [50] . Total reducing carbohydrates were quantified using the anthrone method [51] with glycogen as the standard . Chlorophyll concentrations were measured as described elsewhere [52] , [53] . Amino acid composition was analyzed in acid-phenol hydrolyzed samples prepared using Eldex hydrolysis/derivatization station ( Eldex Laboratories , Inc . , Napa , CA ) [29] . The derivatized samples were resolved on a 4-µm AccQ-Tag Nova-Pak C-18 column ( 3 . 9 mm×150 mm , Waters Corp . , Milford , MA , USA ) , eluted using a linear gradient of acetonitrile ( from 1 . 2% to 4 . 2% over 15 min . ; from 4 . 2% to 6% over 4 min . ; from 6% to 20% over 12 min . ; at 20% over 1 min . ; from 20% to 60% over 1 min . ) with a flow rate of 1 . 0 ml/min at 37°C , and detected at 254 nm ( HPLC system and UV detector by Shimadzu , Tokyo , Japan ) . Cyanophycin was estimated based on relative amino acid values and total protein measurements ( see Text S1 for details ) . Previously developed whole-genome oligonucleotide microarrays of Cyanothece 51142 [7] were manufactured by Agilent Technologies ( Santa Clara , CA ) . RNA isolation , labeling , hybridization , and data analysis were performed by MOgene , LC ( St . Louis , MO ) using published protocols [7] . Cell lysis and tryptic digestion followed a previously described “global protein preparation” scheme [54] . A reference peptide database was prepared using strong cation exchange fractionation ( 10 fractions ) of a portion of each global digest , as previously reported [55]–[57] . The methods for capillary liquid chromatography and mass spectrometry have been described in detail elsewhere [54] , [58] , [59] . Here , the HPLC mobile phase was 0 . 1% formic acid in water ( A ) and 0 . 1% formic acid in acetonitrile ( B ) . A Finnigan LTQ ion trap mass spectrometer ( ThermoFinnigan , San Jose , CA ) was used for MS/MS analysis of SCX fractions and an LTQ-Orbitrap ( Thermo ) was used for high-resolution MS analysis of the global unfractionated samples . Each of the 10 SCX fractions was analyzed once , while each global digest was injected four times . To build an accurate mass and time ( AMT ) tag database , SEQUEST analysis software was used to match the MS/MS fragmentation spectra to sequences from the annotation of the Cyanothece 51142 proteome [6] . Peptide identifications from the SCX fractions were combined with identifications from unfractionated samples to create a reference database of calculated mass and normalized elution time for each identified peptide . This database was used for subsequent high-sensitivity , high-throughput analysis of Cyanothece 51142 samples using the AMT tag approach [60] . LC-MS features from the unfractionated global samples were matched to the rich database built from the fractionated samples to give accurate peptide IDs . The area of each LC elution peak was used as a measure of peptide abundance . Data from the AMT output were imported into the software MDART ( Burnum et al . , unpublished results ) , for filtering using a mass error tolerance of <5 parts per million , delta match score >0 ( a measure of peptide uniqueness ) , match score >−1 , and absolute normalized elution time error <10 , 000 . The resulting 7450 peptides were imported into the software tool DAnTE [61] for further filtering and analysis . Peptide abundances were transformed to log base 2 and mean-centered . A linear regression-based normalization method available in DAnTE was then applied within each replicate category . Peptide abundances were used to infer the corresponding protein abundances through the ‘Rrollup’ algorithm in DAnTE [61] . During the Rrollup step , peptides were excluded if not present in at least 3 of the eight datasets , and Grubbs' outlier test was applied with a P-value cutoff of 0 . 05 to further remove outlying peptides . For increased confidence in protein identifications , each protein was required to be identified by at least 2 unique peptides , resulting in a total of 865 proteins . The minimum observed relative protein abundance value ( 14 , 465 ) was imputed as a crude surrogate for missing data for statistical calculations . Statistical differences between the two samples ( 4 technical replicates of each ) were determined using ANOVA with a P-value cutoff of 0 . 05 ( q<0 . 03 ) in DAnTE [61] . A draft metabolic network of Cyanothece 51142 was reconstructed in SimPheny ( Genomatica , San Diego , CA ) using a previously described automated model-building process [62] . Metabolic reactions and gene-to-protein-to-reaction ( GPR ) associations from other models were incorporated into the reconstruction if good BLAST hits could be found between genes in Cyanothece 51142 and genes in other modeled organisms . Additional reactions were added as necessary to produce known biomass constituents or utilize known nutrients; detailed literature , database , and BLAST searches were then carried out to find genes encoding these reactions in Cyanothece 51142 genome . This resulted in several new GPR associations that were incorporated into the reconstruction . Based on the metabolic reconstruction , a constraint-based metabolic model for Cyanothece 51142 was developed as described in [63] . Fluxes are limited based on several different types of constraints: steady-state mass balance constraints ( Eq . 1 ) , enzyme capacity and thermodynamic constraints ( Eq . 2 ) [10] , given by: ( 1 ) ( 2 ) where S is a stoichiometric matrix for the reaction network , v is a flux vector , and α and β are parameters that limit the capacity and directionality of individual reactions . Flux balance analysis ( FBA ) uses these constraints to identify a flux distribution which maximizes or minimizes an objective function , such as growth rate [10] . Flux variability analysis ( FVA ) can also be used to determine the range of values each flux can take that are consistent with Eq . 1 and 2 , by maximizing and minimizing each flux individually [64] . To further constrain the models based on mRNA or protein expression data , a modified version of the method developed by Shlomi et al . [30] was used . Here , we identified a single flux distribution that best agreed with measured transcriptome and proteome data ( TPD ) and minimized flux usage . Reactions with experimentally measured fluxes belong to set RE ( which included biomass production and exchange fluxes for oxygen , 630 nm and 680 nm photons ) and were constrained to their measured values . Reactions associated with detected proteins were included in the high reaction set ( RH ) . Reactions associated with undetected proteins and genes with low mRNA expression levels ( whose mRNA expression was less than the lowest mRNA expression of detected proteins ) were included in the low reaction set ( RL ) . The method finds a flux distribution that maximizes the number of active reactions ( v≠0 ) and inactive reactions ( v = 0 ) in reaction sets RH and RL , respectively . For reactions in set RH , binary variables x and y indicate whether a reaction is active , meaning its flux is greater than a positive threshold ε ( x = 0 and y = 1 ) , or smaller than a negative threshold -ε ( x = 1 and y = 0 ) for reversible reactions . If both x and y are zero then the reaction is inactive and its flux value is zero . Likewise , a binary variable z is used for reactions in set RL such that if z = 1 then the reaction is inactive ( v = 0 ) . The original method [30] has alternate solutions , which can contain unrealistically high flux values due to the presence of cycles ( e . g . , futile cycles and circulations ) in the network . To identify a solution that minimizes the use of these cycles , the objective function was modified to also minimize the sum of squared fluxes through the network . The mixed integer quadratic programming formulation to identify a flux distribution that best matches TPD while minimizing flux magnitude is given below ( Eq . 3 ) . ( 3 ) Additionally , to find the flux ranges consistent with the TPD , flux variability analysis ( FVA ) was performed by minimizing and maximizing the flux through each reaction in the network . In these FVA simulations , the same constraints described above were included and the binary variables ( x , y , and z ) were further constrained by their optimal values ( xopt , yopt , and zopt ) found in the original problem ( formulation below , Eq . 4 ) . In this study , all model simulations were performed in GAMS software ( General Algebraic Modeling System , GAMS Development Corporation , Washington , D . C . ( 4 )
|
Cyanobacteria have been promoted as platforms for biofuel production due to their useful physiological properties such as photosynthesis , relatively rapid growth rates , ability to accumulate high amounts of intracellular compounds and tolerance to extreme environments . However , development of a computational model is an important step to synthesize biochemical , physiological and regulatory understanding of photoautotrophic metabolism ( either qualitatively or quantitatively ) at a systems level , to make metabolic engineering of these organisms tractable . When integrated with other genome-scale data ( e . g . , expression data ) , numerical simulations can provide experimentally testable predictions of carbon fluxes and reductant partitioning to different biosynthetic pathways and macromolecular synthesis . This work is the first to computationally explore the interactions between components of photosynthetic and respiratory systems in detail . In silico predictions obtained from model analysis provided insights into the effects of light quantity and quality upon fluxes through electron transport pathways , alternative pathways for reductant consumption and carbon metabolism . The model will not only serve as a platform to develop genome-scale metabolic models for other cyanobacteria , but also as an engineering tool for manipulation of photosynthetic microorganisms to improve biofuel production .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"Methods"
] |
[
"biology",
"computational",
"biology"
] |
2012
|
Genome-Scale Modeling of Light-Driven Reductant Partitioning and Carbon Fluxes in Diazotrophic Unicellular Cyanobacterium Cyanothece sp. ATCC 51142
|
We investigate spatiotemporal dynamics of human interphase chromosomes by employing a heteropolymer model that incorporates the information of human chromosomes inferred from Hi-C data . Despite considerable heterogeneities in the chromosome structures generated from our model , chromatins are organized into crumpled globules with space-filling ( SF ) statistics characterized by a single universal scaling exponent ( ν = 1/3 ) , and this exponent alone can offer a quantitative account of experimentally observed , many different features of chromosome dynamics . The local chromosome structures , whose scale corresponds to that of topologically associated domains ( ∼ 0 . 1 − 1 Mb ) , display dynamics with a fast relaxation time ( ≲ 1 − 10 sec ) ; in contrast , the long-range spatial reorganization of the entire chromatin ( ≳O ( 10 2 ) Mb ) occurs on a much slower time scale ( ≳ hour ) , providing the dynamic basis of cell-to-cell variability and glass-like behavior of chromosomes . Biological activities , modeled using stronger isotropic white noises added to active loci , accelerate the relaxation dynamics of chromatin domains associated with the low frequency modes and induce phase segregation between the active and inactive loci . Surprisingly , however , they do not significantly change the dynamics at local scales from those obtained under passive conditions . Our study underscores the role of chain organization of chromosome in determining the spatiotemporal dynamics of chromatin loci .
The three dimensional ( 3D ) structures of chromosome vary with the developmental stage [1] and cell types , which implies that knowledge of chromosome structure and dynamics is key to understanding their link to gene regulation [2] . A well-designed chromosome structure can facilitate long range transcriptional regulation by keeping two distal genomic loci of enhancer and promoter in proximity [3–5] . Hierarchical organization of chromosomes are inferred from the patterns of Hi-C maps which measure mean contact frequencies of cross-linking between DNA segments based on an ensemble of millions of fixed cells . Chromosomes at ∼ 5 Mb resolution are partitioned into alternating A and B type compartments that are enriched with active and inactive loci , respectively [6] . Hi-C data at submegabase resolution offer glimpses into the structure of TADs ( topologically associated domains ) , the functional building blocks of interphase chromosome [7 , 8] . Genome-wide Hi-C maps at even higher resolution of ∼ O ( 10 ) Kb suggests that each type of compartment is associated with distinct epigenetic pattern , further segregating into six sub-compartments [9] . In addition , fluorescence images visualizing real-time chromatin dynamics in vivo [10–13] allow us to decipher the link between structure , dynamics , and function [14–16] . Along with the above-mentioned knowledge from measurements , extensive effort has also been made in developing polymer models for the 3D organization of chromosomes [17–23] and their dynamics [24–30] . For example , ‘strings and binders switch ( SBS ) ’ model , originally proposed to explain many generic behaviors of chromatin within living cells [19] , has recently been further extended to explore the hierarchical chromosome structures [31] and the effects of structural variants on chromatin architecture [23] . More recently , chromatins have been modeled as a block polymer condensed by bivalent or multivalent binding factors , mimicking the binding of transcription factors; while mainly focusing on structural properties , the model has shown how an extended chain is collapsed , and discuss how domains are formed [32] . The loop extrusion polymer model [20 , 21] , based on the knowledge of the convergent orientation of the CTCF-binding motifs , has been used to explain the formation of TADs and predict the contact maps of edited genomes upon deletion of CTCF-binding sites [20 , 21] . There is also a growing trend to integrate the data from Hi-C , fluorescence in situ hybridization ( FISH ) , and epigenetic states into a block copolymer-type model in order to more realistically design 3D chromosome structures and their role in biological function [33–39] . However , homopolymer models with geometrical and topological constraints alone [6 , 28 , 40–43] may suffice in capturing some of the physical bases of chromosome organization . The primary aim of this study is to elucidate the principles underlying the intra-chromosomal dynamics in space and time , which has been underappreciated in theoretical and computational studies than the problem of inferring chromosome structure from Hi-C data . A heterogeneous population of conformational ensemble of chromosomes was generated by using one of the recently proposed heteropolymer models , —Minimal Chromatin Model ( MiChroM ) —whose parameters were trained for the Hi-C data of chromosome 10 ( Chr10 ) from human B-lymphoblastoid cell [22] . To study dynamics of chromosomes we modified the original MiChroM , which is partially self-avoiding with an energetic penalty for each crossing , by imposing a strict self-avoidance constraint and performed Brownian dynamics simulations . Discussing their dynamic properties using various correlation functions , we show that the basic features of the chromatin dynamics reported in the recent experiments [44 , 45] can be explained quantitatively by the crumpled , hierarchical , territorial , summarized as space-filling organization of chromatin chain . Finally , by incorporating active noises onto active loci , we investigate the contribution of activity to the dynamic properties of the interphase chromatin .
We use MiChroM [22] , a 3D coarse-grained heteropolymer model , to study chromosome dynamics at genomic scales greater than 50 Kb . In the model one of the 6 subcompartment types ( B3 , B2 , B1 , NA , A1 , and A2 ) ( see the color barcode above Fig 1A ) , determined based on the correlation between the distinct patterns of interchromosomal contacts and epigenetic information [9] , is assigned to each monomer representing 50 Kb of DNA segment . In the Hi-C map , potential binding sites for CTCF [20] display higher contact frequencies than their local background . The interactions for chromosome are implemented in the model in terms of the energy potentials of ( i ) a homopolymer , ( ii ) monomer type dependent interactions , ( iii ) attractions between loop sites , and ( iv ) genomic distance-dependent condensation energies ( See SI for details ) . We note that due to intra-chromosomal interactions , the effect of the confining sphere used in this model , which gives rise to a volume fraction of 10% ( ϕ = 0 . 1 ) , is not significant enough to alter the chromosome structure and dynamics [28] . To generate a conformational ensemble of chromosomes , we used the low friction Langevin simulation [46] ( see S1 Text ) and sampled the folded conformations of chromosome by collapsing an ensemble of extended chromatin chains . The conformational ensemble of Chr10 , resulting from the enhanced sampling of chromosome conformation , produces a checkerboard pattern which resembles that of the Hi-C contact map [9] ( Fig 1A ) , and it displays the hallmark of space-filling ( SF ) statistics , i . e . , the characteristic scaling of contact probability P ( s ) ∼ s−1 over the intermediate range of genomic distance 1 < s < 10 Mb ( S1B Fig ) . The distribution of Alexander polynomial , |Δ ( t = −1 ) | [47] ( S1D Fig ) , which characterizes the amount of chain entanglement , has the highest mode at |Δ ( t = −1 ) | ≈ 0 , which indicates that the majority of chromosome conformations are free of knots . According to the radial distributions of monomers belonging to the different subcompartment types [22 , 48] , the condensed and transcriptionally inactive loci are buried inside the chromosome , whereas the open and active loci are distributed near the chromosome surface , which is presumably needed to increase the accessibility to transcription factors ( S1E and S1F Fig ) . Because of the nature of frustrated interactions in the heteropolymer model , substantial heterogeneity is expected for the structural ensemble; thus rigorous conformational sampling is not easy to achieve . Nonetheless , the resulting heterogeneity of conformational ensemble can be visualized using clustering analysis over the structures generated . In order to quantify the ( dis ) similarity between two conformations and to perform the clustering analysis for the structures , we use the distance-based root-mean-square deviation ( DRMS , D ) , D α , β = 2 N ( N - 1 ) ∑ i > j ( r i , j α - r i , j β ) 2 . ( 1 ) If DRMS of two distinct chromosome structures , say α and β , is smaller than a cut-off value D c such that D α , β < D c , we consider them similar and group them together into the same cluster . By repeating this grouping process with increasing value of D c we clustered the chromosome structures hierarchically; the result is summarized into a dendrogram ( Fig 1B and S2 Fig ) . When D c reaches 〈 D 〉 ≈ 4 . 5 a , which corresponds to the average DRMS , the distinction between the structures belonging to different clusters or between their contact maps becomes clear ( Fig 1B ) . We will show that the transformation of a conformation in one cluster to those in another cluster beyond the value of DRMS greater than 〈 D 〉 is dynamically a very slow process . Partitioning of the conformations into distinct clusters is a first indication that the configurational space of chromosome is rugged , suggestive of the cell-to-cell variability discovered in the recent single-cell Hi-C data [5 , 49 , 50] . The time-averaged mean square displacement ( MSD ) is a routinely calculated quantity in analyzing the dynamics of cellular constituents in live cell imaging experiments as well as in chromosome studies [44 , 51 , 52] . The time-averaged MSD for i-th locus is defined as MSD ¯ i ( t ) = 〈 | r → i ( t 0 + t ) - r → i ( t 0 ) | 2 〉 t 0 = 1 τ max - t ∫ 0 τ max - t d t 0 | r → i ( t 0 + t ) - r → i ( t 0 ) | 2 , where τmax ( = 4 × 104τBD ∼ 0 . 5 hour: see Methods ) is the longest simulation time . The loci-averaged MSD is then obtained by summing over the loci as MSD ( t ) = ∑ i = 1 N MSD ¯ i ( t ) / N . Substantial dynamical heterogenetiy is present in MSD ¯ i ( t ) for different i ( the inset of Fig 2A and S3 Fig ) . As a result , the dynamics of individual loci is characterized with a different scaling exponent β at long time ( see S3 Fig ) . Dynamics of individual locus , quantified in terms of MSD ¯ i ( t ) depends on the position of locus and varies from one trajectory to another . Nevertheless , the diffusion of chromatin loci is on average characterized by three different time regimes ( Fig 2A ) . ( i ) At short times ( t < 10−2τBD ) , the loci diffuse freely with MSD ∼ t . ( ii ) At the intermediate times , corresponding to the Brownian time t ∼ τBD ∼ a2/D , each locus starts to feel the influence of adjacent loci . ( iii ) For t > 103 τBD , a subdiffusive behavior of MSD ∼ tβ with β ≈ 0 . 4 , spanning at least 2–3 orders of time interval , is observed ( Fig 2A ) . This exponent is in line with the reported values of β = 0 . 38 ∼ 0 . 44 [45] and β = 0 . 4 ∼ 0 . 7 [13] from live human cells . As discussed in other studies [45 , 53] , the exponent β = 0 . 4 of loci-averaged MSD at t > 103τBD can be rationalized using the following argument . The spatial distance ( R ) between two loci separated by the curvilinear distance , s , satisfies R ( s ) ∼ sν , where ν , the scaling exponent [42 , 54] , is ν = 1/2 for the ideal chain obeying the random walk statistics , and ν = 1/3 for the space-filling ( SF ) chain for crumpled globules . Notice that the MSD of a locus in a chain segment of arc length s scales with time t as MSD ∼ tβ ∼ D ( s ) × t ∼ Do × t/s , where the scaling relationship of the diffusion constant of freely draining chain D ( s ) ∼ Do/s is used . Meanwhile , the space taken up by the chain segment of arc length s is described by the relation of MSD ∼ R2 ( s ) ∼ s2ν . These two relations of MSD allow us to relate s with t as s ∼ tβ/2ν , and it follows that MSD ∼ tβ ∼ t1−β/2ν , which leads to β = 2ν/ ( 2ν + 1 ) [45 , 53] . Thus we obtain MSD ( t ) ∼ t 2 ν 2 ν + 1 . ( 2 ) The SF organization of chromosome at intermediate scales ( 1 ≪ s < N2/3 ) implies ν = 1/3 , and hence β = 0 . 4 . A similar argument was used to explain the growth of MSD ( t ) in an entirely different model [39] . Other theories [45 , 55] and a modeling study [26] , which consider interactions to maintain the compactness of the chain structure , lead to the same conclusion . Meanwhile , a high-throughput measurement of chromatin motion tracking has shown MSD ∼ t0 . 5 for yeast chromosomes [11] . Evidently , MSD ∼ t1/2 for ν = 1/2 from Eq 2 , and it is well known that yeast chromosomes obey the random walk statistics ( R ( s ) ∼ s1/2 and P ( s ) ∼ s−3/2 ) , indicative of ν = 1/2 . Therefore , the diffusion exponent of chromosome loci reflects the effect of chain organization of chromatin in chromosome structure [45 , 53 , 55] . The loci-averaged MSD ( t ) is used as a handy probe for chromatin dynamics in experiments [45 , 53 , 55] . However , when a polymer is extraordinarily long just like in the problem of chromatin chain , MSD ¯ i ( t ) of the i-th locus of even a homopolymer depends critically on the position of the locus and its motion exhibit its characteristic scaling behavior at different time regimes with various crossovers [26 , 27 , 56–58] . The scaling behavior of MSD ¯ i ( t ) for different loci ( different i ) at different time regimes can be used to disentangle the dynamics of a polymer chain , e . g . , the diffusion time along the tube that can be hypothesized in melt-like dense polymer environment ( τ e = N e 2 / W ) , Rouse relaxation time ( τR = N2/W ) , and reptation time ( τN = N3/NeW ) , where Ne and W denotes the entanglement length and diffusivities of polymer segments , respectively . A test polymer chain of length N in a highly entangled equilibrium melt ( Ne < N ) [56–58] , exhibits scale-dependent dynamics with multiple crossovers: MSD ¯ N / 2 ∼ t 1 / 2 , MSD ¯ com ∼ t for t < τ e ; MSD ¯ N / 2 ∼ t 1 / 4 , MSD ¯ com ∼ t 1 / 2 for τ e < t < τ R ; MSD ¯ N / 2 ∼ t 1 / 2 , MSD ¯ com ∼ t for τ R < t < τ N ; MSD ¯ N / 2 ∼ t , MSD ¯ com ∼ t for τ N < t . ( 3 ) where the behaviors of time-averaged MSDs were given for the mid-point monomer ( i = N/2 ) and the center of mass ( i = com ) . Our chromosome model differs from polymer melts and thus the above scalings of MSD ¯ i ( t ) for an ideal test chain ( ν = 1/2 ) in polymer melts in principle do not apply to our chromosome model comprised of non-ideal subchains ( ν = 1/3 ) . Nevertheless , the crossover behaviors at distinct characteristic times ( τe , τR , τN ) discussed in Eq 3 is still be of great use to illuminate the dynamics of our chromosome model . Two points are worth making . ( i ) The distribution of Alexander polynomial indicates that our chromatin chain is rarely entangled ( S1D Fig ) . Thus τe is not a quantity relevant to our chromosome model . Furthermore , MSD ¯ com ∼ t for the entire simulation time ( Fig 2C ) , which is also an indication of the absence of the crossover . ( ii ) For an ideal Rouse chain , the chain relaxation time ( the Rouse time , τR ) can be estimated from MSD ¯ com = 〈 R e e 2 〉 at t = 3τR/4 , where 〈 R e e 2 〉 is the mean square end-to-end distance of the chain [57 , 58] . In our case , MSD of ‘com’ still has not reached 〈 R e e 2 〉 even at the maximum simulation time , i . e . , MSD ¯ com ( t = τ max ) < 〈 R e e 2 〉 , which indicates that the total simulation time of our study is still shorter than the Rouse relaxation time ( τmax < τR ) . Taken together , the two critical time scales for equilibration , the reptation and Rouse relaxation times , of our model are substantially longer than the typical time scales relevant for cellular processes such as cell doubling times ( see below ) . The global dynamics of chromosomes are not only heterogeneous but also are too slow for a full equilibration . Thus , it is reasonable to view that chromosome dynamics is sluggish , glass-like and occurs out of equilibrium . Correlation functions are a general tool to study the dynamics of complex systems [59] , and have been used in experimental analysis of genomes or chromosomes [10 , 12 , 60 , 61] . Here , we adopt this strategy to study the spatio-temporal dynamics of our chromosome model . Recently , displacement correlation spectroscopy ( DCS ) using fluorescence has been employed to study the dynamics of whole chromosomes in the nucleus , revealing that coherent motion of the μm-sized chromosome territories could persist for μs to tens of seconds [10] . We adopted the same approach used in DCS and studied the spatial correlation in the intra-chromosomal dynamics generated from our simulations . The spatial correlation between chromatin loci is evaluated using CsΔt ( r ) =〈 ∑i>j[ Δr→i ( t;Δt ) ·Δr→j ( t;Δt ) ]δ ( ri , j ( t ) −r ) ∑i>jδ ( ri , j ( t ) −r ) 〉t , ( 4 ) which quantifies the displacement correlations between loci separated by the distance r over the time interval Δt . C s Δ t ( r ) decays more slowly with increasing Δt . The correlation length calculated using l c = ∫ 0 ∞ [ C s Δ t ( r ) / C s Δ t ( a ) ] d r , shows how lc increases with Δt ( Fig 3B ) . To demonstrate an image of displacement correlation over the structure , we project the displacement vectors of the monomers near the equator of the confining sphere ( −a ≤ z ≤ a ) onto the xy plane , and visualize the dynamically correlated loci moving parallel to each other by using the vector field with a similar color ( see Fig 3C ) . If Δt < 100 τBD , the spatial correlation of loci dynamics is short-ranged and the displacement vectors appear to be random . In contrast , multiple groups of coherently moving loci that form substantially large domains ( ∼ 5a ≈ 0 . 75 μm ) emerge at a longer waiting time ( Δt > 500 τBD ) . We also calculated C s Δ t ( r ) for the Rouse chain as a reference ( see SI ) . Just like our chromosome model , C s Δ t ( r ) for the Rouse chain decays more slowly over the distance r with increasing Δt ( S4A Fig ) , and the correlation length lc increases monotonically with Δt as well ( S4B Fig ) . However , this very feature differs from the one observed in the experiment [10] where lc displayed nonmonotonic change with Δt . In fact , the experimentally observed nonmonotonic change of lc is obtained by incorporating active noise to the model , which will be discussed in the section that follows ( see below , Effects of active noise on chromosome dynamics ) . In parallel to the spatial correlation functions calculated above , a time-correlation function that can potentially characterize the chromatin dynamics has recently been proposed [12 , 60] for the displacement vectors of the same locus or two distinct loci for varying lag times . However , we find the resulting time-correlation function ( mean velocity auto-correlation function ) is not so informative in the sense that it is barely discernible from that of the ideal Rouse chain ( see S1 Text and S5 Fig for details ) . Diffusion of heterochromatin-rich loci is slower than euchromatin-rich loci [45] . The time-averaged MSD ( MSD ¯ i ) exhibits substantial dispersion among different loci ( Fig 2A inset and S3 Fig ) , and the overall mobility of loci depends on the subcompartment types ( see Fig 4A ) . In our chromosome model we find that the A-type loci , which are less condensed and distributed closer to the chromosome surfaces , diffuse faster than the B2 and B3 type loci . The dispersion of MSD ¯ i shown in the inset of Fig 2A is the outcome of both different sub-compartment types and different genomic positions of loci . Although the diffusivity is greater for the active loci , they still have the same β = 0 . 4 for t > 103τBD ( Fig 4A inset ) . The relation β = β ( ν ) = 2ν/ ( 2ν + 1 ) suggests that the exponent ν representing the chain organization is the sole determinant of the diffusion exponent ( β ) characterizing the global motion . We will show that this conclusion holds good even in the presence of active noise incorporated into the chromatin dynamics ( see below ) . Decomposing the spatial correlation C s Δ t ( r ) into A and A , B and B , or A and B type loci ( S6A Fig ) , we find that the corresponding correlation length lc of A-type loci is greater than B-type loci for Δt ≳ τBD ( Fig 4B ) . This suggests that the motion of A-type loci is more coherent; however , this picture changes completely when “activity” is incorporated into the model ( see below ) . The time evolution of the averaged mean square deviation of the distances between two loci with respect to the initial value ( see Fig 5A and the caption for the definition of δ ( t ) ) was calculated to discuss the dynamical stability of chromosome structure . Within our simulation time τmax , the largest value δmax ( = 4 . 0 ± 0 . 3 a ) is smaller than the value , D c = 4 . 5 a , which was chosen to define different conformational clusters in Fig 1B . An extrapolation of δ ( t ) to δ ( τ c ) = D c gives an estimate of τc ≈ 105 × τBD ≈ 1 . 4 hours , which is a long time scale considering that most cells of adult mammals spend about 20 hours in the interphase [62] . From the definition of , δ ( t ) , it follows that limt→∞〈δ ( t ) 〉 = δeq . Here , δeq is finite , and 〈⋯〉 is an ensemble average , meaningful only if the equilibrium is reached . We estimate δeq assuming that the long time limit of the mean deviation of the distance between two loci is approximately the mean end-to-end distance between the loci . Thus , lim t → ∞ 〈 ( r i j ( t ) - r i j ( 0 ) ) 2 〉 ∼ R i j 2 where Rij is the mean end-to-end distance between ith and jth loci . For |i − j| ≫ 1 , we expect that R i j 2 ∼ a 2 | i - j | 2 ν . Consequently , δeq can be calculated using δ eq 2 = 2 N ( N - 1 ) ∑ s = 1 N - 1 ( N - s ) R 2 ( s ) = 2 a 2 N ( N - 1 ) ∑ s = 1 N - 1 ( N - s ) s 2 ν . For N = 2712 , and with ν = 1/3 we estimate δeq ≈ 9 . 4 a , which is greater than the value ( δmax ≈ 4 . 0 a ) reached at the longest times ( Fig 5A ) . An upper bound of δeq for an ideal Rouse chain is 16 . 4 a ( see SI ) . These considerations suggest that the chromosome dynamics falls short of equilibrium on the time scale of a single cell cycle . Relaxation dynamics of chromatin domain should be scale-dependent , which is quantified using the time evolution of intermediate scattering function Fk ( t ) [59 , 63] , the van Hove correlation function in Fourier space , calculated at different length scale ( ∼ 2π/k ) ( Fig 5B ) : Fk ( t ) =〈 〈 1N∑meik→·r→m ( t+t0 ) ∑ne−ik→·r→n ( t0 ) 〉|k→| 〉t0 , ( 5 ) where 〈 〈 … 〉 | k → | 〉 t 0 is an average over t0 and over the direction of vectors k → with magnitude k ( = | k → | ) . Two points are worth making for Fk ( t ) at varying k . ( i ) The chromatin chains at high wave number ( at local scale ) relax fast , which implies that chromatin chains are locally fluid-like ( 2π/k ≲ a ) . Although the structure of TAD is highly coarse-grained in our study ( TADs , whose median size is 880 Kb [7] , is represented by only 18 beads ) , this fluid-like dynamics at local scale is in accord with the recent experimental finding on the structural deformation of chromatin fibers within TADs [8 , 64] . ( ii ) The spatial organizations of chromatin chains over intermediate to global scales ( 2π/k ≫ a ) are characterized by slow relaxation dynamics . This scale-dependent relaxation time is reminiscent of a similar finding in random heteropolymers [65] . Relaxation time ( τ ) of a subdomain of size ξ = 2π/k is estimated using τ k = ∫ 0 ∞ [ F k ( t ) / F k ( 0 ) ] d t , which can in turn be related to the number of coarse-grained monomers comprising the subdomain as ξ ∼ 2π/k ∼ sν . Since the chromosome domain loses memory of the initial conformation by spatial diffusion ( instead of reptation ) , the relaxation time τ is expected to obey τ ∼ ξ2/Deff ∼ ( sν ) 2/ ( D0/s ) ∼ s2ν+1 , thus τ ∼ s5/3 for the chromosome structure that obeys SF statistics ( ν = 1/3 ) . The size-dependent relaxation times calculated for our chromosome model indeed scales with the domain size as τ ∼ s5/3 ( cyan symbols and solid line in Fig 6C ) . Effects of biological activities on the chromosome structure , such as ATP hydrolysis-driven non-conservative forces exerted by cohesins [20] , are only implicit in the original MiChroM in terms of the differential energy parameters for the loci of A , B subcompartment types . Thus , it could still be argued that such a model misses the most critical component of living systems . Live cells abound in a plethora of biological activities such as replication , transcription , and error-correcting dynamics . While these processes produce local directionality , when mapped onto our model that has coarse-grained 50 Kb of DNA into a single bead , the effects of vectorial forces on the surrounding environment at length and time scales greater than the correlation length and time of active noises can be assumed isotropic . This is supported by Javer et . al . [66] who also pointed out , by performing an experimental study of locus-dependent diffusion coefficient in E . coli . , that the contribution of “ballistic” motion to MSD beyond the time scale of seconds is negligible . We study how an increased noise strength on the active loci ( A1 and A2 ) occupying 40% of loci population for Chr10 , which resuts in the breakdown of fluctuation-dissipation theorem [67 , 68] , affects the dynamical properties of entire chromosome . To model the active noise , we increased the noise strength from 〈 R → i ( t ) · R → j ( t ′ ) 〉 = 6 D i 0 δ i j δ ( t - t ′ ) to 〈 R → i ( t ) · R → j ( t ′ ) 〉 = 12 D i 0 δ i j δ ( t - t ′ ) , following the recent literature [69 , 70] . The model that incorporates active noises as described above has led to two important results . ( i ) The disproportionate increase in the mobility of A and B type loci promotes the phase segregation of the two loci types ( see Fig 6B , S3 Fig , and compare S1 and S2 Movies ) . The active noises push A-type loci towards the surface of the chromosome , and B-type loci are pulled towards the center to offset this effect . ( ii ) More quantitatively , we find that the average MSD of A1 loci exhibits ∼ 70% increase relative to the passive case ( Fig 6A left panel ) , while the diffusion exponent ( β ≈ 0 . 4 in MSD ∼ tβ ) remains unaltered ( Fig 6A right panel and S3 Fig ) . The finding that the inclusion of active noises increases the amplitude of the MSD without altering the diffusion exponent ( β ≈ 0 . 4 ) is in accord with an experiment on bacterial chromosomes performed with and without ATP depletion [44] . In addition , the finding is consistent with the MSD data reported for a live human Hela cell [45] , where chromatin loci at the nuclear periphery and interior , corresponding to the heterochromatin and euchromatin , displayed diffusion exponents β = 0 . 39 and 0 . 41 , respectively , although the MSD of the euchromatin was significantly greater . We however also note that the diffusion exponent β = 0 . 32 ± 0 . 03 was reported for the whole genome of ATP-depleted HeLa cells [10] , which is qualitatively different from β ≈ 0 . 4 ( see S1 Text and S8 Fig for detailed analyses of the experimental data reported in [10] ) . In terms of Fk ( t ) , the active noises mainly influence the chain relaxation associated with the low frequency modes . For the high frequency modes or at local length scales ( k ≳ 2π/3a ) , Fk ( t ) is practically indistinguishable between active and passive cases ( S9 Fig ) . The chromatin segments in the presence of active noise , on average , relax faster when the size of the segment is greater than the sub-Mb . A comparison of the relaxation times in Fig 6C under passive and active conditions highlight this difference . Similarly , the effect of active noise on the correlation length ( lc ) is evident only at a large lag time ( Δt ) . We find that in contrast to the passive case , lc changes nonmonotonically with Δt . There is no distinction between the effects of passive and active noises on lc for small Δt; however , deviation between the two cases becomes evident for Δt ≳ 103τBD ≈ 50 sec ( Fig 6D ) . Importantly , a similar dependence of correlation length on Δt has been discussed in DCS measurement on genome-wide dynamics of live cell [10] . To dissect the contribution from the loci of each subcompartment type in the presence of active noises , we again calculated the spatial correlation C s , A B Δ t , C s , B B Δ t , C s , A A Δ t ( S6B Fig ) and the corresponding correlation lengths ( lc ) ( Fig 6E ) . At short time scale ( t < 500τBD ) , A-type loci display slightly stronger self-correlations than B-type loci . In stark contrast to the passive case ( Fig 4B ) , however , at Δt > 500τ active noises disturb the spatial correlations between active loci , which subsequently reduces the correlation of entire structure . Compared to the thermal noise ( Fig 4B ) , the active noises randomize the global structure of chromatin chain more efficiently , which shortens the correlation length at sufficiently large lag time .
Despite a great amount of complexity inherent to its size and heterogeneous interactions that give rise to various dynamic behaviors at different time and length scale and crossovers , chromatin chain folded into a heterogeneous ensemble of chromosome conformations via protein mediated interactions can be viewed from a perspective of polymer physics as a very long heteropolymer chain collapsed in a poor solvent condition [56–58] . Our study highlights the importance of chromosome architecture in determining the subdiffusive behavior and dynamic correlations between distinct loci . Most importantly , we have shown that structure alone explains many of the dynamical features observed in live cell experiments [10 , 13 , 44 , 45] . In other words , conformational properties of chromatin chain dictate the dynamics of chromosome . Remarkably , several static and dynamic properties of the model , including R ( s ) ∼ sν , P ( s ) ∼ s−3ν , MSD ( t ) ∼ t2ν/ ( 2ν+1 ) , τ ( s ) ∼ s2ν+1 , and 〈 X p 2 〉 ∼ p - ( 1 + 2 ν ) ( X → p is the p-th Rouse mode . See S1 Text and S5D Fig for the details ) are fully explained by the SF organization characterized by the single scaling exponent ν = 1/3 , offering a unified perspective on the link between the structure and dynamics of chromosomes . The relaxation time ( τ ) of the chromatin domain spans several orders of magnitude depending on its genomic length ( s ) , satisfying the scaling relation τ ∼ s5/3 ( Fig 6C ) . To be more concrete ( see Fig 6C ) , while local chromatin domains of size s≲O ( 1 ) Mb , a scale corresponding to TADs , reorganize on the time scale of t < 10 3 τ BD ∼ O ( 1 ) seconds , it takes more than hours to a day for an entire chromosome chain ( ≳ 100 Mb ) to lose its memory of the initial conformation as long as the chromosomes are in the interphase with no significant vectorial active noises . This timescale of relaxation is expected to increase even further at higher volume fractions [28] . Under in vivo conditions , with 46 chromosomes segregated into chromosome territories , the time scale for relaxation would be considerable . The effects of active noise on chromatin dynamics [10 , 44] deserve further discussion . While the isotropic active noises modeled in this study enhance chain fluctuations and structural reorganization , their effect on chromatin domain manifests itself only on length scales greater than 5 . 5 a ( ≈ 0 . 8 μm ) , and on a time scale greater than 50 sec ( Fig 6D ) . Our finding is reminescent of the microrheology measurements on active cytoskeletal network [71] , where the effect of myosin activity could be observed only at low frequency regime of the loss modulus . Of course , the active noise in live cell nuclei is still not a scalar , and thus it remains a challenge to model their vectorial nature in the form of force dipole or vector force in the context of chromatin dynamics [69] . Vector activities promote super-diffusive motion ( MSD ¯ i ∼ t β with 1 < β < 2 ) , and could in principle elicit a qualitative change in the dynamical scaling relations as in the mitotic phase . Still , the dynamic scalings discussed in this study ( e . g . , MSD∼ t0 . 4 ) are in good agreement with those observed in interphase chromatins of live cells [13 , 45] . There could be many different explanation for this observation , but we reason as follows . In terms of power generated in a cell , the passive ( thermal ) power Wp ∼ kBT/ps is many orders of magnitude greater than the active power ( e . g . , molecular motors , Wa ∼ 20 kBT/10 ms [62] ) . At least in the interphase , the gap between the total passive and active power is substantial; the number of active loci ( Na ) is smaller than the number of passive loci ( Np ) , rendering the total passive power much greater than the active power ( NpWp ≫ NaWa ) . Thus , the total energetic contribution of the biological activities during the interphase to the chromosome structure would be insignificant compared to thermal agitation . Taken together , even in the presence of biological activities , as long as the scaling exponent ν = 1/3 characterizing the chromosome structure is unaltered , the various dynamical scaling behaviors remain intact . To recapitulate , we have shown that the SF organization ( ν = 1/3 ) adopted by a block-copolymer type model of chromosome alone suffices to explain many of the experimentally observed loci dynamics of human interphase chromosomes . The average behaviors of chromatin dynamics that we have drawn here should not depend critically on the details of the chromosome model being used . One should be able to confirm them as long as a chromosome designed using those models maintains crumpled architectures displaying SF statistics with ν = 1/3 . On one hand , despite seemingly a daunting problem at first sight , many aspects of chromosome dynamics can be quantitatively explained and predicted using purely physical argument based on the basic concepts of polymer physics . This means that if care is taken , even the dynamics of a highly complex biological object like chromosome can be deciphered using the physical law as far as the global dynamics averaged over the large ensemble is concerned . On the other hand , experimental measurement should either be made at a higher resolution in space and time or be specific to genomic loci in individual cells , if one were to extract dynamical information relevant for specific biological function of chromosomes beyond the fractal dimension of chain organization .
To build the model of chromosome 10 of human lymphoblastoid cell and study its dynamical behaviors , we used the energy potentials and parameters of MiChroM , a type of block-copolymer ( heteropolymer ) model . The coarse graining of chromatin leads to N = 2712 loci with the diameter of each locus being a ≈ 150 nm , so that a single locus represents 50 Kb of DNA . The inverse mapping of the Hi-C map to the ensemble of chromosome structures was carried out by sampling the conformational space using low-friction Langevin simulations [46] . The generated structures exhibit the characteristic scaling of the contact probability , P ( s ) ∼ s−1 , and show the spatial distribution of A/B compartment as well as the plaid pattern noted in Hi-C experiments . Whereas the original study of MiChroM allowed the chain-crossing with an energetic penalty for the purpose of sampling the conformations whose population reproduces the Hi-C map , we imposed a strict chain non-crossing constraint on the chromosome structures and performed Brownian dynamics simulations to study the dynamics of chromatin when the conformational sampling was completed . The mapping from simulation times to the physical times deserves a few remarks . The apparent viscosity of nuclear environment varies among different experimental reports within an order of magnitude: η = 1–3 cP [72] , 3 cP [73] , 7 cP [74] , and 10 cP was assumed in modeling chromosome dynamics [32] . In the model employed in this study , each monomer represents 50 Kb genomic region , which is mapped to the diameter of a = 150 nm . Assuming that the nuclear viscosity η = 7 cP , the Brownian time of single particle τBD = 3πηa3/kBT ≈ 50 ms . Therefore , the longest simulation time in this study τmax = 4 × 104 τBD corresponds to 0 . 5 hour . At 0 . 5 second , MSD measured in the nucleus of HeLa cells is in the range of 0 . 01–0 . 015 μm2 in the experiment ( see Fig 2E in Ref . [45] ) ; correspondingly , at t = 10 × τBD ≈ 0 . 5 second , we get MSD ≈ 0 . 96 a2 ≈ 0 . 022 μm2 in our simulation ( see Fig 2A ) . Clearly , they are within the same order of magnitude . Thus , the estimate of physical time from our simulation results is sufficient for the present purpose of our study , given that the model itself is significantly coarse-grained . In comparison to the time scale estimates for chromosome dynamics in other studies [25 , 29] , the Brownian time τBD , albeit a large uncertainty , is roughly mapped to 50 ms in physical time ( τBD ≈ 50 ms , which is the value estimated from η ≈ 7 cP and monomer size a = 150 nm . ) . Further details of the energy function and simulation algorithm are provided in the Supporting Information ( S1 Text ) .
|
Chromosomes are giant chain molecules made of hundreds of megabase-long DNA intercalated with proteins . Structure and dynamics of interphase chromatin in space and time hold the key to understanding the cell type-dependent gene regulation . In this study , we establish that the crumpled and space-filling ( SF ) organization of chromatin fiber in the chromosome territory , characterized by a single scaling exponent , is sufficient to explain the complex spatiotemporal hierarchy in chromatin dynamics as well as the subdiffusive motion of the chromatin loci . While seemingly a daunting problem at a first glance , our study shows that relatively simple principles , rooted in polymer physics , can be used to grasp the essence of dynamical properties of the interphase chromatin .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"relaxation",
"time",
"chromosome",
"structure",
"and",
"function",
"monomers",
"chromosome",
"mapping",
"materials",
"science",
"epigenetics",
"molecular",
"biology",
"techniques",
"macromolecules",
"chromatin",
"research",
"and",
"analysis",
"methods",
"polymers",
"polymer",
"chemistry",
"relaxation",
"(physics)",
"gene",
"mapping",
"chromosome",
"biology",
"gene",
"expression",
"chemistry",
"molecular",
"biology",
"genetic",
"loci",
"physics",
"cell",
"biology",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"materials",
"chromosomes"
] |
2018
|
Chain organization of human interphase chromosome determines the spatiotemporal dynamics of chromatin loci
|
In mammalian cells , AU-rich elements ( AREs ) are well known regulatory sequences located in the 3′ untranslated region ( UTR ) of many short-lived mRNAs . AREs cause mRNAs to be degraded rapidly and thereby suppress gene expression at the posttranscriptional level . Based on the number of AUUUA pentamers , their proximity , and surrounding AU-rich regions , we generated an algorithm termed AREScore that identifies AREs and provides a numerical assessment of their strength . By analyzing the AREScore distribution in the transcriptomes of 14 metazoan species , we provide evidence that AREs were selected for in several vertebrates and Drosophila melanogaster . We then measured mRNA expression levels genome-wide to address the importance of AREs in SL2 cells derived from D . melanogaster hemocytes . Tis11 , a zinc finger RNA–binding protein homologous to mammalian tristetraprolin , was found to target ARE–containing reporter mRNAs for rapid degradation in SL2 cells . Drosophila mRNAs whose expression is elevated upon knock down of Tis11 were found to have higher AREScores . Moreover high AREScores correlate with reduced mRNA expression levels on a genome-wide scale . The precise measurement of degradation rates for 26 Drosophila mRNAs revealed that the AREScore is a very good predictor of short-lived mRNAs . Taken together , this study introduces AREScore as a simple tool to identify ARE–containing mRNAs and provides compelling evidence that AREs are widespread regulatory elements in Drosophila .
Gene expression is extensively regulated at both transcriptional and posttranscriptional levels . In the cytoplasm , numerous mechanisms act on mRNAs to ensure their proper localization , translation and stability [1] . Together with the rate of transcription , the lifespan of an mRNA is a key determinant of the level at which any given gene is expressed . Half-lives differ widely between transcripts , ranging in human cells from 5 minutes to >10 hours [2] , [3] . In yeast , mRNAs are degraded more rapidly and their half-lives range from 3 to >90 minutes [4] . AU-rich elements ( AREs ) are well-characterized cis-acting regulatory sequences that strongly accelerate the degradation of mammalian mRNAs . AREs were initially discovered in 3′ untranslated regions ( UTRs ) of short-lived transcripts encoding cytokines [5] , [6] , and since have been proposed to reside in 5–8% of all transcripts [7] . However , the frequency of functional AREs in a given cell type is certainly lower because genome-wide measurements of mRNA decay rates showed that the presence of AU-rich sequences correlates only to a limited extent with rapid mRNA decay: In primary human T-cells , only about 25% of mRNAs with AU-rich sequences were found to decay rapidly [2] , and in the hepatocellular carcinoma cell line HepG2 this proportion was only 10–15% [8] . Although there is no strict consensus sequence for AREs , the following key motifs have been identified: AUUUA pentamers that frequently occur in multiple copies , which may overlap or localize in close proximity [6] , [9]; a related nonameric motif UUAUUUAUU or UUAUUUA ( U/A ) ( U/A ) , which is strongly linked to rapid mRNA decay [10] , [11] , [12] , [13]; and a generally U-rich or AU-rich context required for maximum efficiency of either pentamers or nonamers [9] . AREs can be distinguished according to different deadenylation kinetics , which gave rise to a widely used classification published by Shyu et al . [14] . Class I AREs ( e . g . c-myc , c-fos ) contain a few scattered pentamers within a larger U- or AU-rich context , and mediate synchronous deadenylation indicative of a distributive exoribonuclease . Class II AREs ( e . g . GM-CSF , IL-3 and TNFα ) have a cluster of 4–7 partially overlapping pentamers within a U-rich context , and mediate asynchronous deadenylation indicative of a processive exoribonuclease . Class III AREs ( e . g . , c-jun ) lack pentamers and have been less well characterized . Khabar et al . proposed an alternative classification of AREs into five groups based on the number of overlapping AUUUA pantamers [15] . This classification has been used to mine databases for the occurrence of ARE-regulated genes [7] , yet the functional implication of this classification has not been thoroughly tested . ARE-mediated mRNA decay ( AMD ) depends on specific RNA-binding proteins ( BPs ) that recognize AREs and target the mRNA for rapid degradation [16] . The Tis11 zinc finger proteins are ARE-BPs with a major role in AMD . The mammalian Tis11 family comprises TTP [17] , BRF1 [18] and BRF2 [19] , all of which are potent inducers of mRNA degradation . These proteins share a highly conserved tandem C3H zinc finger domain required for RNA binding . TTP is the best characterized member of this family and acts as a suppressor of inflammation in mice by controlling the expression of tumor necrosis factor-α ( TNFα ) [17] . Further studies showed that TTP causes the degradation of many additional mRNAs related to the immune response ( reviewed in [20] ) . TTP induces the degradation of its target mRNAs by recruiting the components of the general RNA degradation machinery such as the exosome [21] , the decapping complex [22] and the Ccr4-Caf1-Not deadenylation complex [23] . Moreover , TTP is regulated through phosphorylation by the p38-MAPK – MK2 kinase cascade . Direct phosphorylation by MK2 causes binding of 14-3-3 adaptor proteins and decreases the activity of TTP [24] , [25] by interfering with the ability of TTP to recruit the Ccr4-Caf1-Not deadenylation complex [26] , [27] . In turn , the protein phosphatase 2A dephosphorylates TTP and thereby activates AMD [28] . Very little is known about AMD in Drosophila melanogaster . So far , only the mRNAs encoding CecA1 and bnl were shown to contain a functional ARE [29] , [30] , [31] . The CecA1 ARE binds to Tis11 , the homologue of TTP in D . melanogaster , which in turn promotes rapid degradation of CecA1 mRNA by enhancing deadenylation [29] . Interestingly , expression of mammalian TTP could compensate for the knock down of Tis11 in Drosophila cells [30] , suggesting evolutionary conservation . While the regulation of CecA1 mRNA degradation has been well characterized , there is no experimental study addressing more generally the role of AMD in D . melanogaster . Here we report the development of AREScore , a software application by which mRNAs can be assessed for the presence of AREs . After validating the AREScore using half-life measurements of human and mouse mRNAs , the transcriptome-wide AREScore distribution was analyzed across 14 metazoan species . The AREScore was then applied to the analysis of AMD in Drosophila SL2 cells . By combining biochemical and bioinformatic approaches , we provide evidence for a specific set of mRNAs regulated by Tis11 , and for the broader role of AREs in controlling mRNA degradation in D . melanogaster .
With the aim to identify genes containing AREs in any given set of sequences , we developed an algorithm termed AREScore , schematically depicted in Figure 1A . Its purpose is to provide a numerical measure of the potential strength of an ARE , and assess the occurrence of AREs on a transcriptome-wide level . The AREScore is based on quantifying three typical features of AREs: the number of AUUUA pentamers , the proximity between pentamers , and the presence of a region with high AU content surrounding AUUUA pentamers . The UUAUUUA ( U/A ) ( U/A ) nonamer was not counted as a separate parameter because it largely corresponds to two overlapping AUUUA pentamers or a pentamer within a region of high AU content . The algorithm first counts AUUUA pentamers and attributes a fixed value of 1 for each pentamer to generate a basal score . It then calculates the distance between neighboring pentamers , and adds a value to the basal score if pentamers are close to each other . Likewise , a value is added if pentamers are located within a region of high AU content , herein termed an AU-block . To increase the flexibility of AREScore , users can change the values that are added to the basal score , and alter the settings that define an AU-block . Thereby users can adapt the algorithm to their needs and particular questions . A web-based version of AREScore is available at http://arescore . dkfz . de/arescore . pl . To validate the algorithm , we calculated the AREScore for every human mRNA in the RefSeq database with a 3′UTR length ≥10 nucleotides ( nt ) , whereby many falsely annotated 3′UTRs could be excluded from the analysis . In Figure 1B , the AREScore was then compared to previously measured mRNA half-lives in human DG75 B-cells [32] . The AREScore shows a slight , but statistically highly significant , negative correlation with mRNA half-life ( Spearman rank correlation coefficient RS = −0 . 155 , p<0 . 0001 ) . The correlation was more apparent when mRNAs were classified into groups with similar AREScores and the average half-life was plotted for each group ( Figure 1C ) . We then used Receiver Operating Characteristic ( ROC ) analysis to assess the predictive power of the AREScore in this dataset ( Figure 1D ) . Every possible AREScore value was tested for its ability to discriminate the 10% most short-lived mRNAs from the 10% most long-lived ones . For instance , mRNAs with an AREScore ≥3 . 9 make up 53% of the short-lived mRNAs ( true positive rate ) , but only 14% of the long-lived mRNAs ( false positive rate ) . By plotting true positive rate against false positive rate for every possible AREScore , the ROC curve is obtained . The area under this curve ( AUC ) corresponds to the probability that a random short-lived mRNA has a higher AREScore than a random long-lived mRNA . With a value of 0 . 75 , the AUC is well above that of a random predictor ( AUC = 0 . 5 ) . In a similar manner , we compared the AREScore of mouse mRNAs with half-lives measured previously in mouse NIH3T3 fibroblasts [33] . This analysis again showed a weak but highly significant negative correlation between AREScore and mRNA half-life ( Figure 1E , RS = −0 . 147 , p<0 . 0001 , and Figure 1F ) . With an AUC of 0 . 73 , the ROC curve of the mouse dataset ( Figure 1G ) is very similar to the curve of the human dataset ( Figure 1D ) . Taken together , the comparison of AREScores with genome-wide measurements of mRNA half-lives showed that mRNAs with a high AREScore are more likely to be short-lived , both in human and mouse cell lines . To further validate the AREScore , we analyzed a set of transcripts that we had previously identified as TTP-associated mRNAs in mouse macrophages using RNA-immunoprecipitation [13] . Figure 1H shows that the AREScore is very high among the 135 TTP-associated mRNAs ( median 7 . 8 ) compared to the entire mouse transcriptome ( median 1 . 3 ) or a more stringent control set of randomly chosen , concatenated 3′UTR sequences ( median 4 . 65 ) whose lengths were matched to the lengths of the TTP-associated 3′UTRs . To test whether the AREScore distribution of the TTP-associated mRNAs was statistically different from the controls , we compared the frequency of mRNAs with an AREScore <4 and ≥4 in 2×2 contingency tables ( Tables S1 and S2 ) . P-values were calculated either by χ2-test or Fisher's exact test , and found to be <0 . 0001 for both comparisons . Thus , the AREScores of the TTP-associated mRNAs were significantly higher than the AREScores of both control groups . This confirmed that the AREScore is a useful tool to identify ARE-containing mRNAs . Having the AREScore at hand as a numerical tool to estimate the abundance and strength of AREs in any given genome , we calculated the AREScore of all annotated transcripts with a 3′UTR length ≥10 nt for Homo sapiens and four important metazoan model organisms , Caenorhabditis elegans , D . melanogaster , Danio rerio , and Mus musculus . The analysis shows that in all five species , the vast majority of mRNAs have a score below 4 ( Figure 2A ) . Differences became apparent when frequencies were plotted on a logarithmic scale ( Figure 2B ) . The highest AREScore is 17 . 4 in C . elegans and 34 . 3 in D . melanogaster , whereas in the two mammalian species AREScores go beyond 60 . These differences are also visible in the plot of cumulative frequencies ( Figure 2C ) , which shows the highest prevalence of low AREScores in C . elegans and of high AREScores in H . sapiens . It was interesting to note that the 3′UTR length follows a similar pattern ( Figure 2D ) , with C . elegans having the by far shortest 3′UTRs ( median: 140 nt ) , followed by D . melanogaster ( 207 nt ) and D . rerio ( 402 nt ) , and considerably longer 3′UTRs in the two mammalian species ( 704 nt in mouse , 804 in man ) . This analysis shows that mRNAs with high AREScores as well as long 3′UTRs are more abundant in the two mammalian species , which likely reflects the need for additional elements regulating gene expression . To test whether AREs are truly enriched in any of the transcriptomes we analyzed , we compared the AREScore distribution in different species with sets of randomized sequences that have identical A/T/G/C contents and length distributions ( Figure 3 and Figure S1 ) . This comparison revealed that mRNAs with high AREScores ( ≥10 ) are overrepresented in the transcriptome of H . sapiens ( Figure 3A ) . In D . melanogaster , the enrichment already starts at an AREScore of 4 ( Figure 3B ) , whereas there is no enrichment of mRNAs with higher AREScores in C . elegans ( Figure 3C ) . We then expanded this analysis to the transcriptomes of 11 additional species , covering most of the major branches of metazoan evolution ( Figure S1 ) . Only for Annelida and Crustacea , no properly annotated transcriptomes were available . In the 14 species analyzed , the frequency of mRNAs with an AREScore ≥10 was compared to the frequency of AREScores ≥10 in sets of randomized control sequences ( Figure 3D ) . mRNAs with an AREScore ≥10 were found to be overrepresented in the transcriptomes of H . sapiens , M musculus , Gallus gallus ( chicken ) , Danio rerio ( zebrafish ) and D . melanogaster . This is reflected by a positive Φ coefficient , a measure for how strongly AREScores ≥10 are associated with the actual transcriptome as compared to the randomized control . In all these cases , the difference was significant as determined by χ2-test . mRNAs with an AREScore ≥10 were also more abundant in Ixodes scapularis ( deer tick ) , although in this case the difference was statistically not significant . In the 8 other species analyzed , mRNAs with an AREScore ≥10 were less abundant than in the randomized control sequences . Thus , our analysis suggested that AREs were selected for during the evolution of several vertebrate species ( Xenopus laevis being the exception ) as well as D . melanogaster . Given that we found AREs to be overrepresented in the D . melanogaster transcriptome ( Figure 3B ) and that little is known about the general importance of AMD in this organism , we decided to experimentally address the functional relevance of the AREScore in Drosophila . We first established an assay to measure AMD in D . melanogaster SL2 cells by generating firefly luciferase ( FL ) reporter genes containing the ARE of mouse interleukin ( IL ) -3 in the 3′UTR ( Figure S2A , IL3 ARE sequence depicted in Figure S3 ) . Expression of the FL reporter gene was found to be strongly suppressed by the IL3 ARE in SL2 cells , both at the protein ( luciferase activity ) and mRNA level , and suppression was due to accelerated degradation of the reporter mRNA ( Figure S2B–S2D ) . We then tested several factors for their involvement in Drosophila AMD by knocking down their expression using dsRNAs . Whereas knock down ( kd ) of Tis11 and Not1 , a core protein of the cytoplasmic Ccr4-Caf1-Not deadenylation complex , caused elevated expression of the ARE-reporter , other proteins such as Rox8 , AGO1 , AGO2 , LSm1 and pcm did not affect reporter gene expression ( Figure 4A ) . Since Drosphila Not1 is important for mRNA deadenylation in general [34] , [35] , we focused on Tis11 as an ARE-specific RNA binding protein . Our goal was to examine the AREScore of mRNAs regulated by Tis11 . We first confirmed that the dsRNA against Tis11 potently suppressed the expression of Tis11 mRNA ( Figure 4B ) , and that Tis11 kd stabilizes the FL-mIL3-ARE reporter mRNA ( Figure 4C ) . To identify Drosophila mRNAs regulated by Tis11 , we determined the mRNA expression profile in SL2 cells after knocking down Tis11 or , as a control , GFP . Since direct targets of Tis11 are expected to show higher expression levels after Tis11 kd , we concentrated on the 53 mRNAs that we found to be upregulated by a factor of at least 1 . 41 ( 0 . 5 log2-transformed ) after Tis11 kd in the microarray analysis ( Figure 5A ) . 20 out of these mRNAs were chosen for confirmation by qPCR , and for 18 of them we could verify that Tis11 kd causes a an increase in expression of minimally 1 . 41-fold ( Figure S4 ) , indicating that our microarray dataset was reliable . The Vir1 mRNA , which was strongly upregulated by Tis11 kd ( Figure S4 ) , has an AREScore of 5 . 6 and a readily detectable ARE ( Figure S3 ) . Indeed , an FL reporter mRNA containing the ARE of Drosophila Vir1 was stabilized by kd of Tis11 ( Figure 4C ) . Out of the 53 mRNAs sensitive to Tis11 kd , we then determined the AREScore for those 49 transcripts whose annotated 3′UTR length is ≥10 nt . In comparison to the AREScore distribution of the entire D . melanogaster transcriptome , the Tis11-sensitive mRNAs showed an increased abundance of AREScores ≥4 ( Figure 5B ) . By χ2-test , this increase was statistically significant with a p-value of 0 . 0011 ( Table S3 ) , suggesting that target mRNAs of Drosophila Tis11 share characteristics with mammalian AREs . After applying the AREScore to the subgroup of Tis11-sensitive mRNAs , we wanted to assess the importance of AREs in regulating Drosophila gene expression more generally . If AREs are wide-spread elements that promote mRNA degradation but do not affect transcription , a first prediction is that , on average , mRNAs with a high AREScore should be expressed at lower levels . A second prediction is that these mRNAs should have shorter half-lives . We tested the first prediction by comparing the expression levels of 6657 mRNAs , derived from our microarray analysis in SL2 cells , with their AREScores ( Figure 6A ) . Indeed , we observed a tendency for mRNAs with high AREScores to be expressed at lower levels . We further grouped the mRNAs into 9 categories according to their AREScore , and compared the average expression levels of each group to the overall average ( Figure 6B ) . The two groups with very high AREScores ≥12 showed average expression levels that were more than 3-fold ( 1 . 6 log2-transformed ) below the overall average , and the reduction in expression was already significant above an AREScore of 8 . We then compared the 3′UTR lengths with the AREScores of all 6657 mRNAs ( Figure 6C ) . As expected , we found a very strong correlation between these two parameters ( RS = 0 . 57 , p<0 . 0001 ) . Thus , it was important to assess whether the 3′UTR length on its own had an influence on mRNA expression levels ( Figure 6D and 6E ) . Two opposing correlations were apparent: mRNAs with very short 3′UTRs <100 nt , and mRNA with long 3′UTRs ≥1000 nt were expressed at significantly reduced levels , whereas mRNAs with 3′UTRs of intermediate length ( 100–999 nt ) showed the highest expression levels . In fact , the 3′UTR length appeared to have a stronger influence on the expression level than the AREScore , as mRNAs with 3′UTRs ≥2000 nt were expressed more than 5-fold ( 2 . 4 log2-transformed ) below the overall average . To examine whether the predictive power of the AREScore is independent of 3′UTR length , we chose to analyze a subgroup of 1781 mRNAs with 3′UTRs between 200 and 499 nt ( pink bars in Figure 6E ) . In this group , the length of the 3′UTR per se does not negatively correlate with mRNA levels ( Figure 6F ) , whereas mRNAs with higher AREScores do show a trend towards reduced expression levels ( Figure 6G ) . Indeed , the 35 mRNAs that have an AREScore ≥8 within this group had a more than 3-fold ( 1 . 6 log2-transformed ) reduced average expression level compared to the 1746 mRNAs that have an AREScore between 0 and 7 . 99 ( Figure 6H ) , and this difference was highly significant ( p<0 . 0005 ) . In contrast , the average expression level of the 35 mRNAs with the longest 3′UTRs was only 1 . 4-fold ( 0 . 5 log2-transformed ) below the expression level of the remaining 1749 mRNAs with the shorter 3′UTRs . From this comparison we concluded that a high AREScore correlates with lower mRNA expression levels independently of the 3′UTR length . Finally , we tested the second prediction that mRNAs with higher AREScores should undergo more rapid decay . To this end , we measured the half-lives of 26 mRNAs with high accuracy by qPCR ( Table 1 , Figure S5 ) . 12 mRNAs were chosen from the group of Tis-11 sensitive mRNAs , and 14 from the large pool of mRNAs that are not affected by Tis11 kd . To cover the entire range , 5 mRNAs had a high AREScore ≥8 , 8 mRNAs had a medium AREScore between 4 and 7 . 99 , and 13 had a low AREScore <4 . In Figure 7A , we plotted the half-lives of these mRNAs against the AREScore . The most striking observation was that 9 out of 10 mRNAs with an AREScore of 0 degraded very slowly with half-lives >240 minutes . On the other side , the two mRNAs with the highest AREScore ( CG115435 from the group of Tis11-sensitive mRNAs and Reck and from the control group ) also had the shortest half-lives . In our analysis of 26 mRNAs , the Spearman's rank correlation coefficient RS between the two parameters equals −0 . 73 , and this correlation was highly significant ( p<0 . 001 ) . We also compared the half-lives of these 26 mRNAs with their 3′UTR length ( Figure 7B ) , and found a weaker correlation ( RS = −0 . 61 , p<0 . 001 ) . ROC analysis was then applied to test the ability of both AREScore and 3′UTR length to discriminate labile mRNAs with half-lives <140 minutes from stable mRNAs with half-lives >240 minutes ( Figure 7C ) . The AREScore performed extremely well in this test with an AUC of 0 . 95 , better than 3′UTR length with an AUC of 0 . 87 . Clearly , the AREScore identifies short-lived mRNAs in D . melanogaster , showing that AREs are general regulatory elements in this organism .
In this report , we developed AREScore as an algorithm to identify AREs and provide a measure for their potential strength ( Figure 1 ) . The AREScore was validated using genome-wide mRNA half-life measurements in human DG75 B-cells [32] and mouse NIH3T3 fibroblast [33] . Although the correlation between AREScore and mRNA half-life was weak ( RS = −0 . 155 and −0 . 147 in the two data sets , respectively ) , it was statistically highly significant . To our knowledge , this is the best correlation observed so far between any parameter and mRNA half-lives on a genome-wide scale . The potential of the AREScore could be further demonstrated with a set of TTP-associated mRNAs that we had previously identified by RNA-IP in mouse macrophages [13] . AREScores were much higher in this set of mRNAs than in the two control groups ( Figure 1H ) . Among the Tis11-sensitive mRNAs that we identified in Drosophila SL2 cells , we also observed an increased frequency of mRNAs with higher AREScores ( Figure 5 ) , suggesting that Drosophila Tis11 recognizes AREs with sequence features similar to mammalian AREs . Khabar et al . used bioinformatic tools to generate the ARE-database ( ARED ) , a comprehensive list of potential AREs in the human , mouse and Drosophila genome [7] , [31] , [36] . The principle behind ARED is that it classifies AREs according to the number and density of AUUUA pentamers and surrounding AU-rich sequences , which correlates , to some degree , with the potential strength of the ARE . In contrast to ARED , the purpose of AREScore is not to make categories , but rather generate a single score that provides a measure for the likelihood and potential strength of an ARE . It is important to emphasize that in the absence of experimental validation , neither ARED nor AREScore is able to predict with absolute certainty whether a given mRNA contains a functional ARE . For the AREScore , the false positive rate was visualized by ROC analysis , whereby the AREScore is tested for its ability to discriminate between the 10% most short-lived and the 10% most long-lived mRNAs ( Figure 1D and 1G ) . For ARED , the false positive rate is not known . An advantage of AREScore is that it can be applied easily to any genome or set of sequences . Thus , we were able to compute the AREScore distribution for the transcriptomes of 14 species representing all but two of the major branches of metazoan evolution ( Figure 2 and Figure S1 ) . The analysis showed that mRNAs with high AREScores are most abundant in man and mouse , the two mammalian species analyzed . Comparison to randomized control sequences revealed that mRNAs with high AREScores ( ≥10 ) are overrepesented in man , mouse , chicken , zebrafish and the fruit fly ( Figure 3 ) . This suggests that AREs were under positive selection pressure during the evolution of these organisms . On the other hand , high AREScore mRNAs are underepresented in the sponge A . queenslandica , the freshwater cnidarian H . magnipapillata , the mollusc A . californica and the nematode C . elegans , suggesting that AREs did not expand in the genomes of metazoans with simpler body plans . Alternatively , the element corresponding to the ARE might have different sequence features in these organisms . Given that very little is known about AREs in D . melanogaster , we then made use of the AREScore to address the role of AMD in Drosophila SL2 cells . Using an FL-based reporter assay , we first tested several factors and found that knocking down Tis11 or Not1 caused inhibition of AMD , whereas the kd of Rox8 , AGO1 , AGO2 , LSm1 or pcm had no effect ( Figure 4 ) . The requirement of Tis11 for AMD is in good agreement with the well documented role of TTP in mammalian AMD [37] as well as previous reports demonstrating that Tis11 participates in AMD in Drosophila cells [29] , [30] , [31] , [38] . The requirement for Not1 may be linked to our recent finding that mammalian TTP recruits the Caf1 deadenylase through its association with Not1 [23] . Not1 is the scaffold protein of the Ccr4-Caf1-Not deadenylase complex that plays a key role in cytoplasmic mRNA turnover . In Drosophila , Not1 was shown to be important for bulk mRNA deadenylation and , more specifically , for the rapid deadenylation of Hsp70 mRNA [34] , [35] . A previous report had suggested that AGO1 and AGO2 are required for the rapid degradation of a reporter mRNA containing the ARE of mammalian TNFα in Drosophila S2 cells [38] . In our assay , kd of the argonaute proteins AGO1 and AGO2 did not affect expression of the reporter gene containing the ARE of mouse IL-3 ( Figure 4A ) , indicating that AGO proteins are not generally required for AMD . As potential substrates of AMD , we then identified 53 mRNAs whose expression was elevated after kd of Tis11 ( Figure 5A ) . The AREScore of these Tis11-sensitive mRNAs was found to be higher in comparison to the distribution in the entire D . melanogaster transcriptome ( Figure 5B ) , and the difference was statistically significant for mRNAs with AREScores ≥4 ( Table S3 ) . CecA1 mRNA , previously identified as a target of Tis11 [29] , [30] , [31] , did not come up as Tis11-sensitive simply because this mRNA is not represented on the Affymetrix Drosophila Genome 2 . 0 array that we used for our study . We then compared the expression levels of 6657 mRNAs in SL2 cells with their AREScore ( Figure 6A–6E ) , and observed that mRNAs with high AREScores have reduced expression levels . However , this effect may be indirect because the AREScore strongly correlates with 3′UTR length . Indeed , when grouping mRNAs according to their 3′UTR length , we again observed that mRNAs with long 3′UTRs have lower expression levels . The impact of 3′UTR length was in fact stronger than the impact of the AREScore . Long 3′UTRs are likely to correlate with low expression levels through the presence of different suppressive elements including AREs and miRNA-binding sites . Moreover , the distance between the stop codon and the poly ( A ) tail is a determinant of nonsense-mediated mRNA decay [39] and may thereby as well contribute to mRNA suppression . We also noted that mRNAs with very short 3′UTRs <100 nt are expressed below the overall average . A possible explanation is that very short 3′UTRs might lack stabilizing elements , although there is little experimental evidence that such elements are abundant . To examine the impact of the AREScore independently of its correlation with 3′UTR length , we chose a group of mRNAs with intermediate 3′UTRs ( Figure 6F–6H ) . Within this group we could observe that mRNAs with an AREScore ≥8 had a more than 3-fold reduced average expression level compared to the mRNAs with AREScores <8 . Since 3′UTR length had a much smaller effect on mRNA levels in this group , we concluded that the AREScore is an independent parameter that correlates with suppressed mRNA levels . Given the multitude of factors that affect mRNA stability and transcription rates , it is remarkable that the AREScore alone has a detectable influence on mRNA expression levels . Finally , we measured the decay rates of 26 mRNAs in Drosophila SL2 cells ( Table 1 ) . Indeed , we observed a very strong , negative correlation between mRNA half-life and the AREScore ( RS = −0 . 73 , Figure 7 ) , which was higher than the correlation with 3′UTR length ( RS = −0 . 61 , ) . Since we measured mRNA half-lives both in control GFP and Tis11 kd cells , we could also identify three mRNAs that are significantly stabilized by the absence of Tis11 . These mRNAs encode for peroxidasin ( Pxn ) , CG15435 , a C2H2 zinc finger protein of unknown function , and CG7115 , a protein phosphatase of the PP2C family . Taken together , our analysis provides compelling evidence that AREs are functional regulatory elements in D . melanogaster cells whose suppressive effect can be detected on a transcriptome-wide level . Interestingly , we found two short-lived mRNAs with a high AREScore ( Reck and CG32512 ) in our control group of mRNAs that are not sensitive to Tis11 kd . This indicates that in addition to Tis11 , other proteins also participate in regulating AMD . It is clear that we have only begun to understand the posttranscriptional regulatory network that controls gene expression through mRNA turnover in D . melanogaster .
Plasmid pRp128-RL ( p2933 ) [40] contains the Drosophila RNA polymerase III 128 kDa subunit promoter to drive RL expression , and was kindly provided by Michael Boutros ( German Cancer Research Center , Heidelberg ) . For pRp128-FL ( p2934 ) , pRp128-RL was digested with SpeI/NheI and religated to remove part of the polylinker . In the resulting construct , the RL-containing HindIII/XbaI fragment was replaced with the FL-containing HindIII/XbaI insert from pGL3-Basic ( Promega ) . For pRp128-FL-mIL3-ARE ( p2935 ) , the mouse IL-3 ARE sequence ( NM_010556 . 4 , nt 680–744 ) was amplified by PCR using primers G1090/G1091 ( Table S4 ) and inserted into the XbaI site of pRp128-FL . The control vector pRp128-FL-mIL3-INV ( p2936 ) was constructed in the same way with the IL-3 ARE inserted in the opposite orientation . To generate pAc5-FL-mIL3-ARE ( p2937 ) , a 3 . 8 kb FL-containing HindIII ( blunt ) –BglII fragment was excised from plasmid pRp128-mIL3-ARE and ligated to KpnI ( blunt ) - BglII fragment ( 2 . 4 kb ) with Ac5 promoter obtained by digestion of pAc5 . 1b-EGFP-dmDcp1 ( p2450 ) ( kindly provided by Elisa Izaurralde , Max Planck Institute for Developmental Biology , Tübingen , Germany ) . For pAc5-FL-Vir1-ARE ( p2938 ) , the D . melanogaster Vir1 3′UTR ( NM_165011 . 2 , nt 1521–1830 ) was first amplified by RT-PCR using primers G1673/G1674 . An ARE-containing 191 nt long fragment ( NM_165011 . 2 , nt 1640–1830 ) was re-amplified by PCR using XbaI site-containing primers G1681/G1679 and inserted into the XbaI site of pRp128-FL to generate pRp128-FL-Vir1-ARE . Finally , the Ac5 promoter was excised as a SapI–BglI fragment from pAc5-FL-mIL3-ARE and cloned into the SapI/BglI sites of pRp128-FL-Vir1-ARE , thereby replacing the pRp128 promoter . pAc5-FL ( p2939 ) was generated in a similar manner by cloning the SapI–BglI fragment from pAc5-FL-mIL3-ARE into the SapI/BglI sites of pRp128-FL . Drosophila SL2 cells were cultivated at 26°C under atmospheric CO2 in Schneider's Drosophila Medium ( Invitrogen-Gibco , Cat . No . 11720-034 ) supplemented with 10% foetal bovine serum ( Biochrome Superior FBS , Cat . No . S0615 ) , 50 U/ml penicillin and 0 . 05 mg/ml streptomycin ( both Pan Biotech ) . All DNA transfections were performed using Effectene reagent ( Qiagen , Cat . No . 301425 ) according to the manufacturer's instructions . When combined with RNAi , cells were first treated with dsRNA for two days , followed by DNA transfection for two additional days . For luciferase assays , 10 . 000 cells were seeded per well of a 384-well plate ( Greiner ) , treated with 250 ng of dsRNA and transfected with 7 ng of FL-encoding and 3 ng of RL-encoding plasmids . Where indicated , ActinomycinD ( Applichem , Cat . No . A1489 ) was used at a concentration of 5 µg/ml . Unless noted otherwise , cell lysis and RNA extraction were performed with Genematrix universal DNA/RNA/Protein purification kit ( Eurx ) , according to manufacturer's instructions . DNA templates for in vitro transcription were amplified by PCR using primers containing Sp6 promoter sequences , as specified in Table S5 . DNA templates were gel-purified using a gel extraction kit ( QIAGEN , Cat . No . 28706 ) . In vitro transcription reactions were assembled in a total volume of 50 µl containing 50–75 ng/µl DNA template , 3 mM NTPs each ( Promega ) , 40 U RNasin ( Promega , N2111 ) , 0 . 5 U yeast pyrophospatase ( Sigma , Cat . No . I1891 ) , 200 U Sp6 RNA polymerase ( Fermentas , EP0133 ) , 80 mM HEPES-KOH pH 7 . 5 , 32 mM MgCl2 , 2 mM spermidine and 40 mM DTT . Reactions were incubated for 4 hours at 37°C . DNA was then digested by the addition of 1 U/µl DNase RQ1 ( Promega ) for 15 minutes at 37°C . The synthesized RNA was purified by gel filtration using pre-packed Sephadex G-50 columns ( Roche , Cat . No . 11274015001 ) . Strands were annealed by heating the purified RNA to 65°C and allowing it to slowly cool to room temperature . For RNAi , Drosophila cells were grown in 6 cm-dishes and incubated with 50 µg of dsRNA per 4 ml of medium for a minimum of 4 days . Total RNA was extracted using the Genematrix universal RNA purification kit ( Eurx ) . 5–12 µg of RNA was resolved by 1 . 1% agarose/2% formaldehyde/MOPS ( morpho-linepropanesulfonic acid ) gel electrophoresis and blotted over night with 8× saline-sodium citrate ( SSC , 1× contains 0 . 15 M NaCl and 0 . 015 M sodium citrate ) buffer onto Hybond-N+ Nylon membranes ( Amersham , GE Healthcare ) . Membranes were hybridized overnight at 55°C with digoxigenin-labelled RNA probes synthesized in vitro using Sp6 polymerase ( Fermentas ) and DIG RNA labelling mix ( Roche ) . 500 ng RNA probe was diluted in 10 ml hybridization buffer containing 50% formamide , 5× SSC , 5× Denhard's solution , 5 mM EDTA , 10 mM PIPES pH 7 . 0 , 4 mg torula yeast RNA ( US Biological ) and 1% SDS . Membranes were washed twice with 2× SSC/0 . 1% SDS for 5 minutes , and twice with 0 . 5× SSC/0 . 1% SDS for 20 minutes at 65°C . Alkaline phosphatase-coupled anti-digoxigenin Fab fragments and CDP-Star substrate ( both Roche ) were used for detection according to the manufacturer's instructions . Sequences of primers that were used to generate templates for digoxigenin-labelled RNA probes are provided in Table S6 . For qPCR , total RNA was extracted by Genematrix RNA purification kit ( Eurx ) and subjected to DNase treatment using RQI DNase ( Promega , 1 . 5 U/column ) . cDNA was synthesized from 5 µg of total RNA using oligo-dT18 ( Invitrogen ) and M-MLV H ( - ) reverse transcriptase ( Promega ) . 1∶40 volume of a cDNA reaction was used for PCR . PCR reactions were assembled in 384-well plates , 15 µl/well final volume . DNA SYBR Green I Master kit ( Roche Cat . No . 04707516001 ) was used according to manufacturer's instructions . Quantitative PCR was performed with the Lightcycler 480 system ( Roche ) . Gene-specific primers sequences are given in Table S7 . FL and RL activities were measured using the Dual Luciferase Reporter Assay system ( Promega ) or reagents developed in the lab of Michael Boutros ( German Cancer Research Centre , DKFZ , Heidelberg ) . Chemiluminescence was measured using a Mithras LB940 plate reader . Drosophila SL2 were treated with either dsRNA-GFP or dsRNA-Tis11 for 4 days . Total RNA was prepared using RNEasy kit from Qiagen ( Cat . No . 74106 ) . Efficiency of Tis11 knockdown was confirmed separately by Northern blot and qPCR analyses . Expression profiling was carried out on GeneChip Drosophila Genome 2 . 0 Arrays ( Affymetrix , Cat . No . 900531 ) at European Molecular Biology Lab Genecore facility ( Heidelberg , Germany ) . Microarray data were deposited at NCBI GEO , accession GSE28147 . The RMA algorithm was used for normalization of raw data ( RMAExpress software , http://rmaexpress . bmbolstad . com/ ) . Further statistical analyses were performed in the multi-experiment viewer of the TM4 microarray software suite [41] or with R software ( http://www . r-project . org ) . Beside standard Student t-test , we also used microarray-oriented Rank products test [42] to identify significant changes in expression . The AREScore algorithm was written in Perl and is accessible online at http://arescore . dkfz . de/arescore . pl . AREScore uses either a set of sequences provided in FASTA format or retrieves the 3′UTR sequences of properly annotated transcripts if Refseq IDs are entered . The algorithm first generates a basal score by adding a fixed value of 1 for each AUUUA pentamer identified . It then calculates the distance between neighboring pentamers , and adds a value to the basal score if pentamers are in close proximity . A value is also added when pentamers are located within a region of high AU content , termed an AU-block in AREScore . In its standard setting , AREScore adds a value of 1 . 5 for overlapping pentamers , 0 . 75 if pentamers are 0–3 nt apart , 0 . 4 if pentamers are 4–6 nt apart , 0 . 2 if pentamers are 7–9 nt apart , and 0 . 3 if pentamers are within an AU-block . By default , an AU-block starts when a sequence of 20 nt ( word size ) has an AU content of ≥80% . The block ends when the AU content drops below 55% within the chosen word size . To increase the flexibility of AREScore , users can change the values that are added to the basal score , and alter the settings that define an AU-block . For the transcriptome-wide AREScore analysis , transcripts with properly formated feature fields ( Genbank GeneID and Accession identifiers ) were downloaded from Refseq , and the AREScore was determined for every 3′UTR ≥10 nt in length . If several transcripts map to the same gene locus ( identical GeneIDs ) , only the mRNA with the highest AREScore was taken into the analysis . To generate a fully matched set of randomized control sequences , nucleotides were randomly chosen from the pool of all 3′UTRs analyzed , and assembled into sequences identical in length to the original 3′UTRs . The AREScore was then determined for the randomized control sequences as well .
|
Many genes are regulated at the posttranscriptional level by factors that influence the stability of the messenger RNA . In mammals , AU-rich elements are known to cause rapid degradation of messenger RNAs and thereby suppress gene expression . In order to identify such elements on a genome-wide scale , we developed a bioinformatic tool with which we can score messenger RNAs for the presence of AU-rich elements . Using the AREScore algorithm , we observe that AU-rich elements correlate with reduced messenger RNA stability and expression levels . We then used the AREScore to compare the transcriptomes of 14 metazoan species and found that messenger RNAs with high AREScores are enriched in several vertebrates and the fruit fly Drosophila melanogaster . We identified messenger RNAs whose levels are regulated by the Drosophila Tis11 protein , which binds to AU-rich elements . Our study introduces the AREScore as a means to globally assess AU-rich elements and predict short-lived messenger RNAs . Furthermore , it demonstrates the regulatory role of AU-rich elements in suppressing gene expression by accelerating messenger RNA degradation in D . melanogaster cells .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"gene",
"regulation",
"rna",
"stability",
"eukaryotic",
"cells",
"genome",
"analysis",
"tools",
"molecular",
"genetics",
"gene",
"expression",
"biology",
"molecular",
"biology",
"transcriptomes",
"biochemistry",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"genetic",
"screens",
"genetics",
"cellular",
"types",
"genomics",
"molecular",
"cell",
"biology",
"computational",
"biology",
"genetics",
"and",
"genomics"
] |
2012
|
Genome-Wide Assessment of AU-Rich Elements by the AREScore Algorithm
|
Francisella tularensis is a potent mammalian pathogen well adapted to intracellular habitats , whereas F . novicida and F . philomiragia are less virulent in mammals and appear to have less specialized lifecycles . We explored adaptations within the genus that may be linked to increased host association , as follows . First , we determined the genome sequence of F . tularensis subsp . mediasiatica , the only subspecies that had not been previously sequenced . This genome , and those of 12 other F . tularensis isolates , were then compared to the genomes of F . novicida ( three isolates ) and F . philomiragia ( one isolate ) . Signs of homologous recombination were found in ∼19 . 2% of F . novicida and F . philomiragia genes , but none among F . tularensis genomes . In addition , random insertions of insertion sequence elements appear to have provided raw materials for secondary adaptive mutations in F . tularensis , e . g . for duplication of the Francisella Pathogenicity Island and multiplication of a putative glycosyl transferase gene . Further , the five major genetic branches of F . tularensis seem to have converged along independent routes towards a common gene set via independent losses of gene functions . Our observations suggest that despite an average nucleotide identity of >97% , F . tularensis and F . novicida have evolved as two distinct population lineages , the former characterized by clonal structure with weak purifying selection , the latter by more frequent recombination and strong purifying selection . F . tularensis and F . novicida could be considered the same bacterial species , given their high similarity , but based on the evolutionary analyses described in this work we propose retaining separate species names .
Francisella tularensis is probably best known , and most feared , for its potential as a bacterial biological weapon [1] . As such this pathogen was grown and stockpiled in large quantities during the Cold War by both the U . S . and the former Soviet Union . Today , the most virulent Francisella strains are among the six biological agents considered to pose the greatest potential public health threats if used by terrorists [2] . Strains of F . tularensis subsp . tularensis can be lethal to humans , doses as low as 10–25 bacteria can be infective , and transmission can occur via skin inoculation or aerosols [1] . However , in addition to its potentially destructive applications , the genus Francisella provides interesting models for studying processes whereby quite harmless environmental bacteria may become transformed into host-restricted and highly virulent human pathogens . In this respect Francisella bacteria appear to be in a state previously attributed to several other human pathogens ( e . g . Shigella flexneri , Salmonella enterica serovar Typhi and Yersinia pestis [3] ) that are in intermediate stages of a genome-erosion process associated with early stages of host-restriction . Francisella strains are attractive ( as model organisms ) since they span a broad range of functional diversity , from strains with high metabolic capacity that are easily grown on artificial media and exhibit low disease potential in humans , to specialized , highly pathogenic bacteria with reduced metabolic capacities that ( hence ) require very rich culture media [4] . Besides F . tularensis , the genus Francisella includes two accepted species , F . philomiragia and F . novicida , both of which are isolated from environmental samples [5] , [6] . In contrast to F . tularensis , F . philomiragia and F . novicida are metabolically competent and thus much less fastidious in their growth requirements . They are only rarely human pathogens , the only diseased individuals they have been isolated from were nearly drowned or suffered from a weakened immune system [4] . F . tularensis is the cause of tularemia and is characterized as a facultative intracellular pathogen . Tularemia is a typical zoonosis; it is frequently arthropod vector-borne , transmissible to humans , and its usual host is a non-human animal [7] . However , the term “usual host” is somewhat arbitrary for F . tularensis; according to a recent review known susceptible species include 190 mammals , 88 invertebrates , 23 birds , and three amphibians [8] . At present , three F . tularensis subspecies are accepted and known to cause infections in humans [9] . Two of these are clinically important: subsp . tularensis ( type A ) is found exclusively in North America and may cause severe and life-threatening infections , while subsp . holarctica ( type B ) occurs throughout the Northern Hemisphere and is associated with milder clinical symptoms [10] . Both subspp . have been implicated in the production of biological weapons , although the use of subsp . tularensis as a weapon would be more serious . Two major subpopulations among type A strains are designated A1 and A2 , respectively [11] . The third F . tularensis subspecies , subsp . mediasiatica , has only been isolated in areas of central Asia and is reported to exhibit comparable virulence to that of the holarctica subspecies [12] , [13] . In a rabbit model , strains of F . tularensis subspp . holarctica and mediasiatica kill at a dose of >106 microbial cells , while a lethal dose of subsp . tularensis , is 1–10 cells [14] . However , previous studies indicate that subsp . mediasiatica is , in evolutionary terms , the closest neighbor to subsp . tularensis [15]–[17] . We hypothesized , therefore , that comparing the genome of subsp . tularensis to that of subsp . mediasiatica , which exhibits lower disease potential for humans , would be valuable for determining factors that cause high pathogenicity in subsp . tularensis . Sequencing a genome of a subsp . mediasiatica strain would additionally allow for multiple genome comparisons to understand the evolution of human pathogenic strains of genus Francisella .
Previously , genome sequences have been reported for two of three recognized F . tularensis subspecies . The third subspecies , F . tularensis subsp . mediasiatica , has been isolated only in dry areas of Central Asia . We describe here the genome of strain FSC147 , isolated in the Alma-Ata region of Kazakhstan in 1965 , from the rodent species Meriones meridianus ( Midday gerbil ) . The genome is composed of a single circular 1 , 893 , 886 bp chromosome with an average G+C content of 32 . 25% ( Table 1 ) . It contains 1 , 470 predicted protein-coding genes and 263 pseudogenes . As in previously characterized representatives of subspp . tularensis and holarctica [18]–[21] , we found three rRNA operons , 38 tRNA genes with 30 anticodons for 20 amino acids , and seven types of insertion sequence ( IS ) elements . ISFtu1 and ISFtu2 were the most abundant elements , with 59 and 17 copies , respectively ( Table 1 ) . All predicted genes in F . tularensis subsp . mediasiatica strain FSC147 were found either in the tularensis subspecies ( SCHU S4 ) or F . novicida ( strain U112 ) . Rhomer et al . suggested that six genes predicted to be functional in an 11 . 1 kb region ( loci FTT1066-FTT1073 ) , together with three other genes ( loci FTT1308c , FTT1580c , FTT1791 ) , may promote the high virulence of subsp . tularensis since these genes appeared to be specific to strain SCHU S4 when compared with F . novicida strain U112 and two strains of F . tularensis subsp . holarctica [22] . However , all open reading frames in the 11 . 1 kb region appear to be present and intact in the mediasiatica genome of strain FSC147 , indicating that these genes by themselves cannot explain the different degrees of virulence . In addition , FTT1308c is inactivated by a frameshift mutation in mediasiatica and is not present in F . tularensis subsp . holarctica , but the intact gene is present in the F . novicida isolate 3548 , suggesting that this gene is probably not responsible for the high virulence of subsp . tularensis SCHU S4 either . Furthermore , FTT1580c encodes a protein of unknown function and contains a frameshift mutation close to the N-terminal part in F . tularensis subsp . mediasiatica FSC147 that could disrupt the function of the encoded protein . A role for FTT1580c in enhancing virulence is also disputed since the ortholog is intact in F . tularensis subsp . holarctica FSC022 , which exhibits low virulence . Finally , the hypothetical protein referred to as FTT1791 has a nonsense mutation in mediasiatica . This gene is also missing in the isolate WY-96 of the type A2 clade of subspecies tularensis . Thus , if the isolate WY-96 possesses the high level of virulence ascribed to other strains of subsp . tularensis this would exclude FTT1791 as a likely cause . Using 1 , 104 , 129 aligned genomic nucleotide sites , we estimated a whole-genome phylogeny ( Figure 1 ) for one F . philomiragia , three F . novicida or “novicida-like” , and 13 F . tularensis strains ( Table S1 ) . Both the neighbor-joining and maximum likelihood methods identified an identical topology , with maximal support for all nodes in bootstrap analyses . The tree was consistent with the previous assumption that F . tularensis and F . novicida constitute sister groups , and confirmed that F . tularensis subspp . mediasiatica , holarctica , and tularensis form a monophyletic group [7] , [16] . The data extend previous findings by demonstrating the monophyly of F . novicida and the novicida-like isolates that were included . Substantial differences in branch lengths were found in the phylogenetic reconstructions ( Figure 1 ) , indicating that historic mutation rates have differed among F . tularensis genetic lineages . Exemplifying the extremes , mutations in the F . tularensis subsp . tularensis A2 lineage have occurred much less frequently than in the subsp . holarctica lineage ( Figure 1 ) . Increased rates of mutation may occur because some DNA repair functions are lost; therefore we scrutinized corresponding genes in the F . philomiragia , F . novicida and F . tularensis genomes . Among 37 potential DNA repair genes we found only one possible candidate , a deoxyribodipyrimidine photolyase gene , phrB , that appears to be functional in WY-96 , but disrupted in other strains ( Table S2 ) . The Phr protein enhances repair of UV-light induced DNA damage and is lacking in many bacterial species because they live in environments where they are not exposed to UV light [23] . The F . novicida , novicida-like isolates , and F . tularensis isolates were found to display considerable overall genetic relatedness . Pairwise analyses of average nucleotide identities ( ANI ) [24] demonstrated that all combinations had ANI values ≥97 . 7% . If only isolates of F . tularensis were considered , ANI values ≥99 . 2% ( Table S3 ) were obtained , thus demonstrating a striking level of genetic monomorphism within this monophyletic group of strains , which includes three separate subspecies: tularensis , mediasiatica , and holarctica . In contrast , comparisons that included the F . philomiragia isolate ATCC 25017 provided significantly lower ANI estimates , ranging between 80 . 6% and 81 . 2% . The presence and extent of recombination in Francisella was rigorously investigated using several strategies: visual exploration of genomic data , estimation of recombination and mutation parameters using the ClonalFrame [25] and Rm [26] methods , and estimation of the proportions of genes potentially affected by recombination using a combination of the MaxChi2 [27] and Phi [28] tests . To assess the possibility that recombination has occurred among the highly similar F . tularensis lineages , we also used the Maynard Smith and Smith [29] homoplasy test . Visual exploration of colour-coded nucleotide plots revealed indications of numerous past recombination events among metabolically independent Francisella lineages ( the F . philomiragia , F . novicida , and ancestral F . tularensis lineages; Figure 2 ) . Abundant tracts containing incongruent sites were found , and loci with increased numbers of informative sites , suggestive of recombination between sister lineages . However , we found no evidence to support the occurrence of past recombination between the metabolically independent Francisella and any modern lineage of F . tularensis . The few genomic regions potentially indicative of such events could be dismissed after close examination as being due to other evolutionary events , such as gene conversion ( recombination within the genome ) or incomplete lineage sorting ( differential loss of previously duplicated genes ) . The impact of recombination on F . philomiragia , F . novicida , and the ancestral branch of F . tularensis was further quantified by applying the Clonalframe algorithm to whole genomes and , for comparison , by analyzing five collinear genomic regions to estimate minimum numbers of recombination events and segregating sites . Clonalframe analysis of 1 , 527 , 362 sites estimated the 95% credibility region of rho/theta , the ratio of absolute numbers of recombination and mutation events , to be 0 . 079–0 . 089 . This indicates that mutation , despite an abundance of recombination footprints , has clearly been the predominant evolutionary process . The 95% credibility region of r/m , indicating the probability of recombination versus mutation per individual nucleotide site , was 0 . 78–0 . 89 , illustrating that the impact of recombination on genetic diversity has been significant . Furthermore , minimum numbers of recombination events ( Rm ) were estimated , as described by Hudson and Kaplan [26] , for segregating sites in five 75-kb locally collinear sequence blocks . This analysis suggested a rate of recombination an order of magnitude lower than that of mutation ( Table 2 ) . However , since Rm represents a lower bound and ClonalFrame models only recombination “imports” , both methods likely underestimate the true number of recombination events We also assessed the proportions of genes affected by recombination in the environmental lineages by a combined analysis in which individual gene alignments were tested by the maximum chi-squared method [27] and the Phi method [28] . This combination of methods was used to increase sensitivity , since they detect different , complementary recombination signals . The maximum chi-squared method and the phi method indicated 223/1251 and 101/1251 genes to have been affected by recombination , respectively ( **p<0 . 01 ) . Using either method , significant indications of recombination were obtained for19 . 2% of the genes ( 240/1251 ) tested . Because of the limited diversity of F . tularensis , few tests can be used to assess recombination within this species . However , in nucleotide alignments representing 13 F . tularensis genomes 21 apparent homoplasies were recorded . Therefore we applied Maynard Smith and Smith's homoplasy test [29] , which assesses whether there is an excess of homoplasies , compared to expected numbers derived by mutation in the absence of recombination . The null hypothesis of clonality was not rejected using any reasonable estimate of Se . Thus , we found no evidence to support the hypothesis that recombination has occurred among F . tularensis lineages . Measuring dN/dS might provide some information on how ecology constrained the evolution of Francisella at population levels . We calculated by maximum likelihood methodology [30] lineage-specific dN/dS estimates of selection pressures across the Francisella phylogeny ( Figure 3 ) , taking care to avoid overfitting by using information-rich genome-wide datasets of aligned codons , and employing a genetic algorithm that optimizes model complexity [30] . To increase the accuracy of our calculations , we focused on F . tularensis representatives in one analysis , and on F . novicida and F . philomiragia in a separate analysis . Overall , our results revealed distinct differences between the environmental lineages ( F . novicida , F . philomiragia ) and F . tularensis , since we found high dN/dS ratios for all F . tularensis branches , and considerably lower ratios for the environmental lineages ( Figure 3 ) . Thus , the data indicates that slightly deleterious mutations have been inefficiently removed following the formation of the F . tularensis species . This lends support to the hypotheses that effective population sizes of these strains have been low , and successful recombination events among them have been rare or non-existent , consistent with evolution in a favorable intracellular environment . However , there were statistically significant dN/dS differences among F . tularensis lineages , which might reflect ecological differences within the species , notably estimates of dN/dS ratios in the F . tularensis subsp . holarctica clade were consistently higher than in the clade consisting of subspp . tularensis and mediasiatica . It has been found that among closely related strains or species dN/dS ratios can be elevated in a time-dependent fashion [31] , [32] , and thus not reflect longer term selection pressures . This is because such comparisons are akin to the study of de novo mutations that have not yet been eliminated at the population level . Since all F . tularensis isolates are highly similar , we anticipated that there was a high likelihood that such effects would be observed . By plotting intergenic distances against dN/dS ratios determined in pairwise comparisons of 13 F . tularensis isolates , a negative correlation was indeed found between intergenic distance and dN/dS for the most closely related genomes , in support of non-stationarity ( Figure 4 ) . The correlation disappeared , however , at intergenic distances >0 . 15% , beyond which dN/dS ratios asymptotically approached a value of ∼0 . 5 , suggesting that stationarity of dN/dS was reached . These findings indicate that , because of time dependence , dN/dS ratios will be inflated for branches between very closely related genomes ( e . g . OSU18 , FTA , LVS ) . The lineage-specific analyses of dN/dS ratios , however , were performed using less closely related genomes , a precaution taken to limit effects of time dependence . In analyses of genome-wide mutational biases , we noted a correspondence between a relative surplus of G+C→A+T mutations and a reduced level of purifying selection ( Table S4 and Figure 3 ) . Among seven complete Francisella genome sequences ( U112 , FSC147 , SCHUS4 , WY-96 , LVS , FTA , OSU18 ) there is an overall evolutionary pattern of step-by-step degradation of genes , which are ultimately deleted . The majority of gene disruptions found in modern F . tularensis strains occurred independently along the five major genetic branches of F . tularensis ( Table S5 , S6 ) . That is , we see a converging evolutionary scenario among the F . tularensis lineages towards a common functional gene set . The strain F . novicida U112 has the least degraded , and largest , of all the analyzed genomes , containing 1 , 731 protein-coding genes and only 14 pseudogenes , according to recent annotation by Rhomer at al . [22] . A common set of 1 , 162 genes was identified that appeared to be functional in all seven genomes . This set of genes represents functions that have been preserved amongst F . novicida strain U112 and the six F . tularensis strains . Next , we identified gene function losses across the reconstructed phylogeny using a parsimony criterion . The absence of a full-length gene in two terminal branches was taken to indicate an absence in the nodes connecting the branches . We identified 798 gene function losses , in total , across the phylogeny depicted in Figure 5 , of which only ca . 62% ( 495/798 ) show inactivation patterns that are congruent with the inferred SNP phylogeny . In reality this is an overestimate of the proportion of congruent events , since in some cases there are likely to have been several independent disruptions of the same gene , which will remain undetected along internal branches including the branch from F . novicida to the last common ancestor of F . tularensis ( 279 congruent gene losses are indicated in Figure 5 ) . We counted 166 gene losses and 109 predicted pseudogenes , in comparison with the F . novicida U112 genome . An additional four genes were either absent or are pseudogenes in the six F . tularensis strains , making the total of 279 gene function losses . The function of these genes was presumably lost after the divergence of F . novicida and F . tularensis from a common ancestor . An alternative scenario , of gene acquisition in the F . novicida branch , can easily be dismissed by inspecting intact and disrupted genes in the genomes , since we identified only 11 genes that are absent in F . novicida compared with the common F . tularensis-F . novicida gene set . Mapping of gene function losses on the SNP tree shows additional distinct losses along each of the major genetic branches of F . tularensis . The virtual absence of recombination among the F . tularensis lineages makes further analysis fruitful . It is clear that the majority of gene function losses have occurred independently along the branches ( Figure 5 ) . We explored further by simulation possible characteristics of a putative dispensable gene set . A simplistic model was used with genes randomly sampled for inactivation under a uniform distribution model along branches according to frequencies previously identified by parsimony-mapping . Using this method , expected numbers of apparently homoplastic gene disruptions were inferred for different effective gene set sizes . The number of homoplasies in the real data was found to greatly exceed the numbers expected for any gene sample size , with a maximum of ∼80 homoplasies at an optimal gene sample size of 400 ( Figure S1 ) . An additional simulation was performed with increased numbers of homoplasies to account for the possibility that the true number of homoplasies may be greater than apparent numbers , but again similar results were obtained ( data not shown ) . In agreement with previous suggestions , we found strong indications that rearrangements in F . tularensis have been mediated by IS elements after divergence from a common F . tularensis ancestor [21] , [22] . Mapping of flanking sequences of ISFtu1 and ISFtu2 in three completed F . novicida and eight F . tularensis genomes identified a most parsimonious scenario for the pattern and order in which the IS elements have been inserted during the course of evolution ( Figure 6 ) . The ISFtu2 distribution in the genomes was consistent with largely independent past increases in ISFtu2 numbers in F . novicida and F . tularensis . We counted 17 ISFtu2 elements in F . novicida U112 , but found that only a single element had a corresponding flanking nucleotide sequence in a F . tularensis genome ( at 246 , 100 bp in F . novicida U112 ) . Thirteen ISFtu2 elements were likely inserted into ancestral F . tularensis taxa before the formation of the major genetic lineages of F . tularensis , since they share one or both flanking sequences in all F . tularensis genomes . The rate of expansion of ISFtu2 was reduced before the formation of the tularensis-mediasiatica genetic clade , seen as an extensive positional conservation of sequences flanking 16–18 ISFtu2 elements in both branches ( Figure 6 ) . During formation of the holarctica clade , ISFtu2 continued to be inserted at novel positions , resulting in a total of 42–44 occurrences in this subspecies . The ISFtu1 element appears to have a different evolutionary history . Mapping the flanking sequences of ISFtu1 shows that nearly all ISFtu1 elements found in F . tularensis strains ( N = 44 ) were inserted before formation of subsp . holarctica , but later than the divergence of F . novicida ( Figure 6 ) . The patterns of IS elements also indicates that there were no individual expansions of ISFtu1 in the A1 or A2 clades of F . tularensis subsp . tularensis . No ISFtu1 border was unique comparing strain SCHU S4 with strain WY-96 . Six unique ISFtu1 element insertions occurred along the F . tularensis subsp . mediasiatica lineage . Using Multiple Genome Rearrangements , an inversion metric software package [33] , to analyze the order of 53 local collinear sequence blocks ( LCBs ) in seven completed and aligned sequences , we observed large rearrangement distances despite close genetic relationships at the SNP level among strains ( Figure 7 ) . No less than 78 inversions were required to explain the gene order data . In reconstructions without imposing topological constraints , the algorithm was unable to recover the SNP-based tree . Assuming an ancestral gene order close to that of the species with the lowest number of IS elements ( F . novicida U112 ) or a reconstructed order at its closest internal node , strain WY-96 displayed the lowest rearrangement distance ( in accordance with the shortest SNP distance; Figure 7 ) . The LVS lineage was inferred to have a shorter inversion distance than both SCHU S4 and FSC147 . The gene orders of WY-96 , SCHU S4 and FSC147 are highly divergent , resulting in large rearrangement distances despite their intimate relationships as determined by SNP analysis . Increased rates of rearrangements are therefore apparent in the mediasiatica FSC147 and tularensis A1 SCHU S4 lineages . A set of 16–19 genes denoted the Francisella pathogenicity island ( FPI ) is critical for phagosomal escape , and is present in duplicate in F . tularensis strains , but in a single copy in the less virulent F . novicida [34] , [35] . Given the importance of the FPI for intracellular replication , it is likely that the duplication represents an adaptation to a more restricted niche . Our analysis confirmed that the FPI exists in duplicate in all analyzed F . tularensis genomes ( F . tularensis subspp . holarctica , mediasiatica and tularensis ) and showed there were single copies of homologous genes in all the analyzed environmental Francisella genomes , i . e . Francisella philomiragia ATCC 25017 , F . novicida U112 , and the two novicida-like strains GA 99-3548 and GA 99-3549 ( Figure S2 ) . Its ubiquitous presence among strains of genus Francisella challenges the description of the FPI as a classic pathogenicity island , i . e . , a mobile locus promoting pathogenicity with specific presence in pathogens but absence in benign relatives [36] . However , in agreement with its designation as a pathogenicity island we reaffirm that the region likely has a lateral origin and was inserted into an ancestor of Francisella . Further , in Francisella , FPI genes appear to be part of the core genome . Analysis of proteins encoded within and outside the FPI demonstrated that the most over-represented amino acids within the FPI correspond to those encoded by the most GC-rich codon families , i . e . alanine , glycine , proline , arginine , tryptophan and cysteine . Accordingly , the most underrepresented amino acids are encoded by the most GC-poor codon families , i . e . isoleucine , tyrosine , aspargine , lycine and phenylalanine . Since the composition of encoded amino acids has been found to be strongly influenced by G+C content [37] , this finding indicates that the FPI was originally acquired from an organism with a higher G+C content . In agreement with long presence of the FPI in Francisella we found no significant differences at third codon positions in G+C composition between the FPI and other parts of the genome ( 17% for non-ribosomal proteins encoded outside the FPI , and 16% for those encoded within the FPI ) . The presence of an ISFtu1 insertion sequence element at one flank of the FPI has previously been assumed to have mediated its lateral transfer . However , irreconcilably with such a role , this ISFtu1 copy appears to have a more recent origin than the FPI itself , since it is present exclusively in members of F . tularensis . We infer that the flanking ISFtu1 instead likely played a role in the duplication of the FPI . As outlined in Figure 8 , the most parsimonious evolutionary scenario appears to be that insertions of an ISFtu1 element adjacent to FPI genes and a set of rRNA genes occurred in a common F . tularensis ancestor , followed by duplication of FPI genes by unequal recombination . An ancestral FPI unit position can also be inferred from the observed conservation of surrounding genes in F . novicida U112 , F . tularensis subsp . tularensis WY-96 and subsp . holarctica sequences . From its unique position among LCBs in different strains , it is also evident that the second copy has subsequently acted as an independent rearrangement unit in all F . tularensis lineages ( data not shown ) . The putative glycosyl transferase may represent a second example of gene multiplication in Francisella coinciding with a function that is central to the ecology of the bacterium . A gene which encodes a putative glycosyl transferase ( FTT0354 , FTT0378 , FTT1263 ) is found in two to four copies in all F . tularensis genomes , but not among environmental Francisella genomes ( F . novicida and F . philomiragia ) . This gene , unlike the FPI , may therefore have arisen in the F . tularensis lineage via a lateral gene transfer event . Again , flanking ISFtu1 elements seem to have subsequently mediated homologous recombination and gene multiplication events ( not shown ) . Under the hypothesis that the gene acquisition and amplification reflect adaptive processes , we analyzed non-synonymous/synonymous mutation ratios , and found significant evidence of positive selection according to codeml estimates [38] . The M2a model indicated a dN/dS ratio of 2 . 344 , and that 2 . 38% of the nucleotide sites have been positively selected . Likelihood ratio tests using both codeml model pairs M1a/M2a and M7/M8 indicated that the null hypothesis of neutral evolution should be rejected ( p = 0 . 0242 and p = 0 . 0282 , respectively ) .
We have performed a broad comparison of 17 Francisella genomes to infer past evolutionary events and possible ecological adaptations that distinguish the primarily human pathogenic F . tularensis from its less virulent opportunistic neighbors , F . novicida and F . philomiragia . Our analysis provides support for a proposed evolutionary scenario of events that formed F . tularensis . The analysis also offers important clarifications of several uncertainties and ambiguities regarding F . tularensis , more specifically concerning: ( i ) the phylogenetic origin of F . tularensis subsp mediasiatica , ( ii ) the suggested dependence of virulence on the mere occurrence of specific genes , ( iii ) the occurrence of genetic recombination in F . tularensis , and ( iv ) the previously suggested lateral mobility of the FPI in F . tularensis . In addition , the results provide evolutionary data indicating that strains of F . novicida should continue to be regarded as a separate species . Based on our results , we propose the following sequence of events . F . tularensis emerged from a recombining Francisella population that was relatively unrestricted or free-living , then an ancestral F . tularensis variant invaded a novel and more host-restricted niche . This event led to clonal evolution under a reduced purifying selection pressure with ensuing genome degradation and proliferation of insertion sequence elements . We further propose that this series of events provided important prerequisites for further alterations of genomic architecture , and possibly increased adaptability of F . tularensis . In support of this scenario , we found apparent differences in the evolutionary mode of basal Francisella lineages ( including F . philomiragia , F . novicida and an ancestral F . tularensis lineage ) , and F . tularensis lineages . These apparent differences include frequent genetic exchanges among basal Francisella lineages , but not among F . tularensis lineages , strong purifying selection pressure in basal lineages but weak levels in F . tularensis , and increasing frequencies of adenine and thymine nucleotides in F . tularensis but not in F . novicida genomes ( Table S3 ) . The pronounced increase in G+C→A+T mutations in F . tularensis supports a link to weak purifying selection , allowing for the fixation of slightly deleterious mutations . In agreement with this interpretation , Balbi et al . recently found inefficient purifying selection to be intimately connected with adenine and thymine enrichment in Shigella spp . [39] . Differences in genomic architecture were also apparent . While genome erosion appears to have occurred in all F . tularensis genomes , the representatives of basal lineages have maintained genomes tightly packed with genes . Corroborating a previous genomic analysis of F . novicida strain U112 , the unfinished genomic sequences of F . novicida strain GA99-3548 and strain GA99-3549 , as well as the completed F . philomiragia strain ATCC 25017 , were found to contain few IS elements and pseudogenes . The overall gene synteny was extensive among the three available F . novicida genomes ( data not shown ) . In contrast , all F . tularensis genomes are crowded with IS elements and pseudogenes , and display highly rearranged gene orders , each corresponding to a subspecies or a major genetic lineage . The findings in this study thus indicate that F . novicida has remained relatively unchanged over a long period with respect to gene content , presence of IS elements , and gene order . If so , the genomic architecture of the ancestor of F . tularensis must have more closely resembled F . novicida than any current F . tularensis isolate . Genomic data therefore indicate that the deviating evolutionary patterns in F . tularensis represent a derived state . The greater metabolic competence of F . novicida compared to F . tularensis , and the abundance of IS elements in F . tularensis ( but not F . novicida ) , provide additional indirect support for a change of living habitat . The genomic erosion identified in F . tularensis is consistent with its occupation of a habitat that supplies nutrients , making some metabolic functions superfluous . Host-pathogen or recent symbiotic restrictions appear to have been similarly associated with genome erosion and proliferation of IS elements in several other organisms , e . g . Yersinia pestis [40] , Bordetella pertussis [41] , and bacterial endosymbionts of insects [42] . Generally , IS element expansions in host-restricted bacteria are considered to be consequences of reductions in effective population size and relaxed purifying selection , which provide opportunities for insertions [3] . Supporting this hypothesis in F . tularensis is the bacterium's exceptionally high infectiousness , 10–25 cfu being sufficient to cause disease in humans , a trait consistent with repeated population contractions during infection of hosts . Assuming that IS elements proliferate as a result of reduced selection pressure , it follows that this is a neutral process that in itself provides no advantage for the bacterium [43] . A neutral random insertion of IS elements likely provided the necessary raw materials for secondary pathoadaptive mutations in F . tularensis . Out of the two genetic loci that were found to be multiplied in all F . tularensis genomes by an IS element-mediated process , both were found to represent functions of central importance to the pathogen . The first locus corresponds to the FPI , a critical virulence determinant recognized for its importance for phagosomal escape [44] . The other locus contains a hypothetical glycosyltransferase gene , which we here demonstrate has been under strong adaptive selection . F . tularensis may therefore provide an example of an organism for which random genetic drift , with consequent fixation of many neutral or slightly deleterious mutations , provided novel evolutionary opportunities . Although not providing definitive proof , we propose that secondary gene multiplications enabled by past random IS element insertions represent examples of adaptively selected traits of the bacterium . In line with arguments recently advanced by Lynch [45] , our data suggest that an accumulation of mutations that were originally neutral or slightly deleterious to the organism in the short term proved to be fruitful in the long term when exploited by natural selection . As mentioned above , the data presented here also offer possible clarifications of several uncertain aspects and ambiguities regarding F . tularensis . Finally , the results provide compelling arguments in favour of continuing to regard strains of F . novicida as belonging to a separate species . In agreement with a previous proposal by Hollis et al . , based on DNA-DNA re-association [50] , our ANI analysis indicates that strains belonging to F . novicida meet formal requirements for classification as a F . tularensis subspecies . All pairs of isolates classified as F . novicida , novicida-like , and F . tularensis demonstrated ANI values well above 95% ( Table S1 ) , a limit proposed as the threshold for classification into different bacterial species [51] . According to the method-free species concept recently outlined by Wagner and Achtman [52] , however , species should be regarded as “metapopulation lineages” where separate designations are warranted if population lineages evolved separately despite a close relatedness . Our comparisons of environmental lineages ( F . novicida , F . philomiragia ) and F . tularensis show a typical example of such evolutionary separation . In addition to distinct population structures with regard to recombination , we also found substantial differences in overall dN/dS between environmental Francisella and F . tularensis ( Figure 3 ) , lending support to smaller effective population sizes in the latter . Other differences between environmental Francisella and F . tularensis include differences in metabolic competence , which is higher among environmental strains , and signs of ongoing genome erosion , which is pronounced among F . tularensis strains but not among the analyzed F . philomiragia and F . novicida strains . It is also clear that tularemia caused by F . tularensis is a distinct clinical disease entity with little similarity to the bacteraemia caused by F . novicida [50] , [53] . Moreover , tularemia is a classical vector-borne zoonosis while F . novicida is not known to be transmitted among vertebrate species , and F . tularensis is considered a biothreat agent while F . novicida is not . A fuzzy distinction between these quite different organisms may therefore complicate clinical decisions . Based on the evolutionary analyses described in this work , their distinct epidemiological features , and on clinical grounds: even though their average nucleotide identities exceed 97% , we propose that the species boundary between F . tularensis and F . novicida should be retained .
DNA for genomic sequencing of F . tularensis subsp . mediasiatica FSC147 was prepared as described in Text S1 . The genome was sequenced at the Joint Genome Institute using small ( 2–3 kb ) and medium ( 6–8 kb ) insert plasmid libraries . The Phred/Phrap/Consed software package was used for sequence assembly and quality assessment [54] . During the manual finishing process , possible mis-assemblies were corrected by transposon bombing ( Epicentre Biotechnologies ) of bridging clones . Gaps between contigs were closed by editing in Consed , by custom primer walks , or by PCR amplification . The nucleic acid sequence and annotation of F . tularensis subsp . mediasiatica strain FSC147 was deposited in GenBank under accession no . CP000915 . 1 . Automatic annotation using TIGR's annotation engine for gene prediction , GO classification , EC numbers , and protein functions was performed . The annotation was then manually curated with the aid of previous annotations of F . novicida strain U112 and F . tularensis subsp . tularensis strain SCHUS4 . Information on these and other genomes used in this study is presented in Table S1 . All multiple alignments of genomic sequences were performed using Mauve v . 2 . 2 [55] with the progressive alignment option under default parameters . All alignments were visually inspected and potentially incorrectly aligned regions were removed before further analysis . Phylogenetic analyses were conducted using MEGA version 4 [56] and Phyml v . 2 . 4 . 4 [57] . Using 1 , 104 , 129 aligned sites from all 17 taxa ( Table S1 ) , Mega was used for Neighbor-joining based estimates of the Francisella phylogeny and 1 , 000 bootstrap pseudo-replicates were performed . The evolutionary distances were computed using Tamura's three-parameter distance model . For maximum likelihood-based analysis , a subset of 566 , 154 randomly selected aligned nucleotide sites were used to avoid software crashes using the Phyml package . The GTR model with a proportion of invariant sites and six gamma-distributed discrete rate categories was used , estimated from the data . Non-parametric bootstrapping was performed using 100 pseudo-replicates in the maximum likelihood analysis . Average nucleotide identity ( ANI ) estimates were obtained by whole-genome sequence comparisons . Using the Perl script language and the NCBI blastall package v . 2 . 2 . 17 , we implemented the algorithm and performed analyses as previously described elsewhere [51] . Analysis of recombination in basal parts of the phylogeny was assessed using the ClonalFrame algorithm [25] . Default parameters were used except that the topology was fixed to that estimated from phylogenetic analyses to increase the computational speed . Mauve-alignments for the analysis were based on genomic sequences of the strains GA99-3548 , GA99-3549 , F . novicida U112 , and F . tularensis LVS ( Table S1 ) . After manual curation to remove poorly aligned regions , 1 , 527 , 362 nucleic acid sites in 128 local collinear blocks were retained for the analysis . Markov chain Monte Carlo iterations were run for 500 , 000 generations The proportion of genes affected by recombination was assessed by the MaxChi2 method [27] and the Phi method , both implemented in the Phi Package [28] ( **p<0 . 01 ) . Recombination among F . tularensis lineages was analyzed using the homoplasy test as proposed by Maynard Smith and Smith [29] , implementing the algorithm as a Perl script ( available upon request ) on genomic information from 13 F . tularensis isolates ( Table S1 ) . The effective number of mutable sites ( Se ) , required by the method , was calculated as previously described [58]:where pS is the probability that two independent substitutions in the gene occur at the same site . Since F . tularensis lineages deviate significantly from mutational equilibrium , we avoided using the “outgroup method” proposed by Maynard Smith and Smith [29] . This approach would have overestimated Se in our case because of the strong deviation from stationary nucleotide frequencies observed in F . tularensis ( Table S3 ) . Instead , we calculated pS using the number of sites ( nA , nC , nG , nT ) , by their probability of mutation ( pA , pC , pG , pT ) as follows . Given that two independent mutations occur . For all sites ( i , j , k , l ) :WhereEstimates of numbers of sites and their probability of mutation were obtained from inferred ancestral nucleotide frequencies and mutations along the F . tularensis LVS branch from its division from the F . tularensis subsp . tularensis lineage , represented by the subsp . tularensis SCHU S4 strain . The reconstruction was performed using a maximum parsimony method in which only two-fold degenerate sites were used and the genomic sequence of F . novicida U112 was included as an outgroup . Sites with variable amino acids , changes at the first site of a codon , and sites coding for tryptophan and methionine residues were excluded from analysis to reduce the impact of selection . Analysis of recombination was also performed by visual examination along genomic alignments of segregating sites , parsimony-informative sites and homoplasies . Genome-wide plots to support the analysis were generated using in-house Perl scripts . Assessment of positive selection for the multiplicated glycosyl transferase gene ( corresponding to locus tags FTT0354 , FTT0378 , and FTT1263 in the F . tularensis SCHU S4 genomic annotation ) was performed using Codeml in the Paml 4b package [38] . The probability of positive selection was assessed by likelihood ratio tests ( LRT ) for the hierarchical model pairs M1a vs . M2a and M7 vs . M8 . The HYPHY package [30] was used to assess branch-specific selectional regimes in Francisella . Two sets of analyses were performed . In one , entirely local models were fitted to the data and estimates of dN and dS were allowed to vary freely for each branch . Confidence intervals were here determined using the asymptotic normality of the maximum likelihood estimates . In the other , branch estimates of dN/dS were obtained using a genetic algorithm [30] , ensuring that the data had not been overfitted . In all analyses , the Muse-Gaut 94 ( MG94 ) 3×4 model [59] crossed with the general time-reversible ( GTR ) model was used , justified by parametric bootstrapping ( in comparison with the Goldman-Yang 94 , GY94 , model [60] ) and by likelihood ratio tests . For analysis of time-dependence of dN/dS estimates , pairwise estimates of synonymous , non-synonymous , and intergenic evolutionary distances between isolates was estimated using MEGA version 4 [56] using the Nei-Gojobori method [61] . A modified version of the Psi-Phi package [62] was applied using U112 as the reference genome to identify pseudogenes . The default settings for Psi-Phi were used except for the merging distance , which was set to 1350 to allow for ISFtu1 insertion events . A parsimony criterion was applied to determine the functional status of genes in internal nodes of a whole genome SNP phylogeny . The absence of a full length gene in two terminal branches was taken to indicate an absence in the nodes connecting the branches . The sum of gene deletions and pseudogenes constituted the total amount of gene function loss . Both congruent and homoplastic gene function losses were considered , using the whole genome SNP tree as a reference . An alignment including 1 , 762 , 117 nucleotides in F . novicida U112 was used to explore genome-level rearrangements . Duplicated sequences including ribosomal RNA genes , the 30–34 kb duplicated sequence , and all IS elements were masked prior to rearrangement analysis . The LCB-weight was set to remove any very short collinear blocks , since we reasoned that these LCBs may be prone to duplication followed by random deletion events , which could lead to incorrect reconstructions . For calculating rearrangement scenarios based on inversions , the five gene orders representing strains U112 , SCHU S4 , WY-96 , and the common gene orders for LVS , FTA and OSU18 were analyzed using MGR software [33] , run in the circular genomes mode both with and without a fixed tree topology . Locus tags referred to in the text correspond to those used in the annotation of the F . tularensis SCHU S4 genomic sequence ( AJ749949 ) : FTT0354 , FTT0378 , FTT1066-FTT1073 , FTT1263 , FTT1308c , FTT1580c , FTT1581-FTT1582 and FTT1791 . Completed genomic sequences with accession numbers used in this work are: U112 ( CP000439 ) , ATCC 25017 ( CP000937 ) , WY96-3418 ( CP000608 ) , FSC147 ( CP000915 ) , FTA/FTNF002-00 ( CP000803 ) , OSU18 ( CP000437 ) , LVS ( AM233362 ) , SCHU S4 ( AJ749949 ) . Preliminary sequence data were obtained from the MIT Broad Institute website at www . broad . mit . edu for the following Francisella strains , GA99-3549 , GA99-3548 , FSC033 , FSC022 , and FSC257 , and from the Baylor College of Medicine Human Genome Sequencing Center website at www . hgsc . bcm . tmc . edu for the following Francisella strains: ATCC 6223 , KO 97-1026 , MI 00-1730 and OR 96-0246 .
|
The intracellular bacterium Francisella tularensis causes the disease tularemia in various mammals , including humans , and is highly infectious ( so infectious that highly virulent forms of the pathogen were developed as biological aerosol weapons during the Cold War ) . Little is known about where F . tularensis resides in nature and how it evolved but , intriguingly , closely related Francisella bacteria are less dangerous . Therefore , we have explored the evolutionary events that shaped F . tularensis by analyzing 17 Francisella genome sequences . Its evolution appears to have involved many losses of metabolic functions and random mutations , with little exchange of genetic material among F . tularensis strains . Furthermore , increased host association appears to have irreversibly separated F . tularensis populations from other populations of Francisella bacteria . This study provides new information on the processes whereby relatively harmless Francisella bacteria evolved into aggressive invaders of mammalian cells . Our findings support previous proposals that identification of distinct population lineages provides meaningful species boundaries among bacteria .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"infectious",
"diseases/bacterial",
"infections",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"microbiology/microbial",
"evolution",
"and",
"genomics",
"genetics",
"and",
"genomics/population",
"genetics"
] |
2009
|
Molecular Evolutionary Consequences of Niche Restriction in Francisella tularensis, a Facultative Intracellular Pathogen
|
The gene mutated in Bloom's syndrome , BLM , is important in the repair of damaged replication forks , and it has both pro- and anti-recombinogenic roles in homologous recombination ( HR ) . At damaged forks , BLM interacts with RAD51 recombinase , the essential enzyme in HR that catalyzes homology-dependent strand invasion . We have previously shown that defects in BLM modification by the small ubiquitin-related modifier ( SUMO ) cause increased γ-H2AX foci . Because the increased γ-H2AX could result from defective repair of spontaneous DNA damage , we hypothesized that SUMO modification regulates BLM's function in HR repair at damaged forks . To test this hypothesis , we treated cells that stably expressed a normal BLM ( BLM+ ) or a SUMO-mutant BLM ( SM-BLM ) with hydroxyurea ( HU ) and examined the effects of stalled replication forks on RAD51 and its DNA repair functions . HU treatment generated excess γ-H2AX in SM-BLM compared to BLM+ cells , consistent with a defect in replication-fork repair . SM-BLM cells accumulated increased numbers of DNA breaks and were hypersensitive to DNA damage . Importantly , HU treatment failed to induce sister-chromatid exchanges in SM-BLM cells compared to BLM+ cells , indicating a specific defect in HR repair and suggesting that RAD51 function could be compromised . Consistent with this hypothesis , RAD51 localization to HU-induced repair foci was impaired in SM-BLM cells . These data suggested that RAD51 might interact noncovalently with SUMO . We found that in vitro RAD51 interacts noncovalently with SUMO and that it interacts more efficiently with SUMO-modified BLM compared to unmodified BLM . These data suggest that SUMOylation controls the switch between BLM's pro- and anti-recombinogenic roles in HR . In the absence of BLM SUMOylation , BLM perturbs RAD51 localization at damaged replication forks and inhibits fork repair by HR . Conversely , BLM SUMOylation relieves its inhibitory effects on HR , and it promotes RAD51 function .
Homologous recombination ( HR ) is a high-fidelity DNA repair mechanism that functions to rejoin double-strand breaks ( DSBs ) and restart broken replication forks . A major outcome of the repair of replication fork damage by HR is the generation of sister-chromatid exchanges ( SCEs ) , which result from resolution of Holliday junctions during HR repair [1] , [2] . Predictably , a large number of agents that cause DNA damage increase the frequencies of SCEs [3]–[5] . Bloom's syndrome ( BS ) is the only clinical entity in which increased levels of SCE are a prominent cellular feature [6] . It is an autosomal recessive disorder , which is characterized by proportional dwarfism , photosensitivity , immunodeficiency , hypogonadism , and predisposition to a wide range of different types of cancer [7] . BS is caused by biallelic null mutations of the BLM gene [8] . The BLM gene encodes a DNA helicase of the RecQ family , which is an evolutionarily conserved group of enzymes that operates at the interface of DNA replication , HR , and DNA repair [9] . The RecQ helicases are DNA-dependent ATPases that can translocate on single-stranded DNA ( ssDNA ) with 3′ to 5′ directionality [10] . In vitro , they preferentially unwind DNA substrates that resemble recombination intermediates , including G4 tetrahelical DNA , Holliday junctions , double Holliday junctions , and D-loops . A complex consisting of BLM , topoisomerase IIIα , BLAP75 , and BLAP18 ( BLAPs are BLM-associated proteins ) can “dissolve” a substrate representing a double Holliday junction—a late intermediate in HR-mediated repair of DSBs—in such a way that crossing over would not occur between DNA strands [11]–[13] . This activity could provide an explanation for the increased SCEs in BS cells; however , recent genetic and biochemical studies have shown that BLM also has activities in upstream parts of the HR pathway . Because BLM is recruited to damaged replication forks early in the repair process [14]–[16] , it could suppress the formation of aberrant recombination intermediates at the replication fork . Such a mechanism has been proposed for Sgs1 , the yeast homolog of BLM [17] , [18] . BLM interacts directly with the RAD51 recombinase , which is the enzyme that catalyzes homology-dependent strand invasion [19] , [20] , and in vitro it can displace RAD51 from ssDNA and unwind the invading DNA strand of a D-loop formed by RAD51 [21] , [22] , suggesting that BLM regulates the formation of D loops . Finally , BLM and Sgs1 each collaborate with exonucleases that process DSBs to generate ssDNA with a 3′ tail , which is the substrate for RAD51 [23]–[26] . Collectively , these data show that BLM has both pro- and anti-recombinogenic functions in HR . A key question that emerges from these studies is how are these different functions of BLM in HR regulated ? Modification by the small ubiquitin-related modifier ( SUMO ) has emerged as an important regulator of HR [27] . In response to replication fork damage in the budding yeast Saccharomyces cerevisiae , the polymerase processivity factor PCNA ( proliferating cell nuclear antigen ) is SUMOylated , and PCNA SUMOylation recruits the DNA helicase Srs2 to the fork , which functions to prevent aberrant recombination events between sister chromatids [28]–[31] . Mutants of the SUMO-specific E3 ligase gene MMS21 accumulate RAD51-dependent cruciform structures at damaged replication forks [32] , which are aberrant structures that also accumulate in sgs1 deletion mutants [33] . These studies indicate that SUMO modification can play important roles in response to damaged forks; however , the role of SUMO in regulation of HR is not fully understood , and these mechanisms have not been studied in mammalian cells . We have previously shown that BLM is SUMOylated and that failure to SUMOylate BLM results in changes in BLM's nuclear distribution [34] . Expression in cells of SUMO-mutant BLM , containing lysine to arginine mutations at residues 317 and 331 that prevent SUMOylation , induces excess γ-H2AX foci—a marker for DNA damage and repair—in the absence of exogenously induced DNA damage [34] . Despite the presence of excess γ-H2AX foci and micronuclei in cells that expressed SUMO-mutant BLM , there was insufficient evidence to conclude that SUMO-mutant BLM generated excess DNA damage . Because SUMOylation is known to regulate the localization of proteins in the nucleus [35] , we hypothesized that the accumulation of BLM in γ-H2AX foci could result from a kinetic defect in recruitment of BLM back to the promyelocytic leukemia ( PML ) nuclear bodies ( PML-NBs ) . In the present study , we aimed to determine how SUMOylation regulates BLM's function in maintaining genomic integrity . We hypothesized that cells that express SUMO-mutant BLM have a DNA repair defect . Characterization of cells that expressed SUMO-mutant BLM revealed that SUMOylation of BLM regulates its association with RAD51 and its function in HR-mediated repair of damaged replication forks . Our data support a model in which SUMOylation of BLM acts as a switch to regulate its effects on recombination .
Because BLM functions at damaged replication forks and γ-H2AX is a marker for DNA damage , we hypothesized that SUMO-mutant BLM is defective in repair of damaged replication forks . To gain insight into this question , we introduced GFP-BLM expression constructs into the BS cell line GM08505 , isolated clones that stably expressed either normal BLM ( BLM+ cells ) or SUMO-mutant BLM ( SM-BLM cells ) , and studied these clones for responses to replication fork damage . We treated SM-BLM and BLM+ cells with 0 . 5 mM hydroxyurea ( HU ) for 24 h , which stalls replication forks , and quantified the production of γ-H2AX by immunofluorescence , Western blot , and flow cytometry analyses ( Figure 1 ) . The 24-h HU treatment blocks approximately 80% of the cells in S phase , simultaneously providing a primary synchronization of the cells and stressing replication forks through nucleotide deprivation . As we reported previously , untreated SM-BLM cells exhibited higher levels of γ-H2AX foci compared to untreated BLM+ cells ( 35 . 9 vs . 16 . 4 , respectively; Figure 1A ) . Treatment of SM-BLM and BLM+ cells with HU resulted in a larger increase in γ-H2AX foci per cell in SM-BLM compared to BLM+ cells ( a gain of 56 . 8 vs . 42 . 7 foci ) . Particularly notable were the presence of SM-BLM cells that stained brightly with γ-H2AX ( Figure 1B ) . HU induced twice the numbers of γ-H2AX-bright nuclei in SM-BLM cells than in BLM+ cells ( Figure 1C ) . Consistent with the immunofluorescence analysis , by Western blot and flow cytometry analyses , the levels of γ-H2AX in HU-treated SM-BLM cells were higher than in HU-treated BLM+ cells ( Figure 1D and 1E ) . Because the results could have been influenced by cell-cycle effects , we analyzed nuclear DNA content and measured BrdU incorporation at different times after release from the HU block by flow cytometry . After treatment with HU , in both SM-BLM and BLM+ clones , the majority of cells were blocked in early S phase , and they progressed to mid-S phase by 6 h after release from the HU block ( Figure S1 ) . These data indicated that differences in position in the cell cycle or irreversibility of the S phase block did not explain the differences in the accumulation of γ-H2AX after HU treatment of SM-BLM and BLM+ cells . In summary , these experiments showed that SM-BLM cells exhibited excess phosphorylated H2AX in both untreated and HU-treated conditions . The presence of increased levels of spontaneous and HU-induced γ-H2AX strongly suggests the presence of excess DNA damage . In order to obtain direct evidence for the presence of DNA damage , we analyzed HU-treated and untreated cells for DSBs by pulsed-field gel electrophoresis ( PFGE ) ( Figure 2 ) . In the absence of treatments , SM-BLM cells exhibited over 1 . 5 times more DSBs compared to BLM+ cells ( Figure 2B ) . This result confirmed the presence of increased numbers of DSBs consistent with the higher numbers of γ-H2AX foci . After a 24-h treatment with HU , SM-BLM cells again exhibited 1 . 5 times more DSBs compared to BLM+ cells . Because 80% of the cells are blocked with stalled forks , the DSBs detected in HU treatment conditions likely originate from fork breakage . After release from the HU block , DSBs accumulated over time , with the total number of DSBs observed in SM-BLM cells being greater at each time point than the total number in BLM+ cells ( Figure 2B ) . A 24-h HU treatment induced more DSBs in BS cells compared to no treatment , and BS cells also accumulated more DSBs after release from the HU block compared to either SM-BLM or BLM+ cells ( Figure S2 ) . Consistent with the PFGE analysis , the total number of HU-induced micronuclei was greater in SM-BLM compared to BLM+ cells ( Figure 2D ) . We also analyzed the numbers of DSBs in cells treated with camptothecin ( CPT ) , which generates replication-associated DSBs [36] , [37] . Treatment with different concentrations of CPT for 3 h generated two times more DSBs in SM-BLM compared to BLM+ cells , showing that breaks accumulate at an accelerated rate in SM-BLM cells ( Figure 2C ) . Altogether , these data were consistent with the hypothesis that SM-BLM cells have a defect in the repair of replication-associated DSBs . Because SM-BLM cells exhibited higher levels of DSBs induced by replication damage , we expected SM-BLM cells to be hypersensitive to DNA damage encountered during S phase . To test this hypothesis , we compared the levels of cell death in cells exposed to replication damage , using a standard flow cytometry assay ( Figure 3 ) . In the absence of HU or etoposide treatment , BLM+ and SM-BLM cells exhibited similar levels of cell death . A 24-h treatment with HU induced a 3% increase in cell death in BLM+ cells as compared to a 10 . 6% increase in SM-BLM cells ( p = 0 . 01 ) , demonstrating that SM-BLM cells have increased sensitivity to HU treatment alone . Similarly , a 24-h treatment with etoposide induced an 11 . 3% increase in cell death in BLM+ cells compared to a 20 . 7% increase in SM-BLM cells ( p = 0 . 003 ) , demonstrating that SM-BLM cells are also hypersensitive to etoposide treatment compared to BLM+ cells . After HU pretreatment , etoposide induced a 13 . 5% increase in cell death in BLM+ cells , which was similar to the level of cell death observed without HU pretreatment ( 11 . 3% ) ( p = 0 . 84 ) , whereas after HU pretreatment , etoposide induced a 43 . 1% increase in cell death in SM-BLM cells , which was 2-fold greater compared to the level observed without HU pretreatment ( 20 . 7% ) ( p<0 . 001 ) . As expected , BS cells that lack BLM protein are also hypersensitive to DNA damage encountered during S phase . In corroboration of these results , in colony survival assays , we also observed increased sensitivity of SM-BLM cells to CPT compared to BLM+ cells ( Figure S3 ) . These data indicated that SM-BLM cells are more sensitive than BLM+ cells to DNA damage generated during S phase , again consistent with a defect in the repair of replication-associated DNA damage . Replication-associated DSBs are repaired by HR , which generates increased numbers of SCEs [1] . Because SM-BLM cells exhibited excess HU-induced DSBs , we hypothesized that HR is impaired in SM-BLM cells . Therefore , we tested whether replication stalling induced fewer SCEs in SM-BLM compared to BLM+ cells ( Figure 4 ) . Untreated BLM+ and SM-BLM cells showed similar numbers of SCEs ( 17 . 4 vs . 16 . 7 SCEs/46 chromosomes , respectively ) . However , whereas HU treatment induced a 2-fold increase in SCEs in BLM+ cells ( from 17 . 4 to 29 . 6 SCEs/46 chromosomes; p<0 . 001 ) , HU treatment had almost no effect on the levels of SCEs in SM-BLM cells ( from 16 . 7 to 18 . 5 SCEs/46 chromosomes; p = 0 . 32 ) . The numbers of HU-induced SCEs in BLM+ compared to SM-BLM cells was significantly different ( p<0 . 001 ) . In contrast to SM-BLM cells , HU induced a large increase in SCEs in BS cells ( Figure S4A ) . These data suggest that HR repair is not engaged normally at damaged replication forks , leading to the excess DSBs that are observed in HU-treated SM-BLM cells . It is worth noting that in our earlier report on BLM SUMOylation [34] , we found that the mean number of SCEs in untreated SM-BLM cells was greater than the mean number in BLM+ cells . This difference was caused by the presence in the SM-BLM cultures of 5% of cells with SCE levels equal to levels typically observed in BS cells , whereas we had detected no cells of this type in the BLM+ cultures . In the present study , we did not detect cells with high SCEs in the SM-BLM cultures; consequently , we suggest that these high-SCE cells were produced by extinction of SM-BLM expression in a small fraction of SM-BLM cells . RAD51 is a key enzyme in HR repair , and normally it interacts with BLM at damaged replication forks [19] , [20] . Because SM-BLM cells exhibited a defect in HR-mediated repair after replication stalling , we examined whether RAD51 and BLM colocalize normally at γ-H2AX–marked damage in SM-BLM cells ( Figure 5 ) . In untreated cells , SM-BLM cells contained more RAD51 foci than BLM+ cells ( 26 . 9 vs . 18 . 1 foci/cell ) . Similarly , as previously noted [34] , untreated SM-BLM cells contained more BLM foci ( 19 . 0 vs . 11 . 0 foci/cell ) ( Figure 5B ) . However , after treatment with 0 . 5 mM HU for 24 h , whereas BLM+ cells exhibited a large increase in RAD51 foci from 18 . 1 to 51 . 7 foci/cell , SM-BLM cells exhibited only a modest increase from 26 . 9 to 34 foci/cell ( Figure 5B ) . Consistent with these observations , whereas HU treatment induced substantial increases in RAD51-γ-H2AX and RAD51-BLM colocalization in BLM+ cells , HU treatment induced only a modest increase in these colocalizations in SM-BLM cells ( Figure 5C and Figure S5 ) . In HU-treatment conditions , whereas 66% of the γ-H2AX foci contained RAD51 in BLM+ cells ( vs . 58% in untreated cells ) , only 25% of γ-H2AX foci contained RAD51 in SM-BLM cells ( vs . 43% in untreated cells—a decrease ) . As a positive control , we found that RAD51 showed much higher levels of colocalization with γ-H2AX in HU-treated BS cells compared to BLM+ cells ( Figure S4B ) , as previously reported [19] . To distinguish whether the RAD51 localization defect was an early or late effect of replication stalling , we treated BLM+ and SM-BLM cells with 10 mM HU for 1 h and stained cells with antibodies to RAD51 and PCNA ( to identify cells in S phase ) . Whereas BLM+ cells exhibited a 2 . 3-fold increase in RAD51 foci from 14 . 4 to 33 . 6 foci/cell in PCNA-positive cells , SM-BLM cells exhibited no increase in total RAD51 foci from 19 . 7 to 21 . 0 foci/cell ( Figure 5D ) . Synchronization experiments with mimosine followed by treatment with 10 mM HU for 1 h corroborated these data ( Figure S6 ) . It is worth noting that both BLM+ and SM-BLM proteins localized efficiently with PCNA foci in HU-treated cells , indicating that SUMOylation is not required for normal trafficking of BLM to stalled forks ( Figure S7 ) . Altogether , these data demonstrated that there is a dramatic defect in RAD51's recruitment to and/or retention in repair foci induced by replication stalling . The RAD51 localization defect could explain both the impairment of HR after replication stalling and the excess DSBs observed in SM-BLM cells . To investigate the mechanism that might explain the RAD51 localization defect , we considered the possibility that RAD51 interacts with SUMO noncovalently , which would facilitate interaction between RAD51 and SUMOylated BLM . To test this hypothesis , we assayed for possible noncovalent interactions between RAD51 and SUMO and also for possible effects of covalent SUMOylation of BLM on its interactions with RAD51 ( Figure 6 ) . In an in vitro binding assay , more RAD51 bound to SUMO-coated beads than to controls beads ( Figure 6A ) , showing that RAD51 binds equally well to both SUMO-1 and SUMO-2 . To test whether SUMO modification of BLM affects its interaction with RAD51 , we incubated RAD51-coated beads with either unmodified BLM or with a mixture of SUMO-2–modified and unmodified BLM and analyzed bound proteins by Western blot with anti-BLM antibodies . Consistent with previous findings [20] , unmodified BLM bound specifically to RAD51-coated beads , confirming that BLM and RAD51 interact directly ( Figure 6B , left panel ) . SUMO-2–modified BLM also bound specifically to RAD51-coated beads ( Figure 6B , right panel ) . To evaluate the effect of SUMO-2 modification on BLM's interaction with RAD51 , the results of the binding assays were analyzed quantitatively . This analysis revealed that , whereas the ratio of SUMO-2–modified to unmodified BLM in input fractions and fractions bound to control beads was approximately 1 . 1∶1 and 1 . 7∶1 , respectively , the ratio present in fractions retained on RAD51-coated beads was approximately 5 . 1∶1 ( Figure 6C ) . These binding ratios reveal that SUMO-2 modification of BLM has a strong , positive effect on its binding to RAD51 . Altogether , these data demonstrated that RAD51 contains a SUMO binding site ( s ) and that SUMOylation of BLM can affect its interactions with RAD51 . The data support the hypothesis that SUMOylation of BLM facilitates repair of damaged replication forks by HR by modulating RAD51's recruitment and/or retention at repair sites .
The data presented here demonstrate the importance of BLM SUMOylation in the repair of damaged replication forks . Failure to SUMOylate BLM resulted in excess damage-induced repair foci , DSBs , and hypersensitivity to DNA damage . Importantly , in SM-BLM cells , replication stalling by HU did not stimulate HR as measured by SCEs , suggesting a defect in RAD51 function . Consistent with these data , RAD51 failed to accumulate at stalled forks . SUMO-mutant BLM exerts a dominant effect , because excess γ-H2AX foci were induced by expression of SUMO-mutant BLM in HeLa cells , which express endogenous normal BLM [34] . Moreover , we showed here that several SM-BLM phenotypes differed from BS phenotypes , such as the presence of excess H2AX phosphorylation in untreated cells , the RAD51 localization defect , and the lack of HU-induced SCEs . We found that RAD51 is a SUMO-binding protein , implicating BLM SUMOylation in recruitment of RAD51 to repair sites through a mechanism involving noncovalent SUMO interactions . Our findings demonstrate that BLM SUMOylation regulates the recruitment and/or retention of RAD51 to damaged replication forks , and it is important in HR-mediated repair . The steady-state levels of DSBs in cells are a function of the rate at which DNA damage accumulates and the rate at which it is repaired . We observed that SM-BLM cells exhibited greater numbers of DSBs than BLM+ cells under a variety of conditions ( Figure 2 ) . For example , SM-BLM cells exhibited more HU-induced DSBs , which arise due to breakage of stalled forks , and more CPT-induced DSBs , which arise due to replication runoff at sites where topoisomerase-cleavage complexes are bound to the DNA [38] . Formally , the presence of increased levels of DSBs in SM-BLM cells could result from an increase in the rate at which DNA damage accumulates ( due to excess numbers of replication forks , increased numbers of topoisomerase cleavage complexes , or a failure to process aberrant replication intermediates ) or from a decrease in the rate of DNA repair ( due to a failure to recruit RAD51 ) . We noted that the rate of DSB accumulation after release from the HU block was the same in both SM-BLM and BLM+ cells , indicating that the rate of breakage exceeds the rate of repair under these conditions in both types of cells . Consistent with these observations , the numbers of γ-H2AX foci in both BLM+ and SM-BLM cells increase 6 h after release from the HU block ( unpublished data ) . Recent work has shown that BLM is present on a class of ultrafine anaphase bridges [39] , and it acts to separate interlinked DNA strands especially at loci with intrinsic replication difficulties [40] . BS cells consequently have a defect in separation of sister chromatids , resulting in more anaphase bridges; some fraction of the increased DSBs that arise in BS cells no doubt traces to breakage at sites of underreplicated DNA . BS cells exhibit an inadequate response to replication stress [16] , [41]–[43] , in which additional forks are initiated apparently to compensate for forks that have collapsed [44] , [45] . One possible explanation for the excessive numbers of γ-H2AX foci and DSBs in treated and untreated SM-BLM cells is that these cells have increased replication difficulties , as BS cells do , and they compensate by activating additional replication forks , which are concomitantly more likely to break after replication damage . According to this view , BLM SUMOylation helps prevent the collapse of replication forks in regions with hard-to-replicate DNA , perhaps by stimulation of BLM's activity in subverting aberrant recombination intermediates at stalled replication forks . Alternatively , BLM SUMOylation could promote HR-mediated repair of broken forks through the recruitment and/or retention of RAD51 at damaged forks . RAD51 is the DNA recombinase essential for HR-mediated DNA repair , and previous studies have demonstrated that RAD51 and BLM interact specifically in DNA-damaged cells [19] , [20] . On the basis of immunolocalization studies , we found that the recruitment and/or retention of RAD51 at sites of stalled DNA replication forks is impaired in SM-BLM cells . Whereas BLM+ cells exhibited a 4-fold increase in the number of colocalized γ-H2AX-RAD51 foci upon HU treatment , SM-BLM cells exhibited a <1 . 5-fold increase in these foci . On the basis of this finding , and the finding that HR-mediated DNA repair is defective in SM-BLM cells , we propose that BLM SUMOylation mediates the recruitment and/or retention of RAD51 to sites of DNA damage and thereby facilitates HR-mediated repair processes . Previous cell and biochemical studies have led to the view that BLM has both pro- and anti-recombinogenic functions . Most notably , BLM is important in stabilizing damaged replication forks [14]–[16] and repressing aberrant recombination events , as evidenced by the dramatic increase in levels of SCEs and loss of heterozygosity in BLM null cells [6] , [46] . In contrast , BLM is also predicted to promote HR by facilitating exonucleolytic resection of DSBs [23]–[26] , by stimulating synthesis-dependent strand annealing [21] , [47] , and by promoting noncrossover resolution of Holliday junctions [12] . Both pro- and anti-recombinogenic functions have likewise been proposed for Escherichia coli RecQ helicase [48] . Our finding that cells expressing SUMO-mutant BLM have a defect in the recruitment and/or retention of RAD51 to sites of DNA damage , and that they are defective in HR-mediated repair , supports a model in which SUMOylation of BLM acts as a switch to regulate its effects on recombination ( Figure 7 ) . In the absence of SUMOylation , we propose that BLM binds to stalled replication forks and suppresses aberrant HR by inhibiting excessive accumulation of RAD51 at repair sites . In the event that a stalled replication fork progresses to a DSB , stimulation of HR-mediated repair would be triggered by BLM SUMOylation and , consequently , more efficient recruitment and/or retention of RAD51 at the repair site . SUMOylation of BLM could regulate the recruitment and/or the retention of RAD51 at sites of DNA damage through several different mechanisms . SUMOylation could dissociate BLM from broken DNA ends , where it might otherwise inhibit RAD51 binding and function by displacing RAD51 from ssDNA or by unwinding D-loops [21] , [22] . SUMOylation could limit the binding of BLM to sites of DNA damage by altering its affinity to ssDNA or possibly by triggering its ubiquitin-dependent degradation . SUMO-mediated degradation has been described for PML and other proteins [49]–[51] . In the model that we currently favor , BLM SUMOylation could function to promote RAD51 localization at repair sites by stabilizing its interactions with BLM through a mechanism involving noncovalent SUMO binding . Consistent with this model , we found that RAD51 is a SUMO-binding protein and that it interacts more efficiently with SUMOylated BLM compared to unmodified BLM . RAD51 is recruited to damaged replication forks in BS cells [19] , [20] and to a limited number of sites of DNA damage in SM-BLM–expressing cells . Thus , multiple factors appear to control RAD51 recruitment to sites of DNA damage . Nonetheless , our findings support the hypothesis that in BLM-expressing cells , BLM SUMOylation promotes RAD51 recruitment and/or retention at sites of DNA damage and thereby facilitates HR-mediated DNA repair . SUMOylation of BLM is likely to have multiple roles . In addition to regulating its activity at sites of DNA damage as revealed in the current study , BLM SUMOylation may also be important in mediating the localization of BLM to PML-NBs in undamaged cells . SUMO modification can function to retain proteins in the PML-NBs [35] , and fluorescence recovery after photobleaching studies have shown that BLM rapidly associates and dissociates from the PML-NBs [52] . This on–off process may be mediated primarily through BLM's SUMO interaction motif , which is required for BLM localization to the PML-NBs and for BLM SUMOylation [53] , [54] . The presence of a SUMO binding site ( s ) in RAD51 suggests that its association with the PML-NBs may also be regulated through the SUMO pathway . Our results have broad implications for understanding not only how the integrity of replication forks are maintained under stress but also how SUMO modification regulates its substrates , because many proteins in the DNA repair and signaling pathways are SUMO substrates . Although the role of SUMO in HR function is not yet understood , it is clear that SUMOylation plays multiple roles in regulating the HR pathway through modifications of various HR factors , including Sgs1 [32] , Rad52 [55]–[57] , PCNA [58] , and other recombination-associated factors . sgs1 mutants and mutants of the SUMO-specific E3 ligase gene mms21 accumulate aberrant cruciform structures at damaged replication forks [32] , [33] . This genetic evidence suggests that SUMOylation is important in the regulation of HR , but there has been no direct evidence that SUMOylation occurs at the repair site . Because SUMO-mutant BLM accumulates at HU-induced replication fork damage , the present results indicate that BLM SUMOylation occurs at the sites of damaged replication forks , where it affects stabilization of stalled forks , trafficking of RAD51 to repair sites , and HR repair of damaged forks . Further experiments are now needed to characterize the spatial and temporal regulation of SUMOylation of the different repair factors in HR . In particular , we need to determine what signals activate BLM SUMOylation and how BLM SUMOylation is regulated at damaged forks .
For BLM Western analysis , rabbit polyclonal anti-BLM antibodies raised against the first 431 amino acids of human BLM [59] or commercially available antibodies ( A300-110A , Bethyl Laboratories ) were used . Anti-SUMO antibodies were used as described [54] . For indirect immunofluorescence , we used mouse monoclonal anti–γ-H2AX antibody ( Upstate ) , rabbit polyclonal anti-RAD51 antibodies PC130 ( Calbiochem ) , mouse monoclonal anti-PCNA antibody sc-56 ( Santa Cruz Biotechnology ) , Cy-5–labeled donkey anti-rabbit antibodies ( Jackson Labs ) , Alexa Fluor 594–labeled goat anti-mouse , Alexa Fluor 594–labeled goat anti-rabbit , and Alexa Fluor 647–labeled goat anti-mouse antibodies ( Invitrogen ) . Rat monoclonal anti-Hsc70 antibodies ( Assay Design ) were used as a loading control in Western analyses . The full-length BLM cDNA was cloned into the EGFP-C1 vector ( Clontech ) , which produced a GFP-BLM fusion protein with GFP at the N-terminus of BLM , as described previously [53] . The GFP-BLM construct was used as a template for the construction of BLMs that contain SUMO acceptor-site mutations , by substituting arginine for lysine at amino acid residues 317 and 331 using standard polymerase chain reaction–based methods [34] . The construct used in the experiments reported here contained mutations at both 317 and 331 . We stably expressed the normal BLM and SUMO-mutant BLM constructs in the SV40-transformed fibroblast cell line GM08505 ( BS cells ) and isolated multiple clones expressing each construct , as described previously [34] , [60] . To measure the levels of GFP-BLM expression , we prepared cell lysates in Laemli sample buffer , fractionated proteins by sodium dodecyl sulphate-polyacrylamide gel electrophoresis , and transferred the proteins to nitrocellulose membranes ( Bio-Rad ) . The membranes were then processed for Western blot analysis and probed with anti-BLM antibodies as described earlier [53] . Varying levels of BLM expression were detected in BLM+ and SM-BLM clones . We chose BLM+ and SM-BLM clones that had comparable levels of transgene expression . To measure DNA content and percentage of cells in each phase of the cell cycle , cells were harvested by trypsinization and fixed in 70% ethanol for >3 h at −20°C . After fixation , cells were pelleted and resuspended in a solution of 1× phosphate buffered saline ( PBS; Gibco ) containing propidium iodide ( 10 µg/ml ) and RNase A ( 0 . 1 mg/ml ) . The fluorescence intensities of the propidium iodide–stained cells were measured using a FACScalibur ( Becton-Dickinson ) , and data were analyzed with CellQuest ( Becton-Dickinson ) and WinMDI ( Joe Trotter; http://facs . scripps . edu ) software . To examine the percentage of cells in S phase , cells were treated with HU , released , and then analyzed at times after release using the BrdU Flow Kit ( BD Biosciences ) according to the manufacturer's instructions . BLM+ and SM-BLM cell clones in the logarithmic phase of cell proliferation had comparable proliferation rates , with cell-doubling times equal to ∼30 h for each clone examined . BLM+ and SM-BLM cells were seeded on coverslips and then treated with 10 mM or 0 . 5 mM HU in culture medium for 1 or 24 h , respectively . To achieve cell synchronization through an independent mechanism , BLM+ and SM-BLM cells were seeded onto coverslips and treated with 0 . 5 mM mimosine for 24 h . The cells were released into normal medium for 5 h to allow entry into S phase , then treated or not with 10 mM HU for 1 h . For indirect immunofluorescence , cells were washed and fixed at the end of HU treatment . They were then stained with anti-RAD51 and with anti–γ-H2AX or anti-PCNA antibodies , and counterstained with secondary antibodies labeled with Alexa Fluor ( Invitrogen ) . Fixation and staining was performed as described previously [34] . Coverslips were mounted with Prolong Gold antifade reagent containing 4′ , 6-diamidino-2-phenylindole ( DAPI ) ( Invitrogen ) . Images were captured on a spinning disk confocal microscope ( Carl Zeiss , LSM-510 ) , and data were collected using Slidebook 4 . 1 software . Z-stacks were captured using a 100× oil immersion objective , and the optical slice thickness was 0 . 2 µm . A focus was defined as a defined area of the nucleus greater than the minimum area of optical resolution ( >0 . 125 µm2 ) in at least one Z-stack in which the fluorescence intensity was greater than the background fluorescence intensity of the nucleoplasm . Colocalization was defined as an area of overlap between two foci of different fluorophores . The maximum number of foci that could be counted in these cells was 150 . For purposes of quantification of γ-H2AX foci in the γ-H2AX-bright cells , the γ-H2AX-bright cells were assigned 151 foci , which was one focus more than the maximum countable number . A typical immunofluorescence experiment consisted of assessment of 30–50 cells per condition . The data presented are from two to three independent experiments performed on two or three clones of each type . Immunofluorescence images for figures were created using Image J and Metamorph software ( Molecular Devices ) . Cells were untreated or treated with 0 . 5 mM HU for 24 h . Immediately after treatment , cells were harvested , washed twice with PBS , resuspended in fixation solution ( 2 . 3% paraformaldehyde , 0 . 6% methanol ) at a density of approximately 1×106 cells/ml , and incubated on ice for 20 min . Following the fixation process , cells were washed twice with PBS to remove the fixative and were resuspended in permeabilization solution ( 0 . 25% saponin , 10 mM HEPES [pH 7 . 4] , 140 mM NaCl , 2 . 5 mM CaCl2 ) at a concentration of approximately 5×106/ml . Cells were incubated overnight at 4°C in Alexa Fluor 647–conjugated anti-H2A . X ( BioLegend ) . After incubation , the cells were washed twice with 0 . 1% saponin and 5% fetal bovine serum in PBS and were resuspended in 5% fetal bovine serum in PBS for analysis . Cells were analyzed on a FACScaliber as described above . Data were collected from a minimum of two experiments on five BLM+ clones and six SM-BLM clones . Median fluorescence intensity data was log normalized , and the differences in median intensity were calculated . Log-normalized median intensity differences were tested by Student's t-test . Cells were untreated or treated with 0 . 5 mM HU for 24 h and subsequently released into fresh medium for an additional 0 , 12 , or 24 h , or they were treated with different concentrations of CPT for 3 h . For each damage condition , 4×105 cells were formed into individual 1% agarose plugs ( Cleancut agarose , Bio-Rad ) . The plugs were then incubated in 100 mM EDTA ( pH 8 . 0 ) , 0 . 2% sodium deoxycholate , 1% sodium lauryl sarcosine , and 1 mg/ml proteinase K at 50°C for 24 h . The plugs were washed four times in 10 mM Tris-HCl ( pH 8 . 0 ) and 1 mM EDTA for 30 min at room temperature with gentle agitation . Plugs were loaded onto a 0 . 8% agarose gel ( Pulsed Field Certified Agarose , Bio-Rad ) , and PFGE was preformed on a CHEF DR III ( 96° , 100° , 106° angle ramp , 1 , 200–1 , 800 s switch time , 2 V/cm; Bio-Rad ) for 72 h . The gel was stained with SYBR Gold , visualized under UV light , and analyzed using Quantity One and ImageJ software after contrast adjustment . Each lane on the gel was divided into seven areas , and the intensity in each area was analyzed and weighed according to fragment size . The amount of breakage in each lane was further normalized against total DNA content . Values are given relative to the level of DNA breakage in the untreated control . At least two independent experiments were performed on two clones of each type . For SCE analyses , cells were cultured with 10 µM BrdU ( Sigma-Aldrich ) . After 60 h , the cells were incubated with 0 . 02 µg/ml colcemid ( Invitrogen ) for up to 2 h , harvested and processed as described earlier [60] . The slides were examined under the microscope at 100× , and SCEs were counted from metaphases with an acceptable quality of sister-chromatid discrimination . For measurements of HU-induced SCEs , cells were cultured in 10 µM BrdU for 30 h , washed one time with 1× PBS , and treated with 0 . 5 mM HU for 24 h . Next , the cells were released into medium containing 10 µM BrdU for an additional 20 h . Metaphases were collected in colcemid and processed as described above . Two independent experiments were performed on two clones of each type . To quantify the cytogenetic effects of replication-associated DSBs , cells were treated with HU , and micronuclei formation was assessed using the cytokinesis-block micronucleus assay [61] , [62] . Cells were plated on chamber slides ( Lab-Tek ) for approximately 48 h . Next , cells were treated with 0 . 5 mM HU for 24 h , after which the cells were washed thoroughly with PBS and then incubated in culture medium containing 8 . 7 µM cytochalasin-B ( Sigma-Aldrich ) . After 28 h of cytochalasin-B treatment , cells were fixed on the slides with a 9∶1 methanol:acetic acid solution and then stained with Diff-Quik ( Baxter ) according to the manufacturer's instructions . Using a blinded analysis , we examined cells under a light microscope at 100× . A minimum of 500 binucleated cells were assessed under each condition and categorized as follows: cells with no micronucleus , one micronucleus , more than one micronucleus , and nucleoplasmic bridges . Three independent experiments were performed on three clones of each type . To measure cell viability under different DNA damage conditions , 2×105 cells were seeded onto six-well dishes . Cells were treated or not treated with 0 . 5 mM HU for 24 h and then treated or not treated with 50 µM of etoposide for an additional 24 h . At the end of the second treatment , the cell-growth medium was retained , and the floating cells were combined with adherent cells harvested by trypsinization . We added the ViaCount reagent ( Guava Technologies ) according to the manufacturer's instructions . Live and dead cells were counted with the Guava Cell Analyzer . A minimum of three independent experiments with three replicates for each condition were performed on three clones of each type . Human RAD51 protein was purified as described previously [63] . The purified hRAD51 was either cleaved from the streptavidin-conjugated agarose beads ( Ultralink , Pierce ) using tobacco etch virus protease or left on the beads and directly used for the binding reactions . SUMO was biotinylated by Sulfo-NHS-Biotin ( EZ-link , Pierce ) according to the manufacturer's instructions . Equal amounts of unbiotinylated and biotinylated SUMO-1 and SUMO-2 proteins were incubated with RAD51 in binding buffer ( PBS with 0 . 1% Triton X-100 ) for 2 h at 4°C . Streptavidin beads were then added to each reaction and incubated for another 1 h at 4°C . After five washes with binding buffer , proteins were eluted by SDS-PAGE buffer and analyzed by Western blotting with RAD51 antibodies . BLM N-terminal fragment ( 1–431 ) was modified by SUMO-2 in vitro as described [54] . The SUMO-modified BLM was then aliquoted equally into two tubes with either RAD51-coated streptavidin beads or biotin-coated beads , and rotated in the binding buffer in the presence of 2% BSA at 4°C for 2 h . Unmodified BLM was used as a control . The eluted protein was analyzed by SDS-PAGE and Western blotting with anti-BLM antibodies . Fujifilm Multi Gauge image analysis software ( Fujifilm Corp . ) was used to determine relative Western blot band intensities and ratios of SUMO-2–modified to unmodified BLM . Because observations within each clone may be correlated , we used mixed effects linear models to test the data for statistical significance . In the mixed effects models , each clone was treated as a random effect , and the experimental variables were treated as fixed effects . Because the foci and micronuclei data were not normally distributed , we first applied a square root transformation to stabilize the variance and normalize the data ( results in the figures were still presented in the original scale ) . For testing changes in the number of foci per cell and the number of colocalized foci per cell , cell type ( BLM or SM-BLM ) , treatment ( with and without HU ) , and interaction terms for cell type by treatment were treated as fixed effects . Similarly , for testing changes in the number of micronuclei per cell and the number of SCEs per 46 chromosomes , cell type , treatment with HU , and their interaction were treated as fixed effects . Finally , for testing changes in percentage of cell death , cell type , treatment with HU , treatment with etoposide , and their interaction terms were included as fixed effects . If the random effects ( i . e . , the clonal variation ) were found to be nonsignificant based on likelihood ratio test , the mixed effect models reduced to traditional analysis of variance .
|
Replication is the process in which cellular DNA is duplicated . DNA damage incurred during replication is detrimental to the cell . Homologous recombination , in which DNA sequences are exchanged between two similar or identical strands of DNA , plays a pivotal role in correcting replication processes that have failed due to DNA breakage and is tightly regulated , because deficient or excess recombination results in genomic instability . Previous studies have implicated the DNA-processing enzyme BLM in the regulation of homologous recombination; BLM is defective in Bloom's syndrome , which is characterized by excess recombination and cancer susceptibility . Here , we show that modification of BLM by the small protein SUMO controls BLM's function in regulating homologous recombination at sites where DNA replication failed . We showed that cells expressing a SUMO-deficient mutant of BLM accumulated more DNA damage and displayed defects in repair by homologous recombination . An enzyme involved in homologous recombination , RAD51 , displayed a defect in localization to sites where DNA replication failed . Our data support a model in which SUMO modification regulates BLM's function in homologous recombination by controlling the localization of RAD51 to failed replication sites .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry/replication",
"and",
"repair",
"genetics",
"and",
"genomics/cancer",
"genetics",
"cell",
"biology/cellular",
"death",
"and",
"stress",
"responses"
] |
2009
|
SUMO Modification Regulates BLM and RAD51 Interaction at Damaged Replication Forks
|
Reverse genetics in the mosquito Anopheles gambiae by RNAi mediated gene silencing has led in recent years to an advanced understanding of the mosquito immune response against infections with bacteria and malaria parasites . We developed RNAi screens in An . gambiae hemocyte-like cells using a library of double-stranded RNAs targeting 109 genes expressed highly or specifically in mosquito hemocytes to identify novel regulators of the hemocyte immune response . Assays included phagocytosis of bacterial bioparticles , expression of the antimicrobial peptide CEC1 , and basal and induced expression of the mosquito complement factor LRIM1 . A cell viability screen was also carried out to assess dsRNA cytotoxicity and to identify genes involved in cell growth and survival . Our results identify 22 novel immune regulators , including proteins putatively involved in phagosome assembly and maturation ( Ca2+ channel , v-ATPase and cyclin-dependent protein kinase ) , pattern recognition ( fibrinogen-domain lectins and Nimrod ) , immune modulation ( peptidase and serine protease homolog ) , immune signaling ( Eiger and LPS-induced factor ) , cell adhesion and communication ( Laminin B1 and Ninjurin ) and immune homeostasis ( Lipophorin receptor ) . The development of robust functional cell-based assays paves the way for genome-wide functional screens to study the mosquito immune response to infections with human pathogens .
Anopheles gambiae is a major vector of Plasmodium falciparum malaria in sub-Saharan Africa and a secondary vector of other parasitic and viral diseases [1] . Differences in vector susceptibility to malaria parasites are partly attributed to the ability of the mosquito immune system to fight infections . The developmental migration of Plasmodium within the mosquito hemolymph , the main carrier of the immune system , presents opportunities for the vector humoral and cellular immune reactions to attack the parasites [2] . Key functions of mosquito hemolymph components include killing of Plasmodium ookinetes as soon as they emerge from the midgut epithelium [3] and sporozoites before they invade the salivary glands [4] . Numerous mosquito agonist and antagonist effectors of Plasmodium and bacteria have been identified , principally by RNAi-mediated reverse genetic tests using dsRNA ( double-stranded RNA ) injections into adult mosquitoes [5] . These factors operate in complex molecular networks that involve pathogen recognition by secreted or membrane bound receptors , activation of immune signaling pathways , and synthesis or activation of effectors that contribute to lysis , melanization or phagocytosis of the invading pathogens [2] , [6] , [7] . Importantly , many of these factors are produced by hemocytes and function in the hemolymph [8] , [9] , [10] . Two hemocyte expression datasets have been reported recently , providing a comprehensive list of hemocyte-expressed genes [8] , [11] . An . gambiae cell lines have been used extensively to study mosquito immune responses [12] , [13] , [14] , [15] . Indeed , these cells are capable of accomplishing complex immune tasks that include phagocytosis of bacteria and beads [16] , as well as expression of immune factors upon microbial challenge . It has been shown that IMD pathway activation in cell lines leads to robust expression of the antimicrobial peptide ( AMP ) gene Cecropin 1 ( CEC1 ) [15] and other immune factors [13] . This pathway is activated when the transmembrane receptor PGRPLC binds peptidoglycan ( PGN ) and induces nuclear translocation of the REL2 transcription factor [15] , [17] , [18] . The development of high-throughput RNAi screens in cultured cells has been a major breakthrough in functional genomics of model organisms , in both basic and applied research [19] , [20] , [21] , [22] . Here we report the development and implementation of RNAi screens in An . gambiae cells to provide insights into the functional immune repertoire of mosquito circulating hemocytes . We have generated a dsRNA library targeting 109 genes specifically or predominantly expressed in circulating hemocytes and then optimized cell-based RNAi screens to investigate the role of these genes in phagocytosis of bacteria and transcriptional activation of immune-related genes . Our results identify novel regulators of the hemocyte immune responses and interactions with pathogens , including regulators of a complement-like pathway component that plays a key role in reactions to malaria parasites . This is a key milestone towards development of genome-wide RNAi screens in An . gambiae cells .
We generated a set of 111 dsRNAs to target 109 genes that exhibit enriched expression in hemocytes , differential regulation by immune challenges , and presence of immune-related InterPro domains and/or signal peptide or transmembrane domains ( Dataset S1 ) . To populate and annotate this dsRNA library we used: two published datasets of genome-wide transcriptional repertoires of An . gambiae circulating hemocytes from naive or Plasmodium infected adult females [8] , [11] , the published expression profile of An . gambiae hemocyte-like cell lines in response to microbial challenges [13] , the VectorBase An . gambiae genome annotation [23] and information about the silencing phenotypes of Drosophila orthologs found in the GenomeRNAi database [24] , [25] . Analysis of AMP expression and dsRNA-mediated RNAi efficiency ( Text S1 ) led us to choose the Sua5 . 1* cell line [12] , [16] , [26] as our model experimental system . We carried out viability screens to assess the levels of dsRNA toxicity as determined by the effect of gene silencing on fundamental housekeeping processes such as cell growth , proliferation and survival , which could hamper true identification of immune regulators . An . gambiae homologs of three genes previously shown to cause lethal or growth-defective RNAi phenotypes in Drosophila cells [22] , [27] were used as controls: inhibitor of apoptosis 1 ( IAP1; AGAP007294 ) , a ubiquitin-like/ribosomal fusion protein ( AGAP008001; 2 different dsRNA fragments ) and the Rho1 small GTPase ( AGAP005160 ) ( Table S2 and Text S1 ) . Two protocols were implemented to assess cell growth and viability: ( a ) image acquisition and quantification by automated fluorescence microscopy , which is time-consuming and technically challenging but allows for more accurate and informative analysis; and ( b ) microplate reader fluorescence quantification , which is quicker and can accommodate large datasets but is less user-responsive . A subset of 37 dsRNAs was screened by automated fluorescence microscopy . Staining cells with Sytox green/Hoechst was used to assess the effect of dsRNAs on cell viability . Protocols in Volocity Improvision and ImageJ softwares were developed to capture images and quantify fluorescent cells , to determine the ratio of dead cells ( Sytox green positive ) over the entire cell population ( Hoechst positive ) ( Figure 1A and Materials and Methods ) . ANOVA statistics followed by Bonferroni's post-test correction revealed that dsIAP1 leads to a significant increase of cell mortality . Silencing Rho1 significantly reduced the number of cells but did not increase cell mortality ( Figure S1 , Figure S2 and Text S1 ) . The cells and their nuclei were much larger in size indicating a defect in cytokinesis and cell cycle progression . A similar phenotype was observed in Drosophila S2 cells after silencing the orthologous Rho1 gene [22] , [27] . Next , we screened the entire collection of 111 dsRNAs using plate reader quantification in conjunction with a CellTiter-Blue Cell Viability assay . Data from three replicates were analyzed and z-score analysis was performed posing thresholds of +/−2 for at least two replicates out of three . Reproducibility among replicates was evaluated by correlation tests as shown in the plots in Figure S3 . Silencing the ubiquitin-like/ribosomal fusion protein gene AGAP008001 led to significant decrease in cell viability ( Figure 1B and Figure S3 ) . No other dsRNA treatment resulted in statistically significant deviation from the average cell count . A general observation was that dsRNA treatment resulted in significant cell mortality that was independent of the targeted gene . We validated the viability data in vivo by injecting dsRNAs of control LacZ , IAP1 and AGAP008001 into newly emerged adult female mosquitoes and then monitoring the survival of KD mosquitoes daily ( Figure S1 ) . Both gene KDs led to statistically significant increase of mortality rates compared to control . In addition to the support that this analysis provided to our experimental approach , the good correspondence between the ex vivo cell-based viability screen and the in vivo phenotypic analysis highlights opportunities for future use of such viability screens in identifying targets of novel mosquito insecticides . Phagocytosis is a highly effective and immediate response against microbial invaders [28] . Mosquito hemocytes can bind and phagocytose bacterial bioparticles and Sephadex beads , as well as malaria sporozoites [4] , [29] , [30] , [31] . We established a fast and reliable cell-based assay in An . gambiae cells using Escherichia coli bioparticles conjugated with pHrodo succinimidyl ester , a pH-sensitive fluorescent dye , to investigate the potential role of genes in the hemocyte-enriched library in bacterial phagocytosis . The phagocytic activity of cells was determined as the increase of bioparticle fluorescence caused by the drop of pH in the acidified phagosomes [32] . Bacterial bioparticles were added to the cells three days after incubation with dsRNA , and the capacity of cells to uptake bioparticles was assessed by fluorescence measurements using a microplate reader . Four time-points were assayed to take into account the kinetics of bioparticle uptake ( 1 , 3 , 6 and 24 h post-challenge ) . The measurements obtained were subtracted from basal level measurements ( 0 h , immediately after challenge ) . The entire dsRNA library was screened three times and z-score values for each of the dsRNA in each of the replicate screens were calculated . We considered positive dsRNA hits those with z-score values above 2 or below −2 in at least two out of the three replicates for each time point ( Figure S4 and Supplemental Table S3 ) . Because our library is strongly biased towards genes that are likely to play a role in immune reactions , the z-score method which compares the effect of each dsRNA with the average effect of all dsRNAs is a very strict condition . Therefore , we also analysed the data using ANOVA followed by Tukey's multiple comparison test , thus comparing each dsRNA with the reference dsLacZ control . The results from both methods revealed a total of 13 positive dsRNA hits , 6 from the z-score and 11 from ANOVA ( Table S3 and Figure 2A–B ) . Four dsRNAs showed significant effects on bioparticle uptake with both methods: 2 of them decreased phagocytosis ( FBN8 and AGAP000095 ) and 2 of them increased phagocytosis ( AGAP006769 and FBN9 ) . Cactus dsRNA that was used as a positive control also led to a significant increase in phagocytosis of E . coli as soon as 1 h after challenge , in consistence with previous observations [29] . We examined the 13 positive hits from the microplate reader analysis using automated fluorescence microscopy and a protocol for quantification of phagocytosed bioparticles developed in the ImageJ software . Bioparticle uptake was monitored 2 h and 20 h after challenge and compared to the dsLacZ control using ANOVA ( Figure 2C–D ) . An overall consistency was observed between the microplate reader and the microscopy analyses . Of the 13 dsRNAs , 6 showed the expected phenotype with statistical significance , 5 showed the expected phenotype but were not statistically significant , and 2 showed no difference with the dsLacZ control and/or a phenotype opposite to the expected , respectively ( Figure 2B ) . As mentioned earlier , imaging analysis can provide additional , more detailed , information when compared to the microplate reader method , but it is technically more challenging and time consuming . The few discrepancies between these two approaches may be due to both technical and biological reasons , for example image analysis cannot quantify the amount of bioparticles in a single cell , while the microplate reader quantifies the intensity of fluorescence . Moreover , as shown in Figure 2D , the distribution of fluorescent bioparticles is not uniform in the cell layer , and this introduces another variable when microscope images are captured and analyzed . Next , we investigated the silencing effect of 8 out of the 13 dsRNAs identified by the microplate reader method on bacterial phagocytosis in vivo . For this , we employed a protocol that was used in an earlier study , in which bacteria injected into the mosquito hemolymph spread rapidly into the cavity and are phagocytosed by hemocytes often found in clusters associated with the tracheal system [29] . We injected 2-day old mosquitoes with dsRNAs and 4 days later re-injected them with E . coli pHRodo-conjugated bioparticles . Mosquitoes were dissected 1 h after bioparticle injection , mounted onto glass slides and immediately observed by fluorescence microscopy . Pictures of different parts of the mosquito abdomens were captured and analyzed using a protocol developed in ImageJ to quantify the numbers of fluorescent particles ( Figure 3 ) . The results revealed strong in vivo phenotypes similar to those of the cell-based analysis for 4 out of the 8 dsRNAs: AGAP000182 , AGAP008492 , AGAP004928 and AGAP002243 , the last one corroborated by a significant statistical evaluation . Three dsRNAs ( AGAP003879 , FBN8 and AGAP006769 ) also showed similar albeit weaker phenotypes compared to the cell-based analysis and only one dsRNA ( AGAP009459 ) did not confirm the expected phenotype . Based on to the microplate reader analysis , the silencing of 7 genes led to a significant decrease of the cellular capacity to phagocytose E . coli bioparticles . Some of these genes were also confirmed with the microscopy and the in vivo analysis , as presented above . These genes encode: a protein of unknown function with a putative signal peptide and a peptidase domain ( AGAP000182 ) ; a protein with a homodimerization BTB/POZ domain , ankyrin repeats and a zinc finger domain ( AGAP002243 ) ; a putative transmembrane v-ATPase ( AGAP003879 ) ; a membrane-bound protein with a zinc finger and a LITAF ( LPS-induced tumor necrosis factor alpha factor ) domain putatively involved in immune signaling ( AGAP004928 ) [33] , [34]; the fibrinogen-domain FBN8 ( also known as FREP57 ) , previously shown to play a role in anti-Plasmodium defense ( AGAP011223 ) [35]; a putative Calcium channel protein ( AGAP000095 ) ; and a three-transmembrane protein of unpredicted function ( AGAP008500 ) . Two of these proteins are likely to play a role in phagosome formation and maturation/acidification . The putative Ca2+ channel protein ( AGAP000095 ) may be involved in the cellular Ca2+ balance that is required for solubilization of the actin meshwork surrounding nascent phagosomes , fusion of phagosomes with granules containing lytic enzymes , or assembly and activation of the superoxide-generating NADPH oxidase complex [36] . The v-ATPase ( AGAP003879 ) is known to play a role in phagosome acidification in other model organisms [37] . Orthologs of AGAP003879 and AGAP004928 in D . melanogaster show similar phenotypes in RNAi screens that investigate host-pathogen interactions ( Dataset S1 ) , as both KDs cause a decrease in intracellular Listeria monocytogenes infection [38] . In contrast , silencing 6 out of the 109 genes leads to a significant increase of E . coli phagocytosis . Proteins encoded by these genes represent potential novel negative regulators of bacterial recognition and phagosome assembly . These include: a secreted protein of unknown function that is strongly expressed in mosquito hemocytes [8] and cultured cells following LPS or PGN challenge ( AGAP006769 ) [13]; a putative tyrosine and serine/threonine kinase ( AGAP009459 ) with homologs described in other mosquitoes ( Aedes aegypti , AAEL008621 , cell division protein kinase 1 , cdk1; and Culex quinquefasciatus , CPIJ001155 , cdk2 ) and D . melanogaster ( CG5363 , cdc2 cell division control protein ) ; AGAP008492 that does not exhibit similarity to any other genes and is regulated during immune challenges [13]; the ortholog of Drosophila laminin B1 chain ( AGAP001381 ) ; and the fibrinogen-domain lectins FBN30/FREP8 and FBN9/FREP13 ( AGAP006914 and AGAP011197 , respectively ) [35] . The negative effect of AGAP009459 silencing in bacterial phagocytosis is possibly related to defects in cytoskeleton regulation . Its Drosophila ortholog , cdc2 , is similarly involved in defense-related processes as highlighted by increased Listeria intracellular infection , reduced Chlamydia infection and decreased Drosophila C virus and influenza virus replication following silencing [38] , [39] , [40] . Similarly , the fruit fly ortholog of Laminin B1 may also play a role in innate immune reactions since its silencing is shown to decrease viability after intestinal infections with Serratia marcescens [41] . FBN9 has been previously shown to be upregulated both by malaria parasite and E . coli infections [42] . The involvement of FBN9 in the defense against bacteria and maintenance of basal immune homeostasis is supported by evidence that the protein is found on the surface of non-challenged cells and strongly co-localizes with bacteria as well as malaria parasites following infection [35] . A specific role of FBN9 as a negative regulator of phagocytosis can be therefore hypothesized considering the fine interplay between different immune processes , where a pattern recognition receptor may specifically promote one in favor of another process . It has been previously shown that the AMP CEC1 is transcriptionally induced in cultured cells following immune challenge and that this induction depends partly on the IMD signaling pathway [13] , triggered after PGN recognition by the PGN Recognition Protein LC ( PGRP-LC ) [18] . To identify novel hemocyte regulators of the A . gambiae IMD pathway and potentially other pathways regulating the expression of AMPs , we developed a luciferase reporter assay in Sua5 . 1* cells to screen the hemocyte dsRNA library . A 660 bp fragment of the CEC1 promoter cloned upstream of the luciferase gene was used [15] . Initial experiments revealed a significant induction of CEC1 promoter activity 7 h after PGN challenge . A dsRNA targeting the NF-κB transcription factor REL2 , previously shown to regulate CEC1 transcriptional activation [15] , was included as a positive control . Changes in luciferace activity were analyzed by calculating the z-score values of ratios of average RLU ( Relative Light Units ) measurements of PGN vs . PBS treatments ( Figure 4 and Table S4 ) . One-way ANOVA followed by Dunnet's multiple-comparison post-test was also performed to compare dsLacZ control and KD values ( Table S4 ) . Silencing AGAP010531 ( FBN12 or FREP2 ) and AGAP006771 led to increased CEC1 expression following PGN challenge . AGAP006771 encodes a putative transmembrane protein orthologous to the Drosophila Tumor Necrosis Factor-like , eiger . Interestingly , Drosophila eiger is also induced during microbial infections and required for both resistance and tolerance to infections , partly by controlling the expression of the AMP Diptericin [33] , [43] . Eiger functions as a negative regulator of AMP expression following PGN challenge and IMD pathway activation by blocking the expression of the NF-κB factor , Relish ( the ortholog of REL2 ) , through the JNK pathway [44] . Our data is consistent with the fruit fly model and identify mosquito eiger as a negative regulator of CEC1 expression . Like FBN9 that is identified as a negative regulator of phagocytosis , FBN12 belongs to a family of putative pattern recognition receptors known to be involved in immune responses and maintenance of homeostasis; however , in contrast to FBN9 , FBN12 is downregulated during bacterial infections [17] , [35] . This is consistent with a role of FBN12 as a negative regulator of CEC1 expression . Silencing AGAP009231 and AGAP009459 reduced PGN-induced CEC1 transcriptional activation . AGAP009231 encodes a transmembrane domain protein of the family of ninjurins , most likely of the A sub-family , a complex class of cell adhesion molecules that are proteolytically processed and shed by matrix metalloproteases ( MMP ) . MMP1 and NinjA genes are co-expressed and upregulated in D . melanogaster S2 cells after LPS challenge [45] and in adult flies after wounding [46] . Proteolytic cleavage of NinjA by MMPs releases an ectodomain involved in cell adhesion and cell-cell communication [47] . MMPs are known to play various roles in inflammation and innate immunity but the identity and function of their substrates and mechanisms are still to be elucidated [48] . AGAP009231 has been previously shown to be highly expressed and localized on the membrane of An . gambiae circulating hemocytes [8] . Our data showing involvement of this protein in AMP expression suggest an important role in mosquito innate immunity , probably in signaling and cell communication . As discussed earlier , AGAP009459 encodes the ortholog of Drosophila cyclin-dependent protein kinase cdc2 that has been implicated in several processes from cell cycle regulation to cytoskeleton remodeling . Importantly , RNAi silencing of cdc2 also leads to decreased STAT92E phosphorylation , suggesting a regulatory role of cdc2 in JAK/STAT signaling [49] . The involvement of JAK/STAT pathway in fruit fly hemocyte differentiation and proliferation is well documented , but its exact role in immune responses such as AMP expression remains unclear [50] . In mosquitoes , the JAK/STAT pathway is activated by immune challenges and is involved in responses against pathogens [51] , [52] , [53] , and our data suggest involvement of the JAK/STAT pathway in CEC1 activation . The function of AGAP009459 in phagocytosis of E . coli bioparticles may be related to the involvement of cdc2 in cytoskeleton regulation , but could also imply a novel role of JAK/STAT in phagocytosis . We investigated the expression of the LRIM1 gene using an approach identical to that described above for CEC1 . LRIM1 is expressed in An . gambiae hemocytes [8] and secreted in the hemolymph in a disulfide-linked complex with the structurally related protein APL1C; there the complex binds and solubilizes TEP1cut , a cleaved , activated form of the complement C3-like factor TEP1 [54] , [55] . This new complex plays a key role in mosquito responses against invading malaria parasites . A previous study showed that LRIM1 is transcriptionally induced following PGN challenge [13] . We used a 1600 bp fragment of the LRIM1 promoter fused to luciferase ( courtesy of M . J . Povelones ) . Our preliminary data showed high luciferase activity in Sua5 . 1* cells but no further upregulation following PGN challenge . We investigated whether the lack of LRIM1 promoter upregulation in Sua5 . 1* cells upon PGN challenge was due to inhibition by other hemocyte factors ( Figure 4 and Table S4 ) . Our screen identified four genes that inhibit transcriptional activation of the LRIM1 promoter following PGN challenge , as silencing AGAP009762 , AGAP007499 , AGAP003473 and AGAP001964 led to increased luciferase expression compared to PBS control . AGAP009762 is highly expressed in circulating hemocytes [8] and encodes a EGF-like domain protein with similarities to Drosophila phagocytosis receptors eater [56] and NimC1 [57] as well as to other members of the Nimrod superfamily [58] . Interestingly , a screen for novel regulators of JNK following IMD pathway activation upon challenge with PGN in Drosophila cells , revealed that the ortholog of AGAP009762 caused an increased P-JNK protein expression , which , in turn , may act to modulate the expression of Relish-controlled effectors [59] . AGAP007499 encodes a chloride channel protein , orthologous to the human chloride channel 7 . It is strongly expressed in circulating hemocytes [8] and upregulated in mosquito cultured cells after hydrogen peroxide treatment [13] . The Drosophila ortholog of AGAP007499 is shown to play a role in the receptor tyrosine kinase ( RTK ) -Ras-extracellular signal-regulated kinase ( MAPK/ERK ) signaling pathway; its silencing leads to increased MAPK phosphorylation following EGF stimulation [60] . It has been previously shown that MAPK ERK signaling plays a role in the mosquito immune response against malaria parasites [61] . AGAP003473 encodes a transmembrane protein with no significant similarity to known proteins . Finally , AGAP001964 encodes a previously uncharacterized member of the clip-domain serine protease subfamily A ( CLIPA ) that lacks protease activity . Importantly , several CLIPAs show Plasmodium infection phenotypes [62] , [63] , mostly by regulating the hemocyte-mediated melanization reaction . Since LRIM1 is involved in malaria parasite melanization and lysis , as well as in bacterial phagocytosis , we hypothesize that the identified proteins function as negative regulators of these reactions some of which ( e . g . melanization ) are potentially costly to the host; thus LRIM1 is induced only when these proteins are downregulated or presumably depleted during these reactions . Silencing AGAP001381 had an opposite effect , reducing luciferase expression driven by the LRIM1 promoter 7 h after challenge with PGN compared to mock PBS challenge . As mentioned previously , AGAP001381 encodes the ortholog of the fruit fly Laminin B1 and its silencing also increases phagocytosis of E . coli bioparticles . These data conform to our hypothesis that a network of negative and positive regulators is involved in induction of LRIM1 expression that follows infection . Intriguingly , silencing of Laminin B1 also resulted in a contrasting increase of the basal LRIM1 promoter activity ( Figure 4C ) , suggesting a dual role of laminin B1 in activating and suppressing LRIM1 expression in the presence and absence of immune challenge , respectively . Two additional genes were identified as negative regulators of LRIM1 basal expression: silencing of AGAP009119 and AGAP002186 led to a significant increase of luciferase expression driven by the LRIM1 promoter ( Figure 4C ) . AGAP009119 encodes a protein with tetratrico peptide structural repeats , involved in protein-protein interactions , and a heat shock chaperonin-binding motif , which is orthologous to the D . melanogaster Hsp70-interacting protein 2 ( HIP-replacement ) . AGAP002186 encodes a receptor of lipophorin ( Lp ) , that is the insect equivalent of low-density lipoproteins and co-ortholog of the Drosophila LpR1 and LpR2 [64] . In An . gambiae , Apolipoprotein I and II , the main components of Lp , have been shown to act antagonistically to TEP1-dependent malaria parasite killing [65] , [66] , [67] . Interestingly , LpR2 ( Lipophorin Receptor 2 ) , the fruit fly ortholog of AGAP002186 , has been recently shown to suppress formation of melanotic tumors [68] . Reduction of the basal LRIM1 promoter activity in cultured cells was detected only after silencing REL2 compared to dsLacZ-treated control cells , consistent with an earlier study showing that REL2 , as well as REL1 , control basal LRIM1 and TEP1 expression [69] . Our results identify 22 novel regulators of the hemocyte immune response in this major African vector of human malaria . As summarized in Figure 5 and in Table S5 , a complex network of positive and negative regulators of immediate ( phagocytosis and basal complement state ) as well as induced ( AMP expression and induced complement state ) responses are revealed . Four membrane-bound or transmembrane proteins are implicated in E . coli phagocytosis , but none of them have domains that could indicate bona fide phagocytic receptors . Two of them , a putative Ca2+ channel protein ( AGAP000095 ) and a v-ATPase ( AGAP003879 ) have predicted functions that point to their involvement in phagosome assembly and maturation . Of the remaining two membrane-bound or transmembrane proteins , AGAP004928 has an LPS-induced tumor necrosis factor alpha factor domain and could be involved in immune signaling [33] , [34] . A very intriguing connection is revealed between phagocytosis and regulation of the basal and induced expression of LRIM1 , a key component of the mosquito complement cascade [54] , [55] and a known facilitator of E . coli phagocytosis [29] . The basement membrane protein Laminin B1 inhibits both phagocytosis and basal expression of LRIM1 , but promotes LRIM1 expression after immune challenge . Therefore , Laminin B1 appears to play a dual role in maintaining the basal levels of complement in the hemolymph and in promoting production of complement components when needed or in preparation of potential reinfections . Whether this function of Laminin B1 is based on cell signaling and hemocyte differentiation , it remains to be investigated . Indeed , it has been suggested that immune priming of the mosquito hemolymph by gut bacteria following midgut invasion by malaria parasites causes hemocyte differentiation and attachment to the midgut basement membrane ( basal lamina ) , which induces transient overexpression of LRIM1 and TEP1 [9] . Indeed complement is a very important defense reaction of the mosquito hemolymph [16] , [29] , [70] . Upon TEP1 maturation cleavage , TEP1cut binds to the LRIM1/APL1C complex , where it remains in an active soluble state until an infection occurs [54] , [55] . However , the last steps of TEP1 binding on the pathogen surface remain unknown . Rono and co-workers [67] have recently suggested that the observed antagonistic effect of Lp ( and Vitellogenin ) on TEP1 binding on malaria parasites may be due to Lp masking or competing for TEP1 binding sites on the parasite surface , directly interacting with TEP1 , or modifying the lipid composition of the parasite membrane . Our finding that the Lp receptor suppresses the basal expression of LRIM1 provides additional insights into the mechanism underlying the effect of Lp on TEP1-mediated parasite killing . Further investigation of the roles of the Lp receptor , Laminin B1 and the additional regulators of LRIM1 expression identified in this screen will shed light into the regulatory mechanisms of mosquito complement and how these impact upon infections with malaria parasites .
An . gambiae G3 strain was maintained according to standard insectary procedures ( www . mr4 . org/ ) . DsRNA injections in 2-days-old mosquitoes were performed as described previously [71] . The An . gambiae cell lines 4a2 , 4a3A , 4a3B , L3-5 , SuaE . 1 , SuaB . 1 , Sua4 . 0 and Sua5 . 1* were maintained as described ( www . mr4 . org/ and [12] ) . Briefly , cells were grown in Schneider's medium supplemented with 10% fetal bovine serum ( heat inactivated ) , 100 U/ml Penicillin and 100 µg/ml Streptomycin at 27°C . Splitting was carried out by shaking flasks to detach cells and freezing/thawing procedures were performed according to standard cell culture protocols . RNAi-mediated gene silencing of cells was carried out in 96-well plates: approximately 105 cultured cells were bathed in 1 µg dsRNA per well dissolved in serum-free Schneider's Medium and 2 h later complete medium was added to obtain a final serum concentration of 10% . Total RNA from wild type , KD females and cultured cells was extracted using Trizol Reagent ( Invitrogen ) . After DNAseI ( Invitrogen ) treatment , first strand cDNA was synthesized using oligo-d ( T ) primers ( Invitrogen ) and Superscript Reverse Transcriptase II ( Invitrogen ) according to the manufacturer's instructions . For dsRNA synthesis , T7-tailed primers ( see Table S1 ) were designed using the E-RNAi web-service at http://www . dkfz . de/signaling/e-rnai3/evaluation . php . PCR products were synthesized using cDNA from female mosquitoes as a template and purified using the QIAquick PCR Purification kit ( QIAGEN ) . DsRNA synthesis was performed according to MEGAscript T7 Kit ( Ambion ) manufacturer's protocol and purification of dsRNA was performed by phenol/chloroform extraction . The quality and quantity of dsRNA were checked by agarose gel electrophoresis and Nanodrop reading , respectively . Quantitative RT-PCR was performed using the SYBR Green PCR mastermix and analyzed using the ABI PRISM 7700 sequence detection system and the manufacturer's instructions . Expression levels were calculated by the relative standard curve method using S7 as endogenous control [72] , [73] . Primers used are listed in Table S1 . Cells were seeded in a 96-well plate at a concentration of 105 cells/well . The next day the medium was removed and cells were treated with 1 µg dsRNA dissolved in 50 µl serum-free Schneider's Medium; 2 h later , complete medium was added to obtain a final serum concentration of 10% in a volume of 100 µl . Four days after dsRNA treatment , cells were stained with 1 µg/ml Hoechst 33342 ( Invitrogen ) , and 500 nM Sytox Green ( Invitrogen ) , by adding 25 µl of both chemicals to obtain a final volume of 150 µl/well . After 30 min incubation in the dark at 28°C , plates were analyzed by fluorescence microscopy . Images of cells were captured using a Zeiss Axiovert 200 widefield fluorescence microscope ( 10× objective ) and the Improvision Volocity Software . Three images per well were taken ( bright-field , DAPI and GFP ) . Images were analyzed using a protocol developed in ImageJ . Briefly , images were grouped in stacks and uniformly handled; stacks were transformed into binary images , and cells labeled with Hoechst and with Sytox Green were separately counted using Find Maxima Process Tool . Numbers obtained were statistically analyzed to quantify cell viability . ANOVA statistical analysis followed by Bonferroni's post-test were applied to groups of four to six pictures per dsRNA treatment . To assess cell size , stacks were further processed to subtract background and enhance contrast and then transformed into binary images; cells were converted into particles , and their number , size and shape were calculated . Viability assay was also performed using Cell Titer Blue kit ( Promega ) . Four days after dsRNA treatment carried out as above , 20 µl/well of CellTiter Blue solution were added to obtain a final volume of 120 µl/well . Plates were incubated for 2 h and then fluorescence produced by reduction of substrate resazurin into resorufin was measured with BMG Labtech FLUOstar OMEGA plate reader . Three replicates were performed and z-score analysis was applied to identify positive phenotypes in each replicate . Thresholds of +/−2 were applied and positive candidates were considered those passing the threshold in at least two out of three replicates . Sua5 . 1* cells were seeded in 96-well plate at a concentration of 105 cells/well . The next day the medium was removed and cells were treated with 1 µg dsRNA dissolved in 40 µl serum-free Schneider's Medium; 2 h later , complete medium was added to obtain a final serum concentration of 10% in a volume of 50 µl . PHRodo E . coli bioparticles ( Invitrogen ) were dissolved in sterile 1xPBS , sonicated according to manufacturer's instructions and 50 µl were added in each well 3 days after gene silencing to obtain a final volume of 100 µl/well . Cytochalasin D was added 20 min before bioparticle challenge at a concentration of 10 µM or 100 µM . BMG Labtech FLUOstar OMEGA plate reader was used to measure fluorescence intensity ( Ex 532 nm/Em 595 nm filter ) immediately after pHRodo bioparticles challenge , and at several time-points up to 24 h , keeping the temperature at 27°C during the entire procedure . Three replicates were performed . Plate reader outcomes were used to calculate z-score values for each replicate . Hits were considered as those dsRNAs producing z-score values above 2 or below −2 in at least two out of three replicates . Numerical values were also pooled and statistically analyzed using ANOVA followed by Tukey's Multiple Comparison Test to assess each dsRNA average value in relation to dsLacZ . Automated fluorescence microscopy was also employed as a separate method to assess fluorescent pHRodo bioparticles uptake at 2 h and 20 h post challenge for those dsRNAs showing a phenotype according to the microplate reader assay . Images were captured using a Zeiss Axiovert 200 fluorescence widefield microscopy and the Volocity Improvision software ( 10× objective ) . Image analysis and quantification were performed using a protocol developed in ImageJ . Briefly , images were grouped in stacks and uniformly handled; background was subtracted , contrast enhanced and stacks were transformed into binary images to separate the fluorescent particles from the cell layer and quantify the spot number and size . Numbers obtained from at least two experiments were statistically analyzed using GraphPad Prism software . Numbers were converted to percent values compared to dsLacZ . After normality test , values of each dsRNA were compared to dsLacZ using student's t-test . In vivo phagocytosis assays were performed on 2-day old mosquitoes injected with dsRNA as described above . Four days later , mosquitoes were injected again in the thorax with 69 nl of pHRodo E . coli bioparticles and allowed to recover and resume phagocytosis for 1 h . Mosquitoes were partially dissected ( wings and legs removed ) and gently compressed between a slide and a coverslip ( using clay to hold them together ) for imaging . Images of mosquito abdomens were captured using a Zeiss Axiovert 200 widefield fluorescence microscope and the Volocity Improvision software . At least two replicates for each dsRNA treatment were carried out , and at least 8 mosquitoes per KD were analyzed by microscopy with no less than four pictures captured per mosquito . Fluorescence images of sections of mosquito abdomens were captured as described above and shown in Figure 3 ( 10× objective ) . Quantification and statistical analysis were carried out as described above for pHRodo bioparticle phagocytosis in cultured cells . Sua5 . 1* cells were seeded in a 96-well plate at a concentration of 105 cells per well . Upon reaching 80% confluence , the medium was removed and cells were treated with 1 µg dsRNA dissolved in 40 µl serum-free Schneider's Medium . Two h later , complete medium was added to obtain a final serum concentration of 10% in a volume of 60 µl . Four days after dsRNA treatment , cells were washed and co-transfected in 50 µl of final volume with CEC1 or LRIM1 promoter firefly luciferace fusion reporter constructs ( Promega pGL3 backbone ) and reference pAct5C-Renilla luciferase construct using Effectene Transfection Reagent ( QIAGEN ) following the manufacturer's instructions . One day after transfection , 25 µl of PGN ( Sigma ) were added to obtain a final concentration of 100 µg/ml . Seven h later , cells were subjected to Dual Luciferase assay ( Promega ) according to the manufacturer's instructions in a final volume of 150 µl per well . Relative light units per second for both firefly and renilla luciferase were measured using the Luminometer mode of BMG Labtech FLUOstar OMEGA plate reader . RLU measurements were obtained dividing the firefly by renilla measurements . RLUs from three replicates were averaged and z-scores calculated after PGN and PBS treatments . One-way ANOVA followed by Dunnet's multiple-comparison post-test was also performed to compare dsLacZ control and dsRNA RLUs . Z-score was calculated using the formula zkj = ( ykj−M ) /S , where ykj is the background subtracted value for the kth well in the jth replicate , and M and S are mean and standard deviation ( SD ) of the distribution of y values , respectively [74] . We considered positive hits those dsRNAs exhibiting z-scores>2 or <−2 , which corresponds to SDs above or below the population mean in a given replicate . Considering that the critical z-score values when using a 95% confidence level are −1 . 96 and +1 . 96 SDs , positive hits have corresponding p values<0 . 05 . To assess z-score correlation between replicates , D'Agostino & Pearson omnibus normality test was applied to examine whether the data follow a Gaussian distribution . Correlations were computed using Spearman r correlation test for not normally distributed data and Pearson r correlation test for normally distributed data . GraphPad Prism 5 software was used for statistical analyses and graph design .
|
The mosquito immune system relies on innate humoral and cellular reactions to fight infections , including those by malaria parasites that must pass through mosquitoes before they can infect humans . Therefore , a detailed molecular understanding of these reactions could assist the design of new ways to control the spread of malaria and other mosquito-borne diseases . Here we use a technique to silence in mosquito cultured cells genes that are highly and/or specifically expressed in mosquito hemocytes , the equivalent of human white blood cells , as a means to investigate their function in reactions of the mosquito immune system . Our study identifies several novel regulators of immune reactions including phagocytosis , the engulfment and subsequent destruction of bacteria and other pathogens by hemocytes , the production of antimicrobial peptides , which directly kill or inhibit the proliferation of microbes , and the basal and induced production of an important complement regulator . Complement is a robust reaction of mosquitoes against malaria parasites and bacteria through phagocytosis , lysis or melanization ( the enclosure of pathogens in a melanin capsule ) . We also reveal intriguing molecular connections between these reactions such as phagocytosis and regulation of complement . Our study provides novel insights into mosquito immune system and its reactions against infections .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"death",
"complement",
"system",
"immune",
"activation",
"immunology",
"microbiology",
"host-pathogen",
"interaction",
"parasitology",
"gene",
"function",
"anopheles",
"immune",
"defense",
"gene",
"expression",
"biology",
"immune",
"system",
"cell",
"biology",
"immunity",
"vector",
"biology",
"innate",
"immunity",
"genetics",
"molecular",
"cell",
"biology",
"gene",
"networks",
"genetics",
"and",
"genomics"
] |
2013
|
Comprehensive Genetic Dissection of the Hemocyte Immune Response in the Malaria Mosquito Anopheles gambiae
|
The dissacharide trehalose is an important intracellular osmoprotectant and the OtsA/B pathway is the principal pathway for trehalose biosynthesis in a wide range of bacterial species . Scaffolding proteins and other cytoskeletal elements play an essential role in morphogenetic processes in bacteria . Here we describe how OtsA , in addition to its role in trehalose biosynthesis , functions as an osmotic stress sensor to regulate cell morphology in Arthrobacter strain A3 . In response to osmotic stress , this and other Arthrobacter species undergo a transition from bacillary to myceloid growth . An otsA null mutant exhibits constitutive myceloid growth . Osmotic stress leads to a depletion of trehalose-6-phosphate , the product of the OtsA enzyme , and experimental depletion of this metabolite also leads to constitutive myceloid growth independent of OtsA function . In vitro analyses indicate that OtsA can self-assemble into protein networks , promoted by trehalose-6-phosphate , a property that is not shared by the equivalent enzyme from E . coli , despite the latter’s enzymatic activity when expressed in Arthrobacter . This , and the localization of the protein in non-stressed cells at the mid-cell and poles , indicates that OtsA from Arthrobacter likely functions as a cytoskeletal element regulating cell morphology . Recruiting a biosynthetic enzyme for this morphogenetic function represents an intriguing adaptation in bacteria that can survive in extreme environments .
Trehalose ( a-D-glucopyranosyl ( 1 , 1 ) -a-D-glucopyranoside ) is a non-reducing disaccharide that functions as an important intracellular protectant against a variety of stress conditions including desiccation , dehydration , heat , cold , and oxidation [1] . At least four different pathways for trehalose biosynthesis have been reported , described as OtsA/B , TreY/Z , TreS , and TreT [2] . The OtsA/B pathway is the principal pathway for trehalose biosynthesis and is widely distributed in bacteria , fungi and plants ( trehalose-6-phosphate synthase and trehalose-6-phosphate phosphatase in Arabidopsis thaliana ) . OtsA utilises UDP-glucose and glucose-6-phosphate to synthesize trehalose-6-phosphate ( T6P ) and subsequently OtsB converts T6P into Pi and trehalose . As a signaling molecule , T6P is important as an ‘energy checkpoint’ during development in eukaryotes . For example , in Saccharomyces cerevisiae , T6P controls the decision to proceed through cell division [3] and in plants it regulates flowering both in the leaf and in the shoot apical meristem [4] . Osmotic stress can inhibit the growth rate and affect the morphology of bacteria belonging to several genera , including Arthrobacter species , Rhodococcus species [5] and Aeromonas hydrophila [6] . However , the relationship between osmotic stress , trehalose biosynthesis and the regulation of cell division and morphogenesis is unclear . Cell division and dynamic reorganisation of cell morphology depends on the internal cytoskeleton or scaffolding elements . In most bacteria , the tubulin homolog FtsZ is critical for driving binary fission [7] . The FtsZ protein assembles into protofilaments that are bundled together to form the Z-ring at the site of cell division , usually at the mid-cell . Other components of the division machinery are then recruited to form a multi-protein divisome complex responsible for mid-cell peptidoglycan synthesis . Contraction of the Z-ring also drives constriction of the cell envelope to form the septum [8] . Other cytoskeletal elements that have a role in determining cell morphology in typical rod-shaped eubacteria include MreB , an actin-like ATPase cytoskeletal proteins , that in vitro can polymerize into filaments in the presence of ATP or GTP [9 , 10] and guide lateral wall peptidoglycan synthesis [11 , 12] . The actinobacteria include species of contrasting morphologies , including coccoid-shaped Rhodococci , rod-shaped Mycobacteria and Corynebacteria , filamentous spore-forming Streptomyces and pleomorphic Arthrobacter . Actinobacteria studied thus far grow by apical extension , with new peptidoglycan synthesized and added at the cell poles . This contrasts with other eubacteria that insert new peptidogylcan in their lateral walls , with the poles being inert [13–15] . Apical growth in actinobacteria is independent of MreB . In fact , the genomes of actinobacteria that adopt bacillary-type growth lack mreB homologs , and filamentous Streptomyces use an MreB protein only during sporulation [16–18] . Apical growth is guided by the protein DivIVA [13–15]; DivIVA assembles to form an internal cytoskeletal element at the cell poles that appears to function to recruit proteins for apical cell-wall synthesis . Actinobacterial Arthrobacter species typically inhabit soil ecosytems and are fascinating for their pleomorphism . During exponential growth , rod-shape cells elongate and undergo cell division at the midcell region; the two daughter cells remain joined forming a V shape and subsequently separate by snapping apart [19] . A reversible transition from rod-shaped cells to non-separating multi-cellular , branching myceloids is induced in some species by osmotic stress and this is documented to be an adaptive response to promote bacterial survival through altered metabolism and increased resistance to environmental stress [20 , 21] . Indeed , Arthrobacter typically exhibit high resistance to , among other stresses , cold , heat and dessication [22 , 23] . Stress resistance is likely related to their pleomorphic behaviour , making them an interesting model for analysis of environmentally triggered developmental switches although , to date , there has been a paucity of molecular characterization of these bacteria . Arthrobacter sp . strain A3 ( hereafter referred to as Arthrobacter A3 ) , a psychrotrophic bacterium , was isolated from the alpine permafrost of the Tianshan Mountains in China [24] . It has an optimal growth temperature of 20 0C , but can survive and grow at near-freezing temperatures as low as -4 0C . Its stress tolerance is in part due to synthesis of trehalose catalyzed by OtsA/B [25] . As in Escherichia coli [26] , the otsAB genes in Arthrobacter A3 are arranged as an operon [25] . This organisation allows for efficient co-regulation of both genes [27] . OtsA of E . coli contains an N-terminal loop , located between Arg9 to Gly22 , based on the crystal structure . This N-loop is located in the catalytic centre of the OtsA enzyme and interacts with the phosphate moiety of glucose-6-phosphate and the distal phosphate of UDP-glucose , respectively [28] . Furthermore , both ends of the amino acid sequence of the N-loop are conserved in many microorganisms . During the catalytic reaction , the N-loop undergoes significant conformational changes [29] , suggesting that the N-loop is directly related to the catalytic efficiency of OtsA . The enzymatic activity of OtsA of Arthrobacter A3 at low temperatures is due to a very flexible N-loop containing the active site [25] , a key feature that distinguishes the protein from its E . coli counterpart . Here we demonstrate that depletion of OtsA or T6P results in constitutive myceloid growth . Further analyses indicate that OtsA doubles as a novel self-assembling morphogenetic protein . OtsA , acting as an osmotic stress sensor together with T6P , mediates the switch to myceloid growth during osmotic stress . Recruiting a biosynthetic enzyme for this morphogenetic function represents an intriguing adaptation in bacteria that can survive in extreme environments .
An otsA deletion mutant ( Ar0002 ) has significantly reduced intracellular trehalose levels compared to the wild-type strain , Ar0001 ( S1A Fig ) [24] . We also observed that the mutant exhibits an apparent markedly slower growth rate in low osmolarity medium as determined by optical density ( OD600; Fig 1A ) . Whereas the doubling time for the wild-type was 2 . 5 h , for the mutant it was 3 . 3 h , approximating the 3 . 2 h doubling time of the wild-type grown in salt-amended Luria broth ( LB ) . When early log-phase cells were examined by phase-contrast microscopy , we observed extensive aggregate formation by the mutant , similar to previously reported myceloids formed by other Arthrobacter species after osmotic stress [30] . Hence the mutant fails to grow by snap division , but instead adopts non-separating myceloid growth , characteristic of the morphological switch of the wild-type when subjected to salt stress . Consequently , OD600 measurements do not necessarily reflect slower growth of the mutant or wild-type subject to salt stress , but simply the growth of cell aggregates . We subsequently employed OD600 measurements as a proxy for measuring the formation of myceloids , verifying the presence of cell aggregates in early log-phase cultures by phase-contrast microscopy . Quantification and analysis of aggregate dimensions of early log-phase cells of the otsA mutant grown in chemically-defined minimal medium revealed 58% of colony-forming units existing as myceloids with a maximum dimension between extremities of the aggregates of >4 μm ( as viewed in two dimensions under the microscope ) , with the average value being 4 . 34 μm ( n = 385; Fig 1B ) . In contrast , the proportion of wild-type cell aggregates of greater than 4 μm maximum dimension formed was 21% of the total , with 79% of colony forming units being single cells or small multiples of 2 to 4 cells still joined prior to snap division ( Fig 1B ) . The aggregates of the mutant resembled those formed after 16 h growth by osmotically stressed cultures of the wild-type which have an average maximum dimension of 6 . 4 μm ( n = 200 ) , with over 80% of aggregates being > 4 μm in maximum dimension ( Fig 1B ) . Similar proportions of cell aggregates were observed during growth in LB medium with or without addition of salt ( S2 Fig ) . Cells of the wild-type , salt-induced wild-type myceloids and constitutive myceloids of the mutant were examined by scanning electron microscopy ( Fig 1C–1E ) . Although many single or small chains of non-separated cells of the wild-type grown in LB appeared normal ( Fig 1C , bottom left ) , we also detected a small proportion of cells in chains with branches emerging from their lateral walls ( e . g . Fig 1C , top panel , arrowed ) . The much larger constitutive or salt-induced myceloids consisted of networks of extensively branched chains of cells ( Fig 1D and 1E ) . Genetic complementation of the mutant with a single copy of otsA under control of its native promoter sequence restored both the normal growth rate and the largely bacillary morphology of log-phase bacterial cells grown in LB ( S3 Fig ) , whereas complementation with the flanking genes , otsB or dsbA , or E . coli otsA ( otsAEc ) did not restore bacillary growth ( S3 Fig ) , although the latter gene was biochemically functional ( see below ) . To examine a possible link between intracellular trehalose and growth morphology , cultures of the ΔotsA mutant were subjected to a biochemical complementation test . Addition of between 0 . 5 mM and 4 . 0 mM trehalose , which is effectively taken up by the bacteria ( [24] , S1A Fig ) had no effect on growth rate ( Fig 1A ) or myceloid formation . In addition , when a trehalase enzyme encoded by treF was over-expressed in the wild-type ( strain Ar0008 ) , resulting in a more than 5-fold reduction in intracellular trehalose to less than that detected in the ΔotsA mutant ( S1 Fig ) , there was only a modest reduction in growth rate ( Fig 1A ) . The growth rate was reflected in a low proportion ( 24% ) of cell aggregates with maximum dimension > 4 μm ( Fig 1B ) . We also determined that trehalose biosynthesis in the ΔotsA mutant was restored due to complementation by otsAEc ( S1 Fig ) . Consequently , we concluded that the constitutive myceloid formation of the ΔotsA mutant did not reflect a reduction of intracellular trehalose . We also observed that the ΔotsA mutant is osmotic stress- sensitive and whereas this phenotype could be rescued by genetic complementation , it could not be by biochemical complementation with trehalose ( S4 Fig ) . We constructed a strain , wild-type + up-otsA , containing the gene fused with a strong promoter on a multi-copy plasmid . Overexpression was verified by western blot , indicating an approximate 10-fold greater intracellular abundance of the protein relative to FtsZ ( Fig 2A ) . We used a fluorescent derivative of vancomycin ( fluo-vancomycin ) that binds to nascent peptidoglycan to establish firstly if , as in other studied actinobacteria , growth is at the cell poles , and secondly to visualise how overexpression of OtsA affects cell wall biosynthesis . The antibiotic bound to nascent peptidoglycan at the poles , confirming apical growth , and , with more intense fluorescence , at the newly forming septum in the midcell region of rod-shaped wild-type cells ( strain Ar0003 ) growing in LB ( Fig 2B ) . In cells with evidence of septum formation , the distance between the stained poles and midcell was on average 0 . 7 μm . There was evidence for some ‘mini-chains’ of cells , for example the 4 joined cells in the top panel of Fig 2B , due to inefficient snap division . The result of OtsA over-expression ( wild-type + up-otsA; strain Ar0004 ) was a pattern of peptidoglycan synthesis consistent with the formation of multiple septa in very long , enlarged cells with bulbous poles and limited branching ( Fig 2C ) . Staining these cells with both DAPI and fluo-vancomycin revealed that many of the newly-formed compartments possessed less intensely staining nucleoids , with a small proportion lacking detectable DNA ( arrowed in the overlay image , Fig 2C ) . This indicates that overexpression of OtsA affects the coordination of DNA synthesis and chromosome segregation with septum formation . The doubling time of this OtsA overexpression strain was 3 . 1 h ( Fig 1A ) , indicating that these observed abnormalities in cell division can retard growth . These results , indicating that OtsA function has a role on growth and division , prompted us to test the effects of overexpression of OtsAEc . Over-expression of C-terminal His-tagged OtsAEc , verified by western blotting , had no effect on septum formation or the rod-shaped morphology of the strain ( wild-type + up-otsAEc; Ar0010 ) grown in LB ( Fig 2D and 2E ) . We tested the activity of the overexpressed OtsAEc in Arthrobacter A3 by analysis of the trehalose concentration in Ar0010; this strain had almost 5 times greater intracellular trehalose than the wild type strain ( S1C Fig ) . Moreover , whereas the strain overexpressing OtsA was sensitive to osmotic stress , the strain overexpressing OtsAEc was not ( S4 Fig ) . Consequently , we inferred that specific features of the Arthrobacter protein confer its function as a morphogenetic determinant . To investigate the relationship between OtsA enzyme activity and its role as a morphogenetic protein and effector of the osmotic stress response , an amino acid substitution was introduced in the active site , as determined by crystallography of the corresponding E . coli enzyme [29] , replacing the conserved arginine residue ( R36 ) with alanine ( the arginine residue of the E . coli protein is involved in glucose-6-phosphate binding ) . The mutant protein , OtsAR36A , was overexpressed in strain Ar0011 ( wild-type + up-otsAR36A ) at similar levels to OtsA in strain Ar0004 ( Fig 3A ) . However , the Ar0011 strain had a much shorter doubling time compared to strain Ar0004 , and similar to that of the wild-type ( Fig 1A ) . Staining with fluo-vancomycin revealed single septa located at the midcell of dividing bacillary-form cells ( Fig 3B ) , together with evidence of occasional foci located in the lateral walls ( indicated by white arrows in the NCW image ) . The mutant protein was also expressed under control of the native promoter in the ΔotsA mutant ( ΔotsA + otsAR36A; strain Ar0115 ) . It could not restore normal bacillary growth to the mutant ( S1 Fig ) . Moreover , when the mutant protein was overexpressed this did not affect bacillary growth or sensitivity to osmotic stress ( S4 Fig ) . We hypothesised that the loss of a morphogenetic function of OtsAR36A in vivo could reflect an inability to synthesise a threshold concentration of trehalose-6-phosphate ( T6P ) that may be necessary for the morphogenetic function of OtsA . To examine this we monitored intracellular levels of T6P in wild-type cells before and after salt stress . Based on a cell volume of 10−15 l , the T6P concentration can be estimated as ranging from approximately 2 mM in the wild-type ( no salt stress ) to 0 . 2 mM in the otsA mutant . An approximate 40% decrease in the intracellular concentration of the metabolite was noted in wild-type cells 3 h after salt stress ( Fig 3C , bars 2 and 3 , compared to bar 1 ) , coincident with the time when we noted changes in cell morphology ( see below ) . To examine this further , we overexpressed the E . coli treC gene [31] encoding T6P hydrolase in the wild-type ( wild-type + treCEC; strain Ar0012 ) . The recombinant strain grew slowly and formed constitutive myceloids in the absence of salt-stress ( Fig 1A and see below , Fig 4C ) . These cell aggregates had an average maximum dimension between extremities of 5 . 4 μm ( n = 361 ) , similar to those formed by the ΔotsA null mutant strain . Measurements of intracellular T6P revealed a significant depletion ( 17% of the level in the wild-type ) of this metabolite in Ar0012 compared with the wild-type ( Fig 3C , bar 4 ) , indicative of functional activity of TreCEc in Arthrobacter . The reduced level of T6P in Ar0012 was comparable to the amount detectable in the ΔotsA mutant strain ( Fig 3C , bar 7 ) . We also compared the levels of this metabolite in the strains overexpressing OtsA and OtsAR36A . Whereas the former strain , Ar0004 , contained 137% of the amount in the wild-type ( bar 6 ) , reflecting increased synthesis due to amplification of the enzyme , the latter had levels similar to wild-type , reflecting expression of the single-copy wild-type gene in this strain and an absence of metabolic activity due to the active site mutation in the overexpressed enzyme ( Fig 3C bar 5; relative to OtsA , purified OtsAR36A exhibited 7 . 53% +/- 0 . 22 enzyme activity ) . Quantification of T6P in strain Ar0010 overexpressing OtsAEc revealed 118% of the levels found in the wild-type ( Fig 3C , bar 8 ) , implying that an increase in the metabolite in the absence of increased amounts of OtsA protein is insufficient to promote any change in cell morphology . Moreover , expression of otsAEc in the ΔotsA mutant strain restored T6P synthesis ( bar 9 ) but , as described above , did not change the constitutive myceloid phenotype of the mutant . We used fluo-vancomycin to stain myceloids . Due to the extensive three-dimensional structure of myceloids of the ΔotsA mutant ( strain Ar0002 ) , fluorescence microscopy was more challenging and better resolution images were obtained with smaller cell aggregates . In these myceloids we observed irregular peptidoglycan synthesis , with most staining associated with adjacent poles of contiguous non-separated cells , the sites of joined cells being evident as cell envelope constrictions in the corresponding differential interference contrast images ( as indicated by black arrows , S5A Fig ) . In many long cells ( of average 1 . 1 μm length , examples indicated by white arrows in the corresponding overlay image ) , there was no observable nascent peptidoglycan in the midcell region . A similar picture emerged when salt-induced myceloids of the wild-type were examined . Cultures were grown to early log phase ( 20 h ) in LB amended to a final concentration of 0 . 57 M NaCl . Fluo-vancomycin staining revealed irregular patterns of nascent peptidoglycan , with a reduced frequency of midcell peptidoglycan synthesis evident in longer cells ( indicated by white arrows in the overlay image , S5B Fig ) . The myceloids of the strain expressing TreCEc ( strain Ar0012 ) also showed evidence of reduced synthesis of peptidoglycan at the midcell with evidence of longer cells ( indicated by white arrows in S5C Fig ) . In addition , to examine how salt stress affected the transition from bacillary to myceloid growth of the wild-type in a time-course , cultures were grown to early log-phase in LB and subsequently in LB amended to a final concentration of 0 . 57 M NaCl , sampled at successive time-points and stained with fluo-vancomycin . In non-amended medium , after 3h , the proportion of cells scored with midcell peptidoglycan synthesis was 62% ( n = 480 ) , whereas with salt-stress , the percentage was reduced to 49% ( n = 280 ) . In addition , in cells from non-amended medium , we also observed one or two foci of peptidoglycan synthesis in the lateral walls in 39 . 4% of cells ( n = 513 ) ; these foci are likely sites for growth of branches , consistent with the tendency for the wild-type to form occasional emerging branches as observed in SEM images ( Fig 1C ) . After 3 h salt-stress , there was an increase to 66 . 5% ( n = 524 ) of the proportion of cells with foci of nascent peptidoglycan in the lateral walls . Using dynamic light scattering ( DLS ) , we examined the native state of Arthrobacter OtsA expressed and purified from E . coli . This revealed two populations of the protein: smaller assemblies with an average diameter of 68 . 86 nm and much larger assemblies with an average diameter of 1738 . 42 nm ( Fig 4A ) . The protein was reanalyzed after denaturing the multimeric forms in 4 M urea followed by dialysis . If all urea was removed prior to DLS , during the dialysis process ( 16 h ) all the protein self-assembled into two populations of multimeric forms with similar average diameters to those of assemblies prior to denaturation ( S6 Fig ) . After denaturing in 4M urea and subsequent dilution to a final concentration of 1M urea , a much slower self-assembly process could be monitored by DLS in real-time ( Fig 4B ) . We used these conditions to then ask if T6P can promote OtsA self-assembly . We observed that increasing concentrations of this metabolite had a dramatic effect on promoting the rate of OtsA polymerization ( Fig 4B ) . Prior to addition of T6P , the average diameter of OtsA was 64 . 86 nm . After polymerization of OtsA promoted by 500 μM T6P ( a physiologically relevant concentration–see above ) , there was only one population of the protein detected consisting of large assemblies with an average diameter of 2745 . 58 nm . The implication is that T6P can promote the assembly of OtsA into large protein networks . As 1mM MgCl2 was used in these assays and magnesium ions are required for OtsA enzyme activity [28] , we then examined whether magnesium ions have a role in the self-assembly of OtsA . Purified OtsA was denatured as described above and then dialysed against a phosphate buffer containing 1M urea and no magnesium ions . The protein was then incubated with between 0 and 1mM MgCl2 and assembly monitored in real time using DLS . Little or no assembly was observed in the absence of the ion , whereas addition of 1mM or greater MgCl2 promoted assembly ( Fig 4C ) , although the maximum assembly was much less than that observed in the presence of T6P ( Fig 4B ) . We also used DLS to analyse the native state of OtsAR36A , revealing protein structures of 26 . 5 nm average diameter ( S7A Fig ) . Moreover , addition of T6P failed to promote self-assembly of OtsAR36A ( Fig 4B ) . Consequently , we inferred that the lack of any morphogenetic activity of OtsAR36A is not simply due to its loss of enzyme activity but presumably because the amino substitution also affects the protein’s tertiary structure and its ability to both interact with T6P and form large assemblies . DLS analysis of OtsAEc indicated protein structures with a range of sizes , and a modal diameter of approximately 7 nm ( S7A Fig ) . No increase in diameter was observed after addition of T6P ( Fig 4B ) . A feature of the Arthrobacter OtsA is its enzymatic activity at low temperatures due to a very flexible ‘N-loop’ containing the active site [25] , a characteristic that distinguishes the protein from its E . coli counterpart . To test if this flexible N-loop affects assembly formation , we purified and tested the assembly of two more OtsA proteins , OtsAA3mu and OtsAEcmu , which have , respectively , the E . coli N-loop replacing that of the Arthrobacter protein and vice versa . Whereas OtsAA3mu retained the ability to polymerize , albeit less efficiently , OtsAEcmu behaved like OtsAEc with no evidence for self-assembly ( S8 Fig ) . Consequently , we inferred that the N-loop alone is insufficient to promote self-assembly . We used transmission electron microscopy ( TEM ) to examine different states of assembly of OtsA . As urea will interfere with negative staining , we chose to dialyse purified OtsA in phosphate buffer lacking magnesium ions . This resulted in depolymerization as evident in the sizes of imaged protein structures which had an average diameter of approximately 60 nm ( Fig 4D ) , consistent with the dimensions of OtsA depolymerized after urea treatment as determined by DLS . Negative-stained OtsAEc , purified the same way , was most abundant as structures of approximately 10 nm diameter ( S7B Fig ) , again consistent with the size of protein structures determined by DLS analysis ( see above ) . TEM of OtsAR36A revealed structures of approximately 25 nm diameter ( S7B Fig ) , consistent with the DLS data for this protein . We then imaged OtsA after addition of magnesium ions alone or combined with T6P . After incubation for 30 min with 0 . 1mM MgCl2 we observed the appearance of branched protein filaments of varying lengths , up to approximately 200 nm in length ( Fig 4D ) . After 30 min incubation with 1 mM MgCl2 and 500 μM T6P , very large assemblies could be observed but only with low resolution using negative staining . Consequently , we used positive staining to obtain images of better resolution , as exemplified in Fig 5D , indicating assembly of the protein into large networks of greater than 2000 nm diameter . To examine localization of OtsA in vivo , a C-terminal translational fusion with mCherry was expressed in Arthrobacter A3 using the native otsA promoter sequence and a single-copy gene fusion integrated at the chromosomal otsA locus ( strain Ar0007 ) . Cells expressing the fusion protein grew normally . In addition , morphogenetic functionality was indicated both by the ability of the fusion protein expressed under control of the native promoter to restore normal snap-division growth to the ΔotsA mutant ( strain Ar0116; S3 Fig ) and by the promotion of the characteristic multiple-septation phenotype in long , enlarged cells when the fusion protein was overexpressed ( strain Ar0006 , Fig 5B ) . In Ar0007 cells grown in LB , the majority of the protein assembled at the midcell region ( indicated as ‘m’ in the overlay image , Fig 5A ) and some at the cell poles ( indicated as ‘p’ in the overlay image , Fig 5A ) , co-localizing at sites of peptidoglycan synthesis as revealed by fluo-vancomycin staining of the same cells ( Fig 5A ) . To examine if osmotic stress affects protein localisation , strain Ar0007 expressing OtsA::mCherry was grown to early log phase in LB , which was then amended to a final concentration of 0 . 57 M NaCl . After 3 h salt stress , we observed a diffuse distribution of OtsA throughout the cells ( Fig 5A ) , some colocalising with sites of peptidoglycan synthesis at the cell poles , but no longer localized at the midcell region . When the fusion protein was overexpressed it clearly localized to the sites of multiple septum formation , again indicative of a morphogenetic function ( Fig 5B ) .
Coordinating trehalose concentration and morphology requires that pleomorphic Arthrobacter cells can detect osmotic stress and communicate this information to the cell division apparatus . Here we describe an unexpected role for the trehalose synthase protein OtsA , which doubles as a morphogenetic protein , acting as a direct link between trehalose synthesis and cell morphology , and effecting the transition from bacillary growth to the development of myceloids . A product of the OtsA enzyme , T6P , can function as a signaling proxy for osmotic stress but is insufficient itself to direct changes in cell morphology as evidenced from the lack of any morphogenetic function of the otherwise enzymatically active OtsAEc . In vitro , OtsA can self-assemble to form elaborate protein networks , this assembly being promoted by T6P . In vivo , when cells are growing in low medium osmolarity , we hypothesise that these protein networks have a morphogenetic cytoskeletal function in promoting normal cytokinesis leading to a bacillary growth-style . Indeed , in non-stressed cells the protein assembles at the midcell and poles , consistent with this hypothesis . We have analyzed the morphological outcomes of various permutations of the genetic background of Arthrobacter that affect either or both intracellular T6P and OtsA concentrations ( Table 1 ) . Overexpression of OtsA , resulting in increased T6P , leads to increased formation of septa and loss of coordination of cytokinesis . Moreover , the overexpressed protein localizes at the multiple sites of septum formation in filamentous cells . But an increase in intracellular T6P , due to overexpression of OtsAEc , is insufficient itself to cause aberrant cytokinesis . Salt stress leads to a depletion of T6P and this likely affects the dynamics of OtsA self-assembly in vivo , resulting in the observed diffuse cytoplasmic distribution of the protein in salt-stressed cells and the reduction of peptidoglycan assembly at the midcell , but increasing the likelihood of the emergence of branches from lateral cell walls . These effects , coupled with a reduction in snap-division frequency , presumably lead to the growth of myceloids . Additional evidence that T6P can act as an intracellular proxy for salt stress comes from experimentally depleting this metabolite by expression of E . coli T6P hydrolase . A consequence of this depletion is that the strain can only grow with a myceloid morphology , despite expressing normal levels of OtsA . The psychrotrophic nature Arthrobacter A3 is to an extent due to its ability to accumulate trehalose as a cryoprotectant , which in turn is due to the enzymatic activity of OtsA at low temperatures [25] . This activity is due to a very flexible ‘N-loop’ containing the active site [25] , a key feature that distinguishes the protein from its E . coli counterpart . However , exchanging the N-loops of the Arthrobacter and E . coli proteins indicated that this sequence alone is not responsible for the self-assembly characteristic of the former , although the Arthrobacter protein with an E . coli N-loop is less proficient at self-assembly compared to its wild-type counterpart . In the genetic backgrounds in which either OtsAR36A or OtsAEc are overexpressed and T6P levels are normal or increased , these proteins have no effect on cytokinesis . The lack of in vivo morphogenetic and in vitro self-assembly activities of either OtsAR36A or OtsAEc can be rationalized in part by the active site mutation of the former and reduced N-loop flexibility of the latter , both of which likely affect T6P binding . We are currently comparing the structural properties of these variant proteins to identify other features that contribute to the ability of Arthrobacter OtsA to self-assemble . Our analysis of published Arthrobacter genome sequences indicates that these bacteria lack typical actin-like or intermediate filament cytoskeletal proteins found in other bacteria , including some other actinobacteria . The evolutionary recruitment of the principle biosynthetic enzyme involved in synthesis of the osmoprotectant trehalose to a function that regulates morphology in response to osmotic stress is an intriguing adaptation to coping with extreme environments . This adds to a few other known examples of biosynthetic enzymes co-opted for morphogenetic roles in bacteria . The primary enzyme involved in CTP synthesis , CtpS , from Caulobacter , E . coli and several eukaryotic species can self-assemble into linear filaments [32–34] and in Caulobacter this protein has a role in determining cell shape [32] . In the same bacterium , a NAD ( H ) -binding oxidoreductase , KidO , can inhibit Z-ring formation [35] , linking cell division with metabolic status . In Bacillus subtilis the membrane-associated glucosyltransferase UgtP , involved in glycolipid biosynthesis , acts as a metabolic sensor governing cell size . During growth in rich media , when ugtP expression is upregulated , the enzyme localises to the midcell division site and inhibits Z-ring formation at the midcell [36] . In E . coli , a non-homologous glucosyltransferase OpgH , an integral inner membrane protein that is functionally analogous to UgtP of B . subtilis in linking cell size with central metabolism , is believed to inhibit Z-ring formation by a different mechanism involving sequestering FtsZ [37] . In these latter two examples , the bacteria utilise UDP-glucose as an intracellular signal and proxy for nutrient availability . In contrast to these examples , in Arthrobacter the OtsA glucosyltransferase doubles as a morphogenetic protein and determine cell morphology in response to an environmental signal . Although several other bacterial species are known to exhibit morphological plasticity as a stress survival strategy , switching from a rod-shaped to a filamentous morphology [38] , the mechanism for this transition , when known , is quite different . In E . coli , the product of an SOS-induced gene , SulA , binds to FtsZ monomers , inhibiting polymerization [39] . In older Caulobacter cells , depletion of FtsZ leads to filamentation [40] . It will be of interest to determine if OtsA in Arthrobacter interacts with FtsZ and to investigate the properties of OtsA from other actinobacteria , including M . tuberculosis which undergoes filamentation in macrophages [41] .
The growth conditions for Arthrobacter strains and E . coli cultures are given in detail in the Supplemental protocols . The list of all relevant strains and plasmids is provided in S1 Materials and Methods . Details of how each plasmid and strain were constructed , and the primers used for amplification and cloning of DNA sequences are also provided in the Supplemental protocols . All strains were analyzed in exponential growth phase unless otherwise stated . Nascent cell walls were stained using fluo-vancomycin as described previously [42] . Bacterial cells were stained by DAPI ( 0 . 1ug/ml , PBS ) , vancomycin ( 1ug/ml , PBS ) and fluo-vancomycin ( 1ug/ml , PBS ) for 20min . Cultures were then washed in PBS , and suspended in 1 . 6% formaldehyde ( in PBS ) and left on ice for 1 hr . Treated cells were distributed on microscope slides that had been treated with 0 . 1% ( wt/vol ) poly-L-lysine ( Sigma ) . Images were acquired on a confocal laser scanning microscope ( Olympus FV1000 ) . The purification of proteins , immunoblotting and coimmunoprecipitation analysis were carried out as described previously [24 , 25] . The assembly of OtsA was monitored in real-time with a dynamic light scattering assay using a Brookhaven Instruments BI-200SM system ( USA ) . The wavelength of the stable argon ion laser was 532 nm . The assay was performed at 20°C . OtsA was denaturated in 4 M urea for 5 min and then diluted in polymerization buffer O ( 20mM Tris HCl , pH 7 . 5 , 1 mM MgCl2 ) , with different concentrations of T6P , to 10 uM final concentration of OtsA and 1M of urea . To measure the effect of magnesium ions , MgCl2 was excluded from buffer O . The intensity of scattered light was measured at an angle of 90° . Negative staining electron microscopy was used to visualize OtsA . Carbon-coated copper grids ( 400 mesh , Electron Microscopy Sciences ) were glow discharged for 5 s before use . Before applying OtsA , a drop of 0 . 2 mg/ml cytochrome c was pipetted onto the carbon , incubated for 30 s , and then blotted with filter paper . A drop of OtsA solution was then applied to the carbon and incubated for 10 s before the excess was blotted . The grid was immediately rinsed with 3–4 drops 2% uranyl acetate , blotted , and air-dried . For positive staining , 5 ul of protein ( 10 μM ) were placed on carbon coated grids , incubated for 2 min , washed in buffer A ( 20 mM sodium phosphate , pH 7 . 5 and 20 mM NaCl ) , and stained with 2% uranylacetate for 30 s . Protein were visualized and photographed using a Tecnai-G2-F30 electron microscope . To image cells , exponentially growing Arthrobacter strains were fixed in 4% paraformaldehyde , 2 . 5% glutaraldehyde ( PBS , PH 7 . 4 ) , at 25°C for 2 h . SEM of whole cells were carried out as described previously [43] . Trehalose-6-phophate phosphatase was obtained from E . coli as described [44] . To measure trehalose-6-phosphate , cells were broken using ultrasonication at 4 0C and extracts prepared and assayed as described [45] . For each strain , four replicate assays were conducted . Trehalose was assayed as previously described [24] .
|
For free living bacteria , little is known about how environmental cues are perceived and translated into changes in cell morphology . Here we describe how a biosynthetic enzyme involved in synthesis of an important intracellular osmoprotectant doubles as an osmotic stress sensing morphogenetic protein . This protein is involved in an adaptive response involving a growth transition in stress-tolerant bacteria , from growing as individual cells to forming non-separating branched cell aggregates . We demonstrate that the protein can self-assemble into large networks , consistent with its role as a morphogenetic protein , this assembly process being promoted by a metabolic product of the enzyme . Depletion of either this metabolite or the morphogenetic protein results in the inability of the bacteria to grow as individual cells in conditions of low osmolarity .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"fluorescence",
"imaging",
"urea",
"chemical",
"compounds",
"disaccharides",
"arthrobacter",
"cell",
"cycle",
"and",
"cell",
"division",
"cell",
"processes",
"carbohydrates",
"organic",
"compounds",
"materials",
"science",
"macromolecules",
"bacteria",
"materials",
"by",
"structure",
"research",
"and",
"analysis",
"methods",
"polymers",
"polymer",
"chemistry",
"specimen",
"preparation",
"and",
"treatment",
"trehalose",
"staining",
"imaging",
"techniques",
"chemistry",
"actinobacteria",
"cell",
"staining",
"organic",
"chemistry",
"cell",
"biology",
"peptidoglycans",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"osmotic",
"shock",
"organisms"
] |
2017
|
A trehalose biosynthetic enzyme doubles as an osmotic stress sensor to regulate bacterial morphogenesis
|
Severe fever with thrombocytopenia syndrome ( SFTS ) is a tick-borne infectious disease with a high case fatality rate , and is caused by the SFTS virus ( SFTSV ) . SFTS is endemic to China , South Korea , and Japan . The viral RNA level in sera of patients with SFTS is known to be strongly associated with outcomes . Virological SFTS diagnosis with high sensitivity and specificity are required in disease endemic areas . We generated novel monoclonal antibodies ( MAbs ) against the SFTSV nucleocapsid ( N ) protein and developed a sandwich antigen ( Ag ) -capture enzyme-linked immunosorbent assay ( ELISA ) for the detection of N protein of SFTSV using MAb and polyclonal antibody as capture and detection antibodies , respectively . The Ag-capture system was capable of detecting at least 350–1220 TCID50/100 μl/well from the culture supernatants of various SFTSV strains . The efficacy of the Ag-capture ELISA in SFTS diagnosis was evaluated using serum samples collected from patients suspected of having SFTS in Japan . All 24 serum samples ( 100% ) containing high copy numbers of viral RNA ( >105 copies/ml ) showed a positive reaction in the Ag-capture ELISA , whereas 12 out of 15 serum samples ( 80% ) containing low copy numbers of viral RNA ( <105 copies/ml ) showed a negative reaction in the Ag-capture ELISA . Among these Ag-capture ELISA-negative 12 samples , 9 ( 75% ) were positive for IgG antibodies against SFTSV . The newly developed Ag-capture ELISA is useful for SFTS diagnosis in acute phase patients with high levels of viremia .
Between 2007 and 2010 , a severe febrile illness associated with gastrointestinal symptoms , thrombocytopenia , and leukocytopenia caused by an unknown etiological agent was reported in rural areas of Hubei and Henan provinces in Central China [1] . The case-fatality rate of the disease was reported to be between 12%–30% at that time . The disease was named severe fever with thrombocytopenia syndrome ( SFTS ) , or fever , thrombocytopenia and leukopenia syndrome ( FTLS ) [1 , 2] . A novel phlebovirus , termed SFTS virus ( SFTSV and also known as Huaiyangshan virus or Henan Fever Virus ) , has been identified as the causative agent of the disease [1 , 2 , 3] . SFTSV has been detected in two tick species ( Haemaphysalis longicornis and Rhipicephalus microplus ) in epidemic areas , suggesting that these ticks are the most likely vectors for transmission of the virus to humans [1 , 3] . SFTSV antibodies were detected at various rates in goats , cattle , sheep , pigs , dogs , and chickens [4 , 5 , 6 , 7 , 8 , 9] , indicating that these animals were infected with SFTSV . There are no reports describing that the virus causes disease in these animals , suggesting that these animals and some species of ticks are the reservoirs of SFTSV . SFTS is endemic to Japan and South Korea [10 , 11] . SFTS patients show abrupt onset of fever with gastrointestinal tract symptoms in the early phase . Most patients have marked thrombocytopenia and leukocytopenia at this stage . Later stages of the syndrome are characterized by a progressive multiple organ failure in fatal cases or a self-limiting process in survivors [12] . The level of viral RNA in patient sera correlates to the clinical outcome . In fatal cases , viremia increases to 109 viral copies per mL . In contrast , the convalescent stage is characterized by decreasing levels of viremia and normalization of clinical laboratory parameters [13 , 14 , 15] . SFTSV is classified into the genus Phlebovirus , family Bunyaviridae . Tick-borne phleboviruses ( TBPVs ) including SFTSV are globally distributed . TBPVs closely related to SFTSV , such as Heartland virus , Malsoor virus , and Hunter Island Group viruses , have been discovered [16 , 17 , 18] . Phylogenetic and serological studies revealed that TBPVs are classified into four distinct groups , Uukuniemi group , Bhanja group , SFTS/Heartland virus group , and Kaisodi group [19 , 20] . SFTSV is classified into the SFTS/Heartland virus group . SFTSV has 3 segmented , single-stranded RNA genomes and is composed of large ( L ) , medium ( M ) , and small ( S ) segments . The L , M , and S segments encode an RNA-dependent RNA polymerase , a precursor of glycoproteins ( Gn and Gc ) , a nucleocapsid ( N ) protein and a nonstructural ( NS ) protein using an ambisense coding strategy , respectively [1] . The N protein is highly immunogenic and conserved among all isolates in each of the phleboviruses [21 , 22] . Therefore , N protein is often selected as a target of antigen ( Ag ) and antibody detection [23 , 24 , 25] . SFTS and other infectious diseases are difficult to diagnose clinically without microbiological tests , particularly when symptoms are non-specific . Hence , laboratory tests are necessary for SFTS diagnosis . Several genome amplification-based methods for SFTS diagnosis have been reported e . g . , conventional reverse transcription-PCR ( RT-PCR ) , quantitative RT-PCR , reverse transcription-loop-mediated isothermal amplification assay ( RT-LAMP ) , and reverse transcription-cross-priming amplification coupled with vertical flow visualization [2 , 13 , 26 , 27 , 28] . However , genome amplification techniques are limited by their requirement of expensive equipment and technical expertise . Methods for the detection of viral Ags using an Ag-capture sandwich ELISA have been previously described , and the sensitivity of this assay is comparable to that of RT-PCR for the detection of Lassa virus and filoviruses [23 , 24 , 29 , 30 , 31 , 32 , 33] . The assay is highly accurate in identifying viral Ags in a rapid and robust manner; additionally , it has been accepted as a useful method for diagnosis during the acute phase of viral infections . To our knowledge , an Ag-capture sandwich ELISA has not yet been developed for SFTS . In this study , mouse MAbs against SFTSV N protein were generated and characterized . Furthermore , Ag-capture ELISA for detection of SFTSV using the MAb was developed , and its efficacy in SFTS diagnosis was evaluated using sera collected from patients with SFTS .
All samples were collected as part of public health diagnostic activities for SFTS , were pre-existing relative to the start of the study , and were examined as anonymous samples . All protocols and procedures were approved by the research ethics committee of the National Institute of Infectious Diseases for the use of human subjects ( no . 531 ) . The experiments with animals were performed in strict accordance with the Animal Experimentation Guidelines of the National Institute of Infectious Diseases . The protocol of experiments for mice and rabbits were approved by the Institutional Animal Care and Use Committee of the National Institute of Infectious Diseases ( Permit number: 112116 and 111124 , respectively ) . Mouse myeloma cells , P3X63Ag8 . 653 , obtained from the American Type Culture Collection ( Manassas , VA ) , were maintained in RPMI 1640 medium ( Sigma-Aldrich , St . Louis , MO ) supplemented with 10% heat-inactivated fetal bovine serum ( FBS ) and kanamycin sulfate ( Life Technologies , Carlsbad , CA ) . Hybridomas were maintained in Growth Medium E ( Stem Cell Technologies , Vancouver , Canada ) , RPMI 1640 medium supplemented with 10% FBS and kanamycin sulfate , or KBM-343 medium ( Kohjin Bio Co . , Ltd . , Saitama , Japan ) supplemented with antibiotics . BTI-TN-5B1-4 ( Tn5 , High Five; Life Technologies ) insect cells were maintained in TC100 ( Life Technologies ) supplemented with 10% FBS , 2% tryptose phosphate broth ( Difco , Detroit , MI ) , and kanamycin sulfate . Huh7 , Vero , and Vero E6 cells , obtained from the American Type Culture Collection , were maintained in DMEM ( Sigma-Aldrich ) supplemented with 5% FBS and kanamycin sulfate . SFTSV strains YG1 , SPL004 , and SPL010 isolated from serum samples of Japanese patients with SFTS were used [11] . SFTSV strain HB29 was kindly provided by De-Xin Li and MiFang Liang , Chinese Center for Disease Control and Prevention , Beijing , People’s Republic of China . As a negative control antigen , a supernatant of Vero E6 cells infected with Rift Valley fever virus ( RVFV ) strain MP-12 was used [23] . Experiments using infectious SFTSV and RVFV were conducted in a biosafety level ( BSL ) -3 laboratory . Forecariah virus ( FORV ) and Palma virus ( PALV ) , which were kindly gifted from Robert Tesh , University of Texas Medical Branch , USA , were handled in BSL-2 . The infectious dose of the SFTSV and RVFV stock solutions was determined by calculating the 50% tissue culture infectious dose ( TCID50 ) as described previously [11 , 13 , 34] . Preliminary experiments indicated that treatment of sera containing SFTSV ( 107 TCID50/ml ) with 1% triton X-100 in combination with UV-irradiation ( 312 nm , 2 . 5 mW/cm2 ) on a trans-illuminator for 10 min caused complete loss of viral infectivity in cells . Therefore , viruses used for Ag-capture ELISA were treated with UV-irradiation on a trans-illuminator for 10 min and followed by 1% Triton X-100 for the destruction of the virus particle . We asked medical personnel in Japan to inform us on a voluntary basis if they had seen any patients with symptoms similar to those of SFTS , as reported previously [11] . Through the courtesy of prefectural and municipal public health institutes , 63 serum samples were collected from 55 acute phase patients suspected of SFTS in Japan . Viral gene detection by the qRT-PCR and viral antibody detection by IgG ELISA and/or IFA were conducted to diagnose SFTS . From 55 patients , 34 of these were diagnosed as having SFTSV . Serum samples obtained from 18 healthy donors were used to establish the cut-off value of the IgG ELISA . Serum samples used for IgG ELISA were inactivated under the UV light in the biosafety cabinet for 1 h . Serum samples used for Ag-capture ELISA were treated with 1% Triton X-100 for the destruction of the virus particle followed by an UV-irradiation on a trans-illuminator for 10 min . The recombinant baculovirus , Ac-His-SFTSV-N expressing a histidine ( His ) -tagged SFTSV recombinant N ( rN ) protein at C-terminal , was generated as described previously [23 , 35 , 36] . Briefly , the cDNA encoding the N protein of SFTSV strain HB29 ( nucleotide position 43–780 of segment S , GenBank accession No . NC_018137 ) was artificially synthesized ( GeneScript , Piscataway , NJ ) and then was ligated into the BamHI sites upstream of the 8-His tag coding sequence of the transfer vector pAcYM1 [37] . Tn5 cells were transfected with mixtures of the transfer vector pAcYM1-SFTSV-N and BD BaculoGold Linearized Baculovirus DNA ( BD Biosciences , San Jose , CA ) according to the manufacturer’s instructions with the procedures described by Kitts et al . [38] but with modification by Matsuura et al . [37] . A baculovirus ( Ac-ΔP ) , which lacks polyhedrin expression , was used as a negative control virus . SFTSV rN protein [>75% purity as determined by ImageJ analysis ( http://rsbweb . nih . gov/ij/ ) of sodium dodecyl sulfate polyacrylamide gel electrophoresis] was generated as previously described [23 , 35 , 36] . Briefly , Tn5 cells infected with Ac-His-SFTSV-N were incubated at 27°C for 72 h . The cells were then washed three times with phosphate-buffered saline ( PBS ) solution . The cells were lysed in PBS solution containing 1% Nonidet P-40 ( NP-40 ) and sonicated . After the cell lysates were centrifuged at 17 , 800 × g for 10 min at 4°C , the supernatant fractions were collected as a source of SFTSV rN protein for purification . SFTSV rN protein was purified by Ni2+-nitrilotriacetic acid affinity chromatography ( Qiagen , Hilden , Germany ) according to the manufacturer’s instructions . RVFV rN protein was expressed and purified as described previously [23] . The histidine-tag was not removed from the rN protein . The concentration of the purified SFTSV and RVFV rN proteins were determined by the Pierce BCA Protein Assay Reagent ( Life Technologies ) . Antigens were aliquoted and stored at −80°C until use . BALB/c mice were immunized subcutaneously twice with the purified SFTSV rN protein emulsified in TiterMAX Gold ( TiterMax USA , Inc . , Norcross , GA , USA ) . Hybridomas were produced by fusion of myeloma cells with the splenic cells , obtained 4 days after the last immunization , using ClonaCell-HY Hybridoma Kit ( Stem Cell Technologies ) according to the manufacturer’s instructions . The culture supernatants of hybridoma cells were screened for the presence of antibodies against SFTSV antigen by IgG ELISA as described below . Positive hybridoma cells were cloned by limiting dilution . The isotypes of the MAbs were determined using Mouse Monoclonal Antibody Isotyping Kit ( AbD Serotec , Kidlington , UK ) . The MAbs were purified from mouse ascitic fluid ( Unitech Co . Ltd . , Chiba , Japan ) or from the culture supernatant by protein G column chromatography ( MAbTrap Kit , GE Healthcare UK Ltd . , Buckinghamshire , UK ) according to the manufacturer’s instructions . The concentration of each purified MAb was determined using the Pierce BCA Protein Assay Reagent . Two hybridoma clones ( designated as 2D11 and 9D3 ) producing MAbs reactive to SFTSV N protein were obtained . MAb 9D3 and MAb 2D11 were isotypes of IgG1 and IgG2a , respectively . The light chain of these MAbs was κ-type . Polyclonal antibodies to each of the rN proteins of SFTSV and RVFV were raised by immunization of rabbits with the respective rN protein [11 , 23] . Polyclonal antibodies to FORV and PALV were produced by infection of rabbits with FORV and PALV , respectively . Rabbit sera were obtained 7 days after infection . The experiments with animals were performed in strict accordance with the Animal Experimentation Guidelines of the National Institute of Infectious Diseases . The antigens of SFTSV strain YG1 , FORV , PALV , and RVFV were prepared for IFA as previously described [39] . Briefly , Vero cells infected with each virus ( MOI = 0 . 1 ) were cultured , harvested by trypsinization , washed with PBS , and mixed with parent uninfected cells at a ratio of 1:3 . The cells were spotted on to 14-well HT-coated slide glasses ( AR Brown Co . , Ltd . , Tokyo , Japan ) , air dried , and fixed with a mixture of methanol and acetone [1:1 ( v/v ) ] . These IFA antigen slides were stored at -80°C until use . They were thawed and dried immediately prior to use . The IFA was performed by diluting MAbs at the concentration of 1 ng/μl with PBS and were placed on the slides . As a positive control , rabbit sera diluted with PBS at 1:1 , 000 were also placed on the slides . The slides were incubated under humidified conditions at 37°C for 1 h . After washing with PBS , the slides were treated with Alexa Fluor 488 conjugated goat anti-mouse IgG ( H + L ) antibody ( Life Technologies ) or Alexa Fluor 488 conjugated goat anti-rabbit IgG ( H + L ) antibody ( Life Technologies ) diluted with PBS at 1:400 . The slides were incubated under humidified conditions at 37°C for 1 h . After washing , the slides were examined for antigen staining under a fluorescent microscope ( Olympus , Tokyo , Japan ) [11 , 39] . Immunohistochemical analysis was performed as previously described [11] . The mouse MAbs 9D3 and 2D11 were used in the immunohistochemical analysis as the primary antibodies . Lymph nodes of necrotizing lymphadenitis without SFTSV infection were used as negative controls for tissue specimens . The IgG ELISA was performed as previously described , except for antigen preparation [32 , 33 , 35 , 36] . Antigen preparation for IgG ELISA was performed by infecting Huh7 cells with SFTSV strain HB29 ( MOI = 0 . 1 ) and incubated at 37°C for 48 h . The cells were collected and washed with PBS , and then lysed with PBS solution containing 1% NP-40 . The cell lysates were centrifuged at 8 , 000 rpm for 10 min at 4°C , and the supernatant fraction was collected as a source of SFTSV antigen for IgG ELISA . Huh7 cell lysates without infection were treated in the same way as that for SFTSV antigen preparation and were used as a negative control antigen . Nunc-Immuno Plates ( Thermo Fisher Scientific Inc . , Waltham , MA ) were coated with a pre-determined optimal quantity of Huh7 cell lysates prepared from SFTSV-infected or uninfected cells diluted with PBS at 1:800 and incubated at 4°C overnight . The following procedure was performed in the same way as described previously [32] . The cut-off value was set as the average value of the control sera ( healthy donor sera ) plus three times standard deviation ( SD; mean + 3×SD ) . The sample was considered positive if it yielded an OD405 value above the cut-off value . The Ag-capture ELISA was performed as previously described [23 , 24 , 29] . Nunc-Immuno plates were coated with 100 ng of capture MAbs ( 9D3 or 2D11 ) in 100 μl of PBS at 4°C overnight , and then the wells were incubated with a blocking reagent . After removal of the blocking solution , a series of samples ( 100 μl/well ) diluted with PBST-M was added and incubated for 2 h at RT . After the plates were washed , 100 μl of the rabbit anti-SFTSV rN protein sera diluted 1:1 , 000 with PBST-M was added to each well , followed by incubation for 2 h at RT . After washing , HRP-conjugated goat anti-rabbit IgG antibody ( Life Technologies ) diluted 1:1 , 000 with PBST-M were added to each well and incubated for 2 h at RT . After further washing , 100 μl of ABTS [2 , 2azinobis ( 3-ethylbenzthiazolinesulfonic acid ) ] substrate solution ( Roche Applied Science , Penzberg , Germany ) was added and incubated for 30 min at RT . The optical density at 405 nm ( OD405 ) was measured against a reference of 490 nm using a microplate reader ( Model 680 Microplate Reader; Bio-Rad Laboratories Inc . , Hercules , CA ) . The adjusted OD405 value was calculated by subtracting the OD405 value of the negative antigen-coated wells from that of the corresponding wells . The cut-off value was set at the average value of the control sera ( antigen free ) plus three times the standard deviation ( SD; mean + 3×SD ) . The sample , which yielded an OD405 value above the cut-off value , was thus considered positive . Protein or viral quantities detected per 100 μL reaction in 96-well micro-plates in the Ag-capture ELISA were presented as “/100 μL/well . ” The qRT-PCR method using the qRT-PCR primer and probe sets targeted to N protein or glycoprotein genes was performed as described previously [13] . Genome copies obtained from qPCR assays were presented as “/mL” of serum samples . Unpaired t-test with Welch's correction was used to determine significant differences in the data using the GraphPad Prism 6 software program ( GraphPad software , La Jolla , CA ) . A significant difference was considered to be present for any p value <0 . 05 .
Novel MAbs ( 9D3 and 2D11 ) against SFTSV N were generated in this study . SFTSV N protein characterized by a diffuse granular cytoplasmic staining was detected by these MAbs through indirect immunofluorescence ( IFA ) for SFTSV infected Vero cells , but was not detected in Rift Valley fever virus ( RVFV ) infected cells ( Fig 1A ) . Since the serologic relationships between SFTSV and Bhanja group virus , including Forecariah virus ( FORV ) and Palma virus ( PALV ) , have been demonstrated [19] , we also examined the cross-reactivity of MAbs to these phleboviruses . As shown Fig 1A , both MAbs did not react to FORV and PALV . SFTSV antigens were detected in the lymph node specimens obtained from patients with SFTS clearly through immunohistochemistry ( IHC ) staining using the MAbs , but not in that of the patients without SFTS ( Fig 1B ) . The minimum amounts of SFTSV rN protein detected in the Ag-capture ELISA with MAb 2D11 and MAb 9D3 were 40 pg and 10 pg /100 μL/well , respectively ( Fig 2A ) , while levels of up to 2 . 6 ng of RVFV rN protein were not detected ( Fig 2A ) . The Ag-capture ELISA using MAb 9D3 was more sensitive in detection of SFTSV rN protein than the MAb 2D11 . The Ag-capture ELISA using both MAbs ( 9D3 and 2D11 ) as capture antibody was less sensitive than using MAb 9D3 alone ( S1 Fig ) . Therefore , the MAb 9D3-based Ag-capture ELISA was selected for further experiments . Four SFTSV strains , including a Chinese strain ( HB29 ) that we tested were detected in the Ag-capture ELISA . The minimum levels of SFTSV detected in the assay were 1 , 100 , 1 , 200 , 350 , and 540 TCID50 /100 μL/well for SFTSV strains HB29 , YG1 , SPL010 , and SPL004 , respectively ( Fig 2B ) . Because 1 . 0 TCID50 of SFTSV corresponds to approximately 15 . 4 copies of the SFTSV genome [13] , the sera containing theoretical value of more than 5 , 000 copies of the SFTSV genome/100 μL/well could be used for Ag detection in this assay . In contrast , RVFV antigens prepared from virus infected culture supernatants were not detected in this assay . In order to evaluate the efficacy of the Ag-capture ELISA in SFTS diagnosis , these systems were tested using acute phase sera collected from patients suspected of having SFTS . In a total of 63 serum samples , 24 samples were negative by qRT-PCR , and they were also negative for IgG antibodies to SFTSV determined by IgG ELISA and IFA . The patients , whose sera were negative for virus genome by qRT-PCR and IgG antibodies to SFTSV , were confirmed to be patients without SFTS . In a total of 63 serum samples , 27 samples showed a positive reaction in the Ag-capture ELISA ( Table 1 ) . Thirty-nine samples including all the Ag-capture ELISA-positive samples were SFTSV genome positive in the qRT-PCR . All 24 samples containing SFTSV genome copy numbers higher than 105 copies/ml showed a positive reaction in the Ag-capture ELISA , while only 3 of 15 SFTS-genome positive samples with the viral RNA copy numbers of less the 105 copies/ml had a positive reaction in the assay . The sensitivity and the specificity of the Ag-capture ELISA were 69% ( 27/39 ) and 100% ( 24/24 ) , respectively , based on the qRT-PCR results . The viral RNA copy number in the Ag-capture ELISA-positive samples ( mean ± SD: 6 . 548 ± 0 . 227 log10 copies/ml ) was significantly higher than that observed in the Ag-capture ELISA-negative samples ( 4 . 077 ± 0 . 178 log10 copies/ml; p < 0 . 0001; Fig 3A ) . We then determined the antigen titers of each of the serum samples by using serially-diluted serum samples for the Ag-capture ELISA . There was no statistically significant difference in the antigen titers between patients with SFTS with fatal and non-fatal outcomes ( p = 0 . 08; Fig 3B ) . However , high antigen titers ( ≥160 ) were detected in 82% ( 9/11 ) of serum samples collected from patients with fatal outcomes , and only 18% ( 2/11 ) of serum samples collected from patients with non-fatal outcomes were detected ( Table 2 ) . Furthermore , significant high antigen titers ( ≥10 , 240 ) were detected in serum samples collected from three patients with fatal outcomes ( Table 2 ) . We performed the IgG ELISA against SFTSV in the samples to determine the antibody responses , because the presence of the antibodies against SFTSV might inhibit the capture capacity of the MAb for SFTSV N protein . The IgG antibody status was compared between the Ag-capture ELISA positive and negative groups among the total of 39 serum samples positive for qRT-PCR . In 27 Ag-capture ELISA-positive samples , 11 ( 41% ) samples were IgG ELISA-positive , while 9 of 12 ( 75% ) Ag-capture ELISA-negative samples were IgG ELISA-positive . The OD values of the Ag-capture ELISA-positive samples in IgG ELISA ( mean ± SD; 0 . 143 ± 0 . 024 ) were significantly lower than those of Ag-capture ELISA-negative samples in the assay ( mean ± SD; 0 . 401 ± 0 . 087; p < 0 . 05 ) .
We demonstrated that both the novel 2 MAbs ( 9D3 and 2D11 ) generated in this study reacted to SFTSV , but not to RVFV , FORV , and PALV in the genus Phlebovirus ( Fig 1A ) . However , a close antigenic relationship between FORV , PALV and SFTSV was demonstrated by the serological tests [19] . In addition , these MAbs did not react to the recombinant N protein of Heartland virus in IFA ( S2 Fig ) . Therefore , we speculate that the MAbs may not be cross-reactive to Malsoor virus and Hunter Island Group virus , which are closely related to SFTSV . This is because the amino acid sequence homology of N protein of SFTSV strain HB29 with those of Hearland virus , Malosoor virus , and Hunter Island Group virus were shown to be 61 . 6% , 55 . 6% , and 60 . 9% , respectively [20] . As amino acid sequence identities among the N protein of SFTSV strains available from databases are conserved with more than 98% homology , it is thought that the N protein of the Japanese strains and also the Chinese strains and South Korea strains are detectable in the Ag-capture ELISA and IHC using the MAbs developed in the present study . Furthermore , MAbs may be useful for future development of rapid dipstick , flow-through devices that require minimal training and do not require electricity . The rN protein concentration detectable using the Ag-capture ELISA for detecting SFTSV ( 10–40 pg /100 μL well ) was the same level as that for detecting RVFV with the previously developed Ag-capture ELISA [23] . However , the detection limit of the concentration of authentic viral antigens detectable by the Ag-capture ELISA for SFTSV ( 350–1 , 200 TCID50/100 μL/well seems to be higher than that for RVFV ( 7 . 8–31 . 3 pfu /100 μL/well ) [23] . Although it is difficult to simply compare the detection limits between the two ELISAs , a more sensitive detection of RVFV in Ag-capture ELISA in the previous study may be due to an abundant non-virion associated N protein in the viral supernatants collected from infected cell cultures that exhibit an obvious CPE as described by Shafagati et al [40] . In contrast , since SFTSV do not exhibit obvious CPE on infected Vero cells . Therefore , relatively lower detection limits of authentic SFTSV N protein in the Ag-capture ELISA seems to be attributable to a low amount of non-virion associated SFTSV N protein in the viral supernatants , despite virions being lysed by treatment with the detergent , Triton X-100 . The viral RNA level in sera of patients with SFTS was reported to be associated with the outcomes [13 , 14 , 15] . During the first stage of the disease ( day 1 to 7 post-onset of illness , taking the day on which symptoms , fever , first appeared as day 0 ) , the serum viral load is high ( average 105–106 copies/ml ) regardless of the outcomes of fatal or non-fatal cases [41] . During the second stage of the disease ( day 7 to 13 post-onset of illness ) , the serum viral load decreased in non-fatal cases but still remained high in fatal cases ( average 108 copies/ml ) [41] . It has also been reported that the amounts of Ag detected by the Ag-capture ELISA are well correlated with viremia of ebolavirus or Lassa virus following experimental animal infection [32 , 42] . Also , a moderate difference has been demonstrated in the serum ebolavirus Ag levels between patients who died and those who survived [43] . These findings suggested that the patient outcomes were expected from the results of Ag-capture ELISA . Indeed , we found that high antigen titers ( ≥160 ) were detected at a higher rate in serum samples collected from patients with fatal outcomes than from serum samples collected from patients with non-fatal outcomes ( Table 2 ) . However , there was no statistically significant difference in antigen titers between patients with SFTS with fatal and non-fatal outcomes ( p = 0 . 08; Fig 3B ) . This might be due to the small-scale samples used in this study . Thus , further large-scale investigation is required to elucidate the correlation between the results of Ag-capture ELISA and patient outcomes . Among qRT-PCR positive-patients , the Ag-capture ELISA-negative patients showed significantly higher IgG responses than the Ag-capture ELISA-positive patients ( Fig 3C ) . We speculate that the amount of N protein in serum samples collected may be lower than the detectin limit of the Ag-capture ELISA , since these patients had already reached convalescence phase , where IgG antibodies to SFTSV N protein had been induced . The induced antibodies against SFTSV N protein in serum samples may inhibit the reaction of the MAb the N protein in the Ag-capture ELISA . A similar event was reported in the case of development of Crimean–Congo hemorrhagic fever virus ( CCHFV ) N protein detection ELISA system . The presence of antibodies to CCHFV N protein in the samples inhibited the reactivity of MAb with antigens in the CCHFV N protein Ag-capture ELISA [24] . Direct evidence of an inhibitory effect on Ag detection by anti-N Abs has been provided by experiments using mixtures of viremic serum with increasing amounts of immune serum [44] . Our data indicate significantly high IgG levels in the serum samples of Ag-capture ELISA negative patients . Based on these findings , the underlying immune status of patients may be characterized using this assay . It is concluded that the Ag-capture ELISA developed is available for serum samples collected during the early phase of SFTS before antibody responses become detectable . Furthermore , the specific reaction of the MAbs to SFTSV antigens in tissues of patients with SFTS was confirmed ( Fig 1B ) . Therefore , the MAbs were demonstrated to be of use in the detection of SFTSV antigen in the autopsied materials for SFTS diagnosis with IHC . In this study , novel MAbs to SFTSV N protein were generated . The Ag-capture ELISA used for the MAbs in detecting SFTSV in the serum samples of the SFTS suspected patients was developed . Furthermore , MAbs were applied for the detection of SFTSV antigen in autopsied materials . These SFTSV antigen detection methods were useful for SFTS diagnosis .
|
Severe fever with thrombocytopenia syndrome ( SFTS ) is a tick-borne emerging infectious disease caused by a novel bunyavirus , SFTS virus ( SFTSV ) . Since first discovered in China in 2011 , SFTSV has been detected from SFTS patients and ticks with expanding geographic ranges from China to Japan and South Korea . The potential for SFTS spread to other warm or sub-tropical regions makes it a serious concern to the public health . It is of great importance to detect SFTSV rapidly and specifically for the effective control of the disease . For the diagnosis of viral infections , a sandwich antigen ( Ag ) -capture ELISA detecting viral nucleoprotein ( N ) in viremic serum samples has been widely applied to detect the agents , since it is the most abundant viral antigen and has highly conserved amino acid sequence . In this study , using the novel monoclonal antibodies raised against SFSTV-N , an Ag-capture ELISA system was developed , and the validation of this system was performed using sera collected from SFTS-suspected patients . Our data show that the Ag-capture ELISA was useful for the diagnosis of SFTS patients in the acute phase of the disease . This study shows a novel methodology for the diagnosis of SFTS , which may provide helpful information for the effective control of the disease .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"rift",
"valley",
"fever",
"virus",
"enzyme-linked",
"immunoassays",
"immune",
"physiology",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"immunology",
"microbiology",
"viruses",
"rna",
"viruses",
"signs",
"and",
"symptoms",
"antibodies",
"immunologic",
"techniques",
"bunyaviruses",
"research",
"and",
"analysis",
"methods",
"immune",
"system",
"proteins",
"thrombocytopenia",
"proteins",
"medical",
"microbiology",
"immunoassays",
"microbial",
"pathogens",
"recombinant",
"proteins",
"hematology",
"biochemistry",
"immunohistochemistry",
"techniques",
"diagnostic",
"medicine",
"fevers",
"physiology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"histochemistry",
"and",
"cytochemistry",
"techniques",
"serum",
"proteins",
"organisms"
] |
2016
|
Severe Fever with Thrombocytopenia Syndrome Virus Antigen Detection Using Monoclonal Antibodies to the Nucleocapsid Protein
|
Numerous studies of human populations in Europe and Asia have revealed a concordance between their extant genetic structure and the prevailing regional pattern of geography and language . For native South Americans , however , such evidence has been lacking so far . Therefore , we examined the relationship between Y-chromosomal genotype on the one hand , and male geographic origin and linguistic affiliation on the other , in the largest study of South American natives to date in terms of sampled individuals and populations . A total of 1 , 011 individuals , representing 50 tribal populations from 81 settlements , were genotyped for up to 17 short tandem repeat ( STR ) markers and 16 single nucleotide polymorphisms ( Y-SNPs ) , the latter resolving phylogenetic lineages Q and C . Virtually no structure became apparent for the extant Y-chromosomal genetic variation of South American males that could sensibly be related to their inter-tribal geographic and linguistic relationships . This continent-wide decoupling is consistent with a rapid peopling of the continent followed by long periods of isolation in small groups . Furthermore , for the first time , we identified a distinct geographical cluster of Y-SNP lineages C-M217 ( C3* ) in South America . Such haplotypes are virtually absent from North and Central America , but occur at high frequency in Asia . Together with the locally confined Y-STR autocorrelation observed in our study as a whole , the available data therefore suggest a late introduction of C3* into South America no more than 6 , 000 years ago , perhaps via coastal or trans-Pacific routes . Extensive simulations revealed that the observed lack of haplogroup C3* among extant North and Central American natives is only compatible with low levels of migration between the ancestor populations of C3* carriers and non-carriers . In summary , our data highlight the fact that a pronounced correlation between genetic and geographic/cultural structure can only be expected under very specific conditions , most of which are likely not to have been met by the ancestors of native South Americans .
The way a certain habitat is first colonized by humans creates a primordial pattern of genetic variation that is subsequently attenuated by various demographic processes , including migration , population bottlenecks , fissions and fusions . A popular ramification of this paradigm is that most changes of the original genetic ‘make-up’ of a particular region follow trajectories established by geography and language [1] because , in addition to climatic conditions , the latter are the main conductors of gene flow . As a consequence , the type and degree of correlation observed between the genetic structure of an extant population on the one hand , and its linguistic and geographical structure on the other , should provide valuable information about the history of that population . Dissenting processes such as the adoption of a new language without substantial gene flow into the adopting population , for example , by ‘elite dominance’ are usually conceived as exceptions to the rule [2] . Following this line of arguments , any concordance between genetic , linguistic and geographic data should be indicative of steady settlement , isolation by distance and constant population growth whereas discordances suggest abrupt demographic changes such as major contractions or relocations [3] . A plethora of culture anthropologic and population genetic studies have corroborated the above viewpoint for various geographical regions , with Europe providing a most illustrative example . Thus , genetic markers of different time depths , including rapidly mutating short tandem repeats ( STRs ) and slowly mutating single nucleotide polymorphisms ( SNPs ) , have been employed to trace the settlement of Europe over the last 40 , 000 years , from the Middle Paleolithic through the Neolithic into historical times [4]–[8] . Studies of Y-chromosomal markers , in particular , revealed a substantial correlation between genetic and geographic as well as linguistic patterns for parts of Europe and Asia that did not however become similarly apparent with mitochondrial or autosomal markers [5] , [7] , [9]–[13] . Hence , an analysis of molecular variance ( AMOVA ) of Y-STRs identified clearly distinguishable sub-clusters of western and eastern European Y chromosomes that were largely congruent with the Slavic and Romance language domains [7] , [9] , [11] , and Y-chromosomal genetic discontinuities throughout the continent were found to coincide with linguistic boundaries [7] , [14]–[16] . Other in-depth studies in Asia [17]–[20] , Africa [21] , [22] , Melanesia [23]–[27] and globally [28] , [29] also suggested that paternal ( i . e . Y-chromosomal ) but not maternal ( i . e . mtDNA ) lineage formation was strongly related to language dispersal [30] . These findings notwithstanding , it must be noted that at least some instances of male-mediated gene flow over major linguistic barriers have been inferred as well , for example , in Iberia [31] and in the Balkans [32] . Therefore , the correlation between genetics , language and geography may vary , particularly across a highly differentiated region such as Europe , depending upon the local effective population size and the time-depth of the DNA markers used . South America was the last continent colonized permanently by modern humans . The popular “Out of Beringia” model [33] purports that , after the initial peopling of Beringia by North Asians >16 , 000 years ago , this proto-American population expanded , migrated southward along the coast and through an ice-free interior corridor , and differentiated either into a few large groups corresponding to the Eskimo-Aleut , Na-Dene and Amerindian linguistic families [34] or , in the view of many scholars , into hundreds of independent Paleo-Indian groups . On their way south , human populations may have experienced their first notable expansion in Central America [35] . However , the narrow Isthmus of Panama is a likely bottleneck that should have allowed only a comparatively small number of nomadic hunters , fishermen and gatherers to enter the northern Andean region via the Cauca and Magdalena rivers [36] . Another major population expansion is thought to have occurred in the Amazon basin from which the diverse habitats of the continent were then rapidly colonized [35] . As yet , this scenario could not be confirmed by any archeological evidence but is instead based mainly upon geographic reasoning as well as linguistic and molecular genetic data . Today , most researchers agree that the initial human settlement of the Americas was a relatively swift process , a view spawned mainly by the dating of archeological findings from Paleo-Indian populations in North and South America . In particular , the Manis site in North America [37] has an estimated age of 13 , 860–13 , 765 calendar years before present ( YBP ) while the Monte Verde site in Chile dates back to an estimated 14 , 220–13 , 980 YBP [38] . Additional South American excavation sites in the north ( Taima-Taima , Falcon , Venezuela ) , east ( Lagoa Santa , Minas Gerais , Brazil ) , west ( Pikimachay , Ayacucho Valley , Peru ) and south ( Los Toldos , Santa Cruz , Argentina ) as well as in Amazonia ( Pedra Pintada , Pará , Brazil ) also indicate that human populations of high technological standard were scattered all over the continent by 12 , 000 YBP and were contemporary with the North American Clovis culture [39] , [40] . As an alternative to terrestrial expansion , it has been suggested that some early settlers entered South America by sea [41] . Such a scenario would point towards trans-Pacific links with Polynesia [42] and the ancient Jōmon culture of Kyushu ( Japan ) [43] , but no genetic evidence for a large-scale pre-Columbian immigration into the south other than via Central America has been found so far . In summary , it thus appears as if no consistent model for the peopling of South America has been established yet . Greenberg's influential tripartite classification of the American language phyla [34] , [44] , in which the Amerind group reflects a single wave of colonization and ties together all extant native American languages , is no longer considered adequate [45]–[47] . Many historical linguists even doubt whether a credible relationship can be established at all between the long-diverged Amerind languages [48] . Moreover , Hunley and coworkers [49] highlighted an apparent incongruence between the American mtDNA pool and popular linguistic classifications . In general , molecular genetic data ( i . e . blood groups , HLA , mtDNA , Y-chromosomal and autosomal STRs and SNPs ) poorly fit the available linguistic , archeological and paleo-anthropological data from South America , and different types of genetic markers even gave contradictory results [50] . Undoubtedly , the hegemony of the pre-Columbian Inca empire in the 15th Century and the contact with Europeans soon thereafter erased much of the previous relationship between genes , geography and culture through the elimination of some South American populations , and through the reduction of genetic diversity in others [39] , [51] . Most population genetic studies so far have confirmed an Asian origin of all native American populations and provided clues as to the timing of the colonization and subsequent differentiation processes [52]–[54] . Reduced genetic diversity and strong differentiation prevail among South American natives , and do so more markedly in the east than in the west of the continent [50] , [55] , [56] . Strong genetic drift must have accompanied the dispersal process . However , while most researchers assume that , before the arrival of the Europeans , most communities in South America were rather small [50] , [57] , new evidence suggests that not only the Andes but also the Amazon region were home to large integrated populations between 1250 and 1600 AD , and that the small and isolated contemporary groups are only remnants of these [58] . Y-chromosomal studies of the peopling of South America are abundant because South American natives possess a founder Y chromosome defined by Y-SNP M3 , a haplogroup commonly classified as Q1a3a [57] , [59]–[61] . Some carriers of Q1a3a have also been found in Siberia , probably reflecting reverse gene flow from Alaska into Asia . On the other hand , sub-lineages downstream of Q1a3a and different members of upstream paragroup Q1a3* ( defined by Y-SNP M346 ) have seldom been observed in South America [62]–[65] . In addition to haplogroup Q , only Asian haplogroup C is known also to occur in American natives where its presence is however restricted to sub-clades C3b and C3* . Lineage C3b , defined by Y-SNP P39 , is apparently confined to North America [65]–[67] . In contrast , more ancient lineage C3* has only been detected unquestionably in four chromosomes from the northwest of South America as yet [68] and , more recently , also in a Tlingit individual from Southeast Alaska of self-reported indigenous ancestry [67] . No C haplogroups have been found in Central American natives as yet [66] , [69] . In the present study , we examined whether the view that human genetic variation predominantly follows geographic and linguistic trajectories also holds for South America . To this end , we initiated the largest population genetic study of Y-chromosomal genetic variation in South American natives to date . We also report upon the discovery of a distinct cluster of C3* carriers in different albeit closely neighboring tribal and linguistic groups from Ecuador . At the same time as supporting our conclusion that the extant genetic structure of South American native populations , if any , is largely decoupled from the continent-wide linguistic and geographic relationships , this finding lends credit to the possibility of coastal or trans-Pacific migration that left no traces in North and Central America .
A total of 1011 DNA samples from native South American males , ascertained at 81 different sites , were available for analysis . The sample size per site ranged from 1 to 57 , with a median of 8 and an interquartile range ( IQR ) of 2 to 17 . Sampling sites were located between −78 . 18° and −46 . 17° longitude , and between −46 . 08° and 12 . 07° latitude . Places of origin ranged from the Guajira peninsula of Venezuela in the north and the Maranhão province of Brazil in the east to the Chubut province in Argentina in the south and the Pastaza province in Ecuador in the west ( Figure 1 ) . Additional information including the haplogroup , ethnic and linguistic assignment of each sample as well as site-specific measures of Y-STR haplotype variation is provided in Table S1 . Our samples included members of 50 different ethnic groups , with the number of males per group ranging from 1 for the Borjano , Embera , Movima , and Pastos to 90 ( 9% of the total ) for the Toba ( median number per group: 14 . 5 , IQR: 6 . 3 to 26 . 5 ) . Some 26 different language groups were present . Predominant groups included Mataco-Guaicuruan ( n = 200; 20% ) , Tupi-Guaraní ( n = 158; 16% ) and Quechua speakers ( n = 97; 10% ) . In contrast , Movima and Pano-Tacana were each spoken by a single individual only . Following Ruhlen [44] , all samples were assigned to one of the four Amerindian language classes , namely Equatorial-Tucanoan ( n = 399; 40% ) , Ge-Pano-Carib ( n = 327; 32% ) , Andean ( n = 166; 16% ) and Chibchan-Paezan ( n = 79; 8% ) , except for 40 Waorani males ( 4% ) from two villages in the Pastaza province of Ecuador who spoke a Wao Tiriro isolate language . For the geographic distribution of language groups and classes , see Figure S1 . Five different Y-SNP haplogroups were present in the sample ( Figure 1 ) , with the large majority ( n = 927; 92% ) being Q-M3 ( Q1a3a ) . This haplogroup was found throughout South America . Another 58 samples ( 6% ) carried Q-M346 ( Q1a3 ) , a haplogroup previously reported from Argentina , Chile and Bolivia , and sporadically from North America [62] and Siberia [70] . Q1a3 carriers in our study originated from the north of South America and from Central Bolivia . The only sublineage of Q1a3a present in our study was Q-M19 ( Q1a3a1 ) , which was observed in six samples from the Argentinian Chaco province [71] . This sublineage has only been reported before from the Ticuna ( Upper Amazon ) and the Wayuu ( Caribbean coast ) [57] , [72] . Another six samples from a single Gaviao village belonged to a Q1a3a-del haplogroup . This haplotype is characterized by a large deletion at the proximal azoospermia factor ( AZFa ) region that is responsible for the failure to amplify five of the STRs included in the YFiler kit , namely DYS389 I and II , DYS437 , DYS439 and DYS635 . However , other markers from distal AZFa could be analyzed in these samples , namely DYS438 , M194 , M242 and M199 . Since both the USP9Y gene and the DBY gene are also located in the distal portion of AZFa [73] , the deletion is not expected to affect fertility , which would be compatible with a spread of Q1a3a-del in South America . We did not observe other sublineages downstream of Q1a3a , such as Q-M194 ( Q1a3a2 ) or Q-M199 ( Q1a3a3 ) [65] . We did not test for the presence of recently published branch Q-SA01 ( Q1a3a4 ) which had been detected before in Peru and Bolivia [64] . Notably , we found 14 individuals , all from Ecuadorian ethnic groups , who carried a C3* ( xC3a-f ) haplogroup . While three of these were Waorani and belonged to the Wao-Tiriro language isolate , the other 11 males belonged to the Kichwa speakers from the Pastaza province and lived close to the Waorani settlements . No carriers of C-P39 ( C3b ) were detected , a Y-SNP haplogroup that is apparently confined to North America [66] , [74] . We analyzed two sets of Y-STRs , namely a small set ( DYS19 , DYS389I , DYS389II , DYS390 , DYS391 , DYS392 and DYS393 ) and a large set ( additionally including DYS437 , DYS438 , DYS439 , DYS448 , DYS456 , DYS458 , DYS635 and YGATAH4 ) . Since both sets yielded similar results , we will only report upon the small marker set unless indicated otherwise . The haplotype diversity for the small marker set ranged from 0 . 0 to 1 . 0 per site , with a median of 0 . 88 and an IQR of 0 . 24 to 0 . 98 , but these calculations were partially based upon small sample sizes and should therefore be considered with some caution . Bearing in mind the uneven and sometimes sparse distribution of sampling sites , we refrained from performing any geographical interpolation of haplotype diversity . In an analysis of molecular variance ( AMOVA ) , variation between sampling sites was found to account for 21% of the total Y-STR genetic variation with the large marker set , and for 28% with the small marker set . These figures increased to 26% and 32% , respectively , when only Q1a3a carriers were analyzed . Next , sampling sites were combined into geographic clusters and analyzed jointly ( Figure S2 , Table S1 ) . Clusters were defined manually taking altitude ( high versus low ) , barriers ( e . g . north versus south of the Amazon River ) and distance between sites into account . We considered two types of geographic clusters . Fine clustering ( A ) assigned each sampling site to one of six clusters , namely A1 ( ‘Patagonia’; 6% of samples ) , A2 ( ‘Central South America’; 33% ) , A3 ( ‘El Beni/Rondonia’; 9% ) , A4 ( ‘Northwest South America’; 24% ) , A5 ( ‘Northern Amazon’; 8% ) and A6 ( ‘Southern Amazon’; 20% ) . Coarse clustering ( B ) resulted in three clusters , namely B1 ( ‘Highland’; 12% ) , B2 ( ‘North Lowland’; 57% ) and B3 ( ‘South Lowland’; 31% ) . Only a minor fraction of the total Y-STR variation of the small marker set was explained by differences between type A clusters ( 4% in a one-factor model and 1% in a two-factor model also including sampling site ) or between type B clusters ( 3% and 1% , respectively ) . Virtually no systematic inter-cluster Y-STR variation was observed with the large marker set . Restricting the analysis to Q1a3a samples increased the above figures to 9% and 6% for type A clusters , respectively , and left them virtually unchanged for type B clusters . Multidimensional scaling ( MDS ) analysis of pair-wise RST between sampling sites also gave no indication of systematic population differences ( Figure 2A and 2B ) . The first two MDS components ( C1 and C2 ) explained 9 . 2% and 6 . 7% of the RST-defined genetic variation , respectively , for type A clustering , and 8 . 4% and 7 . 0% for type B clustering , respectively . Next , sampling sites were grouped for AMOVA and MDS according to their local Y-SNP haplogroups . The 18 sites harboring more than one Y-SNP haplogroup were split into homogeneous haplogroup-defined subgroups . As was to be expected , grouping by Y-SNP haplogroup explained a large proportion ( ∼85% ) of the total Y-STR variation in the AMOVA . However , RST-based MDS ( Figure 2C ) indicated only a modest separation between sites of C3* , Q1a3a1 and Q1a3a-del carriers , but not between sites of Q1a3 and Q1a3a carriers . The first two MDS components explained 6 . 8% and 5 . 3% of the RST-defined genetic variation , respectively . One possible explanation for these apparently discrepant results may be that >90% of chromosomes belonged to haplogroup Q1a3a . Thus , the bulk of genetic variation occurred within one group and was therefore well explicable by group affiliation in an AMOVA but cannot , as a matter of principle , exhibit much structure in a sampling site-wise MDS . To forestall concerns that the observed lack of genetic structure among native South American males was due to our sampling scheme , we performed an AMOVA of 100 random subsamples taken from the European portion of the Y Chromosome Haplotype Reference Database ( YHRD ) , adopting the same number of sampling sites and the same distribution of sample size per site as in the South American data ( see Materials and Methods ) . Over the 100 subsamples , geographic cluster affiliation ( Figure S8 ) was found to account for 5 . 5% to 12 . 8% of the genetic variation in a two-factor analysis , with a median of 8 . 8% and an IQR of 8 . 0% to 9 . 6% . These figures were very similar to the proportion explained when all 214 European sampling sites were taken into consideration ( mean: 8 . 9% , standard error: 0 . 2%; see Materials and Methods for details of the estimation procedure ) . Spatial autocorrelation analysis ( SAA ) revealed a substantial positive correlation between Y-STR haplotypes only for sites that were at most a few hundred kilometers apart ( Figure 3 ) . With the exception of A1 cluster ‘Patagonia’ , SAA confined to geographically defined clusters revealed a similar pattern , most convincingly for clusters B2 ‘North Lowland’ , A6 ‘Southern Amazon’ and A4 ‘Northwest South America’ ( Figure S3 ) . When the SAA was confined to Q1a3a carriers , however , no significant autocorrelation of Y-STR haplotypes was observed apart from a modest correlation within sites , i . e . at zero distance ( Figure 3 ) . After removing samples that carried at least one deletion or duplication , median-joining network analyses were performed for chromosomes with haplogroup Q or C3* . The resulting haplogroup Q network was characterized by star-like offshoots and little structure . No correlation was observed between any of the linguistic or geographic classifications and Y-STR haplotypes belonging to haplogroup Q ( Figures S4 , S5 , S5 , S7 ) . The C3* network structure will be alluded to in more detail below . We also assessed the degree to which Y-STR variation correlated with spoken language . To this end , we considered two types of linguistic categories , namely the 26 narrowly defined language groups and the four more broadly defined language classes of Ruhlen [44] ( Andean , Chibchan-Paezan , Equatorial-Tucanoan and Ge-Pano-Carib ) plus a language isolate for 40 Waorani samples ( Table S1 ) . A highly significant association was observed between haplogroup and both language class ( Cramer's V = 0 . 20 , p<10−8 , Table 1 ) and language group ( V = 0 . 41 , p<10−8 , Table S3 ) . In view of their uncertain linguistic relationship with the remaining samples , the Waorani haplotypes were excluded from subsequent AMOVA and MDS analysis . Sampling sites comprising more than one language group ( n = 7 ) or class ( n = 2 ) were split into homogeneous subgroups . Language class was not found in an AMOVA to explain much of the Y-STR genetic variation ( <0 . 5% for both marker sets ) . In contrast , differences between the more narrowly defined language groups explained 12% and 8% of the variation for the large and the small marker set , respectively . MDS analysis did not indicate any strong genetic differences between language classes , except for a weak clustering of sites belonging to the Chibchan-Paezan class ( Figure 2D ) . The first two MDS components explained 8 . 0% and 6 . 6% of the RST-defined genetic variation . We observed a considerable number of C3* haplogroup carriers in our study ( n = 14 ) . These were confined to the northwest where C3* was found at substantial frequency in two culturally very distinct native groups from Ecuador , namely the Kichwa ( 26% ) and the Waorani ( 7 . 5% ) . The C3* haplogroup was absent from all other samples . Previously published data [20] , [25] , [66] , [67] , [70] , [75]–[77] indicate that C3* occurs at a high frequency throughout continental East Asia ( Figure 4 ) and is most prevalent in Kamchatka ( 38% in Koryaks ) and in Outer and Inner Mongolia ( 36% and 38% , respectively ) . At the Pacific coast , the average C3* frequency is higher in Korea ( 10% ) than in Japan ( 3% ) , with the notable exception of 15% for the Ainu from Hokkaido , representing the aboriginal people of Japan . In striking contrast , this haplogroup is apparently absent from the whole of North and Central America , with the exception of a single C3* carrier of self-reported indigenous ancestry from Southeast Alaska [67] , as well as from Melanesia east of Borneo and Polynesia . We performed a median-joining network analysis of the Y-STR haplotypes of the 14 C3* carriers in our study and of 396 carriers identified in previous reports [66] , [67] , [70] , [78]–[83] . In the resulting network ( Figure 5 ) , native South American C3* carriers from the present study ( marked in red in Figure 5 ) belonged to separate and rather distant clusters at the periphery of the network , suggesting that the time of the last contact between these two groups predated the time of the initial colonization of the Americas . The Alaskan Tlingit C3* haplotype H166 ( marked in pink ) is between four and five steps away from the Ecuadorians , but is connected to the same frequent haplotype , H21 via a quasi-median . The most frequent Ecuadorian C3* chromosome H7 ( occurring eight times in the Kichwa ) shared an identical 8-locus haplotype with two Koryak samples from Kamchatka . The other three Kichwa haplotypes were related to this prevalent type by a one-step mutation at DYS391 ( H162; occurring twice ) and by two steps at DYS391 and DYS439 ( H163; occurring once ) . This cluster was connected to the core of the network via a quasi-median , thereby highlighting its substantial distance to common Asian types . The three identical Waorani haplotypes differed from three identical Mongolians by a single step mutation only and grouped together with these in haplotype H22 for the small marker set plus DYS439 . The putative C3* haplotypes of the Colombian Wayuu [66] were only distantly related to the Ecuadorians ( H165 ) . The median TMRCA estimate for the haplogroup C3* chromosomes in our data was between 168 and 206 generations ( 1st quartile: 113 to 135 generations; 3rd quartile: 252 to 317 generations ) , depending upon the employed population growth model and prior population size . Assuming a generation time of 30 years , this would imply an MRCA for the 14 C3* chromosomes at 5040 to 6180 YBP . Similarly , the posterior mean of population parameters Nc and Na fell short of their prior counterparts by a factor of 5 to 25 , judged by the respective median ( Table 2 ) . Inclusion of the single Tlingit C3* chromosome from Alaska [67] increased the TMRCA estimate to a median between 202 and 249 generations ( 1st quartile: 140 to 168 generations; 3rd quartile: 294 to 734 generations ) . The model-specific change in median TMRCA estimate ranged from 34 to 43 , corresponding to a time difference of 1020 to 1290 years . If haplogroups Q and C3* both entered the American continent from Asia at the same time 15 , 000 YBP , then C3* would have been expected to be more widespread than has been reported so far . We employed three simplified models of population divergence ( see Materials and Methods for details ) to ascertain which migration rates likely prevailed between the subpopulations preceding C3* carriers ( designated SA/C+ ) and non-carriers ( SA/C− and NA/C− , see Figures S10 , S11 , S12 ) . When considering South America alone ( scenario SA ) , the median of the migration rate into population SA/C− was 0 . 023 ( inter-quartile range 0 . 001–0 . 036 ) , while that into SA/C+ was 0 . 092 ( 0 . 072–0 . 147 ) . Similar results were obtained when South American and North American non-carriers were collapsed into one population ( scenario AA ) , with a median migration rate into SA/C− and NA/C− of 0 . 027 ( 0 . 010–0 . 065 ) , and into SA/C+ of 0 . 112 ( 0 . 073–0 . 154 ) . Assuming a ten-fold larger effective population size , albeit in a smaller number of simulations , ( see Figures S13 and S14 ) led to very similar results , with a median migration rate into SA/C− ( scenario SA-10x ) of 0 . 034 ( 0 . 024–0 . 045 ) and into SA/C− and NA/C− ( scenario AA-10x ) of 0 . 011 ( 0 . 004–0 . 018 ) . Consideration of three populations ( scenario BA ) yielded median migration rates from the other two populations combined of 0 . 079 ( 0 . 075–0 . 121 ) into SA/C− , 0 . 058 ( 0 . 019–0 . 107 ) into NA/C− , and 0 . 140 ( 0 . 095–0 . 157 ) into SA/C+ . The median migration rate into the common ancestral population of SA/C− and NA/C− was 0 . 117 ( 0 . 067–0 . 164 ) . It should be remembered , however , that all migration rates into one of the non-carrier ancestral populations included migration from the other non-carriers , which explains why these rates are substantially higher than with the two-population models .
The presence of Y-SNP haplogroup C-M217 ( C3* ) in the northwest of South America , and its concomitant absence from most of North and Central America , are intriguing in view of the high prevalence of this haplogroup in Central , East and Northeast Asia . Given the large population size of native North Americans , it appears unlikely that the early settlers of America carried C3* with them , and that the haplogroup got lost by genetic drift in the north , but not in the south . In fact , the locally confined occurrence of C3* in South America would require migration rates into the ancestral C3* and Q carrier populations that are so low ( most likely only 2 . 5% out of the C3* carrier population ) that they are hardly compatible with a long period of joint immigration from Asia . Instead , an independent introduction of C3* into South America appears plausible not the least because it would be consistent with the observed pattern of locally confined Y-STR autocorrelation as well . This view is further supported by the comparatively recent coalescence of the 14 C3* haplotypes from the present study , which appears to have occurred some 200 generations ago , corresponding to 6000 years . The above notwithstanding , inclusion of an isolated Tlingit C3* haplotype found in Alaska prolonged the coalescence time estimate by no more that approximately 40 generations which means that a North American origin of the Ecuadorian C3* haplotypes , albeit less likely prima facie , cannot be ruled out . In view of the above , two scenarios for the introduction of C3* into Ecuador seem credible: ( i ) one or more late migratory waves that quickly passed North and Central America without leaving a trace of C3* , and ( ii ) long-distance contact with East Asia . As regards the second scenario , there appears to be at least some archaeological evidence for a pre-Columbian contact between East Asia and South America [43] . In particular , the similarity of ceramic artifacts found in both regions led to the hypothesis of a trans-Pacific connection between the middle Jōmon culture of Kyushu ( Japan ) and the littoral Valdivia culture in Ecuador at 4400–3300 BC . In view of the close proximity of the spotty C3* cluster to the Valdivia site , which was considered at the time to represent the earliest pottery in the New World [40] , it may well be that C3* was introduced into the northwest of South America from East Asia by sea , either along the American west coast or across the Pacific ( with some help by major currents ) . The considerable differences between the extant Y-STR haplotypes of Ecuadorian and Asian C3* carriers would clearly be explicable in terms of their long divergence time . The differences between C3* chromosomes carried by different ethnic groups in Ecuador , on the other hand , highlight that population splits followed by limited gene flow are characteristic of the genetic structure of South American natives [88] . Of the 14 C3* haplotypes observed , 11 belonged to Lowland Kichwa today living in geographic proximity or even in the same villages as the Waorani of which three men from different families carried identical C3* haplotypes . It is important to realize that the Waorani , who were known for their extreme ferocity against invaders , were the only human inhabitants of a region of approximately 20 , 000 km2 east of the Andes between the Napo and Curaray rivers before their first peaceful contact with outsiders in 1958 [89] . A post-contact introduction of the C3* haplogroup from a Kichwa ancestor to the Waorani families can be excluded according to the genealogical record . The difference of 10 to 16 mutational steps between the Waorani haplotype and the Kichwa cluster comprising four more closely related haplotypes ( see Table S5 ) corroborates our view that even geographically neighboring ethnic groups survived for a long time in isolation from each other . In summary , our study revealed that the Y-chromosomal genetic variation of South American natives lacks a clear structure that could be related to the continent-wide geographic and linguistic relationship , suggesting a history of rapid peopling and subsequent evolution in small groups . Moreover , it appears unlikely that the South American natives are descendants of a single terrestrial wave of migration . Instead of being confined to a major founding lineage of the Q branch of the human phylogeny , as has been widely held to be the case in the past [60] , the continent hosts other Asian haplogroups as well ( e . g . those belonging to the basal C3* clade ) . Further characterization of their distribution is likely to provide new insights into the demographic history of South America .
The present study complies with the ethical principles of the 2000 Helsinki Declaration of the World Medical Association . The current study was approved by the institutional review board of the Institute of Legal Medicine and Forensic Sciences Berlin under protocol authorization number 11-2010/02 . Buccal swabs , liquid saliva and capillary blood spotted on FTA cards were obtained from 1452 South American males . Of these , 441 individuals carried Y-SNP haplogroups other than Q or C ( including clades B , E , G , I , J , R , and T as well as subgroups within these clades ) . Such samples were considered as being of post-Columbian European or African origin and were excluded from further analysis . Typical East Asian or Oceanian lineages other than C ( e . g . D , O or M ) were not detected . Only a single Y chromosome with haplogroup R1a , common in many parts of Eurasia , was observed and was excluded from further analysis because of the impossibility to distinguish a pre- from a post-Columbian origin . The remaining 1011 males were deemed of native American ancestry adopting the widely accepted view that all Paleoindian ancestors originated from East Asia [90] . These individuals were ascertained at 81 sampling sites in seven South American states , namely Argentina , Bolivia , Brazil , Colombia , Ecuador , Peru and Venezuela ( Figure 1 ) . Samples were gathered by the authors and their collaborators for research purposes only . All samples were fully anonymized . Written informed consent was obtained from all participants prior to the study . Data collected in the field included tribal affiliation , population size , geographic location of the home settlement , language spoken and patrilineal relationships . The latter information was used to exclude likely close relatives and individuals of mixed ancestry . Linguistic group classifications were based upon Campbell's scheme [46] whereas language classes ( Andean , Chibchan-Paezan , Equatorial-Tucanoan , Ge-Pano-Carib ) were assigned following Ruhlen [44] ( see Table S1 ) . All wet lab analyses were performed at the authors' institutions in Porto ( Portugal ) , Buenos Aires ( Argentina ) , Belém ( Brazil ) and Berlin ( Germany ) . DNA was extracted following standard protocols using either Chelex or phenol-chloroform-based methods or magnetic bead extraction with QIAsymphony ( Qiagen , GmbH Hilden , Germany ) . A total of 919 samples ( 91% ) were typed for the 17 Y-STRs of the AmpFlSTR Yfiler kit , following the manufacturer's instructions ( AmpFlSTR Yfiler PCR Amplification kit , Applied Biosystems , Foster City , USA ) . Although duplicated marker DYS385ab was typed in the whole sample , the respective genotypes were disregarded in all subsequent analyses due to ambiguous allele assignment . The remaining 15 markers will henceforth be referred to as the 'large marker set' ( DYS19 , DYS389I/II , DYS390 , DYS391 , DYS392 , DYS393 , DYS437 , DYS438 , DYS439 , DYS448 , DYS456 , DYS458 , DYS635 and YGATA H4 ) . Complete genotype information on these markers was available for 874 samples ( 86% of the total ) . Another 92 older samples ( 9% ) could not be retyped for the large marker set for logistic reasons . They had been characterized before in Buenos Aires and Belém for the so-called ‘minimal haplotype’ , either by two multiplex PCRs [91] or according to protocols published by Palha and coworkers [92] , [93] . The corresponding seven markers ( DYS19 , DYS389I/II , DYS390 , DYS391 , DYS392 and DYS393 ) will henceforth be referred to as the ‘small marker set’ here . PCR product separation and detection was carried out on different ABI genetic analyzers ( Applied Biosystems , Foster City , USA ) including ABI310 , ABI3100 and ABI3130 machines . Alleles were identified by means of the reference ladder provided with each kit and following recommendations by the DNA Commission of the International Society of Forensic Genetics ( ISFG ) [94] . In the present study , allele designations for DYS389I and DYS389II refer to the repeat numbers at individual loci rather than the repeat numbers revealed by the multiplex genotyping method . Comprehensive marker genotype information is provided in Table S5 . All samples were subjected to a first round of Y-SNP analysis using mainly the ABI PRISM SNaPshot system ( Applied Biosystems , Foster City , USA ) to genotype a panel of phylogenetic markers that define the most frequent haplogroups in South America ( M42 , M207 , M242 , M168 , M3 , M145 , M174 , M213 , RPS4Y711 , M45 , P170 and M9 ) [68] , [95] . Those 441 samples belonging to Y-SNP haplogroups other than Q and C were excluded from further analysis ( see above ) . In the Porto laboratory , all remaining samples were typed by the SNaPshot multiplex reaction ( ‘multiplex Q’ ) that allows genotyping of Y-SNPs M242 , P36 . 2 , M346 , M3 , M19 , M194 and M199 . In Berlin , a similar SNaPshot multiplex ( ‘SA SpecQ’ ) was used to subtype haplogroup Q for Y-SNPs M19 , M194 , P292 , M3 and M199 [68] . Since haplogroup C occurs in both Asia and North America ( in the form of C3b* , defined by Y-SNP P39 ) , we assumed that some C lineages were autochthonous as well . Therefore , another multiplex ( ‘SA SpecC’ ) was established in Berlin to allow further subtyping of haplogroup C ( mainly as subclade C3* , defined by Y-SNP M217 ) . The full SNaPshot panel includes M407 , M48 , P53 . 1 , M217 , P62 , RPS4Y711 , M93 , M86 and P39 . Instead of the SNaPshot assay , the Buenos Aires institute used real time PCR followed by High Resolution Melting Analysis for Y-SNP genotyping [96] . The Y-SNP haplogroup nomenclature used here follows the recommendations by the Y Chromosome Consortium [65] . The data were checked for invalid alleles , duplicates and missing genotypes using in-house scripts for the R software v2 . 14 . 0 [97] . See Figure S9 for an illustration of the phylogenies of haplogroups C3* and Q* . All genotypes generated in the present study are provided in Table S5 and were logged in the online Y Chromosome Haplotype Reference Database ( YHRD ) , maintained by the Institute of Legal Medicine and Forensic Sciences , Charité - University Medicine , Berlin , Germany ( www . yhrd . org ) . For some analyses , we also drew upon external data resources . Thus , information on the C3* haplogroup frequency in Asia and the Americas was obtained from 6 , 562 haplotypes reported in 109 different population genetic studies [20] , [24] , [66] , [69] , [70] , [75]–[77] . In fact , 316 of these Y chromosomes ( 4 . 8% ) were reported to carry haplogroup C3* . The corresponding Y-STR haplotypes were also extracted from the literature as well as from YHRD data submissions [66] , [70] , [78]–[83] to allow comparative network analyses jointly with our own data ( see Table S6 ) . For network analyses and estimation of the time to the most recent common ancestor ( see below ) , we included the genotype data of a recently published C3* haplotype from Southeast Alaska with self-reported indigenous ancestry [67] ( sample #2 from Table S8 of the original publication ) . Allele and haplotype frequencies were estimated by counting . Gene diversity and haplotype diversity were calculated following Nei [98] , [99] . The R software v2 . 14 . 0 [97] was used for statistical analysis unless indicated otherwise , and to create graphs . The association between haplogroup and language group or class was measured by Cramér's contingency coefficient V [100] and tested for statistical significance using Fisher's exact test with 100 million simulations , as implemented in the fisher . test function of R . R software: http://www . r-project . org/ map , mapdata , mapproj , plotrix packages: http://cran . r-project . org/ NETWORK software: http://www . fluxus-engineering . com/sharenet . htm Arlequin software: http://cmpg . unibe . ch/software/arlequin35/ BATWING software: http://www . mas . ncl . ac . uk/~nijw/ popABC software: http://code . google . com/p/popabc/ YHRD: http://www . yhrd . org/ CIA World Data Bank II: http://www . evl . uic . edu/pape/data/WDB/
|
In the largest population genetic study of South Americans to date , we analyzed the Y-chromosomal makeup of more than 1 , 000 male natives . We found that the male-specific genetic variation of Native Americans lacks any clear structure that could sensibly be related to their geographic and/or linguistic relationships . This finding is consistent with a rapid initial peopling of South America , followed by long periods of isolation in small tribal groups . The observed continent-wide decoupling of geography , spoken language , and genetics contrasts strikingly with previous reports of such correlation from many parts of Europe and Asia . Moreover , we identified a cluster of Native American founding lineages of Y chromosomes , called C-M217 ( C3* ) , within a restricted area of Ecuador in North-Western South America . The same haplogroup occurs at high frequency in Central , East , and North East Asia , but is virtually absent from North ( except Alaska ) and Central America . Possible scenarios for the introduction of C-M217 ( C3* ) into Ecuador may thus include a coastal or trans-Pacific route , an idea also supported by occasional archeological evidence and the recent coalescence of the C3* haplotypes , estimated from our data to have occurred some 6 , 000 years ago .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genetic",
"mutation",
"genetics",
"molecular",
"genetics",
"population",
"genetics",
"biology",
"population",
"biology"
] |
2013
|
Continent-Wide Decoupling of Y-Chromosomal Genetic Variation from Language and Geography in Native South Americans
|
Recent advances in the identification of susceptibility genes and environmental exposures provide broad support for a post-infectious autoimmune basis for narcolepsy/hypocretin ( orexin ) deficiency . We genotyped loci associated with other autoimmune and inflammatory diseases in 1 , 886 individuals with hypocretin-deficient narcolepsy and 10 , 421 controls , all of European ancestry , using a custom genotyping array ( ImmunoChip ) . Three loci located outside the Human Leukocyte Antigen ( HLA ) region on chromosome 6 were significantly associated with disease risk . In addition to a strong signal in the T cell receptor alpha ( TRA@ ) , variants in two additional narcolepsy loci , Cathepsin H ( CTSH ) and Tumor necrosis factor ( ligand ) superfamily member 4 ( TNFSF4 , also called OX40L ) , attained genome-wide significance . These findings underline the importance of antigen presentation by HLA Class II to T cells in the pathophysiology of this autoimmune disease .
Narcolepsy is a life-long sleep disorder caused by the autoimmune-mediated loss of 70 , 000–90 , 000 hypocretin ( orexin ) -producing neurons in the hypothalamus . Prevalence is approximately 0 . 02–0 . 03% in Caucasian populations , and somewhat higher in Japanese ( 0 . 16% ) . Family and twin studies support the importance of genetic ( 10–40 fold increased risk in first degree relatives ) as well as environmental factors ( 25% concordance in identical twins ) [1] . Onset is typically around puberty and displays a seasonal pattern of incidence , with highest rates in spring and summer . Likely triggering factors are influenza A , notably the pandemic H1N1 2009 variant , and Streptococcus Pyogenes infections [2]–[5] . Unique among autoimmune diseases , the condition is almost completely associated with Human HLA DQ0602 , a heterodimeric protein encoded by the DQA1*01:02-DQB1*06:02 haplotype ( 90% versus 25% frequency in European ancestry cases and controls , respectively ) . The overwhelming effect of this haplotype on risk suggests the importance of antigen presentation by DQ0602 . As seen in other autoimmune diseases , additional HLA alleles carried in trans of this haplotype also confer modulatory effects [6] , [7] . Most notably , DQA1*01:02-DQB1*06:02 homozygosity increases predisposition by 2–4 fold . Further , DQA1 and DQB1 alleles known to heterodimerize with DQA1*01:02 or DQB1*06:02 reduce susceptibility , likely through allelic competition with DQ0602 [8] . However , non-HLA related genes play important roles . In addition to these well-established HLA class II effects , recent genome-wide association studies ( GWAS ) have identified variants in the T Cell receptor alpha locus ( TRA@ ) , on chromosome 14q11 . 2 , and in the region containing P2RY11-DNMT1 on chromosome 19p13 . 2 , as additional susceptibility loci . Finally , exome sequencing in families with a rare autosomal dominant syndrome including cerebellar ataxia , narcolepsy and deafness ( ADCA-DN ) also indicate an important role for DNMT1 in survival of hypocretin neurons [9] . Based on the recognition that considerable overlap exists in risk loci for various autoimmune diseases , the Immunochip consortium was formed to create a single nucleotide polymorphism ( SNP ) array for targeted finemapping of these loci [10] , [11] . The ImmunoChip was designed for deep replication of signals from large-scale meta-analyses in nine autoimmune diseases , and for finemapping of loci reaching genome-wide significance ( 2009 ) . Approximately 200 , 000 rare and common variants were selected to cover intervals with established genome-wide significant association to autoimmune and seronegative diseases , and at selected loci of known importance in major immune-related diseases , including the major histocompatibility ( MHC ) and KIR/LILR loci . Using this platform , we conducted a GWAS to identify genetic risk factors for narcolepsy in addition to HLA DQ0602 . We analyzed 111 , 240 high quality SNP markers of minor allele frequency ≥1% , located outside the extended HLA region on chromosome 6 , in 1 , 886 narcolepsy cases and 10 , 421 controls of European ancestry sampled across global populations including the European Union , Canada and North America ( Table 1 ) . To test for potential confounding effects of population stratification in our study cohort , we performed principal component analysis ( PCA ) of cases and controls ( Figure S1 ) . Control samples showed clear separation into distinct European countries in plots of the first and second principal components , with good overlay of case samples , and plots of observed versus expected association results showed no inflation of signal ( λ = 1 . 004; Figure 1 ) .
One previously reported , and two novel non-HLA loci surpassed genome wide significance ( gws ) P<5×10−8 in this study ( Figure 1 and Table 2 ) . The strongest association was with rs1154155 ( MAF = 0 . 15 , P = 8 . 87×10−30 OR = 1 . 72 ) in the T cell receptor ( TCR ) alpha ( TRA@ ) locus , on chromosome 14 , replicating signal previously reported using smaller samples [7] , [12] , [13] . The TCR protein is comprised of alpha and beta chains . As for immunoglobulin loci , TCR loci undergo somatic DNA recombination during T cell development , generating a large number of possible proteins specific to individual T-cell clones . T cells bearing specific recombinants are then negatively or positively selected , allowing adaptation of the immune system to past environmental history . The T cell receptor binds foreign or self-peptides presented by Class II MHC proteins ( such as the DQ alpha/beta heterodimer ) , allowing initiation and regulation of immune responses . It is thus the natural receptor of DQB0602 . The TRA@ locus was sparsely covered on the ImmunoChip ( 15 SNPs within a 1 Mb window of rs1154155 , none with r2 above 0 . 5 ) precluding fine mapping or haplotype analysis , although providing robust replication of the previously reported findings . SNP rs1154155 is located close to the J10 segment region of the locus , with linkage disequilibrium ( LD ) data suggesting the involvement of a specific J segment in the narcolepsy pathophysiology . The association with TRA@ is unique to narcolepsy , as no other autoimmune diseases have been associated with this locus . Two SNPs rs34593439 , and rs34843303 , located in intron 1 of Cathepsin H ( CTSH ) , a papain-like cysteine protease , reached gws ( MAF = 0 . 11 , P = 1 . 78×10−8 OR = 1 . 34 and MAF = 0 . 11 , P< = 2 . 79×10−8 OR = 1 . 35 , respectively ) . Another SNP located in intron 1 , rs3825932T has been previously reported to be associated with type 1 diabetes [14] , [15] . Although in close proximity , this marker is in weak LD with rs34593439 and rs34843303 ( r2 = 0 . 23 and 0 . 23 respectively ) and shows no significant association in the present sample ( P = 0 . 01 ) . The local region of LD surrounding these markers encompasses exon 1 , where 4 potentially functional polymorphisms have been identified . One of which , SNP rs2289702T ( p . Gly11Arg , MAF = 0 . 11 ) , is in tight LD with our markers ( pairwise r2 = 0 . 96 and 0 . 98 respectively , 1000genomes data , phase 1 release V3 ) , and could be the culprit behind this association . Following imputation in a 1 Mb window surrounding CTSH , SNPs rs2289702 and rs34593439 were the two most highly associated variants ( respectively ) ( Figure 2 ) . The Arg allele of rs2289702 also underlies a minor histocompatability antigen restricted by HLA-A*3101 and HLA-A*3303 , causing selective lysis of hematopoetic cells by cytotoxic lymphocytes [16] . p . Gly11Arg is located within the signal peptide sequence of CTSH , where the introduction of a highly charged arginine could affect trafficking or cleavage , as predicted by some but not all signal peptide predicting programs ( see Methods ) . Cathepsins are primarily located within the lysosomal/endosomal compartment and typically activated by low pH . These enzymes play diverse and important roles including cellular recycling of proteins , activation of selected preprohormones , antigen processing , and loading of peptides onto MHC class II proteins . Eleven family members are known . Cystatins and other endogenous inhibitors are known to regulate cathepsin activity , and the balance of these activities has been proposed to be the major selector for the repertoire of surface peptide –MHC II complexes [17] . Deficiencies in selected cathepsins impair immune cell development ( NKT cells in Cathepsin S or L deficient mice , thymocytes and T cell repertoire in cathepsin L deficient mice ) , and produce defects in immune cell effector functions ( cytotoxic T cell , neutrophil and mast cell defects in cathepsin C deficient mice; see [18] ) . Cathepsin H is somewhat unique in that it can have both exopeptidase and endopeptidase activities , depending on the presence of a bound mini-chain ( a remnant of the pro-enzyme ) within the active cleft . Although ubiquitously expressed , CTSH expression is especially high in type II pneumocytes , where it plays a key role in the maturation of lung surfactant protein B [17] , [19] . CTSH is also highly expressed in MHC class II positive immune cells such as B cells , monocytes and dendritic cells , but not T cells , notably in the presence of inflammation . For example , CTSH enzyme activity increases in parallel with proinflammatory cytokines during the development of autoimmune inflammation in a NOD mouse model of Sjögren's syndrome [20] . One hypothesis may be that decreased CTSH activity reduces antigen processing resulting in an altered repertoire presented by DQB0602 and resulting in increased risk of narcolepsy . SNPs located in the Tumor Necrosis Factor ( ligand ) Superfamily member 4 ( TNFSF4; also called OX40L or CD252 ) are strongly associated with narcolepsy . SNP rs7553711 reached gws ( MAF 0 . 29 , P = 4 . 08×10−8 and OR = 1 . 33 ) . No other SNP was more strongly associated with narcolepsy following imputation in a 1 Mb window around this locus , although additional strongly associated variants were identified ( Figure 2 ) . TNFSF4 is known to be strongly associated with systemic lupus erythematosis ( SLE ) [21] and systemic sclerosis [22] , [23] and SNPs in this region were densely represented on the ImmunoChip . Two distinct haplotypes composed of SNPs upstream of the gene confer susceptibility or resistance to SLE , whereas our most significantly associated SNP markers in narcolepsy are downstream of the gene in a separate haplotype block . The SNPs associated with SLE and narcolepsy are in weak LD , and rs844648 , an established marker of SLE , is not strongly associated with narcolepsy ( p = 0 . 016 ) . Interestingly , rs7553711 maps to a potential enhancer site ( H3K4Me1 site , UCSC browser , Layered H3K27 Track ) . The association of narcolepsy with SNPs in TNFSF4 is consistent with a primary role of antigen presentation to T cells in narcolepsy . Like CTSH , OX40L is primarily expressed in MHC Class II-positive antigen presenting cells ( e . g . dendritic and B cells ) . Optimal activation of T cells following the binding of T cell receptor- MHC class II/antigen complex requires the action of additional costimulatory factors , notably involving receptor/ligand pairs from the tumor necrosis superfamily . The interaction of two of these , OX40 receptor ( encoded by TNFRSF4 ) and OX40L ligand ( encoded by TNFSF4 ) , provides an important costimulatory signal supporting Th1 and Th2 responses , promoting expansion and survival of effector T cells and the generation of T memory cells . Although less understood , OX40/OX40L interactions also play a role in the activity and homeostasis of T regulatory cells . Signaling of this pair is tightly controlled , as OX40 is not expressed in resting T cells , only appearing approximately one day following initial activation . Similarly , OX40L is found only at sites of inflammation , first on the surface of antigen presenting cells , but later on diverse cell types including mast cells , suggesting a role distinct from T cell priming or memory cell generation . OX40-OX40L interactions are known to be involved in autoimmune disease , ( e . g . SLE ) likely acting through a disruption of tolerance . OX40 signaling within responding T cells renders them resistant to Treg- mediated suppression , and acts within the Treg cells to inhibit suppressive functions . In addition , sustained inflammatory response may result from excessive OX40-OX40L signaling and consequent increased survival of effector T-cells ( see [24] , [25] ) . Two other regions showed suggestive associations , including SNPs between MIR-552 and GJB5 on Chromosome 1p34 . 3 ( rs10915020 MAF = 0 . 84 , P = 5 . 40×10−07 , OR = 1 . 32 ) , and near ZNF365 on chromosome 10q21 . 2 ( rs10995245 MAF = 0 . 35 , P = 3 . 24×10−07 OR = 1 . 20 ) . ZNF365 is highly expressed in the brain and has been implicated in susceptibility to breast cancer , Crohn's Disease , and more recently , atopic dermatitis [26]–[28] . None of these reached genome-wide significance levels after correcting with the EMMAX [29] procedure in the current study , although nearly reaching or surpassing Bonferroni significance ( P = 4 . 5×10−07 ) ( Table S1 ) . Increased sample size and replication will be needed to confirm these loci . Our study , analyzing 1886 narcolepsy-cataplexy cases of European ancestry , is the largest collaborative cohort study of narcolepsy to date , including samples from across the United States , Canada and Europe , and representing the majority of available case samples of European ancestry . To preserve the statistical power afforded by this sample size , we elected not to split our cases into discovery and replication cohorts , and thus our study is limited by the lack of replication in an ethnically similar population . We identified two novel narcolepsy susceptibility genes , CTSH and TNFSF4 ( OX40L ) , and confirmed strong associations with HLA and TRA@ . The two new loci identified outline with striking clarity that the key pathology underlying narcolepsy likely resides in the interaction between T cells and antigen presenting cells . Although a role of antigen presentation to CD4 T cells is likely the primary susceptibility pathway for the disorder , narcolepsy was not associated with all components of this pathway as represented on the array . For example , we found no association at the p<10−4 threshold with the class II invariant chain , AEP and cathepsin B ( CTSB ) genes or , more surprisingly , with genes encoding other co-stimulatory molecules such as CD28 , cytotoxic T-lymphocyte antigen-4 ( CTLA4 ) and their cognate ligands , CD80 and CD86 ( these have been involved in many other autoimmune disorders ) ( see Table S1 ) . The present results also show limited overlap in susceptibility loci between narcolepsy and loci associated with classical autoimmune disorders , a fact that may be unsurprising based on the lack of readily identifiable autoantibodies , or other clear signs of inflammatory damage in the disease . To date , the TCR locus has only been observed in narcolepsy . Notably , we found no associations with loci widely shared among other autoimmune diseases such as interleukin genes and receptors ( IL2 , IL21 , IL12 , IL2RA , IL23R ) acting in differentiation; PTPN2 and 22 , SH2B3 and TAGAP involved in immune-cell activation and signaling; and IRF5 , TNFAIP3 involved in TNF signaling and innate immunity ( Table S1 ) . Together with findings implicating pandemic H1N1 influenza as a trigger , narcolepsy may offer a unique opportunity , furthering our understanding of how HLA class II presentation of foreign and self-antigens predispose to autoimmunity .
Informed consent in accordance with governing institutions was obtained from all subjects . The research protocol at Stanford was approved by the IRB Panel on Medical Human Subjects . Cases included in this study all met criteria for narcolepsy/hypocretin deficiency ( clear-cut cataplexy and DQB1*06:02 positive , or low cerebrospinal fluid hypocretin-1 ) . Samples included 1301 patients sourced from the Stanford Center for Narcolepsy database ( North America , and worldwide collaborators ) , and 585 samples contributed by the European narcolepsy network ( EU-NN ) . ImmunoChip typing was performed at centers in the US and in Germany . Informed consent in accordance with governing institutions was obtained . Countries of origin included: United States ( 657 ) , France ( 296 ) , Italy ( 157 ) , Germany ( 157 ) , the Netherlands ( 111 ) , Czech Republic ( 104 ) , Canada ( 101 ) , Austria ( 83 ) , Denmark ( 74 ) , Spain ( 51 ) and Norway ( 32 ) . A further 63 cases came from Argentina , Australia , Finland , Israel , Poland , Portugal , Slovakia , Switzerland and Turkey , each with fewer than 20 samples . Control genotypes were contributed through multiple immunochip consortium collaborators including 4289 samples from the United Kingdom 1958 Birth Cohort , 3609 samples from selected European countries including Italy ( 1251 ) , Netherlands ( 1173 ) , Poland ( 529 ) and Spain ( 656 ) , 980 samples from the German KORA cohort; 794 Samples from Cincinnati through CCHMC [30]; and 749 French samples ( 2 collaborators ) . Genotyping of cases was performed following Illumina's recommendation at U Virginia , USA , U of Munich , Germany , and Stanford University , Palo Alto , CA USA . NCBI build 36 ( hg18 ) mapping was used as reference . Illumina manifest file Immuno_BeadChip_1149691_B . bpm was used in the majority of cases . In cases where file Immuno_BeadChip_11419691_A was used , map positions were converted to be consistent with 1149691_B , or omitted from the analysis . Genotypes were called using Illumina GeneExpress ( Illumina GenomeStudio GenTrain2 . 0 algorithm ) , with extensive additional curation . Individuals with call rate under 98% ( 123 controls , 147 cases ) , and samples which were related ( pi hat>0 . 2 ) were excluded from further analysis . Data from all sources were merged in forward-strand format . We identified 142 , 054 high quality SNPs with call rate above 99% ( in both cases and controls separately ) , and passing HWE filtering in controls ( P>1×10−5 ) using the Plink suite of software [31] . We excluded a broad region around the HLA complex ( 7 , 893 markers at Chr 6:24 , 067–35 , 474 kb ) due to the strong LD effects with DQB1*06:02 . This region contained nearly 3000 SNPs associated with narcolepsy at GWA significant levels . We additionally excluded SNPs with minor allele frequency below 1% ( 22 , 921 SNPs ) . Finally 111 , 240 high quality SNPs of MAF≥0 . 01 ( including 91 , 804 MAF≥0 . 05 ) were selected for the analysis presented here . Principal components analysis ( PCA ) was performed to identify 162 outliers ( 133 controls , 29 cases; Golden Helix SVS , v7 ) , and those were removed . Genome wide association analysis was performed using a variance component model implemented in EMMAX [29] . The EMMAX software does not return odds ratio or adjusted allele frequency data after correction for stratification . We therefore calculated OR and MAF ( using Plink ) for our tables based on a more homogeneous subsample of 8474 cases/controls based on principal components ( EV 11 . 21<0 . 004 , EV 4 . 12<0 . 01 , see Figure S1 ) . Linkage disequilibrium ( as r2 ) values and haplotype analysis were calculated using Plink and Haploview [32] using data from our sample and/or from the 1000 genomes dataset [33] . QQ plots were generated using estlambda ( http://www . genabel . org/GenABEL/estlambda . html ) , and Manhattan and PCA plots were made using SVS software . Imputation and phasing of ImmunoChip genotypes were performed using Beagle v3 . 3 [34] against 4 European populations ( 286 individuals from CEU , TSI , GBR , IBS ) in the 1000 genomes integrated data set ( phase 1 release v3 ) within a 1 Mb window of the top hit at the CTSH and TNFSF4 loci . SNPs with an imputation R2 value≥0 . 8 ( representing reliability of imputation ) were considered in the analysis . Pairwise LD was calculated in Plink . Association P values in Figure 2 were calculated with Plink , as EMMAX would be inappropriate in this context , and therefore P values are slightly different than those presented in Table 2 . http://faculty . washington . edu/browning/beagle/beagle . html http://bochet . gcc . biostat . washington . edu/beagle/1000_Genomes . phase1_release_v3/ Sequence used: MWATLPLLCAGAWLL[G/R]VPVCGAAELCVNSLEKFHFKSWTSKHRKTYSTEEYHHRLQTFAS SignalIP: http://www . cbs . dtu . dk/services/SignalP/ Both alleles are predicted to have normal cleavage SigPred: http://bmbpcu36 . leeds . ac . uk/prot_analysis/Signal . html Predicts cleavage unlikely for Arg variant .
|
While there is now broad consensus that narcolepsy-hypocretin deficiency results from a highly specific autoimmune attack on hypocretin cells , little is understood regarding the initiation and progression of the underlying autoimmune process . We have taken advantage of a unique high-density genotyping platform ( the ImmunoChip ) designed to study variants in genes known to be important to autoimmune and inflammatory diseases . Our study of nearly 2000 narcolepsy cases compared to 10 , 000 controls underscored important roles for HLA DQB1*06:02 and the T cell receptor alpha genes and implicated two additional genes , Cathepsin H and TNFSF4/OX40L , in disease pathogenesis . These findings are particularly important , as these encoded proteins have key roles in antigen processing , presentation , and T cell response , and they suggest that specific interactions at the immunological synapse constitute the pathway to the disease . Further studies of these genes and encoded proteins may therefore reveal the mechanism leading to this highly selective and unique autoimmune disease .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] |
[
"genetics",
"immunology",
"biology",
"genetics",
"and",
"genomics"
] |
2013
|
ImmunoChip Study Implicates Antigen Presentation to T Cells in Narcolepsy
|
The serotonin 2C receptor ( 5-HT2CR ) –a key regulator of diverse neurological processes–exhibits functional variability derived from editing of its pre-mRNA by site-specific adenosine deamination ( A-to-I pre-mRNA editing ) in five distinct sites . Here we describe a statistical technique that was developed for analysis of the dependencies among the editing states of the five sites . The statistical significance of the observed correlations was estimated by comparing editing patterns in multiple individuals . For both human and rat 5-HT2CR , the editing states of the physically proximal sites A and B were found to be strongly dependent . In contrast , the editing states of sites C and D , which are also physically close , seem not to be directly dependent but instead are linked through the dependencies on sites A and B , respectively . We observed pronounced differences between the editing patterns in humans and rats: in humans site A is the key determinant of the editing state of the other sites , whereas in rats this role belongs to site B . The structure of the dependencies among the editing sites is notably simpler in rats than it is in humans implying more complex regulation of 5-HT2CR editing and , by inference , function in the human brain . Thus , exhaustive statistical analysis of the 5-HT2CR editing patterns indicates that the editing state of sites A and B is the primary determinant of the editing states of the other three sites , and hence the overall editing pattern . Taken together , these findings allow us to propose a mechanistic model of concerted action of ADAR1 and ADAR2 in 5-HT2CR editing . Statistical approach developed here can be applied to other cases of interdependencies among modification sites in RNA and proteins .
The serotonin receptor 2C ( 5-HT2CR ) is widely distributed within the central nervous system [1] , [2] , where it mediates diverse neurological processes that affect feeding behavior , sleep , sexual activity , anxiety and depression [reviewed in [3] , [4]] . The 5-HT2CR protein belongs to the G-protein-coupled receptor ( GPCR ) superfamily and potentiates multiple signal transduction pathways via several different G proteins ( Gαq/11 , Gα12/13 and Gαi ) to modulate effector molecules such as phospholipases C , D and A2 , as well as the extracellular signal-regulated kinases 1 and 2 [reviewed in [5] , [6]] . The 5-HT2CR protein exhibits functional variability that is derived from editing of its pre-mRNA by site-specific adenosine deamination ( A-to-I pre-mRNA editing ) [6] . Editing of 5-HT2CR can produce inosine from adenine at up to five closely-spaced ( within a 15 nucleotide segment ) position that have been named A , B , E ( also known as C' ) , C , and D sites . Because inosine is read as guanosine by the translational machinery , editing can alter codons for three amino acids in the second intracellular loop of the receptor [7] , [8] , a region involved in coupling with G-proteins [9] . Combinatorial editing at the five positions can generate up to 32 mRNA variants encoding 24 different receptor isoforms ( sites A and B as well as sites E and C are situated in the same codons ) . The extent of editing is inversely correlated with 5-HT2CR functional activity such that the more highly edited isoforms are less active than less extensively edited ones [reviewed in [6]] . The unedited Ile156-Asn158-Ile160 ( INI ) isoform possesses considerable constitutive and agonist-stimulated activity . In contrast , when the 5-HT2CR is edited , its coupling to G-proteins and its affinity for serotonin are drastically reduced . Specifically , experiments in heterologous expression systems have shown that , compared to the INI , the fully-edited Val156-Gly158-Val160 ( VGV ) 5HT2CR isoform ( which is edited at all five editing sits ) has a 40-fold decreased serotonergic capability to stimulate phosphoinositide hydrolysis due to reduced Gq/11-protein coupling efficiency and decreased coupling to other signaling pathways [7] , [10] . In addition , cells expressing more highly edited 5HT2CR isoforms such as VGV demonstrate considerably reduced ( or absent ) constitutive activity compared to cells expressing the non-edited INI isoform [10] . This reduction in coupling efficiency and constitutive activity derives from a difference in the ability of edited 5-HT2CR isoforms to spontaneously isomerize to the active conformation ( R* ) , a form of the receptor that efficiently interacts with G-proteins in the absence of agonist [11] . A-to-I editing is catalyzed by specific editing enzymes , RNA-specific adenosine deaminases ADAR1 and ADAR2 [reviewed in [12] , [13]] . A-to-I editing most frequently occurs in repetitive RNA sequences ( e . g . , Alu sequences ) located within introns and 5′ or 3′ untranslated regions ( UTRs ) . Although the biological significance of non-coding A-to-I RNA editing remains uncertain , the overall editing levels are higher in human compared to primate brains , thus suggesting a possible contribution of editing to the development of higher brain function [14]–[17] . Site-specific edited substrates have been identified in only a few transcripts , including 5-HT2CR mRNA , most of which are expressed in the central nervous system ( CNS ) and encode proteins involved in neurotransmission [6] . In these protein-coding transcripts , several adenosines are targeted within an imperfect RNA fold-back structure . The features that make RNA prone to site-specific editing are not fully understood , but it is thought that internal mismatches and bulges within double-stranded RNA ( dsRNA ) are important for the specificity of the ADARs [18]–[20] . Although the specificities of ADAR1 and ADAR2 toward different editing sites often overlap , some sites are edited entirely by one enzyme or the other , and the two enzymes display somewhat different preferences for nearest neighbors of the specific editing sites [19] , [21] . Experiments on mouse models with null mutations in one or both ADARs suggest that , within 5-HT2CR mRNA , the A site is predominantly edited by ADAR1 and the D site is mostly edited by ADAR2 [22]–[24] . The other sites have the potential to be edited by both ADAR1 and ADAR2 . In addition , it has been proposed that there is crosstalk between ADAR1 and ADAR2 , and therefore the relative expression of the different ADARs might ultimately influence the pattern of editing [reviewed in [6]] . The mechanism underlying the putative crosstalk is unclear , but because the five 5-HT2CR editing sites are closely spaced , editing at one site might lead to perturbation of the dsRNA structure that , in turn , would facilitate further editing at other site ( s ) . Indeed , apparent interdependence of editing among the sites has been previously reported for rodent brain [25] , [26] . Serotonin signaling , including 5-HT2CR , has been implicated in the etiology of behavioral and psychiatric disorders , and 5-HT2CR is considered an important target for pharmacologic intervention [4] . Several groups have recently reported an association between 5-HT2CR editing and suicide [27]–[31] . Specifically , our studies suggest that in the three major psychiatric diseases ( schizophrenia , bipolar disorder , and major depression ) that comprise ∼75% of suicides , suicide is associated with enhanced levels of editing ( and by inference , with lower activity ) of 5-HT2CR in the prefrontal cortex independent of the contributions of the underlying disease [30] , [31] . The biological mechanisms that contribute to higher 5-HT2CR editing ( and therefore , hypoactive receptors ) in suicide compared to non-suicide psychiatric patients remain unclear . However , it seems likely that because enhanced editing decreases 5-HT2CR activity , the resulting reduction in the receptor function might predispose some individuals to suicide by altering 5-HT2CR-dependent signal transduction in critical brain regions . Thus , altered editing mechanisms might be linked to liability for suicide , and detailed understanding of these mechanisms could facilitate the development of unique pharmacological strategies that target suicidal behavior . Alteration of the 5HT2CR function via editing has also been reported in response to spinal cord injury ( SCI ) in rats [32] . Muscle paralysis after SCI is partly caused by a loss of all brainstem-derived neurotransmitters ( including serotonin ) , which normally modulate motoneuron excitability . Murray et al . examined how motoneurons in the spinal cord of the SCI rats compensated for lost brain-derived neurotransmitters to regain excitability and found that changes in 5-HT2CR mRNA editing led to increased expression of the 5-HT2CR isoforms that are active without serotonin n [32] . Such constitutive receptor activity restored excitability of the motoneurons in the SCI rats in the absence of serotonin , helping motoneurons recover their ability to produce sustained tail muscle contractions . Accordingly , blocking constitutively active 5-HT2CR with specific drugs ( SB206553 or cyproheptadine ) largely eliminated these calcium currents and muscle spasms , providing a new rationale for antispastic drug therapy . Recently , we applied the Massively Parallel Sequencing ( MPS ) technology to quantify 5-HT2CR editing in the postmortem human brain and the rat spinal cord specimens [31] , [33] . The traditional cloning and sequencing approach [7] , [34] relies on sampling a limited population of cloned transcripts ( ∼20–100 ) , thus producing significant sampling errors that can obscure differences between experimental groups . The use of MPS , which analyzes several hundred thousand 5-HT2CR transcripts per specimen , not only allowed us to detect all 32 mRNA variants of 5-HT2CR in both species , but substantially increased precision and sensitivity in measuring 5-HT2CR editing frequencies for all these mRNA variants . Specifically , a comparison between MPS ( over 730 , 000 reads per subject ) and the traditional method ( 46 clones per subject ) , performed for the same human subjects and the same brain region , has shown that the mean coefficient of variation of the editing frequencies of all variants in the NGS analysis was approximately one-third that of the traditional method [31] . Here we use the MPS data generated in these recent studies on 5-HT2CR editing in the human and rat CNS specimens to comprehensively characterize the dependencies among the 5 different editing sites in the 5-HT2CR mRNA . The extremely high number of sequenced transcripts combined with the use of a newly developed rigorous statistical procedure allowed us to elucidate the fine structure of these interactions and compare them between the two species as well as among individuals .
5-HT2CR mRNA editing was measured in the specimens obtained from the human dorsolateral prefrontal cortex and rat spinal cord . The 101 human subjects comprised 45 individuals diagnosed with major depression , and 56 normal controls [31] . The 19 rats comprised 7 controls and 12 rats whose spinal cord was transacted six weeks prior to the data collection [33] . In these rat specimens , the mRNA levels are assumed to be unaffected by the transaction , being collected from a region above it . Overall , the analysed data included 56 , 690 , 398 human reads ( an average of 561 , 291 per subject ) and 5 , 659 , 108 rat reads ( an average of 297 , 848 per rat ) ( Supplementary Table S1 ) . Each measurement ( mRNA molecule ) is represented by a binary vector indicating the editing states of the five sites A , B , E , C , and D . For example , a measurement in which sites A , B , and D are edited but E and C are not is represented by the binary pattern 11001 . For a collection of measurements , we denote the editing pattern as the vector , where is the number of binary vectors whose decimal representation is . First , we tested whether the editing patterns of all human normal controls were statistically indistinguishable from the editing patterns of all subjects with major depression . To this end , we conducted a conservative randomization test , whereby the -test statistic was repeatedly computed on modified data . In each repetition , we randomly assigned subjects as normal or as depressed , keeping the total number of normal controls and the total number of subjects with major depression fixed . For each repetition , we computed the test statistic of the -test , where and are the editing patterns of normal and depressed samples , respectively , and and are the total number of measurements from normal controls and from subjects with major depression , respectively . This procedure was repeated 106 times , and the p-value of the test was computed as the number of random test statistics that were larger than the true test statistic . A similar procedure was used to compare normal rats with transacted ones . We found that the editing pattern in normal human controls was indistinguishable from the editing pattern in subjects with major depression ( P = 0 . 80 ) , and that the editing pattern in normal rats was indistinguishable from that in transacted rats ( P = 0 . 65 ) . This result justifies pooling together all human subjects and all rats for further analysis . Using a similar randomization procedure , we found that the editing pattern in humans is very different from that in rats ( P<10−6 ) ( Supplementary Figure S1 ) . Next , we tested for each pair of sites whether their editing patterns were correlated . To this end , we computed the φ-coefficient ( which , for binary data , is simply the correlation coefficient; see Methods ) , and found that all pairs of sites are correlated , either positively or negatively , except for the pair ( D , E ) in human , and the pair ( A , E ) in rat ( Supplementary Table S2 ) . In order to obtain more detail on the level of dependence between different sites , we followed Ensterö et al . [26] and clustered the editing sites ( Figure 1 ) . We used the Jaccard distance coupled to single linkage hierarchical clustering ( see Methods ) , but using Dice distance following Ensterö et al . [26] had no significant effect on the clustering ( Supplementary Figure S2 ) . In order to assign confidence level to the clusters , we repeated the clustering for each individual and measured the fraction of cases in which the cluster was supported ( see Methods ) . In both human and rat the strongest association was observed to exist between sites A and B , to which site D joins next . Sites C and E were more weakly associated with the rest of the editing sites , at least in human , and the order by which they join the dendrogram changed from human to rat . Clustering , by nature , identifies groups of associated sites . However , to obtain finer resolution of the relationship between the sites , we resorted to more elaborate methods . The ultimate description of the dependency between the editing sites would be their joint probability distribution . For five editing sites , there are 8 , 782 possible joint distribution functions . We enumerated all the 8 , 782 functions , and ranked them according to how well they fit the data using both maximum-likelihood and Bayesian inference ( see Methods ) . In both human and rat , and for both maximum-likelihood and Bayesian inference , the best model was the maximally-dependent joint probability distribution , . We graphically represent probability models by a pDAG ( partial Directed Acyclic Graph ) , which is a Bayesian network containing a mixture of directed and undirected edges ( see Methods ) . The pDAG of this maximally-dependent model is simply the fully connected undirected graph ( Figures 2 , 3 ) . This result is consistent with our previous finding that all pairs of sites are significantly dependent . Such result is expected given the large size of the data . In order to find which edges in the graph are more strongly supported by the data , we divided all the 8 , 782 probability models into 11 groups according to the number of edges in the corresponding pDAG . The first group consists of all models with zero edges ( which is simply the single model ) , the second group consists of all ( ten ) models with one edge , etc . Then , we computed the best-fitting probabilistic model within each group . Hereinafter , denotes the best-fitting model from within the group of models with edges . The results for the maximum-likelihood Bayesian Information Criterion ( BIC ) score ( see Methods ) in human are shown in Figure 2 . Adding an edge to a model always improves its score , , but the improvement becomes smaller as increases . The estimated parameters of each of the best-fitting models are given in Supplementary Table S3 . To further explore the relative impact of the different edges , we ranked the edges by the order in which they first appear in the sequence of models . Specifically , the rank of an edge is the smallest integer for which contains this edge ( Table 1 ) . Importantly , an edge in need not necessarily be included in , which is the reason why two edges – ( B , E ) and ( E , C ) – have the same rank , and why no new edge appeared in . To account for the possibility that edges may disappear or reappear as the number of edges grows , we define for each edge its support . The support of an edge with rank is the fraction of the models that contain this edge ( Table 1 ) . Clearly , the higher the support , the more confident we are that the edge has a unique contribution to making the respective model better fitting the data . The edge ( A , B ) is the first to appear ( ) , and has a full support ( it appears in all the models through ) , indicating that the dependence between A and B is obviously the strongest among all pairs of sites , in accord with the findings described above . Next appear edges ( B , D ) and ( A , C ) that both also have full support . This observation is consistent with the clustering analysis results ( Figure 1 ) but provides more detail on the interdependencies among A , B , and C , D . The next edge to appear is ( A , E ) , but it does not have full support which lowers our confidence in its unique contribution to the score of the best model . The edges that appear in models and – ( C , D ) and ( E , D ) –make ( at least qualitatively ) only marginal contributions to the score . Repeating the analysis with the maximum-likelihood Akaike Information Criterion ( AIC ) scores , or with Bayesian scores , gave the same series of best-fitting models to ( Supplementary Figure S3 ) . We conducted the same analysis for the rat data . Here , too , adding edges kept improving the BIC score of the model ( Figure 3 ) . The estimated parameters of the best-fitting models are given in Supplementary Table S4 . For rat , all edges have full support , which means that an edge with rank appears in all the models to ( Table 2 ) . In rat , the two edges with the lowest rank – ( A , B ) and ( B , D ) – have the same rank as in human . However , the edge with in rat is ( B , C ) as opposed to ( A , C ) in the equivalent human model . Similarly , the edge with is ( A , E ) in human , but it is ( B , E ) in rat . This suggests that the central role of site A in governing the editing state of sites E , C , and D in human is taken by site B in rat . Indeed , referring collectively to site A or site B as F , human and rat show very similar edge rankings ( compare Tables 1 and 2 ) . Repeating the analysis with Bayes scores yielded identical series of best-fitting models for rat ( Supplementary Figure S4b ) . Using AIC scores produced only a single difference , in model . The edge ( C , D ) , which is present in the BIC and Bayes scores , was replaced by the edge ( E , C ) for the AIC score ( Supplementary Figures S4a and S5 ) . However , as we have seen , these edges anyway have marginal contribution to the best-fitting model . On the whole , the information contribution of additional edges dropped much faster for the rat data than it did for the human data ( compare Figures 2 and 3 ) , suggestive of a more complex pattern of dependencies among editing sites and accordingly more subtle regulation of the editing process in human brain . The above analysis lacks measure of score variance , thus hindering quantitative evaluation of the significance of each edge to the total score . To overcome this , we repeated the analysis for each individual separately , for both human and rat . In this way , each individual provides its own sequence of best fitting models , and for each number of edges ( ) , there is now a sequence of best models , where is the total number of individuals . For a certain , let individuals support different best-models , , such that is supported by individuals . Let us further assume that we have sorted the sequence according to the level of support , such that . Next , we define a set of models that are equally supported by the different individuals ( ) . To this end we make a Bonferroni-corrected series of proportion tests , asking whether is supported significantly more than the other best-fitting models . is the first model whose support is significantly lower than that of . The results for the BIC scores in human at significance level 0 . 05 are given in Table 3 . The results for the AIC and Bayes scores are similar , and are given in Supplementary Tables S5 and S6 . As an example , in the BIC score analysis , out of 101 individuals 78 ( 77 . 2% ) support the single-edge best-fitting model ( A→B ) ( Table 3 ) . The second-supported model is supported by 21 individuals ( 20 . 8% ) , which is significantly lower than the support for and so in this case and the set of best fitting models is simply ( ) . As another example , is supported by 14 individuals ( 13 . 9% ) , but this level of support is not statistically different from the support by 3 individuals ( 3 . 0% ) of the model , and in this case . Overall , there is a good agreement between this individual-based analysis and the pooled analysis . The pooled best-fitting model for each is marked by asterisk in Table 3 and Supplementary Tables S5 and S6 , and it is always within the group of models that are equally supported by the different individuals . Very similar results had been obtained for rat ( Supplementary Tables S7 , S8 , S9 ) . Here too , the pooled best-fitting model for each is always within the group of models that are equally supported by the different individuals . The individual-based approach can be used not only to re-evaluate the support for the different graphical models but also to perform an edge-by-edge analysis . To this end , we can look at each edge , and count how many times ( in either direction ) it appears in the sequence . These counts are binomial random variables , so if an edge appears , overall , in the best-fitting model of individuals out of a total of individuals , its variance is . The support of each edge for any in human is given in Figure 4 . Consider , for example , . Overall , the 101 individuals support different best-fitting models . Yet , the edge ( A , B ) appears in all of them , and thus is supported by all the individuals . The edge ( B , D ) is supported by 98 individuals , or by 97% of the best-fitting models . For each , we can take the first most-supported edges as the basic set of edges in the model . Then , we can check how unique is this set of edges by testing ( using proportion test ) whether the e'th supported edge is significantly more supported than the next edges ( Table 4 ) . From Table 4 and Figure 4 we see that the first three edges ( A , B ) , ( B , D ) , and ( A , C ) are clearly more supported than all other edges , in this order . However , the next edges ( B , E ) , ( E , C ) , and ( A , D ) all have approximately the same support and no one is more significant than the others . The results are almost identical when using AIC or Bayes scores ( Supplementary Figures S6 and S7 ) . Similar analysis for rat shows , in accord with our previous results , a more hierarchical relationship between the edges ( Figure 5 and Table 5 ) . Here , the order of importance is clear for the first six edges: ( A , B ) , ( B , D ) , ( B , C ) , ( B , E ) , ( A , C ) , and ( A , D ) . The edges ( C , D ) , ( E , D ) , and ( E , C ) all have approximately the same support and no one is more significant than the others . The results are almost identical when using AIC or Bayes scores ( Supplementary Figures S8 and S9 ) .
Here we analysed interdependencies among editing sites within mRNA of 5-HT2CR . The studies were performed using available data sets for the human dorsolateral prefrontal cortex and rat spinal cord tissues . Alterations in 5-HT2CR editing in these particular species and CNS regions were reported in connection to completed suicide and in response to SCI , respectively [30]–[32] . Thus , detailed understanding of editing mechanisms in these particular areas of the human and mouse CNS are expected to aid in the development of unique pharmacological strategies that target suicidal behavior as well as SCI-related spasticity . The dependencies among editing sites described here allow us to propose a hypothetical mechanistic model for the concerted action of ADAR1 and ADAR2 in 5-HT2CR editing . Given that the dependence between sites A and B was by far the strongest revealed ( see Figures 1–5 ) and that these sites are adjacent in 5-HT2CR mRNA , we speculate that ADAR1 that is known to be responsible for editing at A [22] also edits B . Moreover , the strong connection between sites A and B mechanistically might stem from editing of both sites by the same ADAR1 molecule without dissociation of the enzyme from the mRNA ( Figure 6 ) . Given that site D , known to be edited by ADAR2 [22] , is next after sites A and B in terms of the strength of the dependency , followed by site C , we further speculate that editing of sites A and B by ADAR1 affects the RNA structure such that binding of ADAR2 followed by editing at site D and possibly the two remaining sites is enhanced ( Figure 6 ) . A more far reaching implication is that the apparent primary role of ADAR1 in 5-HT2CR editing makes it the most attractive target for pharmacological intervention in the associated psychiatric disorders . It is worth noting that such intervention would not interfere with the essential editing of the GluR2 subunit of the AMPA receptor that is primarily dependent on ADAR2 [35] . The results reported here show that for both human and rat 5-HT2CR , the editing states of the physically proximal sites A and B are highly dependent . In contrast , the editing states of sites C and D , which are also physically close , seem not to be directly dependent , but rather indirectly linked through the dependencies of C and D on sites A and B , respectively . The results also reveal pronounced differences between the editing patterns in humans and rats: in humans site A has the key role in determining the editing state of the other sites whereas in rats this role belongs to site B . Although not detected by the simple analysis of the dependencies among the editing sites , computing the best-fitting probabilistic models shows that the editing state of site E is strongly dependent on the state of site A in human and on the state of site B in rat ( Tables 1 and 2 ) . Furthermore , the structure of the dependences between the editing sites is simpler in rats than it is in human implying more complex regulation of 5-HT2CR editing and by inference function in human brain . Mechanistically , the differences between the emerging patterns of editing regulation in humans and rats could be underpinned by the notable differences in the predicted secondary structures of the respective pre-mRNA regions [6] . To conclude , the results of the exhaustive analysis of 5-HT2CR editing patterns described here indicate that sites A and B strongly depend on each other in both human and rat , and that the editing state of these two sites is a key determinant of the editing state of the other three sites , and hence the overall editing pattern . The direct dependencies among the editing states of sites E , C , and D are much weaker , and the observed dependencies are probably an indirect effect of the dependency of those three sites on editing in sites A and B . Taken together , these findings allowed us to propose a mechanistic model of concerted action of ADAR1 and ADAR2 in 5-HT2CR editing . Methods of statistical inference developed here can be applied to other cases of interdependencies among multiple modification sites in RNA and proteins .
We tested whether the editing state of a pair of sites and is correlated by computing the contingency tablefor each individual , where is the number of mRNA molecules in which and ( ) in that individual . Then , the φ-coefficient was computed . This computation was repeated for all possible pairs in all individuals . Grouping the values from all individuals , the mean and standard deviation were computed for the φ-coefficient for each pair of sites , and the z-test was used to test for significance . We defined the distance between sites and as the Jaccard distance between their binary patterns , This distance was computed for all pairs of editing sites , and the distance matrix served as input for a single linkage ( shortest-distance ) hierarchical clustering . Using the Dice distance , as in [26] , had a negligible effect on the results ( Figure 1 , Supplementary Figure S2 ) . Edges were given support between 0 and 1 according to the number of individuals in which they are supported . The editing state of a site is a random variable . Thus , the joint probability distribution is the ultimate description of the dependencies among the five serotonin receptor editing sites . Any joint probability distribution can be decomposed in many different ways as a product of conditional and marginal probabilities , where each decomposition may represent different dependencies among the sites . For example , the joint probability distribution of two random variables can be decomposed in three ways: , , and . The first two models represent dependency between and , whereas the third model represents independence between and . There is a recursive formula for computing the number of possible decomposition for any given number of random variables [36] . In our case , the joint probability distribution of five random variables can be decomposed in 29 , 281 different ways . Importantly , many of these decompositions are redundant in the sense that several decompositions can describe essentially the same probabilistic model . In the two-random variable example above , Bayes law renders equivalence between the first two decompositions , . A set of equivalent decompositions is denoted equivalence class . There is no known general formula to compute the number of equivalence classes for a given number of random variables . However , there is an algorithm allowing one to tell , given two decompositions , whether they belong to the same equivalence class or not [37] . Here we propose a technique to find the joint probability distribution that fits best to the data . This technique , being exponential with the number of editing sites , is useful when there is a small number of editing sites , as in the present case and in several other functionally important cases of mRNA editing ( e . g . , kainate 2 glutamate receptor or CaV1 . 3 channel ) [38] , [39] . In a nutshell , we scanned through the entire set of 29 , 281 possible decompositions , and constructed the full set of equivalence classes . Then , we tested which of the equivalence classes fits the data best ( see details below ) . In order to enumerate all the possible decompositions of , we used Steinsky's ranking algorithm , that allows for a one-to-one mapping between the set of all decompositions and the integers 0 , 1 , 2 , … , [36] . Then , we scanned through the list of decompositions by a series of pairwise comparisons , and kept only a single decomposition from each equivalence class . In this way , we found that the joint probability distribution of five random variables can be decomposed into 8 , 782 equivalence classes ( Supplementary Table S1 ) . A Bayesian network provides a compact graphical representation of a decomposition . It is a directed acyclic graph ( DAG ) in which the nodes are the random variables , and an edge leading from a node to each of its children ( a parent of a node is a node upon which is conditionally dependent in the decomposition ) . In the context of Bayesian networks , the collection of DAGs that represent equivalence class is called Markov equivalence class . For convenience , we shall hereinafter use probabilistic model as a synonym to equivalence class or to Markov equivalence class . Bayesian networks have been proved as a very efficient tool to facilitate calculations on probabilistic models . In order to score how well each probability model fits the observed data , we used two alternative scoring methods . The first is based on a maximum-likelihood ( ML ) procedure , and the second is based on Bayesian inference . Below , we describe both methods . The Bayesian learning formalism requires assumptions about the prior probability of the parameters . We used the Dirichlet priors , which is the standard choice of priors in this kind of problems because it bears desirable properties such as global and local parameter independence [40] . For each node , and for each editing state of its parents , the Dirichlet priors are specified by two parameters that we denote and . The use of these priors can be conceived as adding another pseudo-measurements to the observed measurements , where is the number of pseudo-measurements in which and the editing state of is , and is the number of pseudo-measurements in which and the editing state of is . We denote by the number of pseudo-measurements in which the editing state of is , . The Bayesian Score ( BS ) of a probabilistic model is given bywhere the summation is over all the nodes , andwhere the product is over all possible editing statees of , and is the gamma function [40] . We generated the set of pseudo-measurements to consist exactly one of each of the possible editing statees of the five editing sites . That is , the pseudo-measurements consist a single measurement 00000 , a single measurement 00001 , etc . If we denote the number of editing sites by ( ) , then the set of pseudo-measurements consists of measurements . If we denote the number of parents of node by , then , and . This giveswhich is justIf node has no parents , then , , , and the formula further simplifies to A whole Markov equivalence class can be described by a partial-directed acyclic graph ( pDAG ) [40] , [41] , which is a graph made of both directed and undirected edges . If an edge can be oriented differently in DAGs belonging to the same Markov equivalence class , it would be undirected . In this work , whenever a probabilistic model is visualized as a Bayesian network , the pDAG representation is employed .
|
The serotonin receptor 2C is a key regulator of diverse neurological processes that affect feeding behavior , sleep , sexual behavior , anxiety and depression . The function of the receptor itself is regulated via so-called pre-mRNA editing , i . e . site-specific adenosine deamination in five distinct sites . The greater the number of edited sites in the serotonin receptor mRNA , the lower the activity of the receptor it encodes . Here we used the results of extensive massively parallel sequencing from human and rat brains to elucidate the dependencies among the editing states of the five sites . Despite the apparent simplicity of the problem , disambiguation of these dependencies is a difficult task that required development of a new statistical technique . We employed this method to analyse the dependencies among editing in the 5 susceptible sites of the receptor mRNA and found that the proximal , juxtaposed sites A and B are strongly interdependent , and that the editing state of these two sites is a major determinant of the editing states of the other three sites , and hence the overall editing pattern . The statistical approach we developed for the analysis of mRNA editing can be applied to other cases of multiple site modification in RNA and proteins .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"molecular",
"neuroscience",
"biology",
"neuroscience"
] |
2012
|
Dependencies among Editing Sites in Serotonin 2C Receptor mRNA
|
The molecular clock and its phylogenetic applications to genomic data have changed how we study and understand one of the major human pathogens , Mycobacterium tuberculosis ( MTB ) , the etiologic agent of tuberculosis . Genome sequences of MTB strains sampled at different times are increasingly used to infer when a particular outbreak begun , when a drug-resistant clone appeared and expanded , or when a strain was introduced into a specific region . Despite the growing importance of the molecular clock in tuberculosis research , there is a lack of consensus as to whether MTB displays a clocklike behavior and about its rate of evolution . Here we performed a systematic study of the molecular clock of MTB on a large genomic data set ( 6 , 285 strains ) , covering different epidemiological settings and most of the known global diversity . We found that sampling times below 15–20 years were often insufficient to calibrate the clock of MTB . For data sets where such calibration was possible , we obtained a clock rate between 1x10-8 and 5x10-7 nucleotide changes per-site-per-year ( 0 . 04–2 . 2 SNPs per-genome-per-year ) , with substantial differences between clades . These estimates were not strongly dependent on the time of the calibration points as they changed only marginally when we used epidemiological isolates ( sampled in the last 40 years ) or three ancient DNA samples ( about 1 , 000 years old ) to calibrate the tree . Additionally , the uncertainty and the discrepancies in the results of different methods were sometimes large , highlighting the importance of using different methods , and of considering carefully their assumptions and limitations .
In 1962 , Zuckerland and Pauling used the number of amino-acid differences among hemoglobin sequences to infer the divergence time between human and gorilla , in what was the first application of the molecular clock [1] . Although many at the time found it “crazy” [2] , soon the molecular clock was incorporated in Kimura’s neutral theory of molecular evolution [3] , and found its place in the foundations of evolutionary biology . Thanks to the improvements of sequencing technologies and statistical techniques , it is now possible to use sequences sampled at different times to calibrate the molecular clock and study the temporal dimension of evolutionary processes in so called measurably evolving populations [4] . These advancements have been most relevant for ancient DNA ( aDNA ) , and to study the evolutionary dynamics of pathogen populations , including one of the deadliest human pathogens: Mycobacterium tuberculosis . In 1994 , Kapur and colleagues pioneered molecular clock analyses in MTB: they assumed a clock rate derived from other bacteria and used genetic polymorphisms to infer the age of divergence of different MTB strains [5] . Since the publication of the MTB reference genome [6] , whole genome sequence ( WGS ) data of MTB strains is becoming available at increasing speed , and especially in the last five years , studies using large WGS data sets allowed for precise estimates of the MTB genetic diversity and of the molecular clock rate . Phylogenetic analyses with a molecular clock have been used to estimate the timing of the introduction of MTB clades to particular geographic regions , the divergence time of the MTB lineages , and the age of the most recent common ancestor ( MRCA ) of the MTB complex [7–13] . Clock models , together with phylodynamic models in a Bayesian setting have been used to characterize tuberculosis epidemics by determining the time at which outbreaks began and ended [14–18] , establishing the time of origin and spread of drug resistant clades [11 , 14 , 19–20] , and correlating population dynamics with historical events [9 , 12 , 20 , 21–22] . One example of the potential of molecular clock analyses is the study of Eldholm and colleagues [20] , where the collapse of the Soviet Union and of its health system was linked to the increased emergence of drug-resistant strains in former Soviet Republics , thus providing insights into the evolutionary processes promoting drug resistance . A key aspect about estimating evolutionary rates and timescales in microbial pathogens is assessing their clocklike structure . All molecular clock analyses require some form of calibration . In many organisms , this consists in constraining internal nodes of phylogenetic trees to known divergence times ( for example , assuming co-divergence with the host , or the fossil record ) , but in rapidly evolving pathogens and studies involving ancient ( a ) DNA , it is also possible to use sampling times for calibrations [23] . In the latter approach , the ages of the tips of a tree , rather than those of internal nodes are constrained to their collection times . Clearly , the sampling time should capture a sufficient number of nucleotide changes to estimate the evolutionary rate , which will depend on the evolutionary rate of the organism and the extent of rate variation among lineages . Some popular methods to assess such clocklike structure are the root-to-tip regression and the date randomization test ( DRT ) . While many of the studies inferring evolutionary rates for MTB reported support for a molecular clock [10–11 , 13–14 , 16 , 18 , 20 , 22] , some found a lack of clocklike structure [7 , 17–18] , and others assumed a molecular clock without testing whether the data had a temporal structure [9 , 12 , 15 , 19 , 21] . In all studies where the calibration was based on the sampling time ( tip-dating ) , the clock rate estimates spanned roughly an order of magnitude around 10−7 nucleotide changes per site per year . This was in contrast with the results of Comas et al . 2013 [7] , where the clock was calibrated assuming co-divergence between MTB lineages and human mitochondrial haplotypes ( i . e . internal node calibrations ) , and was estimated to be around 10−9 nucleotide changes per site years . Additionally , the age estimate of the MRCA of MTB obtained by Comas et al . 2013 ( ~70 , 000 years ) is similar to the one of Wirth et al . 2008 [24] ( ~ 40 , 000 years ) , which was based on the molecular clock analysis of 24 variable tandem repeat loci , but about ten times older than the one of study that used WGS of aDNA samples and tip-dating to calibrate the tree ( ~ 6 , 000 years ) [8] . The diverging results obtained by Comas et al . 2013 [7] , Wirth et al . 2008 [8] and studies based on tip-dating are therefore most likely caused by the adoption of different assumptions and methodologies . Studies based on WGS and tip-dating found some differences in the clock rate of different lineages: some lineage 2 ( L2 ) data sets [20] were found to have a faster clock rate compared to lineage 4 ( L4 ) data sets [11 , 14 , 16 , 21] , while others showed lower clock rates , comparable with L4 [13 , 18] . Studies based on aDNA produced slightly lower clock rate estimates [8 , 10] compared to studies based on modern strains , thus suggesting support for the phenomenon of time dependency of clock rates in MTB [25–26] . All these results indicate that different MTB lineages and populations might have different clock rates , and that the age of the calibration points could influence the results of the analyses . Comparing the results of different studies has however a main limitation: the observed differences could be due to different rates of molecular evolution among MTB populations , to methodological discrepancies among studies , or a combination of both . Here we assembled a large genomic data set of sequences belonging to epidemiological contemporary strains ( sampled in the last 40 years ) , including sequences from all major lineages of MTB ( 6 , 285 strains in total , belonging to six human adapted lineages , L1-L6 , and one lineage predominantly infecting cattle , M . bovis ) . We then applied the same set of methodologies to the whole data set , to individual lineages and sub-lineages , and to selected local outbreaks , thus disentangling the temporal signal of multiple nested data sets with different structures and complexity , and ensuring the comparability of the results among different bacterial clades and epidemiological settings . Finally , in a separate analysis we investigated the time-dependency hypothesis by including three aDNA sequences about 1 , 000 years old [8] . With this systematic approach , we addressed the following questions:
Finding evidence of temporal structure is the first step when performing molecular clock analyses [27] . If there is not enough genetic variation between samples collected at different times , these cannot be used to calibrate the molecular clock , i . e . the population is not measurably evolving . To test the temporal structure of MTB data sets , we identified 6 , 285 contemporary strains ( sampled in the last 40 years ) that passed our quality filters ( average coverage > 15X , they were not identified as mixed infection etc . see Methods for details ) , and for which the date of isolation was known ( S1 Table ) . We used root-to-tip regression to evaluate the temporal structure of the whole MTB complex and of the individual lineages ( L1-L6 and M . bovis ) [28] . The root-to-tip regression is a regression of the root-to-tip distances as a function of sampling times of phylogenetic trees with branch lengths in units of nucleotide changes per site , where the slope corresponds to the rate . Under a perfect clock-like behavior , the distance between the root of the phylogenetic tree and the tips is a linear function of the tip’s sampling year: recently sampled strains are further away from the root than older ones , such that the R2 is the degree of clocklike behavior [29] . We obtained very low values of R2 for all lineages ( maximum 0 . 1 for M . bovis ) , indicating a lack of strong clock-like behavior ( S1 Fig ) . Additionally , we found a weak negative slope for L1 , L5 and L6 , normally interpreted as evidence for a lack of temporal structure , or overdispersion in the lineage-specific clock rates [28] ( S1 Fig , S2 Table ) . A negative slope of the regression line can be caused by an incorrect placement of the root [30] . To address this potential problem , we repeated these analyses rooting the trees with an outgroup . We found a negative slope for L1 and L6 and a positive slope for L5 , although with an extremely low value of R2 ( < 0 . 01 ) . These results indicate that the negative slope of L1 and L6 and the low R2 values of the three data sets are not due to an incorrect placement of the root ( S2 Fig ) . Since root-to-tip regression can be used only for exploratory analyses and not for formal hypothesis testing [28] , we performed a date randomization test ( DRT ) . The DRT consists in repeatedly reshuffling the date of sampling among taxa and then comparing the clock rate estimates among the observed and reshuffled data sets [27] . If the estimate obtained from the observed data does not overlap with the estimates obtained from the randomized data sets , we can conclude that the observed data has a stronger temporal signal than expected by chance , such that there is statistically significant clocklike structure [27] . Usually the DRT is implemented in a Bayesian phylogenetic setting , however , considering the size and the number of data sets included in this study , an excessive amount of computation would be required . To overcome this problem , we estimated the clock rate with the least-squared dating method ( LSD ) [31] . The advantage of this method is that it is orders of magnitude faster than fully Bayesian approaches , and can therefore be used on data sets with thousands of taxa and with more randomizations compared to the 10–20 typically used in a Bayesian setting [32] . A limitation of least squares dating is that it typically assumes a single tree topology and vector of branch lengths , and a strict clock ( i . e . all branches have the same clock rate ) . However , a simulation study showed that maximum likelihood trees produced similar estimates compared to the true topology , and that it is robust to uncorrelated variation of the clock rate among branches in the phylogeny [31–33] . For each data set ( the MTB complex and the seven analyzed lineages ) , we reshuffled the year of sampling among tips 100 times and estimated the clock rate of observed and randomized data sets with LSD . All eight data sets except L5 and L6 passed the DRT ( Methods , S1 Fig , S2 Table ) . L5 and L6 are the two lineages with the lowest sample size , 117 and 33 strains , respectively . Moreover most strains were sampled in a short temporal period compared to the other lineages ( S3–S8 Figs ) . It is likely that with additional strains sampled across a larger time period , L5 and L6 will also show evidence for a molecular clock . We complemented the analysis described above with a Bayesian phylogenetic analysis in BEAST2 [34] . Since this is computationally expensive , we reduced the large data sets ( MTBC , L1 , L2 , L4 and M . bovis ) to 300 randomly selected strains . For each data set , we selected the best fitting nucleotide substitution model identified with jModelTest 2 [35] . For this first analysis , we assumed a coalescent constant population size prior , used a relaxed clock model , and a 1/x prior for the clock rate , constrained between 10−10 and 10−5 nucleotide changes per site per year . This interval spans the range of clock rates proposed for M . tuberculosis and for most other bacteria [20 , 36] . We observed that for all data sets the posterior was much more precise ( with a narrow distribution ) than the prior , thus indicating that the data was informative [37] . Again , the only exceptions were L5 and L6 , where the posterior distribution was flat , ranging between 10−10 and 10−7 nucleotide changes per site per year , confirming the lack temporal structure of these two data sets ( S1 Fig ) . We repeated these analyses on 23 sub-lineages and 7 outbreaks and local populations to test whether we could detect a temporal structure also in smaller , less diverse data sets . With this sub-sampling scheme , we could compare the results among different clades , among outbreaks with different epidemiological characteristics , and among local outbreaks and global data sets ( see Methods ) . We found that 11 sub-lineages and 5 local populations passed the DRT ( S2 Table , S3–S6 and S9–S11 Figs ) . All the data sets that failed the DRT had less than 350 genomes , or were composed of strains sampled in a temporal range of 20 years or less . Additionally , only two of the ten data sets sampled across less than 15 years , and three of the twelve data sets with less than 100 strains passed the DRT ( Fig 1; S2 Table ) , indicating that large sample sizes and wide temporal sampling windows are necessary to obtain reliable estimates of evolutionary rates and timescales in MTB . Conversely , the number of polymorphic positions and the genetic diversity measured with Watterson’s estimator did not correlate with the outcome of the DRT ( S12 Fig ) . Among the three methods generally used to study the temporal structure of a data set , the root-to-tip regression resulted in a negative slope , and therefore failed to detect the temporal structure of some of the data sets that passed the DRT ( i . e . L1 , L4 . 1 . 2 and L1 . 1 . 1 ) . Nevertheless , root-to-tip regression can be useful to identify data sets where the temporal signal comes from a single strain , or a few strains ( see below ) . Comparing prior and posterior distributions of the clock rates was also useful to detect the presence of temporal structure , although this was not always in agreement with the results of the DRT: some of the data sets that did not pass the DRT ( e . g . L2 . 2 . 1_nc2 , Trewby 2016 [38] ) had a posterior distribution of the clock rate more distinct from the prior than some of the data sets that passed the DRT ( e . g . L1 . 1 . 1 , L1 . 2 . 1 and L1 . 2 . 2 ) ( S3 , S5 and S6 Figs , S2 Table ) . A possible reason for this could be that LSD and BEAST have different statistical power with different data sets . Additionally , in some cases the deviation of the posterior distribution of the clock rate from the prior could be an artifact caused by tree prior misspecification , and not the result of genuine temporal structure [39] . In Bayesian analyses , different models and priors are based on different assumptions about the evolutionary processes , and can thus influence the results [40] . Often different sets of assumptions are tested with the Bayes factor ( the ratio of the marginal probabilities of two competing models ) , and the most likely model is then chosen to estimate the parameters of interest [40] . Given the size and number of the data sets considered in this study , it was not possible to assess the relative fit of many competing models for all data sets . However , model misspecification can result in biased estimates . It was therefore important to investigate the robustness of the results to different models and priors . We repeated the Bayesian analysis using a uniform prior instead of the 1/x prior on the clock rate . We ran a BEAST analysis sampling from the priors and found that the uniform prior was biased towards high clock rates and put most weight on rates between 10−6 and 10−5 nucleotide changes per site per year ( S13 Fig ) . For all data sets , we compared the posterior distribution of the clock rate obtained with the two different priors ( S14–S16 Figs , S2 Table ) . Some data sets showed hardly any difference ( e . g . MTBC , L1 , L2 , L3 , L4 etc . ) , indicating that the data was informative and that the data set had a strong temporal structure . However , this did not always correlate with the results of the DRT . For example , the subset of 300 strains of L2 and the data set Trewby 2016 [38] did not pass the DRT but showed a distinct posterior distribution that was not sensitive to the prior choice . Other data sets , including three that passed the DRT by a small margin ( L1 . 1 . 1 , L1 . 2 . 1 and L1 . 2 . 2 ) , were more sensitive to the prior choice and resulted in two distinct posterior distributions , indicating a weaker temporal structure ( S6 Fig ) . An additional assumption of the phylogenetic model that can influence the results of molecular clock analyses is the tree prior ( also known as demographic model ) . We tested the sensitivity to the tree prior by repeating the analysis with an exponential population growth ( or shrinkage ) prior instead of the constant population size . For this analysis , we used the 1/x prior on the clock rate and we considered only the data sets that passed the DRT ( 21 data sets ) . The constant population model is a specific case of the exponential growth model ( when the growth rate is equal to zero ) . Therefore , if the 95% Highest Posterior Density interval ( HPD ) of the growth rate does not include zero , we can conclude that the data reject a demographic model with constant population size . We found that 14 data sets rejected the constant population size model , and that all of them had positive growth rates ( S2 Table ) . The three data sets that were found to be sensitive to the prior on the clock rate were also sensitive to the tree prior , confirming their low temporal structure and information content , while the results for all other data sets were only moderately influenced by the tree prior ( S17 and S18 Figs , S2 Table ) . Overall , we found that , except for three data sets ( L1 . 1 . 1 , L1 . 2 . 1 and L1 . 2 . 2 ) , the clock rate estimates were robust to different priors of the clock rate and to different demographic models . To compare the clock rates of different data sets , we report the analysis with the 1/x prior on the clock rate because the uniform prior can bias the estimates upward . For data sets that showed evidence against the constant population size model ( 95% HPD of the growth rate not including zero ) , we report the results of the analysis with the exponential population growth , and for the others , we report the results of the analysis with constant population size . We found that the point estimates of all data sets where we detected temporal structure ranged between 2 . 86x10-8 ( L3 , BEAST analysis ) and 4 . 82x10-7 ( Eldholm 2016 [20] , BEAST analysis ) nucleotide changes per site per year . While some data sets had a low range of the 95% confidence interval ( CI ) , reaching the hard limit imposed by LSD of 10−10 , most of the CI and 95% highest posterior density intervals ( HPD ) were included between 10−8 and 5x10-7 ( Fig 2 , S2 Table ) . This range encompasses previous estimates obtained with epidemiological samples and aDNA , and is among the lowest in bacteria , thus supporting our conclusion from above: tip-dating with MTB requires samples collected over a long period of time because of the slow clock rate . There was one notable exceptions to the pattern described above: the data sets L4_nc which showed a much higher clock rate estimate compared to all other data sets included in this study ( ~10−6; S2 Table ) . However , this is most likely an artifact: 1 ) L4_nc is the smallest among all considered data sets , with 32 strains . 2 ) Most strains are identical or nearly so , collected in the same year , and form a monophyletic clade ( S7 and S19 Figs ) . It is known that data sets with a high degree of temporal and phylogenetic clustering can pass the DRT also when they do not have temporal structure [41] . 3 ) The root-to-tip regression suggests that the temporal signal comes from one single strain in L4_nc ( S5 Fig ) . We therefore excluded the L4_nc data set from further analyses . Our results suggest that different lineages of MTB have different clock rates . For example most L1 data sets had point estimates higher than most L4 data sets , although the CI and HPD were often overlapping . The point estimates indicate that the clock rate of L1 is more than double the clock rate of L4: two average L1 strains are expected to differ by 12 SNPs after ten years of divergence , while two average L4 strains will differ by 5 SNPs after the same period of time . This was supported by the results of both LSD , where the 95% CI of L1 and L4 did not overlap , and BEAST , where the 95% HPD overlapped partially , but the two posterior distributions showed distinct peaks ( Fig 2 , S2 Table , S20 Fig ) . A practical implication of these results pertains to the widespread use of SNP distances to identify ongoing transmission in MTB epidemiological studies . Usually , recent transmission is postulated when two or more strains differ by a number of SNPs below a certain threshold [42] . However , this approach will result in systematically lower levels of transmission for clades with faster rates of molecular evolution . For example , a recent study reported low transmission rates of L1 compared to L2 and L4 in Vietnam [43] , which could partially be explained by a faster clock rate of L1 , as opposed to reduced ongoing transmission . When considering the results of BEAST , L2 had a higher clock rate compared to L4 , and all data sets included in the sub-lineage L2 . 2 . 1 showed a faster clock rate compared to the complete L2 data set ( Fig 2 ) . The sub-lineage L2 . 2 . 1 includes the so called “modern Beijing” family , which was shown to be epidemiologically associated with increased transmission , virulence and drug [43–48] , and to have a higher mutation rate compared to L4 strains [49] . However , the LSD estimates for L2 . 2 . 1 and for its sub-lineages , despite showing the same trend of BEAST , support a lower clock rate compared to BEAST , and have large confidence intervals , overlapping with the results of L2 and L4 ( Fig 2 ) . Further evidence of among-lineage variation is provided by the results of the Bayesian analyses , where for most data sets , we obtained coefficients of variation ( COV ) with a median of 0 . 2–0 . 3 , and not abutting zero ( S2 Table ) , thus rejecting the strict clock [37] . Taken together , these results indicate that there is a moderate variability among the current rate of molecular evolution of different MTB lineages , which could be caused by different mutation rates as it was reported for L2 and L4 [49] , and support the idea that the inference of transmission in MTB should move away from the use of SNP distances to methods that incorporate information about the molecular clock [50] . In our analysis , we included two outbreaks caused by strains belonging to the same sub-lineage ( L4 . 1 . 2; Eldholm 2015 [14] , Lee 2015 [15] ) . This gave us the opportunity to compare the molecular clock of clades with a similar genetic background in different epidemiological settings . The Eldholm 2015 data set is a sample of an outbreak in Argentina , in which resistance to multiple antibiotics evolved several times independently [14] . The Lee 2015 data set represents an outbreak of drug-susceptible strains in Inuit villages in Québec ( Canada ) [15] . The clock rates of these two data sets were highly similar ( 95% CI and HPD ranging between 5 . 07x10-8 and 8 . 88x10-8 for all analyses; Fig 2 , S2 Table ) , thus suggesting that for these two data-sets , different epidemiological characteristics , including the evolution of antibiotic resistance , did not have a large impact on the rate of molecular evolution of MTB . Overall , our results are comparable with previously published tip-dating studies of MTB that used WGS data , indicating that in practice , using different sequencing and data analysis pipelines is unlikely to lead to drastically different results . However , further studies are needed to reveal in detail the effect that some steps of the bioinformatic pipeline , such as the exclusion of low confidence SNP calls , repetitive genomic regions , or mixed infections might have on the results of molecular clock analyses . We showed that in the last 40 years , the clock rates of different MTB data sets were moderately divergent . A different question is whether the clock rate was constant during the evolutionary history of the MTB complex . When looking at the phylogenetic tree of the MTB complex , rooted with the genome sequence of M . canettii , one notices that strains belonging to different lineages , despite being all sampled in the last 40 years , have different distances from the root ( Fig 3 ) . For example , since their divergence from the MRCA of the MTB complex , the two M . africanum lineages ( L5 and L6 ) and especially M . bovis , accumulated more nucleotide changes than the lineages belonging to MTB sensu stricto ( L1-L4; Fig 3 ) . Additionally , all methods ( root-to-tip regression , LSD and BEAST ) if used without an outgroup , placed the root on the branch between M . bovis and all other lineages , while rooting the tree with the outgroup M . canettii placed the root on the branch connecting MTB sensu stricto with M . africanum ( L5 and L6 ) and M . bovis . The different root placement affects the clock rate estimation only moderately ( 4 . 16x10-8 LSD analysis without outgroup , 5 . 59x10-8 LSD analysis with outgroup; S2 Table ) , but it is a further indication of the variation of the rate of molecular evolution during the evolutionary history of the MTB complex . The observation that all M . bovis strains , despite having a clock rate similar to all other data sets , have a larger distance from the root of the MTB complex tree compared to other lineages is intriguing , and could be explained by a faster rate of molecular evolution of the ancestors of M . bovis ( Figs 2 and 3 ) . It is believed that M . bovis switched host ( from human to cattle ) [51–53] , and it is possible that during the adaptation to the new host , several genes were under positive selection , thus leading to an increase in the accumulation of substitutions in the M . bovis genome . Another possibility is that the ancestor of M . bovis experienced a period of reduced population size , a bottleneck , and as a consequence , slightly deleterious mutations were fixed by genetic drift , resulting in a faster clock rate compared to larger populations where selection is more efficient in purging deleterious mutations [54–55] . It has been suggested that in MTB , as in other organisms , the clock rate estimation is dependent on the age of the calibration points [7 , 25–26 , 36 , 56] , and that using recent population-based samples could result in an overestimation of the clock rate , because these samples include deleterious mutations that have not yet been purged by purifying selection . However , the validity of the time dependency hypothesis has been contested in general [57] , and for MTB in particular [21] . Here we used an approach similar to Rieux et al . 2014 [58] and tested whether the time dependency hypothesis was supported by our data . We repeated the analyses presented above , only this time we included the aDNA genome sequences of three MTB strains obtained from Precolumbian human remains from Peru [8] . If the clock rate estimates depend on the age of the calibration points , adding ancient genomes should result in lower clock rates . We performed this analysis with LSD , using the complete data set ( 6 , 285 strains ) , and with BEAST , using the sub-sample of 300 randomly selected strains described above , and an additional independent random sub-sample of 500 strains ( Methods ) . With LSD , adding the aDNA samples resulted in a slightly faster clock rate , conversely all the analyses performed with BEAST resulted in marginally slower clock rates when the aDNA samples were included ( Table 1 ) . These results indicate that the effect of the age of the calibration points on the clock rate is modest , and they are corroborated by the observation that MTB mutation rates in vitro and in vivo , estimated with fluctuation assays and resequencing of strains infecting macaques , are remarkably similar to the clock rates obtained in our study ( ~ 3x10-8–4x10-7 ) [59] . The aDNA samples considered in this study are not optimal to test the time dependency hypothesis because they belong to the M . pinnipedii clade of the MTB complex [8] . The modern strains of this lineage are rarely sampled , because they are infecting seals and sea lions rather than humans . The only additional aDNA samples available for MTB are L4 samples isolated from 18th century Hungarian mummies [10 , 60] . However , these samples are a mix of strains with different genotypes , and cannot be easily integrated with the data and pipelines used in this study . While these results suggest that the age of the calibration points has a small effect on the clock rate estimates , they are based on only three aDNA samples . Additional aDNA samples from older periods and belonging to other lineages are necessary to reject ( or confirm ) the time dependency hypothesis in MTB . Recently , Sabin and colleagues [61] reported the sequencing of a high quality MTB genome from the 17th century , this data will contribute to the investigation of the time dependency hypothesis in MTB . In most cases , the goal of molecular clock studies is not to estimate the clock rates , but rather the age of the phylogenetic tree and of its nodes . Conceptually , this means extrapolating the age of past events from the temporal information contained in the sample set . If we exclude the few aDNA samples that are available [8 , 10] , all MTB data sets have been sampled in the last 40 years . It is therefore evident that the age estimates of recent shallow nodes will be more accurate than medium and deep nodes . In part , this is reflected in the larger CI and HPD of the age of ancient nodes compared to more recent ones . Extrapolating the age of trees that are thousands of years old with contemporary samples is particularly challenging , because the observed data captures only a small fraction of the sample’s evolutionary history , and these are the cases where aDNA samples are most valuable . Nevertheless , the age of the MRCA of the MTB complex and of its lineages is highly relevant to understand the emergence and evolution of this pathogen and a debated topic [7–8 , 24] . The LSD analyses on the tree rooted with M . canettii estimated the MRCA of the MTB complex to be between 2 , 828 and 5 , 758 years old ( S2 Table ) . These results are highly similar to the ones of Bos and colleagues ( 2 , 951–5 , 339 ) which were obtained with Bayesian phylogenetics and a much smaller sample size [8] . These estimates should be taken with caution because of the intrinsic uncertainty in estimating the age of a tree that is several thousands of years old , calibrating the molecular clock with the sampling time of modern strains and only three aDNA samples . A more approachable question is the age of the MRCA of the individual MTB lineages . Here we can consider the results of four different analyses: the LSD and BEAST analyses on the individual lineages ( L1-L4 , and M . bovis ) , and the LSD and BEAST analyses on the complete MTB complex ( including the aDNA samples ) , from which the age of the MRCA of the lineages can be extracted ( L1-L6 , and M . bovis ) . When we combined all these results , merging the CI and HPD , we obtained an estimate of the age of the MTB lineages which accounts for the uncertainty intrinsic in each analysis , but also for the differences among inference methods and models , thus providing a more conservative hypothesis . In all our analyses , the point estimates of the age of all lineages resulted to be at most 2 , 500 years old , and the combined CI and HPD extended to a maximum of 11 , 000 years ago for L2 ( 95% CI of the LSD analysis; Fig 4 , S3 Table ) . The large CI of L2 obtained with LSD was maybe due to among-lineage variation of the clock rate in L2 . While L5 , L6 and M . bovis have younger MRCAs and narrower confidence intervals , we should note that for these lineages the sampling is much less complete compared to L1-L4 , and it is possible that further sampling will add more basal strains to the tree , thus resulting in older MRCAs . For the other lineages , where the sampling is more representative of the global diversity , the confidence intervals of the age of the MRCAs extend over several thousands of years , and the point estimates of the four analyses spread over 1 , 000–2 , 000 years . This shows that we should be very careful when interpreting the results of tip-dating in MTB , especially if our goal is to estimate the age of ancient nodes such as the MRCAs of MTB lineages . Conservative researchers might want to use different methods; several model and prior combinations should be formally tested in BEAST , and the final results can be combined in one range providing an estimation of the uncertainty of the clock rate and of the age of some specific node of the tree . Altogether our results highlight the uncertainty of calibrating MTB trees with tip-dating , they nevertheless support the results of Bos et al . 2014 [8] that found the MRCA of the MTB complex to be relatively recent , and not compatible with the Out-of-Africa hypothesis [7 , 24] in which the MTB lineage differentiated in concomitance with the dispersal of Homo sapiens out of Africa , about 70 , 000 years ago . Dating analyses based on DNA samples can only reconstruct the evolutionary history of the data set as far back as the MRCA of the sample . It is possible that in the future new lineages will be sampled , and the MTB phylogeny will be updated moving the MRCA further in the past . Additionally , it is also possible that extinct lineages were circulating and causing diseases much earlier that the MRCA of the strains that are circulating now . This hypothesis is supported by the detection of molecular markers specific for MTB in archeological samples ( reviewed in [62] ) , the oldest of them in a bison’s bone about 17 , 500 years old [63] . Several such studies directly challenge the results of tip-dating presented here because they reported molecular markers specific to MTB lineages in archeological samples that predate the appearance of those lineages as estimated by tip-dating [64–66] . However , there is a controversy regarding the specificity of some of the used markers , and the potential contamination of some of the samples by environmental mycobacteria [67–68] . Whole genome sequences from additional aDNA samples are needed to reconcile these two diverging lines of evidence . Ideally , these samples should represent different lineages , span different periods , and be more ancient than the currently available aDNA from Peruvian human remains . In this systematic study of the molecular clock of MTB , we collected the genome sequences of 6 , 285 strains from all over the world , divided them into multiple data sets and used different tip-dating methods to assess their temporal structure and molecular clock rates . In most cases , the clock rate could be estimated reliably only if the data sets included strains sampled for 15 or more years . We inferred an overall clock rate ranging between 10−8 and 5x10-7 nucleotide changes per site per year , independently of the data set . We explored different methodologies and approaches to molecular clock analyses in MTB , thus providing information and guidance for future studies . Our results support the robustness of molecular clock analyses in MTB under some conditions , but also highlight the challenges of tip-dating with large genome sizes and slow evolutionary rates . The first challenge is to sample strains for a period long enough to calibrate the clock . When this cannot be done , it is possible to include in the analysis additional strains with older dates that are phylogenetically closely related and for which the genome sequence is available in the public domain . Alternatively , for data with no temporal structure , our estimates can be used to calibrate the clock rate at 10−8–5x10-7 nucleotide changes per site per year , thus obtaining a broad ( conservative ) time estimate for the age of the tree and of its nodes . The second challenge is computational . As the cost of sequencing decreases , large WGS data sets with several hundred ( or thousand ) of strains are becoming more common . While large data sets allow for a more precise estimation of the clock rate , they are problematic to analyze with Bayesian phylogenetics . Two possible strategies around this problem are sub-sampling large data sets or reducing the parameter space by using hybrid methods ( e . g . using a fixed tree topology ) [32] . Some of these challenges might be overcome through future developments . For example , the expansion of fast tip-dating algorithms ( such as LSD ) to incorporate relaxed clock models would benefit the analysis of large data sets . Additionally , the use of long-read sequencing data will enable the inclusion of SNPs located in repetitive regions of the genome , thus maybe shortening the sampling period needed to calibrate the molecular clock .
We identified 21 , 734 MTB genome sequences from the sequence read archive ( S4 Table ) . All genome sequences were processed similarly to what was described in Menardo et al . 2018 [69] . We removed Illumina adaptors and trimmed low quality reads with Trimmomatic v 0 . 33 ( SLIDINGWINDOW:5:20 ) [70] . We excluded all reads shorter than 20 bp and merged overlapping paired-end reads with SeqPrep ( overlap size = 15 ) ( https://github . com/jstjohn/SeqPrep ) . We mapped the resulting reads to the reconstructed ancestral sequence of the MTB complex [7] using the mem algorithm implemented in BWA v 0 . 7 . 13 [71] . Duplicated reads were marked by the MarkDuplicates module of Picard v 2 . 9 . 1 ( https://github . com/broadinstitute/picard ) . We performed local realignment around Indel with the RealignerTargetCreator and IndelRealigner modules of GATK v 3 . 4 . 0 [72] . We used Pysam v 0 . 9 . 0 ( https://github . com/pysam-developers/pysam ) to exclude reads with alignment score lower than ( 0 . 93*read_length ) - ( read_length*4*0 . 07 ) ) : this corresponds to more than 7 miss-matches per 100 bp . We called SNPs with Samtools v 1 . 2 mpileup [73] and VarScan v 2 . 4 . 1 [74] using the following thresholds: minimum mapping quality of 20; minimum base quality at a position of 20; minimum read depth at a position of 7X; minimum percentage of reads supporting the call 90%; no more than 90% , or less than 10% of reads supporting a call in the same orientation ( strand bias filter ) . SNPs in previously defined repetitive regions were excluded ( PPE and PE-PGRS genes , phages , insertion sequences and repeats longer than 50 bp ) [53] . We excluded all strains with average coverage < 15 X . Additionally , we excluded genomes with more than 50% of the SNPs excluded due to the strand bias filter , and genomes with more than 50% of SNPs with a percentage of reads supporting the call included between 10% and 90% . We filtered out genomes with phylogenetic SNPs belonging to different lineages or sub-lineages ( only for L4 ) of MTB , as this is an indication that a mix of strains could have been sequenced . To do this , we used the diagnostic SNPs obtained from Steiner et al . 2014 [75] and Stucki et al . 2016 [76] for L4 sub-lineages . We excluded all strains for which we could not find the date of isolation 1 ) in the SRA meta-information , 2 ) in the associated publications , 3 ) from the authors of the original study after inquiry . We divided all remaining strains by lineage ( L1 -L6 and M . bovis ) , and excluded strains with a number of called SNPs deviating more than three standard deviations from the mean of the respective lineage . We built SNPs alignments for all lineages including only variable positions with less than 10% of missing data . Finally , we excluded all genomes with more than 10% of missing data in the alignment of the respective lineage . After all filtering steps , we were able to retrieve 6 , 285 strains with high quality genome sequences and an associated date of sampling ( S1 Table ) . To perform a systematic analysis of the molecular clock in MTB we considered different data sets: For all data sets , we assembled SNPs alignments including variable positions with less than 10% of missing data . We inferred phylogenetic trees with raxml 8 . 2 . 11 [79] using a GTR model ( -m GTRCAT -V options ) . Since the alignments contained only variable positions , we rescaled the branch lengths of the trees rescaled_branch_length = ( ( branch_length * alignment_lengths ) / ( alignment_length + invariant_sites ) ) , Duchene and colleagues [32] showed that this method produced similar results compared to ascertainment bias correction . We then used the R package ape [80] to perform root to tip regression after rooting the trees in the position that minimizes the sum of the squared residuals from the regression line . Root to tip regression is only recommended for exploratory analyses of the temporal structure of a dataset and it should not be used for hypothesis testing [28] . A more rigorous approach is the date randomization test ( DRT ) [81] , in which the sampling dates are reshuffled randomly among the taxa and the estimated molecular clock rates estimated from the observed data is compared with the estimates obtained with the reshuffled data sets . This test can show that the observed data has more temporal information that data with random sampling times . For each dataset , we used the least square method implemented in LSD v0 . 3-beta [31] to estimate the molecular clock in the observed data and in 100 randomized replicates . To do this , we used the QPD algorithm allowing it to estimate the position of the root ( option -r a ) and calculating the confidence interval ( options -f 100 and -s ) . We defined three different significance levels for the DRT: 1 ) the simple test is passed when the clock rate estimate for the observed data does not overlap with the range of estimates obtained from the randomized sets . 2 ) The intermediate test is passed when the clock rate estimate for the observed data does not overlap with the confidence intervals of the estimates obtained from the randomized sets . 3 ) The stringent test is passed when the confidence interval of the clock rate estimate for the observed data does not overlap with the confidence intervals of the estimates obtained from the randomized sets . Bayesian molecular clock analyses are computationally demanding and problematic to run on large data sets . Therefore we reduced the thirteen largest data sets ( MTBC , L1 , L1 . 1 . 1 , L1 . 1 . 1 . 1 , L2 , L2 . 2 . 1 , L2 . 2 . 1 . 1 , L2 . 2 . 1_nc1 , L2 . 2 . 1_nc3 , L4 , L4 . 1 . 2 , L4 . 10 and M . bovis ) to 300 randomly selected strains . For each data set we used the Bayesian information criterion implemented in jModelTest 2 . 1 . 10 v20160303 [35] to identify the best fitting nucleotide substitution model among 11 possible schemes including unequal nucleotide frequencies ( total models = 22 , options -s 11 and -f ) . We performed Bayesian inference with BEAST2 [34] . We corrected the xml file to specify the number of invariant sites as indicated here: https://groups . google . com/forum/# ! topic/beast-users/QfBHMOqImFE , and used the tip sampling year as calibration . We ran four BEAST analyses with different settings: we used a relaxed lognormal clock model [37] , the best fitting nucleotide substitution model according to the results of jModelTest , and two different coalescent priors: constant population size and exponential population growth ( or shrinkage ) . We chose a 1/x prior for the population size [0–109] , two different priors for the mean of the lognormal distribution of the clock rate ( 1/x and uniform ) [10−10–10−5] , a normal ( 0 , 1 ) prior for the standard deviation of the lognormal distribution of the clock rate [0 –infinity] . For the exponential growth rate prior , we used the standard Laplace distribution [-infinity–infinity] . For all data sets , we ran at least two runs , we used Tracer 1 . 7 . 1 [82] to identify and exclude the burn-in , to evaluate convergence among runs and to calculate the estimated sample size . We stopped the runs when at least two chains reached convergence , and the ESS of the posterior and of all parameters were larger than 200 . We analyzed the complete data set of 6 , 285 genomes with the same methods described above . The only difference was that for the LSD analysis , we rooted the input tree using Mycobacterium canetti ( SAMN00102920 , SRR011186 ) as outgroup . We did this because we noticed that without outgroup , all methods placed the root on the branch separating M . bovis from all other lineages , and not on the branch separating MTB sensu stricto from the other lineages . To test the time dependency hypothesis , we repeated the LSD and BEAST analyses on the MTB complex , adding the aDNA genome sequences of three MTB strains obtained from Precolumbian Peruvian human remains [8] . These are the most ancient aDNA samples available for MTB . For LSD , we assigned as sampling year the confidence interval of the radiocarbon dating reported in the original publication . For BEAST , we assigned uniform priors spanning the confidence interval but we failed to reach convergence , therefore we used the mean of the maximum and minimum years in the confidence interval ( SAMN02727818: 1126 [1028–1224] , SAMN02727820: 1117 [1023–1211] , SAMN02727821: 1211 [1141–1280] ) . We ran three different analyses with BEAST: we used the sub-sample of 300 strains with two different priors on the clock rate ( 1/x and uniform ) , and an independent sub-sample of 500 strain , for this last data set ( 500 strains ) we assumed a HKY model and used a uniform prior on the clock rate ( S2 Table ) . To summarize the results of the BEAST analysis with the aDNA samples and retrieve the age of the MRCA of the individual lineages , we considered the analysis performed on the subset of 500 strains: we randomly sampled 5 , 000 trees from the posterior ( after excluding the burn-in ) , and calculated the Maximum clade credibility tree with the software Treeannotator v2 . 5 . 0 .
|
One of the major recent advancements in evolutionary biology is the development of statistical methods to infer the past evolutionary history of species and populations with genomic data . In the last five years , many researchers have used the molecular clock ( i . e . the observation that genomes accumulate mutations at an approximately constant pace ) to study the evolution and epidemiology of Mycobacterium tuberculosis , a bacterial pathogen that causes tuberculosis and is responsible for 1 . 6 million human deaths ever year . Applications of the molecular clock are used to understand when tuberculosis emerged as a pathogen , the evolution of drug resistance , how different strains transmit and spread across the world and how MTB populations are affected by control programs . Here , we performed a systematic analysis of the molecular clock of MTB , analyzing several whole genome sequence data sets with the same set of methodologies . We characterized the rate of molecular evolution ( the pace of the clock ) , and its variation between different MTB populations and lineages . Our results provide an important guideline for future analyses of tuberculosis and other organisms .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Methods"
] |
[
"taxonomy",
"medicine",
"and",
"health",
"sciences",
"tropical",
"diseases",
"bacterial",
"diseases",
"phylogenetics",
"data",
"management",
"extensively",
"drug-resistant",
"tuberculosis",
"phylogenetic",
"analysis",
"population",
"biology",
"bacteria",
"infectious",
"diseases",
"computer",
"and",
"information",
"sciences",
"evolutionary",
"rate",
"tuberculosis",
"molecular",
"evolution",
"actinobacteria",
"evolutionary",
"systematics",
"population",
"metrics",
"population",
"size",
"mycobacterium",
"tuberculosis",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"evolutionary",
"processes",
"organisms"
] |
2019
|
The molecular clock of Mycobacterium tuberculosis
|
Epstein-Barr virus , a B-lymphotropic herpesvirus , is the cause of infectious mononucleosis , has strong aetiologic links with several malignancies and has been implicated in certain autoimmune diseases . Efforts to develop a prophylactic vaccine to prevent or reduce EBV-associated disease have , to date , focused on the induction of neutralising antibody responses . However , such vaccines might be further improved by inducing T cell responses capable of recognising and killing recently-infected B cells . In that context , EBNA2 , EBNA-LP and BHRF1 are the first viral antigens expressed during the initial stage of B cell growth transformation , yet have been poorly characterised as CD8+ T cell targets . Here we describe CD8+ T cell responses against each of these three “first wave” proteins , identifying target epitopes and HLA restricting alleles . While EBNA-LP and BHRF1 each contained one strong CD8 epitope , epitopes within EBNA2 induced immunodominant responses through several less common HLA class I alleles ( e . g . B*3801 and B*5501 ) , as well as subdominant responses through common class I alleles ( e . g . B7 and C*0304 ) . Importantly , such EBNA2-specific CD8+ T cells recognised B cells within the first day post-infection , prior to CD8+ T cells against well-characterised latent target antigens such as EBNA3B or LMP2 , and effectively inhibited outgrowth of EBV-transformed B cell lines . We infer that “first wave” antigens of the growth-transforming infection , especially EBNA2 , constitute potential CD8+ T cell immunogens for inclusion in prophylactic EBV vaccine design .
Epstein-Barr virus ( EBV ) , a human γ-herpesvirus with potent B cell growth-transforming ability , is carried by most individuals as an asymptomatic infection yet has a remarkable potential to cause disease . Thus delayed primary infection of the immune-competent host leads in many cases to infectious mononucleosis ( IM ) , where disease symptoms are coincident with an over-active T cell response [1]; while infection of T cell-compromised or T cell-suppressed patients brings a high risk of EBV-driven B-lymphoproliferative disease ( LPD ) . Equally important , the virus is linked to a number of lymphoid and epithelial malignancies that arise as a consequence of longer-term virus carriage [2 , 3] . Collectively these EBV genome-positive tumours , including endemic Burkitt Lymphoma , many cases of Hodgkin Lymphoma , adult T/NK cell lymphoma , nasopharyngeal carcinoma and a subset of gastric carcinomas , impose a global disease burden of ~200 , 000 new cancer cases per year [4] . EBV infection is also implicated as a major environmental risk factor for the development of various autoimmune conditions , especially Multiple Sclerosis [5] . Viewed in this light , there is a compelling case for a prophylactic vaccine that could either prevent EBV infection altogether or , at least , reduce disease risk by lowering the set-point of long-term virus carriage [6 , 7] . However there is uncertainty about which viral antigens to include in such a vaccine , not least because the very early events of in vivo infection are poorly understood . The virus is orally transmitted and replicates in the oropharynx , probably in mucosal epithelium and/or locally-infiltrating B cells , while colonising the lymphoid system via the expansion of virus-transformed cells before entering a true ( antigen-negative ) latency in the recirculating memory B cell pool [8] . Arguably all EBV-associated diseases depend directly or indirectly on this initial colonisation of the B cell system [9]; thus all available evidence suggests that the virus does not persist at epithelial sites and that all non-B cell-derived tumours arise from infections acquired by reactivation from the latent B cell reservoir . The aim of a prophylactic vaccine must therefore be to prevent or limit the virus’ initial colonisation of the B cell system . Most interest in that regard has focused on the initial process of B cell infection , where viral attachment is mediated by the major envelope glycoprotein gp350 binding to the complement receptor CD21 on the B cell surface . Thereafter viral entry involves binding of a co-receptor ( HLA class II ) by the gp85/gp25/gp42 glycoprotein complex , and a subsequent envelope fusion event requiring gp85/gp25 and another viral glycoprotein gp110 [10] . Vaccine constructs based on gp350 , known to be the dominant target of the neutralising antibody response , have been tested in primate models [11] and more recently in man [12 , 13] . In the only phase II clinical trial , a recombinant gp350 vaccine given to EBV-naïve adolescent volunteers apparently did not lower the chances of becoming infected but did reduce the number of primary infections manifesting as IM [13] . However , the mechanism underlying disease protection ( neutralising antibodies and/or T cell-mediated immunity ) and the impact of vaccination on the long-term viral load in the B cell system were not addressed . A small trial in monkeys studying the EBV homologue , rhesus lymphocryptovirus , found that gp350 vaccination reduced both incidence of infection and long-term viral load after oral challenge . Furthermore , the inclusion of EBNA3A/3B in vaccine constructs induced T cell responses which appeared to mediate additional protection [14] . EBNA3A was also targeted in the sole phase I trial of a CD8+ T cell peptide epitope-based vaccine for IM [15] . We inferred from the work to date that a gp350-based EBV vaccine , even optimised to induce higher neutralising antibody responses [16] might be further improved by eliciting T cell responses , in particular CD8+T cell responses , capable of recognising and killing recently-infected B cells in the very early stages of virus-induced transformation . There are three possible sources of such T cell target antigens: In the present work , we set out to detect CD8+ T cell responses against these “first wave” transforming proteins , to identify their target epitopes and HLA restricting alleles , and to determine how well they recognised B cells in the days following EBV infection , comparing such recognition with that shown by CD8+ T cells against the other two sources of candidate vaccine antigens .
In a first series of experiments we screened a panel of 20 healthy EBV-seropositive donors for T cell responses to EBNA2 , EBNA-LP and BHRF1 . Peripheral blood mononuclear cells ( PBMCs ) were stimulated with peptide pools comprised of overlapping peptides spanning the complete unique amino acid sequence of each of these proteins . Following 7 days of culture in cytokine supplemented medium , the resultant polyclonal T cell populations were screened for recognition of peptide sub-pools; recognition was assessed by IFNγ production measured in Enzyme-linked Immunosorbent assays ( ELISAs ) . In these initial screening experiments we analysed both CD4-selected and CD4-depleted ( i . e . CD8-enriched ) T cell populations to determine the overall pattern of responses . CD8+ T cell responses to EBNA2: Fig 1A and S1A Fig illustrate the screening strategy for EBNA2-specific responses and the characterisation of novel CD8+ T cell epitopes . In the first example , CD8-enriched T cell populations from Donor 17 were screened against 18 peptide sub-pools representing the unique 487 amino acid sequence of EBNA2 ( B95 . 8 strain ) ; a single response was detected to peptide ( s ) within pool 6 . Subsequent screening against individual peptides from this pool identified two overlapping peptides , 6 . 4 and 6 . 5 that mediated T cell recognition ( Fig 1A , upper panels ) . The peptide 6 . 4/6 . 5-specific polyclonal population was cloned by limiting dilution and the resultant CD8+ T cell clones used to determine the HLA restriction of this response . Fig 1A ( middle panel ) illustrates the results obtained for one representative T cell clone screened against HLA class I-matched LCLs , pre-loaded with peptides 6 . 4/6 . 5 to maximise recognition; only LCLs sharing HLA-B7 were recognised . The sequence QPRLTPPQPL located within the overlapping region of peptides 6 . 4 and 6 . 5 conforms well to the defined HLA-B7 peptide motif . This 10-mer peptide was recognised by both the polyclonal CD8-enriched T cell population ( Fig 1A , table ) and specific T cell clones . Results from screening a second donor ( Donor 8 ) against the EBNA2 peptide sub-pools are shown in S1A Fig; here the response mapped to the 9mer TSSPSMPEL presented by HLA-C*0304 . In total , 19 donors were screened against the EBNA2 peptide sub-pools . CD8+ T cell responses for individual donors are detailed in S1 Table; all EBNA2-derived CD8 responses detected are summarised in Table 1 . These include three previously described epitopes , as well as 5 novel CD8+ T cell responses , the HLA class I restriction determinants were identified for three of these responses and two minimal epitopes were defined . CD8+ T cell responses to EBNA-LP: EBNA-LP ( B95 . 8 strain ) contains only 110 residues of unique sequence , being comprised of multiple copies of a 66 amino acid repeat domain upstream of a unique 44 amino acid carboxy-terminus . Screening 18 of the above 20 donors , with a range of HLA class I types , against peptide sub-pools covering the whole EBNA-LP sequence gave only a single positive result . The CD8-enriched T cell population from Donor 20 recognised peptide ( s ) within pool 3 , and this response mapped to peptide 3 . 6 , which encompasses the previously identified SLR epitope [32] , EBNA-LP amino acids 284–292 ( Fig 1B ) , lying within the unique C-terminal domain . Donor 20 indeed turned out to be HLA-A*0203-positive and the response proved to be HLA-A*0203-restricted . CD8+ T cell responses to BHRF1: Prior to this work , the CD8+ T cell response to BHRF1 had not been characterised . BHRF1 is another relatively small protein which , in the B95 . 8 strain of EBV , is comprised of 191 amino acids . Screening 18 donors against 6 BHRF1 peptide sub-pools detected responses to two CD8+ T cell epitopes . As illustrated in Fig 1C , the CD8-enriched T cell population generated from Donor 8 recognised peptide ( s ) contained within pool 1; this response mapped to peptides 1 . 3/1 . 4 . HLA restriction analysis identified B*3901 as the restriction determinant; the peptide SRVHGNGTL contained within the 10 amino acid sequence common to peptides 1 . 3/1 . 4 conforms well to the B*3901 peptide motif and was identified as the minimal epitope . Results for a different donor mapping the second BHRF1-derived CD8 response to the 9mer ETFTETWNR , presented by HLA-A68 , are shown in S1B Fig . T cell clones specific for the two BHRF1-derived epitopes , as well as the two minimally-defined EBNA2 epitopes , recognised their respective proteins expressed from a recombinant vaccinia virus ( rVV ) confirming their antigen as well as epitope specificity ( S1C and S1D Fig ) . BHRF1-specific CD8+ T cell responses are summarised in Table 1 and detailed for individual donors in S1 Table . CD4+ T cell responses: In parallel with the above work , donors were screened for CD4+ T cell responses to these three “first wave” antigens; several new responses were observed . Illustrative results for one donor per protein are shown in Fig 2 . Thus for EBNA2 , the CD4-selected T cell population from Donor 14 recognised peptides from two adjacent sub-pools , 15 and 16; splitting these pools into their component peptides narrowed the response down to two overlapping peptides 15 . 5 and 16 . 1 , which share the sequence VCRNSHTATPNVSPI ( Fig 2A ) . Of 19 donors included in the screening , 17 ( ~ 90% ) had detectable EBNA2-specific CD4+ T cell responses ( S1 Table ) , although such responses were not always characterised beyond the peptide pool ( s ) recognised . The two responses defined at the individual peptide level are included in Table 2 , which summarises all EBNA2-derived CD4 responses detected in this study . We found that , despite its small size , EBNA-LP was also a relatively frequent target of CD4+ T cell responses , often directed against a shared immunodominant epitope . Thus 7/18 donors recognised EBNA-LP sub-pool 2 and , of these , all three donors screened further recognised peptide 2 . 3 , sequence QEPRRVRRRVLVQQE . This is illustrated for one representative donor ( Donor 5 ) in Fig 2B . Responsive donors do not share a common class II allele , suggesting that this peptide is presented by multiple class II antigens . Finally , screening 18 donors against the BHRF1 peptide pools identified one novel and one known CD4 epitope response ( Table 2 ) . For example , as shown in Fig 2C the CD4-selected T cell population from Donor 3 recognised peptides within pools 3 and 5 . Analysis of individual pool 3 peptides mapped the response to peptide 3 . 1 , sequence NSETFTETWNRFITH; the pool 5 response mapped to peptide 5 . 1 , which corresponds to the published PYY epitope [33]; BHRF1-specific CD4 responses for individual donors are summarised in S1 Table . Fig 3 summarises the mapping of EBNA2 , EBNA-LP and BHRF1 responses to single-epitope regions , showing the primary sequence of each protein according to its relative size and identifying the positions of all defined CD8 and CD4 epitopes within each sequence . Focussing initially on EBNA2 , it can be seen that whilst both CD8 and CD4 epitopes are distributed throughout the length of the protein , there are “hotspots” where ( i ) CD8 or CD4 epitopes are co-localised ( e . g . the two newly defined B7-restricted epitopes between amino acids 181 and 200 ) , or ( ii ) CD8 and CD4 epitopes overlap ( e . g . the B38-restricted YHL epitope and the DR4-restricted GQT epitope ) . The one EBNA-LP-derived CD8 T cell epitope lies within the short unique carboxy-terminal sequence whereas the one CD4 epitope is present in each copy of the repeat domain . For BHRF1 , the two defined CD8 epitopes lie towards the amino-terminus; CD4 epitopes are distributed throughout the protein , with the one newly defined epitope encompassing the CD8 A68/ETF epitope . Having mapped the above CD8+ T cell epitopes using in vitro-expanded effector populations , we sought to determine ( i ) the relative size of such responses in PBMCs ex vivo , versus those seen against well-defined EBV epitopes ( derived from latent or lytic cycle antigens [31] ) , and ( ii ) the fraction of donors expressing the relevant HLA class I allele who responded to each epitope . PBMCs were therefore screened in IFNγ Elispot assays against new and previously defined epitope peptides appropriate for each donor’s HLA class I type . Examples of results from individual donors are shown in Fig 4A and the overall results summarised in Fig 4B , with responses categorised as immunodominant ( numerically greater than other measured responses ) , co-dominant ( equivalent to other measured responses ) or subdominant ( smaller than other measured responses ) . Several of the “first wave” protein-derived epitopes induced strong responses . For example , the EBNA2-derived RPT/B*5501 response in Donor 7 ( A2/B55-positive ) was comparable with that to the immunodominant GLC/A2 epitope from the BMLF1 lytic cycle protein [27] ( Fig 4A , second left panel ) . Likewise , in one of the two B57-positive subjects ( Donor 11 , second right panel ) , the EBNA2-derived LAS/B57 epitope induced the strongest response , higher even than that seen against the previously described immunodominant B57/58 epitope , VSF from EBNA3B [32] . Fig 4B not only summarises the strength of responses to EBNA2 , EBNA-LP and BHRF1 epitopes but also shows the frequency with which individuals with the appropriate HLA class I allele make a detectable epitope response . Although there is wide variation in both parameters , there is a general trend for immunodominant or co-dominant responses ( e . g . B*3801/YHL and B*5501/RPT ) to be frequently seen in the appropriate donors , and for numerically sub-dominant responses ( e . g . B7/QPR and Cw3/TSS ) to be less often detectable . As described above , EBNA2 , EBNA-LP and BHRF1 are amongst the first viral proteins expressed in B cells following EBV infection and thus represent potential targets for specific T cell recognition prior to cell cycle entry , blast transformation and the establishment of latency . In subsequent experiments we therefore wanted to determine how closely T cell recognition follows the kinetics of antigen expression . The key question being addressed is whether T cell mediated control can be exercised in the initial 24–48hrs post-B cell infection , before the first cell division and prior to LMP1-mediated activation of antigen processing and presentation pathways [29 , 30] . For the analysis of gene expression , B cells isolated from buffy coat cells were exposed to EBV , cultured and then harvested at time points between 6hrs and 14 days post-infection . mRNA levels were quantified using a 48:48 Dynamic Array IFC-Gene Expression system and a plasmid standard containing a single copy of each of 45 EBV and 3 cellular amplicons [34]; this assay enables the absolute quantification of EBV transcript levels . Focusing initially on EBNA2 , results for one representative experiment , here comparing the kinetics of expression of EBNA2 , EBNA3B and LMP2 , are shown in Fig 5A . EBNA2 transcripts were detected as early as 6hrs post-infection , peaked to a very high level within 12hrs , then fell sharply by day 2 before reaching plateau levels . EBNA3B expression was slightly delayed relative to EBNA2 , becoming first detectable at 12hrs and reaching low maximal levels around days 2–3 . LMP2 transcripts were first detected at very low levels on day 2 and plateaued around day 5 . This temporal sequence of latent gene expression essentially accords with previously published data [34–36]; however , as recently reported [34] and apparent from the different scales used in Fig 5A , there are also major differences in the levels of these latent gene transcripts . Thus , even as early as 6hrs post-infection , EBNA2 transcript levels exceeded the maximum levels detected for both EBNA3B and LMP2 . Indeed peak EBNA2 transcript levels seen at 12–24hrs were ~100-fold higher than the EBNA3B or LMP2 maximum and were still >10-fold higher than these at day 10 with the establishment of latency . To complement the mRNA analysis , protein expression was analysed by immunoblotting ( Fig 5B ) . EBNA2 was detectable 12hrs post-infection , reaching peak levels within 24hrs . EBNA3B expression was again slightly delayed relative to EBNA2 , being detectable at very low levels 12hrs post-infection and maximally around day 2; low level LMP2 expression was detectable from day 3–4 onwards . For T cell recognition assays primary B cells were isolated from healthy adult donor PBMCs of appropriate HLA class I/II type , infected with EBV and then co-cultured with specific T cell clones in medium supplemented with interleukin-2 ( IL-2 ) . Culture supernatant was sampled at regular intervals between days 1 to 14 post-infection and the IFNγ concentration quantified by ELISA as a measure of T cell recognition . Illustrative results for one experiment , evaluating CD8+ T cell recognition of EBNA2 , EBNA3B and LMP2 are shown in Fig 5C ( left panels ) ; recognition of established LCLs is shown for reference ( right panels ) . CD8+ T cells specific for the EBNA2-derived YHL epitope recognised EBV-infected B cells at the earliest assay time point , day 1 post-infection; similar results were obtained using CD8+ clones specific for a second EBNA2-derived epitope ( RPT , Fig 6C ) . By contrast , recognition of EBNA3B ( represented by the IVT epitope ) was first detectable from day 2 post-infection and increased thereafter; CD8+ T cell clones specific for epitopes derived from EBNA3A , 3C and EBNA1 gave similar results to EBNA3B ( S2A Fig ) . Recognition of LMP2 ( represented by the SSC epitope ) was further delayed , only reaching detectable levels on day 6 . Absolute levels of IFNγ production varied between individual new-infection experiments ( for example compare top panels of Fig 5C and 5D ) , and were generally lower than those induced by established LCLs . However , the pattern of CD8+ T cell recognition for individual latent proteins was consistent and broadly correlates with the kinetics of antigen expression . It is interesting to compare such recognition kinetics with those seen using CD4+ T cell clones specific for these antigens , for example EBNA2 . Though EBNA2 is highly expressed within the first 2 days post-infection , CD4+ T cell recognition of this protein is essentially undetectable at such early times and appears only gradually thereafter ( Fig 5D ) . EBNA-LP is also known to be highly expressed at the protein level in newly-infected B cells [36 , 37] , but comparable experiments with CD8+ effectors against the EBNA-LP-derived SLR epitope could not be conducted because of limited access to PBMCs from A*0203-positive subjects . However , we were interested to extend this work to study the kinetics of BHRF1-specific CD8+ T cell recognition in relation to expression of this protein during the B cell transformation process . Note that BHRF1 is traditionally considered to be an early lytic cycle antigen , but can also be expressed as a latent gene product from transcripts initiating from Wp , the promoter that initially drives both EBNA2 and EBNA-LP expression [36] . Fig 6A shows the expression levels of such BHRF1 and EBNA2 mRNAs as determined by 48:48 Dynamic Array QPCR assay in one representative experiment . Latent BHRF1 transcripts were maximally detected on day 1 , then temporarily declined , recovering slightly from day 4 onwards . Though absolute mRNA values differed between individual transformation experiments , peak BHRF1 transcript levels were usually ~10-fold lower than those for EBNA2 . Furthermore at the protein level , immunoblotting showed that ( unlike EBNA2 ) the BHRF1 protein was barely detectable at day 1 and increased only gradually with time . Even at 7 days the protein had not achieved the stable levels that can be detected in established LCLs , whether they carry replication-competent ( wild-type ) or replication-defective ( BZLF1-knockout ) virus ( Fig 6B ) . Results for one representative T cell assay , comparing recognition of newly infected B cells by CD8+ effectors specific for BHRF1 and EBNA2 , are illustrated in Fig 6C . Here , T cell clones specific for the EBNA2-derived RPT epitope recognised EBV-infected B cells from day 1 ( lower panel ) . In contrast , BHRF1-specific T cells showed no recognition at early time points; recognition only becoming detectable , albeit at low levels , by day 14 ( upper panel ) . This late level of recognition is equivalent to that observed for established LCLs , again whether carrying wild-type or BZLF1-knockout virus ( Fig 6D ) , suggesting that such recognition is predominantly mediated by BHRF1 expressed as a latent cycle antigen . We infer that BHRF1 , in contrast to EBNA2 , is a poor target antigen for CD8+ T cell recognition early post-B cell infection , likely reflecting the low level of BHRF1 protein expression seen at these times . In a next series of experiments we analysed the kinetics of T cell recognition of lytic cycle antigens . It has been reported previously that IE/E lytic cycle proteins , transiently expressed in newly infected B cells from viral mRNAs carried within mature virus particles , constitute target antigens for early CD8+ T cell recognition [24–26] . We sought to replicate these results using our experimental protocols . Fig 7A shows results for one 48:48 Dynamic Array QPCR assay , measuring transcript levels for representative IE ( BZLF1 ) , E ( BMLF1 ) and L ( BALF4/gp110 ) lytic cycle antigens . All lytic cycle transcripts were detected at very low levels throughout the 14 day time course , although a small peak of expression was observed around day 5 . Of note , transcript levels on day 1 post-infection were >100-fold lower than those for EBNA2 ( compare Fig 7A with Figs 5A and 6A ) . The same panel of lytic cycle proteins was assessed as target antigens for CD8+ T cell recognition on days 1–14 post-EBV infection of B cells; results for one representative experiment are shown in Fig 7B . T cell recognition was undetectable at early time points , becoming measurable around days 6–7 , and then increasing over time to day 14 . This pattern of recognition was reproduced in experiments using T cell clones specific for four different IE/E lytic antigen-derived CD8 epitopes . Absolute levels of T cell recognition at late time points differed between separate experiments , presumably reflecting variable numbers of B cells spontaneously entering lytic cycle at these later times . Lytic cycle proteins therefore appear to constitute poor targets for immediate recognition by CD8+ T cells following EBV infection of B cells . However , protein components of the viral envelope and capsid may be efficiently processed via the class II pathway for immediate recognition by CD4+ T cells . Indeed , it has previously been shown that glycoprotein-specific CD4+ T cells can recognise B cells at early time-points post-EBV infection [17] . As comparators in the CD8 recognition experiments , we likewise assayed for CD4+ T cell recognition of newly infected B cells using effectors specific for the viral glycoproteins gp85 , gp110 and gp350; representative results are shown in Fig 7C . CD4+ T cell clones specific for epitopes derived from gp85 and gp350 recognised EBV-infected B cells at the earliest assay time point , 1 day post-infection; similar kinetics were seen in other experiments probing with gp110-specific CD4+ effectors ( Fig 8A , bottom right panel ) . Fig 8 summarises all the data for CD8+ and CD4+ T cell recognition of primary B cells 24hrs post-EBV infection . EBNA2 is efficiently processed and presented for CD8+ T cell recognition at this time , whereas BHRF1 and IE/E lytic cycle antigens appear not to be . Furthermore , CD8+ T cells specific for gp110 also failed to detect recently-infected B cells , arguing against cross-presentation of viral envelope glycoproteins to CD8+ effectors . For CD4+ T cells the pattern of results is quite different . CD4+ T cell clones specific for epitopes derived from latent , or IE/E lytic cycle proteins recognised B cells loaded with the appropriate synthetic peptide , but failed to recognise EBV-infected B cells . However , CD4+ T cells specific for epitopes derived from glycoproteins that are components of the incoming virion ( L lytic ) recognise newly infected B cells strongly within 24hrs . A final set of experiments compared the ability of EBNA2 versus other latent antigen-specific CD8+ T cell clones to inhibit B cell transformation and LCL outgrowth following EBV infection in vitro . Primary B cells were isolated from PBMCs , infected with EBV at two virus doses ( neat and 1:100 ) , and then both infected populations co-cultured with specific CD8+ T cells at a range of effector:target ratios ( between 2:1 and 0:1 ) . Cultures were maintained by weekly re-feeding and scored visually for LCL outgrowth at ~4 weeks . In each case the B cell donor was selected to allow HLA class I matching of the same infected B cells with T cell clones against the different latent antigens . Representative results from three such experiments are illustrated in Fig 9A , here comparing inhibition of LCL outgrowth by CD8+ effectors specific for epitopes derived from EBNA2 ( YHL and RPT ) , EBNA3B ( IVT and AVF ) and LMP2 ( TYG ) ; recognition of an established LCL from the B cell donor is included for reference . Combined results for all experiments are shown in Fig 9B . In this experimental setting EBNA2-specific T cells appeared to be as efficient , and often more efficient , than EBNA3-specific effectors in their ability to control B cell transformation and LCL outgrowth , especially at higher virus doses; LMP2-specific effectors , at least as represented by TYG-specific T cell clones , were reproducibly poorer in this regard .
An effective prophylactic EBV vaccine should aim to prevent , or at least limit colonisation of the B cell system , the process that is essential for virus persistence and central to the development of most , if not all , EBV-associated disease . An ideal vaccine might induce both neutralising antibodies to reduce levels of infection and T cell responses to target B cells that do become infected . Proteins that are expressed in the very early phase of B cell transformation , including EBNA2 , EBNA-LP and BHRF1 , constitute potential vaccine immunogens for the induction of CD8+ T cell responses . Here we have ( i ) determined to what extent natural EBV infection elicits CD8 responses to these “first wave” proteins , including identification of target epitopes/HLA restricting alleles and assessment of relative immunodominance; ( ii ) assayed the ability of such T cells to recognise B cells expressing these proteins in the very early phase of infection , prior to cell cycle entry; ( iii ) compared such recognition with that shown by CD8+ T cells against other potential targets of early detection i . e . IE/E lytic antigens , that may be expressed from mRNAs contained within the virion [22–26] , and virus structural proteins delivered into B cells upon infection; and ( iv ) assessed the ability of such effectors to inhibit EBV-induced B cell transformation and LCL outgrowth . To characterise the immunogenicity of “first wave” proteins , PBMCs from healthy seropositive donors were stimulated with peptides representing EBNA2 , EBNA-LP or BHRF1 , and T cell responses assessed by cytokine production following a brief period of in vitro expansion to maximise detection sensitivity . The relative magnitude and frequency of “first wave” responses were then compared with previously defined latent/lytic cycle epitope responses using ex vivo assays . This work clearly identified EBNA2 as a strong CD8 immunogen in certain HLA contexts , for example B*3801 and B*5501 . Likewise EBNA-LP and BHRF1 , although relatively infrequently targeted by CD8+ T cells among donors in our panel , both contain epitopes that are immuno-/co-dominant in the appropriate HLA context . This mirrors the situation with respect to another latent protein , EBNA1 , which despite the influence of its internal glycine-alanine repeat domain elicits strong CD8 responses in the context of particular HLA alleles such as HLA-B*3501 and -B53 [38] . It would therefore appear that all endogenously expressed EBV latent antigens ( with the possible exception of LMP1 [39 , 40] ) are accessible to the MHC I antigen processing pathway and can induce strong CD8 responses at least in a subset of individuals . We infer that the marked immunodominance of the EBNA3A , 3B , 3C family noted in early studies [41 , 42] can , at least in part , be explained by two contributory factors . One is the sheer size of the EBNA3 proteins; thus EBNA3A , 3B and 3C together provide around 60% of the total unique amino acid sequence content of all the latent proteins , compared to ~9% for EBNA2 , ~2% for EBNA-LP and ~4% for BHRF1 . Another is the fact that most studies of EBV-induced CD8+ T cell responses to date have involved Caucasian subjects and some of the strongest responses restricted through common Caucasian HLA I alleles ( e . g . HLA-B7 , -B8 and -B44 ) happen to involve EBNA3-derived epitopes . By comparison , CD8+ T cell responses to “first wave” proteins that are restricted through common Caucasian class I types are often subdominant and immunodominant responses are restricted through less common alleles . We also note that , with one exception , all of the novel CD8+ T cell epitopes described herein are restricted through HLA-A or -B alleles . Indeed the exception , the HLA-C*0304-restricted TSS epitope , is only the second EBV latent antigen-derived response found to be restricted through an HLA-C allele [43]; by contrast there are at least six known EBV lytic antigen-derived responses presented by HLA-C alleles [20 , 21 , 31] . The apparently higher incidence of HLA-C-restricted responses to lytic cycle antigens may reflect the influence of the immune-evasin BILF1 , whose expression during early lytic cycle selectively down regulates antigen presentation via HLA-A and -B , but not HLA-C [44] . Whilst the main focus of this study was CD8+ T cell responses , CD4+ responses to the three “first wave” antigens were analysed in parallel . In agreement with previously published work [45] , EBNA2 was found to be a frequent target antigen for CD4+ T cell responses ( 17/19 donors responsive ) . Although there are at least nine defined EBNA2-derived CD4 epitopes ( Fig 3 ) , a significant part of this response is focused on a single epitope ( PRS ) , which was detected in more than half of subjects , and is presented through multiple common class II allotypes including DR7 and DR52b [45] . In the present work , a single novel CD4 epitope was identified from EBNA-LP ( QEP ) and detected in up to 40% of donors . This response was not fully characterised but , like the PRS response , appears to be promiscuously presented , being detectable in donors of disparate class II type . The QEP epitope derives from the repeat region of EBNA-LP; therefore multiple copies are present within the protein , which may also contribute to the observed immunodominance . Our next key objective was to examine T cell recognition of “first wave” proteins during the process of B cell transformation , using well-established target antigens such as EBNA3B and LMP2 as comparators . To achieve this we used a short-term co-culture protocol which ( i ) minimises the number of B cells required , thus allowing the analysis of T cell responses restricted through less common HLA class I types , as well as combinations of multiple class I/II alleles; and ( ii ) removes counting errors inherent within the set-up of individual assays on a daily basis . One potential disadvantage of the assay design is that T cells of certain specificities ( e . g . LMP2 ) are in co-culture for several days before their cognate antigen is expressed . However , since T cell clones are routinely maintained with only periodic re-stimulation , we consider it unlikely that this would result in effectors becoming unresponsive within the assay time-frame . Relatively high virus doses were used in these assays to ensure that a majority of B cells were infected and thus had the potential to express genes/proteins of interest and to function as T cell targets . Since the mean number of EBV genomes acquired per cell remains low [46] , this increases the sensitivity of detection , but does not disproportionately increase gene expression/de novo protein synthesis on a per cell basis . IFNγ production was used as the read out for specific T cell recognition , as we find this to be the most sensitive cytokine in the EBV system . To determine how closely T cell recognition follows the kinetics of antigen expression , co-culture assays were combined with new assays quantitating latent/lytic transcript levels and with immunoblotting to assess the appearance of latent proteins . Our previous measurements of EBV gene expression using conventional QRT-PCR assays [35 , 47 , 48] delineated expression patterns for individual transcripts , but could not accurately measure the abundance of different transcripts relative to each other because of using different standards . The plasmid standard , containing a single copy of 45 EBV and 3 cellular control amplicons , used here in the 48:48 Dynamic Array QPCR assay allows absolute quantitation , thus enabling the direct comparison of transcript levels for different genes at a single time point . Using these assays we found EBNA2 to be well recognised by CD8+ effectors very early post-infection; levels of CD8+ T cell recognition at 24hrs were higher than for all subsequent time points mirroring the peak of Wp driven EBNA2 transcription and closely correlating with new EBNA2 protein synthesis . Thus even minimally activated B cells are able to process antigen for CD8 epitope display , prior to the first cell division and in advance of LMP1 expression ( detectable from ~day 3/4; [36] ) and up-regulation of antigen presentation pathways [29 , 30] . The independence of CD8+ T cell recognition from cell cycle entry was further emphasised in experiments using a recombinant virus deleted for EBNA2 . B cells infected with EBNA2-KO virus do not enter cell cycle , but overexpress EBNA1 and the EBNA3 proteins from day 1 post-infection [48] , and indeed we found that these B cells are well recognised by EBNA1- and EBNA3-specific CD8+ T cell clones at these atypically early times ( S2B Fig ) . However , in experiments using wild-type virus , CD8+ T cells specific for EBNA1 and the EBNA3A , 3B , 3C antigens only recognised newly infected B cells at low levels from ~2 days post-infection and this recognition increased with time ( Fig 5C and S2A Fig ) . The apparent delay in EBNA3-specific T cell recognition compared with mRNA expression in the present experiments may relate to the surprisingly low transcript levels for these proteins ( Fig 5A ) . Low expression might lead to increased dependence on B cell activation and the up-regulation of antigen presentation pathways and/or increased susceptibility to the effects of co-expressed immune-evasins [25 , 26] . By contrast , the much higher expression levels of EBNA2 might counteract any such immune-evasion strategies , allowing sufficient representation of EBNA2-derived epitopes on the B cell surface to mediate specific CD8+ T cell recognition . Note that , for CD8+ T cells of all specificities , recognition of established LCLs is superior to that of newly-infected B cells , suggesting that antigen presenting function becomes optimal in the fully growth-transformed state . As a measure of the biological effectiveness of early B cell recognition by EBNA2-specific effectors we compared their ability to inhibit B cell transformation and LCL outgrowth with that of other latent antigen-specific CD8+ T cells . Importantly , the hierarchy of different antigen specificities in their control of early B cell outgrowth ( i . e . EBNA2 ≥ EBNA3 > LMP2; Fig 9 ) closely correlates with the level and timing of target antigen expression in the initial days post-infection . In contrast to the above work with EBNA2-specific CD8+ T cell clones , there was no detectable BHRF1-specific recognition at early time points post-infection , even though CD8+ effectors had high functional avidity as measured by peptide titration assays ( S3 Fig ) . Instead , the T cell data show slow accruement of BHRF1-specific recognition during the transformation process . At later time points some fraction of this may result from small numbers of B cells entering lytic cycle , however , most appears to reflect the low but progressive expression of BHRF1 as a latent antigen . Thus LCLs made with a BZLF1-KO virus continue to express the protein and are recognised by BHRF1-specific CD8+ effectors at similar levels to WT-LCLs . We infer that low levels of protein expression at very early time-points , although able to mediate essential anti-apoptotic effects [22 , 36] , are insufficient to allow detectable display of BHRF1-derived epitopes at the B cell surface . Unfortunately parallel experiments using EBNA-LP-specific CD8+ effectors could not be carried out due to limited availability of A*0203-positive B cells . However , we would predict similar results to those seen for EBNA2 for two reasons: ( i ) EBNA-LP transcripts like those for EBNA2 initially derive from the Wp promoter which reaches peak activity ~12hrs post-infection and then declines [36] , and ( ii ) EBNA-LP protein is highly expressed by day 1 post-infection at levels easily detectable by immunoblotting [36 , 37] . An alternative antigen source for early CD8+ T cell recognition derives from IE/E lytic cycle gene products whose reported expression immediately post-infection [22–25] has recently been attributed to the translation of viral mRNAs transduced during B cell infection [26] . This is of particular interest because several such proteins constitute immunodominant targets for CD8 responses [20 , 27] . However , in our assays we did not detect early CD8+ T cell recognition of IE/E lytic antigens ( including BZLF1 , BRLF1 and BMLF1 ) , despite using high avidity clones that showed good recognition of these antigens when expressed in B cells during lytic replication . The absence of T cell recognition at early time points accords with mRNA quantitation results where the relevant transcripts are detected at very low levels , if at all . Single cell staining suggests that the small peak of BZLF1 expression detected around Day 5 represents a minority of B cells entering lytic cycle , which fraction may be insufficient to mediate T cell recognition at this time . Expression of IE/E lytic cycle proteins immediately post-infection will have opposing effects on CD8 recognition: proteins may be processed and presented via the class I pathway and behave as T cell targets; alternatively , those with immune-evasion functions may inhibit recognition of self and any co-expressed antigens . Thus in a previous study [25] CD8+ effectors recognised B cells infected with a recombinant virus deleted for BNLF2a ( an inhibitor of the transporter associated with antigen processing [49] ) at day 1 post-infection , but not wild-type virus-infected B cells; however , none of the “first wave” proteins were investigated as target antigens . The final potential source of CD8 epitopes for early T cell recognition considered here comprises of viral structural proteins . Opportunities to study this were limited because late antigen-specific CD8+ T cells are rarely detectable in healthy virus carriers , and until recently only a limited number of epitopes were defined [20 , 31] . Cross-presentation of virion component proteins for CD8+ T recognition was not observed for the single gp110-derived epitope tested . However , other virus structural antigens such as capsid proteins , that naturally enter the cytosol during transport to the nucleus , may differentially access the class I processing pathway . Work in our laboratory has recently identified several new responses to L antigens including tegument and capsid as well as envelope proteins [21]; the potential of such antigens to be cross-presented for CD8+ T cell recognition remains to be determined . Immune correlates for protection from EBV-associated disease are not fully understood . To date , vaccines that have shown promise in clinical trials for the prevention of IM have focused on gp350 [12 , 13] , the major target of the neutralising antibody response . However it is not known to what extent such vaccines exert their effects through induction of neutralising antibodies and/or T cell-mediated immunity . CD4+ T cells specific for gp350 ( and other virion glycoproteins ) are capable of recognising B cells at very early time-points post-EBV infection in vitro ( [17] and Figs 7 and 8 ) and may well contribute towards vaccine-induced protection . Vaccine efficacy might be further improved by eliciting CD8+ T cell responses capable of recognising and killing recently-infected B cells at the very early stages of virus-induced transformation . Our data suggest that at least one “first wave” protein EBNA2 , and possibly a second , EBNA-LP , have the potential to serve as such candidate immunogens . This study focussed for the most part on Caucasian HLA types; in the future it would be interesting to look for CD8+ T cell responses restricted through class I types present in non-Caucasian populations ( like the A*0203-restricted EBNA-LP response ) which could increase vaccine range .
All experiments were approved by the West Midlands—Black Country NRES Committee ( 07/Q2702/24 ) . All donors provided written informed consent for the collection of blood samples and subsequent analysis . Overlapping peptides ( 20mers overlapping by 15aa or 15mers overlapping by 10aa ) spanning the complete unique sequences of EBNA2 , EBNA-LP and BHRF1 were dissolved in DMSO . Peptides were combined into pools comprising 5–6 adjacent peptides , each at a concentration of 100μg/ml . Overlapping and minimal epitope peptides were synthesised by Alta Bioscience , Abingdon , UK or Peptide 2 . 0 , Virginia , USA . PBMCs were isolated from whole blood by density gradient centrifugation and incubated with peptide pools at a final concentration of 1μg/ml in RPMI containing 8% human serum ( HuS; TCS Biosciences Ltd , ) for 1½ hours . Peptide loaded PBMCs were washed , and then plated out in 24 well plates at a maximum density of 5 million cells/well in RPMI/HuS supplemented with 10ng/ml IL-7 ( Peprotech ) . IL-2 ( Novartis Pharma ) was added on day 3 at a final concentration of 20IU/ml . Polyclonal T cell populations were screened for peptide specificity after 7 days of culture . Dependent on available cell numbers total and/or CD4-positive as well as CD4-depleted polyclonal T cell populations were screened against peptide sub-pools in duplicate . CD4-positive T cells were isolated using Dynabeads according to the manufacturer’s instructions . The average purity of T cell populations , where sampled , was: CD4-positive <1% CD8+; CD4-depleted <8% CD4+ . 100μl of T cells were plated out in 96V-well plates at cell densities between 50 , 000 and 175 , 000 cells/well; 100μl of peptides were then added at a final concentration of 1μg/ml . Co-cultures were incubated at 37°C overnight . Specific peptide recognition was measured by IFNγ ELISA ( Perbio ) according to the manufacturer’s protocol . T cell clones were obtained by limiting dilution from peptide-stimulated polyclonal populations or from polyclonal cultures stimulated with the autologous LCL , as previously described [50] . Briefly , T cells were plated out at 3 or 0 . 3 cells/well in 96U well plates with 105 PHA-treated irradiated allogeneic PBMCs and 104 irradiated autologous LCL cells or 10ng/ml OKT3 monoclonal antibody . All growing wells were screened for recognition of the appropriate peptide by IFNγ ELISA; specific T cell populations were expanded using feeder cells and LCL and maintained in RPMI medium containing 10% FCS , 1% HuS , 25% MLA 144 supernatant and 20IU/ml rIL-2 . LCLs sharing one or more HLA class I alleles with the donor of interest were pre-loaded with peptide ( 1μg/ml , 1hour incubation followed by washing ) and co-cultured overnight with specific T cell clones ( 20 , 000 LCL and 1–2 , 000 T cells/well ) . T cell recognition was measured by IFNγ ELISA . Minimal epitopes were predicted using the following algorithms: SYFPEITHI ( http://www . syfpeithi . de ) , The Immune Epitope Database and Analysis Resource ( IEDB; http://www . iedb . org ) and Bimas ( http://www-bimas . cit . nih . gov/molbio/hla_bind ) ; T cell recognition of minimal epitope peptides was confirmed by IFNγ ELISA . PBMCs were tested in ex vivo ELISPOT assays of IFNγ release against individual defined epitope peptides ( final concentration 5μg/ml ) as previously described [51] . An equivalent volume of DMSO and 10μg/ml PHA were added to separate wells as negative and positive controls respectively . To analyse events occurring shortly after EBV infection in vitro , primary B cells were isolated from donor PBMCs by positive selection using CD19 beads according to the manufacturer’s instructions ( Dynal ) and exposed to EBV ( either supernatant from the B95 . 8 cell line , centrifuged and filtered to remove cellular debris , or purified recombinant wt2089 virus , quantified as previously described [46] and used at a MOI of 100:1 ) for 2hrs at 37°C , washed once and used as required . Following EBV infection , B cells were co-cultured with specific T cell clones ( 20–40 , 000 B cells and 1–5 , 000 T cells/well ) in 200μl RPMI containing 10% FCS and 40IU/ml IL-2 , in 96U well plates . Uninfected B cells or B cells pre-pulsed for 1 hour with 1μg/ml of the relevant epitope peptide served as negative and positive controls respectively . Supernatant ( ~80μl ) was harvested at time points between 24hrs and 14 days and stored frozen . An equivalent volume of fresh medium was replaced in the co-cultures . At the end of the time-course the supernatant from all time points was assayed concomitantly by IFNγ ELISA . RNA was prepared using a Nucleospin II kit ( Macherey-Nagel ) and subjected to an additional DNAse treatment using a DNAfree kit ( Life Technologies ) ; cDNA was then synthesised using QScript ( VWR ) . All protocols were carried out according to the manufacturer’s instructions . Samples were then analysed using a high throughput 48:48 Dynamic Array IFC-Gene Expression system ( Fluidigm ) . Absolute quantitation was achieved by generating a standard curve for each target gene using a dilution series of the AQ-plasmid standard which contained a single copy of each of 45 EBV and 3 cellular amplicons [34] . Immunoblotting was carried out as described previously [36] using mAbs to EBNA2 ( PE2 ) , LMP2 ( 14B7 ) and BHRF1 ( 5B11 ) and a polyclonal antibody specific for EBNA3B ( Exalpha Biologicals ) . Following EBV infection , B cells ( 20 , 000 cells/well ) were seeded into 96U well plates in RPMI containing 10% FCS and 40IU/ml IL-2 and co-cultured with specific T cell clones at effector to target ratios between 2:1 and 0 . 031:1 ( 40 , 000 to 625 T cells/well ) in triplicate wells . B cells cultured in the absence of T cells were included as a positive control for virus infection and transformation . Cultures were refed weekly by a half change of standard medium without cytokine supplement . Outgrowth was scored visually at ~4 weeks and the B cell identity of growing cultures confirmed by monoclonal antibody staining for CD19 .
|
Epstein-Barr virus infects the vast majority of the world’s population; in most individuals both primary infection and long-term virus carriage are asymptomatic . However , EBV is the major cause of glandular fever , is associated with multiple cancers and is implicated in various autoimmune conditions; thus there is a strong impetus for the development of a prophylactic vaccine . To date , vaccine design has largely focused on the induction of neutralising antibodies to virion structural components which can prevent virus binding and infection . Such strategies may be improved by the inclusion of immunogens to induce T cell responses with the potential to promptly recognise and eliminate cells that do become infected . Here we characterise T cell responses to three proteins , EBNA2 , EBNA-LP and BHRF1 that comprise the “first wave” of de novo viral antigen expression following EBV infection of B cells . Each of these proteins is targeted by strong T cell responses in a subset of donors . Furthermore , CD8+ T cells specific for that at least one of these proteins , EBNA2 , efficiently recognise B cells at very early time-points post-infection , before CD8+ T cells of all other specificities tested , and effectively inhibit outgrowth of B cell lines following EBV infection in vitro . Thus “first wave antigens” , particularly EBNA2 , may comprise suitable candidate immunogens for inclusion in prophylactic EBV vaccine design .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"blood",
"cells",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"enzyme-linked",
"immunoassays",
"immunology",
"microbiology",
"cloning",
"cytotoxic",
"t",
"cells",
"molecular",
"biology",
"techniques",
"immunologic",
"techniques",
"research",
"and",
"analysis",
"methods",
"white",
"blood",
"cells",
"animal",
"cells",
"t",
"cells",
"immunoassays",
"viral",
"replication",
"molecular",
"biology",
"lytic",
"cycle",
"immune",
"response",
"antibody-producing",
"cells",
"cell",
"biology",
"b",
"cells",
"antigen",
"processing",
"and",
"recognition",
"virology",
"biology",
"and",
"life",
"sciences",
"cellular",
"types"
] |
2016
|
Early T Cell Recognition of B Cells following Epstein-Barr Virus Infection: Identifying Potential Targets for Prophylactic Vaccination
|
Diet may be modified seasonally or by biogeographic , demographic or cultural shifts . It can differentially influence mitochondrial bioenergetics , retrograde signalling to the nuclear genome , and anterograde signalling to mitochondria . All these interactions have the potential to alter the frequencies of mtDNA haplotypes ( mitotypes ) in nature and may impact human health . In a model laboratory system , we fed four diets varying in Protein: Carbohydrate ( P:C ) ratio ( 1:2 , 1:4 , 1:8 and 1:16 P:C ) to four homoplasmic Drosophila melanogaster mitotypes ( nuclear genome standardised ) and assayed their frequency in population cages . When fed a high protein 1:2 P:C diet , the frequency of flies harbouring Alstonville mtDNA increased . In contrast , when fed the high carbohydrate 1:16 P:C food the incidence of flies harbouring Dahomey mtDNA increased . This result , driven by differences in larval development , was generalisable to the replacement of the laboratory diet with fruits having high and low P:C ratios , perturbation of the nuclear genome and changes to the microbiome . Structural modelling and cellular assays suggested a V161L mutation in the ND4 subunit of complex I of Dahomey mtDNA was mildly deleterious , reduced mitochondrial functions , increased oxidative stress and resulted in an increase in larval development time on the 1:2 P:C diet . The 1:16 P:C diet triggered a cascade of changes in both mitotypes . In Dahomey larvae , increased feeding fuelled increased β-oxidation and the partial bypass of the complex I mutation . Conversely , Alstonville larvae upregulated genes involved with oxidative phosphorylation , increased glycogen metabolism and they were more physically active . We hypothesise that the increased physical activity diverted energy from growth and cell division and thereby slowed development . These data further question the use of mtDNA as an assumed neutral marker in evolutionary and population genetic studies . Moreover , if humans respond similarly , we posit that individuals with specific mtDNA variations may differentially metabolise carbohydrates , which has implications for a variety of diseases including cardiovascular disease , obesity , and perhaps Parkinson’s Disease .
Diet and an organism’s genes contribute towards its phenotype and impact a range of scientific disciplines that span from the more fundamental disciplines of evolutionary biology and quantitative genetics to the more medically applied fields of nutrigenomics and pharmacogenomics . In nature , the dietary macronutrient balance is a strong selective force within and among populations . The relative proportions of macronutrients in food can fluctuate seasonally , vary when species colonise new habitats and can influence the frequency of alleles in populations [1 , 2] . It is well documented that nutritional responses vary with genotype [3–8] and it has been convincingly argued that the human genome is maladapted to our 21st century diet [9] . Dietary modification is an established treatment for certain diseases including cardiovascular disease , diabetes , and obesity [10] , yet , we still have an incomplete knowledge of how genetic variants that modulate susceptibility to disease are influenced by exogenous factors . This study explores the potential for diet to differentially influence mitochondrial function and the organismal health and fitness of Drosophila melanogaster flies harbouring distinct mtDNA types ( mitotypes ) . Protein and carbohydrate are the two primary energy-yielding macronutrients in fly food , and their ratio has been shown to have profound impacts on various aspects of physiology , behaviour , and biochemistry [11 , 12] . In adult females of D . melanogaster Canton S , a 1:2 Protein: Carbohydrate ( P:C ) ratio of food yielded the highest egg-laying rate and a 1:16 P:C ratio maximised survival [13] . Aw et al . [14] and Towarnicki and Ballard [15] demonstrated a more complex scenario whereby diet interacted with Drosophila mitotype and with other environmental factors such as temperature . For adults , Aw et al . [14] reported sex-specific influences of mitotype and diet on mitochondrial functions and physiological traits in males harbouring the Alstonville and Japan mitotypes . In larvae , Towarnicki and Ballard [15] manipulated food and temperature to study the development of the Alstonville and Dahomey mitotypes . We observed that larvae harbouring the latter mitotype developed more slowly than the former when fed a high protein diet at all temperatures , but more quickly when fed the high carbohydrate diet at higher temperatures . These studies did not determine the magnitude of selection at an organismal level or differentiate the relative importance of the interactions in larvae and adult stages , nor did they provide a mechanism of action . Toward these goals , we constructed laboratory diets that differed in their P:C ratios ( 1:2 , 1:4 , 1:8 and 1:16 P:C ) and also tested natural fruits that differed in their P:C ratio . Laboratory population cage studies are a sensitive method to test for selection in Drosophila and the frequency of each genotype type in cages is taken as an indicator of fitness [16 , 17] . Previous cage studies have provided evidence that distinct mitotypes can influence the frequency of flies in the laboratory , but none of these studies manipulated the diet [17–19] . The cage study paradigm used here does not involve flies breeding until termination of fecundity or lifespan; instead , it enforces a short window for flies to lay eggs . As a consequence , repeatable changes in the frequency of genotypes are caused by differences in immature development time and the fitness of young adults during the period that larval fat body remains [20] . Other experimental methods that have been utilised to estimate fitness of flies harbouring different mitotypes include in vivo competition and assaying mitotype frequencies of wild-caught animals [e . g . , 21 , 22 , 23] . Ma and O’Farrell [21] created fly lines with multiple mitotypes and utilised the uniparental mode of inheritance in mitochondria to test for selection . They observed that non-coding differences in the origin of replication region could cause the frequency of individuals harbouring a genome with a detrimental mutation to increase , but then lead to population death after several generations . The mechanism for this is still unknown . Thermal selection has been proposed to shape the pattern of mtDNA variation in eastern Australian D . melanogaster , but no experimental information has been provided on which mutation ( s ) may be driving these data [22 , 24] . Here , we chose the population cage paradigm for its high sensitivity and have quantified the frequencies of four globally sourced D . melanogaster mitotypes ( Alstonville , Dahomey , Japan and w1118 ) fed our four P:C diets . We then directly compete two mitotypes fed two diets . Mechanistically , provisioning of dietary macronutrients to mitochondria may be influenced by genetic variations that influence the activity of the electron transport system , organelle retrograde signalling to the nuclear genome , anterograde signalling to the mitochondrion and epigenetic modifications [12 , 25] . These variations may result from mtDNA mutations , mito-nuclear interactions and nuclear-encoded differences [7 , 22 , 26–28] . Mitochondria produce energy by utilising electrons harvested from oxidisable dietary substrates and O2 to build up a proton-motive force by pumping protons from the mitochondrial matrix into the intermembrane space . The subsequent backflow of protons to the matrix across complex V ( ATP synthase ) of the inner membrane drives the synthesis of ATP . Functional differences in mitochondrial energy production influence evolutionarily important physiological and organismal traits . In Drosophila , these traits include development time and egg production , and in humans , they include inherited disease and the decline in mitochondrial function with advancing age [25 , 29–31] . Here , we identified functionally significant differences between mitotypes by carefully controlling the nuclear genetic background , modelling quaternary and secondary structures , conducting multiple independent in vitro assays , adding electron transport system inhibitors to the diets , and assaying independently collected mitotypes [32–35] . Is it possible that a given mtDNA mutation could be slightly deleterious in one environment but advantageous in another ? If a mtDNA mutation is functionally deleterious , and linked mutations are neutral or nearly neutral , current models predict that the mitotype will have a selective disadvantage , causing it to decline in frequency in nature and population cage studies . Slightly deleterious mutations have been reported in Drosophila [26 , 36–38] , purifying selection has been demonstrated in the mouse female germline [39 , 40] , and deleterious mtDNA mutations are well-known in humans [41–43] . However , as dietary stress increases , genotype-specific mitochondrial responses may trigger flexible and broad cytosolic and nuclear reactions that have collectively been termed mitohormesis [44] . Remarkably , rather than being harmful , these changes caused by low levels of stress can result in a reconfiguration of metabolism , which in turn can enable increased production of ATP , increased evolutionary potential , and decreased susceptibility to disease [12 , 45] . Again , the mechanisms for this are not well understood . Various mechanisms by which stressed mitochondria may signal outward to the cytosol and the nucleus have been proposed . These include regulation of ATP levels , altering mitochondrial membrane potential to allow recruitment and assembly of signalling molecules , and the production of reactive oxygen species ( ROS ) . These are , however , not the only available pathways in the mitochondrial repertoire [44] . For instance , calcium signalling from the endoplasmic reticulum likely influences a multitude of mitochondrial functions [46 , 47] . Above a genotype-specific threshold , increasing the level of a specific stressor is expected to be deleterious and disease-causing , with the distribution of the response determined by the “norm of reaction” [48] . The norm of reaction describes the pattern of phenotypic expression across a range of environments and may be entirely different for two mitotypes . To investigate the possibility of functional compensation through mitohormesis , here , we conducted transcriptomics and metabolomics studies . We then experientially examined the mechanisms involved by manipulating dietary sugars and inhibiting specific metabolic pathways . Our series of studies show that a diet by mitotype interaction mediated the remodelling of carbohydrate metabolism in two Drosophila mitotypes . When fed the high protein 1:2 P:C diet , the slightly deleterious ND4 mutation in complex I of Dahomey mtDNA caused the mitotype to have a selective disadvantage compared to those harbouring Alstonville mtDNA . Complex I is the primary entry point for electrons into the mitochondrial electron transport system and is a site of electron leak to oxygen and the generation of ROS [49] . In contrast , when fed the high carbohydrate 1:16 P:C diet , mitotypes differentially remodelled energy metabolism , and this resulted in an evolutionary advantage to Dahomey . Were the same mechanisms found to occur in humans , the enhanced lipogenesis in individuals with slightly deleterious complex I mutations could make them more susceptible to obesity when eating a high carbohydrate diet , however , for those individuals with a predisposition to Parkinson’s disease , which has been linked to defects in lipogenesis [50] , this diet could delay onset or rate of decline .
To test the hypothesis that the fitness of mitotypes can be differentially influenced by diet [12] , we fed four diets varying in P:C ratio to four Drosophila mitotypes and assayed their frequency in population cages over 12 generations . The nuclear genome was standardised to w1118 and the microbiome controlled each generation by adding a ground homogenate of laboratory males . Given random mating of Drosophila harbouring distinct mitotypes [18] , population cage studies are a sensitive method to detect positive selection [17 , 18] . We investigated whether differences in immature development or adult fitness best described the observed mitotype frequencies on the four diets . Demonstrating that natural selection acts on mitochondrial genes is now firmly established [e . g . , 14 , 28 , 52–56] , but the specific life history stages and exogenous conditions through which mtDNA variations benefit the organism have rarely been experimentally identified . When all else is equal , reduced immature development time is beneficial in nature as it reduces exposure to predators and limited food supply [57] . It is also advantageous in population cages if a higher proportion of females of a specific mitotype develop into adults and more eggs are laid . The fitness of young females is experimentally determined by assaying fecundity and fertility . Female fecundity is sensitive to dietary changes and is experimentally measured as the number of eggs laid [13 , 14] . Fertility is a central determinant of an animals inclusive fitness and is quantified as the number of offspring , per female [58] . We conducted three additional cage studies to corroborate the hypothesis that the fitness of the Alstonville and Dahomey mitotypes was differentially influenced by diet [12] . In the first set of cage studies , we permute the diet to determine whether the mitotype specific responses are generalisable . Here , we include the 1:2 P:C diet for generations 1–4 , swapped to the 1:16 P:C diet for generations 4–20 , and then returned to the 1:2 P:C diet for generations 20–26 . In a second set of cages , we include fruits with ~1:2 and ~1:16 PC ratios . Fruits have previously been used to validate laboratory diets as they effectively control for artificial differences in amino acids , lipids , and micronutrients [63] . We include passionfruit ( ~1:2 P:C ) and banana ( ~1:16 P:C ) . In the third set of cages , we compete the two D . melanogaster mitotypes independently against Drosophila simulans ( Wolbachia uninfected with the siIII mitotype collected from Kenya [64] ) . These species are sympatric through large parts of their range and compete for similar resources . We assay immature development time and test for reproducibility , permute the nuclear genome , replace the laboratory diets with natural fruits and include the microbiome from orchard fed flies . To test for reproducibility , development time was assayed at ~6-month intervals . Mito-nuclear interactions have been shown to influence a range of molecular and organismal traits in insects , crustaceans , fish , and mice [7 , 65–68] . Here , we substituted the w1118 nuclear genome with Oregon R and with Canton S using balancer chromosomes and then conducted five generations of backcrossing before our experiments [7] . The w1118 nuclear genome diverged from the wild caught Oregon R line in 1984 , and they have been separated for more than 800 generations . The Canton S line was collected before 1916 in Canton , Ohio [69] . To corroborate the cage studies that included fruit , we test the development times of the mitotypes fed passionfruit and banana . Host-associated microbiota can impact metabolism and gene expression at cellular and organism-level scales [70 , 71] . Adair et al . [72] quantified the bacterial communities associated with natural populations of D . melanogaster and found microbes were predominantly of two to three taxa . Here , we focus on levels of Acetobacter and Lactobacillus as they dominated the microbial communities in our populations . To predict whether a nonsynonymous change , an RNA mutation , or variation in A+T repeat number was likely to be functionally significant we generated quaternary and secondary structure models and then assayed repeat number variation [79–82] . There are three nonsynonymous difference between Alstonville and Dahomey mtDNA [66] and we modelled each complex harbouring a change—complex I ( V161L , ND4 subunit ) , complex IV ( D40N , COIII subunit ) and complex V ( M185I , ATP6 subunit ) [79–81] . There are also three rRNA differences ( two srRNA and one lrRNA ) [83] and 52 A+T-rich region variations [84] ( S1 Table ) . Towarnicki and Ballard [15] mapped the two srRNA mutations on the human mitoribosome and proposed that they are unlikely to influence mitochondrial function [83 , 85] so they are not considered here . Selection has rarely been shown to act on the mitochondrial A+T rich or control region [but see , 21 , 27] , and no differences were identified in secondary structures or the central T-stretch between the Alstonville and Dahomey mitotypes [15] . However , differences in the number of repeats have been recently shown to influence mitochondrial functions [86] . We tested hypotheses generated from the modelling by extracting mitochondria and assaying organelle function , independent of cellular interaction . These cellular assays included electron transport system complex activity assays , in vitro mitochondrial oxygen consumption , Western blots and native protein gels . Activity was included because it has previously been employed to corroborate the influence of a mtDNA mutation [87 , 88] . We predicted that functionally significant mutations would reduce the activity of the complex . In vitro mitochondrial oxygen consumption is increasingly recognised as a fundamental measure of mitochondrial function [89 , 90] and we assayed the rate from extracted mitochondria using complex I and II substrates [91] . Complex I substrates assay the combined mitochondrial functions of complexes I , III and IV , while complex II substrates assay the collective functions of complexes II , III and IV . Western blots were used to measure expression of complex I and complex V and native protein gels to determine native protein masses of complex I and its protein–protein complexes . Chemical impairment of complex I reproduces the observed flip in development rates . Here , complex I inhibitors were added to the diet to create phenocopies in Alstonville of the Dahomey ND4 mutation . Goldschmidt [100] coined the term “phenocopy” to describe morphological alterations in Drosophila that could be induced by the imposition of stress during development . Thus , a phenocopy is produced environmentally and shows features characteristic of a genotype other than its own . Chemically induced phenocopies in Drosophila are well studied with the production of eyeless mutants by feeding food containing borate [101] and production of bithorax mutants by treating embryos with diethyl ether [102] . We added rotenone to the diet to phenocopy the ND4 mutation in Dahomey because it inhibits electron transfer from the iron-sulphur centres in complex I , leading to a partial blockade of oxidative phosphorylation with reduced synthesis of ATP [103] . We then quantify the rate of development . If the slightly deleterious V161L ND4 mutation in Dahomey was driving the differences in development time ( Fig 1B ) , we predicted that Alstonville larvae fed food containing rotenone ( the phenocopy ) would develop more slowly than untreated larvae on the 1:2 P:C diet , but faster on the 1:16 P:C food . In contrast , we predicted that Dahomey larvae fed rotenone would develop more slowly when fed both diets as the complex I dysfunction would be the combined effects of the mutation and the inhibitor . We then tested the generality of the rotenone result with paraquat . Paraquat is a common herbicide that has been proposed to cause mitochondrial dysfunction by complex I toxicity following lipid peroxidation of the mitochondrial inner membrane [104] . Next , we tested whether dietary addition of rotenone influenced complex I activity , superoxide dismutase ( SOD ) activity , and larval dry weight . Complex I activity in larvae fed the standard diets was measured in Study 3 ( Fig 4A ) . It was included here to test whether the ND4 mutation and the dietary addition of rotenone had similar effects on the complex . SOD constitutes the first line of defence in the antioxidant enzyme network [105 , 106] , is the primary scavenger of the ROS superoxide [107] , and total activity was assayed . Larval weight was assayed as an organismal trait that can influence development time [reviewed in 108] and patterns of adult reproductive investment [109] . To test whether the flip in immature development time was generalisable to a second pair of mitotypes , one of which harboured the V161L ND4 mutation , we compared the immature development times of flies harbouring Madang ( Papua New Guinea ) and Victoria Falls ( Zimbabwe , Africa ) mtDNA [84] . Madang mtDNA has the same ND4 , and lrRNA mutations and differs from Dahomey by 27 A+T rich region mutations ( S1 Table ) . Victoria Falls does not harbour either the ND4 or the lrRNA mutations . It has three nonsynonymous ( ND2 , ATP6 and COIII ) , two sRNA and 49 A+T rich region differences from Alstonville ( S1 Table ) . For experimentation , both mitotypes were harboured in the w1118 nuclear genetic background and the microbiome was controlled . To gain mechanistic insight into the processes underpinning the flip in the development times of the mitotypes we include transcriptomics and metabolomics studies . Next-generation RNA sequencing has permitted the mapping of transcribed regions of the genomes of a variety of organisms [120–122] . Studies of Drosophila reveal a transcriptome of high complexity that is expressed in discrete , tissue- and condition-specific mRNA and ncRNA transcript isoforms [120] . This enables a dynamic ensemble of transcript isoforms that gives rise to substantial diversity . Recently , Crofton and colleagues [123] asked whether D . melanogaster mothers who experience poor nutrition during their own development change their gene product contribution to the egg . They find an increase in transcripts for transport and localization of macromolecules and for the electron transport chain . In this study flies were raised for at least two generations on instant Drosophila food . Eggs were then transferred to each diet and a standard microbiome added after 2 d . A limitation of the technique is that not all transcripts currently have a known function . Metabolomic profiling provides an additional layer of knowledge for the most complete representation of the phenotype of the animal , revealing the combined contributions of gene expression , enzyme activity , and environmental context [124] . Here , we include gas chromatography-mass spectroscopy ( GC/MS ) , which is capable of measuring small molecules with a mass <500 Da . One constraint of the GC/MS method for metabolomics studies is that distinct molecules may have similar retention times and it is necessary to validate results with standards [125] . In this section , we explore the disadvantage to Dahomey on the 1:2 P:C diet and begin to investigate the mitohormetic responses of the mitotypes fed the 1:16 P:C diet . We assay basal ROS production and expression of two Glutathione S-transferase ( GST ) genes because SOD activity was higher in Dahomey than Alstonville larvae and higher in mitotypes fed the 1:2 P:C diet ( Fig 5C ) . Basal mitochondrial ROS gives the levels produced at the resting state and are an indicator of mitochondrial coupling efficiency in respiration [143] . ROS production and detoxification are tightly balanced , and numerous stress response mechanisms have evolved [144] . GSTs are a large supergene family of an ancient detoxifying enzyme and respond to endogenous and exogenous substrates through glutathione conjugation [145] . Transcriptomic data showed that Dahomey larvae fed the 1:2 P:C diet exhibited an elevation in cytochrome P450 metabolism ( Fig 7A ) , and had higher expression levels of GstE1 and GstE5 ( S2A Table ) . Here , we perform quantitative reverse transcription PCR ( RT-qPCR ) to confirm that the genes that were identified in the original RNA-seq were also altered as expected . Next , mtDNA copy number and levels of ATP were assayed . Copy number is regulated by ROS in yeast [146] , is positively linked to levels of ATP [147] , and is crucial for maintaining cellular energy supplies [147 , 148] . In Drosophila , mtDNA copy number is proposed to impact the organismal phenotype by influencing the respiratory membrane and the efficiency of oxidative phosphorylation [86] . ATP production has been shown to influence many cellular processes and evolutionary important physiological parameters including development rates [29 , 149] . We posited that the polyol pathway was mechanistically involved in the mitohometic response in Dahomey larvae due to the elevation of sorbitol levels in the metabolomics data and predicted that including dietary sugars in the pathway would be beneficial . If true , we hypothesised that the addition of the polyol pathway inhibitor Epalrestat would mitigate the net benefit . We then assayed the number of flies eclosing in 3 d , quantified the expression of Notch ( N ) and Cyclic-AMP response element binding protein B ( CrebB ) , and determined food consumption . In the non-disease context the polyol pathway is essential for cellular osmoregulation but , in the context of diabetes , it is associated with tissue-damage during hyperglycaemia [151] . In the pathway , glucose is reduced to sorbitol , via the action of the enzyme aldose reductase , and then oxidized to fructose . O-fucose and O-glucose are essential for normal Notch signalling [152] and their levels are regulated by derivatives of the polyol pathway including fructose , sorbitol and mannose , while xylose negatively regulates signalling [153 , 154] . Notch regulates the cAMP responsive element binding protein ( CREB ) [155 , 156] , and experientially blocking CREB activity in Drosophila fat body has been shown to increase food intake [157] . N and CrebB were differentially expressed in the transcriptomics data ( S2B Table ) . The polyol pathway does not produce ATP so could not adequately account for the similarities in ATP levels between the mitotypes fed the 1:16 P:C diet . Here , we test the hypothesis that rates of β-oxidation differed between the mitotypes and add Etomoxir to the diet . β-oxidation of fatty acids generates NADH and FADH2 and thereby partially bypasses complex I of the electron transport system [91] . Etomoxir inhibits entry of long-chain fatty acids into the mitochondrion via the carnitine shuttle and we predicted its addition would result in loss of the selective advantage to Dahomey . We then quantified development , triglyceride levels , expression of elongase F ( eloF ) and brummer ( bmm ) , β-oxidation activity , acetyl-coA enzyme activity , NAD+/NADH ratio and starvation survival . Metabolomic data showed high levels of stearic and palmitic acid in Dahomey larvae so we assayed triglycerides . To test for increased lipogenesis , we assayed the expression of eloF and bmm . elofF is a female-biased elongase involved in long-chain hydrocarbon biosynthesis [158] . bmm is a lipase which promotes fat mobilisation and is responsible for channelling fatty acids toward β-oxidation [159] . Both , elofF and bmm were differentially expressed in the transcriptomics data ( S2B Table ) . β-oxidation was directly quantified using 14C-labelled palmitic acid . Acetyl-CoA was measured because the breakdown of carbohydrate influences its levels . NAD+ is required for fatty acid metabolism and the NAD+/NADH ratio was assayed . Starvation resistance was tested as a significant organismal trait [160] . When a larva is not feeding , energy can only come from the metabolism of existing resources [161] , which occurs when fruits are small , when food quality declines and also in a fluctuating environment [162] . Transcriptomic data discussed in Study 5 actively support the result that a general increase in mitochondrial gene expression is part of rewiring in Alstonville on the 1:16 P:C diet . Furthermore , we hypothesised that glucose-6-phosphate was differentially metabolized in Alstonville due to the observed elevation in gluconate . Glucose 6-phosphate can be converted to store glycogen through the action of glycogen synthase and so we assayed levels of glycogen . Glycogen synthase and insulin-like receptor ( Inr ) are elevated in Alstonville ( S2 Table ) . Glycogen is a primary source of energy for adult muscle function [170 , 171] and the ubiquitous activation of Inr has previously been shown to cause larvae to feed less and to wander off the food [172] . Therefore , we assay development time and movement . Glucose 6-phosphate is also metabolized by the pentose phosphate pathway and D-Gluconate can be phosphorylated to 6-phospho-D-gluconate to enter the oxidative phase of the pathway [173] . Here we quantified the expression of Zwischenferment ( Zw ) and assayed glucose-6-phosphate dehydrogenase ( G6PD ) activity . Zw was differentially expressed in the transcriptomics data ( S2B Table ) . Zw catalyses the oxidation of glucose-6-phosphate ( G6P ) to 6-phosphogluconate . G6PD is the rate-limiting enzyme of the pentose phosphate pathway [174 , 175] . We then assayed one aspect of insulin signalling . The insulin/insulin-like growth factor signalling pathway controls a wide variety of biological processes in metazoans [176] and stimulates glucose metabolism via the pentose phosphate pathway in Drosophila cells [177] . The most upstream central players in this pathway are members of the insulin-like peptide ( ILP ) family , which includes insulin and insulin-like growth factors in mammals [178] , as well as multiple ILPs in worms and insects [179] . ILPs are regulated by nutritional status and Insulin-like peptide 2 ( Ilp2 ) is essential for maintaining normoglycemia [180] . We assayed Ilp2 to corroborate the results from the transcriptomics data ( S2B Table ) . Here , we replaced sucrose ( control ) with gluconate as the dietary sugar , but did not include any blockers because we considered this the wild-type pathway on the 1:16 P:C diet . Over the past decade , it has become clear that diet is an evolutionary force that has immediate implications for our understanding of health and disease . Here , we provide substantial evidence to suggest that a single mtDNA encoded nonsynonymous mutation can differently influence the regulation of dietary metabolites and have significant phenotypic consequences . When fed the 1:2 P:C diet , Alstonville larvae had a relative advantage as the V161L ND4 mutation in Dahomey caused an increase in ROS production , which resulted in oxidative stress and a decrease in mitochondrial functions leading to reduced mtDNA copy number and ATP levels . When fed the 1:16 P:C diet , Dahomey larvae had the relative advantage with multiple linked pathways working in a synergistic mitohormetic response that enabled larvae to eat more and develop more quickly . The remodelled pathways in Dahomey included upregulation of the polyol pathway , which fed back to increase food consumption and fuelled increased β-oxidation of fatty acids . Each cycle of β-oxidation results in the donation of electrons to the quinone pool downstream of complex I in the electron transport system , thereby bypassing the V161L , ND4 subunit mutation [189] . This process maintains levels of the quinone pool , which has been shown to be functionally important [190] . In Alstonville , mitochondrial gene expression was higher , glycogen metabolism increased and larvae were more active . We postulate that the greater physical movement in Alstonville larvae on the 1:16 P:C diet caused a reallocation of ATP away from cell division and growth , thereby slowing development . ATP drives many cellular processes and constrains development rates [29 , 149] . An alternative explanation is that upregulation of Notch and/or FOXO signalling in Dahomey may be responsible for driving mitotype-specific differences in development [127 , 128] . These data further question whether mtDNA can be assumed to accurately reflect species or population-level demographic processes when the dietary protein to carbohydrate ratio varies over time or space . It is now well documented that purifying selection affects the variability of mtDNA encoded genes , and the purging of deleterious variants will result in the removal of linked variants through background selection . In humans , deleterious mtDNA mutations are well-known [41–43] , and evidence for a profound effect of accumulated mutations on men’s health has been reported [191] . Purifying selection has been demonstrated in the female mouse germline [39 , 40] and in Drosophila slightly deleterious mutations have been reported [26 , 36–38] . Evidence of positive selection on mitogenomes has been reported [27 , 52] , but to our knowledge , no specific mutation has been experimentally shown to result in an evolutionary advantage . Our observation that distinct mitotypes reached high frequency when fed different macronutrient ratios in population cages suggests that diet may also be a strong selective force in nature . Here , we advocate future studies test for selection on mtDNA within and among naturally occurring populations where macronutrients change over time and space . The influence of diet is extensively studied in the literature but few studies investigate genotype-by-diet interactions and fewer still that have unravelled the underlying mechanisms [3–7 , 51 , 143 , 192] . One prediction from these data is that experimentally increasing the P:C ratio ( i . e . , 1:20 P:C ) may further increase the dietary-induced metabolic stress and cause increased mortality in larvae harbouring Dahomey mtDNA . Conversely , development time in Alstonville larvae may decrease if the polyol pathway is upregulated . Experimentally , such a dietary perturbation would be outside the range of P:C ratios encountered by Drosophila in nature , but perhaps would reflect the human genomes clash with modern life and the vending machine . Most common mtDNA mutations are thought to be deleterious and involved in a variety mitochondriopathies and complex diseases like diabetes , cardiovascular disease , gastrointestinal disorders , skin disorders and elevated blood pressure [e . g . 193 , 194–197] . Further , the accumulation of somatic mtDNA mutations likely influences primary cancers and the ageing process [198 , 199] . The data presented here suggest that it is also possible that slightly deleterious mtDNA mutations may confer an advantage in certain situations . Our data , therefore , support matching an individual’s diet to their mitotype as an approach to treating mitochondriopathies , complex diseases or even for optimising health in non-disease populations . For example , were the same mechanisms found to occur in humans the enhanced lipogenesis in individuals with mild complex I mutations could make them more venerable to obesity when eating a high-carbohydrate diet , yet less susceptible to Parkinson’s disease , which has been linked to defects in lipogenesis [200] .
Here , RNA and metabolites were extracted from female third instar wandering larvae sourced from the side of the bottle that had developed on 1:2 and 1:16 P:C diets [108 , 208] . In uncrowded conditions , on a fixed light/dark regimen , larval wandering is highly synchronous and begins some 24 h before pupation ( at 25° C ) [108] . To test specific hypotheses , we replaced sucrose as the dietary sugar . The 1:16 P:C diet was prepared without the addition of sucrose . Then , 200 ml of food was combined with 1 . 87 g of either sucrose ( Sigma S0389 ) as the control , sorbitol ( Sigma S1876 ) , fructose ( Sigma F0127 ) , mannose ( Sigma M6020 ) , fucose ( Sigma F2252 ) or xylose ( Sigma X3877 ) . Each new diet was poured into 8 bottles . Equal amounts of eggs harbouring Alstonville or Dahomey mtDNA were added to each food and microbiome was added after 2 d . Flies were kept at 23° C , 50% humidity on a 12 h light/dark cycle . Emerging adult female flies were counted over 3 d , and percentage eclosion of each mitotype was calculated ( dx . doi . org/10 . 17504/protocols . io . rtfd6jn ) . For inhibitors , freshly prepared aldose reductase ( polyol pathway ) inhibitor ( Epalrestat , Sigma SML0527 ) and carnitine palmitoyltransferase-1 inhibitor ( Etomoxir , Sigma E1905 ) were solubilised in water to make a 5 mM stock . The stock solutions were added to the 1:16 P:C diet to final concentrations of 25 μM Epalrestat , 12 . 5 μM Etomoxir . Methodology followed that described above for 2-Deoxy-D-Glucose . Unless otherwise stated , all data are biological replicates and statistically analysed by ANOVA followed by Student’s t-tests to determine difference ( JMP software 12 , SAS Institute , NC , USA ) . Biological replicates are parallel measurements of biologically distinct samples . Where the numbers of Dahomey larvae eclosing in 3 d was compared between dietary sugars ( sorbitol , fructose , mannose , fucose , xylose and gluconate ) and the control diet with the diets supplemented with an inhibitor ( Epalrestat , and Etomoxir ) we conducted Dunnett’s tests . Data were checked for normality using a Shapiro-Wilks W test and outliers removed before statistical analyses using box plots . Values that were greater than ± 1 . 5 interquartile range were categorised as an outlier and excluded from the data set . No statistical methods were used to predetermine sample size .
|
The detection and quantitation of mtDNA polymorphisms in populations and across whole habitats continues to be used as a central investigatory tool in evolutionary genetics . But , the approach is laden with assumptions about selection that are rarely examined . We present a series of studies that traverse the genotype to the phenotype . The studies were designed to experimentally test the interaction between diet and mitotype in Drosophila flies and provide a mechanism by which selection occurs . We start with population cage studies that include four laboratory diets and four mitotypes . We then directly compete two mitotypes ( Alstonville and Dahomey ) on a high protein and a high carbohydrate diet and show a flip in their relative fitness that is driven by differences in larval development . Next , we identify a single naturally-occurring point mutation , which drives the cage results . We track the ripple effects up to the level of the organelle ( mitochondria ) , through the labyrinth of metabolic pathways and on to the phenotype . Notably , when flies were fed the high carbohydrate diet , energy metabolism was extensively remodelled in both mitotypes causing increased physical activity in Alstonville flies . These data invite an extensive experimental re-evaluation of the assumption that mtDNA inescapably evolves in a manner consistent with a strictly neutral equilibrium model . It also motivates investigation of genotype-specific dietary manipulation as an integrative treatment of human disorders involving mitochondrial metabolism and offers the potential for future therapeutic strategies .
|
[
"Abstract",
"Introduction",
"Results",
"and",
"discussion",
"Materials",
"and",
"methods"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"mitochondrial",
"dna",
"computational",
"biology",
"diet",
"animals",
"invertebrate",
"genomics",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"nutrition",
"experimental",
"organism",
"systems",
"genome",
"analysis",
"forms",
"of",
"dna",
"mitochondria",
"dna",
"bioenergetics",
"cellular",
"structures",
"and",
"organelles",
"drosophila",
"research",
"and",
"analysis",
"methods",
"food",
"genomics",
"animal",
"studies",
"life",
"cycles",
"insects",
"animal",
"genomics",
"arthropoda",
"biochemistry",
"eukaryota",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"transcriptome",
"analysis",
"biology",
"and",
"life",
"sciences",
"energy-producing",
"organelles",
"larvae",
"organisms"
] |
2018
|
Genotype to phenotype: Diet-by-mitochondrial DNA haplotype interactions drive metabolic flexibility and organismal fitness
|
Alphaviruses , such as chikungunya virus , and flaviviruses , such as dengue virus , are ( re ) -emerging arboviruses that are endemic in tropical environments . In Africa , arbovirus infections are often undiagnosed and unreported , with febrile illnesses often assumed to be malaria . This cross-sectional study aimed to characterize the seroprevalence of alphaviruses and flaviviruses among children ( ages 5–14 , n = 250 ) and adults ( ages 15 ≥ 75 , n = 250 ) in western Kenya . Risk factors for seropositivity were explored using Lasso regression . Overall , 67% of participants showed alphavirus seropositivity ( CI95 63%–70% ) , and 1 . 6% of participants showed flavivirus seropositivity ( CI95 0 . 7%–3% ) . Children aged 10–14 were more likely to be seropositive to an alphavirus than adults ( p < 0 . 001 ) , suggesting a recent transmission period . Alphavirus and flavivirus seropositivity was detected in the youngest participants ( age 5–9 ) , providing evidence of inter-epidemic transmission . Demographic variables that were significantly different amongst those with previous infection versus those without infection included age , education level , and occupation . Behavioral and environmental variables significantly different amongst those in with previous infection to those without infection included taking animals for grazing , fishing , and recent village flooding . Experience of recent fever was also found to be a significant indicator of infection ( p = 0 . 027 ) . These results confirm alphavirus and flavivirus exposure in western Kenya , while illustrating significantly higher alphavirus transmission compared to previous studies .
Arthropod-borne viruses ( arboviruses ) , such as the alphavirus chikungunya ( CHIKV ) , and the flavivirus dengue ( DENV ) , represent a multi-dimensional , ongoing threat for current and future generations[1–6] . Sudden and pervasive outbreaks have become an increasingly regular occurrence over the last decade , illustrating the intensity at which arboviruses can spread and affect naïve populations[1 , 7] . Many alphaviruses and flaviviruses are primarily transmitted by the same vector , the Aedes aegypti mosquito[8 , 9] , which is found in most regions of Kenya , in both rural and urban sites [10–12] . Due to the shared primary vector species , DENV and CHIKV are now co-endemic in many regions of the world , including Asia , Africa , South and Central America , and the Caribbean[13–15] . Acute symptoms of many alphavirus and flavivirus infections are generally representative of nonspecific and mild febrile disease , with the addition of a possible rash , arthralgia , and arthritis[16–18] . For this reason , accurate differential diagnosis is necessary for determining appropriate symptom-specific treatment , and avoiding non-specific clinical diagnoses that often lead to inappropriate treatments , most commonly those used for malaria[19] . Many previous studies describing virus-specific prevalence in African countries report conflicting results[20–23] . This may be due to regional distribution of vectors [12 , 24] , seasonal fluctuations in climate and flooding , parallel sylvatic transmission cycles , demographics associated with previous exposure and acquired immunity , and other factors involved in aiding viral spread[25] . In this study , we aimed to increase the knowledge regarding seroprevalence and factors associated with increased exposure to alphaviruses and flaviviruses in a population of children and adults living in western Kenya .
The study area spans an approximately 3 , 200 Km2 semi-circle centered in the town of Busia[26] . This area is largely representative of the wider Lake Victoria Crescent ecosystem , which includes regions in Kenya , Uganda , and Tanzania . The study area is a rural area of approximately 1 . 4 million people [27] , with the majority of people involved in mixed farming of crops and livestock[26] . Serological samples and demographic data were collected by weighted and stratified random sampling of 416 homesteads between August 2010 and July 2012 as part of a cross-sectional study of zoonotic infections in western Kenya[26 , 28–30] . Sera were analyzed to determine alpha- and flavivirus seroprevalence among children and adults . Random sampling of homesteads within the original study was stratified within sub-locations , the smallest administrative unit in Kenya . The number of homesteads to select per sub-location ( between 1 and 8 ) was proportional to the expected cattle population , so that more homesteads were sampled in sub-locations with larger cattle populations . A household was defined as all people identified by the head of household as being an occupant at the time of recruitment , to the extent that food is regularly shared from the household pot within the past 4 weeks . GPS coordinates were obtained for every homestead in the study using a handheld Garmin GPS unit . A maximum of 25ml of venous blood was collected for the original study . Biobanked aliquots of serum stored at -80°C were used for this study . Individuals aged 5 years or older from consenting homesteads were included in our study population . A subset of 250 samples per age category ( adults ( ages ranging from 15 to ≥ 75 years ) and children ( ages ranging from 5 to 14 years ) ) was selected as a representative sampling from the original cohort of 2 , 106 subjects ( 879 adults and 1 , 227 children ) . The 250 samples from adults and children were selected from a randomized sorting of the original study samples in Excel . Relevant demographic and health information was collected from questionnaire data regarding health and vaccine history , behavioral lifestyle and practices within the homestead and were linked to seropositivity . The original data pertaining to this study are available in an online repository . Exclusion criteria included subjects with severe anemia and those in their third trimester of pregnancy . Sera were tested by indirect ELISA for the presence of anti-CHIKV and anti-DENV IgG antibodies , as described previously[2 , 31] . Nunc-immuno 96-well plates were coated with CHIKV antigen ( derived from the 181/25 vaccine vector strain , and supplied by Dr . Mark Heise , University of North Carolina School of Medicine , Chapel Hill , NC 27599 ) , in a carbonate coating buffer , or DENV1–4 antigen ( derived from four serotypes[32–34]: DENV1 , Western Pacific 74; DENV2 , S16803; DENV3 , CH53489; and DENV4 , TVP360 , and provided by Dr . Eva Harris , University of California Berkeley , Berkeley , CA 94720 ) in a PBS/0 . 01% NaN3 buffer ( 1:400 , 50μl/well ) . Plates were coated overnight at 4°C . Plates were washed with a PBS/0 . 01% Tween-20/0 . 01% NaN3 wash buffer , blocked with blocking buffer ( 5% powdered milk in PBS ) for two hours at 37°C , and washed prior to adding the samples . Serum samples ( 1:200 in blocking buffer ) were added in duplicate ( 50μl/well ) and incubated overnight at 4°C . Plates were washed , coated with goat anti-human IgG conjugated to alkaline phosphatase ( AP ) ( 1:1000 in blocking buffer , 50μl/well ) , and incubated for one hour at 37°C . Following incubation , plates were washed and AP substrate in PnPP buffer ( 1ug/1ml ) was added ( 100μl/well ) . After a 30-minute incubation at 37°C , optical density was read at 405nm . Cut-off values for seropositive and seronegative results were determined against plaque-reduction neutralization tests ( PRNT ) -confirmed serum samples for CHIKV and DENV , by calculating three times the negative control , and at least half of the positive control reading . All statistical analysis was conducted using R programming language open-source software[35] . Prevalence was calculated by determining the percentage of serologically positive samples within each age category . Bivariate analysis of each of the potential predictors of either alphavirus or flavivirus exposure was performed using the Chi-square or Fisher’s Exact test . Demography and health data derived from the participant questionnaires were analyzed using LASSO ( least absolute shrinkage and selection operator ) generalized linear mixed models regression ( the glmmLasso package ) [36] to determine factors associated with increased risk of alphavirus and flavivirus seropositivity . The glmmLasso package was used because it allows the inclusion of a random effect ( household ) in the LASSO variable selection process . With many potential predictors of exposure available in this study and limited prior knowledge of which are definitively predictive of DENV/CHIKV infection , the LASSO procedure shrinks unimportant variable estimates to 0 and allows selection of the most correlated variables based on this dataset . The LASSO procedure is typically used when sample size is relatively small when compared to the number of variables of interest . Akaike’s Information Criteria ( AIC ) and Bayesian Information Criteria ( BIC ) were used to run iterations of glmmLasso to identify statistically predictive variables for alphavirus and flavivirus infection . The optimal lambda tuning parameter was chosen by the minimum AIC value . A cross-validation ( CV ) method was also used to run iterations of glmmLasso to identify variables significantly related to the increased odds of alphavirus and flavivirus exposure . Models produced using CV are created using “train” datasets and the resulting optimal model is validated using “test” datasets . Variables included in the glmmLasso model were selected based on their relevance to arbovirus and vector exposure , including basic demographic data such as age group , sex , community , education level , and occupation; proxies for time spent outdoors such as frequency of hunting , fishing , and grazing behaviors; and animal contact such as whether livestock and poultry have access to the homestead buildings; whether wildlife , including rats , were observed; village flooding and drought history; water sources used during wet and dry seasons; and health-related data such as recent illness , smoking behavior , and vaccination history . All variables were subject to a random effect of homestead , defined by shared exact northing and easting GPS coordinates . Kernel density analysis was used to map the distribution of CHIKV and DENV seropositive events within the geographic limits of the study region[37] . Samples that tested seropositive for previous alphavirus or flavivirus exposure were mapped within the study area based on GPS coordinates collected at the time of sample collection . Maps were created using ArcGIS software by ESRI ( ESRI ArcGIS Desktop: release 10 . 3 . 1; Redlands , CA ) . For each case , case level data was projected using Arc 1960 Universal Trans Mercator Zone 37s ( 21037 ) and geographic coordinate system GCS_Arc_1960 . Kernel density analysis was performed using the geoprocessing tools within ArcGIS 10 . 3 . 1 ( ESRI ) using a bandwidth of 9 , 000 meters based on incremental spatial autocorrelation analysis to select a distance band reflecting maximum spatial autocorrelation . Various methods are suggested for selecting a bandwidth based on the biological bases of underlying disease mechanisms and quantitative criteria[38] . Human movement also impacts transmission patterns[39] , and estimates of the biological distance at which dengue and chikungunya transmission occurs vary widely[40–43] . Resulting kernel density raster files were contoured and the top 50% of contours selected and mapped . Spatial scan statistics were performed using SatScan[44] with Bernouli distribution [45] . All data points were separated by infection status for cases versus controls for each DENV and CHIKV . SatScan searched for high clusters using a maximum spatial cluster size of 50 percent of population at risk and a circular window shape . Secondary clusters were only reported without geographic overlap . Resulting clusters were mapped using ArcGIS software by ESRI ( ESRI ArcGIS Desktop: release 10 . 3 . 1; Redlands , CA ) . Maps utilized for kernel density and spatial scan statistics were produced using basemaps OpenStreetMap Data from mapbox ( https://www . mapbox . com/ ) , an open source mapping resource . Bodies of water in the study area were mapped , and a near table was generated by calculating the shortest path based on a spheroid ( geodesic ) to the nearest body of water using the ArcGIS geoprocessing toolbox . Median distance to nearest water body was compared across infected and non-infected subjects using the Wilcoxon rank-sum test . Serological samples and demographic data were collected by weighted and stratified random sampling of 416 homesteads between August 2010 and July 2012 as part of a cross-sectional study of zoonotic infections in western Kenya . Ethical approval for study was granted by the Kenya Medical Research Institute ( KEMRI ) Ethical Review Board ( SC1701 ) ; participants of the original study consented to long term storage and further use of their anonymized samples . Subsequent analysis of biological material as described here was approved by the Stanford University Institutional Review Board ( R01: IRB-31488 ) ; all participants and/or legal guardians provided written informed consent .
Of the 500 samples tested , 66 . 9% ( n = 335 , CI95 62 . 7%–70 . 9% ) of all participants were seropositive for previous alphavirus exposure , as indicated by the presence of anti-CHIKV IgG , and 1 . 6% ( n = 8 , CI95 0 . 8%–3 . 1% ) were seropositive for previous flavivirus exposure , indicated by the detection of anti-DENV IgG . Comparatively , adults ( age groups 15–24 through 75+ ) had a higher rate of alphavirus seropositivity ( 78 . 7% ) than children ( age group 5–14 ) ( 42% ) . Flavivirus seropositives were only identified in individuals under the age of 45 , with the highest percentage ( 5 . 2% ) of positives in individuals aged 15–24 years ( n = 76 ) . Only 1 of the 2 children seropositive for flavivirus IgG was also seropositive for alphavirus IgG , whereas all six of the adults that were seropositive for flavivirus IgG were also seropositive for alphavirus IgG . Of the flavivirus seropositives ( n = 8 ) , 62 . 5% ( n = 5 ) had not been vaccinated against yellow fever , indicating that false positives were not a concern for the vaccinated population . History of vaccination against yellow fever was not significantly correlated with alphavirus infection ( p = 0 . 06 ) . Although this may suggest weak evidence of an inverse relationship between vaccination against yellow fever and flavivirus infection , the yellow fever vaccination variable was not selected as a significant variable by further multivariable modeling . Biological sex was not a statistically significant factor for alphavirus or flavivirus seroprevalence in adults ( p = 0 . 16 ) . Seroprevalence for alphaviruses in children was nearly equal for female ( n = 70 ( 57 . 4% ) ) , and male ( male: n = 71 ( 55% ) ) participants . Two female children tested positive for anti-DENV IgG , resulting in 0 . 8% ( CI95 0 . 1%–2 . 9% ) seroprevalence for flaviviruses in females . No flavivirus positive cases were identified in male child participants . Using Chi-square test and fisher’s exact test , a range of variables was assessed for their influence on the risk of alphavirus or flavivirus infection ( Table 1 ) . Given the low number of flavivirus positives , statistical analysis described in Table 1 details infection , inclusive of either alphavirus or flavivirus positives , compared to no infection , inclusive of all seronegatives . Occupation type between ‘infection’ and ‘no infection’ groups was significantly different ( p < 0 . 001 ) . Variables relating to keeping livestock , including feeding and sources of water for livestock in wet and dry seasons , husbandry practices , meat and dairy consumption , and slaughtering practices were not indicators of previous alpha- or flavivirus infection , with the exception of grazing in the last 12 months ( p < 0 . 001 ) . Proximity of wildlife to the home was correlated with infection ( p = 0 . 05 ) . Recent flooding was also significantly correlated with previous alpha- or flavivirus infection ( p < 0 . 001 ) , whereas drought was not significant ( p = 0 . 628 ) . Fishing in the last 12 months was also found to be significant ( p = 0 . 018 ) . Water collection during dry seasons was not significant , regardless of water source , yet obtaining water from a well during the wet season was found to correlate with infection ( p = 0 . 05 ) . Variables selected by the lowest AIC glmmLasso analysis predictive of an outcome of either alphavirus or flavivirus infection included education level , specifically pre-school ( as compared to none ) ( OR = 0 . 36 , p = 0 . 014 , CI95: 0 . 16–0 . 81 ) , occupation status of student ( as compared to having no occupation ) ( OR = 0 . 36 , p = 0 . 008 , CI95: 0 . 17–0 . 77 ) , grazing in the last 12 months ( OR = 2 . 19 , p < 0 . 001 . CI95: 1 . 38 to 3 . 45 ) , recent village flooding ( OR = 2 . 49 , p = 0 . 005 , CI95: 1 . 31–4 . 73 ) , and recent fever ( OR = 0 . 52 , p = 0 . 0275 , CI95: 0 . 29–0 . 93 ) ( Table 2 ) . Significant variables selected for education level and occupation may also be used as a proxy for age . Similarly , statistically significant variables selected by CV analysis by glmmLasso with increased odds of exposure to alphavirus or flavivirus infection included sex ( male , OR = 0 . 47 , p = 0 . 009 , CI95: 0 . 27–0 . 83 ) , grazing in the last 12 months ( OR = 2 . 49 , p < 0 . 001 , CI95: 1 . 45–4 . 26 ) , and recent village flooding ( OR = 2 . 75 , p = 0 . 005 , CI95: 1 . 35–5 . 63 ) . Kernel density analysis was performed to examine spatial variation in the distribution of alphavirus or flavivirus exposure in the study area ( Fig 1A and 1B ) . To quantify the spatial clustering , a spatial scan was performed to identify geographic clusters of higher than expected case counts in the distribution of alphavirus or flavivirus exposure in the study area ( Fig 1C ) . Two clusters of flavivirus exposure were identified but were not statistically significant ( Relative Risk ( RR ) inside cluster as compared to outside; RR = 12 . 5; p > 0 . 5 and RR = 16 . 4; p > 0 . 5 ) . Strong evidence for spatial autocorrelation was identified for individual risk of alphavirus exposure with six clusters ( RR range 1 . 4–1 . 5 ) , and one statistically significant cluster ( RR = 1 . 4; p = 0 . 05 ) . This alphavirus exposure cluster was identified in the south-west corner of the study region , proximate to the Lakes Victoria , Kenyaboli , and Sare , and wetlands . Distance to nearest water body was associated with infection exposure ( median distance = 1 . 2 km; IQR = 0 . 3–2 . 4 ) compared to non-infected ( median = 1 . 7 km; IQR = 0 . 8–3 . 2 ) ( Wilcoxon rank-sum p-value = 0 . 004 ) .
Alphavirus seroprevalence was significantly higher than that of flavivirus , which is consistent with previous studies conducted in Busia[22] . Rates have not changed substantially in recent years , despite documented CHIKV and DENV outbreaks . A similar study by Mease et al . reported 59 . 91% seroprevalence for CHIKV , and 1 . 96% seroprevalence for DENV in 2004[22] . Flavivirus and alphavirus prevalence in Kenya is highly variable between regions . The low exposure to flaviviruses reported in our study is similar to that of previous surveys conducted within Busia[3] and other western parts of Kenya[20 , 21] . However , other studies have found low CHIKV seroprevalence in Western Province[20] . Flavivirus exposure is much more likely in southern[46] and coastal Kenya[3 , 21] . In a study by Sutherland et al . in 2011 , 20% of inland subjects and 37% of coastal subjects tested positive for CHIKV by IgG ELISA[21] . A 2009 study by LaBeaud et al . reported prevalence of 26% for alphaviruses along coastal Kenya[2] . The regional variability of seroprevalence for flaviviruses and alphaviruses throughout Kenya suggests exposure is likely dependent on fluctuations in climate and weather patterns , environmental features , vector abundance , and mobility and access to transportation . Seroprevalence may also be linked to areas with a higher percentage of rural versus urban villages . However , risk factors for exposure to arboviruses are not reported in all of the studies previously conducted in Kenya , making it difficult to compare factors associated with exposure for each study population . Our study represents a rural , predominantly poor population that keeps small numbers of livestock on an individual homestead , and works outside during the day . Experience of recent fever was found to be significantly relevant to IgG positives for either alphavirus or flavivirus infection , described as “arbovirus infection” , by AIC models , suggesting ongoing , interepidemic transmission of alphaviruses and flaviviruses in western Kenya . Adults aged ≥ 45 showed no exposure to flaviviruses by DENV IgG ELISA , suggesting either a more recent emergence of flaviviruses in this area , or a lower risk due to advanced age or behavioral factors . This may suggest exposure may vary by location , as students may be exposed at school , as opposed to in the homestead . The youngest seropositive participant was aged 5 , which alludes to ongoing transmission of flaviviruses , despite lack of reported outbreaks in this region . The minimal number of flavivirus seropositives overall do not indicate flaviviruses as an emergent threat to the Busia region . Other regions of Kenya , such as central and coastal villages , have reported outbreaks of flaviviruses , resulting in high seroprevalence and risk for future outbreaks[2 , 20] . Marginal flavivirus prevalence may be due to vector behavior , yet the extensive alphavirus prevalence suggests a regular , strong vector presence . Our data suggest very early and common exposure to alphavirus infections in western Kenya . The youngest seropositive participant was 5 years , demonstrating ongoing and persistent exposure throughout life in this region , or persistent antibody from an early infection in childhood . Of note , older children were more likely to be seropositive than adults , which suggests more recent exposure to alphaviruses in this region within the last decade . The CV test model suggested females were at greater risk of exposure , indicating further investigation should be dedicated to arboviral disease disparities between males and females , especially relating to behavioral or occupational differences defined by gender roles defined in communities . Kernel density and SatScan analysis of seropositives illustrated spatial clustering in the distribution of exposure to alphaviruses and flaviviruses in the study region . Areas of higher relative exposure to alphavirus and flavivirus transmission overlapped , with exposure appearing to cluster around regions with direct access to Lake Victoria . Regions such as Teso district , Bungoma county , and northern areas of Busia , which only showed high risk for alphaviruses ( represented by CHIKV in Fig 1 ) , are more than 100km away from Lake Victoria . Regions associated with high risk for both alphaviruses and flaviviruses were located primarily in households within close proximity to Lake Victoria , in the southwestern area of the study site . Proximity to bodies of water , such as Lakes , wetlands , or larger rivers , were also found to be associated with seroprevalence . Proximity to bodies of water may be related to occupation or homestead behaviors and activities , as fishing within the last 12 months was found to be significantly correlated with previous infection . Ecological variations throughout Kenya influence overall presence and abundance of vector species , as well as species diversity within specific environments[12 , 24] . This contributes to variation in risk for specific arboviruses based on the presence and abundance of the vector . Regions within close proximity to Lake Victoria and smaller bodies of water are susceptible to flooding during wet seasons , which may increase potential environments for mosquito breeding . Studies in Thailand have shown populations of vector eggs increase exponentially between the beginning to the peak of rainy seasons , directly affecting arbovirus transmission[47] . Many believe flooding is an indicator of infection risk that results from fluctuating climate extremes , such as periods of prolonged drought followed by heavy rainfall[48–51] , lead to rewetting of environments and increased water pooling and innocuous water collection around homesteads . Our results show that flooding is significantly correlated to infection , whereas drought is not , contributing further evidence for the importance of rewetting and subsequent flooding as a result of heavy rainfall in arbovirus transmission . Regions with differential exposure to alpha- or flavivirus transmission are of interest for future risk analyses . Grazing was consistently found to be significant in univariate and multivariate analyses . Activities involving animal exposure are not typically considered risk factors for alphaviruses such as CHIKV , and flaviviruses such as DENV , as transmission in humans is limited to mosquito bite . However , as climate and environment fluctuate between drought and flooding , individuals herding and grazing livestock may have an increased risk of mosquito exposure as breeding habitats expand[12] . Variables describing recent hunting , fishing , and grazing behaviors may act as proxies for time spent outdoors , where Aedes aegypti breeding sites are more common . It is also likely that factors such as time outdoors and travel to or through high transmission areas that are not located near the homestead , through grazing , fishing , and herding activities , may increase risk of exposure to alphaviruses and flaviviruses . Overlapping areas of high exposure may be due to behavior of the vector that is shared between alphaviruses and flaviviruses[24] , due to environmental factors that support mosquito proliferation , or behavioral practices that impact risk of exposure to arboviral diseases . Given the extremely low number of participants seropositive for previous flavivirus exposure , we believe the overlapping seropositives for both alphaviruses and flaviviruses could provide evidence of saturation of alphavirus exposure in this area . Co-infection of specifically CHIKV and DENV have been reported in mosquitoes and humans during a number of overlapping outbreaks[23 , 52–54] , and in mosquitoes via artificial oral exposure in a laboratory setting[55] . Yet , others suggest that competitive suppression occurs when cultured mosquito cells are co-infected with CHIKV and DENV[56] , which may explain the regional variability of alphavirus and flavivirus prevalence that has been continually reported throughout Kenya[2 , 20–22] . No outbreaks of alphaviruses have been described in this region in the last 10 years , despite common exposure . It is likely that some of the exposure represented here did not manifest clinically; however , any participants who suffered clinical disease , likely did not garner a diagnosis of arboviral infection . The clinical presentations of arboviral diseases are often highly non-specific , with the exception of severe and persistent arthralgia associated with CHIKV infections experienced by 7–79% of patients [16] . The non-specific , febrile clinical presentation of arboviruses is commonly indistinguishable from each other and from malaria[19] , which may have an effect on the accuracy of diagnoses and regional incidence reports . There are currently no diagnostics for arboviruses being routinely implemented in health centers in the study area[57] , therefore clinical cases would rarely be identified outside of research studies or referral hospitals outside of the region . Additionally , persons experiencing non-specific or mild symptoms may not seek medical attention , whether due to low-impact illness , or to other boundaries that can restrict access to medical care , such as cost , limited access to transportation , distance required to travel to a medical facility , occupational or childcare responsibilities , or corruption , which may cause inaccuracies or biases in the estimation of alphavirus and flavivirus burden in many areas . There are some limitations of this study that should be considered . All samples were tested for previous exposure to alphaviruses and flaviviruses by indirect IgG ELISA against CHIKV and DENV1–4 , which has limited specificity against cross-reactivity within each viral genus[58–60] . Viral specificity can be identified by PRNTs , which were not performed . The questionnaire utilized during sample collection was not designed to test the hypothesis , which may explain why so few variables were chosen by the glmmLasso analysis . Questionnaires did not include questions about regular exposure to mosquitoes , homestead structures that may support mosquito breeding or access , or mosquito abatement behaviors , whether on an individual or community level , which limits our ability to assess the risk factors directly related to vector behavior and exposure . Individual-level mosquito exposure is a specific area that deserves a more detailed investigation in order to fully understand the risks of arboviral transmission in the study region . Mosquito behavior was not available for integration into the kernel density analysis , which also limited our ability to definitively link mosquito abundance and likelihood of exposure to positive event densities . Due to low number of cases identified by this study , we could not present odds ratio by sub-region . Neither was the population density available at a finer scale to support estimates adjusted for population density . The cases represent a random sample of a weighted and stratified random sampling of households in the study area . The sampling and thus distribution of cases may over-represent geographic regions with greater cattle populations as a result of the sampling strategy . Our findings indicate minimal flavivirus exposure and significant alphavirus exposure in the Busia region of western Kenya . Despite our intentions of surveying prevalence and prior exposure , experience of recent fever was found to be significant to arbovirus infection , suggesting recent exposure and acute disease , and possible interepidemic transmission in this area of Kenya . Alphavirus exposure is common and occurs early on in childhood , which may have important , yet undetermined health implications . The high prevalence of alphavirus reported here , in combination with the extensive spatial clusters of autocorrelation with alphavirus exposure , indicates that vector populations are consistently prolific in western Kenya , suggesting that there are other factors influencing DENV exposure . Kernel density analysis results indicate overlapping regions of exposure for alphavirus and flavivirus transmission in western Kenya , especially in homesteads located close to Lake Victoria , suggesting environmental , behavioral , or demographic factors influence differential exposure , despite transmission by the same vectors . Further research is required to accurately determine the burden and impact of arboviruses in different localities . There is a need to increase surveillance for these infections amongst patients presenting with fever in health facilities . The presence of arboviruses in Kenya is undisputable , yet the prevalence data currently available does not accurately represent the severity of exposure , infection , and disease as it varies by region .
|
There are many examples of recent emergence of mosquito-borne viruses , such as chikungunya virus outbreaks throughout the Caribbean in 2013 , Zika virus outbreaks throughout Southern and Central America in 2015 , and yellow fever virus in Brazil in 2017 . Each outbreak draws attention to the limits associated with predicting future outbreaks . This study expands our understanding of risk factors for exposure to two common genera of mosquito-borne viruses , alphaviruses and flaviviruses . Risk factors identified include simple demographic factors , such as age or sex , and behaviors associated with occupation or livelihood around the home . Behaviors enhancing or limiting contact with mosquitoes are also significant predictors , as mosquitoes drive transmission . More clearly defining the epidemiology of these infections within a population can elevate the accuracy and efficacy of public health initiatives that fuel community education and awareness , and outbreak prediction and monitoring can be elevated to a new level of accuracy and efficacy .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"togaviruses",
"pathogens",
"geographical",
"locations",
"microbiology",
"alphaviruses",
"viruses",
"chikungunya",
"virus",
"rna",
"viruses",
"animal",
"behavior",
"flooding",
"zoology",
"africa",
"hydrology",
"infectious",
"diseases",
"grazing",
"medical",
"microbiology",
"behavior",
"microbial",
"pathogens",
"arboviral",
"infections",
"people",
"and",
"places",
"kenya",
"arboviruses",
"flaviviruses",
"viral",
"pathogens",
"earth",
"sciences",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"organisms"
] |
2017
|
Serological and spatial analysis of alphavirus and flavivirus prevalence and risk factors in a rural community in western Kenya
|
Several protein structure classification schemes exist that partition the protein universe into structural units called folds . Yet these schemes do not discuss how these units sit relative to each other in a global structure space . In this paper we construct networks that describe such global relationships between folds in the form of structural bridges . We generate these networks using four different structural alignment methods across multiple score thresholds . The networks constructed using the different methods remain a similar distance apart regardless of the probability threshold defining a structural bridge . This suggests that at least some structural bridges are method specific and that any attempt to build a picture of structural space should not be reliant on a single structural superposition method . Despite these differences all representations agree on an organisation of fold space into five principal community structures: all-α , all-β sandwiches , all-β barrels , α/β and α + β . We project estimated fold ages onto the networks and find that not only are the pairings of unconnected folds associated with higher age differences than bridged folds , but this difference increases with the number of networks displaying an edge . We also examine different centrality measures for folds within the networks and how these relate to fold age . While these measures interpret the central core of fold space in varied ways they all identify the disposition of ancestral folds to fall within this core and that of the more recently evolved structures to provide the peripheral landscape . These findings suggest that evolutionary information is encoded along these structural bridges . Finally , we identify four highly central pivotal folds representing dominant topological features which act as key attractors within our landscapes .
The vast repertoire of proteins which exist in nature are testament to billions of years of evolutionary change . The nature of their relationships and how these have evolved are questions which continue to fascinate the scientific community [1–4] . Protein structure classification schemes such as SCOP [5] and CATH [6] partition the protein universe into different structural units known as folds or topologies . Yet relationships between these folds and topologies , and how they sit relative to each other in a global structure space , are largely undiscussed by these schemes . For example , it is highly unlikely that the current repertoire of folds evolved independently of each other [7] . The evolutionary trajectory of new folds may well be through the adaptation of already existing structures . In fact , recent studies have uncovered such distant relationships between different fold units [8] . This concept has implications for protein structure classification , and more broadly within the field of protein design . A global view of the protein universe which incorporates inter-fold relationships can provide examples of efficient and evolutionary viable transitions between very different structures . More particularly , how this universe has , and continues to , evolve can be used to simulate directed evolution approaches to protein design [9 , 10] . Different techniques have previously been explored in order to generate global representations of protein structure space ( see , for example , [11] ) . Commonly , these approaches utilise structural similarities between protein domains , which produce complex , multi-dimensional data structures . The process of deriving a global landscape from these data can vary and will inevitably involve assumptions about the nature of the underlying relationships and the extent to which structural alignments can reproduce them . For example , using multi-dimensional scaling or principal component analysis can produce lower dimensional embeddings of an array of similarity scores [12–21] . In these spaces , two or three dimensional maps can be visualised which approximate the similarity between any two structures as closely as possible by their distance on the reduced axes . An alternative is to use networks to capture relationships resulting from significant alignments [13 , 22–30] . Unlike multidimensional scaling approaches , network constructions do not assume that structural similarity between protein domains is transitive [11] . On the other hand , they do require a score threshold to be set: above which an alignment is considered significant . Networks and embeddings can both be built using a variety of inputs to the similarity score [14 , 27 , 30] . For example , sequence information can be used to provide the similarity score or supplement structural alignments [12 , 14 , 15 , 29 , 31] , as can functional annotations [24 , 25 , 32] . Despite the fact that the above studies construct visualisations of the protein universe using a wide range of different methods , they present a generally consistent picture of this space . In particular , a striking partition between structures based on their secondary structure content is immediately evident [12–14 , 17 , 21 , 26 , 29 , 32] . Broadly speaking , this partition agrees with the class level classifications of SCOP and CATH , which consist of all-α domains , all-β domains and mixed αβ domains . In SCOP , mixed domains are further split into the parallel stranded α/β class and the anti-parallel α + β class . Protein space representations reveal interesting relationships within and between these groups . Several studies comment on the densely clustered group of α/β structures [26 , 31 , 32] , and the more dissipated α + β structures [12 , 29] , with all-α and all-β domains tending to congregate in between these two extremes [26] . Secondary structure seems to be one of the dominant forces within these spaces . Even when sequence signal alone is used to determine the global landscape , the main secondary structure classes , as well as other classes such as membrane proteins , divide along secondary structure lines [29] . Functional studies also point to the importance of secondary structure . For example , a recent paper exposed a functionally diverse region , at the centre of structure space , which largely overlaps with the α/β cluster [32] . Visualisations of structure space can also allow us to consider the distribution of protein domains across this space . In particular , highly dense and connected portions of the space can be used as evidence for a continuous landscape [28 , 33–35] . This conclusion has implications for the very foundations of our understanding of proteins and their evolution . A continuous protein space indicates that a partitioning of protein structures into folds and topologies is itself meaningless , as these represent discrete units of structure . The initial motivation behind the concept of a discrete fold derives from the fact that structure is a highly conserved property during evolutionary change [36] . As highly correlated to its function , the structure of a protein tends to constrain the variation which is tolerable for that domain to remain operational . The abundant structural similarities between different folds however , have stimulated a debate as to whether this is truly the case [33 , 37 , 38] . A third view has also developed , that protein space displays both discrete and continuous characteristics [31 , 33 , 34 , 39] . In particular , it has been argued that discrete and continuous paradigms of fold space do not necessarily contradict one another but form complementary descriptions of the evolutionary and structural landscapes respectively [39] . This distinction between structural and evolutionary relationships has also been implemented in the new SCOP2 prototype [40] , which separates the hierarchical structure of traditional SCOP where evolutionary units superfamilies are contained within structural units of folds into two distinct categories . It is this dual view of fold space that we will adopt here , and in particular , supplement the discrete classification of domains into folds with summaries of their geometric similarities to other folds to establish a global landscape within which these folds sit . In doing so , it is important to note the assumptions this model makes . The first , which has been stated above , is that there is a duality within the underlying dynamics of the space , where both fold classifications and structural alignments between different folds are meaningful . The second is that the methods we use to capture the discrete units and the continuous relationships are correct . We use the SCOP classification to capture the collapse of the domain universe to discrete folds . While SCOP is well established in the literature , it is by no means the only such scheme . As we have mentioned , CATH describes a complementary scheme [6] . There are also other structural schemes such as FSSP [41] , and purely sequence-based classifications , such as Pfam [42] . Similarly , there is no single established method for structural alignment and there are still many unsolved problems in the field [43 , 44] . In this paper we present several possible sets of inter-fold relationships , which we term bridges through fold space . Each set of bridges is visualised as a network over 631 SCOP folds . To build these networks we have used four different structural alignment algorithms and with each several different thresholds of similarity . We find that with all methods , the resultant organisation of structure space is at least a partial relic of the alignment algorithm used to generate it . While structural alignment programs have continued to improve in quality over recent years , generating relevant alignments consistently remains an unsolved problem [44] . By examining the areas of consensus between these maps we can more easily identify features with a higher confidence than relying on a single method in isolation . We show that such consensus spaces are vital to an appreciation of the underlying structural relationships between folds as , even at stringent threshold , a proportion of edges in a network will always be an artefact of the alignment method . Nevertheless , the different networks agree on a well defined partition of fold space into five principal community structures: all-α , all-β sandwiches , all-β barrels , α/β and α + β . We have previously used a phylogenetic analysis of fold usage to estimate an evolutionary age for different folds [45 , 46] . Age estimates relate to the emergence of a fold’s structural ancestor and are guided by its prevalence on a diverse set of completely sequenced genomes from across the tree of life . In a previous publication we found that different age estimates demonstrated particular preferences in terms of the properties exhibited by their fold structures [46] . Projecting these age estimates onto the structure space networks could provide the potential to examine the relationship between structure and evolution in a more global way . To explore this hypothesis , we examine properties of the nodes and edges of these networks in the context of their estimated age . In particular , we examine the difference in the age of two folds connected by bridges . We also look at the distribution of fold ages across the networks and how these relate to the centrality of a fold in the network’s architecture . Finally , we examine four highly pivotal ancient folds , each of which exhibit different topological properties which act as structural attractors between disparate regions of the network spaces .
Domain coordinate files for structures from the four main SCOP classes ( all-α , all-β , α/β and α + β ) were taken from the ASTRAL database ( version 1 . 75 ) and filtered to < 40% sequence identity [47] . To ensure these structures were of sufficient quality we removed any file with an assigned aerospaci score of < 0 . 4 , as suggested by Brenner et al . [47] . Due to the requirements of the structural alignment algorithms the dataset was further refined by omitting structures with only backbone Cα coordinates , and those which contained one or more chain breaks . Chain breaks were assigned using the Bio . PDB module in BioPython where successive Cα atoms were further than 4 . 3Å apart [48] . This resulted in a dataset of 4 , 098 domains , comprising 793 from the all-α class , 948 classified as all-β , 1 , 215 α/β domains and 1 , 142 from α + β . These domains represent a total of 631 folds . Four different methods were used to generate structural alignments between domains in this dataset . These methods have all been previously published and are available as open source programs or code . They are Mammoth ( MAMMOTH ) [49] , jFatcat ( FATCAT ) [50] , TM-align ( TM-ALIGN ) [51] and Elastic shape analysis ( ESA ) [52] . These methods were chosen as computationally efficient yet methodologically dissimilar representatives from the wide array of structural alignment approaches . For each of these methods , 8 , 394 , 753 = ( 4098 2 ) pairwise comparisons were computed . Each method was run using the default parameters . ESA characterised each domain backbone as a curve of N points , where N is the average length of each pair of domains . FATCAT was run in flexible mode and TM-ALIGN used a TM-score normalised by the average length of the domains . TM-ALIGN measures the strength of each alignment through the TM-score , and ESA generates an elastic metric . However , both MAMMOTH and FATCAT alignments produce multiple similarity scores for each alignment . For example , the MAMMOTH program generates a Z-score , E-value , TM score and PSI score . In these cases we chose the score that maximised the area under the ROC curve , when comparing how well each score correctly identified fold siblings under the SCOP classification . As a result of this analysis MAMMOTH alignments were summarised using the Z-score and FATCAT by the p-value ( −ln ( p ) ) . Networks were constructed from the pairwise comparisons by extracting those entries representing strong similarities between different folds . What constituted a strong similarity was determined by examining each score’s distribution and by assessing its ability to discriminate between SCOP folds . We employed a Bayesian analysis to each score , similar to that outlined in [53] . Explicitly , we considered the posterior probability that two domains were representatives of the same fold ( F = 1 ) if their similarity S was measured above a candidate threshold s ¯: P ( F = 1 | S > s ¯ ) = P ( S > s ¯ | F = 1 ) P ( F = 1 ) P ( S > s ¯ | F = 1 ) P ( F = 1 ) + P ( S > s ¯ | F = 0 ) P ( F = 0 ) The prior probabilities P ( F = 1 ) and P ( F = 0 ) were assumed to be the proportion of pairs in the domain dataset representing SCOP fold siblings and unrelated domains respectively . The conditional probabilities P ( S > s ¯ | F ) were estimated as the proportion of either the set of fold siblings or the set of pairs of unrelated domains in the dataset with similarity scores greater than s . Each pairwise alignment was thus associated with a posterior probability between zero and one , based on the relative strength of its score . For example , Fig 1a shows the relationship between the TM-scores of TM-ALIGN alignments and their posterior probabilities . For the purposes of this work we considered only scores associated with a probability ≥ 0 . 5 . We varied this cutoff between 0 . 5 and 0 . 9 to show that both the network behaviour and our results remained robust to this choice . As suggested by the FATCAT team , we calculated the significance of comparisons involving all-α domains separately to those between the other SCOP classes . This resulted in two different thresholds at each posterior probability: one applicable to alignments involving an all-α domain and one for all other alignments . The scores which were used in this analysis and the effective cutoff equivalent to different probabilities are given in Supplementary S1 Table . Networks of fold relationships were built by collapsing the 4098 × 4098 pairwise array of domains to a 631 × 631 array N of folds . Each entry of this array N ( A , B ) is characterised by the structural alignment between the pair of representative domains of folds A and B with the highest posterior probability . As the probability threshold decreased from 1 to 0 . 5 dynamic networks were constructed with folds as nodes and edges between two nodes where any two of their representative domains produced a similarity score which was associated with a probability above the threshold . For probability thresholds of 0 . 5 , 0 . 6 , 0 . 7 , 0 . 8 and 0 . 9 static networks were also built . Furthermore at each of these thresholds we constructed consensus networks built from edges between folds appearing in all four networks at that threshold . In total 25 static networks were constructed representing the four alignment methods and their consensus at the five different probability thresholds . Fig 1b shows a simplified schematic of the fold network construction process from pairs of representative domains whose alignments correspond to a posterior probability of at least 0 . 9 . Weights were added to each edge and were used to represent a measure of the distance between the folds at its endpoints . The MAMMOTH Z-score and the FATCAT p-value are statistical values and we therefore felt they were inappropriate as quantitative distances between two structures . Instead , we used the TM score as edge weights in both the FATCAT and MAMMOTH networks . TM-scores are generated as part of MAMMOTH’s output , and we calculated an approximate TM-score from FATCAT’s opt_rmsd score and the domain lengths . The TM-score was also used as weights in the TM-ALIGN network and the inverse of the elastic metric was used in the ESA network . Weights in the consensus networks were calculated by first centering weights corresponding to an individual alignment method by dividing them by their mean . Consensus weights were then calculated by averaging the respective normalised weights . Networks were visualised using Cytoscape [54] . Dynamic networks as the posterior probability on edges decreased from 1 to 0 . 5 were visualised as animations using the DynNetwork plugin . Static visualisations were calculated using a spring embedded layout , while the prefuse layout was used in the dynamic representations . Community structures were detected using the Louvain method for non-overlapping communities in weighted networks [55] . Network analysis was performed using the tnet package [56] in R [57] . Shortest path lengths d ( i , j ) between nodes i and j were calculated as the minimum sum of reciprocal weights whk along the series of edges connecting the two nodes as proposed by Dijkstra [58]: d ( i , j ) = min ( 1 w i h + … + 1 w h j ) The centrality of a node i was calculated using degree ( CD ( i ) ) , closeness ( CC ( i ) ) and betweenness ( CB ( i ) ) measures in weighted networks as suggested by Opsahl et al . [56] . Respectively , they are defined: C D ( i ) = ∑ j ∈ N w i j C C ( i ) = ∑ j ≠ i 1 d ( i , j ) C B ( i ) = ∑ j , k ≠ i σ j k ( i ) σ j k where N is the set of nodes connected by a single edge to i , wij is the weight along the edge ij , d ( i , j ) is the shortest path length between nodes i and j as defined above , σjk ( i ) is the number of shortest paths between nodes j and k which go through i , and σjk is the total number of shortest paths between j and k . Closeness and Betweenness were calculated for nodes in each connected component separately . Central and peripheral sets of folds were identified for each measure as follows . Central folds were the top 30% of nodes ranked by their centrality measures . Peripheral folds defined by closeness were the bottom 30% of nodes ranked by closeness . Degree and betweenness measures followed a skewed distribution with large numbers of nodes calculated to have very low values and far fewer folds being assigned a high degree or betweenness . Therefore , folds with peripheral degree were those with either one or no neighbour in the network . Similarly , peripheral folds by betweenness were those with a betweenness value of zero . Evolutionary age estimates were calculated for each fold following the method outlined in [45] , and more recently in [46] . These ages use a parsimony algorithm on the predicted fold content of 1014 genomes from across the sequenced tree of life to predict a relative estimate of its structural ancestor . Ages are normalised to lie between zero and one where zero corresponds to a recent ancestor , while an age of one indicates an ancestral fold predicted to exist in the last universal common ancestor . All statistics are calculated assuming an underlying phylogeny of these species as traced from the NCBI taxonomy database [59] . Populations’ age distributions were compared using the Mann Whitney U test [60] .
The landscapes of structural bridges described above represent similarities at the inter- , rather than the intra-fold level . As such , the method involves a collapse from the set of relationships across 4098 protein domains to those between their 631 different SCOP folds . This process of collapse is an important one as it imposes relationships between domains from an external classification scheme . It is also relevant in the context of comparing our networks to other fold space representations in the literature: some of which consider relationships between domains , and others those between folds . In order to illustrate the stages of the network collapse Fig 3 shows network representations of the TM-ALIGN alignments at a posterior probability threshold of 0 . 7 . These relationships are collapsed first to the SCOP family level , then to superfamilies , and finally to folds . Evident at all stages of collapse is the distinction between the different secondary structure classes . Noticeable too is the relative similarity between the superfamily and fold networks , and a more striking visual difference between the network of domains and that of families . The differences between these early stages of collapse potentially derive from the effects of multiple domains representing a small number of families . For each method the structural bridges at each probability threshold collectively determine a landscape for the global organisation of fold space . Some general network statistics relating to each construction can be found in S1 Fig . As the probability threshold increases , networks become less connected . The number of folds connected to another structure decreases ( S1a Fig ) , and even within connected components , shortest path lengths connecting two folds increase ( S1f Fig ) . The number of edges in the landscapes vary from 5 , 571 in the MAMMOTH network at a 0 . 5 threshold to 250 in the ESA network at a threshold of 0 . 9 ( S1b Fig ) . An important observation is the significant differences between the alignment algorithms , as well as their areas of agreement . In a single network generated from ESA , MAMMOTH or FATCAT alignments , about 50% of the edges were only identified by that method . For the TM-ALIGN networks , this proportion was somewhat lower at 20–30% of edges ( S1c Fig ) . Moreover , this figure does not improve with increased stringency ( i . e . increasing values of the posterior probability ) . In fact , as the similarity threshold increases , this proportion remains relatively constant , and even increases in the TM-ALIGN and ESA networks . In other words , networks constructed using different alignments remain the same distance apart regardless of similarity threshold . Taken in isolation , a proportion of edges in these networks will always be an artefact of the alignment method , emphasising the importance of considering a consensus network . As mentioned previously , connected nodes congregate , in most cases , in a single dominant connected component up till a probability threshold > 0 . 8 ( see also S1i Fig ) . The exception to this is the consensus network , where all-α folds are separated from the largest connected component . While there are separate smaller components within the networks ( S1h Fig ) , the vast majority of nodes are either part of a single connected component or are completely unconnected to other structures . This observation supports previous work suggesting that the proportion of unconnected nodes in structure networks sets these structures apart from random models [23] , and that fold space can be partitioned into either highly continuous or highly discrete sections [31] . It is also significant regarding the discussion of traversing fold space . A previous study emphasised the short path lengths between structures in fold space as indicative of a continuous space [28] . Within largest connected components we found that average path lengths were less than 5 . 5 ( S1f Fig ) . While the increase of unconnected nodes is largely responsible for the lack of traversability of networks at higher probability thresholds , it is also interesting to note that , even within connected components , average path lengths between two folds increase . This indicates that the dynamic networks transition from more continuous , connected landscapes at lower probability thresholds to more unconnected spaces at higher thresholds , although both extremes contain densely connected regions and completely unconnected folds . Previous results have indicated that α/β structures dominate the highly connected section of fold space [31] . Our results do not find such a dramatic distinction , with all four classes found within the connected component . We do however find that far fewer unconnected folds are α/β and , within the α/β cluster shortest path lengths are shorter than those of other classes . Despite these differences , several properties remain conserved across every landscape . In general , and in concert with previous observations , the networks partition fold space into the four secondary structure classes . The α/β folds form densely packed clusters , as too , to a lesser degree , do the all-α folds . On the other hand , folds with anti-parallel β sheets , belonging to the all-β and in particular to the α + β classes are more dissipated throughout the space . Nevertheless , applying a community detection algorithm to these landscapes identifies five predominant communities with a higher density of structural bridges within each group , and sparsely connected externally . These communities can be generally defined as all-α , α/β , α + β , all-β sandwiches and all-β barrels by the prevailing population of folds within these clusters . Fig 4 shows the communities in the consensus network which include these five groups along with smaller communities resulting from the smaller connected components of the network . The all-β sandwiches and barrels tend to remain partitioned from each other even at the least stringent probability threshold of 0 . 5 and are often closer in the landscapes to the α + β community than they are to each other . While the majority of previous visualisations of fold space have noted a four class clustering into SCOP classes [17 , 21 , 26] , one study also saw a division between all-β structures [13] . However , in this case β-meanders and β-zigzags were found to form the basis for this distinction . Meander structures connected to α + β structures and zigzags included both sandwiches and barrels . This is markedly different from the clusters we find here , where the basis of the division is strictly delineated by a domain’s characterisation as a barrel or sandwich . While some α + β folds appear within the sandwich and barrel clusters , in these cases they consist of well segregated α and β regions , with the β regions demonstrating the appropriate structural feature . Edges in these networks represent , not simply the phenomenon of structural similarity between proteins , but structural bridges between folds: thought to be distinct and separate structural units . We projected fold age estimates , as calculated in [46] , onto the folds in each network . These ages estimate the emergence , on a tree of sequenced life , of a fold’s structural ancestor . Each age estimate falls between zero and one , where an age of one represents an ancestral fold emerging at the root of the tree , and an age of zero signifies an ancestor at its leaves . We were thus able to consider the difference in age attributed to each of the bridges in our networks . As edges were undirected we considered the absolute difference in age of the endpoint folds to each edge ( bridge ) . We investigated the distribution of age differences , comparing those of structural bridges to a background distribution of random pairs of dissimilar folds . As described above , a large number of these bridges were identified by just a single alignment method so we examined separately the distribution of age differences on edges found on one , two , three and four networks to those found on none . Fig 5a shows a boxplot of these age differences on the set of networks built at a probability cutoff of 0 . 6 . Distributions for the other networks are similar . Not only are the pairings of unconnected folds associated with higher age differences than bridged folds , but this difference increases with the number of networks displaying the edge . Bridges identified by at least two different methods had a median age difference of zero as opposed to the 0 . 25 of unconnected folds . The prominence of each fold within these landscapes was calculated using three different centrality measures: degree , closeness and betweenness . For each measure on each network , we identified two populations: central and peripheral folds , and compared the age distributions of these two populations . In every network , including the consensus networks , and by all three of these measures central nodes were found to be significantly older than more peripheral nodes ( see S2 Fig ) . This tendency was true regardless of how we partitioned the nodes . While the three measures produced rankings for the nodes which correlated positively with each other , they all define the concept of centrality slightly differently . Fig 6 illustrates these differences on the MAMMOTH network at a threshold of 0 . 6 . While these measures interpret the central core of fold space in varied ways they all identify the disposition of ancestral folds to fall within this core and the more recently evolved structures to provide the peripheral landscape . A previous study found that clusters in structure networks could be associated with functional fingerprints [24] . Based on the assumption that older proteins will be represented by more popular clusters within the domain network , they found that older clusters were matched by a greater heterogeneity in function space . This concept of a functionally diverse ancestral core to structure space was also noted by [32] who found that this region of functional diversity largely derived from a cluster of α/β domains , which are known to be older folds [19 , 45] . We show here that the central core to our networks of structural bridges is significantly older than the peripheral nodes . It is interesting to note that these central , older folds are not in fact dominated by the α/β class . In fact , folds from the four classes are almost equally represented in these sets . Despite this , our results agree in noting the importance of ancient protein folds within fold space . The above network centrality analysis further exposed certain pivotal folds , which were calculated as highly central in all networks , including the consensus network . Here we examine four examples of pivotal folds: the long α-hairpin ( a . 2 ) , the Immunoglobulin-like β-sandwich fold ( b . 1 ) , the Flavodoxin-like fold ( c . 23 ) and the Ferrodoxin-like fold ( d . 58 ) . These folds are all ancient , with a fold age of 1 . 0 , and are represented strongly in proteins found right across the tree of life . S3 Fig shows the situation of each pivotal fold within the consensus network at a threshold of 0 . 5 . a . 2 and c . 23 remain strongly central to the communities of all-α and α/β folds respectively . They are evident as central folds at the highest thresholds of the dynamic networks . On the other hand b . 1 and d . 58 together connect much more diverse neighbourhoods within the landscapes . In particular , they have an edge between them , and their shared neighbourhood incorporates 61 structural bridges connecting together four distinct communities in the network: the α/β , all-β sandwiches , all-β barrels and the α/β cluster . We visit these folds in more detail in S1 Text . In particular , specific topological features of each fold are specified as instrumental to their highly central positions . Such features include a left-handed α-hairpin , the greek key motif and the α-β-α switch .
We have proposed and constructed a dynamic network representation of fold space to capture variations in its organisation resulting from different methodologies and similarity thresholds . While a vast array of different techniques have been applied to visualise the structural organisation of the global protein universe , very little has been done to ensure such landscapes are robust to differences in the alignment methodology which generates them . We have shown that , in terms of network representations using four dissimilar methods , there are several disagreements as to where bridges between different folds in the global space lie . We also found that these disagreements cannot be overcome by simply increasing the threshold at which a structural bridge is determined for each method . Nevertheless , the four different methods and their consensus networks do converge on certain properties of fold space . In particular , the consistent division between secondary structure classes into five predominant communities: all-α , α/β , α + β , all-β sandwiches and all-β barrels . Moreover , folds tend to fall either within a dense and easily traversable connected component , or are completely unconnected . As the probability threshold changes , the balance between these two populations shift as expected , although there remain significant numbers of each at both low and high thresholds . Structural bridges could exist for a variety of reasons . It is possible they are the result of a misannotation of fold boundaries , or that fold space is wrongly assumed to be discrete . They may also be the result of convergent evolution to a particularly favourable confirmation . They could also represent the structural relic of an evolutionary transition from one fold to another . Whatever their cause , such inter-fold similarities are deserving of further study , to illuminate the overall structure and dynamic of naturally occurring fold space . Moreover , the significant number of these bridges , especially in consensus networks representing the agreement of all four methods , suggests that structural classification , while an important and useful construct , might be a misrepresention of the true nature of the protein universe . Another feature of the core structure space is the population of ancestral folds at highly central positions within its landscape . Using each different alignment method separately , as well as in consensus , and at different levels of significance , we examined the age distributions of central and peripheral folds . We calculated the centrality of folds as nodes in each network using three different centrality measures , each with a different interpretation of the priority of a node within the landscape . In all these cases , key locations within the landscapes tend to be occupied by older folds than those at the periphery of the space . A previous study identified a functionally diverse core within fold space [61] . This core was predominantly characterised by α/β folds , which have also been identified as predominantly ancient [19 , 45] . The central folds we identify here , on the other hand , represent all four SCOP classes , and form key structural bridges both within and between the class communities . To illustrate this diversity we identified four highly central pivotal folds . These folds represent dominant structural features , such as the greek key motif , the α-β-α switch and the α hairpin which act as key attractors within our landscapes . Structural alignment in general remains an unsolved problem , and much has been written about the inaccuracies of current methodologies . For example a recent study demonstrated a high level of evolutionary inconsistency when comparing several alignment methods , including MAMMOTH , FATCAT and TM-ALIGN [44] . However , despite their limitations , these alignments can give us clues as to a global structure space , in ways in which common classification systems cannot . The representations we have included here cannot be claimed to be accurate depictions of this global space . However , there does appear to be a well defined core to this space where different alignment methods agree on the architecture and general properties of fold space . Moreover , the fact that structural bridges at the heart of this core consensus tend to fall between folds of similar age estimates lends support to the argument that evolutionary information may be encoded along these bridges .
|
Folds are considered to be the structural units which make up the protein universe . Structural classification schemes focus on the assignment and organisation of protein domains into folds . However , they do not suggest how different folds might relate to one another in a global way . We introduce the concept of bridges through fold space: significant similarities between these units . We consider four alignment methods and a dynamic approach to placing these bridges . A greater consensus between these methods cannot be achieved by simply increasing the stringency with which edges are assigned . Instead , we emphasise the importance of considering consensus maps and only report results where there is agreement across all networks . It is possible that a study of the bridges may reveal evolutionary relationships . Based on a phylogenetic analysis of structures , we find that bridges consistently fall between folds which evolved at similar times . Moreover , the landscapes all consist of a core of older folds , with younger structures more often seen at the periphery . Finally we identify four pivotal folds in the landscapes . They contain topological motifs which unite disparate regions of fold space .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Structural Bridges through Fold Space
|
Molecular epidemiology at the community level has an important guiding role in zoonotic disease control programmes where genetic markers are suitably variable to unravel the dynamics of local transmission . We evaluated the molecular diversity of Trypanosoma cruzi , the etiological agent of Chagas disease , in southern Ecuador ( Loja Province ) . This kinetoplastid parasite has traditionally been a paradigm for clonal population structure in pathogenic organisms . However , the presence of naturally occurring hybrids , mitochondrial introgression , and evidence of genetic exchange in the laboratory question this dogma . Eighty-one parasite isolates from domiciliary , peridomiciliary , and sylvatic triatomines and mammals were genotyped across 10 variable microsatellite loci . Two discrete parasite populations were defined: one predominantly composed of isolates from domestic and peridomestic foci , and another predominantly composed of isolates from sylvatic foci . Spatial genetic variation was absent from the former , suggesting rapid parasite dispersal across our study area . Furthermore , linkage equilibrium between loci , Hardy-Weinberg allele frequencies at individual loci , and a lack of repeated genotypes are indicative of frequent genetic exchange among individuals in the domestic/peridomestic population . These data represent novel population-level evidence of an extant capacity for sex among natural cycles of T . cruzi transmission . As such they have dramatic implications for our understanding of the fundamental genetics of this parasite . Our data also elucidate local disease transmission , whereby passive anthropogenic domestic mammal and triatomine dispersal across our study area is likely to account for the rapid domestic/peridomestic spread of the parasite . Finally we discuss how this , and the observed subdivision between sympatric sylvatic and domestic/peridomestic foci , can inform efforts at Chagas disease control in Ecuador .
Chagas disease , caused by the protozoan Trypanosoma cruzi , is the most important parasitic infection in Latin America [1] . An estimated 10 million people carry the infection , while another 90 million live at risk [2] . This vector-borne zoonosis causes severely debilitating and potentially deadly disease in more than a third of infected people [3] . Mucosal or abrasion contact with the infected faeces of hematophagous triatomine bugs constitutes the major mode of transmission [2] . Chagas disease is endemic to several regions in Ecuador , including the warm inter-Andean valleys of the southern province of Loja , where the main vectors are Rhodnius ecuadoriensis , Triatoma carrioni , Panstrongylus chinai , and Panstrongylus rufotuberculatus [4] , [5] . Loja Province is currently targeted by the Ecuadorian Chagas Disease Control Program . Complementing disease prevention efforts , recent progress has been made in understanding local vector dynamics [5]–[7] . However , parasite molecular epidemiology could also play a role in guiding effective intervention measures . Molecular diversity was first recognised in T . cruzi in the early 1970s [8] . Six major genetic subdivisions , known as discrete typing units ( DTUs ) , are currently recognized ( TcI–TcVI [9] ) , with distributions loosely defined by geography , transmission cycle , and ecology [1] . TcI predominates in northern South America , causes significant human disease [10] , [11] and occurs in both domestic and sylvatic cycles of parasite transmission . Of major interest to those planning sustainable control strategies in this region is the extent to which these cycles are connected [12]–[14] . The provision of such data relies on the evaluation of molecular diversity ‘hidden’ at the sub-DTU level [15]–[17] . Hypervariable molecular markers , like microsatellites , have given new and unprecedented insight into the population genetics of other important parasitic zoonoses [18]–[22] . For the first time , specific hypotheses regarding parasite dispersal and reproduction can be addressed . However , the validity of molecular epidemiological data depends heavily on study design . Numerous confounders , including biased sampling ( e . g . , sampling only one host in a heteroxenous transmission system [23] ) , population subdivision in both space and time ( leading to Wahlund effects [24] ) , and low sample size all influence the estimation of key population genetic parameters . Historically , such biases have acted as an impediment to obtaining useful epidemiological information from parasite molecular data , and , particularly in T . cruzi , to resolving the frequency of sex in natural populations . Here we present microsatellite data for 10 variable loci amplified from a large number of TcI isolates collected from domestic , peridomestic , and sylvatic hosts and vectors in and around several adjacent communities in Loja Province , Ecuador . We evaluate evidence for genetic subdivision between transmission cycles , anthropogenic dispersal of parasites between communities , and panmixia among a subset of strains .
Sixteen communities in Loja Province , southern Ecuador , were sampled ( Figure 1 ) . These communities were located at altitudes less than 2 , 200 m and were representative of the ecological diversity of the province . Trypanosomes were isolated from triatomines and small mammals ( rodents and opossums ) captured at domestic ( within dwellings ) , peridomestic ( near dwellings and/or associated with human activities , e . g . , crop stores , chicken roosts , wood and rock piles ) , and sylvatic ( more than 20 meters from dwellings ) foci ( Table S1 ) . Written informed consents from the head of the houses for domiciliary bug searches and capture of mammals near houses were obtained . These documents have been approved by the institutional review board from National Institute of Health ( NIH ) , Ohio University ( OU ) and Pontifical Catholic University of Ecuador ( PUCE ) . Vertebrates were euthanized to obtain samples; all procedures were carried out in strict accordance with the protocol approved by the Ohio University Institutional Animal Care and Use Committee ( IACUC ) . The Ohio University IACUC adheres to the guidelines in the United States Government Code of Federal Regulations ( CFR ) , Title 9 , Chapter 1 , Subchapter A- Animal Welfare Parts 1–3 and the United States Health Research Extension Act of 1985 , Public Law 99–158 “Animals in Research” . Trypanosome species was determined by PCR amplification of the kinetoplast minicircle region as in Vallejo et al . [25] . Discrete Typing Units ( DTU ) genotyping was achieved by assaying a combination of three nuclear loci as described by Lewis et al . [26] . Ten previously identified polymorphic microsatellite loci were studied ( Table S2 ) [16] . These loci are distributed across seven T . cruzi chromosomes and include two groups of physically linked markers [27] . Allelic products were amplified using previously described reaction conditions [16] . Allele sizes were determined using an automated capillary sequencer ( AB3730 , Applied Biosystems , UK ) in conjunction with a fluorescently tagged size standard and were manually checked for errors . All isolates were typed “blind” to control for user bias . By reference to a representative panel of strains , no cross reactivity was identified between T . rangeli and the microsatellite primers used in this study . Population-level genetic diversity was assessed using sample size corrected allelic richness ( Ar ) in FSTAT 2 . 9 . 3 . 2 [28] and number of private ( population specific ) alleles per locus ( PA ) . FIS , a measure of the distribution of heterozygosity within and between individuals , was estimated per locus per population in FSTAT 2 . 9 . 3 . 2 . FIS can vary between −1 ( all loci heterozygous for the same alleles ) and +1 ( all loci homozygous for different alleles ) . FIS = 0 indicates Hardy-Weinberg allele proportions . The extent of population subdivision between isolates from different transmission cycles was estimated using ( FST ) in ARLEQUIN v3 . 1 and statistical significance assessed via 10 , 000 random permutations of alleles between populations [29] . Similarly , within-population subdivision was examined in ARLEQUIN v3 . 1 , in this case using a hierarchal Analysis of Molecular Variance ( AMOVA ) . Population-level heterozygosity indices were also calculated in ARLEQUIN v3 . 1 and associated significance levels for p values derived after sequential Bonferroni correction to minimise the likelihood of Type 1 errors [30] . Individual-level pair-wise distances were estimated using DAS ( 1-proportion of shared alleles at all loci / n ) [31] under an IAM and δμ2 [32] under an SMM in MICROSAT [33] . DAS values form the basis of the dendrogram in Figure 2 . To accommodate multi-allelic loci , a script was written in Microsoft Visual Basic to make multiple random diploid re-samplings of each multilocus profile ( software available on request ) . Individual-level genetic distances were calculated as the mean across multiple re-sampled datasets . A Mantel's test for the effect of isolation by distance within populations ( pair-wise genetic vs . geographic distance ) was implemented in Genelax 6 using 10 , 000 random permutations [34] . Linkage disequilibrium indices , pair-wise ( RGGD ) and multilocus ( IA ) , were calculated in LINKDOS [35] and MULTILOCUS1 . 3b [36] , respectively . Multiple diploid re-samplings were also made to evaluate the influence of multi-allelic loci on IA , the results of which are shown in Table 1 . Assignment of individuals to populations was made by reference to the topology of the DAS derived tree . Secondarily , this model-free population assignment was corroborated using STRUCTURE ( Figure S1 ) [37] . Sample affiliations are listed in Table S1 .
We evaluated patterns of clustering and subdivision among parasite strains in the Loja samples based upon their microsatellite profiles . To identify genetically distinct groups we relied on three lines of evidence: neighbor-joining analysis based on pair-wise genetic distance; model-based population assignment ( STRUCTURE ) ; and the statistical significance of the fixation index FST . The deepest and most robust ( 56 . 5% ) internal branching within the neighbor-joining tree constructed from pair-wise genetic distance values ( DAS ) supported the delineation of two populations ( Figure 2 and Table 1 ) . No pattern or diversification by host or vector was observed within these populations . The observed bipartite subdivision was unaffected by the presence of multi-allelic loci ( 100% congruence , Figure 2 ) and was used as a means to define the populations examined in later analyses ( See Table 1 ) . Sample allocation between these two populations was exactly corroborated by the optimal number of clusters ( k ) derived using STRUCTURE software as defined by Evanno et al . [37] by Δk ( Figure S1 ) . One population , henceforth called LOJADom/Peri , was predominantly composed of isolates from domestic and peridomestic foci , the other , henceforth LOJASylv , of isolates from the sylvatic environment . Estimates of genetic subdivision ( FST ) between a priori populations ( transmission cycle defined ) corroborated this pattern of dispersal . No evidence for subdivision existed between domestic and peridomestic isolates ( FST = 0 . 027 , p = 0 . 354 ) , whereas subdivision between these populations ( grouped ) and sylvatic samples was pronounced ( FST = 0 . 212 , p<0 . 0001 ) . Naturally , reassignment of outliers to their “correct” genetic groups according to neighbor-joining and STRUCTURE analyses further inflated the latter estimate ( FST LOJADom/Peri−LOJASylv = 0 . 397 , p<0 . 0001 ) . These outliers are evidence for some , albeit limited , parasite dispersal between domestic/peridomestic transmission cycles and sylvatic transmission cycles as evident in Figure 1 and 2 . Following the identification of two genetically distinct groups of parasite strains circulating in this endemic area , the genetic diversity of each was evaluated and compared . Estimates of allelic richness ( Ar ) did not demonstrate dramatic difference between LOJADom/Peri and LOJASylv ( Table 1 ) ; both populations showed considerable genetic diversity . More private alleles per locus ( PA ) were found in the larger and marginally more diverse sylvatic population ( PA = 2 . 0; Table 1 ) . In conjunction with its apparent genetic distance from other South American TcI populations ( Figure 2 ) , the lack of private alleles within LOJADom/Peri ( PA = 0 . 8 ) suggests diversification of this population from a local source . In light of the role played by transmission cycles in structuring the local parasite population , we compared the rate of parasite dispersal within LOJADom/Peri with that within LOJASilv . This rate is inversely proportional to the amount of spatial structure ( or isolation by distance ( IBD ) ) in the population . Interestingly , tests for IBD among individuals from LOJADom/Peri and LOJASylv showed statistically significant and epidemiologically important differences between these two populations . Infinite allele models ( IAMs ) of microsatellite mutation intrinsically overestimate genetic distances between closely related isolates as compared to stepwise mutational models ( SMMs ) . To circumvent possible bias we chose to test for IBD under both . Strong evidence for spatial structure in LOJASylv existed regardless ( DAS−RXY = 0 . 265 , P<0 . 0001; δμ2−RXY = 0 . 177 , p = 0 . 001 ) . Among isolates from LOJADom/Peri , no spatial structure was evident from either measure ( DAS−RXY = 0 . 100 , p = 0 . 164; δμ2−RXY = −0 . 05 , p = 0 . 384 ) . Results are summarised in Figure 3 and strongly suggest more rapid parasite dispersal among domestic and peridomestic foci than that occurring between sylvatic locales at the same spatial scale . Several approaches were employed to estimate the rate of genetic recombination within the parasite populations identified in Loja . Multiple indicators suggested frequent sex among trypanosomes of LOJADom/Peri . Pair-wise inter-locus linkage ( RGGD ) was infrequent ( 5 . 5%; Table 1 ) even among physically linked loci ( 3/4 physically linked locus pairs , those on the same chromosome , were not statistically linked ) and despite abundant allelic diversity available within this population for inter-correlation ( the statistical power of RGGD drops dramatically with decreasing population-level genetic diversity ) . Infrequent pair-wise linkage is consistent with the lack of significance attributable to the index of association ( IA ) ( median p = 0 . 13 , P≥0 . 05 in 89% of 1000 resampled diploid datasets; Table 1 ) , and with the null hypothesis of random mating that must be assumed . Additionally , tests for deficit or excess of heterozygosity in this population showed no significant deviation from Hardy-Weinberg expectations , reflected by mean values for the inbreeding coefficient ( FIS ) across loci that approximate zero ( Table 1 ) . Finally , repeated multilocus genotypes , indicative of clonal reproduction , were absent from this population while present in LOJASylv . Other aspects of LOJASylv diversity pointed to predominant clonality , especially strong pair-wise ( 38 . 5% of locus pairs ) and multilocus linkage ( IA P<0 . 001 ) in all diploid resampled datasets ( Table 1 ) , but also strong deviation from Hardy-Weinberg levels of heterozygosity under all three measures employed ( Table 1 ) . Consistent with spatial structure identified in this population , however , an AMOVA conducted across isolates from San Jacinto and Bramaderos , which make up the majority of LOJASylv strains ( Figure 1 and Table S1 ) , did demonstrate significant but weak FST ( FST = 0 . 173 , P<0 . 0001 , 16 , 000 permutations ) , evidence that a Wahlund effect could be depressing heterozygosity . Correspondingly , estimates of linkage disequilibrium might also be somewhat inflated by subdivision in this population [38] , and it is difficult to reject the possibility that recombination may occur in the sylvatic populations at a micro-geographic scale .
This study constitutes a first attempt to understand the population dynamics of T . cruzi at a local scale using high-resolution molecular markers . The sample includes isolates from different transmission cycles , vectors , hosts , and adjacent communities . This arrangement aims to minimise sample bias and maximise the resulting molecular epidemiological inference . However , all field studies are affected by the natural abundance of hosts and vectors in different transmission cycles , and we cannot claim a perfect dataset . Nonetheless , we can report strong evidence for parasite diversification by transmission cycle , human involvement in parasite dispersal , and the possibility of sex in one parasite population . The presence of the T cruzi lineage I in southern Ecuador is consistent with reports of this DTU throughout northern South America [10] , [39] , [40] . In our study , as in other studies , sub-DTU level diversity of the parasite occurred independently of vector and host [16] , [17] , [41] . Instead , we found evidence that transmission cycle ( domestic , peridomestic , or sylvatic ) is likely to be the major driver behind parasite differentiation , apparently a phenomenon common to T . cruzi populations across much of northern South America [15] , [16] but never before studied on a local scale . On the basis of our data , we suggest that widespread , internationally distributed TcI subgroups associated with specific transmission cycles may not exist . A lack of connectivity between LOJADom/Peri and domestic TcI from Venezuela , VENDom , ( Figure 2 ) exemplifies this . Furthermore , clear cross-propagation of parasites between transmission cycles ( Figure 2 ) and few private alleles in LOJADom/Peri ( Table 1 ) suggests that these domestic groups are likely to emerge and diversify from local sylvatic sources . T . cruzi is the only stercorarian trypanosome of medical importance [42] . Natural transmission efficiency by this route ( contamination with vector feces ) is very low . The rate of transmission from infected Triatoma infestans to humans in Argentina , for example , is estimated at approximately one in 650 bites [43] . As with R . prolixus in Venezuela [14] , R . ecuadoriensis , a major disease vector in Loja , occurs at high frequency in both domestic and sylvatic locales [7] . Our data suggest that even if vector invasion from sylvatic foci is common , as in Venezuela [14] , associated transmission of parasites to domestic foci is too infrequent to break up population subdivision . Where cross-propagation does occur , circumstantial evidence incriminates synanthropic mammals as the link between transmission cycles . Didelphis marsupialis infected with parasites from the LOJADom/Peri group were found at both peridomestic ( Isolate Numbers ( IN ) 9 and 13 , Figure 2 ) and sylvatic locales ( IN 6 and 17 , Figure 2 ) . Furthermore , a R . rattus individual captured at a peridomestic site was found infected with a LOJASylv strain ( IN 31 , Figure 2 ) . Finally P . chinai and T . carrioni adults and nymphs , so far thought to be exclusively domestic and peridomestic triatomine species in Loja ( IN 27 , 28 , 58 , 68 and 81 , Figure 2 ) [5] , were found infected with a LOJASylv strain , likely as a result of contact with invasive sylvatic mammals . This blurring of the lines between transmission cycles is likely to mirror local environmental change , where human activity is driving land-use transformation . Parasite sampling in Loja was undertaken across an area only 50 km in radius ( Figure 1 ) . However , this area encompassed several ecological zones punctuated by high mountains ( >2 , 500 m in elevation ) and deep interconnecting valleys . Spatial genetic diversification among sylvatic isolates is an expected outcome given such barriers to host and vector migration ( Figure 1 and 3 ) . Conversely , parasites belonging to the LOJADom/Peri group lack this signature , a finding possibly linked to rapid anthropogenic dispersal in the form of infected individuals , livestock , or passively transported vectors and/or small peridomestic mammals . T . cruzi has , until recently , been a paradigm for clonal population structure in pathogenic organisms [44] , [45] . However , the presence of naturally occurring hybrids [46] , mitochondrial introgression [46] , and a capacity for genetic exchange in the laboratory [47] has challenged this dogma . The frequent observation of linkage disequilibrium in T . cruzi may partially stem from cryptic population subdivision ( temporal , spatial , and/or genetic ) to which linkage statistics are intrinsically sensitive [38] . Frustratingly , if assignment software with intrinsic Hardy-Weinberg assumptions ( e . g . , STRUCTURE [48] or BAPS [49] ) is used to account for subdivision prior to linkage analysis , the resulting populations will be sorted to maximise adherence to Hardy-Weinberg allelic frequencies , with artifactual sexuality a possible result [21] , [50] . Fortunately , in our study , the status of LOJADom/Peri as a stable deme is corroborated by distance-based , model-free assignment , as well as STRUCTURE . In conjunction with Hardy-Weinberg allele frequencies at individual loci , we consider , therefore , that linkage equilibrium among isolates from LOJADom/Peri represents strong evidence for frequent genetic exchange among field isolates of T . cruzi . We believe that the relatively small sample size of LOJADom/Peri ( n = 18 ) does not affect this conclusion , partly due to the ample genetic diversity present in this popualtion , but also because the lack of spatial subdivision in this group suggests frequent contact and opportunity for mixis . Thus it constitutes exactly the group of strains between which genetic exchange might be expected to occur . We cannot rule out the possibility that genetic exchange may also occur in the sylvatic cycle , if the role that substructure found in LOJASylv played in inflating linkage statistics IA and RGGD could be taken into account . However , more focused high-density sample collection from multiple individual localities would be required to address such a hypothesis . Furthermore , we cannot infer the cellular mechanism of genetic recombination events on the basis of these data . Hardy-Weinberg allelic allelic frequencies are consistent with classical meiosis . However , the lack of haploid life stages so far observed in T . cruzi are not consistent with classical meiosis , nor are the genetic exchange events so far observed in vitro [47] . Molecular epidemiology at this scale has an important guiding role to play in Chagas disease control programmes . Future efforts in Loja province must account for inter-domiciliary and inter-community parasite dispersal . This includes sustained surveillance and coordinated region-wide spraying campaigns to eliminate local vector re-invasion sources and community education to target passive triatomine dispersal routes . It is also clear that the role of synanthropic mammals cannot be overlooked as these represent an important potential link between sylvatic and domestic foci . We have shown that microsatellite markers , adequate sample sizes , and associated population statistics give fundamental insight into the genetic exchange in T . cruzi . Our results , skewed toward samples from the vector , intuitively imply that the vector may be a site of genetic exchange , as is the case for T . brucei [51] and Leishmania major [52] . The data also indicate , not surprisingly , that the majority of events probably occur within a T . cruzi lineage between epidemiologically linked strains , and these events have therefore historically been difficult to detect . The intriguing mechanisms of genetic exchange in T . cruzi warrant further investigation of their functional , adaptive , and epidemiological significance .
|
Trypanosoma cruzi is transmitted by blood sucking insects known as triatomines . This protozoan parasite commonly infects wild and domestic mammals in South and Central America . However , triatomines also transmit the parasite to people , and human infection with T . cruzi is known as Chagas disease , a major public health concern in Latin America . Understanding the complex dynamics of parasite spread between wild and domestic environments is essential to design effective control measures to prevent the spread of Chagas disease . Here we describe T . cruzi genetic diversity and population dynamics in southern Ecuador . Our findings indicate that the parasite circulates in two largely independent cycles: one corresponding to the sylvatic environment and one related to the domestic/peridomestic environment . Furthermore , our data indicate that human activity might promote parasite dispersal among communties . This information is the key for the design of control programmes in Southern Ecuador . Finally , we have encountered evidence of a sexual reproductive mode in the domestic T . cruzi population , which constitutes a new and intriguing finding with regards to the biology of this parasite .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases/protozoal",
"infections",
"infectious",
"diseases/neglected",
"tropical",
"diseases"
] |
2010
|
Sex, Subdivision, and Domestic Dispersal of Trypanosoma cruzi Lineage I in Southern Ecuador
|
Programs for control of soil-transmitted helminth ( STH ) infections are increasingly evaluating national mass drug administration ( MDA ) interventions . However , “unprogrammed deworming” ( receipt of deworming drugs outside of nationally-run STH control programs ) occurs frequently . Failure to account for these activities may compromise evaluations of MDA effectiveness . We used a cross-sectional study design to evaluate STH infection and unprogrammed deworming among infants ( aged 6–11 months ) , preschool-aged children ( PSAC , aged 1–4 years ) , and school-aged children ( SAC , aged 5–14 years ) in Kibera , Kenya , an informal settlement not currently receiving nationally-run MDA for STH . STH infection was assessed by triplicate Kato-Katz . We asked heads of households with randomly-selected children about past-year receipt and source ( s ) of deworming drugs . Local non-governmental organizations ( NGOs ) and school staff participating in school-based deworming were interviewed to collect information on drug coverage . Of 679 children ( 18 infants , 184 PSAC , and 477 SAC ) evaluated , 377 ( 55% ) reported receiving at least one unprogrammed deworming treatment during the past year . PSAC primarily received treatments from chemists ( 48 . 3% ) or healthcare centers ( 37 . 7% ) ; SAC most commonly received treatments at school ( 55 . 0% ) . Four NGOs reported past-year deworming activities at 47 of >150 schools attended by children in our study area . Past-year deworming was negatively associated with any-STH infection ( 34 . 8% vs 45 . 4% , p = 0 . 005 ) . SAC whose most recent deworming medication was sourced from a chemist were more often infected with Trichuris ( 38 . 0% ) than those who received their most recent treatment from a health center ( 17 . 3% ) or school ( 23 . 1% ) ( p = 0 . 05 ) . Unprogrammed deworming was received by more than half of children in our study area , from multiple sources . Both individual-level treatment and unprogrammed preventive chemotherapy may serve an important public health function , particularly in the absence of programmed deworming; however , they may also lead to an overestimation of programmed MDA effectiveness . A standardized , validated tool is needed to assess unprogrammed deworming .
Soil-transmitted helminth ( STH ) infections affect approximately 2 billion persons worldwide [1] , with school-aged children generally having the highest-intensity infections and highest prevalence of infection [2–6] . Improper disposal of human feces contaminated with helminth eggs exposes humans to infection following ingestion of eggs ( Trichuris trichura , or whipworm , and Ascaris lumbricoides , or roundworm ) or skin contact with larvae that hatch from eggs ( Ancystoloma duodenale and Necator americanus , or hookworm ) . A wide array of physical effects have been attributed to intestinal STH infections , including anemia ( primarily from hookworm infection ) [7–9] , Vitamin A deficiency [10] , decreased physical fitness [11] , decreased cognitive function [12 , 13] , decreased growth [12 , 14–16] , and intestinal obstruction [17] . Morbidity is directly related to infection intensity [18] . Without meaningful improvements in sanitation infrastructure in low-resource settings , elimination of STH infections is likely not feasible . Because of this , and because most morbidity is attributable to high- and moderate-intensity infections , the World Health Organization ( WHO ) recommends , rather than elimination , reduction of worm burden in individuals [19] . In 2001 , WHO set the goal of providing regular preventive deworming chemotherapy to at least 75% of at-risk school-aged children by 2010 , and urged endemic countries to develop programs to administer these drugs through schools and primary healthcare systems [20] . In 2012 , pharmaceutical companies GlaxoSmithKline and Johnson & Johnson agreed to donate billions of doses of anthelminthic drugs to countries in need [21] , enabling the expansion of existing government-sponsored deworming programs , most of which have not yet reached the 75% target . Many STH-endemic countries , including Kenya , are now planning or actively implementing national school-based deworming programs , provided as mass drug administration ( MDA ) conducted by Ministries of Health , Ministries of Education , and nongovernmental organization ( NGO ) partners [22] . Based on STH prevalence mapping and spatial modeling in Kenya [23–25] , a phased-in approach to school-based deworming was planned for selected districts in five of the country’s eight provinces , excluding Nairobi , Rift Valley , and North Eastern Provinces [22 , 26]; this program is in the process of being implemented [26] . However , smaller-scale deworming programs are also frequently carried out by other in-country partners [27] , who may use different regimens for deworming . In addition , deworming drugs are widely available from clinics , drugstores , and other sources . This ‘unprogrammed deworming’—deworming outside of the context of a nationally-administered STH control program—is frequently neither documented nor reported to health officials [27] . There is currently great interest in evaluating the effectiveness of national deworming MDA programs and progress towards the WHO treatment coverage targets for 2020 [28–31] . However , accurate reporting of deworming is required to monitor this progress . The unknown extent and patterns of unprogrammed deworming challenge both the monitoring and evaluation of STH control programs . We describe unprogrammed deworming in Kibera , an urban slum in Nairobi , Kenya . Nairobi is not included in the national deworming program , due to an overall low prevalence of STH infection [32] .
This study was approved by Institutional Review Boards at the Kenya Medical Research Institute ( KEMRI ) and the U . S . Centers for Disease Control and Prevention ( CDC ) . Written informed parental or guardian consent was required for all participants . Written assent was additionally obtained from all participants aged 13 years or older .
Of the 692 children included in the initial study , 679 had data on past-year deworming , including 18 infants , 184 PSAC , and 477 SAC . Of the 679 , 377 ( 55 . 5% ) reported receiving deworming drug treatments during the previous year . Past-year deworming occurred approximately equally among PSAC ( 62 . 0% ) and SAC ( 54 . 1% ) ( p = 0 . 07 ) , a median of three months ( range , 0–12 months ) before the interview for both groups . Five ( 27 . 7% ) infants received deworming medication during the previous year . We examined frequency of past-year deworming by one- and two-year age categories; when infants were excluded , there were no differences between these age groups ( p = 0 . 12 ) ( Fig . 1 ) . Among the 377 children dewormed during the previous year , the most common source of the most recent deworming medication was school ( 39 . 5% ) , a chemist ( independently-owned commercial drug kiosks ) ( 27 . 9% ) , or a clinic/hospital or health center ( 26 . 3% ) ( Table 1 ) . Among PSAC who were dewormed , the chemist ( 48 . 3% ) and clinic/hospital or health center ( 37 . 7% ) were the most common source of deworming medications; most SAC who were dewormed ( 55 . 0% ) received the drugs at school . The five infants who were dewormed received drugs from a clinic/hospital or health center . Of the 477 SAC , 442 ( 92 . 7% ) normally spent the day at school , while 85 ( 46 . 2% ) of the 184 PSAC spent the day at a nursery school or early childhood learning center . At least 150 different schools were named by respondents as being attended by the SAC included in our study; school name was not recorded for eight SAC . Of the 71 schools for which data were available , 45 ( 63 . 4% ) were informal , 16 ( 22 . 5% ) were private , and 10 ( 14 . 1% ) were public . Of the 393 school-aged children with data on schools attended , 194 ( 49 . 4% ) attended a public school , 140 ( 35 . 6% ) attended an informal school , and 59 ( 15 . 0% ) attended a private school . Among 383 SAC with data on both school type and deworming , those who attended a public school were more likely to have taken deworming drugs in the past year ( from any source ) ( 120/190 , 63 . 2% ) than children who attended a private school ( 29/58 , 50 . 0% ) ( p = 0 . 07 ) or an informal school ( 61/135 , 45 . 2% ) ( p = 0 . 001 ) . Among all SAC who were dewormed , public-school children were non-statistically-significantly more likely to have received their most recent deworming medications at school ( 78/120 , 65 . 0% ) than private-school children ( 16/29 , 55 . 2% ) or children attending an informal school ( 30/61 , 49 . 2% ) ( p = 0 . 11 ) . Of the 679 children in this analysis , 268 ( 39 . 5% ) were infected with at least one STH , including Trichuris ( n = 176; 25 . 9% ) , Ascaris ( n = 156; 23 . 0% ) , and hookworm ( n = 1; <1% ) . Past-year deworming was associated with reduced frequency of any STH infection ( 34 . 8% vs . 45 . 4% , p = 0 . 005 ) . When data were limited to the three most common sources of deworming medications ( school , chemist , and clinic/hospital/health center ) , SAC whose most recent deworming medication was sourced from a chemist were more frequently infected with Trichuris ( 19/50 , 38 . 0% ) than those whose most recent deworming medication was obtained from a clinic/hospital/health center ( 9/52 , 17 . 3% ) or school ( 33/143 , 23 . 1% ) ( p = 0 . 05 ) . The presence of other STH infections and any STH infection were not significantly different by treatment source among SAC or PSAC . Four NGOs were identified as providing school-based deworming medications to children in our study area and were contacted for follow-up . Local government in Nairobi County had partnered with one of these NGOs to deworm children at selected schools ( independent from the Kenyan national school-based deworming program ) ; the other three NGOs worked independently from the government and directly with schools . The NGOs reported deworming at 47 schools in Kibera during 2012; 44 ( 93 . 6% ) schools received deworming treatment once and three were dewormed twice ( a total of 50 deworming events ) . Of the schools with two deworming events during 2012 , two had both events administered by the same NGO , and one received deworming treatments from two different NGOs , during February and December 2012 . Calculated deworming coverage among enrolled pupils at these events ranged from 72%–130% ( median 98 . 7% ) . Of 31 schools with data on school type , 19 were informal , eight were public , and four were private .
Unprogrammed deworming , defined as treatment with deworming drugs outside the context of a nationally-administered STH program , is increasingly recognized in many areas that are endemic for STH infections . Our data indicate that more than half of all preschool- and school-aged children in two villages of the informal settlement of Kibera , Kenya received unprogrammed deworming treatments during 2012 . These treatments were obtained from a wide variety of sources , which differed by age group: while school-aged children most often obtained treatments in school , frequently through the efforts of NGOs , preschool-aged children more often received treatments from independent suppliers , such as clinics and chemist shops . The median time since last deworming was three months , suggesting that children may be treated several times each year . Because our survey questions were designed to identify only the most recent source of deworming medication , these data likely underestimate the true frequency of unprogrammed deworming events . Although unprogrammed deworming is rarely reported , it is likely widespread . A recent evaluation of MDA in Bangladesh described high levels of unprogrammed deworming among school-aged children living in an area already receiving programmed school-based deworming [34] . The relatively low cost of such drugs also enhances their accessibility [34 , 35]: at the local medical clinic in Kibera , albendazole and mebendazole cost approximately $0 . 02 USD for a single mebendazole tablet to approximately $0 . 22 USD for albendazole suspension ( O . Mogeni , personal communication ) . Although individual-level deworming is indicated in specific circumstances , such as for children with palmar pallor in Integrated Management of Childhood Illness ( IMCI ) programs [36 , 37] or in certain maternal health settings , mass unprogrammed deworming , when carried out in an area that already has programmed deworming , may represent wasted treatment in areas with already-limited resources . In addition to the potential for wasted treatments , unprogrammed deworming may complicate evaluations of effectiveness of nationally-implemented STH control programs . Nonreporting of deworming events to Ministries of Health or to WHO , as previously reported , [27] , compromises the monitoring of progress towards WHO-recommended anthelmintic coverage targets [18 , 20] . Typically , changes in prevalence and intensity of STH infection are assumed to be due to the effectiveness of drugs delivered through programmed MDAs [38–42] , occasionally in combination with improvements in sanitation or hygiene [43] . However , high levels of unprogrammed treatment may inflate the apparent effectiveness of programmed MDA . In our study , 55% of children received unprogrammed deworming drugs , a proportion not far below the 75% coverage target set by the WHO [20] . Unprogrammed deworming at this level would almost certainly have an impact on the evaluated effectiveness of a nationally-run MDA program . While the mutual influences of unprogrammed deworming and programmed MDA on each other’s administration frequencies are unknown , in the Bangladesh study , unprogrammed deworming continued at a high rate despite the presence of a national school-based deworming program: 38 . 7% ( 95% CI 51 . 9–64 . 4 ) of school-aged children living in a district already receiving two programmed school-based deworming events during 2009 [during which surveyed coverage was 52 . 3% ( 95% CI 43 . 6–61 . 1% ) and 54 . 3% ( 95% CI 44 . 8–63 . 8% ) ] reported that they had additionally obtained deworming drugs during that year from other , non-school sources [34] . The lack of coordinated timing of these unprogrammed treatments , even where >1 treatment per year is appropriate ( for example , in very high STH-prevalence settings ) [18] , could also influence their effectiveness . The effectiveness of unprogrammed deworming on the control and transmission of STH infections is unclear . Nairobi is not considered a high-prevalence region for STH infections [32] , and therefore Kibera is not included in the Kenyan national school-based deworming program . Receipt of unprogrammed deworming was associated with reduced STH infection in our study area , and heavy infections were very rare [33]; this may be due in part to the unprogrammed deworming events reported by participants . However , unprogrammed deworming may expose infected individuals to drugs of suboptimal quality [41] . Our data indicated that school-aged children who received their most recent deworming treatment from a chemist were more likely to be infected with Trichuris spp than those who obtained their treatment from a clinic or at school . While this may reflect the use of different ( and variably effective ) drug brands or drug qualities—for example , levamisole , widely available in this setting , is less effective against Trichuris spp than albendazole or mebendazole [44] , and its use in individual but not school-based deworming may have led to these differences—it may also be reflective of the spectrum of reasons individuals take deworming medications . For example , drugs at school may have been administered as MDA , without regard to actual infection status , while drugs from chemists may have been purchased with the intention of treating a known infection . If this is indeed the case , children whose parents purchased drugs from the chemist may be at higher risk for infection . Information about how and why parents obtain deworming medications for their children outside of MDA would be useful in answering these questions . Beyond the potential individual effects of suboptimal treatment , use of suboptimal drugs on a population level may promote the selection of worms resistant to anthelminthic treatments [41 , 45] . Veterinary data demonstrate the rapid spread of benzimidazole resistance in animal populations treated with the drug ( reviewed in [45] ) . While there are few data to suggest that anthelminthic resistance is a widespread problem in humans at present , our inability to accurately determine the true frequency of deworming makes it more difficult to monitor drug efficacy and assess the potential for anthelminthic resistance [46] . Surveys of medications available and doses recommended by chemists would provide information on the occurrence and frequency of suboptimal drug treatment , and allow for opportunities to correct common problems . As indicated from our data , in addition to NGO- and other-entity-driven deworming , individual treatment may also be common . Although reporting of individual treatment is impractical , individual treatment may prove particularly important in the‘endgame’ of STH control , when STH infections may still be present , but at too low a prevalence to warrant MDA [18] . In such settings , continued suppression of STH transmission will likely require the availability of high-quality and low-cost deworming drugs , health-seeking behavior to access these drugs on an individual basis , and improved sanitation and hygiene . In addition , individual treatment will remain important for infected subpopulations who often are not targeted for routine STH treatment , such as adult men , and , currently , women of child-bearing age , who are at increased risk of hookworm-related anemia [47–49] . Understanding the factors that drive self-treatment can serve to prepare public health officials for the STH’endgame . ’ Despite the challenges associated with unprogrammed deworming , it currently fills a need in places where programmed deworming is not occurring . Although Nairobi is not eligible for school-based MDA due to its overall low prevalence of STH infection in school-aged children , Kibera , an impoverished slum inside the city limits , clearly represents a pocket of high prevalence of STH infection , likely due to its very poor water quality and poor sanitation [50] . For school-aged children in this area , the unprogrammed deworming carried out by the local government in partnership with NGOs fills an otherwise unmet need . In addition , for preschool-aged children in Kibera , there is no programmed treatment , including MDA to treat lymphatic filariasis ( although IMCI deworming guidelines are implemented in clinical settings ) [37] . Unprogrammed treatment partially fills this gap , at no cost to national governments . It is important to encourage further understanding of how , how much , and why unprogrammed treatment occurs , to assist in evaluating its contribution to STH control both inside and outside the context of national STH control programs . To this end , the STH community must develop validated , comprehensive , and flexible tools to evaluate the frequency , source , and impact of unprogrammed deworming received whenever MDA coverage or effectiveness surveys are implemented . These tools should include questions about the number of times deworming medication was received outside of the context of MDA during the past year , where the drug was sourced from ( if known ) , the type of drug obtained , and , among persons who choose to obtain their own deworming medications , when and why they opted to do so . Sources of drugs provided by NGOs should also be investigated and a sample of locally-available deworming medications tested , to evaluate their effectiveness at delivering the promised results . Different approaches may be required in different settings to confirm the extent of unprogrammed deworming and the source of drugs for different target groups . The variety of drug sources and informants involved in providing and reporting unprogrammed deworming adds a layer of complexity to its accurate assessment . It was not possible in our study to ascertain the veracity of parent-reported deworming; it is possible that other locally-available ( and perhaps locally specific ) deworming medication sources exist , and setting-specific tools should be developed to assess these during formal evaluations of STH MDA programs , perhaps including components of record review from NGOs and schools as well as individual recall . In addition to possible recall challenges , potential limitations to our analysis include drug misclassification by respondents: what respondents thought was a deworming drug may have been something else , and what they thought was a drug for a different purpose may have been an anthelminthic . Future studies should verify agreement between the answer to deworming questions without visual aids and the answer when mothers are shown the different deworming drugs and preparations available in the area and asked which one , if any , their child received . In addition , this study took place in an area where a national government-sponsored deworming program was not occurring; were such a program taking place , the frequency of unprogrammed deworming , particularly by NGOs partnering with local government , might have been lower . However , non-school-based receipt of deworming medications accounted for approximately 60% of deworming drugs in this study , suggesting that individual deworming might continue . Partially as a result of the findings in this study , there are now discussions about implementing school-based anthelminthic MDA as part of the national program in parts of Nairobi . Should this occur , a repeat of this evaluation at that time will help shed light on the frequency of unprogrammed treatment while national government-sponsored MDA is occurring . Finally , as an urban African slum , Kibera is unusual in that it serves as a setting for multiple studies and NGO-based interventions . Due to the abundance of NGOs in Kibera , children living there may be more likely to have received NGO-sponsored unprogrammed deworming compared with children living in other urban slums . In summary , unprogrammed deworming is substantial in Kibera , Kenya . Anecdotal and limited published evidence suggests that unprogrammed deworming is both prevalent and widespread . STH control programs must develop ways to determine the extent , impact , and patterns of unprogrammed deworming to inform guidelines and rational approaches to STH control .
|
In countries with endemic soil-transmitted helminth infections , deworming medications are widely available from multiple sources , including over the counter . However , in many countries , national programs already provide deworming medications in mass drug administrations to primary school students , as part of World Health Organization recommendations . Evaluations of the effectiveness of such medications at reducing worm burden in children is based solely on the national program’s distribution schedules , primarily because little is known about how frequently deworming medications are obtained from other sources . We investigated sources of deworming medications received by children in a Kenyan slum , finding that more than half of school-aged and preschool-aged children received deworming medications outside of a national school-based deworming program . These drugs were received from multiple sources , including chemists , healthcare centers , and at schools , via the efforts of non-governmental organizations . These data strongly indicate a need to collect data on all sources of deworming medications when evaluating the effectiveness of national school-based deworming programs .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
Unprogrammed Deworming in the Kibera Slum, Nairobi: Implications for Control of Soil-Transmitted Helminthiases
|
Zoonotic Echinococcus spp . cestodes ( E . canadensis and E . multilocularis ) infect domestic animals , wildlife , and people in regions of Canada and the USA . We recovered and quantified Echinococcus spp . cestodes from 22 of 307 intestinal tracts of wild canids ( 23 wolves , 100 coyotes , 184 red and arctic foxes ) in the state of Maine and the province of Québec . We identified the species and genotypes of three Echinococcus spp . cestodes per infected animal by sequencing mitochondrial DNA at two loci . We further confirmed the absence of E . multilocularis by extracting DNA from pools of all cestodes from each animal and running a duplex PCR capable of distinguishing the two species . We detected E . canadensis ( G8 and G10 ) , but not E . multilocularis , which is emerging as an important human and animal health concern in adjacent regions . Prevalence and median intensity of E . canadensis was higher in wolves ( 35% , 460 ) than coyotes ( 14% , 358 ) . This parasite has historically been absent in Atlantic regions of North America , where suitable intermediate hosts , but not wolves , are present . Our study suggests that coyotes are serving as sylvatic definitive hosts for E . canadensis in Atlantic regions , and this may facilitate eastward range expansion of E . canadensis in the USA and Canada . As well , compared to wolves , coyotes are more likely to contaminate urban green spaces and peri-urban environments with zoonotic parasites .
Recent findings suggest that two species of Echinococcus are emerging as threats to public health in regions of Canada and the USA [1] . These include range expansion of E . multilocularis at its western and eastern distributional limits in Canada , and the novel identification of E . canadensis G8 in moose in Maine [2 , 3] . The Echinococcus genus is a group of cestodes maintained by specific predator-prey host assemblages in a wide range of geographic and climatic locales around the world [4] . Generally , wild or domestic canids act as definitive hosts , harboring adult cestodes in the small intestines ( Fig 1 ) that shed infectious eggs via feces into the environment . Intermediate hosts are specific to each Echinococcus species but can include sheep/goats , cattle , swine , cervids , horses , camels , as well as various rodent species . These hosts develop fluid-filled cysts containing the larval protoscolices that are infective to carnivore definitive hosts when ingested . People are infected when they accidentally ingest Echinococcus eggs of canid fecal origin that contaminate food , water , or the environment [5] . Over time , the ingested eggs develop into space-occupying cysts in organs such as the liver and lungs . The prognosis and medical costs for such patients is highly variable , and depend largely on the parasite species ingested , the immune status of the person , how early the infection is detected , and the level of access to medical services [5] . The species of Echinococcus present in Canada are E . canadensis ( genotypes G8 and G10 ) , also known as the cervid or sylvatic strain , and various strains of E . multilocularis[6 , 7] . In the USA , both of these sylvatic species are present , as well as one other—E . granulosus sensu stricto , also known as the domestic or sheep strain [6] . Echinococcus canadensis circulates in cervid-canid host assemblages ( e . g . moose [Alces alces]-wolf [Canis lupus] ) , and is distributed across Canada , except for the high Arctic Islands and the Atlantic provinces [8] . The distribution of E . canadensis is not as well characterized in the USA , but G8 and/or G10 genotypes have been reported in Alaska , Washington , Minnesota , and , recently , Maine [6] . Echinococcus multilocularis was traditionally considered endemic in two distinct regions: the Northern Tundra Zone ( NTZ ) in northwestern Alaska and the Canadian Arctic , and the North Central Region ( NCR ) in northcentral USA and southcentral Canada [5] . However , these two regions are no longer discontinuous and the parasite has a broader host and geographic distribution than previously suspected [9] . The hypothesis that human risk of echinococcosis is increasing in North America is supported by recent reports documenting E . canadensis and E . multilocularis outside traditional geographic and host boundaries . In 2014 , E . canadensis G8 was identified for the first time in the state of Maine in a sylvatic moose population [2] . At the same time , E . multilocularis was reported in Canadian wolves and coyotes outside of the NTZ and NCR [3 , 10] . The first canine case of alveolar echinococcosis in North America was detected in 2009 , and subsequently at least 15 cases have been detected across most of western Canada and Ontario [11 , 12] . Human cases of echinococcosis are rare in Canada , and infection origin is often not determined . Between 2002 and 2011 , the Canadian Institute for Health Information recorded 251 cases of echinococcosis ( species undetermined ) , 48 cases of cystic echinococcosis and 16 cases of alveolar echinococcosis , but did not report whether these cases were domestically acquired [13] . One of the five alveolar echinococcosis cases diagnosed in Alberta between 2013 and 2018 has so far proven autochthonous , prompting renewed public health attention to this parasite [14] ( S . Houston pers comm . ) . A key barrier to evaluating the importance of these findings is the lack of baseline data in eastern Canada and the USA . Therefore , we aimed to inform future public health threat assessments by ( i ) identifying the definitive host ( s ) of E . canadensis in Maine , and ( ii ) determining if E . multilocularis had spread from Ontario to Québec , and ( iii ) developing baseline data on Echinococcus distribution and wildlife hosts in Québec . The province of Québec was chosen because it borders on Maine , is adjacent to Ontario where E . multilocularis is rapidly emerging as a concern in canids [3] , and because surveillance data on Québec canids dates to the 1980s [15] .
Hunters and trappers in Québec and the Maine Department of Inland Fisheries and Wildlife provided carcasses of wild wolves , coyotes and foxes ( red and arctic ) that were harvested for non-research purposes over the winter of 2016/17 . Intestinal tracts were removed and stored at -80˚C for at least 5 days to inactivate infectious Echinococcus eggs as per World Health Organization standards [5] , and at -20˚C otherwise . Echinococcus spp . cestodes were collected from small intestines by the scraping , counting , and filtration method [16] after thawing the tracts at room temperature . Average infection intensity was estimated by suspending all of these cestodes in 100 mL of dH2O and counting the scolices in two 10% aliquots pipetted into grid-lined petri dishes examined under a dissecting microscope . Cestodes were stored in 90% ethanol at room temperature prior to molecular analysis . To identify Echinococcus species and genotypes , we randomly selected three intact cestodes from each infected canid and extracted DNA using a thermocycler tissue lysis technique [17] . We conducted Polymerase Chain Reaction ( PCR ) with two primer sets capable of differentiating genotypes: nicotinamide adenosine dinucleotide dehydrogenase subunit 1 ( ND1 ) and cytochrome c oxidase subunit 1 ( CO1 ) , to amplify two separate regions of mitochondrial DNA [18 , 19] . PCR products were resolved by electrophoresis on a 1 . 5% agarose gel . Single cestode PCR products were purified using the QIAquick PCR Purification Kit ( Qiagen Inc . , Valencia , California , USA ) , and sequenced ( Macrogen Inc . , Seoul , Korea ) . Forward and reverse sequences were trimmed , aligned , and then identified using the BLASTn tool to compare the similarity of sample sequences to reference sequences in the nucleotide database of GenBank [20] . Similar to Santa et al [7] , we then pooled all the Echinococcus spp . cestodes remaining from each infected animal and extracted DNA from these pools ( up to 100 mg per reaction ) using the QIAamp Fast DNA Stool Mini Kit ( Qiagen Inc . , Valencia , California , USA ) . We modified the manufacturer protocol to increase tissue disruption by shaking the cestodes at 4 m/s for 20s in lysis matrix tubes containing a garnet matrix and ¼ inch spherical beads ( MP Biomedicals; Solon , Ohio , USA ) in the initial step and eluting 100 μL of DNA in the final step . To detect E . granulosus/canadensis and E . multilocularis in these pooled samples we conducted a duplex PCR with two gene targets: ND1 and the small subunit of ribosomal of RNA ( rrnS ) [21] . All data were analyzed using R version 3 . 4 . 3 [22] . Host species were compared to infection status by a 2-sided Fisher’s exact test , using a significance threshold of 0 . 05 . The Mann-Whitney U-Test was used to evaluate infection intensity differences among canid host species . The distribution of infected and uninfected canids was mapped by entering the geographic coordinates ( latitude , longitude ) of trap sites into ArcGIS ( v10 . 2 . 2; Esri , Redlands , CA , USA ) . According to Canadian Council on Animal Care guidelines , this research was exempt from Animal Research Ethic Board review in Canada because all tissues were sourced from animals harvested for non-research purposes . This research was found to be exempt from Institutional Animal Care and Use Committee approval in the USA because all tissues were sourced from animals harvested as part of an existing coyote management program .
We examined the intestinal tracts of 23 wolves , 77 coyotes , 181 red foxes , and 3 arctic foxes submitted by Québec trappers , and 23 coyotes submitted from Maine ( Total N = 307; Fig 2 ) . Altogether , Echinococcus infection prevalence was significantly higher in wolves ( 35% of 23 , 95% CI = 16–57 ) versus coyotes ( 14% of 100 , 95% CI = 8–22; X2 ( df1 ) = 4 . 18 , p-value = 0 . 032; Table 1 ) . The prevalence difference in coyotes from Maine ( 22% of 23 , 95% CI = 7–44 ) versus Québec ( 12% of 77 , 95% CI = 5–21 ) was not significant ( X2 ( df1 ) = 0 . 77 , p-value = 0 . 30 ) . No Echinococcus spp . cestodes were detected in red or arctic foxes ( 0% of 184 , 95% CI = 0–2 ) . The overall median infection intensity was 358 ± 1508 cestodes/canid ( range: 5–6038 cestodes/canid ) , with no significant difference between wolves ( 460 ± 1110 cestodes/wolf ) and coyotes ( 358 ± 1750 cestodes/coyote; p-value = 0 . 80 ) . In most cases , we obtained mitochondrial DNA sequences for three Echinococcus spp . cestodes per host . No E . multilocularis was detected . Overall , single E . canadensis G8 infections were most common ( 11/22 , 50% ) , followed by mixed E . canadensis G8/G10 infections ( 8/22 , 36% ) and single E . canadensis G10 infections ( 3/22 , 13% ) . In Maine , we observed three canids with G8 infections and two with mixed G8/G10 infections . Our PCR duplex of pooled Echinococcus samples confirmed the results of the single cestode analysis , and detected DNA of E . canadensis but not E . multilocularis . Single cestode DNA sequences were 98–100% similar to reference sequences in the nucleotide database of GenBank . Sample sequences were most similar to complete G8 and G10 genome sequences from moose in the USA ( accession number: AB235848 ) and in Finland ( accession number: AB745463 ) , respectively [23] . There were no disagreements in species or genotype identity obtained by CO1 versus ND1 sequence data; DNA sequences from pooled samples were not sequenced to determine genotype ( s ) . High quality sequences of suitable length were trimmed ( COI- 374 bps , ND1-485 bps ) , submitted to Genbank , and assigned accession numbers ( G8: MG561268-72 , MG574822-7 MG582994-MG583003; G10:MG583004-19 ) . Infected canids were distributed from east to west across Québec and Maine with no cases observed in foxes from northern regions of Québec or in coyotes from southern coastal regions of Maine ( Fig 2 ) . Although most infected animals were captured in rural/remote areas , infected wolves and coyotes were observed close to urban centers ( e . g . Sherbrooke and Val d’Or , Québec ) . Infected wolves were observed in each locale sampled .
This study confirms the presence of E . canadensis ( G8 and G10 strains ) in wild canids on both sides of the Canada-USA border in eastern North America . In Maine , where this parasite was thought to be absent due to the historical absence of wolves , we identified coyotes as a sylvatic definitive host , and identified a previously unreported genotype to the area ( G10 ) . Previous wild canid studies have reported the G8 genotype in wolves from Canada , Russia , and the USA [1 , 24] . The G10 genotype has been reported in wolves from Canada , Russia , Estonia , Mongolia , and the USA , as well as in red foxes and coyotes in Canada [1 , 7 , 24] . A few infection clusters occurred in close proximity to urban centers in Québec , indicating that there is a need for human health professionals and veterinarians to collaboratively increase awareness about this parasite . This is especially important in light of recent human cystic echinococcosis cases caused by the G8 strain in QC ( pers . comm . C . Yansouni , McGill University Health Centre , June 1 , 2018 ) . We did not detect E . multilocularis in wild canids , despite widespread sampling and recent detection in canids from the neighbouring province of Ontario . We also did not detect the livestock variant , E . granulosus , which is endemic to certain states in the USA . Our finding of E . canadensis G8 and G10 in Maine coyotes builds upon the first published report of E . canadensis G8 in 39% of 54 moose sampled in 2014 as part of a lungworm survey , by extending the sampling area farther south [2] . Infected coyotes were only detected in the north and west of the state , suggesting that infected wildlife is likely crossing the Canada-USA border . Public health messaging in Maine should emphasize the importance of prophylactic cestocidal treatment of domestic dogs with access to cervid carcasses , such as those used for hunting , as dogs can act as bridging hosts between wildlife and people [25] . Livestock producers should be aware that E . canadensis is a risk for captive cervids [26] , but not domestic livestock . A comprehensive survey of domestic and wildlife hosts along the southern and western state limits would complete this initial assessment of Echinococcus prevalence in Maine , and allow for a more informed assessment of public health risk . As well , it might identify definitive and intermediate hosts other than coyotes and moose , as E . canadensis has previously been detected in a range of ungulate hosts in Canada , including cervids and muskoxen [8] . Within Québec , we extended the known distribution of E . canadensis infected wolves beyond the last published report ( La Verendrye Provincial Game Reserve ) in the 1980s , to include wolves trapped near Québec City towards the east and near Val d’Or towards the northwest [15] . The current infection prevalence ( 35% , N = 23 ) is lower than that previously reported ( 60% , N = 25 ) in southwestern Québec , but is similar to the 37% ( N = 191 ) prevalence reported in wolves from western and northern Canada in 2016 [10 , 15] . Although the cervid strain of Echinococcus was endemic to Québec in the 1980s , it should be noted that the molecular methods required to differentiate E . canadensis from E . granulosus did not exist at that time . Furthermore , increased sample sizes of wolves tested would improve the degree of confidence associated with prevalence estimates . Higher prevalence of E . canadensis in wolves versus coyotes might indicate that wolves predate upon intermediate hosts more frequently than coyotes . Known intermediate hosts for E . canadensis in Québec are moose , muskoxen ( Ovibos moschatus ) and caribou ( Rangifer tarandus ) , but could reasonably also include elk ( Cervus canadensis ) and deer ( Odocoileus spp . ) , as these have been reported elsewhere in Canada [8] . We did not detect E . canadensis north of Val d’Or . This is likely because we collected only foxes in northern Québec , and they are not considered common definitive hosts for E . canadensis ( Fig 2 ) in comparison to wolves and coyotes . Surveillance of Inuit and Cree communities in the north of the province report sero-prevalence to echinococcosis ( cystic or alveolar ) ranging from 0 . 7% in James Bay to 8 . 3% in Nunavik [27 , 28] , although it is unclear whether human cases have occurred . This suggests that infected wolves and/or coyotes continue to maintain the sylvatic lifecycle in northern Québec , and that people remain at risk of zoonotic transmission . We did not detect E . multilocularis in Québec , despite sampling several potential definitive host species ( coyotes , red/arctic foxes , wolves ) within a few hundred kilometers of health regions in Ontario where infected canids were detected [3] . Although this distance is well within the migratory limits of such canids [29] , it is possible that our availability sampling technique was not ideal for detection , as we collected no canids directly along the provincial border . An alternative explanation is that few canids are moving eastward from endemic hotspots in Ontario , due to the presence of three large urban centers ( i . e . Ottawa , Montréal , and Sherbrooke ) and their connecting roadways , or due to other geographic or ecological barriers . We believe our detection protocol was comprehensive , as it included morphological and molecular assessment of three Echinococcus spp . cestodes from each animal , which previously detected mixed E . canadensis/E . multilocularis infections in wolves [10] , as well as an analysis of pooled samples of all Echinococcus spp . cestodes from each positive animal . Improved antemortem diagnostic tests for detecting and differentiating Echinococcus species in canids are needed to better determine prevalence and distribution , as E . multilocularis is emerging as a threat to human and animal health in North America . Neither human nor animal echinococcosis cases are federally notifiable in Canada or the USA , although it recently became notifiable in Ontario , and surveillance of Echinococcus is outdated or nonexistent in some regions [1] . This negatively impacts efforts to track changes in Echinococcus distribution and incidence , or to accurately assess risks to animal and human health . Further work to characterize the geographic distribution , incidence , and health significance of these parasites across North America is warranted . Cross-border studies such as this one are important because emerging pathogens and their wildlife hosts do not observe political boundaries , and the movements of dogs across international borders is a known mechanism of spreading important zoonotic pathogens such as Echinococcus .
|
Echinococcosis is a zoonosis caused by ingestion of tapeworm eggs in feces of wild or domestic canids ( e . g . foxes , wolves , coyotes , and dogs ) . In North America , the number of new human echinococcosis cases reported annually is low; however , recent reports of these parasites in unusual presentations , in new locations , and in wildlife near urban areas have caused renewed interest by veterinary and human health professionals . In a cross-border collaboration , we examined the intestines of wolves ( Canis lupus ) , coyotes ( C . latrans ) and foxes ( Vulpes vulpes , V . lagopus ) trapped in Québec ( Canada ) and neighboring Maine ( USA ) , using genetic tools to identify Echinococcus tapeworms . We did not detect E . multilocularis , a serious threat to human health that has recently emerged in southern Ontario . We did identify E . canadensis in wolves and coyotes , in both Quebec and Maine . The presence of this parasite in coyotes is especially concerning because coyotes are more likely to come into close proximity with human communities . This information is relevant to veterinarians who should promote regular fecal examination and/or deworming of high risk dogs ( dogs that scavenge or hunt cervids , such as moose ) , to physicians who might encounter this relatively rare disease , and to public health agencies who should promote surveillance and develop precautions for high risk people .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"united",
"states",
"invertebrates",
"canada",
"cestodes",
"helminths",
"geographical",
"locations",
"wolves",
"vertebrates",
"animals",
"mammals",
"echinococcus",
"north",
"america",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"foxes",
"bioinformatics",
"flatworms",
"maine",
"biological",
"databases",
"people",
"and",
"places",
"coyotes",
"eukaryota",
"sequence",
"databases",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"amniotes",
"organisms"
] |
2018
|
Echinococcus in wild canids in Québec (Canada) and Maine (USA)
|
Leishmaniasis remains a worldwide public health problem . The limited therapeutic options , drug toxicity and reports of resistance , reinforce the need for the development of new treatment options . Previously , we showed that 17- ( allylamino ) -17-demethoxygeldanamycin ( 17-AAG ) , a Heat Shock Protein 90 ( HSP90 ) -specific inhibitor , reduces L . ( L . ) amazonensis infection in vitro . Herein , we expand the current knowledge on the leishmanicidal activity of 17-AAG against cutaneous leishmaniasis , employing an experimental model of infection with L . ( V . ) braziliensis . Exposure of axenic L . ( V . ) braziliensis promastigotes to 17-AAG resulted in direct dose-dependent parasite killing . These results were extended to L . ( V . ) braziliensis-infected macrophages , an effect that was dissociated from the production of nitric oxide ( NO ) , superoxide ( O−2 ) or inflammatory mediators such as TNF-α , IL-6 and MCP-1 . The leishmanicidal effect was then demonstrated in vivo , employing BALB/c mice infected with L . braziliensis . In this model , 17-AAG treatment resulted in smaller skin lesions and parasite counts were also significantly reduced . Lastly , 17-AAG showed a similar effect to amphotericin B regarding the ability to reduce parasite viability . 17-AAG effectively inhibited the growth of L . braziliensis , both in vitro and in vivo . Given the chronicity of L . ( V . ) braziliensis infection and its association with mucocutaneous leishmaniasis , 17-AAG can be envisaged as a new chemotherapeutic alternative for cutaneous Leishmaniasis .
Leishmaniasis is a widespread group of parasitic diseases caused by protozoa of the genus Leishmania , that is transmitted by the bite of female sand flies . Currently , about 12 million people are at risk of leishmaniasis and there are an estimated 1 . 5–2 million new cases each year [1] . There are two main clinical manifestations: visceral leishmaniasis , affecting mainly the spleen and liver and cutaneous leishmaniasis , affecting the skin . CL caused by Leishmania ( V . ) braziliensis is particularly distinguished from other leishmaniasis by its chronicity , latency and tendency to metastasize in the human host [2] . In 1–5% of patients , mucocutaneous leishmaniasis may develop due to the ability of L . ( V . ) braziliensis to persist within lesion scars after spontaneous or chemotherapy-mediated healing and to its ability to metastasize to the nasal mucosal [3] , [4] . In this case , extensive tissue destruction is observed , resulting from the potent cell-mediated immune response triggered by parasite replication [5] . More rarely , parasite invasion of the bloodstream results in disseminated skin lesions [6] . Brazil along with nine other countries account for 70–75% of the global estimated CL incidence [7] . The drugs of first choice for leishmaniasis chemotherapy are Pentavalent Antimonials ( Sb+5 ) [8] , which interfere with the oxidative metabolism of intracellular Leishmania [5] , [9] , [10] . These compounds are significantly toxic and have been associated with drug resistance [11] , [12] . Amphotericin B and Paramomycin , two other drugs available [13]–[15] , also display limitations with regards to toxicity , cost and/or duration of treatment [16] . In the current scenario , the identification of new chemotherapeutic compounds is urgently needed , especially since vaccines against leishmaniasis are not yet available . Heat Shock Proteins ( HSPs ) form complexes that act as chaperones , binding other proteins , denominated client proteins . These multimolecular complexes are involved in regulating protein folding , intracellular protein transport and repair or degradation of proteins partially denatured due to stress , for example [17] , [18] . Among the HSPs , HSP90 is one of the most abundant cellular chaperones and many of its client proteins are involved in cell signaling , proliferation and survival [19] . It is essential for oncogenic transformation and exploited by malignant cells to support cancer-associated kinases and transcription factors [20] . HSP90 also plays an important role in protozoans such as Leishmania and Trypanosoma , which critically rely on HSP90 for survival in alternating environments associated with their complex life cycles [21] . Therefore , HSP90-inhibitors become interesting candidates for leishmaniasis chemotherapy . Treatment of L . donovani parasites with geldanamycin ( GA ) , a HSP90-specific inhibitor , arrested promastigote growth and differentiation into amastigotes [22] . It also reduced gluthathione levels , increasing the production of reactive oxygen species ( ROS ) and promoting apoptosis [23] . Recently , we reported on the effects of 17- ( allylamino ) -17-demethoxygeldanamycin ( 17-AAG ) on L . ( L . ) amazonensis [24] . 17-AAG is a HSP90-specific inhibitor analogous to geldanamycin ( GA ) [25] . Macrophages infected with L . ( L . ) amazonensis and treated with a low dose of 17-AAG displayed significantly smaller parasite loads , an effect that was not mediated by activation of the macrophage inflammatory response [24] . In the present work , we expanded our previous observations to the effects of 17-AAG on L . ( V . ) braziliensis , the etiological agent of both cutaneous and mucocutaneous leishmaniasis in Brazil . Experiments were performed in vitro and in vivo , employing an experimental model [26] . 17-AAG was efficient at reducing L . ( V . ) braziliensis promastigote growth and macrophage infection . More importantly , 17-AAG was equally efficient in vivo , highlighting its potential as a novel chemotherapy agent against CL caused by L . ( V . ) braziliensis .
Female BALB/c mice , 6–8 weeks of age , were obtained from CPqGM/FIOCRUZ animal facility where they were maintained under pathogen-free conditions . All animal work was conducted according to the Guidelines for Animal Experimentation of the Colégio Brasileiro de Experimentação Animal and of the Conselho Nacional de Controle de Experimentação Animal . The local Ethics Committee on Animal Care and Utilization ( CEUA ) approved all procedures involving animals ( CEUA-L001/12-CPqGM/FIOCRUZ ) . 17-AAG ( 17- ( allylamino ) -17-demethoxygeldanamycin ) ( Invivogen ) was dissolved in Dimethyl sulfoxide ( DMSO ) ( SIGMA ) to a 5 mM stock solution , stored at -20°C in aliquots . For in vitro use , the stock solution was diluted in cell culture medium to the desired concentration at the time of use . For in vivo treatments , a stock solution was prepared at 100 mg/ml and diluted to 20 mg/kg at the time of use . Amphotericin B ( Fungizone , Life Technologies ) was dissolved in DMEM medium to a 250 ug/ml stock solution . The stock solution was diluted in cell culture medium to the desired concentration at the time of use . L . ( V . ) braziliensis ( MHOM/BR/01/BA788 ) [26] was cultured at 26°C in Schneider's insect medium ( Invitrogen ) supplemented with 10% inactive Fetal Bovine Serum ( FBS ) , 2 mM L-glutamine , 100 U/ml penicillin , and 100 mg/ml streptomycin ( all from Invitrogen ) . Axenic L . ( V . ) braziliensis promastigotes ( 1×106 parasites/ml ) , cultivated in supplemented Schneider medium , were treated with increasing concentrations of 17-AAG ( 25 , 75 , 125 , 250 , 500 or 625 nM ) . After 48 h , parasite viability was evaluated by direct counting of live motile parasites using a Neubauer chamber . In some experiments , promastigotes were treated with the half maximal inhibitory concentration ( IC50 ) ( 65 nM ) . After 48 h , promastigotes were washed three times with PBS and were further cultured for 24 and 48 h in supplemented Schneider medium , devoid of 17-AAG . The number of viable promastigotes was determined by direct counting . BALB/c mice were injected i . p . with 3% thioglycolate . Five days after injection , peritoneal lavage was performed using 8 ml DMEM medium supplemented with 10% Fetal Calf Serum ( FCS ) , 2 mM L-glutamine , 100 U/ml penicillin and 100 µg/ml streptomycin ( all from Invitrogen ) . To obtain monolayers , cells ( 6×105 cells/ml ) were place into glass coverslips within the wells of a 24-well plate and were left to adhere for 2 h , at 37°C and 5% CO2 . Non-adherent cells were removed by gentle and extensive washing with PBS; purity was routinely above 99% . Remaining cells ( 3×105 cells/ml ) received 3×106 cells/ml of stationary-phase L . ( V . ) braziliensis promastigotes and were incubated at 37°C in supplemented DMEM medium . After 24 h of infection , glass coverslips containing infected macrophages were washed to remove non-internalized parasites and cells were treated with different concentrations of 17-AAG ( 25 , 100 , 250 and 500 nM ) for 12–72 h . Control groups were incubated in supplemented DMEM medium containing DMSO only . Glass coverslips were washed and stained with H&E and the intracellular amastigotes were counted by light microscopy . The results are shown as the percentage of infected cells and the number of intracellular amastigotes was counted in 400 macrophages . Cultures were performed in quintuplicate . Alternatively , infected macrophages were washed extensively and the medium was replaced with 0 . 5 ml of supplemented Schneider medium , devoid of 17-AAG . Cells were cultured at 26°C for an additional 5 days and the number of viable parasites was determined by direct counting . In some experiments , infected macrophages were treated with the half maximal inhibitory concentration ( IC50 ) ( 220 nM ) of 17-AAG or with amphotericin B ( 0 . 25 µg/ml; 0 . 27 µM ) for 24 h . Parasite viability was determined by direct counting . Macrophages ( 2×105 cells/ml ) , obtained as above , were treated increasing concentrations of 17-AAG ( 39–20 , 000 nM ) or with DMSO for 72 h . Next , cultures were washed twice cells were incubated with supplemented DMEM containing 10% AlamarBlue ( Invitrogen ) . Cells were incubated for another 4 h and reagent absorbance was measured at the wavelengths of 570 nm and 600 nm using a spectrophotometer ( SPECTRA Max 340 PC ) . Ethanol-fixed cells were used as positive controls . Macrophages ( 3×106 cells/ml ) , obtained as above , were stimulated with IFN-γ ( 100 UI/ml ) ( Sigma ) and were infected with L . ( V . ) braziliensis ( 3×107 cells/ml ) for 24 h . Macrophage cultures were then washed to remove non-internalized parasites and fresh culture medium containing IFN-γ and 220 nM of 17-AAG was added . Cultures supernatants were collected 48 h later . Griess reaction was used to measure nitric oxide ( NO ) production by determining concentration of its stable reaction product nitrite ( NO2− ) [27] . Superoxide ( SO ) production was determined by adding hydroxylamine ( Sigma ) ( 0 . 5 mM ) [28] , [29] to infected macrophages . Hydroxylamine converts superoxide into nitrite , which is then be quantitated by the Griess reaction , as described above . Background levels of nitrite generated by the release of NO were determined in parallel , without the addition of hydroxylamine . Production of TNF-α , IL-6 , IL-10 and CCL2/MCP-1 was evaluated using an inflammatory Cytometric Bead Array ( BD Biosciences ) following the manufacturer's instructions . Data were acquired and analyzed using a FACSort flow cytometer and FCAP Array ( V . 3 . 0 ) ( BD Biosciences ) . Intradermal infection with L . ( V . ) braziliensis and in vivo treatment with 17-AAG BALB/c mice were inoculated with stationary-phase L . ( V . ) braziliensis promastigotes ( 105 parasites in 10 µl of saline ) in the left ear dermis using a 27 . 5-gauge needle . Four weeks post-infection , mice ( n = 10 ) were treated 3 times/wk for 3 weeks with 17-AAG ( 20 mg/kg of 17-AAG diluted in DMSO i . p . ) . The control group ( n = 10 ) received i . p . injections of DMSO in the same concentrations used in 17-AAG treated animals ( n = 10 ) . Lesion size was monitored weekly for 10 weeks using a digital caliper ( Thomas Scientific ) . Parasite load was determined using a quantitative limiting-dilution assay as described elsewhere [26] . Briefly , infected ears and lymph nodes draining the infection site were aseptically excised six weeks post-infection and homogenized in Schneider medium . Homogenates were serially diluted in supplemented Schneider complete and seeded into 96-well plates . The number of viable parasites was determined from the highest dilution at which the promastigotes could be grown after up to 2 weeks of incubation at 26°C . The half maximal inhibitory concentration ( IC50 ) of 17-AAG on L . braziliensis promastigotes and on intracellular L . braziliensis amastigotes were determined from sigmoidal regression of the concentration-responses curves , respectively , using Prism ( GraphPad Prism V . 6 . 0 ) . The selectivity index of 17-AAG was calculated as the ratio between the CC50 for murine macrophages and the IC50 for intracellular L . braziliensis amastigotes . Data are presented as the mean ± standard error of the mean . Kolmogorov-Smirnov was used for normality analysis . Parametric ( One-way ANOVA followed by Dunnett's Multiple Comparison Test or post-test for linear trend or by Bonferroni ) or non-parametric analysis ( Mann-Whitney ) tests were also performed using Prism ( GraphPad software , V . 6 . 0 ) To evaluate disease burden in mice , ear thickness of mice treated with 17-AAG or DMSO was recorded weekly for each individual mouse . The course of disease for 17-AAG-treated and control mice was plotted individually , and the area under each resulting curve was calculated using a non-parametric test ( Mann-Whitney ) . p-values ≤ 0 . 05 were considered significant .
Initially , we investigated the effects of 17-AAG on axenic L . ( V . ) braziliensis promastigotes . Parasites were incubated with increasing concentrations of 17-AAG for 48 h and viability was quantified by direct counting . All 17-AAG-treated cultures showed a significantly lower number of parasites in comparison to the control , treated with vehicle ( DMSO ) alone ( Figure 1A ) ( One-way ANOVA , p<0 . 001 ) . After 48 h of treatment , 17-AAG ( 25 nM ) reduced parasite viability by 13% ( Figure 1A ) ( compared to DMSO-treated cultures ) and increasing concentrations of 17-AAG ( 75–625 nM ) , maximized killing effects . At the highest concentration tested ( 625 nM ) parasite viability was reduced by 98% when compared to DMSO-treated cultures ( Figure 1A ) . These results also indicate that parasite viability was reduced in a dose-dependent effect ( One-way ANOVA , p<0 . 001 followed by test for linear trend p<0 . 001 ) . The DMSO concentration used was not toxic as parasite viability was similar in cultures left untreated ( Lb ) or treated with DMSO only ( Figure 1A ) . Based on these results , IC50 , after 48 h of 17-AAG treatment , was established at 65 nM ( Figure S1 ) . To evaluate whether the effect of 17-AAG on L . ( V . ) braziliensis promastigotes was reversible , parasites were treated with 17-AAG ( 65 nM ) for 48 h , washed and subsequently re-incubated in 17-AAG-free medium , for an additional 24 and 48 h . Parasite numbers were significantly reduced ( p<0 . 01 ) in cultures kept for both 24 h ( Figure 1B ) and 48 h ( Figure 1C ) . These results show that the effect of 17-AAG on L . ( V . ) braziliensis promastigotes is irreversible . Next , we investigated the effect of 17-AAG on intracellular L . ( V . ) braziliensis amastigotes . BALB/c macrophages were infected with L . ( V . ) braziliensis and cells were treated with a range of 17-AAG concentrations ( 25–500 nM ) for 12–72 h . At each time point , cells were fixed and the parasite load was assessed by light microscopy . At the initial time points ( 12 and 24 h ) , we did not detect significant alterations in treated cultures versus control cultures ( DMSO-treated ) ( Figure 2A and B ) . After 48 h , 17-AAG ( 25 nM ) reduced the infection rate to 85% ( Figure 2A ) and intracellular amastigotes were reduced to 82% ( compared to the percentages obtained in DMSO-treated cultures ) ( Figure 2B ) . With increasing concentrations of 17-AAG ( 100–500 nM ) , these effects became more pronounced and , at 500 nM , the percentage of infected cells was reduced to 63% ( Figure 2A ) and of intracellular amastigotes to 43% ( Figure 2B ) ( One-way ANOVA , p<0 . 001 ) ( again compared to the percentages obtained in DMSO-treated cultures ) . After 72 h of treatment , these effects were maximal: 500 nM of 17-AAG decreased the infection rate to 20% ( Figure 2A ) whereas intracellular amastigotes were reduced to 11% ( Figure 2B ) . The absolute percentages of infection , following treatment with different concentrations of17-AAG and the absolute numbers of amastigotes/100 cells over time are shown in Figures S2A and S2B , respectively . As with promastigotes ( Figure 1A ) , effects observed with 17-AAG on intracellular macrophages were dose-dependent ( One-way ANOVA , p<0 . 001 followed by test for linear trend p<0 . 001 ) . IC50 , after 72 h of 17-AAG treatment , was determined as 220 nM ( Figure S3 ) ; additionally , 17-AAG employed at different concentrations ( 125 , 220 and 500 nM ) did not compromise macrophage viability as assayed by MTT ( Figure S4 ) . Cytotoxicity against murine macrophages was determined upon treatment of non-infected macrophage cultures with 17-AAG with a calculated CC50 of 3 . 6 nM . The selectivity index of 17-AAG was established at 16 . 6 . Although 17-AAG treatment of L . ( V . ) braziliensis-infected macrophages for 24 h did not significantly modify the infection rate ( Figure 2 ) , we asked whether it would alter amastigote viability . Cells were infected and treated with a range of 17-AAG concentrations ( 25–500 nM ) . Intracellular parasite survival was determined following the replacement of DMEM for Schneider culture medium and direct counting of surviving L . ( V . ) braziliensis . Five days after medium replacement , treatment with increasing concentrations of 17-AAG for 24 h significantly reduced the number of viable L . ( V . ) braziliensis parasites ( Figure 3A ) ( One-way ANOVA , p<0 . 001 ) , indicating once more a dose-dependent effect ( test for linear trend p<0 . 001 ) . This effect was also time-dependent as exposure to 220 nM ( IC50 ) of 17-AAG for longer periods ( 48 and 72 h ) also significantly decreased the number of L . ( V . ) braziliensis promastigotes ( Figure 3B ) . Therefore , exposure of infected macrophages to 17-AAG negatively impacted on the survival of L . ( V . ) braziliensis . Macrophage activation and production of nitric oxide and superoxide are key steps towards elimination of intracellular Leishmania [30] . In macrophages infected with L . ( V . ) braziliensis and treated with 220 nM 17-AAG ( IC50 ) , production of nitric oxide ( Figure 4A ) and superoxide ( Figure 4B ) were lower compared to cells exposed to DMSO . 17-AAG-treatment also down-modulated the production of TNF-α ( Figure 4C ) , IL-6 ( Figure 4D ) and CCL2/MCP-1 ( Figure 4E ) by L . ( V . ) braziliensis-infected macrophages . Therefore , the leishmanicidal effect of 17-AAG is uncoupled from the production of microbicidal molecules and from the production of pro-inflammatory cytokines . Next , we tested the in vivo effect of 17-AAG against CL caused by ( V . ) L . braziliensis . These experiments were performed in an mouse model that reproduces aspects of the natural infection such as the presence of an ulcerated lesion , parasite dissemination to lymphoid areas and the development of a Th-1 type immune response [26] . BALB/c mice were inoculated with L . ( V . ) braziliensis in the ear dermis and lesion development was monitored weekly . Four weeks after infection , mice were treated with 17-AAG or with vehicle ( DMSO ) alone , three times a week , for three weeks . The ear thickness of mice treated with 17-AAG was significantly smaller compared to controls ( Figure 5A ) . Disease burden , calculated by the area under the curve ( AUC ) obtained for the two experimental groups , was also significantly ( p<0 . 05 ) different ( Figure 5B ) , demonstrating the ability of 17-AAG to control lesion development , in vivo . 17-AAG treatment significantly ( p<0 . 05 ) reduced the parasite load at the infection site , six weeks post inoculation ( Figure 5C ) . However , treatment with 17-AAG was not able to reduce the parasite load within draining lymph nodes ( dLN ) ( Figure 5D ) . In order to further characterize 17-AAG as an anti-leishmanial , its effect on infected macrophages was compared to that exerted by amphotericin B . Macrophages were infected with L . ( V . ) braziliensis and treated with 220 nM 17-AAG ( IC50 ) or with amphotericin B ( AMB ) ( 0 . 27 µM ) . Forty-eight hours after medium replacement ( Figure 6 ) , cultures treated with either 17-AAG or amphotericin B displayed significantly lower parasite numbers ( p<0 . 01 ) in comparison with controls , treated with vehicle ( DMSO ) alone .
HSP90 is a molecular chaperone fundamental for the life cycle of a variety or protozoa [31] and , as such , inhibitors of HSP90 have been suggested as novel chemotherapeutic agents against malaria [32] , filariasis [33] , [34] and schistosomiasis [34] . Recently , we showed that 17-AAG , a HSP90 inhibitor , reduced L . ( L . ) amazonensis infection in vitro [24] . Herein , we investigated the potential of 17-AAG as a chemotherapeutic agent against L . ( V . ) braziliensis , the main etiological agent of CL and MCL in Brazil [35] . We confirmed the effects of 17-AAG against this L . ( V . ) braziliensis promastigotes and we extended these findings to a pre-clinical model of CL . Initially , we investigated the in vitro effects of 17-AAG , against both axenic promastigotes and intracellular amastigotes . Treatment of L . braziliensis promastigotes with the lower dose of 17-AAG ( 25 nM ) already decreased promastigote viability . Herein , the IC50 determined for L . ( V ) . braziliensis was comparable to that described for L . ( L ) . amazonensis ( 65 nM ) whereas in experiments performed with L . ( L ) . major , the IC50 was established at 80 nM [24] . 17-AAG was equally effective at reducing intracellular amastigote numbers and the viability of surviving L . ( V ) . braziliensis promastigotes . These effects were not associated with an increase in the microbicidal functions of macrophages as levels of NO , superoxide and TNF-α were diminished in the presence of 17-AAG . These results are in accordance with our previous report [24] . Additionally , the lack of amastigote replication in control macrophages could be attributed to innate microbocidal properties of macrophages that allow L . ( V ) . braziliensis killing , as observed with L . ( V ) . guyanensis and L . ( L . ) major [36]–[38] . In CL patients , an exacerbated inflammatory immune response is associated with the development of mucocutaneous leishmaniasis ( rev . in [39] ) whereas subclinical patients , who do not develop the disease , have a more controlled immune response [40] . Therefore , the possibility of selectively inducing parasite killing without contributing to overt inflammation is an important advantage for the treatment of CL using 17-AAG . Of note , macrophages treated with 17-DMAG alone displayed reduced production of IL-6 , TNF-a and NO [41] whereas 17-AAG prevented iNOS expression upon stimulation with LPS or IFN-g [42] . Geldanamycin ( GA ) , a HSP90 inhibitor analogous to 17-AAG , induces an anti-oxidative and attenuated inflammatory response in sepsis [43] , autoimmune encephalitis [44] , experimental atherosclerosis [45] and endotoxin-induced uveitis [46] . The proposed mechanism for these effects is the reduced nuclear translocation of NF-κB , reflecting in decreased production of IL-6 , TNF-α and NO [41] . Although we cannot extrapolate the complexities of in vivo situations cited above , L . ( L . ) amazonensis-infected macrophages treated with 17-AAG displayed parasite killing , in spite of a diminished production of inflammatory mediators [24] . It has been shown that in BALB/c mice infected with L . ( V ) . braziliensis , the density of INOS+ cells was higher when compared to L . ( L ) . amazonensis-infected mice [47] . So , different responses to NO between L . ( V . ) braziliensis and L . ( L . ) amazonensis could also impact on the killing effect exerted by 17-AAG . In our previous work , [24] , we showed that L . ( L ) amazonensis amastigotes displayed structural alterations following exposure of infected macrophages to 17-AAG . Visible alterations in the cytoplasm of parasites such as the presence of myelin figures , vesicles with double-layered membranes and mitochondrial segments inside membrane-bounded structures were the suggestive indications of autophagy , a process that naturally occurs in Leishmania and which plays an important role in the transition from promastigote to amastigote [48] . It is possible that inhibition of HSP90 activity interferes with cell cycle progression , blocking differentiation or expression of stage specific protein and , consequently , affecting survival in the intracellular environment . 17-AAG was also effective in vivo: mice infected with L . ( V . ) braziliensis and treated with 17-AAG showed a significantly smaller disease burden in parallel to a smaller parasite load at the infection site . However , 17-AAG was not able to alter parasite load at the draining lymph nodes ( dLN ) , a site where L . ( V . ) braziliensis parasites persist following lesion healing [26] . In this experimental model , parasite persistence is associated with the presence of regulatory T cells ( Tregs ) that accumulate within dLNs of L . ( V . ) braziliensis-infected mice [49] and these Tregs control Th1 responses by IL-10-dependent mechanisms [50] . Although 17-AAG treatment controlled parasite replication at the infection site and promoted lesion healing , parasite persistence within distal sites such as the dLNs may have important effects with regards to maintenance of immunity to Leishmania [51] and/or development of mucotutaneous leishmaniasis , deserving further investigation . Currently , the drugs available for the treatment of CL are limited and among them , pentavalent antimonials have been the choice for over 60 years . However , treatment is long ( 20-30 days ) , patients develop several side effects and , in the recent years , the number of cases refractory to treatment has increased [52] , [53] . In the case of therapeutic failure , second-line drugs such as amphotericin B can be employed as well as combination of two available drugs [54] . Advantages of a combination treatment include increased efficacy , less drug resistance , lower drug dosage and a general decrease in side effects [55] . Herein , 17-AAG was as effective as amphotericin B at decreasing the parasite load within infected macrophages . Experimentally , the treatment of L . infantum and L . panamensis promastigotes with 17-AAG plus edelfosine improved the anti-leishmanicidal activity of the latter [56] . In vitro synergism was also observed for the combinations of paramomycin and amphotericin B against L . ( V . ) braziliensis [57] . In vivo , association of tamoxifen with amphotericin B yielded an additive effect in mice infected with L . ( L . ) amazonensis [58] . The combination of GA with fluconazole showed synergistic activity against Candida albicans isolates resistant to fluconazole alone [59] . Thus , we propose that combinations of 17-AAG and amphotericin B may be further investigated for the treatment of CL caused by L . ( V . ) braziliensis . Herein , we reported that 17-AAG reduces L . ( V . ) braziliensis infection in vitro and in vivo . 17-AAG shows excellent bioavailability when given to mice by the i . p . route [60] . At 60 mg/kg , 17-AAG caused no changes in appearance , appetite , waste elimination , or survival of treated animals as compared to vehicle-treated controls . We employed with 20 mg/kg , a dose well below that reported as having any harmful effects , as those decribed by Solit et al . ( equal or above 75 mg/kg ) [61] . Given that HSP90 inhibitors , analogous to 17-AAG , have entered clinical trials with cancer patients [62] , we propose that 17-AAG could be further investigated as a novel target for chemotherapy against cutaneous leishmaniasis .
|
Antimony-containing compounds are the main drugs used to treat leishmaniasis but the severe associated side effects pose the need for alternative chemotherapeutic options . Herein , we evaluated the ability of 17-AAG ( a Heat Shock Protein 90 inhibitor ) to kill Leishmania ( Viannia ) braziliensis parasites , a species that causes both cutaneous and mucocutaneous Leishmaniasis in Brazil . Heat Shock Protein 90 ( HSP90 ) is associated with important biological processes; inhibition of this molecule interferes with parasite survival and , hence , it can be exploited as a chemotherapeutic target . We show that exposure to 17-AAG induced killing of L . brazilensis parasites in both its extracellular and intracellular forms . This effect was not dependent on the activation of the host cell . More importantly , treatment of mice infected with L . ( V . ) braziliensis also modulated lesion development and decreased parasite growth at the infection site . Collectively , our results show that targeting HSP90 is a promising alternative for development of novel chemotherapeutic options for leishmaniasis .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases",
"veterinary",
"diseases",
"zoonoses",
"medicine",
"and",
"health",
"sciences",
"leishmaniasis",
"neglected",
"tropical",
"diseases",
"biology",
"and",
"life",
"sciences",
"tropical",
"diseases",
"protozoan",
"infections",
"parasitic",
"diseases",
"veterinary",
"science"
] |
2014
|
Chemotherapeutic Potential of 17-AAG against Cutaneous Leishmaniasis Caused by Leishmania (Viannia) braziliensis
|
Identification of common mechanistic principles that shed light on the action of the many chemically diverse toxicants to which we are exposed is of central importance in understanding how toxicants disrupt normal cellular function and in developing more effective means of protecting against such effects . Of particular importance is identifying mechanisms operative at environmentally relevant toxicant exposure levels . Chemically diverse toxicants exhibit striking convergence , at environmentally relevant exposure levels , on pathway-specific disruption of receptor tyrosine kinase ( RTK ) signaling required for cell division in central nervous system ( CNS ) progenitor cells . Relatively small toxicant-induced increases in oxidative status are associated with Fyn kinase activation , leading to secondary activation of the c-Cbl ubiquitin ligase . Fyn/c-Cbl pathway activation by these pro-oxidative changes causes specific reductions , in vitro and in vivo , in levels of the c-Cbl target platelet-derived growth factor receptor-α and other c-Cbl targets , but not of the TrkC RTK ( which is not a c-Cbl target ) . Sequential Fyn and c-Cbl activation , with consequent pathway-specific suppression of RTK signaling , is induced by levels of methylmercury and lead that affect large segments of the population , as well as by paraquat , an organic herbicide . Our results identify a novel regulatory pathway of oxidant-mediated Fyn/c-Cbl activation as a shared mechanism of action of chemically diverse toxicants at environmentally relevant levels , and as a means by which increased oxidative status may disrupt mitogenic signaling . These results provide one of a small number of general mechanistic principles in toxicology , and the only such principle integrating toxicology , precursor cell biology , redox biology , and signaling pathway analysis in a predictive framework of broad potential relevance to the understanding of pro-oxidant–mediated disruption of normal development .
Determining whether chemically diverse substances induce similar adverse effects at the cellular and molecular level is one of the central challenges of toxicological research . If the structural diversity of different toxicants , and of potential toxicants , means that each works through distinctive mechanisms then this creates a potentially unsolvable challenge in developing means of screening the many tens of thousands of different chemicals for which little or no toxicological information exists . In contrast , the identification of general principles that transcend the specific chemistries of individual substances has the potential of providing broadly relevant insights into the means by which toxicants disrupt normal development . If such principles were found to apply to the analysis of toxicant levels frequently encountered in the environment , this would be of even greater potential importance in providing efficient means of analyzing this diverse array of chemicals . Of all of the effects associated with toxicant exposure , one of the few that appears to be common to multiple chemically diverse substances is the ability of these agents to cause cells to become more oxidized . The range of toxicants reported to alter oxidative status is very broad , and includes metal toxicants such as methylmercury ( MeHg; e . g . , [1–6] , lead [Pb] [6–9] , and organotin compounds [1 , 2 , 5 , 10 , 11] ) , cadmium [12 , 13] , and arsenic [12 , 14] . Ethanol exposure also is associated with oxidative stress [15] , as is exposure to a diverse assortment of agricultural chemicals [16] , including herbicides ( e . g . , paraquat [17 , 18] ) , pyrethroids [19–21] , and organophosphate and carbamate inhibitors of cholinesterase [22–26] ) . Thus , the ability to cause cells to become more oxidized is shared by many toxicants , regardless of their chemical structure . The observations that chemically diverse toxicants share the property of making cells more oxidized is of particular interest in light of the increasing evidence that oxidative regulation is a central modulator of normal physiological function . Although increases in oxidative status in a cell have been most extensively studied in the context of their adverse effects ( in particular , the induction of cell death or of cell senescence ) , multiple studies have demonstrated that changes in redox state as small as 15%–20% may be critical in regulating such normal cellular processes as signal transduction , division , differentiation , and transcription ( reviewed in , e . g . , [27–31] . Although the mechanistic basis for such regulation is frequently unclear , the importance of redox status in modulating cell function makes convergence of different toxicants on this physiological parameter a matter of considerable potential interest . Despite the observations that many toxicants share the property of making cells more oxidized , multiple questions exist regarding the relevance of such observations for the understanding of toxicant function . First , there is considerable uncertainty about the relative importance of effects on redox state in the analysis of individual toxicants , and it is generally believed that the major effects of toxicants on cellular function are distinct from any effects on oxidative status . For example , in the context of agents analyzed in the present studies , MeHg-mediated effects on cellular function generally are thought to be mediated through binding to cysteine residues , thus disrupting function of microtubules and other proteins , but may also involve disruption of Ca2+ homeostasis ( e . g . , [32–34] ) . In contrast , Pb does not bind to cysteine residues and instead is thought to exert its functions through altering normal calcium metabolism by mimicking calcium action and/or by disrupting calcium homeostasis ( e . g . , [35 , 36] ) . This would lead to alterations in function of multiple proteins , of which the most extensively studied have been members of the protein kinase C ( PKC ) family of enzymes ( e . g . , [37 , 38] ) . A further concern is the general lack of knowledge about whether , or where , oxidation induced by different means would mechanistically converge . For example , MeHg has been suggested to cause oxidative stress by a variety of mechanisms , including by binding to thiols , by causing a depletion in glutathione levels , or by impairing mitochondrial function [39 , 40] , whereas Pb is thought to disrupt mitochondrial function through its effects on calcium metabolism ( e . g . , [35 , 36 , 41–45] ) The organic herbicide paraquat ( the third agent examined in the present studies ) is another example of a toxicant with pro-oxidant activities , but in this case , resulting from initiation of a cyclic oxidation/reduction process in which paraquat first undergoes one electron reduction by NADPH to form free radicals that donate their electron to O2 , producing a superoxide radical; upon exhaustion of NADPH , superoxide reacts with itself and produces hydroxyl free radicals ( e . g . , [17 , 18] ) . Whether these different means of altering oxidative state would have different mechanistic consequences is unknown . A further concern regarding the hypothesis that changes in redox state represent an important convergence point of toxicant action is whether oxidative changes are even associated with toxicant exposure at levels frequently encountered in the environment . For example , although several studies have documented the ability of MeHg to cause cells to become more oxidized , effective exposure levels employed in these studies have generally ranged from 1–20 μM [2–5] , which is 30–600 times the upper range of average mercury concentrations found in the bloodstream of as many as 600 , 000 newborn infants in the United States alone [46] . Similar concerns apply to the analysis of multiple toxicants , for which pro-oxidant effects have largely been studied at exposure levels much higher than those with broad environmental relevance . In addition , a more general concern regarding the search for general principles of toxicant action is whether such convergence , if it exists , would occur only at exposure levels that induce cell death or whether common mechanisms might be relevant to the understanding of more subtle effects of toxicant exposure , particularly during critical developmental periods . Because development is a cumulative process , the effects of small changes in , e . g . , progenitor cell division and/or differentiation , that are maintained over multiple cellular generations could have substantial effects on the organism . Such changes are poorly understood , however , at both cellular and molecular levels . Our present studies have led to the discovery of a previously unrecognized regulatory pathway on which environmentally relevant levels of chemically diverse toxicants converge to compromise division of a progenitor cell isolated from the developing central nervous system ( CNS ) . We found that exposure of cells to low levels of MeHg , Pb , or paraquat is sufficient to make cells more oxidized and to activate Fyn kinase , a Src family member known to be activated by increased oxidative status . This first step activates a pathway wherein Fyn activates c-Cbl , a ubiquitin ligase that plays a critical role in modulating degradation of a specific subset of receptor tyrosine kinases ( RTKs ) . c-Cbl activation in turn leads to reductions in levels of target RTKs , thus suppressing division of glial progenitor cells . The effects of all three toxicants are blocked by co-exposure to N-acetyl-L-cysteine , which is widely used to protect against oxidative stress . We also provide evidence that our in vitro analyses successfully predict previously unrecognized effects of developmental MeHg exposure at levels 90% below those previously considered to represent low-dose exposure levels .
The progenitor cells that give rise to the myelin-forming oligodendrocytes of the CNS offer multiple unique advantages for the study of toxicant action , particularly in the context of analysis of toxicant effects mediated by changes in intracellular redox state . These progenitors ( which are referred to as both oligodendrocyte-type-2 astrocyte [O-2A] progenitor cells ( [47] and oligodendrocyte precursor cells , here abbreviated as O-2A/OPCs ) are one of the most extensively studied of progenitor cell populations ( reviewed in , e . g . , [48–52] . They also are among a small number of primary cell types that can be analyzed as purified populations , and at the clonal level , and for which there is both extensive information on the regulation of their development and also evidence of their importance as targets of multiple toxicants ( including such chemically diverse substances as Pb [38 , 53] , ethanol [e . g . , [54–57]] , and triethyltin [10 , 58] ) . Another important feature of O-2A/OPCs , in regard to the present studies , is that their responsiveness to small ( ∼15%–20% ) changes in the intracellular redox state provides a central integrating mechanism for the control of their division and differentiation [59] . O-2A/OPCs purified from developing animals on the basis of the cell's intracellular redox state exhibit strikingly different propensities to divide or differentiate . Cells that are more reduced at the time of their isolation undergo extended division when grown in the presence of platelet-derived growth factor ( PDGF , the major mitogen for O-2A/OPCs [60–62] ) , whereas those that are more oxidized are more prone to undergo differentiation [59] . Pharmacological agents that make cells slightly more reduced enhance self-renewal of dividing progenitors , whereas pharmacological agents that make cells more oxidized , by as little as 15%–20% , suppress division and induce oligodendrocyte generation . Moreover , cell-extrinsic signaling molecules ( e . g . , neurotrophin-3 [NT-3] and fibroblast growth factor-2 [FGF-2] ) that enhance the self-renewal of progenitors dividing in response to PDGF cause cells to become more reduced . In contrast , signaling molecules that induce differentiation to oligodendrocytes ( i . e . , thyroid hormone [TH] [63 , 64] ) or astrocytes ( i . e . , bone morphogenetic protein-4 [BMP-4] [65 , 66] ) cause cells to become more oxidized [59] . The ability of these signaling molecules to alter redox state is essential to their mechanisms of action , because pharmacological inhibition of the redox changes they induce blocks their effects on either division or differentiation of O-2A/OPCs . Thus , multiple lines of evidence have demonstrated that responsiveness to small changes in redox status represents a central physiological control point in these progenitor cells ( as summarized in Figure 1 ) . We initiated our studies of toxicant effects on O-2A/OPCs with an examination of MeHg , which has been previously studied for its effects on neuronal migration , differentiation , and survival , and on astrocyte function ( e . g . , [67–74] ) . Little is known about the effects of MeHg on the oligodendrocyte lineage , despite the fact that there are several reports over the past two decades documenting decreases in conduction velocity in the auditory brainstem response ( ABR ) of MeHg-exposed children [75–78] and rats [79] . Such a physiological alteration has long been considered to be indicative of myelination abnormalities in children whose development has been compromised by iron deficiency ( see , e . g . , [80 , 81] ) . We found that exposure of O-2A/OPCs ( growing in chemically defined medium supplemented with PDGF ) to environmentally relevant levels of MeHg makes these cells approximately 20% more oxidized ( Figure 2A ) , a degree of change similar to that previously associated with reductions in progenitor cell division [59] . Exposure to MeHg inhibited progenitor cell division as determined both by analysis of bromodeoxyuridine ( BrdU ) incorporation ( Figure 2B ) and by analysis of cell division in individual clones of O-2A/OPCs ( Figure 2C–2E ) . These oxidizing effects of MeHg were seen at exposure levels as low as 20 nM , less than the 5 . 8 μg/l or more ( i . e . , parts per billion [ppb] ) of MeHg found in cord blood specimens of as many as 600 , 000 infants in the US each year [46] and 0 . 3% or less of the exposure levels previously found to induce oxidative changes in astrocytes [4] . Exposure to 20 nM MeHg was sufficient to cause an approximately 25% drop in the percentage of O-2A/OPCs incorporating BrdU in response to stimulation with PDGF . When examined at the clonal level , MeHg exposure was associated with a reduction in the number of large clones and an increase in the number of small clones , as seen for other pro-oxidant stimuli [59] . Increasing MeHg exposure levels above 50 nM was associated with significant lethality , but little or no cell death was observed at the lower concentrations used in the present studies ( unpublished data ) . Thus , division of O-2A/OPCs exhibits a striking sensitivity to low concentrations of MeHg . One possible explanation for the reduced division associated with MeHg exposure would be disruption of PDGF-mediated signaling , and molecular analysis revealed that exposure of O-2A/OPCs to 30 nM MeHg for 24 h suppressed PDGF-induced signaling pathway activation at multiple points from the nucleus back to the receptor . One pathway stimulated by PDGF binding to the PDGF receptor-α ( PDGFRα ) leads to sequential activation of Raf-1 , Raf-kinase , and extracellular signal-regulated kinase 1 and 2 ( ERK1/2 ) , which further leads to activation of the Elk-1 transcription factor and up-regulation of immediate early-response gene expression , at least in part through activation of the serum response element ( SRE ) promoter sequence [82 , 83] . MeHg exposure was associated with reduced expression of an SRE-luciferase reporter gene ( Figure 3A ) , and reduced ERK1/2 phosphorylation ( Figure 3B ) . PDGFRα activation also stimulates activity of PI-3 kinase , leading to activation of Akt and induction of NF-κB–mediated transcription ( e . g . , [82 , 84 , 85] ) , both of which also were inhibited by MeHg exposure . Expression of an NF-κB-luciferase reporter gene was decreased ( Figure 3C ) , as was phosphorylation of Akt ( Figure 3D ) . Phosphorylation of PDGFRα , indicating receptor activation , was also reduced in cells exposed to MeHg ( Figure 3E ) . Because O-2A/OPCs growing in these cultures are absolutely dependent upon PDGF for continued division ( e . g . , [60 , 61 , 86] ) , the suppression of PDGF signaling would necessarily cause a reduction in cell division . We next found that the effects of MeHg were pathway specific and were associated with reductions in total levels of PDGFRα . O-2A/OPCs exposed to 30 nM MeHg exhibited no reduction in ERK1/2 phosphorylation induced by exposure to NT-3 ( Figure 4A ) , and no reduction in NT-3–induced expression from an SRE-luciferase reporter construct ( unpublished data ) . This result suggested that the site of action of MeHg was upstream of ERK1/2 regulation , prompting us to look directly at the PDGFRα . We found that the reduction in phosphorylated PDGFRα ( Figure 3E ) was paralleled by a reduction in levels of the PDGFRα itself ( Figure 4B ) . In contrast , no reduction in levels of TrkC ( the receptor for NT-3 [87] ) was caused by exposure to MeHg ( Figure 4C ) . One possible explanation for the ability of MeHg to cause a reduction in PDGF-mediated signaling and in total levels of PDGFRα , without affecting NT-3–mediated signaling or TrkC levels , would be that exposure to this toxicant leads to activation of c-Cbl , an E3 ubiquitin ligase that ubiquitylates the activated PDGFRα [88 , 89] , thus leading to its internalization and potential lysosomal degradation [90–92] . Such a possibility is particularly intriguing in light of multiple reports that c-Cbl can be activated by Fyn kinase ( e . g . , [93–96] ) , a Src family kinase that can be activated by oxidative stress [97–100] . O-2A/OPCs are known to express Fyn , which has been studied in these cells for its effects on regulation of RhoA activity and control of cytoskeletal organization [101 , 102] . Because TrkC does not appear to be regulated by c-Cbl , redox-modulated activation of Fyn , leading to c-Cbl activation and enhanced PDGFRα degradation , would provide a potential mechanistic explanation integrating the observations reported thus far . A variety of data support the hypothesis that MeHg exposure activates Fyn , leading to activation of c-Cbl , followed by degradation-mediated reductions in levels of activated PDGFRα . Exposure of O-2A/OPCs to 30 nM MeHg stimulated Fyn activation and c-Cbl phosphorylation ( Figure 5A and 5B ) . Activation of Fyn and c-Cbl was blocked by the Src family kinase inhibitors PP1 ( Figure 5A and 5B ) and PP2 ( unpublished data ) . We next found that exposure to MeHg enhanced ubiquitylation of PDGFRα ( a predicted consequence of c-Cbl activation ) , an increase readily observed even in the presence of markedly reduced levels of the receptor itself ( Figure 5C ) . Co-exposure to ammonium chloride ( NH4Cl , a lysosomotropic weak base that increases lysosomal pH and disrupts lysosomal protein degradation [103–105] ) prevented receptor degradation , and was associated with increased levels of ubiquitylated receptor in treated O-2A/OPCs . The increase in levels of ubiquitylated receptor was as predicted by the lack of effect of NH4Cl on either Fyn activation or c-Cbl phosphorylation ( Figure 5A and 5B ) . Treatment with PP1 , which inhibits Fyn activity ( Figure 5A ) , was also associated with a marked reduction in the amount of ubiquitylated PDGFRα , particularly in comparison with levels of total receptor ( compare upper and lower lanes in Figure 5C ) . As further confirmation that reductions in levels of PDGFRα were due to protein degradation , exposure to MeHg did not have any significant effects on levels of PDGFRα mRNA , as determined by quantitative PCR analysis ( Figure S1A ) . In the presence of cycloheximide , an inhibitor of protein synthesis , MeHg further accelerated receptor loss as compared with that occurring solely due to failure to synthesize new protein ( Figure S1B ) . Collectively , these results indicate that MeHg enhances active degradation of PDGFRα , as contrasted with reducing receptor levels as an indirect consequence of altering transcriptional or translational regulation of receptor levels . Molecular confirmation of the role of Fyn and c-Cbl in the effects of MeHg on levels of PDGFRα was obtained by expression of dominant negative c-Cbl , or small inhibitory RNA ( RNAi ) for Fyn or Cbl , in MeHg-exposed O-2A/OPCs . Expression of the dominant-negative ( DN ) 70z mutant of c-Cbl [106–108] in O-2A/OPCs prevented MeHg-induced reductions in levels of PDGFRα ( Figure 6A ) . Reduction in levels of Fyn protein by introduction of Fyn-specific small interfering RNA ( siRNA ) constructs ( Figure 6B ) also protected against MeHg-induced reductions in levels of PDGFRα ( Figure 6C ) , as predicted by the hypothesis that MeHg-induced activation of Fyn mechanistically precedes reductions in receptor levels . Similar results were obtained using RNAi constructs for c-Cbl , but are presented later in the paper , in the context of analysis of other toxicants . Suppression of Fyn or c-Cbl activity , or overexpression of PDGFRα itself , also protected against the functional effects of MeHg exposure ( Figure 7 ) . Pharmacological inhibition of Fyn activity with PP1 enabled analysis of O-2A/OPC division at the clonal level , and demonstrated that PP1 blocked MeHg-induced suppression of cell division ( Figure 7A ) . O-2A/OPCs expressing DN-70Z-c-Cbl and exposed to MeHg were also protected from effects of MeHg on cell division , as analyzed by BrdU incorporation ( Figure 7B ) . Co-treatment of MeHg-exposed O-2A/OPCs with PP1 or NH4Cl also blocked MeHg-associated suppression of ERK1/2 phosphorylation ( and MeHg-induced reductions in levels of PDGFRα , indicating that ERK1/2 suppression was a secondary consequence of the effects of Fyn and c-Cbl activation ( Figure 7C ) . Overexpression of PDGFRα in MeHg-exposed O-2A/OPCs also protected cells from MeHg-associated reductions in ERK1/2 phosphorylation ( Figure 7D ) . To determine whether effects of MeHg revealed a general mechanism by which chemically diverse toxicants with pro-oxidant activity could alter cellular function in similar ways , we next examined the effects of exposure of dividing O-2A/OPCs to Pb ( a heavy metal toxicant ) and paraquat ( an organic herbicide ) . As discussed in the Introduction , these toxicants both make cells more oxidized , but through mechanisms that differ between them and also from effects of MeHg . Despite their chemical differences from MeHg , and from each other , Pb and paraquat had apparently identical effects as MeHg on ERK1/2 phosphorylation , activation of Fyn and c-Cbl , and reductions in levels of phosphorylated PDGFRα and on total levels of PDGFRα ( Figure 8 ) . O-2A/OPCs were exposed to 1 μM Pb ( equivalent to the level of 20 μg/dl that is known to be associated with cognitive impairment , and a level of Pb previously found to inhibit O2A/OPC division without causing cell death [38 , 53 , 109] ) or to 5 μM paraquat ( an exposure level selected as being in the lowest 0 . 1% of the range of paraquat concentrations studied by others in vitro , which range from 8 μM–300 mM ( e . g . , [110–114] ) . Pb and paraquat exposure at these levels did not cause cell death , but did make O-2A/OPCs approximately 20% more oxidized , as determined by analysis of cells with the redox-indicator dyes dihydro-chloromethyl-rosamine or dihydro-calcein-AM ( unpublished data ) . Both Pb and paraquat exposure were associated with activation of Fyn ( Figure 8A ) , increased phosphorylation of c-Cbl ( Figure 8B ) , reduced levels of ERK1/2 phosphorylation , and reduced levels of phosphorylated and total PDGFRα ( Figure 8C ) . As for MeHg , the effects of Pb and paraquat on PDGFRα levels were prevented by expression of RNAi for c-Cbl ( Figure 8D ) , DN ( 70Z ) c-Cbl , or RNAi for Fyn ( unpublished data ) . It has previously been suggested that the effects of Pb on O-2A/OPCs are mediated through activation of PKC [38] , a pathway that has not been implicated in the activity of MeHg or paraquat . To determine whether PKC inhibition could distinguish between effects of Pb versus MeHg or paraquat , and to determine if PKC activation was relevant to the effects of toxicants on Fyn or c-Cbl activation or reductions in PDGFRα levels , we next examined the effects of co-exposure of O-2A/OPCs to bisindolylmaleimide I ( BIM-1 , a broad-spectrum PKC inhibitor previously used in the analysis of the role of PKC activation in the effects of Pb on O-2A/OPCs [38] ) . As shown in Figure S2 , we found that co-exposure of O-2A/OPCs to BIM-1 with Pb , MeHg , or paraquat did not prevent toxicant-mediated activation of Fyn ( Figure S2A ) or c-Cbl ( Figure S2B ) . BIM-1 co-exposure also did not protect against MeHg- , Pb- or paraquat-induced reductions in levels of PDGFRα ( Figure S2C ) . If it is correct that Fyn activation , with its consequences , is regulated by the ability of toxicants to make cells more oxidized , then antagonizing such redox changes should prevent Fyn activation . Previous studies have shown that an effective means of preventing the increase in oxidative status and the suppression of cell division caused by exposure of O-2A/OPCs to TH is to treat cells with N-acetyl-L-cysteine ( NAC ) , a cysteine pro-drug that is readily taken up by cells and converted to cysteine [59] . Cysteine is the rate-limiting precursor for synthesis of glutathione , one of the major regulators of intracellular redox status ( e . g . , [115 , 116] . NAC also possesses anti-oxidant activity , has long been used as a protector against many types of oxidative stress ( e . g . , [9 , 117 , 118] ) , and has been shown to confer protection against a wide range of toxicants , including MeHg ( e . g . , [119–121] ) , Pb ( e . g . , [9 , 122 , 123] ) , and paraquat ( e . g . , [17 , 124] ) , as well as such other substances as aluminum [125] , cadmium [126] , arsenic [127] , and cocaine [128] . As predicted by the hypothesis that the pro-oxidant activities of chemically diverse toxicants are causal in Fyn activation , NAC was equally effective at preventing Fyn activation—and its consequences—induced by exposure to MeHg , Pb , or paraquat ( Figures 2–5 , 7 , and 8 ) . For cells grown at the clonal level , NAC blocked the suppressive effects of MeHg on cell division ( Figure 2 ) . NAC also blocked all effects of MeHg on PDGF-mediated signaling , and rescued normal level of activity of SRE and NF-kB promoter-reporter constructs and levels of phosphorylation of ERK1/2 , Akt , and PDGFRα ( Figure 3 ) . Consistent with the hypothesis that Fyn is activated when cells become more oxidized [97–100] , NAC also blocked MeHg-induced activation of Fyn and phosphorylation of c-Cbl ( Figure 5 ) , and prevented MeHg-induced reductions in levels of PDGFRα ( Figure 4 ) . Critically , for the hypothesis that Pb and paraquat effects also were mediated by changes in redox state , NAC also blocked the effects of Pb and paraquat on Fyn activation and c-Cbl phosphorylation , and protected against effects of these toxicants on ERK1/2 phosphorylation and levels of PDGFR ( Figure 8 ) . Levels of PDGFRα were also protected by exposure of O-2A/OPCs to procysteine ( Figure 8 ) , a thiazolidine-derivative cysteine pro-drug that differs from NAC in having no intrinsic anti-oxidant activity [129] . Although it is conceivable that the ability of cysteine pro-drugs to protect against the effects of MeHg , Pb , and paraquat is due to enhanced toxicant clearance associated with elevated levels of glutathione , analysis of Pb uptake with Leadmium Green AM ( a fluorescent indicator of Pb levels ) showed no significant difference in Pb levels between cells exposed to Pb as compared with cells exposed to Pb and NAC ( Figure S3 ) . The ability of NAC to block toxicant-induced activation of Fyn raises the question of whether this is due to a true prevention of the effects of toxicant exposure on activation of this kinase or , alternatively , is due to an ability of NAC to independently suppress Fyn activity to such an extent that the apparent block of toxicant effects instead represents the summation of two opposing influences of equivalent magnitude . To evaluate these two possibilities , O-2A/OPCs were exposed to 1 mM NAC in the absence of toxicants , and Fyn and c-Cbl activation were evaluated as in Figure 5 . We found that NAC exposure had only a slight , and nonsignificant , effect on the levels of basal Fyn activity in O-2A/OPCs ( Figure 9A ) . In agreement with this outcome , NAC exposure did not have any marked effect on levels of c-Cbl phosphorylation ( Figure 9B ) . Thus , it appears that NAC-mediated counteraction of the effects of toxicants on Fyn activation is far greater in its magnitude than its direct effects on basal levels of Fyn activity . If the hypothesis is correct that exposure of O-2A/OPCs to toxicants causes activation of the Fyn/c-Cbl pathway , then other c-Cbl targets should be affected similarly to the PDGFRα . One member of the c-Cbl interactome [92] known to be expressed by O-2A/OPCs is c-Met [130] , the receptor for hepatocyte growth factor ( HGF; [131 , 132] ) . Oligodendrocytes also have recently been reported to be responsive to epidermal growth factor ( EGF ) application with morphological changes [133] , and microarray analysis confirms that the EGF receptor ( EGFR ) is expressed by O-2A/OPCs ( C . Pröschel and M . Noble , unpublished results ) . The EGFR is perhaps the most extensively studied RTK target of c-Cbl [90 , 96 , 107 , 134–137] , but c-Met regulation by c-Cbl appears to follow similar principles [106 , 138] . As shown in Figure 10 , exposure of O-2A/OPCs to MeHg was associated with reductions in levels of c-Met ( Figure 10A ) and EGFR ( Figure 10B ) . As predicted by the hypothesis that Pb and paraquat converge with MeHg on activation of the Fyn/c-Cbl pathway , levels of C-Met and EGFR were also reduced in O-2A/OPCs exposed to these additional toxicants . Consistent with the hypothesis that such changes were associated with the ability of toxicants to make cells more oxidized , NAC protected both c-Met and EGFR levels from reductions associated with exposure to MeHg , Pb , or paraquat . Further support for the Fyn/c-Cbl hypothesis of toxicant convergence was provided by observations that neither Pb or paraquat caused a reduction in levels of TrkC ( Figure 10C ) , just as observed for MeHg ( Figure 4C ) . Although the central goal of the present studies was the identification of mechanistic pathways on which chemically diverse toxicants converge , it is important to also consider whether any aspects of our in vitro findings are predictive of in vivo outcomes . Although detailed in vivo investigations will be the subject of future studies , we have tested three of the key findings of our present work for which previous studies are not predictive of likely experimental outcomes . The three questions we examined in vivo were whether toxicant exposure is associated with specific reductions in RTKs that are c-Cbl targets , whether this occurs at levels of toxicant exposure approximating the effects of environmental exposure , and whether such exposure can be shown to cause subtle changes in O-2A/OPC function . These experiments were conducted entirely with MeHg for several reasons . First , there is already extensive evidence that Pb exposure in vivo has adverse effects on myelination and on O-2A/OPCs ( e . g . , [38 , 43 , 53 , 139–142] ) . In contrast , evidence that MeHg exposure may have any effects on myelination thus far comes only from observations of increased latencies in ABRs [75–79] , with no studies examining effects of this toxicant on the function of cells important for myelination ( i . e . , oligodendrocytes or their ancestral O-2A/OPCs ) . Third , previous studies on mice have not been conducted using levels of exposure of broad environmental relevance . Instead , such studies have defined a low exposure range as being exposure of animals to MeHg in their drinking water at a concentration of one or more parts per million ( e . g . , [143–147] ) , an exposure level considerably higher than what our studies would predict as being necessary to affect progenitor cells of the developing CNS . Thus , the question of whether MeHg exposure levels of broader environmental relevance would have any effects at all in vivo appears to be largely unaddressed . To test the hypothesis that environmentally relevant levels of MeHg exposure can perturb the developing CNS in subtle ways , we exposed SJL mice to 100 or 250 ppb MeHg in their drinking water throughout gestation , and maintained this exposure until sacrifice of pups at 7 and 21 d after birth . As discussed in Materials and Methods , these exposure levels enabled us to approximate the predicted mercury levels in the CNS of 300 , 000–600 , 000 infants in the US . The exposure levels examined in our studies are 75%–90% below what has otherwise been considered to be low-dose exposure in mice . We found that developmental exposure of mice to MeHg at either 100 ppb or 250 ppb in the maternal drinking water was associated with clear and significant reductions in levels of PDGFRα and EGFR , but not of TrkC ( Figure 11 ) . Treatment of SJL mice with 100 or 250 ppb MeHg in the drinking water during gestation and suckling was associated with reductions in levels of PDGFRα and EGFR in the cerebellum , hippocampus , and corpus callosum when brain tissue was sampled at 7 and 21 d after birth . In contrast , levels of the NT-3 receptor TrkC were not reduced in these animals , as predicted by our in vitro analyses . It was particularly striking that exposure even to 100 ppb MeHg in the drinking water was enough to have significant effects on levels of PDGFRα and EGFR . These changes , and the lack of effect of MeHg exposure on TrkC levels , are as predicted from our in vitro analyses . Analysis of BrdU incorporation revealed that these low levels of MeHg exposure also were associated with statistically significant reductions in the division of O-2A/OPCs in vivo . In these experiments , postnatal day 14 ( P14 ) animals were treated as for analysis of receptor levels except that BrdU was administered 2 h before sacrifice . Sections then were analyzed with anti-BrdU antibodies to identify cells engaged in DNA synthesis and with antibodies to olig2 to identify O-2A/OPCs ( as in [148] ) . Olig2 is a transcriptional regulator expressed in oligodendrocytes and their ancestral precursor cells ( e . g . , [50 , 149–152] . In white matter tracts of the CNS , BrdU+ cells that express Olig2 are considered to be O-2A/OPCs [153 , 154] ) . In our studies , greater than 90% of all BrdU+ cells in the corpus callosum were also Olig2+ . When we analyzed the number of Olig2+/BrdU+ cells found in the corpus callosum of control and experimental animals ( see Materials and Methods for details of analysis ) , we found a 20% reduction in the number both of total BrdU+ cells and of Olig2+/BrdU+ cells ( Figure 11 ) , an outcome in agreement with the results of our in vitro studies ( Figure 2B ) .
Our studies demonstrate that chemically diverse toxicants converge on activation of a previously unrecognized pathway of cellular regulation that leads from increases in oxidative status to reductions in levels of specific RTKs . Analysis of effects of MeHg on O-2A/OPCs dividing in response to PDGF first demonstrated suppression of PDGF-induced signaling , but no reduction in NT-3–induced phosphorylation of ERK1/2 . Further analysis demonstrated that MeHg exposure enhanced degradation of PDGFRα as a consequence of the sequential activation of Fyn and c-Cbl . As predicted by the hypothesis that MeHg exposure activates the redox/Fyn/c-Cbl pathway , exposure to this toxicant was also associated with reductions in levels of EGFR and c-Met ( which are c-Cbl targets ) , but not in levels of TrkC ( which is not a c-Cbl target ) . The redox/Fyn/c-Cbl pathway was also activated by Pb and paraquat , leading to negative modulation of RTK-mediated signaling by regulating receptor degradation and causing reductions in levels of PDGFRα , EGFR , and c-Met , but not of TrkC . Developmental exposure to MeHg was also associated with reduced levels of PDGFRα and EGFR , but not of TrkC , consistent with the hypothesis that this same regulatory pathway is activated in association with in vivo toxicant exposure . The results of our studies are novel in a number of ways , beginning with the identification of a previously unrecognized regulatory pathway activated by chemically diverse toxicants . Although the importance of identifying general principles that apply to chemically diverse toxicants is a widely recognized goal of toxicology research , relatively few such principles have been identified . For example , although toxicants may be classified as hormonal mimetics , mutagens , carcinogens , neurotoxins , etc . , relatively few mechanistic pathways have been identified on which chemically diverse substances converge . Our present studies have identified Fyn activation as a common cellular target for the action of chemically diverse toxicants with pro-oxidant activity . Whether oxidative changes are by themselves sufficient to induce sequential activation of Fyn and c-Cbl will be a subject of continued analysis , but existing data make it difficult to imagine a compelling alternative hypothesis to explain our results . Fyn is well established as being activated when cells become more oxidized [97–100] , and there is no evidence for any other unifying feature of MeHg , Pb , and paraquat that would cause Fyn activation . Activation of Fyn , and the effects of activation of the Fyn/c-Cbl pathway , were blocked by NAC ( which antagonizes oxidative changes in O-2A/OPCs [59] ) as effectively as by expression of Fyn-specific RNAi constructs or by pharmacological inhibition of Fyn activity . NAC protects against physiological stress in two ways , both as an anti-oxidant itself and by providing increased levels of cysteine , the rate-limiting precursor in glutathione biosynthesis ( e . g . , [115 , 116] ) . The ability of ProCys ( which has no intrinsic anti-oxidant properties [129] ) to confer similar protection as NAC suggests that it is through their enhancement of glutathione production that these two cysteine pro-drugs exert their protective effects . The relatively small effect of NAC exposure by itself on basal Fyn activity in the experimental conditions used indicates that , at least in these experiments , NAC's protective effect was more likely to be due to protection against increases in oxidative status than due to a direct suppression of Fyn activity to an extent that would neutralize the activating effects of toxicant exposure . Although increased glutathione levels theoretically could also protect against the effects of toxicants by enabling enhanced cellular export of physiological stressors ( reviewed in , e . g . , [155 , 156] ) , analysis with Leadmium Green AM ( which can detect intracellular Pb in the nM range ) revealed no apparent effect of NAC treatment on cellular levels of Pb ( Figure S3 ) . Further support for the hypothesis that transport of xenobiotics is not a likely explanation for the protective effects of NAC is also provided by ongoing studies demonstrating that TH and BMP-4 ( both of which cause O-2A/OPCs to become more oxidized [59] ) also cause activation of Fyn and c-Cbl , with associated reductions in PDGFRα levels ( Z . Li and M . Noble , unpublished data ) . NAC blocks the effects of TH and BMP on differentiation , and also prevents TH- and BMP-mediated activation of Fyn and c-Cbl ( [59]; Z . Li and M . Noble , unpublished data ) . Changes in intracellular redox state , and the predicted ability to protect with NAC , are the common features linking the activation of Fyn with MeHg , Pb , paraquat , TH , and BMP . Although Fyn has multiple targets , it seems most likely that activation of c-Cbl provides the explanation for the effects of MeHg on PDGF-mediated signaling . Suppression of c-Cbl activity by expression of DN ( 70Z ) c-Cbl or RNAi protected against the effects of MeHg on cell division and reductions in levels of PDGFRα . Moreover , the induction of PDGFRα ubiquitylation by MeHg , the lack of effects of MeHg on PDGFRα mRNA levels , the rescue of receptor levels by disrupting lysosomal function , and other observations all strongly indicate the importance of c-Cbl regulation in understanding the effects of toxicant exposure . The importance of Fyn in activation of c-Cbl is supported by the ability of expression of Fyn-specific RNAi , or pharmacological inhibition of Fyn activity , to protect against the effects of toxicant exposure . Because Fyn activation in O-2A/OPCs also leads to activation of Rho-GTPase , leading to inhibition of Rho kinase activity [102 , 157] , we also examined the effects of treatment of cells with the Rho kinase inhibitor Y23762 ( Figure S4 ) . Although this agent inhibited Rho kinase activity in O-2A/OPCs , it neither protected against nor exacerbated the effects of MeHg on progenitor cell division ( as determined by BrdU incorporation ) . Thus , although it will be of interest to examine the effects of toxicant exposure on other Fyn targets , it currently seems that Fyn-mediated activation of c-Cbl is central to understanding the effects of toxicants on O-2A/OPCs . The discovery of sequential activation of Fyn and c-Cbl by pro-oxidants provides a new means of integrating the effects of changes in intracellular redox state with the control of the cell cycle . Although the ability of Fyn to be activated by increases in oxidative status [97–100] , the functional interaction of Fyn with c-Cbl ( e . g . , [93–96] ) , and the regulation of degradation of specific RTKs by c-Cbl ( e . g . , see [88–90 , 96 , 106 , 107 , 134–138] ) have all been subjects of study by multiple laboratories , our studies appear to provide the first integration of all of these components into a regulatory pathway of obvious relevance to the regulation of cell function by redox status . This regulatory pathway , summarized in Figure 12 , offers a number of clear predictions , some of which have been tested in our present studies . Several studies on different cell types have confirmed our own finding [59] that making dividing cells more oxidized can suppress division and induce differentiation [158–160] , and it will be of interest to determine the contribution of the redox/Fyn/c-Cbl pathway in these other cell systems , as well as in modulating other changes in cellular function that have been attributed to increased oxidative status ( e . g . , [73 , 161–166] ) . It is particularly striking that the changes we observed were seen at environmentally relevant exposure levels for both MeHg and Pb . As many as 600 , 000 newborn infants in the US each year have cord blood mercury levels greater than 5 . 8 ppb [46] ( i . e . , ∼30 nM ) . It is reported that the blood:brain ratio for humans may be as high as 1:5 to 1:6 . 7 [167 , 168] , therefore , in vivo levels in brain may be still higher than those we have studied . It is also noteworthy that levels of MeHg exposure at which selective reduction in PDGFRα expression was readily observed in vivo were 90% or more lower than exposure levels generally considered to constitute low-to-moderate exposure ( e . g . , [143–147] ) . Blood Pb levels may be of concern at levels as low as 10 μg/dl ( e . g . , [169–176] ) , which is equivalent to 0 . 48 μM , but which may be increased to micromolar in the brain by mechanisms relevant to Ca2+ transport [171] ) . Even given equivalence in blood:brain Pb levels , a concentration of 1 μM is equivalent to the approximately 20 μg/dl blood Pb levels known to be associated with cognitive impairment ( e . g . , [169–177] , an exposure level of particular concern in countries where leaded gasoline is still used and in which mean blood lead levels in schoolchildren may be as high as 15 μg/dl [178] . The study of environmentally relevant levels of toxicant exposure is a great challenge , both in vitro and in vivo , and it may be that analysis of stem and progenitor cell populations will be critical in furthering such analysis . In vitro , O-2A/OPCs appear to offer a particularly useful target cell for such studies , in part due to their sensitivity to environmentally relevant exposure levels of toxicants , but also due to the ability to use clonal analysis in quantitative studies on the cumulative effects of small changes in the balance between division and differentiation [179–181] . Such studies have shown that even such potent physiological regulators as TH may only increase the probability of oligodendrocyte differentiation at each progenitor cell cycle from approximately 0 . 5 to 0 . 65 [179] . Thus , although their cumulative effects over time may be readily observable , analysis of subtle effects in acute assays may fail to identify important alterations in progenitor cell function . In addition , it will be important to extend analysis on differentiation to other precursor cell populations , as indicated by recent observations that neuronal differentiation of neuroepithelial stem cells may be compromised by MeHg exposure levels as low as 2 . 5 nM [182] . In vivo , the 20% reduction in number of dividing O-2A/OPCs observed in animals exposed to 100 ppb MeHg during development was of particular interest , as such relatively subtle changes might be predicted to reduce myelination in ways that require equally subtle analysis to detect functional outcomes . Analysis of conduction velocity in the auditory system may offer one such analytical tool , and the sensitivity of O-2A/OPCs to toxicant exposure may provide an explanation for the consistency with which increases in ABR latency suggestive of myelination abnormalities are associated with exposure to a variety of toxicants and physiological stressors , including MeHg [75–79] , Pb [183 , 184] ) , cocaine [185 , 186] , and carbamazepine [187] . The general importance of the signaling pathways regulated by Fyn and c-Cbl suggests that the ability of chemically diverse toxicants to converge on this pathway may be of broad relevance to the understanding of toxicant action . Such c-Cbl targets as PDGFRα , EGFR , and c-Met play critical roles in processes as diverse as cell proliferation , survival , and differentiation , cortical neurogenesis , maintenance of the subventricular zone , astrocyte development , development of cortical pyramidal dendrites , motoneuron survival and pathfinding , sympathetic neuroblast survival , and hippocampal neuron neurite outgrowth , as well as having extensive effects on development of kidney , lung , breast , and other tissues ( e . g . , [60 , 61 , 130 , 188–195] ) . Indeed , the range of targets of c-Cbl [92 , 135] offers a rich fabric of potentially critical regulatory molecules that would be affected by changes in activity of this protein , with the importance of particular proteins being dependent on the cell type and developmental stage under consideration . In addition , Fyn regulation of the Rho/ROCK signaling pathway could be of relevance in understanding toxicant-mediated alterations on such cytoskeletal functions as cell migration , neurite outgrowth , and development of dendritic morphology ( e . g . , [196–198] ) . Our studies predict that any toxicant that makes cells and/or tissues more oxidized would activate Fyn , a list that includes substances as chemically diverse as MeHg ( e . g . , [1–6] , Pb [6–9] , and organotin compounds [1 , 2 , 5 , 10 , 11] ) , cadmium [12 , 13] , arsenic [12 , 14] , ethanol [15 , 16] , and various herbicides ( e . g . , paraquat [17 , 18] , pyrethroids [19–21] , and organophosphate and carbamate inhibitors of cholinesterase [22–26] ) . In summary , our studies provide a new general principle and evidence of a new regulatory pathway that may be relevant to the understanding of the action of a large number of chemically diverse toxicants and other modulators of oxidative status . Because the outcomes we have identified occur at quite low toxicant exposure levels , they may provide a particularly useful unifying principle for the analysis of toxicant effects . Our present studies , combined with our previous analysis of the central importance of intracellular redox state in modulating progenitor cell function [59] , lead to the prediction that any toxicant with pro-oxidant activity will exhibit these effects . Although toxicants of differing chemical structures will also have additional activities , the convergence of small increases in oxidative status on regulation of the redox/Fyn/c-Cbl pathway provides a specific means by which exposure to low levels of a wide range of chemically diverse toxicants might have similar classes of effects on development . Our findings also provide a strategy for rapid identification of such effects by any of the estimated 80 , 000 to 150 , 000 chemicals for which toxicological information is limited or nonexistent , thus enabling a preliminary identification of compounds that would need to be examined in vivo . The sensitivity of O-2A/OPCs to environmentally relevant levels of MeHg and Pb provides a great advantage over established cell lines and other such neural cells as astrocytes , for which these low exposure levels may have little effect , and the importance of understanding the effects of toxicants on progenitor cell function provides a direct link between our studies and the broad field of developmental toxicology . In addition , the ability of NAC to protect progenitor cells against the adverse effects of chemically diverse toxicants raises the possibility that this benign therapeutic agent may be of benefit in protecting children known to be at increased risk from the effects of toxicant exposure during critical developmental periods . Finally , the principles indicated by our findings appear likely to have broad applicability in understanding the regulation of cell function by alterations in redox balance , regardless of how they might be generated .
O-2A/OPCs were purified from corpus callosum of P7 CD rats as described previously to remove type 1 astrocytes , leptomeningeal cells , and oligodendrocytes [59 , 64 , 199] . Cells were then grown in DMEM/F12 supplemented with 1-μg/ml bovine pancreas insulin ( Sigma , St . Louis , Missouri , United States ) , 100-μg/ml human transferrin ( Sigma ) , 2 mM glutamine , 25-μg/ml gentamicin , 0 . 0286% ( v/v ) BSA pathocyte ( ICN Biochemicals , Costa Mesa , California , United States ) , 0 . 2 μM progesterone ( Sigma ) , 0 . 10 μM putrescine ( Sigma ) , bFGF-2 ( 10 ng/ml; PEPRO Technologies , London , United Kingdom ) , and PDGF-AA ( 10 ng/ml; PEPRO ) onto poly-l-lysine ( Sigma ) coated flasks or dishes . Under these conditions , O-2A/OPCs derived from the corpus callosum of P7 rats are predominantly in cell division and do not generate large numbers of oligodendrocytes during the time periods utilized in this analysis . To generate sufficient numbers of cells for biochemical analysis , cells were expanded through one to two passages in PDGF + FGF-2 before replating in the presence of PDGF alone . When cells achieved approximately 50% confluence , MeHg , Pb , or paraquat was added to their medium at concentrations indicated in the text . Doses for the toxicants were chosen on the basis of dose-response curves to identify sublethal exposure levels ( unpublished data ) , as a reflection of blood and brain toxicant levels of these compounds and , where applicable , on the basis of previous reports . All toxicant concentrations examined were confirmed to cause death of less than 5% of cells over the time course of the experiment . For analysis of the effects of potential inhibitors of toxicant action , cells were exposed to the blocking compound of interest 1 h before addition of toxicant . The concentrations of inhibitors used are listed as following: 0 . 5 μM BIM-1 ( PKC inhibitor ) , 0 . 5 μM PP1/PP2 ( Src family kinase inhibitors ) , and 10 mM NH4Cl ( lysosome inhibitor ) ; and the concentrations of toxicants used are listed as following , except when mentioned specifically: MeHg ( 20 nM ) , Pb ( 1 μM ) , and paraquat ( 5 μM ) . To examine the degradation of PDGFRα , O-2A/OPCs were treated with MeHg ( 20 nM ) for different durations with or without cycloheximide ( CHX; 1 μg/ml ) added 1 h before MeHg . The cells were then collected and lysed for Western blotting . For example , in the multi-toxicant analysis of Figures 8–10 , for analysis of PDGFRα , O-2A/OPCs were exposed for 24 h to MeHg ( 20 nM ) , Pb ( 1 μM ) , and paraquat ( 5 μM ) for 24 h in the presence of 0 . 5 μM bisindolylmaleimide 1 ( BIM-1 ) , 0 . 5 μM of PP1 , 1 mM NAC , or 1 mM procysteine , which had been added 1 h prior to toxicant addition . Cells were lysed for Western blot analysis using anti-PDGFRα ( pY742 ) antibody . The membranes were de-probed and then re-probed with antibody against total PDGFRα and anti–β-tubulin antibody . For analysis of Fyn activity and c-Cbl phosphorylation , progenitors were exposed to MeHg ( 20 nM ) , Pb ( 1 μM ) , and paraquat ( 5 μM ) for 3–4 h in the presence of 0 . 5 μM BIM1 , 0 . 5 μM PP1 , or 1 mM NAC ( each of which was added 1 h before addition of toxicant ) . Cells were deprived of PDGF-AA for 5 h before re-exposure to PDGF-AA ( 10 ng/ml ) for 1 h for Western blot or 6 h for luciferase assays of pathway activation . Transient transfection was performed using Fugene6 ( Roche , Basel , Switzerland ) transfection solution according to the manufacturer's protocol . For the luciferase assay , cells seeded in 12-well plates were transfected with a reporter plasmid SRE-Luc ( firefly ) or NFκB-Luc ( firefly ) ( BD-Clontech , Palo Alto , California , United States ) and an internal control plasmid pRLSV40-LUC . Analyses of luciferase activity were performed according to the protocol of the Dual Luciferase Assay System ( Promega , Madison , Wisconsin , United States ) , which uses an internal control of Renilla luciferase for quantification , and relative light units were measured using a luminometer . Anti-phosphorylated ERK monoclonal , anti-ERK monoclonal , anti-TrkC polyclonal , anti-Fyn polyclonal , anti-EGFR polyclonal , anti–c-Met polyclonal , anti–phospho-tyrosine monoclonal , and anti-PDGFRα polyclonal antibodies were obtained from Santa Cruz Biotechnology ( Santa Cruz , California , United States ) . Anti–c-Cbl monoclonal antibody was obtained from BD PharMingen ( San Diego , California , United States ) . Anti-phosphorylated Akt monoclonal and anti-Akt polyclonal antibodies were obtained from Cell Signaling Technology ( Beverly , Massachusetts , United States ) . Anti-phosphorylated PDGFRα polyclonal antibody was obtained from Biosource ( Carlsbad , California , United States ) . The cell culture samples were collected and lysed in RAPI buffer , whereas dissected tissue samples were sonicated in RAPI buffer . Samples were resolved on SDS-PAGE gels and transferred to PVDF membranes ( PerkinElmer Life Science , Wellesley , Massachusetts , United States ) . After being blocked in 5% skim milk in PBS containing 0 . 1% Tween 20 , membranes were incubated with a primary antibody , followed by incubation with an HRP-conjugated secondary antibody ( Santa Cruz Biotechnology ) . Membranes were visualized using Western Blotting Luminol Reagent ( Santa Cruz Biotechnology ) . All analyses of signaling pathway components were conducted in the presence of ligand for the receptor pathway under analysis ( either PDGF-AA for PDGFRα , NT-3 or TrkC , HGF for c-Met , or EGF for EGFR ) . Cell proliferation was assessed by bromodeoxyuridine ( BrdU ) incorporation and by using the mouse anti-BrdU mAb IgG1 ( 1:100; Sigma ) to label dividing cells . Stained cells on coverslips were rinsed two times in 1× PBS , counterstained with 4′6-diamidino-2-phenylindole ( DAPI; Molecular Probes , Eugene , Oregon , United States ) and mounted on glass slides with Fluoromount ( Molecular Probes ) . Staining against surface proteins was performed on cultures of living cells or on cells fixed with 2% paraformaldehyde . Staining with intracellular antibodies was performed by permeabilizing cells with ice-cold methanol for 4 min or by using 0 . 5% Triton for 15 min on 2% paraformaldehyde–fixed cells . Antibody binding was detected with appropriate fluorescent dye–conjugated secondary antibodies at 10 μg/ml ( Southern Biotech , Birmingham , Alabama , United States ) or Alexa Fluor–coupled antibodies at a concentration of 1 μg/ml ( Molecular Probes ) , applied for 20 min . Anti-BrdU monoclonal antibody was obtained from Sigma . Cells were plated in 96-well microplates and grown to about 60% confluence . Prior to treatment , cells were washed twice with Hank's buffered saline solution ( HBSS ) , loaded with 20 μM H2DCFDA ( in HBSS 100 μl/well ) , and incubated at 37 °C for 30 min . Cells were then washed once with HBSS and growth medium to remove free probe . Then , fresh growth medium was added and a baseline fluorescence reading was taken prior to treatment . For NAC pre-treatment , NAC was added into media 1 h before further addition of MeHg , and both compounds remained in the medium during the incubation period with H2DCFDA . Fluorescence was measured in a Wallac 1420 Victor2 multilabel counter ( PerkinElmer ) using excitation and emission wavelengths of 485 nm and 535 nm , respectively , at different time courses as indicated in the figures . Results are presented as the value change from baseline by the formula ( Ftexp − Ftbase ) /Ftbase normalized with the control group , where Ftexp = fluorescence at any given time during the experiment in a give well and Ftbase = baseline fluorescence of the same well . We further determined whether pre-treatment with NAC altered levels of intracellular Pb by analysis with the Leadmium Green AM dye ( Molecular Probes ) , according to the manufacturer's instructions . In five separate experiments , we found no significant difference between O-2A/OPCs treated with 1 μM Pb versus [Pb + NAC] ( unpaired t-test ) , and the values for both Pb-treated samples were several-fold higher than control values . All of these data strongly support the hypothesis that the major effect of NAC is to antagonize cellular oxidation . For the co-immunoprecipitation assay , anti-c-Cbl monoclonal antibody ( BD PharMingen ) or anti-PDGFRα polyclonal antibody ( Santa Cruz Biotechnology ) was added to the pre-cleared cell lysates ( 250 μg of total protein ) , and the mixtures were gently rocked for 2 h at 4 °C . A total of 30 μl of protein A/G agarose was then added to the mixture followed by rotating at 4 °C overnight . The protein A/G agarose was then spun down and washed thoroughly three times . The precipitates were resolved on an 8% SDS-PAGE gel and then were subjected to Western blot analysis using an anti-p-Tyr ( for c-Cbl phosphorylation assay ) or ubiquitin ( for PDGFR ubiquitination assay ) antibody ( Santa Cruz Biotechnology ) . Fyn kinase activity was quantified using the Universal Tyrosine Kinase Assay Kit ( Takara , Madison , Wisconsin , United States ) . O-2A/OPCs exposed to different treatments were solubilized with an equal volume of the extraction buffer provided with the kit for 15 min , and the resulting lysates were centrifuged at 13 , 000 × g for 15 min at 4 °C; 250 μg of total cell lysates were immunoprecipitated with anti-Fyn antibody ( Santa Cruz Biotechnology ) . Following immunoprecipitation , Fyn immune complexes were washed four times with extraction buffer , and then Fyn kinase activities of each sample were assayed using the kit according to the manufacturer's instructions . Rho kinase activity was quantified using the CycLex Rho-Kinase Assay kit ( MBL International , Woburn , Massachusetts , United States ) as described . Cells were lysed and about 500 μg of total cell lysates were immunoprecipitated with anti-ROCK1 antibody ( Sigma ) , and the precipitates were re-suspended with kinase reaction buffer provided in the kit . Rho kinase activities of each sample were assayed using the kit according to the manufacturer's instructions . siRNA target sites were selected by scanning the cDNA sequence for AA dinucleotides via siRNA target finder ( Ambion , Austin , Texas , United States ) . Those 19-nucleotide segments that start with G immediately downstream of AA were recorded and then analyzed by BLAST search to eliminate any sequences with significant similarity to other genes . The siRNA inserts , containing selected 19-nucleotide coding sequences followed by a 9-nucleotide spacer and an inverted repeat of the coding sequences plus 6 Ts , were made to double-stranded DNAs with ApaI and EcoRI sites by primer extension , and then subcloned into plasmid pMSCV/U6 at the ApaI/EcoRI site . The corresponding oligonucleotides for the Fyn and c-Cbl RNAi's are listed in Table 1 . Several nonfunctional siRNAs , which contain the scrambled nucleotide substitutions at the 19-nucleotide targeting sequence of the corresponding RNAi sequence , were constructed as negative controls . All of these plasmids were confirmed by complete sequencing . pJEN/neo-HA-70z-c-Cbl plasmids were generously provided by Dr . Wallace Langdon . The pBabe ( puro ) -HA-70z-c-Cbl plasmids were constructed by transferring the BamH1-digested HA-70z-c-Cbl from pJEN/neo-HA-70z-c-Cbl into the BamH1 digested pBabe ( puro ) vector . The pBabe ( puro ) -HA-70z-c-Cbl , pMSCV/U6-Fyn-RNAi , pMCV/U6-c-Cbl-RNAi , and the corresponding scrambled RNAi plasmids and the empty plasmids were transfected into Pheonix Ampho cells by Fugene6 ( Roche ) transfection solution according to the manufacturer's protocol . Twenty-four hours after transfection , medium was changed to DMEM/F12 ( SATO , but with no TH ) supplemented with 10-ng/ml PDGF-AA and bFGF . Virus supernatant was collected 48 h post-transfection , filtered through 0 . 45-μm filter to remove non-adherent cells and cellular debris , frozen in small aliquots on dry ice , and stored at −80 °C . Twenty-four hours prior to infection , O-2AOPCs were seeded . The following day , the culture medium was aspirated and replaced with virus supernatant diluted 1:1 in the O-2A growth media . Medium was then changed into O-2A/OPC growth medium after 8 h or overnight . Twenty-four hours after infection , the cells were collected by trypsinization and reseeded in the selective medium ( growth medium + 200-ng/ml puromycin ) . By the next day , all noninfected cells were floating and presumably dead or dying . The infected cells were allowed to proliferate for 2 d , and then collected and re-seeded for the following experiments . Total RNA was isolated using TRIZOL reagent ( Invitrogen , Carlsbad , California , United States ) according to the manufacturer's protocol . A total of 1 μg of RNA was subjected to reverse transcription using Superscript II ( Invitrogen ) . The reactions were incubated at 42 °C for 50 min . The FAM-labeled probe mixes for rat PDGFRα and Fyn , and the VIC-labeled GAPDH probe mix were purchased from Applied Biosystems ( Foster City , California , United States ) . For multiplex real-time PCR , reactions each containing 5 μl of 10-fold–diluted reverse transcription product , 1 μl of interest gene probe mix , 1 μl of GAPDH probe mix , and 10 μl of TaqMan Universal PCR Master Mix were performed on an iCycler iQ multicolor real-time PCR system ( Bio-Rad , Hercules , California , United States ) and cycling condition was 50 °C for 2 min and 95 °C for 10 min , followed by 40 cycles of 95 °C for 15 sec and 60 °C for 1 min . Each sample was run in triplicate . Data were analyzed by iCycler iQ software ( Bio-Rad ) . O-2A/OPCs purified from P7 rat optic nerve were plated in poly-L-lysine–coated 25-cm2 flasks at clonal density with DMEM medium in the presence of 10-ng/ml PDGF as previously described [59 , 64 , 199] . After 24-h recovery , cells were treated with different toxicants , each for 3 d , until visual inspection and immunostaining was performed . NAC was added 1 h before exposure to other toxicants for NAC pretreatment , and NAC co-exists throughout the culture period . The numbers of O-2A/OPCs and oligodendrocytes in each clone were determined by counting under fluorescent microscope . The three-dimensional graph shows the number of clones containing O-2A/OPC cells and oligodendrocytes . Experiments were performed in triplicate in at least two independent experiments . Six-week-old female SJL mice were treated with MeHg in their drinking water at a concentration of 100 or 250 ppb for 30–60 d prior to mating , and then throughout pregnancy and gestation . This is a level of treatment that is 75%–90% below levels generally considered to be low to moderate and is below levels that have been associated with gross defects in adult or developing animals ( e . g . , [143–147] ) . The exposure levels used in our studies were first determined as candidate exposures from the results of two different previous studies on the relationship between MeHg exposure and levels of toxicant in the brain . Studies by Weiss and colleagues [143] demonstrated that mice exposed to MeHg in their drinking water for up to 14 mo have brain mercury levels roughly equivalent to that in the water . In these studies , mice exposed to MeHg in their drinking water from conception at a concentration of one part per million ( ppm ) had brain levels of MeHg of 1 . 20 mg/kg ( i . e . , ppm ) at 14 mo of age , whereas those exposed to MeHg at a concentration of 3 ppm had brain levels of 3 . 66 mg/kg at this age . It has also been shown , however , that mercury levels in the brain of pre-weanling animals exposed to MeHg via the mother's drinking water throughout gestation and suckling drop rapidly to one-fifth of the levels found at birth , presumably due to reduced MeHg transfer in milk [200] . As an estimated 300 , 000 to 600 , 000 infants in the US have blood cord mercury levels of 5 . 8 μg/l or more [46] , and because the human brain concentrates MeHg 5- to 6 . 7-fold over the concentration occurring in the bloodstream , our goal was to achieve postnatal brain mercury levels of 30 ppb ( i . e . , ng/g ) or less . In practice , we found that exposure of female mice to MeHg in their drinking water at a concentration of 250 ppb prior to conception , and maintenance of this exposure during suckling , was associated with brain mercury levels in the offspring ( examined at P14 ) of 50 ng/g , a fall that was precisely in agreement with predictions based on prior studies on the fall of mercury levels occurring during this period in suckling mice [200] . In offspring of dams exposed to MeHg at a concentration of 100 ppb in the drinking water , brain mercury was below the levels of detection of the Mercury Analytical Laboratory of the University of Rochester Medical Center . The exposure levels of 100 and 250 ppb are 75%–90% below what has otherwise been considered to be low-dose exposure in mice . At the time of sacrifice , mice were anesthetized using Avertin ( tribromoethanol , 250 mg/kg , 1 . 2% solution; Sigma ) and were perfused transcardially with 4% paraformaldehyde in phosphate buffer ( pH 7 . 4 ) following the removal of the blood by saline solution washing . The brains were removed and stored in 4% paraformaldehyde for 1 d , and then changed to 25% sucrose in 0 . 1 M phosphate buffer . Brains were cut coronally as 40-μm sections with a sliding microtome ( SM/2000R; Leica , Heidelberg , Germany ) and stored at −20 °C in cryoprotectant solution ( glycerol , ethylene glycol , and 0 . 1 M phosphate buffer[ pH 7 . 4] , 3:3:4 by volume ) . All animal experiments were conducted in accordance with National Institutes of Health guidelines for the humane use of animals . To analyze DNA synthesis in vivo , mice were injected with a single dose of 5-BrdU ( 50 mg/kg body weight ) , dissolved in 0 . 9% NaCl , filtered ( 0 . 2 μm ) , and applied intraperitoneally 2 h prior to perfusion . After removal and sectioning of brains , 40-μm free-floating sections were incubated for 2 h in 50% formamide/2× SSC ( 0 . 3 M NaCl and 0 . 03 M sodium citrate ) at 65 °C , rinsed twice for 5 min each in 2× SSC , incubated for 30 min in 2N HCl at 37 °C , and rinsed for 10 min in 0 . 1 M boric acid ( pH 8 . 5 ) at room temperature . Several rinses in TBS were followed by incubation in TBS/0 . 1% Triton X-100/3% donkey serum ( TBS-plus ) for 30 min . Sections were then incubated with monoclonal rat anti-BrdU antibody ( 1:2 , 500; Harlan Sera-Lab , Loughborough , United Kingdom ) and polyclonal rabbit anti-Olig2 ( a generous gift from Dr . David H . Rowitch ) in TBS-plus for 48 h at 4 °C . Sections were rinsed several times in TBS-plus and incubated for 1 h with donkey anti-rat FITC and donkey anti-rabbit TRITC ( Jackson ImmunoResearch Laboratories , West Grove , Pennsylvania , United States ) . After several washes in TBS , sections were mounted on gelatin-coated glass slides using Fluoromount-G mounting solution ( Southern Biotech ) . Quantification of BrdU+ cells was accomplished with unbiased counting methods by confocal microscopy . BrdU immunoreactive nuclei were counted in one focal plane to avoid oversampling . In corpus callosum , BrdU+ cells were counted in every sixth section ( 40 μm ) from a coronal series between interaural AP + 5 . 2 mm and AP + 3 . 0 mm in the entire extension of the rostral and medial part of the corpus callosum . Quantitative data are presented as mean percentage normalized to control animals . Error bars represent ± the standard error of the mean . Digital images were captured using a confocal laser scanning microscope ( Leica TCS SP2 ) . Photomicrographs were processed on a Macintosh G4 and assembled with Adobe Photoshop 7 . 0 ( Adobe Systems , Mountain View , California , United States ) . Unpaired , two-tailed Student t-test was used for statistical analysis .
|
Discovering general principles underlying the effects of toxicant exposure on biological systems is one of the central challenges of toxicological research . We have discovered a previously unrecognized regulatory pathway on which chemically diverse toxicants converge , at environmentally relevant exposure levels , to disrupt the function of progenitor cells of the developing central nervous system . We found that the ability of low levels of methylmercury , lead , and paraquat to make progenitor cells more oxidized causes activation of an enzyme called Fyn kinase . Activated Fyn then activates another enzyme ( c-Cbl ) that modifies specific proteins—receptors that are required for cell division and survival—to initiate the proteins' degradation . By enhancing degradation of these receptors , their downstream signaling functions are repressed . Analysis of developmental exposure to methylmercury provided evidence that this same pathway is activated in vivo by environmentally relevant toxicant levels . The remarkable sensitivity of progenitor cells to low levels of toxicant exposure , and the discovery of the redox/Fyn/c-Cbl pathway as a mechanism by which small increases in oxidative status can markedly alter cell function , provide a novel and specific means by which exposure to chemically diverse toxicants might perturb normal development . In addition , the principles revealed in our studies appear likely to have broad applicability in understanding the regulation of cell function by alterations in redox balance , regardless of how they might be generated .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"developmental",
"biology",
"public",
"health",
"and",
"epidemiology",
"cell",
"biology",
"in",
"vitro",
"mus",
"(mouse)",
"neuroscience",
"molecular",
"biology"
] |
2007
|
Chemically Diverse Toxicants Converge on Fyn and c-Cbl to Disrupt Precursor Cell Function
|
Hepatitis C virus ( HCV ) is a single-stranded RNA virus that replicates on endoplasmic reticulum-derived membranes . HCV particle assembly is dependent on the association of core protein with cellular lipid droplets ( LDs ) . However , it remains uncertain whether HCV assembly occurs at the LD membrane itself or at closely associated ER membranes . Furthermore , it is not known how the HCV replication complex and progeny genomes physically associate with the presumed sites of virion assembly at or near LDs . Using an unbiased proteomic strategy , we have found that Rab18 interacts with the HCV nonstructural protein NS5A . Rab18 associates with LDs and is believed to promote physical interaction between LDs and ER membranes . Active ( GTP-bound ) forms of Rab18 bind more strongly to NS5A than a constitutively GDP-bound mutant . NS5A colocalizes with Rab18-positive LDs in HCV-infected cells , and Rab18 appears to promote the physical association of NS5A and other replicase components with LDs . Modulation of Rab18 affects genome replication and possibly also the production of infectious virions . Our results support a model in which specific interactions between viral and cellular proteins may promote the physical interaction between membranous HCV replication foci and lipid droplets .
Hepatitis C virus ( HCV ) is a positive-sense RNA virus in the family Flaviviridae that is estimated to chronically infect up to 170 million people worldwide . The 9 . 6 kb genome encodes three structural and seven nonstructural proteins . One of these nonstructural proteins , NS5A , is an RNA-binding phosphoprotein essential for both viral replication and viral particle assembly [1] . It is composed of a N-terminal amphipathic helix that mediates membrane association [2]–[4] followed by three domains separated by two low-complexity sequences [5] . Domain I is responsible for NS5A dimerization [6] and has been proposed to contribute to RNA binding [7] , [8] . A role for this domain in HCV RNA replication has been supported by the finding that many adaptive mutations that enhance HCV replication in cell culture map to Domain I [9] , [10] , In contrast , the majority of Domain II and the entirely of Domain III are dispensable for RNA replication , while deletion of Domain III virtually abolishes viral particle assembly [1] . How NS5A supports both viral RNA replication and particle assembly remains incompletely understood . NS5A has been reported to interact with the core [11] , NS2 [12] , [13] , and NS5B [14] , [15] viral proteins . Furthermore , it has been reported to interact with many host proteins , such as PI4KA [16]–[18] , VAP-A [19] , VAP-B [20] , and FKBP8 [21] . As NS5A is believed to lack intrinsic enzymatic activity , the main function of NS5A may be to coordinate interactions among viral and host proteins . Indirect evidence suggests that NS5A function might be regulated by its intracellular localization . Only a small fraction of HCV nonstructural proteins appears to be associated with HCV replication complexes in protease-resistant membranes [22] , [23] , raising the possibility that some of the HCV nonstructural proteins have additional functions in the cell outside of the HCV replication complex . In particular , HCV NS5A has been reported to localize to ER and lipid droplet ( LD ) membranes [2] , [24]–[26] . Interestingly , a small-molecule NS5A inhibitor induces NS5A redistribution from ER to LDs [27] , suggesting that the intracellular trafficking of NS5A might be regulated . Lipid droplets are lipid storage organelles composed of a core of neutral lipids , sterols , and sterol esters surrounded by a phospholipid monolayer . In addition to NS5A , HCV core protein localizes to LDs [28] . The core-LD association is thought to be essential for virion assembly [26] , [29] , [30]; however , it remains uncertain whether virion assembly actually occurs on the LD membrane itself or on membranes closely associated with LDs . In addition , it is not known how viral RNA genomes synthesized at replicase complexes are transferred to sites of particle assembly . Recent work has demonstrated that LDs , rather than being simple lipid storage depots , are dynamic , motile organelles that interact with other intracellular membranes such as the ER and possibly also mitochondria [31] . LD functions and interactions with intracellular membranes may be regulated by proteins that specifically associate with the LD surface . For example , the Rab18 small GTPase localizes to LDs , and Rab18-LD association is enhanced by agents that stimulate lipolysis [32] . Furthermore , Rab18 overexpression has been shown to promote the association of LDs with ER membranes [33] . Using a proteomic strategy to identify novel host proteins that interact with HCV NS5A in HCV-infected cells , we have found that NS5A associates with Rab18 . Furthermore , Rab18 modulation affects NS5A localization to LDs . Rab18 also appears to regulate the association of NS5A-positive membranes and other HCV replicase components with lipid droplets , suggesting a model in which binding of Rab18 to NS5A physically recruits sites of HCV replication to LDs .
NS5A , which lacks known intrinsic enzymatic activity , is thought to exert its functions through interactions with host and viral proteins . We therefore sought to identify novel host protein-NS5A interactions , which in turn should yield new insights into the HCV life cycle . As depicted in Figure 1A and described in previously published work [17] , we inserted a small tandem affinity purification tag consisting of a tandem Strep-tag II and a FLAG tag into domain III of NS5A at a site previously shown to be tolerant of heterologous insertions such as GFP [34] . The chimeric genotype 2a Jc1 ( J6/JFH1 ) genome , which is fully infectious in cell culture , was used as the backbone [35] , [36] . This tag was genetically stable and following cell culture adaptation , titers of the tagged Jc1 ( SF ) virus were comparable to those of wild-type Jc1 [17] . We initially attempted to perform tandem affinity purification of tagged NS5A ( SF ) from Jc1 ( SF ) -infected Huh7 . 5 . 1 hepatoma cells using sequential anti-FLAG and Streptactin-Sepharose affinity purification steps [37] . However , the yield of NS5A ( SF ) following anti-FLAG affinity purification was insufficient for downstream mass spectrometry analysis ( data not shown ) , and therefore we used single-step Streptactin-Sepharose affinity purification of NS5A ( SF ) . In pilot experiments , we found that Streptactin-Sepharose failed to bind to NS5A ( SF ) in cellular homogenates of Jc1 ( SF ) -infected cells prepared in the absence of detergent , but did bind to NS5A ( SF ) in detergent lysates ( data not shown ) . Figure 1B demonstrates that a number of proteins specifically copurified with NS5A ( SF ) from Jc1 ( SF ) -infected cells ( left lane ) , but not from wild-type Jc1-infected cells ( right lane ) . On the other hand , we did observe nonspecific interactions with Streptactin-Sepharose from untagged Jc1-infected cells , which was not unexpected from single-step affinity purification . To minimize the false-positive identification of NS5A interacting proteins in subsequent mass spectrometry , we added Stable Isotope Labeling with Amino Acids in Cell Culture ( SILAC; Figure 1C ) [38] . Huh7 . 5 . 1 cells were either cultured in medium with normal “light” amino acids ( LAA ) or with medium containing only “heavy” 13C6 , 15N4 L-arginine and 13C6 , 15N2 L-lysine ( HAA ) , resulting in 97% labeling of the cellular proteome with heavy L-arg and L-lys in pilot studies ( data not shown ) . Cells labeled with LAA were infected with wild-type Jc1 , while cells labeled with HAA were infected with Jc1 ( SF ) . Equal amounts of protein in lysates from Jc1-infected and Jc1 ( SF ) -infected cells were mixed together and then subjected to affinity purification of NS5A ( SF ) and bound proteins using Streptactin-Sepharose . Eluted proteins were separated by SDS-PAGE and stained with Coomassie Blue ( Figure 1D ) ; gel slices were subjected to trypsin digestion and tryptic peptides identified by LC-MS/MS . Peptides containing “heavy” L-lys and/or L-arg from Jc1 ( SF ) -infected cells have a higher mass and can be resolved by mass spectrometry from those from Jc1-infected cells so that their respective intensities can be quantitated . Therefore , the ratios of heavy∶light peptide intensities permits discrimination among proteins specifically bound to NS5A ( SF ) ( heavy∶light peptide ratio >1 ) , proteins nonspecifically bound to Streptactin-Sepharose ( heavy∶light peptide ratio ∼1 ) , and environmental contaminants introduced during protein purification and gel preparation ( e . g . keratins from skin; heavy∶light peptide ratio <<1 ) . As shown in Table 1 , this strategy successfully identified multiple known host proteins known to interact with NS5A or with the HCV life cycle , such as PI4KA , cyclophilin A , casein kinase 1 , VAPA/h-VAP33 , FKBP8 , and B-IND1/PTPLAD1 , thus validating this strategy for the discovery of novel NS5A-interacting cellular proteins . We also identified Rab18 as a previously unknown NS5A-binding protein with a heavy∶light peptide ratio of 8 . 16 , which was higher than those of FKBP8 ( 4 . 83 ) and B-IND1/PTPLAD1 ( 3 . 24 ) and similar to that of cyclophilin A . No peptides derived from viral proteins were identified by mass spectrometry . Rab18 , a member of the Rab family of small GTPases , has not been previously shown to have a function in the HCV life cycle . Rab18 associates with lipid droplets [32] , [33] , which form a cellular compartment for the storage of neutral lipids and cholesteryl esters . Lipid droplets ( LDs ) are believed to be a site of HCV particle assembly [26] , and NS5A interacts with LDs independently of core protein [24] , [25] , [27] . To further evaluate the significance of NS5A binding to Rab18 , we proceeded to examine the localization of NS5A and endogenous Rab18 in HCV-infected cells . We first studied endogenous Rab18 to avoid possible artifacts due to Rab18 overexpression . Triple labeling of JFH-1 infected Huh7 . 5 . 1 cells with antibodies against NS5A , Rab18 , and a fluorescent neutral lipid stain to visualize lipid droplets revealed NS5A staining around Rab18-positive LDs ( Figure 2A ) . We determined that 59±19% of LDs stained positive for endogenous Rab18; of the Rab18+ LDs , 92±13% were also NS5A positive . Endogenous Rab18 and genotype 1 NS5A also displayed substantial colocalization in the OR6 cell line , which harbors a full-length genotype 1b HCV replicon with a Renilla luciferase reporter gene [39] ( Figure S1A ) . The association of JFH-1 core protein with LDs has been described in detail; however , others have reported that Jc1 core protein is more closely associated with ER membranes at steady-state [30] , [40] . We found that core protein appeared to localize around LDs in both JFH-1 and Jc1 ( SF ) -infected cells ( Figure S1B ) . We next examined the distribution of NS5A with respect to Rab18 at the ultrastructural level in Huh7 . 5 . 1 cells stably expressing GFP-Rab18 . Ultrathin frozen sections of control cells or cells infected with HCV for 5 days were double immunolabeled for GFP-Rab18 and for NS5A . Infected cells showed numerous electron-lucent Rab18-positive lipid droplets distributed throughout the cytoplasm ( Figure 2B ) . NS5A and Rab18 showed colocalization on the surface of lipid droplets and on associated membranes , recognizable as rough ER in some areas ( Figure 2C and D ) . In contrast , non-infected control cells showed GFP-Rab18 associated with ER and putative lipid droplets; labeling of GFP-Rab18 surrounding electron-lucent presumed LDs with closely apposed putative ER elements was a common observation in these cells ( Figure 2E , white arrows ) . NS5A immunoreactivity was not observed in uninfected cells . A semiquantitative evaluation of immunogold labeling of LDs in infected cells showed that among a total of 105 analyzed LDs , 36 ( 34% ) had Rab18 labeling . Of these 36 Rab18+ LDs , 34 ( 94% ) were also labeled for NS5A . An additional 27 LDs ( 26% ) were positive for NS5A and negative for Rab18 labeling . Quantitation of an independent labeling experiment gave similar results ( 30% of the LDs in infected cells had Rab18 labeling; of these Rab18+ LDs , 89% were also labeled for NS5A ) . These results suggest that the majority of Rab18+ lipid droplets are associated with HCV NS5A . Given the colocalization of NS5A with Rab18 at lipid droplets , we next proceeded to examine the effect of Rab18 silencing on the HCV life cycle . We first silenced Rab18 in the OR6 replicon cell line . OR6 replicon cells were transduced with lentiviral vectors encoding two independent shRNAs targeting Rab18 ( denoted as shRab18-A and shRab18-B ) , a control nontargeting shRNA ( NTshRNA ) , or a positive control shRNA targeting PI4KA ( shPI4KA ) [41] . As shown in Figure 3A , Rab18 silencing inhibited replication of the genotype 1b replicon compared to cells expressing a nontargeting shRNA ( black bars ) . Cell viability , as monitored by determination of cellular ATP content , was not significantly affected by Rab18 silencing ( Figure 3A , white bars ) . Immunoblotting demonstrated that Rab18 protein levels were markedly lower in cell lines expressing either Rab18 shRNA ( Figure 3B ) . We then studied the effect of Rab18 silencing on the replication of the fully infectious genotype 2a Jc1/Gluc2A virus bearing a secreted Gaussia luciferase reporter [36] . Jc1/Gluc2A replication was inhibited in Huh7 . 5 . 1 cell pools stably expressing Rab18-targeting shRNAs relative to cells expressing a nontargeting shRNA ( Figure 3C , black bars ) . As observed with OR6 replicon cells , Rab18 silencing had no effect on cell viability ( Figure 3C , white bars ) and Rab18 protein knockdown was confirmed by immunoblotting ( Figure 3D ) . Although the observation that two independent Rab18 shRNAs both block HCV replication , we performed a rescue experiment to exclude the possibility that the inhibition of viral replication could be due to off-target effects . We generated two GFP-Rab18 mutants , one with silent mutations at the site targeted by shRab18-A ( denoted as GFP-Rab18-mutA ) and another with silent mutations at the site targeted by shRab18-B ( GFP-Rab18-mutB ) . As shown in Figure 3E , expression of GFP-Rab18-mutA rescued HCV replication in OR6 replicon cells transduced with shRab18-A , while GFP-Rab18-mutB did not . Rab18 protein knockdown and GFP-Rab18 expression were confirmed by immunoblotting ( Figure 3F ) . The localization of Rab18 to lipid droplets and the importance of HCV-LD interactions in virion assembly suggested that Rab18 might play a role in infectious particle production . As Rab18 is known to promote the interaction of ER membranes with LDs , we hypothesized that Rab18 overexpression might enhance the production of infectious particles . Huh7 . 5 . 1 cells stably expressing GFP or GFP-Rab18 were infected with Jc1/Gluc2A , and replication was determined at 72 hr post-infection by measuring Gaussia luciferase activity in the culture supernatant . GFP-Rab18 overexpression did not alter OR6 replication ( Figure 4A ) or Jc1/Gluc2A replication ( Figure 4B ) compared to GFP overexpression . Although HCV replication is impaired by Rab18 silencing , these results suggest that Rab18 levels are not limiting for HCV replication in cell culture . Immunoblotting confirmed expression of both GFP and GFP-Rab18 at similar levels ( Figure 4C ) . As depicted in Figure 4D , relative quantification of released infectious virus was determined by infection of naive Huh7 . 5 . 1 cells followed by measurement of Gaussia luciferase activity at 72 hr postinfection . GFP-Rab18 overexpression enhanced secretion of infectious Jc1/Gluc2A by nearly 2-fold compared to GFP overexpression ( Figure 4E ) . We also observed significantly increased infectious particle release in Huh7 . 5 . 1 cells overexpressing GFP-Rab18 using the JFH-1 strain of HCV . ( Figure 4F ) . As shown in Figure 5A , cell pools stably expressing a nontargeting shRNA or Rab18-targeting shRNA were infected with Jc1/Gluc2A and cell culture medium containing released virus was harvested at 48 hr postinfection . Relative quantification of released infectious virus was determined by infection of naive Huh7 . 5 . 1 cells followed by measurement of Gaussia luciferase activity at 72 hr postinfection . Jc1/Gluc2A infectious particle release was reduced by Rab18 silencing by 69 . 5 and 74 . 9% percent in the shRab18-A and -B cell pools , respectively , compared to the nontargeting shRNA cell pool ( Figure 5B ) . To exclude a possible effect of Rab18 silencing on HCV entry , we transfected in vitro-transcribed Jc1/Gluc2A RNA into Huh7 . 5 . 1 cells stably expressing a nontargeting shRNA or Rab18-targeting shRNAs , and observed similar reductions of HCV particle release by Rab18 knockdown ( Figure 5C ) . We also tested the effect of Rab18 silencing on infectious particle release using the JFH-1 strain of HCV . Absolute quantitation of released infectious virus by a focus-forming unit assay showed that infectious particle release was inhibited by Rab18 silencing by 50 and 67% percent in the shRab18-A and -B cell pools , respectively , compared to the nontargeting shRNA cell pool ( Figure 5D ) . We also quantitated the relative amounts of intracellular and extracellular infectious virus in Jc1/Gluc2A infected stable cell lines . Rab18 knockdown decreased levels of both intracellular and extracellular infectious virus to a similar extent ( Figure 5E ) . Although we cannot definitively determine that Rab18 silencing blocks particle assembly because it also inhibits viral replication , the observation that Rab18 overexpression enhances infectious particle production without affecting replication does suggest that Rab18 may play a role in particle production . Density gradient fractionation demonstrated that the buoyant densities of the fractions with peak extracellular and intracellular infectivity in HCV-infected cells were similar between shRab18 and NTshRNA expressing cell lines ( Figure S2A ) , suggesting that Rab18 knockdown does not grossly alter the physical properties of infectious virions . The specific infectivity of the fractions with peak extracellular and intracellular infectivity was determined; the extracellular and intracellular specific infectivities measured from shRab18-expressing cells were not significantly different from NTshRNA-expressing cells ( Figure S2B ) . These specific infectivities are consistent with previous studies [42] . As the inhibition of HCV production by Rab18 knockdown could simply be due to a block of LD biogenesis , we tested the effect of Rab18 silencing on LD accumulation induced by oleic acid loading . However , Rab18 silencing had no effect on mean LD diameter in Huh7 . 5 . 1 cells or on LD accumulation in oleic acid-loaded cells ( Figure S3 ) compared to cells stably expressing a nontargeting shRNA . Therefore , Rab18 supports HCV infectious particle production by a mechanism other than simple regulation of LD volume . Rab proteins , like other small GTPases , cycle between an inactive , GDP-bound conformation and an active , GTP-bound conformation . We asked whether NS5A binds preferentially to a particular Rab18 conformation . 293T cells were cotransfected with expression plasmids encoding NS5A ( SF ) and either GFP , GFP-Rab18 , the constitutively GDP-bound mutant GFP-Rab18 ( S22N ) , or the constitutively GTP-bound mutant GFP-Rab18 ( Q67L ) . These two mutants have been previously characterized [33] . NS5A ( SF ) and bound proteins were recovered by binding to Streptactin-Sepharose . Expression levels of the GFP-Rab18 constructs were similar as determined by immunoblotting of cell lysates ( Figure 6A ) . As shown in Figure 6A , NS5A ( SF ) bound to GFP-Rab18 but not to GFP alone . In addition , the GDP-bound Rab18 ( S22N ) mutant bound more weakly to NS5A ( SF ) than wild-type Rab18 or the GTP-bound Rab18 ( Q67L ) mutant ( Figures 6A and B ) . These results confirm that Rab18 interacts with NS5A in cells; furthermore , they suggest that NS5A has a higher affinity for the GTP-bound conformation of Rab18 than the GDP-bound conformation . We examined the distribution of NS5A in HCV-infected cells overexpressing wild-type Rab18 or the S22N or Q67L mutants ( Figure 6C ) . As expected , NS5A colocalized with wild-type Rab18 or the activated Rab18 ( Q67L ) mutant around LDs . In contrast , overexpression of Rab18 ( S22N ) , which does not localize to LDs , was associated with reduced NS5A-LD association . Specifically , 94±4 . 5% ( 251 of 267 LDs ) and 95±4 . 7% ( 177 of 186 ) of LDs in GFP-Rab18 ( wt ) and GFP-Rab18 ( Q67L ) expressing cells were NS5A positive , respectively , while 46±5 . 4% ( 65 of 140 ) of LDs in GFP-Rab18 ( S22N ) overexpressing cells were NS5A positive . Quantitation of lipid droplets in oleic acid-loaded Huh7 . 5 . 1 stably expressing GFP or GFP-Rab18 ( wild type , S22N , or Q67L ) demonstrated no significant alterations in the average LD diameter ( Figure S4 ) . As the number of cells overexpressing GFP-Rab18 ( S22N ) in stable cell lines was consistently low despite multiple attempts , we were unable to study the effect of GFP-Rab18 ( S22N ) overexpression on HCV particle release ( data not shown ) . Overexpression of the constitutively active GFP-Rab18 ( Q67L ) mutant did result in enhanced HCV particle release to a degree similar to that seen with wild-type GFP-Rab18 overexpression ( data not shown ) . Given the known ability of Rab18 to approximate ER membranes to lipid droplets , we hypothesized that Rab18 might also help physically tether NS5A-positive membranes to LDs . We tested this hypothesis by first infecting Huh7 . 5 . 1 cell lines stably expressing Rab18 shRNAs or a nontargeting shRNA with JFH-1 . Silencing of Rab18 was associated with a decrease in NS5A association with LDs compared to cells expressing a nontargeting shRNA ( Figure 7A ) . We quantitatively assessed the decrease in NS5A-LD association by Rab18 silencing in cells infected with HCV , and found that Rab18 silencing led to a significant decrease in the proportion of LDs that stained positive for NS5A . In nontargeting shRNA-expressing cells , 85±6% of LDs stained positive for NS5A , while in two independent Rab18 shRNA-expressing cell lines , only 21±15% and 26±13% of LDs were NS5A-positive ( Figure 7B ) . As NS5A is a component of HCV replication platforms , we asked whether the association of other HCV components with LDs could also be modulated by Rab18 . Huh7 . 5 . 1 cells stably expressing a nontargeting shRNA or a Rab18-targeting shRNA were infected with the JFH-1 strain of HCV , homogenized , and the post-nuclear supernatant was subjected to low-speed centrifugation . The supernatant ( S1 ) was then fractionated by density gradient centrifugation to isolate LDs . Similar amounts of HCV RNA from nontargeting or Rab18 shRNA-expressing cells were loaded onto the density gradients ( Figure S5A ) . Fractions from the top of the density gradient ( enriched in LDs ) and from the bottom of the density gradient ( enriched in cytosol and other intracellular membranes ) were immunoblotted for NS5A , NS3 , calnexin , actin , and Rab18 ( Figure 7C ) . Most of the ER membranes ( as shown by immunoblotting for the ER marker calnexin ) , NS5A , and NS3 were removed by the 16 , 000× g centrifugation step . As expected , Rab18 was enriched in the LD fraction from NTshRNA-expressing cells compared to the non-LD fraction . In control cells , nearly a quarter of the NS5A in the S1 supernatant could be recovered in the LD-enriched top gradient fraction , with a smaller amount of NS3 found in the top fraction . On the other hand , we observed substantially lower recoveries of LD-associated NS5A and NS3 in cells silenced for Rab18 . We then quantitated the amount of LD-associated HCV RNA using strand-specific quantitative RT-PCR , and found that cells silenced for Rab18 had decreases in both LD-associated negative-strand and positive-strand RNA compared to cells expressing nontargeting shRNA , with some increase in the amount of negative and positive-strand RNA isolated from non-LD fractions ( Figure 7D ) . These findings indicate that Rab18 promotes not only the association of NS5A , but also other components of the HCV replication complex , with LDs . The association of HCV core protein with LDs is believed to be essential for particle assembly [29] , [43] . We examined the distribution of core protein in JFH-1 infected cells stably expressing nontargeting or Rab18-targeting shRNAs . The association of core with LDs was not altered by Rab18 silencing or overexpression of Rab18 ( wt or mutant ) ( Figures S6 and S7 ) , indicating that core protein association with LDs is not regulated by Rab18 .
In this study , a strategy combining affinity purification with mass spectrometry to identify novel host factors that associate with NS5A in cells infected with a cell-culture infectious HCV strain revealed that Rab18 , a lipid droplet-specific Rab GTPase , interacts with NS5A . A similar strategy using Strep-tag affinity purification of NS5A has been previously used to identify oxysterol-binding protein ( OSBP ) as an NS5A-interacting protein [44] . That study used a subgenomic replicon , as opposed to our use of a fully infectious HCV construct , which may have enhanced our ability to detect host proteins involved in virion production . In addition , we used SILAC to reduce the identification of false-positive protein interactions . Other Rab GTPases have been described to promote HCV replication . The endocytic Rab5 and Rab7 proteins appear to facilitate HCV genome replication [45]–[47] , perhaps by inducing autophagy [48] . A proteomic analysis of detergent-resistant membrane ( DRM ) fractions , which are believed to be enriched in HCV replication complexes , isolated from Huh7 cell lines expressing a full-length HCV replicon cells found that HCV-replicating cells had increased levels of Rab7 associated with DRMs compared to control Huh7 cells [49] . Of note , this study also found that Rab18 was enriched on DRM fractions from replicon-expressing cells , although its role in the HCV life cycle was not directly tested . In addition to Rab5 and Rab7 , other work has suggested a role for Rab1 and TBC1D20 , a Rab1 GTPase-activating protein , in HCV replication . TBC1D20 binds to NS5A [50] , and depletion of TBC1D20 or Rab1 by RNA interference reduces HCV RNA levels [50] , [51] . In the absence of NS5A , neither TBC1D20 nor Rab1 is found on lipid droplets; it appears that NS5A recruits both of these proteins to LDs [25] . In contrast , both endogenous and overexpressed Rab18 localize to LDs in the absence of HCV [32] , [33] . In HCV-infected cells , both endogenous and overexpressed Rab18 colocalize with NS5A around LDs . Rab18 silencing and expression of dominant-negative Rab18 reduce association of NS5A with LDs . NS3 and HCV RNA ( both negative and positive-strand ) cofractionation with LDs is also reduced by Rab18 silencing . Furthermore , immunoelectron microscopy identifies NS5A-positive membrane profiles that are in close association with Rab18-positive LDs in Rab18-overexpressing cells . Taken together , these findings suggest that Rab18 not only recruits NS5A but also membranous sites of HCV genome replication ( also known as “membranous webs” ) to lipid droplets . This is consistent with two previous reports that Rab18 overexpression leads to the close apposition of ER membranes to LDs [33] , [52] . In these studies , electron microscopy of Rab18-overexpressing cells demonstrated wrapping of thin membrane cisternae around LDs , some of which were continuous with the rough ER or themselves studded with ribosomes . Ozeki et al . proposed that LD-associated ER membranes represented a specialized ER region that they termed the lipid droplet-associated membrane , or LAM; this may be analogous to other ER domains that appose other intracellular membranes such as the plasma membrane and mitochondria . In contrast to the effect of Rab18 modulation on the association of NS5A and other membranous web components with LDs , we did not observe any effect of Rab18 modulation on the localization of HCV core protein to LDs , suggesting that the interaction between core protein and LDs is Rab18-independent . The current model proposing that Rab18 recruits ER membranes to LDs may be relevant to the HCV life cycle . While it is generally accepted that HCV replication occurs on ER-derived membranes , it is unclear where progeny HCV genomes are packaged into virions . While LDs appear to play a critical role in the process of HCV assembly , it remains uncertain whether HCV assembly occurs at the LD membrane itself or with closely-associated ER membranes ( reviewed in [53] ) . Assembled HCV virions then traffic through the secretory pathway with glycosylation of the envelope proteins , indicating that HCV virions must then enter the ER lumen during or following assembly . Attempts at identifying the site of HCV virion assembly by microscopic techniques have so far been unsuccessful , likely because the rate of assembly events is likely to be low and because HCV viral particles are heterogeneous in appearance [54] . Nevertheless , an important unanswered question is how HCV RNA replication is physically coupled to virion production and subsequent entry into the cellular secretory pathway . Our observations support a model in which Rab18 helps promote interaction between LDs and HCV membranous webs through a direct association between NS5A and the active , GTP-bound form of Rab18 . A somewhat unexpected finding was that Rab18 silencing inhibited HCV replication without inhibiting LD biogenesis in response to fatty acid loading . While LDs have been shown to be important for HCV particle assembly , there has been relatively little evidence that LDs support HCV genome replication . Indirect support for this possibility comes from studies demonstrating that Rab1 supports HCV replication and that a dominant negative Rab1 mutant suppresses LD formation . In addition , during preparation of this manuscript it was reported that knockdown of the LD-associated protein TIP47 inhibits HCV replication in addition to infectious particle release [55] . Further studies are needed to determine the mechanisms by which cellular LDs assist in HCV replication . We speculate that the Rab18-promoted interaction between HCV membranous webs and LDs may also enhance virion assembly by bringing sites of replication in close physical proximity to sites of virion assembly , either on the LD surface or on membranes apposed to the LD . Because Rab18 silencing also affects HCV genome replication , the decrease in intracellular and extracellular virus titers seen in Rab18 silenced cells cannot be definitively ascribed to inhibition of particle assembly . However , the observation that Rab18 overexpression increases infectious virus production without altering replication does support a possible role for Rab18 in virus production . Note that this model does not exclude the possibility of additional protein-protein interactions that might mediate LD-membranous web interactions . For example , the HCV core protein also appears to play a key role in recruiting HCV nonstructural proteins to LDs [26] . On the other hand , we and others have found that HCV NS5A localizes to LDs when expressed alone in the absence of core protein [24] , [25] , [27] and ( Salloum and Tai , unpublished observations ) , indicating that core protein is not necessary for NS5A-LD interaction . Intriguingly , a small-molecule inhibitor of NS5A relocalizes NS5A from the ER to LDs in cells expressing a genotype 1b replicon [27] , suggesting that the interaction between NS5A and LDs may also be regulated by NS5A conformation . It has been proposed that phosphorylation of NS5A may direct a switch from viral replication to virion assembly [56]; future studies will examine the potential role of Rab18 interaction with NS5A in modulating this transition .
Huh7 . 5 . 1 cells [57] , a subline of Huh7 human hepatoma cells that is highly permissive for HCV replication , and 293T cells were grown in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% fetal bovine serum ( FBS ) , nonessential amino acids , 100 U/mL of penicillin , and 100 µg/mL of streptomycin . OR6 cells containing a full-length genotype 1b HCV replicon with a Renilla luciferase reporter gene have been described elsewhere [39] . Antibodies used included those directed against calnexin ( mouse monoclonal; Sigma-Aldrich , St Louis , MO ) , FLAG epitope tag ( mouse monoclonal clone M2; Sigma-Aldrich ) , GFP ( rabbit monoclonal , Cell Signaling Technology , Beverly , MA ) , HCV core ( mouse monoclonal clone 6G7; Dr . Harry Greenberg , Stanford , CA ) , HCV NS5A ( mouse monoclonal clone 9E10; Dr . Charles Rice , Rockefeller University , New York , NY ) , Rab18 ( rabbit polyclonal; Proteintech , Chicago , IL ) , and β-Actin ( mouse monoclonal; Sigma-Aldrich ) . HCS LipidTOX Deep Red neutral lipid stain , BODIPY 493/503 , DAPI , and Alexa Fluor conjugated secondary antibodies for microscopy experiments were purchased from Invitrogen ( Carlsbad , CA , USA ) . HCV clones used in this work include the cell culture infectious genotype 2a clone JFH-1 [58] and Jc1/Gluc2A [36] , which encodes a Gaussia luciferase reporter gene in the J6/JFH-1 chimeric genome known as Jc1 [35] . Jc1 ( SF ) refers to Jc1 with insertion of a tandem affinity purification tag ( Strep-tag and FLAG ) into NS5A described elsewhere [17] . Two independent Rab18 shRNA lentiviral vectors for Rab18 knockdown were obtained from the TRC shRNA library [59] ( TRCN0000021981 and TRCN0000021983 , Sigma-Aldrich ) . A nontargeting shRNA in the same pLKO . 1 backbone ( Addgene plasmid 1864 ) was used as a negative control . GFP-Rab18 lentiviral expression constructs were generated from plasmids encoding EGFP-human Rab18 and the mutants S22N and Q67L [33] by PCR amplification of the GFP-Rab18 coding sequences using the primer pair 5′-CGCGTGCCGCCACCATGGTGAGCAAG-3′ , and 5′-CGTACGTTATAACACAGAGCAATA ACCACCACAGG-3′ . The amplicons were subcloned using MluI and BsiWI restriction sites into the pSMPUW-IRES-blasticidin lentiviral expression vector ( Cell Biolabs , San Diego , CA ) . We modified this vector to include the MluI and BsiWI restriction sites in the multiple cloning site; details of vector construction are available upon request . EGFP was subcloned into the same vector as a negative control . All constructs were confirmed by sequencing . The GFP-Rab18-mutA construct was generated by introducing three silent mutations into the TRCN0000021981 shRNA target region using mutagenic primers 5′-CAGGCCTTCATTACGGTCTACTTCACGATTTTCCTTATCG -3′ , and 5′-CGTGAAGTAGAC CGTAATGAAGGCCTGAAATTTGC -3′ , while the GFP-Rab18-mutB construct contained silent mutations in the TRCN0000021983 shRNA target region introduced using mutagenic primers 5′-GACATCATAAACAAGAATTACACCCTGTGCACCTCTATAATAG -3′ and 5′- GGTGCACAGGGTGTAATTCTTGTTTATGATGTCACAAGAAG -3′ . Lentiviral expression constructs encoding full-length NS5A ( SF ) , NS5A ( SF ) lacking the N-terminal amphipathic helix , and NS5A ( SF ) lacking domains I , II , or III were generated by PCR amplification from the JFH-1 NS5A sequence and subcloning into our modified pSMPUW-IRES-blasticidin vector using MluI and EcoRI restriction sites . The domain III-deleted construct had a C-terminal SF tag immediately following domain II . All constructs were confirmed by sequencing . Huh 7 . 5 . 1 cells were cultured in SILAC medium ( DMEM with high glucose , sodium pyruvate , without L-arginine or L-lysine , Cambridge Isotope Laboratories , Andover , MA ) with 10% dialyzed FBS , 1% penicillin-streptomycin , and either “light” L-arg and L-lys ( LAA ) or “heavy” L-arg-13C6 , 15N4 and L-lys-13C6 , 15N2 ( HAA ) at concentrations of 146 mg/L and 84 mg/L , respectively . Unlabeled L-proline ( 200 mg/L ) was also added to inhibit arginine-to-proline conversion [60] . All amino acids were obtained from Sigma-Aldrich . Cells were grown for 6 days ( two passages ) in “heavy” or “light” medium for complete metabolic labeling of the cells grown in “heavy” medium prior to viral infection . Pilot experiments confirmed 97% incorporation of heavy lysine and arginine into the proteome of Huh7 . 5 . 1 cells grown under these conditions with negligible arginine to proline conversion ( data not shown ) . Cells were then plated at a density of 106 cells per 10 cm dish 24 hr before infecting LAA-treated cells with Jc1 and HAA-treated cells with Jc1 ( SF ) virus at a MOI of 1 . 2 days post-infection , Jc1 and Jc1 ( SF ) -infected cells were split 1∶5 into medium containing “light” or “heavy” medium , respectively . 4×107 cells were harvested from each condition at day 6 postinfection and lysed in 50 mM Tris pH 7 . 5 , 150 mM NaCl , 1 mM EDTA , and 0 . 5% Triton X-100 with Halt protease inhibitor cocktail ( Pierce , Rockford , IL ) . After centrifugation at 21 , 000× g for 15 min at 4°C to remove cell nuclei and insoluble material , 1 . 3 mg of “light” and “heavy” labeled protein isolated from Jc1 and Jc1 ( SF ) infected cells , respectively , were mixed together and then incubated with Streptactin-Sepharose ( IBA , Göttingen , Germany ) for 1 hr at 4°C . Unbound material was removed by 3 washes in lysis buffer with 0 . 5% Triton X-100 followed by 3 washes in lysis buffer with 0 . 1% Triton X-100 . Bound proteins were eluted with lysis buffer with 0 . 1% Triton X-100 and 4 mM biotin at 4°C . To remove Triton X-100 and concentrate the eluted protein , proteins were precipitated by deoxycholic acid/trichloroacetic acid followed by acetone washes and then were solubilized in SDS-PAGE loading buffer . Proteins were separated on SDS-PAGE and visualized by Coomassie Blue staining . The stained gel was excised into ten equally sized segments . In-gel digestion was performed by washing with 25 mM ammonium bicarbonate followed by acetonitrile , reducing with 10 mM dithiothreitol at 60°C followed by alkylating with 50 mM iodoacetamide at RT . Samples were digested with trypsin at 37°C for 4 hr followed by formic acid quenching and direct analysis without further processing . Each gel digest was analyzed by nano LC-MS/MS with a Waters nanoACQUITY UPLC system ( Waters , Milford , MA ) interfaced to a LTQ Orbitrap Velos ( Thermo Fisher Scientific , Waltham , MA ) . Peptides were loaded on a trapping column and eluted over a 75 µm analytical column at 350 nL/min; both columns were packed with Jupiter Proteo resin ( Phenomenex , Torrance , CA ) . The mass spectrometer was operated in data-dependent mode , with MS performed in the Orbitrap at 60 , 000 FWHM resolution and MS/MS performed in the LTQ . The fifteen most abundant ions were selected for MS/MS . MaxQuant software v1 . 0 . 13 . 13 was used to recalibrate MS data , filter database search results at the 1% protein and peptide false discovery rate , and calculate SILAC heavy∶light ratios . Protein and peptide identifications were performed using Mascot ( Matrix Science , Boston , MA ) , which was configured to include Carbamidomethyl ( C ) as a fixed modification and the following variable modifications: Oxidation ( M ) , Acetyl ( N-term ) , Pyro-Glu ( N-term Q ) , Deamidation ( N , Q ) , ( 13C6 , 15N2 ) K and ( 13C6 , 15N4 ) R . Data were searched with a peptide mass tolerance of 10 ppm and fragment mass tolerance of 0 . 5 Da . The database used was a combination of Uniprot human and virus databases that were reversed , concatenated and appended with common contaminants . VSV-G pseudotyped lentiviral vectors ( for shRNA or protein expression ) were produced by cotransfection of 293T cells with the packaging vectors ( Addgene plasmid 12260 ) and pMD2 . G ( Addgene plasmid 12259 ) . Lentiviral supernatants were harvested at 48 and 74 hr post-transfection , 0 . 45 µm filtered , and stored at −80°C . Target cells were transduced with lentiviral particles for 4 hr in the presence of 8 µg/mL polybrene ( Sigma ) . For generation of knockdown Rab18 cell lines , Huh7 . 5 . 1 cells were transduced with lentivirus encoding two independent shRNAs targeting Rab18 or a nontargeting shRNA . Stable knockdown pools were isolated by puromycin selection . For stable expression of GFP-Rab18 , Huh7 . 5 . 1 cells were transduced with lentiviral expression vectors encoding GFP or GFP-Rab18 ( wild-type , S22N , or Q67L ) followed by blasticidin selection . Cells grown on poly-D-lysine coated glass coverslips were rinsed in PBS and fixed in 4% paraformaldehyde in PBS for 15 min at RT followed by quenching in 50 mM NH4Cl in PBS for 10 min at RT . Blocking and antibody incubations were performed in 10% fetal bovine serum and 0 . 1% saponin in PBS for 1 hr at RT . Blocking was followed by incubation with primary antibody in blocking buffer for 1 hr in RT followed by 4 washes in PBS for 5 min each . Secondary antibody detection was performed with Alexa Fluor 488- and 568- conjugated anti-rabbit and anti-mouse antibodies ( Invitrogen ) . Lipid droplets were visualized using either LipidTOX Deep Red or BODIPY 493/503 ( Invitrogen ) according to the manufacturer's directions . After 4 washes in PBS for 5 minutes each , coverslips were mounted with Prolong Gold with DAPI ( Invitrogen ) and viewed on an Olympus FluoView FV500 or Nikon A1 laser scanning confocal microscope with sequential scanning mode to limit crosstalk between fluorochromes . For the purposes of quantitation , we defined NS5A association with lipid droplets as a minimum of 25% of the LD diameter associated with NS5A . Control uninfected Huh7 . 5 . 1 cells stably expressing GFP-Rab18 or cells infected with HCV for 5 days were fixed in 8% paraformaldehyde in phosphate buffer and processed for frozen sectioning according to published protocols [32] . Although glutaraldehyde fixation resulted in superior ultrastructural preservation , immunoreactivity of glutaraldehyde-fixed specimens with two different NS5A antibodies was inadequate ( data not shown ) . Ultrathin frozen sections were labeled with rabbit antibodies to GFP [61] and the mouse 9E10 antibody to NS5A , followed by goat anti-mouse antibodies conjugated to 10 nm gold particles , and goat anti-rabbit antibodies conjugated to 15 nm gold particles ( BBI , Cardiff , UK ) . Sections were viewed in a JEM-1011 microscope ( Jeol , Japan ) . Selected cell lines were incubated with 180 µM oleic acid-BSA ( Sigma ) for 24 hr . Cells were fixed and processed for LD staining with BODIPY 493/503 ( for shRNA-expressing cells ) or HCS LipidTox Deep Red ( for GFP-expressing cells ) with DAPI nuclear counterstaining as described above . Quantitation of LD size and number was performed using NIH ImageJ version 1 . 47 g software . Briefly , images were background-subtracted and thresholded by Otsu's method . LDs in close proximity to one another were separated by watershed segmentation , and their Feret diameters were quantitated using the ImageJ particle analysis function . For measurement of HCV replication , Renilla luciferase activity in OR6 replicon cells was measured using the Renilla Luciferase Assay System ( Promega , Madison , WI ) and a Synergy 2 microplate reader equipped with reagent injectors ( Biotek , Winooski , VT ) using the manufacturer's directions . Cell viability was determined by measurement of cellular ATP content using the CellTiter-Glo Assay ( Promega ) . Cells were infected with JFH-1 or Jc1/Gluc2A at an MOI of 3 for 4 hr at 37°C . Alternatively , Jc1/Gluc2A RNA was transcribed in vitro as described in [62] and the in vitro transcribed RNA was transfected into cells with TransiT mRNA reagent ( Mirus Bio , Madison , WI ) as described in [63] . The cell monolayers were washed with complete medium to remove input virus , and the cell culture medium was harvested at 48 hr postinfection . Secreted JFH-1 virus in cell culture supernatants was quantitated by a focus-forming unit ( FFU ) assay . Each sample was serially diluted tenfold and used to infect naive Huh7 . 5 . 1 cells in 96-well plates . 3 days postinfection , cells were fixed with methanol and stained for HCV core; the titer was calculated by counting infected cell foci and expressed in FFU/mL . For cells infected with Jc1/Gluc2A , cell culture medium collected at 48 hr post-infection was used to infect naïve Huh7 . 5 . 1 cells . Gaussia luciferase activity was measured at 72 hr post-infection as described in [36] . For determination of intracellular versus extracellular infectious virus titer , selected cell lines were infected with Jc1/Gluc2A virus at an MOI of 1 . At 72 hr postinfection , supernatants containing extracellular virus were harvested . Infected cells were washed to remove residual extracellular virus , collected by trypsinization and centrifugation , and the cell pellet was resuspended in a volume of medium equal to the extracellular virus supernatant . After three rounds of freeze-thawing to release intracellular virus particles followed by centrifugation to remove cell debris , the released intracellular virus and the extracellular virus supernatants were used to infect naïve Huh7 . 5 . 1 cells . Gaussia luciferase activity was measured at 72 hr post-infection . Density gradient fractionation of HCV virions was performed as previously described by [42] with the following minor modifications: 1 mL samples were layered on top of a discontinuous gradient of 6%-12%-18%-24%-30%-40% iodixanol and centrifuged at 30 , 000 rpm ( 111 , 000× gavg ) for 18 hr at 4°C in an SW41 rotor . 293T cells were cotransfected with GFP-Rab18 ( or GFP as a negative control ) and Strep/FLAG-tagged NS5A constructs using Fugene HD ( Promega ) according to the manufacturer's directions . 48 hr after transfection , cells were washed with phosphate-buffered saline ( PBS ) and lysed in 50 mM Tris pH 7 . 5 , 150 mM NaCl , 1 mM EDTA , and 0 . 5% Triton X-100 with protease inhibitors . After centrifugation at 21 , 000× g for 15 min at 4°C to remove insoluble material , equal amounts of protein from clarified lysates were incubated with Streptactin-Sepharose ( IBA , Gottingen , Germany ) for 1 hr at 4°C to recover Strep/FLAG-tagged NS5A . Unbound material was removed by washing in lysis buffer followed by washing in lysis buffer with 0 . 1% Triton X-100 . Bound proteins were eluted with 4 mM biotin in lysis buffer with 0 . 1% Triton X-100 at 4°C and then separated on SDS-PAGE for immunoblotting . Quantitation of chemiluminescence signals from immunoblots was performed on an Odyssey Fc imager ( Li-Cor , Lincoln , NE ) . Isolation of lipid droplet-enriched fractions from cells was performed by density gradient centrifugation . Briefly , selected cell lines were infected with JFH-1 and harvested at day 5 post-infection . The cell pellet was resuspended in homogenization buffer ( 10 mM Tris pH 8 . 0 , 135 mM NaCl , 10 mM KCl , 5 mM MgCl2 ) supplemented with protease inhibitors ( Pierce ) . The cell suspension was homogenized with 10 strokes of a ball-bearing homogenizer ( Isobiotec , Heidelberg , Germany; 10 µm clearance ) and then centrifuged at 1 , 000× g for 10 min at 4°C to remove nuclei and unbroken cells . The postnuclear supernatant was then spun at 16 , 000× g for 15 min at 4°C . The pellet ( P1 ) was saved for subsequent immunoblotting . The supernatant ( S1 ) was mixed with an equal volume of 1 . 04 M sucrose in homogenization buffer and transferred to a 5 mL ultracentrifuge tube ( Beckman Coulter , Indianapolis , IN ) . This was overlaid with 1 mL of homogenization buffer and the discontinuous gradient was centrifuged at 45 , 000 rpm for 1 . 5 hr at 4°C in a MLS-50 rotor ( Beckman Coulter ) After the centrifugation , the LD fraction was recovered from the top of the gradient . The collected LD fraction , the bottom fraction , and the S1 supernatant were used for immunoblotting analysis . Quantitation of chemiluminescence signals from immunoblots was performed on an Odyssey Fc imager ( Li-Cor , Lincoln , NE ) . LD-associated HCV RNA was isolated using RNeasy Mini columns ( QIAGEN ) and strand-specific quantitative RT-PCR was performed as previously described [64] with the following minor modifications . The Tag-RC1 and RC1 primers were changed to the sequences GGCCGTCATGGTGGCGAATAAGCCTAGCCATGGCGTTAGTA and GCCTAGCCATGGCGTTAGTA , respectively , to match the JFH-1 5′NTR sequence . Quantitation of the cDNA was performed using the DyNAmo HS SYBR Green qPCR kit ( Finnzyme , Espoo , Finland ) and primers described in [41] .
|
Hepatitis C virus ( HCV ) chronically infects about 170 million people worldwide and can ultimately lead to liver failure and liver cancer . HCV , like other RNA viruses , exploits cellular proteins and membranes to promote their own replication and virion production . In particular , HCV replication occurs at membranes derived from the endoplasmic reticulum , while HCV virion assembly is believed to occur at or near cellular lipid droplets . In this work , we report that Rab18 , a lipid droplet-associated cellular protein , binds to the viral protein NS5A , and that the silencing of Rab18 reduces the association of other HCV replication complex components with lipid droplets . These data are consistent with a model in which Rab18 promotes the physical interaction between sites of viral replication to lipid droplets . We also speculate that Rab18 may help to link sites of viral replication to sites of virion assembly . Understanding how viruses exploit cellular proteins may result in new methods of disrupting viral infection .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"hepatitis",
"c",
"medicine",
"infectious",
"diseases",
"hepatitis",
"host",
"cells",
"rna",
"viruses",
"viral",
"transmission",
"and",
"infection",
"viral",
"classification",
"virology",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"viral",
"packaging",
"viral",
"replication",
"viral",
"replication",
"complex",
"viral",
"diseases"
] |
2013
|
Rab18 Binds to Hepatitis C Virus NS5A and Promotes Interaction between Sites of Viral Replication and Lipid Droplets
|
Chagas disease remains one of the most neglected diseases in the world despite being the most important parasitic disease in Latin America . The characteristic chronic manifestation of chagasic cardiomyopathy is the region’s leading cause of heart-related illness , causing significant mortality and morbidity . Due to the limited available therapeutic options , new drugs are urgently needed to control the disease . Sirtuins , also called Silent information regulator 2 ( Sir2 ) proteins have long been suggested as interesting targets to treat different diseases , including parasitic infections . Recent studies on Trypanosoma cruzi sirtuins have hinted at the possibility to exploit these enzymes as a possible drug targets . In the present work , the T . cruzi Sir2 related protein 1 ( TcSir2rp1 ) is genetically validated as a drug target and biochemically characterized for its NAD+-dependent deacetylase activity and its inhibition by the classic sirtuin inhibitor nicotinamide , as well as by bisnaphthalimidopropyl ( BNIP ) derivatives , a class of parasite sirtuin inhibitors . BNIPs ability to inhibit TcSir2rp1 , and anti-parasitic activity against T . cruzi amastigotes in vitro were investigated . The compound BNIP Spermidine ( BNIPSpd ) ( 9 ) , was found to be the most potent inhibitor of TcSir2rp1 . Moreover , this compound showed altered trypanocidal activity against TcSir2rp1 overexpressing epimastigotes and anti-parasitic activity similar to the reference drug benznidazole against the medically important amastigotes , while having the highest selectivity index amongst the compounds tested . Unfortunately , BNIPSpd failed to treat a mouse model of Chagas disease , possibly due to its pharmacokinetic profile . Medicinal chemistry modifications of the compound , as well as alternative formulations may improve activity and pharmacokinetics in the future . Additionally , an initial TcSIR2rp1 model in complex with p53 peptide substrate was obtained from low resolution X-ray data ( 3 . 5 Å ) to gain insight into the potential specificity of the interaction with the BNIP compounds . In conclusion , the search for TcSir2rp1 specific inhibitors may represent a valuable strategy for drug discovery against T . cruzi .
Chagas disease , caused by the protozoan Trypanosoma cruzi , remains one of the most prevalent and neglected diseases in the world [1] and is the parasitic disease with the highest socio-economic burden in Latin America [2 , 3] . More recently , Chagas disease has also emerged as a significant public health concern imparting a serious burden of disease in some Southern regions of the USA [4] . While the vast majority of the newly acquired cases are asymptomatic , about 20 to 30% of the cases lead to the development , usually decades later , of the characteristic chronic chagasic cardiomyopathy that is the leading cause of non-ischemic heart disease in South America [5 , 6] . Although vectorial transmission has been greatly reduced in recent years , mainly due to improved housing and awareness , other transmission mechanisms like congenital or oral transmission , blood transfusion and organ transplantation still contribute to about 40 , 000 new cases and 12 , 000 deaths every year [7] . What used to be a local concern , has become a worldwide problem due to human migrations to other parts of the world such as North America , Europe and Japan [8] . With no available vaccine for Chagas disease therapies rely upon two drugs introduced more than 40 years ago: benznidazole and nifurtimox . These drugs are far from ideal because of their many toxic side-effects and the requirement of long periods of administration [4 , 9 , 10] . Thus , there is an urgent need for new drugs . Very few candidates have populated the drug discovery pipeline for Chagas disease in the past decade , and most have the same mechanism of action [11] . The recent failure in clinical trials of two of these drugs , posaconazole and E1224 , has exposed the fragilities of the chemotherapeutic pipeline and reinforced the need for constant drug discovery efforts to find novel alternative targets and therapies [12–14] . Sir2 ( silent information regulator 2 ) or sirtuins are class III histone deacetylases that are evolutionary conserved and present across various kingdoms of life , from bacteria to humans [15 , 16] . They catalyse the NAD+- dependent deacetylation of acetylated lysine residues in a polypeptide chain , according to the following reaction: Acetylated-Protein + NAD+ → Deacetylated Protein + 2’-O-AADPr + Nicotinamide Besides the deacetylated lysine , other products are formed in the reaction , like nicotinamide ( an endogenous inhibitor ) and 2’-O-acetyl-ADP-ribose ( 2’-O-AADPr ) , that has been suggested to act as a secondary messenger [17] , promoting the association between Sir3/Sir2/Sir4 involved in yeast gene silencing [18] . Some human sirtuins like SIRT4 and SIRT6 and the protozoan Sir2 related protein 1 ( Sir2rp1 ) from Trypanosoma brucei and Leishmania infantum also display ADP-ribosyltransferase activity [19–22] . However , the biological role of ADP-ribosylation by sirtuins has not been clearly demonstrated and is currently debated to be a non-specific side reaction [23] . In addition , some sirtuins have been characterized to perform demalonylase , desuccinylase and deglutarylase activity [24 , 25] . Sirtuins have been attributed many roles in different organisms , including life span regulation , cell cycle progression , gene transcription , apoptosis , DNA repair and metabolism [26–29] . The human genome , as well as other mammals’ genomes , codify 7 distinct sirtuins ( SIRT1-7 ) [30] , whose proteins are distributed in different cell compartments: SIRT1 , SIRT6 and SIRT7 are located in the nucleus , SIRT2 is cytoplasmic ( but shuttles to and from the nucleus ) , whereas SIRT3 , SIRT4 and SIRT5 are found in the mitochondria [31] . Their different localizations are related with their cellular functions , for instance: nuclear SIRT1 , 6 and 7 are involved in transcription regulation [32] , DNA repair [33 , 34] and chromatin remodeling [35 , 36] , respectively; SIRT2 is a tubulin deacetylase that co-localizes with the cytoskeleton , but is imported to the nucleus , where it participates in cell cycle regulation [37]; and mitochondrial SIRT3 , 4 and 5 , participate in fatty acid metabolism [38] , amino acid metabolism [19] and the urea cycle , respectively [39] . By contrast , parasitic protozoa have fewer sirtuin homologues , but they have also been described as “pro-life” proteins due to their importance for the normal development and functioning of these cells . For instance , in the apicomplexan Plasmodium falciparum , Sir2A and B are involved in antigenic variation , an essential process for immune system evasion [40 , 41] . Trypanosomatids have two to three homologues ( Sir2rp1-3 ) with some of the members varying in their functions and subcellular localizations . T . brucei Sir2rp1 is the most characterized of the three enzymes , and is located in the nucleus and seems to be important for the protection against DNA damage [21 , 42] . TbSir2rp2 and TbSir2rp3 are localized in the mitochondria and little is known about their function , except that they are not essential for parasite survival [42] . Leishmania species also encode for three sirtuin enzymes , but most studies are focused on Sir2rp1 , that has a cytoplasmic localization . Sir2rp1 from L . infantum is required for the normal replication of amastigotes [43] . T . cruzi genomes sequenced so far , have only revealed two homologues , Sir2rp1 and Sir2rp3 , which are localized in the cytoplasm and the mitochondria , respectively [44 , 45] . Overexpression of either TcSir2rp1 or TcSir2rp3 showed them to interfere with epimastigote growth , differentiation into metacyclic trypomastigotes , infectivity of host cells and intracellular amastigote replication [44 , 45] . It should be noted that some contradictory findings have been reported by both research groups , differences that are likely related with the strategy of overexpression ( constitutive [44] versus inducible [45] ) . A clearer role of the participation of Sir2rp1 and Sir2rp3 in T . cruzi life cycle and infectivity by gene knockout studies awaits investigation . Here , we characterize for the first time the enzymatic NAD+-dependent deacetylase activity of TcSir2rp1 and show that as well as being hindered by classic inhibitors , bisnaphthalimidopropyl ( BNIP ) derivatives [46] , some showing specific “on target” trypanocidal activity against T . cruzi epimastigotes are also effective inhibitors . Furthermore , we demonstrate that these inhibitors are also active against the medically relevant stage of T . cruzi , intracellular amastigotes , suggesting that TcSir2rp1 may be a viable drug target to explore for the chemotherapeutic control of Chagas disease . In addition , the in vivo activity of the most potent inhibitor and selective compound towards T . cruzi was evaluated by bioluminescence imaging .
All experiments involving animals were carried out in accordance with the IBMC . INEB Animal Ethics Committees and the Portuguese National Authorities for Animal Health guidelines , according to the statements on the directive 2010/63/EU of the European Parliament and of the Council . BPC and ACdS are accredited for animal research ( Portuguese Veterinary Direction—DGAV , Ministerial Directive 113/2013 ) . DGAV approved the animal experimentation presented in this manuscript under the license number 0421/000/000/2013 , from 1i3S–Instituto de Investigação e Inovação em Saúde , Universidade do Porto , Porto , Portugal; 2IBMC-Instituto de Biologia Molecular e Celular , Parasite Disease Group , Porto , Portugal . Mouse myoblast C2C12 cell line ( ATCC ) and green monkey kidney epithelial Vero cells ( ATCC ) were cultured with high glucose DMEM , supplemented with 10% FBS ( Gibco ) , 25 mM HEPES , 2 mM glutamine , 100 U/mL penicillin , 100 U/mL streptomycin and maintained in a humid 5% CO2 atmosphere at 37°C . T . cruzi epimastigotes of the strain CL-Brener were maintained in the logarithmic phase of growth ( 1 . 0 x 106–1 . 0 x 107 cells / mL ) at 28°C in RTH/FCS medium ( RPMI 1640 supplemented with trypticase , haemin , HEPES , and 10% FBS ( Invitrogen ) ) . For the generation of transgenic cell lines , 15 μg of the constructs were digested overnight ( SpeI for the pTcINDEX construct; NotI for the knockout constructs ) , precipitated with sodium acetate / ethanol , and introduced into 1 . 5 x 107 epimastigote cells in the logarithmic phase of growth via transfection using an AMAXA Nucleofector II device ( program V-33 ) with the Human T-cell nucleofector kit ( Lonza ) . Parasites were suspended in fresh growth medium and incubated for 24-hrs to allow for the expression of drug selectable markers , prior to the addition of appropriate antibiotic ( Blasticidin S ( Thermo Fisher Scientific; 10 μg/mL ) ; Hygromycin B ( Formedium; 100 μg/mL ) ; and Puromycin ( Calbiochem; 5 μg/mL ) . Cells were then incubated for a further four- to six-weeks to allow for the selection of transgenic parasites and then genetically verified by PCR using ORF and drug selection primers . T . cruzi wild-type ( wt ) and luciferase-expressing ( Luc+ ) Y strain trypomastigotes ( made in house ) were maintained by in vitro infection of a monolayer of Vero cells and harvested by collection of the supernatant after 5 to 7 days . The trypomastigotes were used for the in vitro screening and to re-infect new monolayers for up to 10 cycles , after which point the cultures were rejected and restarted from a new frozen stock . Luminescent Y strain parasites were obtained by transfection as previously described [47] . Briefly , 10 μg of the overexpression plasmid pTREX [48] , containing the sequence for firefly luciferase , were transfected into logarithmic Y strain as described above . Twenty-four hours after transfection , 75 μg/mL of G418 was added to the culture and parasites were selected for 6 weeks . Epimastigotes cultures were allowed to enrich for metacyclic trypomastigotes for 3 weeks , after which the parasites were used to infect monolayers of Vero cells in the presence of G418 . The TcSir2rp1 open reading frame ( TriTryp accession number TcCLB . 508207 . 150 ) was amplified from CL-Brener genomic DNA using KOD polymerase ( Novagen ) . The resulting PCR product encoding the full-length ORF was cloned directly into the T . cruzi expression vector , pTcINDEX [49] , using NotI and BamHI restriction sites . To construct gene replacement cassettes , the 5’ and 3’ UTR regions adjacent to the TcSir2rp1 ORF ( ~500 bp ) were PCR amplified from genomic DNA ( S1 Fig for sequences ) , stitched together in a ‘knitting’ PCR reaction before ligating into the pGEM-5Zf ( + ) vector ( Promega ) via NotI restriction sites . Following this , the antibiotic resistance markers blasticidin-S ( BSD ) and puromycin N-acetyltransferase ( PAC ) were cloned into the stitch sequence BamHI / HindIII sites located between the adjoining UTRs . In order to validate the genetically modified T . cruzi cell-lines generated by previously described gene knock -in / -out experiments and further determine the relative gene expression levels of TcSir2rp1; a quantitative reverse transcription-PCR ( qRT-PCR ) study was performed . qPCR primers were designed , and initially normalized against two constitutively expressed , previously published and verified T . cruzi housekeeping / reference genes: histone H2B ( TcCLB . 511635 . 10 ) of the core nucleosome structure and glyceraldehyde-3-phosphate dehydrogenase ( TcCLB . 506885 . 413 ) . DNase treated RNA was used as the template for a two-step reverse transcription reaction to generate complementary DNA , which in turn was utilized in a quantitative PCR reaction with detection through SYBR Green dye fluorescence . The full coding sequence of TcSir2rp1 was PCR amplified using the primers 5’-CCATGGGAATGAATCAAGATAACGCCAACTTT-3’ and 5’-CTCGAGTTTTCGGTCTGTCTGTGTGTACATG-3’ and ligate into pET-28a ( Novagen ) using the NcoI and XhoI restriction sites . Recombinant expression of the C-terminal His-tagged protein was achieved in BL21 ( DE3 ) by induction with 0 . 5 mM IPTG ( isopropyl-β-D-thiogalactopyranoside ) overnight at 18°C . Pelleted cells were suspended in 500 mM NaCl , 20 mM Tris . HCl , pH 7 . 6 and disrupted by probe sonication and cleared by centrifugation . The Ni2+-NTA purified protein was buffer exchanged to PBS via a PD-10 desalting column ( GE Healthcare ) and aliquots were stored at -70°C prior to use for enzymatic assays . Protein was Western blotted using a rabbit polyclonal anti-HisTag antibody in a dilution of 1:1000 . For crystallography , the full length TcSir2rp1 ( M1-K359 ) was cloned into an in-house modified pET-28b vector , where the thrombin cleavage site is replaced by a tobacco etch virus ( TEV ) cleavage site and verified by sequencing . Expression of TcSir2rp1 was performed in BL21 ( DE3 ) in auto induction medium at 18°C overnight . Spun down bacterial cells were suspended in lysis buffer ( 50 mM Tris . HCl ( pH 7 . 5 ) , 500 mM NaCl , 1 mM dithiotreitol ( DTT ) , 2mM n-octyl β-D-glucopyranoside ( n-OG ) and 20 mM imidazole supplemented with protease inhibitor cocktail . Soluble tagged protein was recovered from the supernatant following centrifugation of the cell homogenate by affinity chromatography on His60 Ni Superflow Resin ( Clontech ) . Lysis buffer was used to wash the resin and protein was eluted by buffer A ( 50 mM Tris . HCl ( pH7 . 5 ) , 500 mM NaCl , 1 mM DTT , 2 mM n-OG and 500 mM imidazole ) . Recombinant TEV was used to remove the tag ( overnight incubation at 4°C ) . The second purification step consisted of anion exchange chromatography ( HiTrap Q HP 1 mL ) with an increasing NaCl gradient in 50 mM Tris . HCl ( pH 7 . 5 ) , 2 mM n-OG and 1 mM DTT . Protein eluted at around 150 mM NaCl . Fractions containing pure protein were pooled and the protein was concentrated to ~9 mg/mL . T . brucei Sir2rp1 ( accession number: AAX70528 . 1 ) tagged with a C-terminal hexahistidine was expressed and characterized essentiality as described previously [50] L . infantum Sir2rp1 was expressed and purified as previously described [22] . The deacetylase activity was measured using the commercial SIRT1/Sir2 Deacetylase Fluorometric Assay Kit ( CycLex , Japan ) according to the manufacturer’s recommendations . Briefly , fluorescence emission after digestion of deacetylated substrates by a lysine specific endopeptidase was measured in a fluorometric plate reader ( Synergy HT , BioTek ) with excitation and emission wavelengths set at 340 and 440 nm , respectively , every 30 seconds for 1 hour . The slope of the linear part of the reaction was calculated and used as readout of enzyme activity . Kinetic constants were determined with 0 . 5 μg of TcSir2rp1 . For the determination of the peptide substrate constants , NAD+ concentration was fixed at an excess of 2000 μM while the peptide substrate concentration was varied between 0 . 63 and 40 μM . For the NAD+ constants , the peptide substrate concentration was kept in excess at 40 μM while NAD+ concentration was varied from 15 . 63 to 2000 μM . Initial velocities were measured between 4 and 6 minutes where steady state conditions were assumed . Data was analysed with GraphPad Prism version 6 . 0 software using the built-in enzyme kinetics , Michaelis-Menten equation regression for Km and Vmax determinations . kcat was calculated from the total enzyme concentration [E]t according to the formula in kcat=Vmax[E]t Eq 1 Potential inhibitors , including newly synthesized BNIP derivatives , were tested using 200 μM NAD+ and 10 μM of peptide substrate as previously described [46] . The inhibition is expressed in percentage and was calculated as the ratio of velocity for the linear portion of the reaction , normalized with a no drug control and a reference drug control ( nicotinamide at 2 mM ) . The following compounds were synthesized as previously reported 1–11 [46 , 51] , 6b and 9a [52] , and 6c [53] . Compounds 1a-c , 6a , 7a 12 and 13 are newly synthetized in this study , see supporting information file for details . A variety of the bisnaphthalimidopropyl ( BNIPs ) derivatives , as well as the newly synthetized derivatives were evaluated as potential TcSir2rp1 inhibitors . The enzymatic reactions were performed using a commercially available CycLex SIRT1/Sir2 deacetylase fluorimetric kit ( CycLex Co . Ltd . , Nagano , Japan ) in the absence and presence of the various inhibitors , 200 μM NAD+ and 10 μM of peptide substrate as previously described [46] . The inhibition is expressed in percentage and was calculated as the ratio of velocity for the linear portion of the reaction , normalized with a no drug control and a reference drug control ( nicotinamide at 2 mM ) . Prior to crystallization , purified TcSir2rp1 was incubated with three molar equivalents of acetylated p53 peptide ( 372-KKGQSTSRHK-K[Ac]-LMFKTEG-389 ) . Co-crystallization experiments were carried out using the sitting drop vapor diffusion method in 96-well plates using an Innovadyne nanodrop robot . Crystals were grown in drops composed of 300 nL protein/p53 mixture and 300 nL crystallization condition ( 100 mM MES pH 6 . 5 and 1 . 6 M MgSO4 ) at 4°C . Crystals were flash-frozen in liquid nitrogen in crystallization condition supplemented with 25% glycerol . Diffraction data were collected on Proxima 2 beamline ( SOLEIL , Saclay , France ) on a Dectris Eiger 9M detector . Data were processed with xds [54] and CCP4 software package [55] . An initial TcSir2rp1 structure was obtained by molecular replacement with MrBump [56] using an “in-house” determined LiSir2rp1 structure ( pdb: 5Ol0 ) ( Ronin et al manuscript submitted ) . Refinement of the TcSir2rp1 model was done by iterative cycles of building with coot [57] and refinement with refmac [58] . The TcSir2rp1 3 . 5 Å structure consists of Chain A = TcSir2rp1: H10-I56 + Y63-T251 + C299-G332 + 1 Zinc ion and Chain B = Peptide p53: R379HKK ( Ac ) LMFK386 . ( SPG = P6322 , Rcryst = 21 . 13% / Rfree = 28 . 05% , global high B value ≈ 112 Å2 ) . The 3 . 5 Å crystal structure of TcSir2rp1 was used for docking of compounds 9 , 1a and 1b . Additionally , crystal structures of human SIRT2 were downloaded from the RCSB Protein Data Bank [59] ( 3zgo , [60]; 4rmj , [61]; 4rmg , [62] . AutoDock Tools version 1 . 5 . 6 was used to prepare the protein model for docking [62] . We used the protein models with and without ligands ( an acetylated fragment of p53 for TcSir2 , ADP and nicotinamide for 4rmj , NAD+ and SirtReal for 4rmg ) . Water molecules were removed , polar hydrogens added , charges for the protein calculated and a pdbqt file for each protein model produced . Coordinates for the docking search grid were also determined . Ligands were prepared for docking by drawing their structures in Chem3D , optimising mm2 , and writing the structures to pdb files . AutoDock Tools were then used to produce a PDBQT file for each ligand [63] . Docking was performed with AutoDock Vina [64] . PyMOL ( version 1 . 5 . 0 . 4 , [65] ) was used to visualize the molecular models . T . cruzi epimastigotes were seeded at 1 . 25 x 106 cells/mL in a 96-well plate in the presence of varying concentrations of test compounds . Following a 64-hour incubation , 15 μL of a resazurin solution ( Sigma-Aldrich; 1 . 1 mg/mL in PBS ) . Following an additional 6-hr incubation , cell viability was determined by measuring the fluorescence of each culture with a BioTek FLX800 Fluorescence Microplate Reader ( excitation and emission wavelengths of 540/535 nm and 590/610 nm , respectively ) and associated Gen5 Reader Control 2 . 0 software . EC50 values were determined by nonlinear regression to a sigmoidal dose-response curve using GraFit software ( Version 5 . 0 , Erithacus Software ) . Each assay was performed in at least triplicate in parallel to the known trypanocidal agent nifurtimox . For the evaluation of the activity against T . cruzi amastigotes , high-content screening ( HCS ) was designed . In summary , 100 μL of host C2C12 cells ( ATCC ) were seeded in clear bottom black 96-well plates at a density of 2 . 5 x 104 cells/mL ( 2 . 5 × 103 cells/well ) . After 24 hrs , 50 μL of Y strain tissue culture-derived trypomastigotes were used to infect the host cells at a density of 7 . 5 x 105/mL ( 3 . 75 x 104 parasites/well ) . The parasites were allowed to infect for a 24 hr period , after which compounds were added in a volume of 50 μL . Final concentration of DMSO in the assay did not exceeded 0 . 5% . After 72 hrs of compound incubation , the wells were fixed with 4% paraformaldehyde for 15 to 30 mins . The plates were then washed once with deionized water and stained with a solution of DAPI ( 3 μM ) for 1 hour . Plates were imaged in an INCell 2000 ( GE Healthcare ) high-content analyser by taking 16 pictures per well at 20 X objective amplification . Images were analysed with INCell Developer software ( GE Healthcare ) using the segmentation of host cell nuclei and parasite kinetoplast DNA . The measurement output used was the average number of parasites per cell calculated by the ratio of host cell nuclei/parasite kinetoplast DNA . To evaluate toxicity towards mammalian cell lines ( hepatocytes ( Seralab ) , neurons and MDCK cells ( ATCC ) ; a renal cell line from dogs ) , all the compounds were assayed with MTT [66] . Additionally , for cardiotoxicity evaluation , where toxic effects in cardiomyocytes could be affecting their function before being lethal , a hERG assay was performed according to the manufacturer’s protocol Predictor hERG Fluorescence Polarization Assay ( Invitrogen ) . Briefly , inhibition of hERG channel was analysed by competition of BNIPSpd and a high-affinity red fluorescent hERG channel ligand . BNIPSpd ( 9 ) was mixed with the ligand and membranes containing hERG channel , incubated for 2 hrs at RT and the fluorescence polarization was measured using 530 / 590 nm ex/em filters using the Synergy 2 plate reader ( Biotek ) . In vitro assays were also performed in primary cells ( hepatocytes and neurons ) to evaluate BNIPSpd ( 9 ) potential toxicity: ( a ) host cell nuclei counting after HOECHST staining , ( b ) viability by measuring the metabolism of WST-8 probe , ( c ) apoptosis induction through caspase 3/7 activation , ( d ) evaluation of mitochondrial dysfunction by measuring the membrane potential using TMRM probe , ( e ) membrane integrity by measuring the extracellular LDH , ( f ) DNA damage by histone H2AX phosphorylation ( exclusively in hepatocytes ) and ( g ) imaging of neurite outgrowth ( exclusively in neurons ) . The assays were performed 24 hrs after cell collection and isolation . For apoptosis , TMRM , HOECHST and WST-8 assays , cells were seeded at a density of 5000 cells/well in coated 96-well plates , incubated for 24 hrs with the compound and then stained during 1 hour with the following fluorescent probes: 5 μM CellEvent Caspase 3/7 Green Detection Reagent ( Invitrogen ) for measuring caspase 3/7 activation , 50 μM tetramethyl rhodamine methyl ester ( TMRM probe , Anaspec Inc . ) for measuring mitochondrial depolarization related to transient cytosolic Ca2+ signals , 5 μg/mL of HOECHST ( Sigma-Aldrich ) for nuclei detection and 10 μL/well of WST-8 reagent ( Sigma-Aldrich ) to detect viable cells . Then , absorbance at 450 nm was measured in order to analyse cell viability ( WST-8 ) . Cells were washed three times and analysed using the automatic fluorescence microscope BD Pathway 855 . Pictures were taken using a 20 X objective and 488/515 nm ex/em filters for CellEvent Caspase 3/7 reagent , 555/ 645 nm ex/em filters for TMRM and 380/460 nm ex/em filters for HOECHST . Data were analysed with AttoVision ( Becton Dickinson ) . The LDH assay was performed according to the manufacturer's protocol , Cytotoxicity Detection Kit LDH ( Roche ) . LDH release was measured in mU/L in culture media obtained from cells subjected to treatments for 24 hrs , by measuring the released LDH rate of oxidation of NADH to NAD+ at 340 nm using Synergy 2 from BioTech . For the neurite outgrowth assay , samples were washed with PBS , fixed using methanol for 10 mins at -20°C . The fixed samples were washed three times with PBS and permeabilized with PBS-Triton X-100 0 . 3% for 10 mins , washed three times and then blocked with BSA 0 . 5% in PBS for 30 mis . Anti-tubulin III antibody ( Sigma-Aldrich ) was added at 1/1000 dilution in blocking solution and incubated for 60 minutes at room temperature . After three washing steps , a secondary antibody Alexa 488 was added at 1/100 and incubated for 60 mins . After washing three times with PBS , pictures were taken using the BD Pathway 855 automated fluorescent microscope at 488/515 nm ex/em . Neurite average length was calculated using the neurite outgrowth module of AttoVision software ( Becton Dickinson ) . DNA damage was evaluated by washing cells with PBS and fixation using paraformaldehyde 3% in PBS for 15 mins . After which they were washed three times with PBS , permeabilized with PBS-Triton X-100 0 . 3% for 10 mins , washed again prior to blocking with PBS-BSA 0 . 5% for 30 mins . H2AX antibody ( Abcam ) was added at 1/400 in PBS-BSA 2 . 5% and incubated for 60 mins at RT . After three washing steps , the secondary antibody Alexa 488 was added 1/100 and incubated for 60 mins . After washing three times with PBS , pictures were taken using the BD Pathway 855 automated fluorescent microscope at 488/515 nm ex/em filters . To determine the DNA damage , intensity in the nuclei was analyzed using AttoVision software ( Becton Dickinson ) . Nimesulide ( 400 μM ) was included as a positive toxicity control , and the vehicle as a negative control . The relative percentage of deviation from the negative control was quantified and assigned with a number from 0 to 5 according to the following criteria: 0 ( 0–20% deviation ) , 1 ( 20–40% ) , 2 ( 40–60% ) , 3 ( 60–100% ) , 4 ( 100–1000% ) , or 5 ( >1000% deviation ) . The sum of these values was further ranked to create a combined injury criteria that varied from no injury ( 0 ) , low injury ( 1 to <5 ) , moderate injury ( ≥5-to <12 ) to high injury ( ≥ 12 ) . To evaluate BNIPSpd ( 9 ) efficacy in vivo , five- to six-week-old female BALB/c mice ( Charles River ) and infected with T . cruzi trypomastigotes expressing luciferase . Parasites were collected from the supernatants of a monolayer of Vero infected cells , washed and suspended in PBS-glucose 0 . 2% , and injected intraperitoneally ( 1 x 104 per mouse ) . After 7 days of infection , mice were treated with drugs for 4 consecutive days orally with ( benznidazole in 20% Kolliphor HS 15 ) or intravenously with ( BNIPSpd ( 9 ) in 10% DMSO ) at a dose of 100 or 5 mg/kg/day , respectively . Infection and treatment efficiency were evaluated following subcutaneous injection of 2 . 1 mg of luciferin and through live imaging using an IVIS Lumina LT ( Perkin Elmer ) . Images of the animals were analysed using the Living Image software ( Perkin Elmer ) . The pharmacokinetic profile of BNIPSpd was determined following intravenous injection of 5 mg/kg in BALB/c mice . The blood was collected from tail veins after 0 , 5 , 15 , 30 , 45 min , 1 , 3 , 24 , 48 and 72 h of drug administration . Blood concentrations of the drug were analysed by UHPLC- ESI-MS/MS .
In order to evaluate the role of TcSir2rp1 in T . cruzi survival and infectivity , a genetic validation approach was taken . Multiple attempts to knockout a single allele in epimastigotes ( with various different constructs/drug selection markers ) of TcSir2rp1 failed , indicating a possible lack of mono-allelic viability and thus essentiality . As such , a non-epitope tagged ectopic copy of the gene was successfully introduced into epimastigotes on a pTcINDEX over-expression vector , principally to introduce an ectopic copy of the gene to T . cruzi , and thus did not require any degree of control over the level gene expression . This was followed by successful sequential knockout of both wild-type alleles as confirmed by PCR ( S2 Fig ) , qRT-PCR , revealed that introduction of the pTcINDEX increased the RNA level for TcSir2rp1 by ~30-fold compared to wild-type . The growth and morphology of all of these genetically altered cell-lines were indistinguishable from wild-type cells . These data strongly suggest that TcSir2rp1 is an essential gene in epimastigote T . cruzi . Recombinant TcSir2rp1 was expressed and affinity purified as a C-terminal hexa-histidine tagged protein . The resultant protein was shown to be pure by SDS-PAGE and of the expected size ( ~40 kDa ) ( Fig 1A ) . Deacetylase activity was measured as described previously for LiSIR2rp1 [22] , in a coupled reaction based on the ability of lysylendopeptidase to digest deacetylated lysine , but not acetyllysine , residues , thereby releasing a quencher group from the molecule , allowing the fluorophore to emit fluorescence in a time-resolved manner . Deacetylase activity was dependent on NAD+ confirming TcSir2rp1 to be a NAD+-dependent deacetylase ( Fig 1B ) . Furthermore , when incubated with the histone deacetylase class I and II inhibitor trichostatin A ( TSA ) at 1 μM , TcSir2rp1 still displayed 77 ± 8% of activity ( Fig 1B ) . TSA is a poor inhibitor of sirtuins ( class III deacetylases ) , confirming the classification of the protein as a canonical Sir2 . The steady-state kinetic parameters ( Km , Vmax , kcat and kcat/Km values ) of deacetylase activity were determined ( Table 1 ) . By way of confirmation , the same kinetic parameters were also independently assessed using an electrophoretic mobility shift assay , performed as described previously [67] . Highly similar values ( Km’s of 38 . 7 ± 1 . 5 μM for NAD+ and 32 . 3 ± 6 . 0 μM for substrate ( TSPQPKK-Ac ) ) were obtained using the alternative microfluidic based assay confirming that the two different techniques generated essentially similar results when used to assess TcSir2rp1 NAD+-dependent deacetylase activity . The kinetic data obtained for the NAD+ cofactor is highly similar to that obtained for T . brucei Sir2rp1 whilst the substrate Km values reported here are 2-4-fold lower [68] . When compared to yeast Sir2 and the human SIRT2 , TcSir2rp1 also displays a high deacetylation activity as demonstrated by the catalytic efficiency ( Km/kcat ) constants calculated for the homologue enzymes [69 , 70] . Catalytic efficiencies are the most relevant constant in physiological conditions since they define the rate of the reaction when substrate concentrations are not at saturating levels , as for most cellular enzymatic reactions [70] . The above results were used to establish a standard screening assay ( 10 μM Ac-peptide and 200 μM NAD+ ) . Nicotinamide was tested in order to evaluate if small molecules could inhibit the enzymatic activity described . Nicotinamide is a classic non-competitive inhibitor of sirtuins that has been used to characterize several enzymes [22 , 71] . A dose-response curve of nicotinamide inhibition of TcSir2rp1 indicates an IC50 of 456 ± 44 μM ( Fig 1E ) . This value is significantly higher when compared to the IC50 determined for other sirtuins like the human SIRT1 ( 118 . 3 ± 23 . 6 μM ) , the Plasmodium falciparum Sir2A ( 51 . 2 ± 3 . 0 μM ) or the LiSir2rp1 ( 39 . 4 ± 5 . 0 μM ) previously described [46 , 72] . When a concentration of 200 μM of nicotinamide was tested against TbSir2rp1 , LiSir2rp1 and the human SIRT1 , TcSir2rp1 only had a mild inhibition of 34 ± 1% , whereas the inhibition was 70 ± 5% , 82 ± 0% , 66 ± 10% for TbSir2rp1 , LiSir2rp1 and human SIRT1 , respectively ( Fig 1F ) . Some studies have previously reported the anti-parasitic activity of nicotinamide on T . cruzi , but , to our best knowledge , this is the first report that proves the molecular inhibition of the parasite sirtuin by nicotinamide [45 , 73] . We next evaluated the enzymatic inhibition of TcSir2rp1 by a class of experimental compounds previously characterized as inhibitors of the orthologue enzyme of L . infantum [46] as well as the newly synthetized derivatives . BNIP derivatives are constituted by two naphthalimidopropyl groups separated by a linker that varies in length and functional groups , and have been characterized as NAD+-competitive inhibitors of LiSir2rp1 [74] . In this study , an additional set of compounds was synthetized with the objective of improving cellular target binding by including heteroatoms in the alkyl chain connecting the two naphthalimidopropyl groups , as well as by introducing cyclic structures in the linker ( Fig 2 ) . The results are summarized in Fig 3A and indicate the percentage of inhibition of the NAD+-dependent deacetylase activity for a single dose concentration of 10 μM . Of all the compounds tested , compounds 2 , 3 , 8 , 9 , 13 and 9a had a statistically significant inhibition of the enzyme . In order to verify a concentration dependent effect on enzymatic inhibition , compounds 2 , 8 , 9 and 9a were additionally tested at 50 μM , whereas compound 3 and 13 were tested at 20 μM due to low solubility . Inhibition percentages for this higher concentration were 55 ± 4% , 34 ± 11% , 31 ± 9% , 72 ± 6% , 48 ± 6 and 29 ± 4% for compounds 2 , 3 , 8 , 9 , 13 and 9a , respectively ( Fig 3B ) . Only compound 9 demonstrated a statistically significant dose-dependent increase in inhibition ( 1-way ANOVA , p value <0 . 05 ) . In fact , compound 9 was one of the most active compounds against TcSir2rp1 , with a 10 μM dose inhibiting almost 50% of the NAD+-dependent deacetylase activity , a concentration 45 times lower than the necessary for nicotinamide to achieve the same level of inhibition . Compound 9 is a derivative of compound 5 by substitution of one of the carbons in the linker chain by a nitrogen atom . Since compound 5 was not active against the enzyme , it is possible that the substitution of the nitrogen allows the establishment of additional molecular interactions with the enzyme and/or modifies the rigidity of the linker chain resulting in higher inhibition . Compound 9a is also a derivative from compound 5 with eight atoms in the linker chain separating the two naphthalimidopropyl groups , but with two oxygen atoms in substitution of two carbons . This modification also seems to increase the inhibitory activity against TcSir2rp1 ( 28 ± 13% ) , but it is not as marked as for compound 9 . Although a clear correlation between the length of the linking chain and enzymatic inhibition was established for the L . infantum orthologue [46] , the same was not verified with TcSir2rp1 . Compounds with 5 and 6 carbons in the linker chain also displayed some inhibitory activity against the enzyme ( compound 2 , 42 ± 13% and 3 , 31 ± 8% , respectively ) , but when tested at a higher concentration , the increase was not statistically significant . Compound 3 was only tested at 20 μM due to poor solubility . In general , the most active inhibitors of TcSir2rp1 were also selective towards the T . cruzi enzyme , as demonstrated by inhibition of a human sirtuin homologue , hSIRT1 . In addition to hSIRT1 inhibition previously described for compounds 1–11 [46] , compounds 12-9a were also tested at 10 μM against this enzyme ( S1 Table , see supporting information ) . Compound 6b was the most active against the human homologue enzyme ( 44 ± 8% of inhibition ) , but was a weak inhibitor of TcSir2rp1 ( 8 ± 5% ) at the same concentration of 10 μM . Compound 13 also had a reduced activity against hSIRT1 ( 15 ± 4% ) , but was a much stronger inhibitor of the T . cruzi enzyme , with an inhibition at least 3-fold higher ( 50 ± 8% ) . As the deacetylase activity was inhibited by some BNIP derivatives , they were also tested against T . cruzi epimastigotes ( Table 2 ) . Some of these compounds ( 3 , 6 and 9 ) showed moderate activity ( EC50 ) , while compounds 7 and 2 were just as potent as nifurtimox , the positive control . To investigate if the BNIP derivatives’ trypanocidal activity is correlated with Sir2rp1 inhibition , compounds were also tested against the pTcINDEX overexpressing TcSir2rp1 cell-line , described earlier . As expected the nifurtimox EC50 did not alter significantly compared to the wild-type cell line , nor did that of compound 7 . However , the EC50 values of the other 4 BNIP derivatives did increase 2 . 8–4 . 8 fold , strongly suggesting that these BNIP derivatives were trypanocidal , because they were “on-target” , i . e . inhibiting TcSir2rp1and causing cell death . Once demonstrated that the deacetylase enzymatic activity is inhibited by some BNIP derivatives and they are also on-target against epimastigotes , the compounds were next tested using an in vitro HCS assay on the medically relevant intracellular form of amastigotes ( S3 Fig ) . In this assay , the compounds are incubated for 72 hours with cells previously infected with wild-type trypomastigotes for 24 hours . The readout was done by comparing the average number of amastigotes per cell that developed by the end of the incubation period . The quality of the assay was statistically evaluated by the calculation of the z-factor [75] , that represents the reliability of the assay in distinguishing positive controls confidently from negative controls . Varying from 0 to 1 , z-factor values higher than 0 . 5 are considered acceptable for drug screening . The z-factor obtained was 0 . 68 , validating the use of the assay ( S3B Fig ) . The reference drug benznidazole was assayed for quality control and showed an EC50 of 1 . 23 ± 0 . 30 μM , in accordance with previous studies [76] . In order to prioritize compounds , all the BNIP derivatives were initially screened at a single dose of 2 . 5 μM ( Fig 4A ) , where all of the compounds were fully soluble . The compounds that presented high anti-parasitic activity ( >40% ) and low toxicity [high cell ratio , defined as ( average number of cells for the compound / average number of cell for the DMSO 0 , 5% control ) , >70%] , represented in a blue box , were selected for dose-response curve analysis . A set of compounds that was not toxic at 2 . 5 μM ( green box ) was further tested at 10 μM . It is worth noting that compound 2 , which showed moderate inhibition of TcSir2rp1 and good on-target trypanocidal activity against T . cruzi epimastigote cells ( Table 2 ) , failed this hurdle due to high cytotoxicity issues ( Fig 5A ) . Again , the compounds that had high anti-parasitic activity and low toxicity were selected for dose-response analysis ( blue box , Fig 4B ) . Compounds 10 and 6c were not tested at 10 μM due to poor solubility . Compounds 1 , 6 , 8 , 9 , 12 , 13 , 1a , 1b , 6a , 7a and 9a ( Fig 4C and Table 3 ) were assayed by dose-response curves in the same conditions of the primary screening . Compound 9 , the strongest inhibitor of TcSir2rp1 and for on-target killing of epimastigotes ( Table 2 ) , was also active against T . cruzi amastigotes , presenting an EC50 of 2 . 84 ± 0 . 30 μM , in the range of the reference drug benznidazole ( 1 . 23 ± 0 . 30 μM ) . In addition , it was the least toxic of the compounds analysed by dose-response curve , with a CC50 of 24 . 47 ± 0 . 45 μM , resulting in a selectivity index ( SI ) of 8 . 8 units . Compound 9a , an analogue of 9 with two oxygen atoms in the linking chain instead of one nitrogen atom , was also active and presented a similar EC50 of 3 . 43 ± 0 . 57 μM . However , it was less selective than compound 9 ( 2 . 1 units ) . Compound 5 which is also the same length between the naphthalimidopropyl groups , but only made of methylene moieties , has very little if any activity against either the enzyme or amastigotes . Compound 13 , also an inhibitor of TcSir2rp1 ( 50 ± 8% ) , was the most active compound against T . cruzi amastigotes , with an EC50 of 0 . 59 ± 0 . 23 μM . It was not possible to calculate the CC50 since the highest concentration tested allowed by the solubility of the compound ( 2 . 5 μM ) did not reduce the cell ratio below 50% , making the SI of the compound at least 4 . 2 . The second most potent hit , compound 1b , is a derivative of compound 1 where a cyclohexane moiety was introduced as a spacer between the naphthalimidopropyl groups . The synthesis of this compound resulted in two isomers with a significant difference in their anti-parasitic activity , where the trans isomer ( compound 1b ) presents an EC50 of 0 . 78 ± 0 . 12 μM and the cis isomer ( compound 1a ) an EC50 of 6 . 09 ± 0 . 14 μM . Such a difference suggests this particular stereoisomeric configuration improves anti-parasitic activity . Activity of nicotinamide in the amastigote assay was also assessed , but there was no activity detected up to a concentration of 2000 μM . Compound 9 , BNIPSpd , showed the best enzymatic inhibition , low EC50s against epimastigotes ( potentially on target ) and the medically relevant amastigote stages of the parasite and some degree of selectivity . Based upon these parameters it was decided to follow up its characterization . The compound toxicity was further evaluated against a panel of primary cells to determine tissue specific toxic effects . The results of cytotoxicity against neurons , hepatocytes and MDCK cells , and the cardiotoxicity determination by hERG predictor method are summarized in Table 4 . As shown , BNIPSpd ( 9 ) presents hepatotoxicity and neurotoxicity values very similar to its efficacy concentration ( SI of 1 . 1 and 1 . 4 respectively ) . On the other hand , it seems not to be nephrotoxic ( SI of 35 . 2 ) or cardiotoxic ( hERG channel 99 . 6% functional ) . The toxicity values are higher than the corresponding ones for the C2C12 cells ( Table 3 , SI of 8 . 8 ) , but this is not surprising given the fact that the primary cells are usually more sensitive . Furthermore , the toxic effects at the cellular level were evaluated and quantified by a set of in vitro assays that include mitochondrial dysfunction , membrane integrity , DNA damage , apoptosis and neurite outgrowth on hepatocytes and neurons ( Fig 5 ) . This analysis could not be performed in the MDCK cell line due to the high amounts of DMSO to test the required BNIPSpd ( 9 ) concentration being toxic for these cells . Nimesulide , an approved and widely used anti-inflammatory drug , was included at a concentration of 400 μM as a control of toxicity [77–79] . The results are presented in Fig 5 . In summary , BNIPSpd ( 9 ) showed a low injury score for hepatocytes , with no toxicity up to 50 μM and moderate injury score for neurons , where toxicity was most significant from 12 . 5 μM . These values correspond to the range within the efficacy concentration for this compound ( 2 . 84 μM ) . Despite showing a mild injury score for neurons , it should be kept in mind that further analysis needs to be done in order to determine whether the compound is capable of crossing the blood-brain barrier or not . Indeed , a compound could be very neurotoxic but never been able to reach the neurons . The main mechanisms responsible for the neurotoxicity are the caspase 3/7 activation , a measure of apoptosis and disruption of membrane integrity as assessed by LDH release . As the concentration of drug increases other toxicity mechanisms like mitochondrial dysfunction and neurite outgrowth start to contribute to the overall cytotoxicity score . In conclusion , up to concentrations of 10 μM , BNIPSpd ( 9 ) appears to be safe for further evaluation in an animal model for Chagas disease . Finally , BNIPSpd ( 9 ) in vivo activity was characterized in a murine model of T . cruzi infection and using bioluminescence imaging . A bioluminescent T . cruzi Y strain ( Luc+ ) was obtained like previously described [47] and the in vitro limit of detection in the bioimaging equipment IVIS Lumina LT was determined to be 104 parasites ( t-test , p-value < 0 . 05 ) ( S4A Fig ) . Different numbers of Luc+ trypomastigotes were tested in order to assess which condition yielded the best readout ( S4B Fig ) . When injected intraperitoneally with 104 parasites , the mice developed an infection at first mostly located around the inoculation site ( day 7 ) but that later spread to the whole body ( S4B Fig ) . Similar observations were made for the 105 and 106 inoculates but 104 parasites yielded a higher and more reproducible increase of the whole body bioluminescent signal . A regimen of 5 mg/kg/day of BNIPSpd ( 9 ) was administered intravenously for 4 consecutive days , while a positive control with a dose of 100 mg/kg/day of benznidazole per os and the respective vehicle controls were also performed ( S5A Fig ) . Unfortunately , BNIPSpd ( 9 ) did not exhibit any in vivo activity , since bioimaging analysis demonstrates the infection progressed as in the animals treated with the vehicle only , which showed on average an increase in the luminescence signal by 40-fold . By comparison , animals treated with benznidazole at 100 mg/kg/day displayed , on average , a 700-fold reduction in the bioluminescence signal compared to mice treated with vehicle only ( S5A Fig ) . To investigate whether the lack of anti-parasitic activity in vivo could be due to the pharmacokinetics profile of BNIPSpd ( 9 ) we proceeded to quantify the compound in the blood over time . Analysis of the snapshot pharmacokinetics profile ( S5B Fig ) may explain the lack of in vivo activity of BNIPSpd ( 9 ) . For the first 30 mins after intravenous injection , it never reaches a concentration equivalent to the in vitro EC50 for T . cruzi amastigotes . The highest plasma concentration detected is 1 . 70 μM at the 5 min time-point , with the compound below the level of detection after the 3 h time-point and up to 72 h post administration ( lower limit of quantification is 8 . 1 nM ) . This pharmacokinetic profile indicates that a concentration of drug able to reach and clear parasitized tissues is not achieved in this mouse model . To understand how BNIPSpd ( 9 ) could interact with TcSIR2rp1 , we tried to solve the X-ray structure of the protein . Unfortunately , despite extensive crystallization optimization attempts , crystals with only relative low resolution limit ( 3 . 5Å ) could be obtained . Nevertheless , this allowed a preliminary TcSIR2rp1 model in complex with p53 peptide substrate to be generated . The final 3 . 5Å TcSir2rp1 structure ( Fig 6A ) shows clearly defined main chain ( Cα ) , but only poor density is observed for side chains in some regions , thus , these side chains were removed . TcSIR2rp1 global fold is , as expected , similar to other sirtuins including the human SIRT2 ( S6 Fig ) with a large Rossmann-fold domain ( composed of 6 parallel β-strands , sandwiched between 2 layers of α-helices ) and a small Zinc binding domain . The substrate peptide p53 is bound to the cleft between the small and the large domains ( Fig 6A ) , the co-factor binding loop is ordered in an open conformation similar to hsSIRT2 apo structure [48] or yeast Hst2 in complex with p53 peptide and NAD [60] . D252-I298 is missing in the TcSIR2rp1 final model due to a lack of density , which suggests high flexibility and/or potential disorder in this region . Nevertheless it should not impact upon our docking experiments as not involved in active site delineation and region , located at the bottom part of the Rossmann fold domain ( diametrically opposed to Zinc ion along an axis going through p53 peptide ) . Docking studies were subsequently performed with the TcSir2rp1 structure , where the p53 peptide substrate was removed and compound 9 was used as a possible ligand . Several conformations of compound 9 were found in a putative ligand binding site close to the NAD binding site ( Fig 6B and 6C ) , all with very similar binding affinities 9 ± 0 . 3 kcal/mol . To investigate if this possible interaction between compound 9 and TcSir2rp1 was specific , several human SIRT2 structures were also subjected to similar docking studies , ( S7 and S8 Figs ) , but they were found to have similar binding affinities to those observed for TcSir2rp1 . Likewise , the two isomers , compound 1a and 1b , which showed distinct difference in EC50 activities ( 6 . 09 and 0 . 78 μM respectively ) were also docked with TcSir2rp1 ( S9 and S10 Figs ) , but they very similar binding affinities to each other .
Sirtuins have long been proposed as interesting targets to treat parasitic diseases [80 , 81] , but only recently have these proteins been characterized in T . cruzi [44 , 45] . Even though the localization , expression and some of the important functions of T . cruzi sirtuins have been described , to date no biochemical characterization has been performed . Our results demonstrate that the annotated coding sequence of TcSir2rp1 encodes for an essential ( as genetically validated here ) canonical sirtuin that does not display deacetylation activity in the absence of NAD+ and is mostly insensitive to TSA . Kinetic data obtained for TcSir2rp1 is highly similar to the values previously described for TbSir2rp1 [67] . Despite moderate differences in primary sequence ( 61% ) , a close analysis of the amino acids adjacent to the essential histidine ( H142 ) at the catalytic site demonstrates a 100% similarity of the four amino acids upstream and downstream , which likely participate and/or help shape the catalytic site , and thus may explain the similarity observed . Nonetheless , it should be noted that distant amino acids in terms of primary structure may have an influence on substrate binding and enzymatic activity , as demonstrated in the case of PfSir2 in which the C- and N-terminal removal led to modifications in the enzyme kinetic constants and sensitivity to nicotinamide [72] . An essential step in the evaluation of a drug target is the proof of druggability . Even if a gene product is deemed essential for a given organism , if the activity or function mediated by this protein is not amenable to modulation by small molecule inhibitors , then it is not considered to be a good drug target . To this end , it was sought to modulate TcSir2rp1 with a classic inhibitor of this protein family , nicotinamide [82] . The enzymatic activity described was indeed inhibited by nicotinamide , although in concentrations higher than those previously reported for other members of the family . In particular , the IC50 for LiSir2rp1 and hSIRT1 is 11 and 4-fold lower , respectively , than the one reported here for TcSir2rp1 [46] . Another observation that supports the resistance to nicotinamide is the fact that even when 2 mM of NAD+ is used in reactions for the determination of the kinetic constants for NAD+ , there is no decrease in enzymatic activity unlike previously reported for other enzymes [22] . Such inhibition might be expected since nicotinamide is an endogenous product of the deacetylation reaction . Nicotinamide binds to a distinct C pocket that appears to be conserved among the sirtuins that have been structurally elucidated by co-crystallography [71] . The study of the presence and conservation of this pocket in TcSir2rp1 may shed some light on the observed nicotinamide resistance . Although nicotinamide has some anti-parasitic activity against Plasmodium [83] , Leishmania [84] , T . brucei [85] and also T . cruzi [73] , no confirmation of a sirtuin-mediated mechanism has been clearly established to date . In this study , no amastigote anti-parasitic activity was detected for nicotinamide in concentrations up to 2 mM . The search for additional inhibitors led to the identification of some BNIP derivatives as stronger inhibitors of TcSir2rp1 . This class of compounds was previously described to be active at 10 μM , with few exceptions , against LiSir2rp1 [74] . However , no correlation has been established between the most active compounds for L . infantum and T . cruzi enzymes , suggesting that both enzymes might be targeted differentially by some of the derivatives [46] . In contrast to LiSir2rp1 inhibition , no structure-activity relationship regarding the length of the alkyl chain linking the two naphthalimidopropyl groups was observed for TcSir2rp1 [75] . Although a linker of eight carbons had no activity at 10 μM ( compound 5 ) , when some of the carbons are substituted by heteroatoms like nitrogen ( compound 9 ) and oxygen ( compound 9a ) , the activity towards TcSir2rp1 increased significantly , suggesting that the rigidity of certain conformations and/or the ability to establish additional molecular interactions in this part of the molecule , increase inhibitory activity . More importantly , some enzymatic inhibitors also demonstrated a strong anti-parasitic activity against T . cruzi epimastigotes , which for some was shown to be on-target , due to the increase in EC50 for the overexpressing cell-line . Moreover , amastigotes , the clinically relevant form responsible for infection persistence in humans , were shown to be sensitive to some of the BNIP derivatives . Differences observed between the enzymatic inhibition and the activity in the parasite may be due to distinct interactions between BNIPs and TcSir2rp1 , and/or , possibly , by the contribution of a multi-target mechanism as well as the relative efficacy of entry into the host cell and then into the parasitophorous vacuole . The majority of the compounds demonstrated low selectivity for the parasite and toxicity for host cells , as indicated by their selectivity indexes . A possible explanation is the characteristic DNA intercalation of BNIP derivatives that were originally developed as anti-cancer agents [86 , 87] . While this might explain some of the toxic effects observed for host cells , it should not be excluded that DNA intercalation may be contributing to their reported anti-parasitic activity , especially since trypanosomes are susceptible to intercalating agents [88] . Analysis of TcSir2rp1 by Wregex and cNLS Mapper , bio-computational tools that identify nuclear export signals ( NES ) and nuclear localization signals ( NLS ) , respectively , indicate the presence of non-canonical NES/NLS in the sequence of this sirtuin [89 , 90] . Whether TcSir2rp1 does shuttle to and from the nucleus during specific times of the T . cruzi cell cycle , possibly to perform DNA damage repair like the T . brucei orthologue , remains to be reported . [42] , In fact , if TcSir2rp1 also has a role in DNA repair both sirtuin inhibition and DNA intercalation may be synergistically contributing to the anti-parasitic activity . Activity towards other molecular targets should also not be excluded , especially considering that T . cruzi has another sirtuin , TcSir2rp3 , that has also been described as having important functions in different cellular processes like metacyclogenesis , epimastigote growth and host-cell infectivity and replication [44 , 45] . The docking studies also suggests that compound 9 is unable to distinguish between TcSir2rp1 and human SIRT2 proteins and is able to bind in multiple conformations either in the presence or absence of p53 substrate . This is likely due to the intrinsic hydrophobicity of sirtuin proteins and the hydrophobic nature of the two bisnapthalmidopropyl-moieties and the highly flexible linker between them in the BNIP derivatives . Since BNIPSpd ( 9 ) met some of the biological criteria recommended for T . cruzi drug discovery follow-up , it was decided to further evaluate it as a possible lead compound [91] . The toxicity determination showed that although it has low SI values ( around 1 for hepatocytes and neurons ) , the toxicity values in the μM range could be considered a promising starting point for further optimization . Moreover , the cardiotoxicity evaluation showed that cardiomyocyte function does not seem to be compromised . On the other hand , a more in-depth analysis of the toxicity by HCS showed that there is very limited injury to hepatocytes and neurons . BNIPSpd ( 9 ) showed no hepatotoxicity up to 12 . 5 μM . At 50 μM there is a low injury score mainly due to caspase 3/7 activation ( although the viability -WST8- is also altered ) . In regard to neurotoxicity , as expected , the neurons start to suffer toxicity at lower concentrations; neurons showed no effects up to 0 . 7 μM but the injury score starts to rise in a dose dependent manner up to 12 . 5 μM . At higher concentrations ( 50 μM ) the injury score is maintained at high levels . The mechanism of toxicity in neurons and hepatocytes seems to be the same , as the main parameter was caspase 3/7 activation followed by membrane integrity . At higher concentrations , all the parameters tested were affected . More assays need to be performed to evaluate the capacity of BNIPSpd ( 9 ) to cross the blood-brain barrier . In addition , BNIPSpd ( 9 ) showed decent aqueous solubility ( up to 70% ) and gastric stability ( 100% ) . Assessment of efficacy of BNIPSpd ( 9 ) in mice infected with bioluminescent T . cruzi parasites by live imaging revealed lack of in vivo activity . The results may be explained by the pharmacokinetic profile of the compound , whose concentration in the blood always remained below the in vitro EC50 for amastigotes . The compound was no longer detected 3 hours after administration , even though the drug is administered by intravenous injection , the most bioavailable route . An explanation is that most of the compound is rapidly metabolized in the organism to an inactive form or removed from circulation . Medicinal chemistry modifications of the compound , as well as alternative formulations may improve activity and pharmacokinetics in the host organism .
|
Trypanosoma cruzi is a protozoan parasite belonging to the Kinetoplastida class responsible for Chagas disease , a neglected tropical illness that affects an estimated 6 to 8 million people in Latin America and some Southern regions of the USA , with another 25 million at risk of acquiring the disease and a death toll of 12 , 000 every year . Commonly transmitted from the feces of the kissing bug , the disease is characterized by a nearly asymptomatic acute phase but a problematic chronic phase in which 20–30% of individuals develop serious cardiac and/or intestinal problems . The therapies currently in use were introduced more than forty years ago , and there are important concerns about adverse effects and lower effectiveness with disease progression . There is , therefore , an urgent need to find better alternatives . In this study , we evaluate the potential of a Trypanosoma cruzi sirtuin protein as a novel drug target and its inhibition by novel members of a known class of sirtuin compound inhibitors .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"alkaloids",
"chemical",
"compounds",
"pathology",
"and",
"laboratory",
"medicine",
"enzymology",
"microbiology",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"protozoan",
"life",
"cycles",
"toxicology",
"neuroscience",
"developmental",
"biology",
"toxicity",
"protozoans",
"epimastigotes",
"enzyme",
"inhibitors",
"animal",
"cells",
"life",
"cycles",
"chemistry",
"nicotine",
"amastigotes",
"trypanosoma",
"cruzi",
"biochemistry",
"cellular",
"neuroscience",
"trypanosoma",
"eukaryota",
"cell",
"biology",
"neurons",
"biology",
"and",
"life",
"sciences",
"protozoology",
"physical",
"sciences",
"cellular",
"types",
"organisms"
] |
2018
|
Inhibitors of Trypanosoma cruzi Sir2 related protein 1 as potential drugs against Chagas disease
|
Upstream events that trigger initiation of cell division , at a point called START in yeast , determine the overall rates of cell proliferation . The identity and complete sequence of those events remain unknown . Previous studies relied mainly on cell size changes to identify systematically genes required for the timely completion of START . Here , we evaluated panels of non-essential single gene deletion strains for altered DNA content by flow cytometry . This analysis revealed that most gene deletions that altered cell cycle progression did not change cell size . Our results highlight a strong requirement for ribosomal biogenesis and protein synthesis for initiation of cell division . We also identified numerous factors that have not been previously implicated in cell cycle control mechanisms . We found that CBS , which catalyzes the synthesis of cystathionine from serine and homocysteine , advances START in two ways: by promoting cell growth , which requires CBS's catalytic activity , and by a separate function , which does not require CBS's catalytic activity . CBS defects cause disease in humans , and in animals CBS has vital , non-catalytic , unknown roles . Hence , our results may be relevant for human biology . Taken together , these findings significantly expand the range of factors required for the timely initiation of cell division . The systematic identification of non-essential regulators of cell division we describe will be a valuable resource for analysis of cell cycle progression in yeast and other organisms .
Understanding cell division requires knowing not only how , but also what determines when cells divide . Previous studies identified several components of the machinery that drives the cell cycle . However , it is not clear how cellular pathways impinge on the cell division machinery to initiate cell division . This is a critical gap in our understanding , since this process governs overall proliferation: once cells initiate their division , they are committed to completing it . In proliferating cells , the G1 phase of any given cell cycle lasts from the end of the previous mitosis until the beginning of DNA synthesis . In unfavorable growth conditions , eukaryotic cells typically stay longer in G1 , delaying initiation of DNA replication [1]–[7] . Subsequent cell cycle transitions , culminating with mitosis , are less sensitive to growth limitations , and their timing does not vary greatly , even if growth conditions worsen . Hence , differences in the length of the G1 phase account for most of the differences in total cell cycle , or generation times , between the same cells growing in different media , or among different cells of the same organism . Such fundamental observations support the notion that eukaryotic cells commit to a new round of cell division at some point in late G1 [3] , [4] , [8] , [9] . Budding yeast cells also evaluate their “growth” in late G1 at a point called START , before DNA synthesis in S phase [1] . In favorable growth conditions , and in the absence of mating pheromones ( for haploids ) , or meiotic inducers ( for diploids ) , cells pass through START [1] . Passage through START and commitment to cell division precedes a large transcriptional program and additional events that lead to initiation of DNA replication [10]–[12] . The lack of a detailed view of upstream regulatory networks that govern the timing of START in the yeast Saccharomyces cerevisiae is surprising , given the rich history of the field . The classic cdc screen identified factors essential for START , such as Cdc28p [1] , the main yeast cyclin-dependent kinase ( Cdk ) . However , the cdc screen did not target nonessential regulators , such as the cyclin regulatory subunits of Cdc28p [13] . Other efforts relied on gene-specific suppression [14]–[18] or sensitivity to mating pheromones [19] , [20] . By far , however , most approaches to identify regulators of START interrogated cell size . Almost half a century ago , a relationship between the size or mass of a cell and the timing of initiation of DNA replication was described from bacterial [21] , to mammalian cells [22] . Indeed , a newborn budding yeast cell is smaller than its mother is , and it will not initiate cell division until it becomes bigger [1] . Thus , it appears that there is a critical size threshold for START completion in yeast . Based on this concept of a critical size , the question of “when do cells divide ? ” was reduced to “what size are cells when they divide ? ” Hence , several screens for regulators of START interrogated cell size [23]–[27] . In fact , systematic , genome-wide approaches to find genes required for the correct timing of START relied solely on cell size changes [23] , [24] . Any gene deletion that alters the length of the G1 phase relative to the rest of the phases of the cell cycle will alter the DNA content profile . Thus , the DNA content of a population reports on the relative length of the G1 phase directly , discerning cells with unreplicated genome . In yeast , DNA content analyses measured the effects of gene over-expression on cell cycle progression [28] , [29] , or cycle arrest when essential genes were turned-off [30] . However , the yeast single-gene deletion collections have not been evaluated with this method . To assess cell cycle progression more directly , we evaluated the yeast deletion collection of nonessential genes for altered DNA content , by flow cytometry . We found that most gene deletions that altered cell cycle progression did not change cell size . Our results suggest that evaluating the length of the G1 phase of the cell cycle , instead of cell size , provides a much more accurate view of the contribution of individual gene products to the timing of START and commitment to cell division . We also documented a strong requirement for ribosomal biogenesis for initiation of cell division , and identified numerous factors that have not been implicated previously in cell cycle control mechanisms . One such factor is the metabolic enzyme cystathionine-β-synthase ( CBS; Cys4p in yeast ) . We discovered a novel , non-catalytic role of CBS , in accelerating START . Taken together , the data we present here substantially expand the range of factors that affect initiation of cell division . We discuss the significance of our finding that most gene deletions that change the length of the G1 phase do not alter cell size , in the context of models that center on the role of cell size at START .
We measured the DNA content during exponential growth in rich media ( YPD-2% dextrose [31] , see Methods ) , for several reasons: First , exponential growth in liquid media affords much greater reproducibility [32] . Second , for the haploid deletion strains , cell size measurements during the same growth conditions are available [23] . Third , fitness data during growth in the same rich media are available [33] , providing another parameter for interpreting our findings . We used the homozygous diploid deletion panel to query the nonessential genes , to minimize the effects of aneuploidy found in a substantial portion of haploid deletion strains [34] . We evaluated strains individually ( Figure 1 ) . We quantified each sample in an automated manner , recording the percentage of cells with unreplicated genome ( %G1 , see Methods ) . We did not quantify complex profiles ( e . g . , due to cell separation defects , see Figure S1 ) , and we excluded these strains from further analyses . At the beginning and end of most batches of strains , we measured the reference wild type strain ( BY4743 ) , which was cultured and processed along with the deletion strains . To identify strains with altered cell cycle , we compared the frequency distribution of the deletion strains against a normal distribution fit of the wild type ( 31 . 17%±5 . 20 , n = 250 ) samples ( Figure 2 ) . Deletion strains that had a %G1 greater or less than two standard deviations of the wild-type distribution were considered to differ significantly from wild type , and we evaluated them further ( see Methods ) . From all strains analyzed successfully ( n = 4 , 342; Dataset S1 ) , 152 were in the “High G1” group , but only 16 were in the “Low G1” group . Hence , the majority of gene deletions that affect cell cycle progression lead to a G1 delay ( Figure 2 ) . We expect that additional gene deletions affect cell cycle progression , but were not included in the “High G1” or “Low G1” groups , for at least two reasons: experimental error; and imposition of restrictive cutoffs ( >41 . 57%G1 for the “High G1” group and <20 . 77%G1 for the “Low G1” group ) . An example of the latter is whi5Δ cells , which lack an inhibitor of START [35] , [36] . whi5Δ cells clearly had “Low G1” DNA content , with ∼25% of cells in G1 ( compared to ∼31% for wild type cells ) , but that value was still within 2 sd of the WT mean ( Figure 2 ) . To examine the issue of false negatives in more detail , we determined the timing of START in two strains , which were close to our cutoffs , but not included in the candidate lists . Each of these strains lacked a protein kinase of unknown function: Kns1p [37] -kns1Δ cells had a 27% G1 score; or Tda1p [38] -tda1Δ cells had a 39% G1 score . DNA content measurements from asynchronous cultures only reflect the relative duration of the G1 phase compared to the rest of the cell cycle phases . For example , a given deletion could increase the length of not only the G1 phase , but also subsequent phases . In that case , if the mitotic phases are disproportionately expanded compared to the G1 phase , that strain will display a “Low G1” DNA content , despite its lengthened G1 phase . To address this possibility , we obtained estimates of the absolute length of the G1 phase . The length of the G1 phase of a strain cultured in any given medium can be measured if one knows three parameters: i ) The size of newborn cells ( “birth” size ) . ii ) The “critical size” these newborn daughter cells must attain to initiate cell division . iii ) The rate ( “growth rate” ) at which they grow from their birth size to their critical size . Each of these variables is obtainable in yeast . From cell size distributions of log-phase cultures obtained with a channelyzer , daughter “birth” size was defined as the maximum size of the smallest 10% of cells on the left side of the cell size distribution of each strain . Wild type , kns1Δ and tda1Δ cells had indistinguishable cell size distributions ( Figure S2A ) , and the same birth size ( ∼35 fl ) , in this medium ( YPD-0 . 5% Dextrose ) . To obtain the “critical size” and “growth rate” of these strains , we examined highly synchronous , elutriated cultures [39]–[41] . As a function of time , we measured cell size and the percentage of budded cells ( budding correlates with START completion ) . We found that there was no difference between wild type and kns1Δ cells ( Figure S3 ) . In contrast , tda1Δ cells delay START , not because they have altered critical size ( Figure S3B ) , but because they reach that size slower than wild type cells do ( Figure S3A ) . Hence , our cutoffs exclude some gene deletions with cell cycle effects , such as whi5Δ or tda1Δ cells . Therefore , despite the large number of gene deletions we identified to alter cell cycle progression significantly , we have likely underestimated that number . We found that reduced fitness [33] correlates with altered cell cycle progression to some degree ( Figure 3 ) . Nevertheless , many gene deletions affect cell cycle progression , without affecting fitness . Cells that spend relatively more time in a particular cell cycle phase may not display reduced fitness because reciprocal , compensatory changes in the duration of other cell cycle phases may result in no net change in total generation time . Several known cell cycle mutants behave in this manner ( e . g . , whi5 cells [23] ) . We then compared %G1 values against cell size [23] , [24] . We expected a strong negative correlation between cell size and the fraction of cells with unreplicated genome , since as cells advance in the cell cycle , the bigger the cells become . Remarkably , however , there was only a very weak , negative correlation between %G1 and cell size ( r = −0 . 14 , Figure 4 and Figure S4 ) . Most of the deletion strains displaying a longer G1 ( the “High G1” group ) did not have altered cell size ( Figure 4 , strains between the dashed lines; and Figure S4 ) . Conversely , many strains classified as size mutants [23] , [24] did not have significantly altered DNA content ( Figure 4 , open circles outside the dashed lines , and Figure S4 ) . These data show that changes in cell size are neither necessary nor sufficient for altered cell cycle progression . In the Discussion , we describe the implications of these results in the context of previous attempts to identify cell cycle regulators based on cell size changes . Along with DNA content , we also analyzed the forward scatter ( FSC ) from the same flow cytometry experiments . FSC values often serve as a proxy for cell size , especially in animal model systems [42] , [43] . An overall negative correlation between FSC values and %G1 was present ( r = −0 . 26 , Figure S5 ) . However , we noticed some discrepancies . For example , in the “High G1” group %G1 correlated to some extent with FSC ( r = −0 . 31 ) , but much less with actual cell size ( r = −0 . 09 , Figure 4 ) . We then correlated FSC values to cell size . Surprisingly , for the majority of strains , FSC values do not correlate well with published [23] , [24] cell size values ( Figure S6 ) . These data suggest that inferring cell size phenotypes from FSC measurements may be problematic . We next asked if there is a correspondence between genes that affect cell division when over-expressed , with genes required for normal cell cycle progression . We compared our data set to the genes identified in a systematic over-expression screen , which also relied on DNA content changes [28] . In only one case did over-expression of a non-essential gene have the reciprocal effect of its deletion ( NIP100 , encoding the large subunit of dynactin; Table S2 ) . On the other hand , about half of the deletion strains with a low budding index [44] also had a high %G1 ( Table S3 ) . This is reasonable , since budding correlates with START completion [1] . The “Low G1” group is enriched for “cell cycle” gene ontologies ( Table S4 ) . We point out the sic1Δ strain , which was the 2nd-highest ranked strain of the group . Sic1p is a Cdk inhibitor of Clb/Cdk complexes , which is destroyed before cells initiate DNA replication [13] . Cells lacking Sic1p are not small size mutants [23] , [24] , and Sic1p was identified biochemically , as a Cdk-associated protein [45] . The “High G1” group is enriched for genes involved in “cytoplasmic translation” and “ribosome biogenesis” ( Table S5 ) . This is consistent with protein synthesis and ribosome biogenesis being required for the timely completion of START [2] , [46]–[49] . In our analyses , we considered a high G1 DNA content and a lengthened G1 phase indicative of delayed START . We noticed that some of the genes involved in ribosome biogenesis and protein synthesis that we found with a “High G1” DNA content , were also classified by others as small size mutants with accelerated START [23] , [50] . For example , sfp1Δ cells , which lack a transcription factor important for ribosome biogenesis [23] , [51] , [52] , was the 2nd highest-ranked gene deletion in our “High G1” group ( see Figure S1 and Dataset S1 ) . Yet , although the high G1 DNA content of sfp1Δ cells was noted [23] , because of the small size of sfp1Δ cells , others concluded that START was accelerated in these cells [50] . To resolve these discrepancies , we decided to examine transit through G1 and START completion in sfp1Δ cells . We did these experiments in YPD medium with 2% Dextrose , because Jorgensen et al used the same medium in a similar analysis of sfp1Δ cells [50] . Under these conditions , wild type cells have a “birth” size of 42 . 12±1 . 23 fl ( n = 3 ) and a “critical” size of 61 . 53±0 . 64 fl ( n = 8 ) . We found that sfp1Δ cells had dramatically reduced “birth” ( 16 . 04±0 . 62 fl , n = 3 , P = 6 . 9×10−5 based on a t test , see Figure S2B ) and “critical” ( 39 . 23±0 . 53 fl , n = 6 , P = 2 . 1×10−10 , Figure 5C , Figure S7 ) sizes , and “growth rate” ( Figure 5A and 5B , Figure S7 ) . We calculated the “growth rate” differences between wild type and sfp1Δ cells in two different ways ( see Methods ) , assuming that growth is exponential or linear . If growth is exponential , then sfp1Δ cells grow at ∼50% the rate of wild type cells ( Figure 5B , Figure S7 ) . If growth is linear , then sfp1Δ cells grow at ∼30% the rate of wild type cells ( Figure 5A , Figure S7 ) . For all other comparisons of “growth rates” between different strains that we present in this study , we obtain similar results , regardless of whether we plot size increases in an exponential or a linear manner , because the overall size of those strains is similar to wild type . However , given the strong cell size phenotype of sfp1Δ cells , and since exponential growth incorporates cell size differences ( i . e . , smaller cells grow slower than large cells ) , the growth rate decrease of sfp1Δ cells compared to wild type appears somewhat less if one assumes exponential increase in size . Nonetheless , regardless of whether growth is linear or exponential , it is clear that the G1 phase of sfp1Δ cells is substantially expanded ( ∼4-fold , see Methods for calculations ) . Cells lacking Sfp1p have a very long G1 because they are born very small , and they grow very slowly . Therefore , their small critical size notwithstanding , we conclude that START is severely delayed in sfp1Δ cells . We expand on this interpretation further in the Discussion . To probe the connection between ribosomes and START further , we next evaluated rps0bΔ cells , another mutant with small size [23] , lacking one of the Rps0 variants of the 40S ribosome particle . Cells lacking RPS0B have a high G1 DNA content ( 54% , see Dataset S1 ) . We found that rps0bΔ cells have a reduced “birth” size ( 34 . 53±1 . 89 fl , n = 3 , P = 0 . 007 based on a t test , see Figure S2B ) , an increased “critical” size ( 70 . 06±1 . 90 fl , Figure 5C , Figure S7 ) , and a slow “growth” rate ( Figure 5A and 5B , Figure S7 ) . From these data , we conclude the following: i ) since each of these changes alone would be sufficient to prolong G1 , the combination of all three adequately explain the significant G1 delay of rps0bΔ cells , ii ) “birth” size is not necessarily a predictor of “critical” size , and vice versa , since the two values can be highly discordant , as in rps0bΔ cells , and iii ) DNA content measurements incorporate contributions of all these variables , including growth rate , successfully identifying the long G1 and delayed START of rps0bΔ cells . Next , we examined if there are any patterns in the requirement of ribosomal proteins for the timely completion of START . Intriguingly , although deletion of ribosomal protein subunits delayed START in general , the effect was much greater upon loss of 40S ribosomal proteins ( RPSs ) , compared to the 60S subunits ( RPLs; Figure 6A ) . In contrast , loss of RPSs or RPLs had similar effects on fitness ( Figure 6B ) , or cell size ( Figure 6C ) . Factors with related biological functions show genetic interactions more often than expected by chance [53] . We queried the BioGRID database [54] , for interactions among the genes we identified . Most of the factors of the “Low G1” group have multiple interactions with each other ( Figure 7 ) . In the “High G1” dataset , we also noted several highly connected factors ( Figure 8 ) , including the SR protein kinase Sky1p , similar to human SRPK1 , which is involved in regulating proteins involved in mRNA metabolism . A group of genes in the “High G1” dataset that does not appear to interact with the rest of the group is composed of subunits of the vacuolar ATPase ( Figure 8 , bottom ) . Finally , we also noted an interaction between a metabolic enzyme , Cys4p , and the Cdk Cdc28p [55] . CYS4 encodes the yeast CBS . We focused on Cys4p because we had previously shown that cells with a hypermorphic CYS4 allele accelerate START [39] . Since the loss of Cys4p delays START ( see Dataset S1 ) , we queried the effects of Cys4p over-expression on START . To measure the timing of START , we examined highly synchronous , elutriated cultures . All strains cells had indistinguishable cell size distributions ( Figure S2C ) and the same birth size ( ∼14 fl , Figure S2C ) in this medium ( YPGal-3% Galactose ) . Consistent with Cys4p's metabolic role [39] , we found that over-expression of Cys4p , but not of the catalytically inactive Cys4p-S289D variant [56] , increased growth rate ( Figure 9A ) . Over-expression of Cys4p also reduced the critical size for START ( Figure 9B ) . Hence , wild type Cys4p accelerates START both by increasing growth rate , and by reducing critical size . Taking both of these variables into account , we conclude that over-expression of Cys4p shortens the length of the G1 phase by ∼30% ( see Methods for calculations ) . Remarkably , over-expression of Cys4p-S289D also decreased critical size ( Figure 9B , right ) . These results suggest that Cys4p promotes START in two ways: By promoting cell growth , which requires its catalytic activity; and by reducing critical size , which does not require Cys4p's catalytic activity . Yeast lacking CYS4 can be viable if supplemented with cysteine [57] . In the standard S288c strain background we used here , cys4Δ cells proliferate slower than wild type ( ∼2 to 3-fold ) , even in rich media [33] . In humans , patients with CBS deficiency have high levels of homocysteine . These patients have brain , skeletal and vascular abnormalities [58] . There are more than 130 pathogenic CBS mutations , but not all of them affect the activity of CBS [59] . Cbs−/− mice have high levels of homocysteine ( >200 µM ) and die within weeks after birth [60] . In Cbs−/− mice , cells critical for the development of the cerebellum cannot proliferate [61] . Introducing human CBS alleles that encode inactive enzymes did not reduce the homocysteine levels of these mice , but these transgenes did rescue the neonatal lethality of Cbs−/− mice [62] . Thus , in animals , CBS must have essential , non-catalytic roles . Because of these observations , we asked if the catalytic role of Cys4p is separable from the proliferative defects associated with loss of Cys4p in yeast . We generated strains that express Cys4p-S289D at endogenous levels ( Figure 10A , lanes 3 & 4 ) . These strains are cysteine auxotrophs ( Figure 10B , middle panel ) , consistent with their lack of Cys4p catalytic activity . However , when cysteine is present , they proliferate much better than strains that lack Cys4p altogether ( Figure 10B , lower panel ) . These results are in remarkable agreement with the data in mice: Loss of CBS leads to proliferative and metabolic defects ( homocysteinuria in mice , cysteine auxotrophy in yeast ) . In both organisms , inactive CBS does not suppress the metabolic defects , but it suppresses the proliferative defects .
We think that this likely reflects the fact that cells commit to initiation of cell division in the G1 phase . It is reasonable to expect that extensive regulatory networks contribute to such a critical cellular transition , perhaps more so than for other cell cycle transitions . Interestingly , inactivation of the majority of essential genes also leads to a G1 arrest [30] . Furthermore , the strong requirement of protein synthesis for START completion [1] , [2] , [46]–[48] , [63]–[65] , and the large number of essential and non-essential genes involved in protein synthesis , also partially explains why most gene deletions that affect the cell cycle lead to a G1 delay . This question has been highly debated ( see [66] , [67] for related commentaries ) , especially when yeast is contrasted with animal model systems . Our study does not address this question . The debate about whether there is a critical threshold for initiation of cell division centers on whether cell size increases in a linear , or in an exponential fashion [66]–[69] . In several experiments , we monitored cell size increases as a function of time in synchronous cultures . However , our data points are of insufficient resolution to distinguish between an exponential vs . linear mode of growth ( see Figure S7 and Methods ) . Note that this limitation does not in any way affect our conclusions about the relative rates of growth of different strains . In fact , when we compare strains with similar overall size distributions ( see Figure S2A ) the relative “growth rates” we obtain are the same , whether cells increase in size exponentially or not . Even in the case of strains with very different size distributions ( e . g . , wild type vs . sfp1Δ cells , see Figure S2B and Figure 5 ) , the results are qualitatively similar , regardless of the pattern of growth . Nonetheless , in our study we have monitored and incorporated in our calculations the “critical size” at which cells initiate their division . From these experiments and similar others we published previously ( see Figures S3 , S7 , S8 , S9 and [39]–[41] ) , the “critical size” is a highly reproducible parameter . Hence , in accordance with numerous other reports , it is our opinion that any strain growing in a given medium has to reach a critical size characteristic of that strain and medium . Our genome-wide data unequivocally show little correlation between %G1 and cell size ( see Figure 4 and Figures S4 , S5 ) . Thus , although reaching a critical size threshold for initiation of cell division contributes to the timing of START , the most reasonable conclusion from our data is that genetic determinants of size control mechanisms are neither the sole nor the major factor determining the timing of initiation of cell division in dividing cells . This is a key finding of our study , which stands in marked contrast to previous approaches that used cell size alterations as a means to identify START regulators [23] . In our opinion , monitoring the length of the G1 phase reflects the timing of START far more accurately than monitoring cell size . We expand more on this issue next , when we discuss the role of ribosome biogenesis and the behavior of wild type cells in different nutrients . The behavior of strains lacking genes involved in ribosome biogenesis and protein synthesis exemplifies the different interpretations about the timing of START , depending on whether the focus is on the length of G1 ( this study ) , or on cell size [23] , [50] . We will discuss the phenotypes of sfp1Δ cells , because we examined them ( see Figure 5 ) with the same methods and under the same conditions as in previous studies by Jorgensen et al [50] . The parameters we obtained are in complete agreement with those of Jorgensen et al [50]: sfp1Δ cells divide at a greatly reduced cell size , grow much more slowly than wild type cells , and they are also born very small . Jorgensen et al focused on their small critical size and concluded that START was accelerated in sfp1Δ cells and other strains lacking genes involved in ribosome biogenesis [50] . Instead , we took into account not only their small critical size , but also their extremely slow growth rate and small birth size ( see Figure 5 and Figures S2 , S7 ) . We conclude that START must be severely delayed in sfp1Δ cells , because these cells have such an expanded G1 . If one focuses only on the small critical size of sfp1Δ cells , it may seem that START is accelerated . However , we think it is more accurate to describe these cells simply as small and severely growth-impaired . Loss of Sfp1p delays START to such an extent that during the time sfp1Δ cells spend in G1 , their wild type counterparts would have initiated several new rounds of cell division . Not all gene deletions that affect ribosome biogenesis prolong G1 and those who do may differ quantitatively and qualitatively in their impact ( Figure 5 , Figure 6 ) . Overall , however , there is a prolongation of the G1 phase in many ribosome biogenesis mutants ( see Dataset S1 ) . Because of their lengthened G1 , we conclude that START is delayed in strains lacking non-essential ribosomal components or factors that regulate protein synthesis . This interpretation is consistent with the terminal G1 arrest of essential genes involved in the same processes [30] , and with the strong delay of START upon inactivation of rRNA processing in yeast [48] . For these reasons , we conclude that gene deletions that impair ribosome biogenesis delay START , and that in dividing wild type yeast cells , ribosome biogenesis promotes START . This conclusion also agrees with extensive evidence from animal cells that increased ribosomal biogenesis ( by Myc and other oncogenes ) promotes initiation of cell division [70]–[73] . Obviously , completion of START and commitment to a new round of cell division precedes the actual end of the G1 phase , when cells initiate DNA replication [1] , [11] . Mutants in processes that molecularly link START with DNA replication ( e . g . , cdc34 cells [74] ) , may complete START , but they are unable to initiate DNA replication . These rare exceptions notwithstanding , we see no compelling reason that invalidates using the length of the G1 phase as an accurate metric of the timing of START . This is supported further by the behavior of dividing wild type cells in different growth conditions . Poor growth conditions greatly prolong G1 , whereas the time required to transit the remaining cell cycle phases is unaffected [3] . In steady-state chemostat cultures , where growth rate can be altered independently of nutrient composition , the lower the growth rate is , the longer the cells stay in G1 , delaying START completion [40] , [75] , while cell size remains largely unaffected [75] , [76] . Nutrients also affect the critical size threshold . Cells dividing in poor carbon sources typically are small , but they also have a slow growth rate and a long G1 [77] , resembling ribosome biogenesis mutants with a delayed START . We would like to clarify that , in all of the above examples we discussed , we considered continuously dividing populations , without media changes . In a nutritional up-shift , from poor to rich media , G1 is transiently prolonged , probably until cells reach the new larger “critical size” characteristic of the rich medium [78] . During this short temporal window , in the first cell cycle as cells transit from the poor medium to the new richer one , genetic control of the “critical size” threshold likely prolongs G1 and delays START by increasing the critical size threshold [50] , [79] . In subsequent cell cycles however , despite the larger “critical size” cells have to attain in that richer medium , the cells are born larger and grow faster , resulting in a short G1 and accelerated START . What could be the benefit of the small critical size observed in poor nutrients ? It has been argued that the plasticity of critical size thresholds may allow yeast cells to “best compete for limited and fluctuating resources” [79] . This is reasonable , if one keeps in mind the two competing objectives of all proliferating cells: i ) Ensure that growth requirements are met before initiating a new round of cell division; ii ) At the same time , exploit all the available nutrients to divide as quickly and as many times as possible . Perhaps , with their smaller birth size and slower growth rate , which lengthen G1 , cells in poor nutrients satisfy the first objective . Then , as they progress in G1 , cells have to reach a smaller critical size , alleviating a little bit the overall delay in initiating a new round of cell division in poor nutrients . Overall , our results increase the number of gene deletions that delay G1 , as listed currently in the Saccharomyces Genome Database , by more than 3-fold . Even if one excludes genes involved in ribosome biogenesis , we still uncovered >100 genes required for the timely initiation of cell division ( see Dataset S1 ) . Most of the genes we identified do not affect cell size . As a result , these genes were not identified in previous attempts to find regulators of START . Hence , our findings greatly expand and reshape our view of START . We followed up one such gene we identified in this study , Cys4p ( CBS ) . CBS is a key metabolic enzyme , associated with disease in humans , with conserved functions between yeast and humans . Indeed , human CBS complements yeast lacking Cys4p [80] . Hence , the role of CBS in cell division we described in yeast may be significant for human biology . The systematic identification of non-essential regulators of START we described here will be the basis for further insight into the control of cell division in yeast and other organisms . It enables future studies to define how many pathways affect START , which factors operate within each pathway , and the extent of interactions between pathways .
S . cerevisiae strains used in this study are listed in Table S1 . Unless noted otherwise , we used standard yeast methods [31] . To construct the PGAL-GST-CYS4 strain ( Figure 9 and Figures S8 , S9 ) , we started from a commercially available plasmid containing a PGAL-GST-CYS4 allele ( Open Biosystems , cat#: YSC3869-95169400 ) . However , this plasmid contained a frameshift mutation at nucleotide position 856 of the CYS4 ORF , which we corrected . We then removed a BsrGI-SalI fragment , re-ligated the plasmid , and digested it with StuI . Finally , we integrated this linearized plasmid derivative containing the PGAL-GST-CYS4 allele at the URA3 locus of W303-k699 ( see Table S1 ) . We sequenced a similar plasmid supposed to carry a PGAL-GST-KIP3 allele ( YSC3869-9518649 ) , but we found that it only drives expression of GST , due to downstream mutations . We used this plasmid to construct the negative control PGAL-GST strain ( Figure 9 and Figures S8 , S9 ) , as we described above . From the PGAL-GST-CYS4 plasmid we generated the PGAL-CYS4 ( S289D ) derivative , as follows: We used the PGAL-GST-CYS4 plasmid as a template in a PCR reaction with a forward primer encoding the S289D substitution , and a reverse primer complementary to sequences downstream of the CYS4 ORF . The PCR fragment was then used to gap-repair the PGAL-GST-CYS4 plasmid , which was previously digested with BstEII . The resulting PGAL-CYS4 ( S289D ) plasmid was then used in the same way as above , to construct the PGAL-CYS4 ( S289D ) strain ( Figure 9 and Figures S8 , 9 ) . All plasmids were sequenced and the resulting strains were verified for expression of the proteins of interest . The CYS4-13MYC strain ( Figure 10 ) was made with a single-step PCR replacement [39] . To make the CYS4 ( S289D ) -13MYC strain ( Figure 10 ) , we used genomic DNA of the CYS4-13MYC strain as a template in a PCR reaction with a forward primer encoding the S289D substitution , and a reverse primer complementary to sequences downstream of the CYS4 ORF . For DNA content measurements , strains were cultured standing at 30°C in YPD ( 1% yeast extract , 2% peptone , 2% dextrose ) . Overnight cultures were diluted 1∶500 into 1 ml fresh medium , cultured for 4–5 hrs , collected by centrifugation and fixed in 70% ethanol . To obtain size distributions from asynchronous cultures , overnight cultures of the strain and medium of interest were diluted 1∶500 in fresh medium , and allowed to proliferate for 5–6 h , before we analyzed them . For synchronous cell cycle analyses [39] , strains were cultured and elutriated in YPD medium containing 0 . 5% dextrose ( Figures S3 , S9 ) , 2 . 0% dextrose ( Figure 5 , Figure S7 ) , or YPGal ( 1% yeast extract , 2% peptone , 3% galactose; Figure 9 , Figure S8 ) , as indicated . Cell size was measured with a Beckman Z2 Channelyzer . For each sample we analyzed , we obtained size distributions from two different dilutions of cells . The average of the geometric mean of each size distribution was recorded . We used the Accucomp Beckman software package to obtain the statistics of each size distribution . For isolation of early G1 daughter cells , cultures were grown in the medium indicated in each case at 30°C to a density of ∼1–5×107 cells/ml , then fractionated with a Beckman JE-5 . 0 elutriator as described previously [41] . Early fractions containing predominantly ( >95% ) small unbudded cells were collected by centrifugation , resuspended in the medium indicated in each case and incubated at 30°C . Every 20 min we monitored the percentage of budded cells and cell size . The “critical size” is the size at which 50% of the cells have budded in these experiments , and it was calculated as we described elsewhere [41] . We calculated the rate of size increase , “growth rate” ( in fl/min ) , assuming linear growth , as we described previously [41] . To calculate “growth rate” assuming exponential growth , we plotted the natural log ( ln ) of cell size as a function of time ( in h ) , see Figure S7B . We fit the data to a straight line using the regression function in Microsoft Excel . From the slope of the line , we obtained the specific rate of cell size increase constant ( k , in h−1 ) . The average of all experiments for each strain was then calculated , along with the associated standard deviation . Since sometimes it takes the cells longer to recover from the elutriation , in our growth rate calculations we exclude this “lag” phase . We derived growth rate data only from the linear portion of each experiment . Estimates of the length of G1 were calculated as follows: Assuming linear growth , G1 ( min ) = ( “Critical Size”-“Birth Size” ) /”Growth Rate” . Assuming exponential growth , G1 ( h ) = ln ( “Critical Size”/”Birth Size” ) /k . Fixed cells were stored at 4°C overnight to 14 days . Cells were collected by centrifugation and stained overnight in 1 ml staining solution containing 50 mM sodium citrate pH 7 . 0 , 0 . 25 mg/ml RNaseA , and 500 nM SYTOX Green ( Molecular Probes , OR ) . Samples were stored at 4°C overnight in opaque containers . Cell suspensions were sonicated briefly at the fixing and staining steps and immediately before flow cytometry . Stained cells were analyzed on a FACSCalibur ( Becton Dickinson Immunocytometry Systems , CA ) flow cytometer , using CellQuest ( version 3 . 3; Becton Dickinson Immunocytometry Systems ) acquisition software . Sytox Green fluorescence was collected through a 515/30-nm bandpass filter , and list mode data were acquired for 10 , 000 cells defined by a dot plot of FSC versus SSC . Prior to each experiment , standard beads ( Cyto-Cal Multifluor Intensity Beads , Thermo Scientific , CA ) were used to calibrate the flow cytometer , and photomultiplier tube voltages were adjusted to place the highest intensity bead in the same channel ( +/−3 ) . FACS files were archived at Cytobank [81] . Automated quantification of the DNA content histograms was done with FlowJo 7 . 5 software . To exclude particulate non-yeast events , which had both very low forward scatter ( FSC ) and low fluorescence ( FL1/2-A ) , asymmetrical gates were fitted with the autogating tool . Gates were centered near FSC ∼100 and FL1/2-A ∼300 and contained all events of sufficient contiguity as defined by the default autogating parameters , on average ∼91% of total . From the gated populations , we determined the mean and standard deviation of the FSC parameter . Cell cycle phase subpopulations were computed from the gated population using the Dean-Jett-Fox model without constraints . %G1 was defined as the area of the G1 model peak , divided by the combined areas of the G1 and G2/M peaks . Because the %G1 results represent a continuum , it was necessary to impose cutoffs in order to exclude model fits that did not accurately represent experimental data . This was assessed primarily by root mean square ( RMS ) error , which averaged 11 . 68 ( +/− a standard deviation of 2 . 80 ) across all included experiments . For this reason , we did not analyze experiments that yielded an initial model fit RMS >25 , %G1<5% , or %G1>95% ( since extremes in %G1 were often indicative of poor fit ) , except in a few cases where the model fit was visually inspected and/or manually constrained . Experiments for which the %G1 fell outside two standard deviations of the wild-type distribution were repeated additional times . Experimental data and correlations are provided in the searchable spreadsheet available as Dataset S1 . Raw data files can be freely accessed at Cytobank ( www . cytobank . org ) and are found in the public experiments “Yeast DNA Content Project – DELETION – INCLUDED” , and “Yeast DNA Content Project – DELETION – EXCLUDED” . Non-parametric Spearman tests were done with the Analyze-it software package . In all other cases , statistical calculations were done with Microsoft Excel . Where indicated , t tests were 2-tailed , assuming unequal variance between data sets . Protein extracts for immunoblots were made with the NaOH extraction method [82] . For detection of proteins of interest on immunoblots we used an anti-PSTAIR antibody to detect Cdk ( Figure 10A; Abcam , Cat#: ab10345 ) and an anti-hCBS polyclonal antibody to detect human and yeast CBS proteins ( Figure 10A; SantaCruz , Cat#:46830 ) . Secondary antibodies were from Pierce . All antibodies were used at the dilutions recommended by the manufacturers .
|
What determines when cells begin a new round of cell division also dictates how fast cells multiply . Knowing which cellular pathways and how these pathways affect the machinery of cell division will allow modulations of cell proliferation . Baker's yeast is suited for genetic and biochemical studies of eukaryotic cell division . Previous studies relied mainly on cell size changes to identify systematically factors that control initiation of cell division . Here , we measured the DNA content of each non-essential single gene deletion strain to identify genes required for the correct timing of cell cycle transitions . Our comprehensive strategy revealed new pathways that control cell division . We expect that this study will be a valuable resource for numerous future analyses of mechanisms that control cell division in yeast and other organisms , including humans .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"systems",
"biology",
"model",
"organisms",
"genetics",
"biology",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] |
2012
|
A Systematic Analysis of Cell Cycle Regulators in Yeast Reveals That Most Factors Act Independently of Cell Size to Control Initiation of Division
|
Wastewater irrigation is associated with several benefits but can also lead to significant health risks . The health risk for contracting infections from Soil Transmitted Helminths ( STHs ) among farmers has mainly been assessed indirectly through measured quantities in the wastewater or on the crops alone and only on a limited scale through epidemiological assessments . In this study we broadened the concept of infection risks in the exposure assessments by measurements of the concentration of STHs both in wastewater used for irrigation and the soil , as well as the actual load of STHs ova in the stool of farmers and their family members ( 165 and 127 in the wet and dry seasons respectively ) and a control group of non-farmers ( 100 and 52 in the wet and dry seasons , respectively ) . Odds ratios were calculated for exposure and non-exposure to wastewater irrigation . The results obtained indicate positive correlation between STH concentrations in irrigation water/soil and STHs ova as measured in the stool of the exposed farmer population . The correlations are based on reinfection during a 3 months period after prior confirmed deworming . Farmers and family members exposed to irrigation water were three times more likely as compared to the control group of non-farmers to be infected with Ascaris ( OR = 3 . 9 , 95% CI , 1 . 15–13 . 86 ) and hookworm ( OR = 3 . 07 , 95% CI , 0 . 87–10 . 82 ) . This study therefore contributes to the evidence-based conclusion that wastewater irrigation contributes to a higher incidence of STHs infection for farmers exposed annually , with higher odds of infection in the wet season .
Wastewater use in agriculture has been promoted as part of the concept of sustainable development . In many cities in developing countries , wastewater irrigation is a common reality linked to rapid urbanization . The practice improves farmers’ livelihoods , contributes to the urban food basket and slightly improves the urban environment by diverting wastewater to agricultural fields [1] . In Sub-Saharan Africa ( SSA ) , it is estimated that 10% of the population in cities are involved in the practice of wastewater irrigation , with 50% to 90% of urban dwellers in West Africa reported to consume vegetables irrigated with wastewater or polluted surface water within or close to cities [2] . In Ghana , a significant proportion of untreated wastewater is discharged into drains and nearby water bodies , which is then used by farmers for irrigation . Wastewater irrigation in the cities of Ghana is mainly for the production of vegetables , such as cabbage , lettuce , spring onion and carrots [3] . Although there are many benefits associated with wastewater irrigation , the practice can lead to significant health risk if not undertaken in a safe manner [4] . All enteric pathogens of viral , bacterial and parasitic ( helminthic and protozoan ) origins can be found in wastewater; and can be transmitted to farmers using the wastewater for irrigation , consumers of wastewater-irrigated vegetables and communities close to wastewater irrigated fields [5] . Several studies have shown a significant relationship between Ascaris infection and exposure to wastewater ( either treated or untreated ) [6 , 7 , 8] . This is because soil transmitted helminths ( STHs ) ( such as Ascaris ) can survive for long periods of time under severe adverse environmental conditions [9] contributing to their high risk of infection . STHs are common worldwide with more than a billion people infected [10 , 11] . Estimates suggest that Ascaris lumbricoides infects over 1 billion people , Trichuris trichiura 795 million , and hookworms ( Ancylostoma duodenale and Necator americanus ) 740 million [12] . Farmers in Pakistan using wastewater for irrigation have been reported to be five times more likely to be infected with hookworms than others using canal water [13] and in Dakar , Senegal the reported incidence of amoebiasis and ascariasis is 60% in farmers involved in wastewater irrigation [14] . In a study in Mexico , irrigation with untreated or partially treated wastewater was directly responsible for 80% of all A . lumbricoides infections and 30% of diarrheal disease in farm-workers and their families [15] . The health risk can differ depending on age and gender . An epidemiological study by Habbari et al . [16] undertaken in Morocco to determine possible health risks associated with raw wastewater use in agriculture found ascariasis infection to be approximately five times higher , especially in children in wastewater impacted regions compared to control regions . In another study Fuhrimann et al [35] found that farmers exposed to wastewater in Uganda were more likely to be infected with helminths than slum dwellers and workers involved in sludge collection . However , in Vietnam , Trang et al . [17] found no evidence that rice cultivation with wastewater posed any significant helminth infection risk to farmers , even though they were exposed to wastewater containing 40–200 helminth eggs/L . Prevalence of and risk factors for helminth infections have been studied in Ghana [18] . Although wastewater irrigation has been the practice for many years in Ghana , especially in major cities ( e . g . Accra , Kumasi , Tamale ) , there are no studies that have investigated the epidemiological link between the practice and helminth infections among the farmers . Studies in Ghana have reported a mean helminth ova concentration of 5–10 helminth ova per liter in water used for irrigation by farmers [19 , 20] . In this regard this study aimed at determining the association between STHs ova concentration in wastewater used for irrigation as well as in farm soil that farmers are exposed to and the actual infection loads in order to ascertain the epidemiological link between wastewater irrigation and risk of STHs infection for farmers . The aim of this study was achieved as it was deduced that farmers had a higher probability of infection than non-farmers .
The study was conducted in wastewater irrigated vegetable farms in the Kumasi Metropolitan Area of Ghana ( Fig 1 ) . The Metropolis has two major seasons , the rainy ( April to October ) and the dry one ( November to March ) . Relative humidity ranges from 60–84% with daily minimum and maximum temperatures of 21 . 5°C and 30 . 7°C , respectively [21] . The majority of vegetable farms in Kumasi are irrigated with wastewater which is most predominant in the dry season . Wastewater from domestic and small-scale industrial ( e . g vehicle garages , saw mills , welding shops , tanneries etc ) sources are discharged directly into stormwater drains and streams and collected for irrigation by farmers . An initial survey was carried out in the Kumasi Metropolis to identify the farms using wastewater for irrigation . This included a detailed explanation of the purpose of the study and farmers and control-group who gave consent to be part of the study were recruited . Non-farmer ( control group ) inhabited the same areas as the farmers and recruitment was made concomitantly . The control group thereby constituted members of families of their communities who did not take part in the practice of wastewater irrigation but stayed in the same neighborhood ( as can be seen in Fig 1 below ) . An initial prevalence survey was undertaken after which participants were dewormed and the efficiency of the deworming exercise assessed directly afterwards . The farmer group consisted of 165 ( in the wet season ) farmers and family members dropping to 127 in the dry season , while the control group originally consisted of 100 individuals ( in the wet season ) , dropping to 52 in the dry season . The exclusion criteria was arrived at after the administration of the questionnaires to all participants , afterwards any participant who did not fall within the criteria set ( based on self-reporting ) was not included in the final data used for analysis , hence the dropout rate , but the project team still visited them and administered antihelminthic drugs when needed so as to encourage participation in subsequent studies . Wastewater farmers recruited into the cohort met the following inclusion criteria: a ) did not consume vegetable salad irrigated with wastewater from their farms; b ) used improved toilet facilities at home and at work; c ) did not use protective clothing during farm work; and d ) had access to treated drinking water in their homes/communities . The Committee on Human Research , Publications and Ethics ( CHRPE ) of the Kwame Nkrumah University of Science and Technology ( KNUST ) approved the study ( No . CHRPE/RC/051/12 ) with additional informed oral consent received from all participants . Informed oral consent of parents or guardians was received for all children who participated in the study , which was written on the field questionnaire administered to each person . The purpose and details of the study was explained to all participants in Twi ( a local dialect ) in the presence of a witness and those willing to participate gave their consent , which was noted on the questionnaires . Each participant was given a unique identifier which was used throughout the study for confidentiality . After the initial deworming exercise all participants who became re-infected were treated again with 400 mg of albendazole ( XL Laboratories PVT Ltd ) . Irrigation water was collected from August 2012 to October 2012 to represent the wet season and December 2012 to March 2013 to represent the dry season . Irrigation water samples were taken from storm drains , streams , shallow wells and potable water pipes ( in a few instances ) , which represented the sources of water in use by the farmers . Soil samples were taken from the vegetable beds that were irrigated with the water types sampled as stated above . In all , 214 and 156 samples ( for soil and irrigation water ) were taken during the wet and dry seasons , respectively . All samples were collected in the morning between 06 . 00 and 10 . 00 ( Greenwich Mean Time ( GMT ) on each day of sampling . Irrigation water and soil samples were collected in triplicates into sterile pre-labeled sample bottles ( about 4 L for the wastewater ) and plastic re-sealable bags ( 30 g composite soil sample each ) and kept in a cooling box and transported to the laboratory where they were processed and analyzed for helminth eggs using the Modified EPA Method [22] . The helminths eggs were identified on the basis of their shape and size with the aid of bench aids for the Diagnosis of Intestinal Parasites [23] . Only viable helminth eggs were counted , viability was assessed based on the presence of motile larvae within the eggs . Stool samples were collected from farmers and family members as well as the non-farmer control group and analyzed using the formal-ether concentration method [23] . After three months , stool samples were taken again from the participants for assessment of re-infections . Descriptive analysis was undertaken to assess the mean concentration and distribution of ova in the irrigation water and soil and described by box plots ( Stata , Statacorp , Texas , USA ) . Analysis of variance was performed to determine the statistical difference between the concentrations of Ascaris spp and hookworm in the dry and wet seasons . The relationship between STHs loads in irrigation water/soil and actual STHs ova per gram of faecal matter from the farmers was determined using Poisson regression analysis ( Stata , Statacorp , Texas , USA ) . The odds ratio ( OR ) , its standard error and 95% confidence intervals were calculated according to Altman [24] .
Ova of Ascaris spp , hookworm and Schistosoma spp ( Schistosoma spp was found only in the irrigation water and is not further reported in this article ) were identified in the irrigation water and soil in the vegetable farms . In general ova concentrations were higher in the wet season than the dry season for both irrigation water and soil samples ( Refer to Table 1 ) . Statistically there was difference in the concentration of hookworm ova in the two seasons and Ascaris spp concentration in the soil for between the seasons ( Table 1 ) . Figs 2 and 3 shows the distribution of the ova of Ascaris spp and hookworm in the irrigation water and soil for both the wet and dry seasons . Infection with the two parasites differed between seasons and between the farmers and the control group . The prevalence of Ascaris spp infection in the wet season was 15 . 77% ( n = 165 ) for the farmers and 6 . 00% ( n = 100 ) for the control group . Similarly , the prevalence of hookworm infection in the wet season was 12 . 73% ( n = 165 ) for the farmers and 2 . 00% ( n = 100 ) for the control group . In the dry season prevalence of Ascaris spp . was lower for both groups , the farmers had a prevalence of 11 . 02% ( n = 127 ) and 5 . 74% ( n = 52 ) for the control group . A much lower prevalence was recorded for hookworm infections for farmers with 4 . 72% ( n = 127 ) however same prevalence as reported for Ascaris spp was reported for hookworm infections of the control group , but with different mean infections . Table 2 above shows the details of the range and significant difference and Figs 4 and 5 distribution of infection intensity . The concentration of the ova of the two reported helminths in both the irrigation water and soil and the intensity of infection of the exposed farmers showed a significantly positive relationship ( p < 0 . 05 ) in the wet season ( regression coefficient of 0 . 04; 95% CI: 0 . 203–0 . 69 ) . The opposite was the case in the dry season ( regression coefficient of - . 0023; 95% CI: - . 020 - . 016 ) , however this was not statistically significant ( p > 0 . 05 ) . The probability of farmers getting infected with STHs compared to the control group as a result of exposure to the ova in the irrigation water and the soil was higher in the wet season than in the dry season for the two STHs ( Table 3 ) . In the wet season , farmers exposed to irrigation water and soil were more likely than the control group to be infected with Ascaris spp ( OR = 3 . 99 , 95% CI: 1 . 15–13 . 86 ) and Hookworm ( OR = 3 . 07 , 95% CI: 0 . 87–10 . 82 ) . However , there was lesser probability of infection in the dry season as shown in Table 3 .
Helminth contamination of irrigation water is a serious public health issue , due to their persistence in the environment and their low infective dose . To safeguard human health , the WHO formulated guidelines for the use of wastewater in unrestricted agriculture [4] , with a guideline value of <1 ova/L aimed at reducing the risk of infection . In this study , the mean concentration of STHs ova was higher than the recommended guideline value , especially for Ascaris spp ( 2 . 62 ova/L and 2 . 82 ova/L for dry and wet seasons , respectively ) and hookworm ( 2 . 05 ova/L in the wet season ) , in line with similar results reported from studies in Ghana [3 , 19 , 20 , 29] as well as other countries [25 , 30 , 35] . There was seasonal variation in the mean concentration of the STHs ova in the irrigation water , with the wet season showing higher concentrations than the dry season . Keraita et al . [26] reported similar patterns of helminth ova concentration in irrigation water from studies conducted in Kumasi . This occurrence could be attributed to rainfall and reduced temperatures which extend the survival period of the ova . However in general , helminth ova are resistant to many types of inactivation , with ova of Ascaris spp and Taenia spp having the highest resistance and survival rates [27 , 28] . In addition to the lower temperatures and much lesser UV irradiation in the wet season , ova concentration could be increased during this period of the year due to run-offs from agricultural fields and other surrounding areas . Open defecation in areas close to these wastewater irrigated farms could also potentially lead to higher STHs ova concentrations after rainfalls . Wastewater irrigation does not only increase risk of STH infection due to exposure to the irrigation water but also exposure to the farm soil . Exposure to the farm soil in wastewater irrigated farms may result in higher risk of infection with STHs than risk attributable to the wastewater alone [29 , 30] . This is due to a higher concentration of STHs ova in the soil as was seen in this study . Irrigation with wastewater result in accumulation of STHs ova in the soil , and therefore accounts for the higher concentration of ova . A . lumbricoides eggs have been found to attach to soil particles ( especially clay ) thereby contributing to their high concentrations in the soil samples [31] . In addition , contamination of soils could serve as a source for re-introduction of eggs into the irrigation water channels . The elevated concentrations of STHs ova , above the WHO guideline levels , in the irrigation water and soil pose a risk of infection for farmers involved in the practice of wastewater irrigation . However , there is always a difference between the estimated risk and actual infections . The potential health risk is based on the number of pathogens in the wastewater or soil , while the actual health risk depends on an expansion of this concept , including: i ) the period pathogens survive in water or soil; ii ) the dose in which pathogens are infective to a human host and iii ) host immunity for pathogens circulating in the environment [32] . The seasonal variation in STHs ova concentration in the irrigation water and soil was also apparent in the STHs infection intensity of the farmers , reflected in higher infection frequency in the wet season . There are other factors such as , climate , types of soils and hygiene behavior , which might have also contributed to this variation in infection rate [33] . The interrelationship with other confounding factors was seen with the correlation analysis where there was a weak association between the load or concentration in the irrigation water/soil and the intensity of the STH infection for the farmers , especially in the dry season . To quantify the actual contribution of wastewater irrigation to STHs infection a control group of non-farmers who had no exposure to the irrigation water and soil but used same sanitation and portable water infrastructure as the farmers ( staying in the same suburbs as the farmers ) was needed . The increased probability of infection for farmers was expected due to a higher exposure to STHs ova over the course of the year as compared to the non-farmers . Infection with Ascaris spp and hookworm for the farmers is three times more likely than it is for non-farmers . This clearly indicates that wastewater use in irrigation contributes significantly to the incidence of helminthiases , as reported by many other studies [13 , 29 , 34 , 35] , especially in the wet season ( Table 3 ) . In the dry season the odds of infection for both farmers and non-farmers is not significantly different . This could be attributed to the lower concentrations of ova recorded in the irrigation water and soil during this time of the year . It can be concluded from the results obtained in this study that exposure to STHs ova in irrigation water and soil contributes to infections in farmers and that farmers involved in the practice are three times likely to be infected with Ascaris spp and hookworm than unexposed populations . This is particularly so during the wet season where there is an increase in the concentration of the STHs ova . The results obtained show an epidemiological link between wastewater irrigation and helminth infection in Ghana , therefore emphasizing the need for regulations and interventions aimed at making the practice safer for the farmers which in turn would contribute significantly in breaking the cycle of infection .
|
Wastewater irrigation in agriculture is a common reality in many developing cities , linked to rapid urbanization . Approximately 50%-90% of urban dwellers in West Africa consume wastewater/ polluted surface water irrigated-vegetables within cities with 10% of the population involved in the practice . Viral , bacterial and parasitic pathogens can all be found in wastewater putting exposed populations at risk of pathogenic infections . The biggest risk however is to helminth infections , due to their long survival time in the environment . Wastewater irrigation has been practiced in Ghana for many years , however few studies have investigated the epidemiological link between the practice and helminth infections among the farmers . In this study the authors measured the helminth ova concentration in wastewater used for irrigation and the infection loads of farmers as well as a non-farmer control group in Ghana . They reported high concentrations of helminth ova in the wastewater as well as soil on the farms , above the World Health Organization ( WHO ) guidelines for wastewater irrigation , which resulted in a three times higher probability of infections with helminths for farmers as compared to the non-farmer control group . This research provides information on the direct link between wastewater irrigation and helminth infection in exposed individuals .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"agricultural",
"irrigation",
"medicine",
"and",
"health",
"sciences",
"helminths",
"tropical",
"diseases",
"hookworms",
"parasitic",
"diseases",
"animals",
"seasons",
"farms",
"ascaris",
"agricultural",
"methods",
"neglected",
"tropical",
"diseases",
"agriculture",
"helminth",
"infections",
"earth",
"sciences",
"nematoda",
"biology",
"and",
"life",
"sciences",
"soil-transmitted",
"helminthiases",
"organisms"
] |
2016
|
Contribution of Wastewater Irrigation to Soil Transmitted Helminths Infection among Vegetable Farmers in Kumasi, Ghana
|
The RFAM database defines families of ncRNAs by means of sequence similarities that are sufficient to establish homology . In some cases , such as microRNAs and box H/ACA snoRNAs , functional commonalities define classes of RNAs that are characterized by structural similarities , and typically consist of multiple RNA families . Recent advances in high-throughput transcriptomics and comparative genomics have produced very large sets of putative noncoding RNAs and regulatory RNA signals . For many of them , evidence for stabilizing selection acting on their secondary structures has been derived , and at least approximate models of their structures have been computed . The overwhelming majority of these hypothetical RNAs cannot be assigned to established families or classes . We present here a structure-based clustering approach that is capable of extracting putative RNA classes from genome-wide surveys for structured RNAs . The LocARNA ( local alignment of RNA ) tool implements a novel variant of the Sankoff algorithm that is sufficiently fast to deal with several thousand candidate sequences . The method is also robust against false positive predictions , i . e . , a contamination of the input data with unstructured or nonconserved sequences . We have successfully tested the LocARNA-based clustering approach on the sequences of the RFAM-seed alignments . Furthermore , we have applied it to a previously published set of 3 , 332 predicted structured elements in the Ciona intestinalis genome ( Missal K , Rose D , Stadler PF ( 2005 ) Noncoding RNAs in Ciona intestinalis . Bioinformatics 21 ( Supplement 2 ) : i77–i78 ) . In addition to recovering , e . g . , tRNAs as a structure-based class , the method identifies several RNA families , including microRNA and snoRNA candidates , and suggests several novel classes of ncRNAs for which to date no representative has been experimentally characterized .
Starting with the discovery of microRNAs [1–3] and the advent of genome-wide transcriptomics [4–6] , it has become obvious that RNA plays a large variety of important , often regulatory , roles in living organisms that extend far beyond being a mere intermediate one in protein biosynthesis . The elucidation of the functional roles of the plethora of newly discovered ncRNAs has thus become a central research interest in molecular biology . Recent advances in computational RNomics have resulted in numerous software packages that can be employed to detect ncRNAs with evolutionarily conserved secondary structures [7–12] . Two of these , EvoFold [10] and RNAz [9 , 13] , are efficient enough to be applied to genome-wide surveys in mammals [10 , 13] and other metazoan clades [14 , 15] . Both approaches start from multiple sequence alignments . While EvoFold uses the SCFG approach pioneered by qrna [7] , RNAz is based on evaluating the folding thermodynamics . Both approaches classify input alignments either as unstructured or as possessing a common RNA secondary structure; in the latter case they provide a prediction for the consensus structure of the aligned sequences . Just as in the case of proteins , ncRNA sequences can be grouped into families that are characterized by clear homologies . Usually the members in a family share functional characteristics as well as conserved sequence and structure motifs . Indeed , the RFAM database [16] compiles several hundred families of ncRNAs based on this observation . Examples include the individual snRNAs U1 , U2 , U4 , U5 , and U6 , 5S rRNA , RNAse P RNA , the RNA component of telomerase , more than a hundred families of snoRNAs , and several hundred microRNA families collected in mirbase [17] . In many cases , RNA families can be grouped together , forming a ncRNA class whose members have no discernible homology at sequence level , but still share common structural and functional properties . The best-known classes are tRNAs ( although it is well-established that all tRNAs derive from a common ancestor [18] ) , the two distinct classes of snoRNAs ( box H/ACA and box C/D ) , RNAse P and MRP RNAs , and microRNAs . It is thus natural to ask whether the many ncRNA candidates that have been predicted computationally can be grouped into families or even classes , and in particular , whether there is evidence for novel families and classes for which we have not yet seen experimentally verified representatives . As sequence similarity is often remote even within well-established RNA families , we cannot rely on pure sequence alignment techniques for this task . Indeed , it has been shown that sequence alignments of structured RNAs fail at pairwise sequence identities below about 60% [19] . Several different algorithmic approaches have been introduced in the past to determine structural similarities and to derive consensus structure patterns for RNAs that are too diverse to be alignable at sequence level . The corresponding software tools , such as MARNA [20] , PMmulti [21] , and RNAforrester [22] cannot be applied without modifications to the problem of clustering predicted structures from RNAz or EvoFold surveys , however . The main reason is that these ncRNA detectors are not guaranteed to find the complete ncRNA genes; rather they usually detect particularly conserved substructures and sometimes the predictions are contaminated with spurious predictions in the flanking sequences . Thus , a local structure-based alignment algorithm is necessary . This is already implemented in RNAforrester [22] , which is based on tree-alignment , and in the local sequence–structure alignment approach described [23] , which in addition can also detect structurally local motifs . A related approach detects exact local sequence structure patterns in O ( n2 ) [24] . However , all these approaches require a single known or predicted input structure . Tree-alignment and tree-editing in addition have only limited capabilities to repair incorrect base pairs . Tree-alignment is particularly restrictive in this respect since even broken arcs must be nested . As a consequence , RNAforrester tends to produce many alignment columns that contain mostly gap characters in the multiple alignment mode . In contrast , derivates of the Sankoff algorithm [25] solve the problem of simultaneous folding and alignment , which turned out to be more appropriate . However , the large number of predicted ncRNAs , several thousands in the case of nematode and urochordate genomes and close to 100 , 000 in the case of mammals , calls for more efficient variants of these algorithms . In this contribution we introduce LocARNA ( local alignment of RNA ) , a local pairwise structural alignment algorithm for pseudoknot-free RNA secondary structures , and its multiple version mLocARNA . ( m ) LocARNA is a Sankoff-style algorithm , similar to PMmulti , that is efficient enough to be used for large clustering of predicted ncRNAs . We have successfully tested the LocARNA-based clustering approach on the sequences of the RFAM-seed alignments to demonstrate the feasibility of the approach , and to evaluate the results . Furthermore , we use the data from a survey of the ascidians C . intestinalis and C . savignyi [14] to achieve the following goals . ( 1 ) We search for novel , clade-specific RNA families in Ciona , which is of interest in itself . ( 2 ) In doing so , we can increase the credibility of some of the predicted ncRNAs , since being part of larger family of related RNAs with similar structure reduces the likelihood of being a false positive prediction . ( 3 ) We improve the genome annotation by assigning additional ncRNAs to known families . ( 4 ) The inferred consensus structures of novel families form a starting point for subsequent searches in related organisms .
In this work , we set up a pipeline for automated clustering of ncRNAs ( or ncRNA candidates ) and semi-automated selection of novel , complex clusters of RNAs . The input is a set of RNAs R1 , … , RN , which are given by their sequences , and the output is a hierarchical clustering of these RNAs . In addition , we will generate a fast , presorted , and annotated overview of the clusters for further inspection by an expert . Our pipeline is built from the following steps: 1 . For each of the RNAs , we compute structural information using McCaskill's algorithm , implemented in RNAfold . This algorithm computes a matrix of pair probabilities based on a complete energy model of RNAs . 2 . The next step is to compute all pairwise alignments of the structurally annotated sequences using LocARNA . Note that this requires computation of O ( N2 ) pairwise sequence–structure alignments for determining the distance matrix . Note further that performing all pairwise comparisons cannot be reasonably circumvented or replaced in a full-featured clustering procedure . For genomic-scale datasets , O ( N2 ) comparisons are way too costly for most existing sequence–structure approaches . The computational efficiency remains crucial , even if this computationally most intensive procedure is distributed for parallel computation , which we do in a straightforward manner . As a result , we assign a LocARNA-alignment score ( i , j ) to each pair of RNAs ( Ri , Rj ) . 3 . A cluster-tree is generated by applying the weighted pair group method algorithm ( WPGMA ) , which is also known as average-linkage clustering , to a matrix of pairwise distances of the RNAs . There , the distances d ( i , j ) correspond directly to our LocARNA-scores . Instead of computing distances as maxij-score ( i , j ) , we define distances by where q is the x-quantile ( e . g . , x = 99% ) of all pairwise scores . This decision avoids the fact that exceptionally large scores influence the distance-transformation . In the resulting tree , internal nodes correspond to clusters of RNAs . Their heights correspond to the mean pairwise LocARNA-scores of their constituents and thus give a single-value measure of cluster quality . 4 . A good overview and a true quality assessment of the clusters can be best provided through multiple alignments of each cluster . We simultaneously construct all multiple sequence–structure alignments , i . e . , one for each cluster , by only O ( N ) runs of the pairwise alignment algorithm . This can be done by constructing the multiple alignments progressively , using the already constructed cluster-tree as guide tree . From each of the multiple alignments , we collect information that can guide a quality assessment of the cluster . We compute the mean pairwise sequence identity ( MPI ) and , using RNAalifold , the structure conservation index ( SCI ) , the consensus minimum free energy ( MFE ) , the consensus MFE structure , and the consensus base pair probabilities . Sorting the list of generated clusters by the quantities size of cluster , SCI , MPI , and MFE provide the expert with an automatically proposed order for his manual inspection of the clusters . The multiple alignment itself and the consensus structure information facilitate the selection of “interesting” clusters . This pipeline crucially depends on pairwise sequence–structure alignments . Therefore , we require the following algorithmic components which we describe in some detail in the next two sections , ( 1 ) LocARNA: Efficient Pairwise Local Sequence–Structure Alignment and ( 2 ) Local Multiple Sequence Structure Alignments . In ( 1 ) , we develop an efficient algorithm for high-quality pairwise alignments of RNAs that considers both sequence and structure information . For this purpose , the best results are achieved with Sankoff-style algorithms . We provide the new method called LocARNA , which is much more efficient than current approaches and uses base pair probabilities as structural input . Regarding ( 2 ) , for the selection of clusters , one important subproblem is to extract consensus structure information from the clustered RNAs , which is done by using RNAalifold . For producing the input for RNAalifold , we introduce the local multiple alignment method mLocARNA . For the pairwise alignments of RNA , we use our novel tool LocARNA , which computes local alignments of RNA . It is a Sankoff-style algorithm in the spirit of PMcomp , but goes beyond its ancestor by introducing local alignment and significantly improving the efficiency . The Sankoff algorithm [25] provides a general solution to the problem of simultaneously computing an alignment and the common secondary structure of the two aligned sequences . In its full form , the problem requires O ( n6 ) CPU time and O ( n4 ) memory , where n is the length of the RNA sequences to be aligned . In general , one can distinguish two variants of the Sankoff algorithms: programs such as foldalign [26 , 27] and dynalign [28] implement a more or less complete energy model for the RNA folding part . In contrast , PMcomp [21] assumes that a structure model for the two input sequences is already known and given in the form of weights for the individual base pairs . However , note that such a structure model is reasonably obtained using McCaskill's algorithm [29] , again on the basis of a full-featured energy model . Consider two sequences A and B with associated base pair probability matrices PA and PB , respectively . The goal is to compute a sequence alignment A of A and B together with secondary structure S on A . A consists of a set of ( mis ) matches written as pairs ( i , k ) , where i is a position in A , and k a position in B . The consensus secondary structure S for an alignment A consists of a set of quadruples ( ij;kl ) , where ( i , k ) ∈A and ( j , l ) ∈A are two matches in A , ( i , j ) is a base pair on sequence A , and ( k , l ) is a base pair on sequence B . Furthermore , denote by As the single-stranded part of the alignment; i . e . , if ( i , k ) ∈As then there is no pair ( j , l ) such that ( ij;kl ) ∈S or ( ji;kl ) ∈S . The goal is to determine the pair ( A , S ) that maximized the score function where and are base pair scores ( see below ) , σ:{A , C , G , U}2 → R is the similarity score for ( mis ) matches , γ is the gap score parameter , and Ngap is the number of insertions and deletions in the alignment A . Both PMcomp and our novel tool , LocARNA , use base pair scores that are derived from the base pairing probability matrices of the two individual sequences . More precisely , we use here where Pij is the equilibrium pairing probability as computed by McCaskill's algorithm [29] , p0 is the expected probability for a pairing to occur at random , and p* is the cutoff probability , below which the arcs are ignored . Formally , this is expressed by assigning −∞ as the weight in this case . The term is the log-odds score for having a specific base pairing against the null model of a random pairing , and is a normalization factor that transforms the weights to a maximum of 1 . The reason for this normalization is just that it is easier to balance the sequence score against the structure score . LocARNA improves the PMcomp approach in several ways . First of all , it uses a modified dynamic programming approach that allows us to utilize the fact that typically the number of significant base pairs does not grow with O ( n2 ) , i . e . , that the base pair probability matrices PA and PB are usually sparse . In particular , if p is constant for different n , then each base can take part in at most 1/p* , and thus O ( 1 ) base pairs . Hence , there are only m = O ( n ) significant entries in P . We define Dij;kl as the maximal similarity score of an alignment for the subsequences A[i . . j] and B[k . . l] with the additional condition that ( i j;k l ) is part of consensus secondary structure . To profit from the reduced number of significant base pairs in time and space complexity , we calculate and store only Dij;kl that correspond to significant base pairs . Due to this modification , we need to take special care to avoid redundant computation . Therefore , we compute the entries Dij;kl by fixing i and k and varying only j and l . We introduce the notation Di•;k• to denote the matrix slice where i and k are fixed . The efficient calculation of Di•;k• in O ( n2 ) time requires an auxiliary matrix M , where the entries Mij;kl are the optimal similarity score of subsequences A[i + 1…j] and B[k + 1…l] , and leads to a computation order that differs from PMcomp . Finally , the dynamic programming recursion for M and D takes the usual form of a Sankoff-style algorithm: The important observation is that the last , computationally most expensive , alternative in the M recursion needs to be evaluated only for PAj′j ≥ p* and PBl′l ≥ p* , and , analogously , D needs to be stored only for matching base pairs . We observe that Di•;k• depends only on Mi•;k• , which in turn can be computed from other Mi•;k• entries . Thus , we only need to store the entries of M for the current values of i and k , i . e . , O ( n2 ) entries . The recursion can therefore be evaluated in O ( m2 + n2 ) memory and O ( n2 ( n2 + m2 ) ) time . From the matrices M and D , we can now compute the score of the best global alignment as well as the score of the best local alignment . In our study , we are only interested in the latter . Global alignment is only explained for better understanding and for comparison to the global alignment algorithm PMcomp . The score of the global alignment can be computed by evaluating the recursion for M0j;0l , i . e . , the optimal global alignment score is M0|A|;0|B| . Recall that our main goal is to apply the procedure to the prediction of ncRNA detectors ( e . g . , such as RNAz ) as generated by genome-wide screens . These detectors are not guaranteed to find the complete ncRNA genes and usually detect conserved substructures . Moreover , the predictions can be contaminated with spurious predictions in the flanking sequences . Hence , we need local sequence–structure alignment . Concerning local alignment , in a Sankoff-style approach usually we compute a four-dimensional matrix of alignment scores for each pair of subsequences A[i…j] and B[k…l] . In this case , we could trivially obtain the best local alignment score by searching for the maximal score . In our case , however , we cannot apply this simple method , since we do not compute entries for all possible pairs of subsequences . Rather , we compute only scores for subsequences that are closed by ( significant ) base pairs or prefixes of them . Those scores are either stored in Dij;kl ( in the case of closing a base pair match ) or in Mij;kl . Instead , we will borrow , slightly tailored for our purpose , the trick of standard sequence alignment , which is to add an additional zero entry in the recursion for cutting off dissimilar prefix-alignments . Note that this assumes that the score parameters yield a score greater than zero only for similar subsequences . The best local alignment is then obtained as the maximal entry of the matrix . However , note that we must not change the recursion equations for all Mij;kl , which serve for computing some entry of D . Only for alignments of subsequences A[i…j] and B[k…l] , where at least one of the subsequences is not enclosed by a ( significant ) base pair , is it correct to cut off dissimilar prefix-alignments . All these cases are accounted for when considering the alignments of all pairs of prefixes of A and B , which are stored in the M0•;0• slice . Therefore , we use for local alignment the following variant of Equation 3 that extends Equation 3 only for the slice M0•;0• . Note that the entries M0j;0l will not be needed to compute any entry Di′j′;k′l′ . By adding a 0-entry in the calculation of M0j;0l , we ensure that entries in M0•;0• are nonnegative . Since negative scores are considered dissimilar , we thereby remove prefix-alignments that do not belong to the local alignment . The optimal local alignment score is then maxjl ( M0j;0l ) . The corresponding optimal alignment and consensus secondary structure can now be obtained by backtracing , i . e . , for local alignment we start from the maximal entry in M0j;0l and stop when similarity drops to its minimal value of zero . In addition , for every pair ( i j;k l ) in the consensus structure we have to recompute the Mi•;k• at a cost of O ( n2+m2 ) . Since there are at most O ( n ) pairs in the consensus structure , the cost of backtracing stays negligible . LocARNA is implemented in C++ , which results in a further performance gain relative to the Perl implementation of PMcomp . While it fully exploits speed and memory reductions that can be obtained by limiting possible consensus structures , additional performance gains are possible by restricting the possible sequence alignments . This is done , e . g . , in stemloc [30] by using “alignment envelops” . A similar but more easily implemented technique is used by CONSAN [31] , where high confidence matches ( “pins” ) are derived from local sequence alignments . The algorithm then considers only alignments that contain all pins . Based on the pairwise LocARNA algorithm , we construct a progressive multiple alignment method , mLocARNA , which is similar in spirit to PMmulti , the PMcomp-derived multiple alignment tool [21] . mLocARNA differs from PMmulti in the algorithm for computing the consensus base pairing probability matrix PA∘B for the combined alignment of A and B from the base pairing probability matrices of the subalignments ( or sequences ) A and B . For a pair of columns p , q in the alignment of A and B , PMcomp defines the combined base pair weight by where ip and iq are the positions corresponding to p and q in the subalignment A , respectively . kp and kq are defined analogously for subalignment B . This has the effect that whenever one subalignment contains a gap at p or q or has a very low base pair probability , then the structural information between p and q from the other subalignment is effectively lost . In consequence , PMmulti tends to remove most base pairs when aligning many sequences . To avoid this undesired effect , we introduce the new definition where and is defined analogously . As usual , the order of pairwise alignments is directed by a guide tree . We use for that purpose the sub-trees produced by the hierarchical clustering .
To evaluate the quality of our clustering approach , we have applied our procedure to the sequences in the RFAM seed alignments . Our test set consists of all seed sequences that have no more than 80% sequence identity and do not exceed 400 nt in length , resulting in 3 , 901 sequences from 503 families . Normally , quality measures such as sensitivity and specificity are defined for binary classification problems , while here we face the problem of comparing our hierarchical clustering with the family assignment in RFAM . In principle , there are two ways of looking at the problem , namely globally ( considering the complete set of clusters ) , and locally ( considering the quality for each family separately ) . Concerning the global view , the complete RFAM defines a partition of the set of all sequences into families ( or clusters ) , and we can compare the degree of agreement between the partition defined by our clustering with the partition defined by RFAM . Since we have a hierarchical clustering , different sets of clusters can be defined by cutting the tree at different thresholds ϑ , and we have to compare all these thresholds to find the set of clusters with the best agreement . The problem of comparing the partition defined by a given set of clusters ( generated by cutting the tree at some specific level ) with the partition defined by RFAM is now transformed into a classification problem as follows . We consider all possible pairs of sequences , and define the number of true positives ( ss ) as the number of sequence pairs from the same family that lie in the same cluster . Analogously , the number of false positives , false negatives , and true negatives are given by the number of pairs from different families but same cluster ( ds ) , same family but different clusters ( sd ) , and different families and different clusters ( dd ) , respectively . Sensitivity and specificity are then defined as usual , namely spec = dd/ ( dd + ds ) and sens = ss/ ( ss + sd ) . The receiver operating characteristic ( ROC ) , obtained by plotting the sensitivity against the false positive rate ( 1-specificity ) for different values of the cutoff ϑ , is shown in Figure 1 . A problem in the comparison with RFAM families is that different families exhibit very different diversity: some families consist only of closely related sequences while others accommodate significant variation in sequence and structure . Therefore , one should not expect that the RFAM family division can be modeled by using one fixed threshold ϑ for all families . We therefore consider a local , family-wise , criterion for the clustering quality . For a given RFAM family R and a cluster C , we define the recall r ( R , C ) as the fraction of members from R contained in C , i . e . , r ( R , C ) = |R∩C|/|R| . For each family and a given minimum recall 0 . 5 < r ≤ 1 , we can always determine the minimal threshold ϑ such that there is a unique cluster C with r ( R , C ) ≥ r . A measure of how well the clustering reconstructs the family R is then the associated precision p ( R , C ) = |R∩C|/|C| . An equal assessment of precision and recall is given with the F-measure: Averages are weighted by family size . Families that are only represented by one sequence do not contribute to the average as their precision is always 1 . Table 1 shows the average precision and F-measure weighted by family size for different minimum recall levels between 0 . 5 and 0 . 95 . If we require that least 70% of a family ( = minimal recall level ) are grouped within the same cluster level , we get in fact on average a recall of 80% . In this case , we observe on average 32% false positive sequences within this cluster . Of course , we have much better values for some families such as 5S rRNA , where we have a precision of 100% at a recall level of 95% . The results also show that we are able to correctly cluster larger RNAs as well . For the members of the SSU_rRNA_5 family ( accession code RF00177 ) included in our test set ( recall that we have restricted the length to at most 400 ) , 72 . 46% of them were clustered together in one single , pure cluster containing no other sequences . The sequences in this cluster have an average identity of 56 . 31% , and an average sequence length of 257 . The complete RFAM tree constructed with our method is given in File Collection S1 . Concerning the formation of classes comprising several families , this mainly makes sense for classes such as tRNAs and miRNAs which have a similar structure , but , for example , not for ribosomal RNAs where there are four structurally different families . The best classification is observed for the class of all tRNAs . They still have a precision of 96% at a recall level of 95% . Concerning the class of all miRNAs , they are ( not surprisingly ) grouped in several separate clusters . However , we have a large cluster comprising 85% of all 213 miRNAs and only 18% false positive sequences . An RNAz screen of six related gammaproteobacteria resulted in an ncRNA candidate set of 123 unique loci of the reference organism Escherichia coli . The screen follows the same pipeline as in [14 , 15] but includes a new approach to build multiple alignments . Only alignments with homolog sequences of at least three genomes , with maximal pairwise blast e-values of 1e-10 and a minimal length of 40 nt were retained for input to the RNAz pipeline . That the majority of ncRNA candidates could be annotated with known E . coli ncRNAs ( labeled with EC[ . . . ] in Figure 2 ) is not surprising , as the screen was set up with a restrictive e-value for the initial blast search . Further , only candidates with homologs in at least three gammaproteobacteria genomes are reported . This provides us with a second ncRNA candidate set to validate the clustering approach , which in contrast to the RFAM seed sequences in the earlier section , Evaluation of the Clustering Procedure , was detected by RNAz . A candidate was annotated to be a known E . coli ncRNA if their genomic regions overlap to at least 70% . If such an annotation was not available , a blast search against the RFAM database ( E < 1e-6 ) identified further homolog ncRNAs . In Figure 2 , the complete WPGMA tree is depicted . It is nicely seen that again tRNAs get grouped in one separate cluster . Even tRNAs coding for the same amino acid are mostly found within the same subclusters . At first glance , the distribution of rRNAs in Figure 2 is disappointing . Different families of rRNAs appear in several separate clusters; however , RNAz predictions for 16S and 23S do not fall into a single cluster . This distribution results from a shortcoming of the RNAz input screen rather than from a weakness of the clustering method . Since RNAz scores alignments in relatively short slices , large structured RNAs are in general not detected as a single contiguous locus . Rather , several substructures are recognized for both 16S and 23S RNAs , which to a certain extent depend on the exact location of sequence windows that are used for the RNAz scoring . As demonstrated in Figure 3 and in detail in File Collection S1 , corresponding substructures ( including features up to 800 nt in length ) from the different rRNA loci in the E . coli genome are correctly clustered together . The dataset resulting from the RNAz-based survey for conserved ncRNAs in the genomes of the ascidians C . intestinalis and C . savignyi [14] consists of 3 , 332 predicted structured RNAs , of which only about 500 could be annotated as members of well-known RNA families . The overwhelming majority of the known RNAs are the 301 tRNAs recognized by RNAz . Figure 4 summarizes the results of the clustering procedure . At first glance , the result might look disappointing as we find a large number of predictions that do not belong to any tight cluster . This is not surprising , however , given that we expect a very high noise level in this dataset . ( 1 ) The RNAz screen has an estimated false discovery rate of about 18% . ( 2 ) No attempts have been made to correct the fairly unreliable strand prediction of RNAz , which has an error rate up to 30% [32] . ( 3 ) We can expect that a significant fraction of structured elements have been predicted only partially . ( 4 ) Thermodynamic consensus structure predictions based on only pairwise alignments are far from perfect [19 , 33] . It is thus not surprising that only a fraction of the input data can be assigned to meaningful clusters . As expected , the largest and most prominent cluster comprises tRNAs . As discussed in some detail in [34] , this tRNA cluster is composed of subclusters corresponding to homologous tRNAs with common anticodons . Several other well-known multigene families are easily identifiable as structural clusters , including the U5 snRNAs , U3 snRNAs , and 5S rRNAs . Several families of multicopy genes with common secondary structure are present in the Ciona genomes [34] . Most of them are also readily identifiable in the structural cluster tree . Since these subclusters are already easily detectable on the sequence level , they are of little interest for the structured-based approach pursued here . A more interesting example is a cluster , Figure 5 , that contains two paralogs of mir-124 and one copy of let-7 microRNAs that were previously described in computational screens of C . intestinalis [35 , 36] , as well as good candidates for mir-126 and mir-7 . The other members of the cluster have no sequence similarity with known microRNA families compiled in miRBase release 9 . 0 ( blast E ≤ 0 . 001 ) . Both mir-124 candidates occur within introns of known mRNAs of C . intestinalis ( JGI2 . 0 ) , while mir-126 and mir-7 do not seem to be located in an intron . That a large cluster of known and putative miRNAs was detected demonstrates that annotation of ncRNA candidates is highly improved by structure-based clustering . The majority of cluster members could not be identified as miRNA candidates by sequence comparison alone [14] . Further , a comprehensive comparative screen for miRNAs across the metazoan species identified only a few homologs with high sequence similarity within the urochordates [36] , raising the question if there may exist a group of yet unknown miRNA families within the urochordates . Figures 6 and 7 highlight two novel clusters of structurally similar predictions for which no functional or class assignment is available . The neighbor-net graphs in the insets show the sequence distance within the example cluster . Since the sequence distance is on average larger than 0 . 5 , this confirms that the clusters are defined essentially based on structural similarities . While our examples usually contain some subsets of related sequences , overall there is little or no detectable sequence conservation so that the clusters could not have been detected by sequence similarity alone . Since many ncRNAs , in particular snRNAs , tend to form multigene families ( often evolving under some form of concerted evolution that keeps the family members nearly identical ) , a moderate copy number in the genome can be interpreted as supporting the hypothesis that the candidate is indeed a true ncRNA . In cluster 1384 , Figure 6 , for example , sequences with a well-conserved secondary structure but low sequence similarity are grouped . Nine of 11 sequences of cluster 1384 could be exactly mapped to the new C . intestinalis assembly JGI2 . 0 . The structural cluster contains three subclusters , 1378 , 1381 , and 1383 , that have overall structural features in common . All subclusters have three stem loops originating from one single multiloop as consensus structure . But their length and number of internal loops differ . Their grouping into the superclusters 1382 and 1384 are justified by compensatory mutations . Two sequences of subcluster 1378 and one of subcluster 1381 appear within an intron of Ciona-mRNA AK113484 . Whereas the two sequences of subcluster 1378 appear within the same copy of mRNA AK113484 on chr01p , the sequence in subcluster 1381 occurs in a copy on chr04q . Six different genomic copies of AK113484 exist in JGI2 . 0 , but none of the intronic regions where the ncRNA candidates are found are associated with repeats . This allows the conclusion that those ncRNA candidates are indeed functional ncRNAs as their sequences are highly diverged , whereas they share common structural features and appear within the same Ciona-mRNA . One sequence of subcluster 1383 occurs in an exon of the known protein coding Ciona-mRNA AK114007 . All other elements are intergenic or at least the corresponding mRNAs are not yet known . Cluster 1249 is also composed of highly divergent sequences but similar secondary structures . Two sequences of subcluster 1247 appear within an intron of the Ciona-mRNA AK174830 . Subclusters 1238 and 1245 contain one sequence occurring in an intron of Ciona-mRNA AK222260 and AK116291 , respectively . Clusters 1384 and 1249 are good candidates for novel classes of urochordate-specific ncRNAs , since none of the sequences has detectable ncRNA homologs in vertebrates .
Genome-wide studies , both experimental and computational , have uncovered tens of thousands of transcripts in higher Eukaryotes that have little or no protein-coding capacity . For a large subset of these , there is evidence for selection acting to preserve secondary-structure motifs . Many classes of functional RNAs , on the other hand , can be recognized based on structural similarities . It is thus natural to ask if the available data contain evidence for novel families and classes of structured RNAs , for which so far no representative has been well-characterized experimentally . To answer this question , it is necessary to cluster the candidate RNAs based on their structural features , a task that is computationally much harder than clustering based on sequence similarity . We present here a new tool , LocARNA , which implements a novel , more efficient variant of the Sankoff algorithm . We have demonstrated that LocARNA is fast enough to make structure-based clustering of thousands of putative structured RNAs feasible . The main reason for its superior efficiency is due to the prefiltering of the base pairs by their probability , and an efficient computation scheme that is able to profit from the reduced number of base pairs considered . The method is also robust enough to find significant clusters in fairly noisy , realistic data that contain a substantial fraction of false positive predictions . We have successfully tested the tool on the sequences of the RFAM seed alignments . The LocARNA implements a local sequence structure alignment method , which is required when applied to candidate ncRNA sequences where the exact region of interest is not exactly known ( of course , the tool can also be applied to global alignment problems ) . Clearly , there is a length dependency in the scores , which has several sources , one being the calculation of pair probabilities . This influences both pairwise alignment and the clustering , which implies that the ncRNAs to be clustered should not diverge too much in length . This is the case in many applications like the clustering of predicted ncRNAs . A more precise treatment of the different kinds of dependencies ( such as GC content ) is planned for a future version . Since perfect predictions and experimentally determined structures are not available , it is imperative to have a method that can identify clusters on imperfect structure prediction , although this also implies that we cannot hope for perfect pure clusters—some “contamination” and some sequences that fall elsewhere in the clustering are thus unavoidable . The ROC curve in Figure 1 shows that LocARNA indeed achieves this goal . The application of the tool to a dataset of more than 3 , 000 predicted structured RNAs in urochordates showed that the clustering approach not only recovers known RNA families and classes such as tRNAs , but also predicts several candidates for novel ncRNA classes . In some cases we find that additional sequences are identified as structural relatives of known RNA families . In this way we have identified , for example , a mir-126 and a mir-7 homolog that were not detected in previous computational studies . More importantly , however , we also find structure-based clusters that are candidates for novel , presumably urochordate-specific , RNA classes . We find that these clusters often contain subclusters consisting of multicopy sequences . Comparing this with the characteristics of several well-studied ncRNA families , in particular tRNAs , the snRNAs associated with the major spliceosome , and SL RNAs , lends further credibility to the hypothesis that these sequences indeed form a bona fide RNA class .
Accession numbers starting with RF ( e . g . , RF00177 ) are taken from the RNA families database RFAM ( http://www . sanger . ac . uk/Software/Rfam/ ) and denote different RNA families . Accession numbers starting with AK ( e . g . , AK113484 ) describe Ciona intestinalis mRNAs in the GenBank database ( http://www . ncbi . nlm . nih . gov/entrez/query . fcgi ? db=Nucleotide ) .
|
For a long time , it was believed that the control of processes in living organisms is almost only performed by proteins . Only recently , scientists learned that a further class of molecules , namely special RNAs , plays an important role in cell control . In consequence , research on such RNAs enjoys increasing attention over the last few years . These RNAs were called noncoding RNAs ( ncRNA ) , because , unlike most other RNAs , these molecules do not code for proteins . Due to recent research successes , one can predict a lot of potential new ncRNAs by comparing the genomes of related organisms . Technically , comparing such RNAs is challenging and computationally expensive , since related ncRNAs often show only weak similarity on the sequence level , but share similar structures . In the paper , we present the new method LocARNA for fast and accurate comparison of RNAs with respect to their sequence and structure . Using this method , we define a distance measure between pairs of ncRNAs based on sequence and structure . This is then used for combining RNAs into a cluster for identifying groups of similar RNAs in large unorganized sets of RNA . The final aim of such a comparison is to identify new classes of ncRNAs . We applied our clustering procedure to a previously published set of 3 , 332 predicted ncRNAs in the C . intestinalis genomes . In addition to rediscovering known classes of RNAs , e . g . , tRNAs , the method predicts microRNA candidates , and suggests several novel , experimentally uncharacterized classes of ncRNAs . For verification , we clustered about 4 , 000 RNAs of RFAM , which is a large database that contains RNAs with an already known classification into families . Our results show good performance of the presented structure-based clustering approach .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion",
"Supporting",
"Information"
] |
[
"ascidians",
"(sea",
"squirts)",
"eukaryotes",
"computational",
"biology",
"molecular",
"biology",
"animals"
] |
2007
|
Inferring Noncoding RNA Families and Classes by Means of Genome-Scale Structure-Based Clustering
|
Elucidation of the biological role of linker histone ( H1 ) and heterochromatin protein 1 ( HP1 ) in mammals has been difficult owing to the existence of a least 11 distinct H1 and three HP1 subtypes in mice . Caenorhabditis elegans possesses two HP1 homologues ( HPL-1 and HPL-2 ) and eight H1 variants . Remarkably , one of eight H1 variants , HIS-24 , is important for C . elegans development . Therefore we decided to analyse in parallel the transcriptional profiles of HIS-24 , HPL-1/-2 deficient animals , and their phenotype , since hpl-1 , hpl-2 , and his-24 deficient nematodes are viable . Global transcriptional analysis of the double and triple mutants revealed that HPL proteins and HIS-24 play gene-specific roles , rather than a general repressive function . We showed that HIS-24 acts synergistically with HPL to allow normal reproduction , somatic gonad development , and vulval cell fate decision . Furthermore , the hpl-2; his-24 double mutant animals displayed abnormal development of the male tail and ectopic expression of C . elegans HOM-C/Hox genes ( egl-5 and mab-5 ) , which are involved in the developmental patterning of male mating structures . We found that HPL-2 and the methylated form of HIS-24 specifically interact with the histone H3 K27 region in the trimethylated state , and HIS-24 associates with the egl-5 and mab-5 genes . Our results establish the interplay between HPL-1/-2 and HIS-24 proteins in the regulation of positional identity in C . elegans males .
Linker histone H1 and heterochromatin protein HP1 are involved in numerous processes ranging from stabilizing heterochromatin condensation to the regulation of gene expression [1]–[5] . As has been reported , a methylation mark on vertebrate histone H1 is specifically recognized by the chromodomain of HP1 . However , the exact biological role of HP1 binding to linker histone has not been determined [6] . The functions of HP1 and H1 proteins are mainly dependent on the cell type in which particular variants are expressed . Although the number of H1 ( 11 ) and HP1 variants ( 3 ) presents difficulties in studying the effect of H1 and HP1 depletion in mice , some data has emerged [3] , [7]–[10] . For example , loss of HP1β results in defective development of neuromuscular junctions and the cerebral cortex [10] , whereas depletion of three of eleven H1 genes causes lethality connected with a very broad range of defects in mice [11]–[12] . In ES cells , the lack of three somatic H1 variants leads to changes in nucleosome spacing and local chromatin compaction , and this is correlated with decreased levels of H3K27 trimethylation [11] . Additionally , H1 is necessary to establish and maintain the DNA methylation pattern in a subset of genes including the reproductive homeobox ( Rhox ) gene cluster [13] . C . elegans possesses eight linker histone variants and two HP1 homologues , HPL-1 and HPL-2 [14]–[16] . Mutation of hpl-2 results in defective vulval and germline development at elevated temperatures [15]–[17] . hpl-1 , in contrast to hpl-2 , does not have visible effects on C . elegans development at different temperatures , however , hpl-1 acts redundantly with hpl-2 to control larval development , somatic gonad development and vulval cell fate determination [17] . Our previous study revealed that HPL-1 recognizes the linker histone variant HIS-24 when it is mono-methylated at lysine 14 ( HIS-24K14me1 ) , similar to the situation in vertebrates [16] . Additionally , we showed that HIS-24 interacts with H3K27me3 [18] . The H3K27me3 modification correlates with a repressive chromatin state that inhibits expression of many developmentally regulated genes . This is consistent with studies of Hox loci demonstrating that enrichment of H3K27me3 recruits the binding of Polycomb group proteins ( PcG ) [19] . The Hox genes encode conserved homeodomain-containing transcription factors that control the positional identities of cells along the anterior–posterior axis [20]–[21] . The expression pattern of Hox genes appears to be regulated by two evolutionarily conserved PcG complexes , the ESC/E ( Z ) complex and the PRC1 complex . Both have been identified in flies and mammals and are linked to modulation of repressive chromatin structures [21] . The C . elegans Hox cluster consisting of lin-39 , ceh-13 , mab-5 and egl-5 ( orthologs of Drosophila Scr , labial , ftz and Abd-B , respectively ) is quite degenerated in comparison to Hox clusters in other species [22] but , as in mammals , is also globally repressed by Polycomb group ( PcG ) proteins [20] , [23] . Mutations in mes-2 and mes-6 , which encode the C . elegans ESC/E ( Z ) complex , result in ectopic expression of Hox genes [23] . A similar phenotype has also been observed in the absence of sop-2 or sor-1 genes . SOP-2 and SOR-1 form another C . elegans PcG-like complex which shares many structural and functional properties with the Drosophila PRC1 , and is involved in the global repression of Hox gene expression . Loss of sop-2 and sor-1 results in gross homeotic transformations [24]–[25] . To elucidate the function of H1 and HP1 related proteins in C . elegans , we decided to generate double and triple mutants , since hpl-1 , hpl-2 and his-24 deficient nematodes are viable , and since HIS-24K14me1 is recognized by HPL-1 [16]–[17] , [26] . We performed global transcriptional analyses of single , double and triple mutant animals , and we found that HPL-1/-2 and HIS-24 regulate a relatively small number of genes . We provide evidence that the methylated form of HIS-24 ( HIS-24K14me1 ) and HPL-2 are involved in the regulation of mab-5 and egl-5 expression by binding to H3K27me3 , although HIS-24K14me1 does not interact with HPL-2 [16] . Furthermore , we observed that HIS-24 and HPL-2 act in parallel pathway as MES ( PcG ) proteins , and loss of their activity causes defects of male tail structures . Overall , our data suggest a common and dual role for C . elegans H1 and HP1 , functioning both as chromatin architectural proteins and at the same time as modifiers of a small subset of genes . Furthermore , we provide the first direct evidence for redundant functions of H1 and HP1 in metazoan development .
C . elegans contains two related HP1 proteins ( HPL ) and eight linker histone variants [14]–[15] . Only one of the eight linker histone variants , HIS-24 is important for germline development , with its absence resulting in reduced fertility and de-repression of extrachromosomal transgenic arrays in the germline [14] . As we previously reported , the absence of HIS-24 did not affect protein levels of the other histone variants , in contrast to the mammalian H1 subtypes which are sufficient to compensate for the loss of a single linker histone [7] , [16] . Furthermore , we showed that C . elegans heterochromatin protein 1 variant , HPL-1 recognizes and binds the methylated form of HIS-24 [16] . Given the physical interaction of HPL-1 with HIS-24 mono-methylated at lysine 14 and their role in chromatin silencing and germline developmental processes [15]–[17] , we decided to study HPL and HIS-24 function in transcriptional regulation in C . elegans . It was of great interest to determine how the HPL subtypes ( HPL-1 and HPL-2 ) and HIS-24 affect gene expression . To determine the contribution of HIS-24 and HPL-1/-2 to the control of gene transcription , we compared the gene-expression profiles of single null mutations in the hpl-1 , his-24 and hpl-2 as well as profiles of hpl-1his-24 , and hpl-2; his-24 double , and hpl-2; hpl-1his-24 triple mutant animals in L4 larval stages grown at 21°C . We decided to use L4 larval stages because HIS-24 is the most abundant linker histone H1 variant at this stage according to mass spectrometry-based protein expression data ( Figure 1 ) . By microarray we observed very few changes in the gene expression profiles of either single , double , or triple mutants when compared with wild type animals at L4 larval stages . Among the 16 , 278 target probe sets assayed , we identified only modest changes in expression of just a small number of genes ( Figure 2A–2H , Table 1 ) . The majority of genes exhibiting changes were upregulated ( 6 . 5% ) in the absence of the three heterochromatin components HIS-24 , HPL-1 and HPL-2 , in contrast to 3 . 7% downregulated genes from a total of 16 , 278 genes analyzed ( FDR<0 . 05 ) suggesting that HPL-1/-2 and HIS-24 are not global repressors of transcriptional activity ( Table 1 ) . The deletion of both hpl-1 and hpl-2 genes caused up-regulation of 4 . 5% genes and downregulation of 2 . 1% of a total 16 , 278 genes when compared to wild type ( WT ) animals . As previously reported , HPL-2 binds to HIS-24K14me1 through its association with HPL-1 , and the heterochromatin proteins HPL-1 and HPL-2 play redundant roles in C . elegans development [16]–[17] . Considering these observations we compared transcriptional profiles between hpl-2 ( tm1489 ) ; hpl-1 ( tm1624 ) double mutants and hpl-2 ( tm1489 ) ; hpl-1 ( tm1624 ) his-24 ( ok1024 ) triple mutant animals . We found that 464 up-regulated ( 2 . 9% of 16 , 278 ) and 195 down-regulated ( 1 . 2% of 16 , 278 ) genes were commonly affected ( FDR<0 . 05; p-value<0 . 000001 for all pair-wise comparisons by hypergeometric tests ) ( Figure 2 , Table S1 ) . Among the 464 up-regulated genes we identified some significantly enriched in GO terms associated with growth regulation ( Fisher exact test ( FET ) P = 4×10−6 ) , determination of adult life span ( FET P = 2×10−6 ) , locomotion ( FET P = 0 , 003 ) , protein phosphorylation ( FET P = 0 , 04 ) , reproduction ( FET P = 0 , 05 ) and lipid storage ( FET P = 0 , 05 ) . The 195 genes that are down-regulated are enriched in GO terms associated with oxidation reduction ( FET P = 0 , 003 ) , embryonic development ( FET P = 0 , 002 ) and metabolic process ( FET P = 0 , 04 ) . We identified common response proteins including heat shock proteins ( HSP-12 . 3 , -12 . 6 , -16 . 2 and -17 ) , enzymes ( cytochromes ) of the P450 family involved in protection against toxins ( CYP-13A12 , CYP-33C4 , CYP-33D3 , CYP-34A2 , CYP-34A4 and CYP-34A9 ) , metabolic enzymes such as the fatty acid-coenzyme A ( CoA ) synthetase ACS-1 and the fatty acid/retinol binding proteins FAR-5 , -7 ( Table S1 ) . Furthermore , we observed the induction of oxidative stress proteins such as glutathione S-transferases ( GST ) and genes commonly associated with increased stress resistance – for example , the mitochondrial sod-3 superoxide dismutase gene ( Table S1 ) . In conclusion , deletion of the different HPL variants and HIS-24 caused an alteration in the expression of a limited number of genes , different in each HPL variant and HIS-24 . Most of the genes are affected by a single HPL variant and HIS-24 , supporting the theory that HPL isoforms or HIS-24 play specific roles in gene expression . Nonetheless , a proportion of genes are altered by more than one HPL variant as well as HIS-24 , suggesting redundant roles for HIS-24 and HPL variants , and for HPL-1/-2 may also exist . In parallel to microarray analysis we investigated the biological role of HIS-24 and HPL proteins in C . elegans . For morphological defects we scored hpl-1 ( tm1624 ) his-24 ( ok1024 ) , and hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) double mutants as well as hpl-2 ( tm1489 ) ; hpl-1 ( tm1624 ) his-24 ( ok1024 ) triple mutant animals . In particular , we focused on germline nuclei morphology , hermaphrodite vulval development and the somatic patterning of the male tail since these tissues are known to be affected by mutations in chromatin factors , and HPL-2 influences vulval cell fate specification in the synMuv ( synthetic multivulva ) pathway [14]–[15] , [27] . We found that the deletion of hpl-2 ( tm1489 ) together with his-24 ( ok1024 ) results in synergistic non-lethal defects of vulval cell fate specification ( everted vulva , multivulva ) and sterility at 21°C , and at 25°C ( Table 2 ) . While the observed phenotypic effects at 21°C were minor in contrast to the situation at 25°C , it is tempting to speculate that the effects can be also modulated through unknown mechanisms , environmental cues ( temperature ) , which in itself may also lead to significant side-effects . Additionally , decreased brood sizes were observed in hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) double and hpl-2 ( tm1489 ) ; hpl-1 ( tm1624 ) his-24 ( ok1024 ) triple mutant animals grown at 21°C ( Figure 3 ) . The brood size of the hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) was strongly decreased by 35% of wild type worms , and was further decreased to about 50% in the hpl-2 ( tm1489 ) ; hpl-1 ( tm1624 ) his-24 ( ok1024 ) triple mutant animals ( Figure 3 ) . These results were consistent with our microarray data analysis that revealed differential expression of genes involved in the embryonic development or reproduction ( Table S1 ) . Furthermore , we observed several defects in the morphology of the somatic gonad of hpl-1 ( tm1624 ) his-24 ( ok1024 ) double mutant animals grown at 21°C . In wild type , single mutant and hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) double mutant the gonad arms form an U-shaped structure ( Figure 4A–4D ) . In contrast , in the double mutant hpl-1 ( tm1624 ) his-24 ( ok1024 ) 25% of gonad arms ( 161 of 642 ) form a loop ( Figure 4E ) . These results suggest that both proteins HIS-24 and HPL-1 are involved in the somatic gonad development whereas HIS-24 and HPL-2 influence vulva cell fate specification and reproduction ( Table 2 ) . Since HPL-2 and HIS-24 are required for germline development and for the chromatin based germline-specific silencing mechanism [14]–[15] , [26] , we asked whether they influence the structure of nuclei . In-depth analysis revealed that the germline nuclei of hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) double mutants differ in size and morphology when compared to single mutants or to wild type worms grown at 21°C ( Figure 4F–4I , 4L ) . The observed chromatin of 86% of gonad arms ( 36 of 42 ) had a more open , relaxed structure suggesting that HIS-24 and HPL-2 play a function in chromatin condensation in the germline ( Figure 4J , 4M ) . To assess the specific requirements for HIS-24 among the H1 isoforms , we also tested hpl-2 ( tm1489 ) ;hil-3 ( ok1556 ) double mutant strain to determine if the observed changes in the chromatin compaction is linker histone variant specific ( Figure 4K ) . As shown , loss of hpl-2 and linker histone variant hil-3 did not cause defects in chromatin compaction in contrast to hpl-2; his-24 strain . In addition , we also did not observe involvement of HPL-2 and HIL-3 on brood size ( Figure 3 ) . To determine if the loss of HIS-24 and HPL proteins also influence chromatin histone modifications as well as core histone H3 level , we performed western blot analysis of mutant animals . No gross changes were observed in the methylation and core histone H3 levels using antibodies directed against H3K9me3 , H3K27me3 , and H3 ( Figure S1 ) . In addition , we did not detect changes in chromatin modification marks on a cellular level by immunofluorescence ( data not shown ) indicating that the observed effects of chromatin compaction are not correlated with alterations of histone modifications in hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) double mutant animals . Loss of hpl-1 , -2 and his-24 function results in changes of transcriptional regulation of genes encoding nuclear hormone receptor family genes ( nhr-60 , nhr-156 ) , transcription factors ( miz-1 , zip-3 , zip-8 , madf-2 ) , homeobox ceh-82 and homeodomain lim-7 genes ( Table S1 ) . Moreover , hpl-2 regulates lin-39 Hox gene expression in vulval precursor cells ( VPCs ) [27] . Therefore we tested whether hpl-1 , -2 and his-24 genes are involved in the regulation of Hox gene expression during the somatic patterning of the male tail . The wild type male tail possesses nine pairs of bilateral sensory rays that function in locating and mating with hermaphrodites . Normally , the posterior hypodermal blast cells V5 and V6 produce six pairs of rays ( ray 1 to ray 6 ) , while the blast cell T gives rise to the three rays ( rays 7–9 ) [23]–[25] . We found that mutations in both his-24 and hpl-2 ( 37% , 51 of 73 males with defected rays ) as well as in his-24 , hpl-1 and hpl-2 ( 83% , 76 of 107 males with defected rays ) cause abnormalities in patterning of blast cells V that result in fused and atypical ( under-developed ) rays , while the single and hpl-1; hpl-2 and hpl-1 his-24 double mutations have normal development of rays ( Table S2 , Figure 5A–5E , 5G ) . Although hpl-1 mutation alone or in combination with his-24 or hpl-2 had no visible effect on the male tail at 21°C ( Figure 5C , 5G–5H ) , it appeared to be partially redundant in combination with hpl-2 and his-24 double mutations . As Figure 5J and Table S2 show , the number of under-developed rays is significantly increased ( up to 42% , 39 of 107 males ) in the hpl-2 ( tm1489 ) ; hpl-1 ( tm1624 ) his-24 ( ok1024 ) triple mutant compared to the hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) double mutant males ( 13% , 18 of 73 males ) ( Figure 5I ) . This synergism suggests that hpl-1 only in combination with his-24 and hpl-2 plays functions in the patterning of the male tail . We also tested hil-3; hpl-2 double mutant animals for the mail tail phenotype . We did not observe any defects in the patterning of the male tail of hil-3; hpl-2 double mutant animals in contrast to hpl-2; his-24 animals suggesting that HIS-24 ( in combination with HPL-2 ) specifically affects the patterning of the mating structures in C . elegans ( Figure 5F–5I ) . In agreement with previous observations we analyzed the ability of his-24 , hpl-1 and hpl-2 genes to regulate mab-5 and egl-5 expression [28]–[29] . Interestingly , these two Hox genes are required for V ray development [23] and mab-5 was slightly up-regulated in our microarray analysis of hpl-2 ( tm1489 ) ; hpl-1 ( tm1624 ) his-24 ( ok1024 ) mutant animals ( data not shown ) . We compared expression of egl-5::gfp and mab-5::gfp reporter genes in wild type animals and in combination with his-24 ( ok1024 ) , hpl-1 ( tm1624 ) and hpl-2 ( tm1489 ) background mutations ( Figure 6 ) . We observed that mab-5::gfp reporter is ectopically expressed in approximately 30% of early L3 stage of hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) double mutant males scored ( n = 100 ) ( Figure 6A , 6B ) . Altered expression of this reporter was also detected in adult males . Similarly , about 80% of L3 stage of hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) double mutant males ( n = 100 ) displayed ectopic expression of EGL-5::GFP protein in two daughters of ray precursors anterior to R4 , R5 and R6 sublineages ( Figure 6C–6E ) . We did not observe any significant enhancement of the ectopic expression of mab-5 and egl-5 Hox genes in hpl-1 depleted hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) double mutant animals . For the crossing with mab-5::gfp transgenic strain we did not use triple mutant animals due to hpl-1 his-24; hpl-2 phenotype ( sterile worms , worms with everted vulva or multivulva; Table 2 ) . To verify HPL-1 depletion directly and to examine the extent of HPL-1 knockdown we tested the hpl-1 depleted hpl-2; his-24 mutant animals for presence of HPL-1 on the western blot . We found that hpl-1RNAi strongly reduces HPL-1 level compared to the controls ( Figure S2 ) . Since mutations in hpl-2 and his-24 affect transgene expression in C . elegans [14]–[15] we assessed the expression level of the endogenous EGL-5 in hpl-2; his-24 double mutant males . Western blot of hpl-2; his-24 double mutant males probed with EGL-5 antibody revealed an increased level of endogenous EGL-5 protein of predicted size ( 26 kDa ) compared to EGL-5 level of wild type C . elegans and egl-5::gfp transgenic line ( Figure 6F ) [29] . Altogether , these results suggest that HIS-24 and HPL-2 silence the Hox gene cluster , either by general repression of the transcriptional activity , or through a specific biochemical and structural function in Hox gene silencing . Since HIS-24 and HPL-2 are required for inhibiting the ectopic expression of mab-5 and egl-5 Hox genes , we tested if HIS-24 and HPL-2 bind directly to their promoters in vivo and therefore regulate egl-5 and mab-5 transcription . The primer sets used for quantitative ChIP-PCR ( qChIP-PCR ) analysis were directed to the promoters , introns and 3′UTR regions of mab-5 and egl-5 genes . Remarkably , mab-5 and egl-5 are tightly clustered on chromosome III , suggesting that chromatin structure coordinately regulates the expression of these genes ( Figure 7A ) . qChIP-PCR analysis revealed that HIS-24 is indeed associated with the promoters and introns of mab-5 and egl-5 genes ( Figure 7B ) . In contrast , we did not see any HIS-24 binding to 3′UTR regions ( Figure 7B ) . However they are occupied by H3 ( Figure 7D ) . As shown , the anti-HIS-24 antibody binds with higher affinity to egl-5 and mab-5 genes than the anti-HIL-4 antibody , which is cross-reactive to C . elegans linker histone variants [14] ( Figure 7C ) . Next , to verify the specificity of the HIS-24 binding to Hox genes , we tested the HIS-24 binding to mab-5 gene ectopically expressed in sor-1 background mutation . As previously reported , SOR-1 ( together with SOP-2 ) shares many structural and functional properties with the PRC1 complex , and is involved in the global repression of egl-5 or mab-5 Hox gene expression [25] . As shown , we detected a significantly decreased level of HIS-24 at this region compared to the situation in wild type animals , implicating that HIS-24 enables mab-5 transcriptional repression , thereby influencing its expression ( Figure 7D ) . Additionally , we observed lower levels of histone H3 occupancy at the mab-5 promoter in sor-1 background mutation than in wild type animals , suggesting that the difference in H3 levels could be due to the nucleosome free region that forms at high levels of expression ( Figure 7D ) . In addition , mab-5 promoter and intron regions in the his-24 mutant animals showed decreased enrichment of the histone H3 than in wild type animals , suggesting that binding of H3 and HIS-24 can be positively correlated at regulatory regions . In comparison , the H3 changes at 3′UTR region of mab-5 in sor-1 and his-24 background mutation were relatively mild than in wild type animals ( Figure 7D ) . Unfortunately , we have failed so far to detect HPL-2 at this region using direct ChIP approach . Hox genes are transcriptionally repressed by the evolutionally conserved Polycomb group ( PcG ) proteins through the H3K27me3 mark in a lineage specific fashion [30]–[31] . In Drosophila , a member of the Polycomb group ( PcG ) , the H3K27 histone methyltransferase E ( Z ) has been identified as a stable repressor of Hox genes [32] . In C . elegans , orthologs of the PcG chromatin repressors E ( Z ) and ESC , namely MES-2 and MES-6 influence expression of Hox genes and male tail development [23] . Since Polycomb group ( PcG ) proteins ( MES-2/3/6 complex ) are involved in the repression of Hox genes , we performed genetic epistasis analysis of mes-2- and mes-3-depleted triple mutant animals [23] , [33] . Interestingly , hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) double as well as hpl-2 ( tm1489 ) ; hpl-1 ( tm1624 ) his-24 ( ok1024 ) triple mutant males on mes-2 or mes-3 feeding plates showed an increased number of ectopic rays ( ∼2-fold ) and defective rays in comparison to mes-3- or mes-2 - depleted double mutant males ( Figure 8A , 8C; Table S2 ) . As shown , loss of HPL and HIS-24 together with depletion of mes-2 or mes-3 resulted in additive defects implying that HPL and HIS-24 act in parallel pathway as MES-2 or MES-3 . We also phenotyped hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) double and hpl-2 ( tm1489 ) ; hpl-1 ( tm1624 ) his-24 ( ok1024 ) triple mutant males on sop-2 feeding plates ( Figure 8B , 8C , Table S2 ) . As previously reported , SOP-2 forms a novel PcG-like complex that may function analogously to PRC1 in C . elegans and regulates expression of Hox genes [25] . Homologs of SOR-1 and SOP-2 are not found in other organisms , including even the very closely related C . briggsae suggesting a C . elegans specific mechanism on an essential global gene regulatory system [25] . Remarkably , we did not observe any influence of SOP-2 depletion in the hpl-2; his-24 double and hpl-2; hpl-1 his-24 triple mutant background suggesting that sop-2 appears to be epistatic to hpl-2; his-24 deletion . Recently , we have reported that HIS-24 specifically interacts with H3K27 trimethylated and H3K27 unmodified peptides [18] . While HPL-1 and HPL-2 were able to pull down native HIS-24K14me1 , and HPL-2 failed to bind either modified or unmodified HIS-24 peptides in vitro , we asked whether HPL-2 and HIS-24K14me1 repress the transcription of egl-5 and mab-5 genes by binding to H3K27me3 [16] , [18] . By peptide pull down assay ( PD ) we observed that HIS-24K14me1 interacts preferentially with H3K27me3 peptide when compared to the unmodified , mono- or di-methylated H3K27 peptides , and conversely , native H3K27me3 binds only the methylated form of HIS-24 peptide ( Figure 9A ) . Furthermore , we found strong preference of HPL-2 for the trimethylated form of H3K27 , as well as for H3K27me2 and H3K9me2/3 as previously reported ( Figure 9A ) [16] . No interaction was observed between H3K9me0/1 or H3K27me0/1 . We confirmed the results obtained from peptide pull down ( PD ) by an immunoprecipitation ( IP ) approach using antibodies raised against different chromatin modification marks and lysates of wild type animals ( Figure 9B ) . Additionally , we were able to pull down native H3K27me3 using a GFP-antibody directed against GFP-tagged HPL-2 and HIS-24 ( Figure 9C ) . As a control we used GFP expressed protein under the his-24 promoter to demonstrate the specificity of HPL-2 and HIS-24 binding to H3K27me3 ( Figure 9C ) . To confirm that HPL-2 and HIS-24 indeed display H3K27me3 binding , we expressed HPL-2 and HIS-24 in E . coli . We did not detect the interaction of HPL-2 with H3K27me3 in contrast to HIS-24 , suggesting that additional factors ( transcription factors , RNAi machinery , post-translational modifications of HPL-2 ) are involved in the mediation of HPL-2 binding to H3K27me3 ( Figure 9D , 9F ) . In the case of HIS-24 we detected strong preference for H3K27me peptides apart from H3K27me1 ( Figure 9D ) . The differences in the binding to H3K27me3 between bacterially expressed HIS-24 and native HIS-24 can be explained by the fact that bacterially expressed proteins are not methylated and only the methylated form of HIS-24 binds specifically the H3K27me3 . Finally , to exclude that the binding of HPL-2 to H3K27me3 takes place via interaction with the C . elegans HIS-24 , we repeated the pull downs using extracts obtained from his-24 ( ok1024 ) mutant animals ( Figure 9E ) . We detected a preference of HPL-2 for H3K27me3 independently of HIS-24 however this binding was reduced compared to binding of HPL-2 to H3K27me3 in the presence of HIS-24 ( Figure 9B , 9E ) . To assess whether the methylated form of HIS-24 has a causal role in the observed changes of the male tail morphology , we generated his-24::gfp and his-24K14A::gfp transgenic worms in the hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) mutant background . We observed that the restoration of HIS-24 levels by expression of HIS-24::GFP rescued the male phenotype and the fused/missing rays were down nearly to zero in the transgenic line ( Figure 10 ) . Importantly , the nonmethylatable HIS-24K14A::GFP mutant failed to rescue the wild type rays development in hpl-2; his-24 animals , suggesting that HIS-24 methylation at lysine 14 is necessary to regulate male tail development ( Figure 10B , 10D , 10E ) . These results also imply that , at 21°C , hpl-2 and his-24 play a redundant role in the regulation of positional identity in the C . elegans males . Importantly , the analysis of transgene expression at the cellular level by immunostaining and immunoblotting of the rescued hpl-2 ( tm1489 ) ; his-24 ( ok1024 ) animals verified that the exogenous HIS-24K14A::GFP mutated form was expressed at a level comparable to that in animals carrying HIS-24::GFP wild type form ( Figure 10C ) .
Our microarray analyses support a role of H1 and HP1 in specific gene regulation , rather than a general repressive function [34]–[36] . Despite global changes in chromatin compaction and synergism of HIS-24 and HPL in aspects of many developmental processes we observed very few and slight changes in gene expression profile of mutants when compared with wild type animals . We detected a set of shared up- and down-regulated genes by HIS-24 and HPL suggesting that redundant roles for HIS-24 and HPL also exist . The relatively small number of regulated genes in observed triple mutant animals may indicate that HPL proteins and HIS-24 serve to fine-tune the regulation of key genes during development or differentiation . This model can be explained by the fact that the sequential arrangement of the linker histone HIS-24 and HPL-2 on the chromatin fibre might influence higher-order chromatin structure and effect nucleosome positioning , and stability [36] . It is possible that the different HPL subtypes and HIS-24 confer subtle differences in the properties of the chromatin fiber which allow for quantitative modulation of gene expression [34] , [35] . Although the changes in gene transcription are subtle , we think that even 1 . 5-fold differences in expression can contribute to the marked phenotypic consequences we observed . We demonstrated that HIS-24K14me1 , together with HPL-2 , has a causal role in transcriptional silencing of egl-5 and mab-5 . We propose that HPL-2 and HIS-24K14me1 may serve as essential protein components in establishing and/or maintaining the repressive chromatin state at the selected Hox genes through their interactions with H3K27me3 . While we did not observe any phenotypic effects on male tail development either in hpl-2; hpl-1 nor in hpl-1 his-24 background , we speculate that HPL-2 acts redundantly with HIS-24K14me1 to regulate the positional identity in the C . elegans males . Loss of the two heterochromatin components , HIS-24K14me1 and HPL-2 , causes significant changes in chromatin structure affecting Hox gene expression in C . elegans . However , since no interaction of HPL-2 and HIS-24K14me1 was observed in immunoprecipitation experiments , it is possible that HPL-2 together with HIS-24K14me1 might be a part of the same protein group involved in the regulation of Hox gene expression . The high degree of redundancy between his-24 and hpl-2 in Hox gene regulation might indicate that these two proteins are the only readers acting in parallel to perform the same role in translating the effects of histone H3K27 trimethylation . However , since we have failed so far to detect HPL-2 at the Hox gene region using direct ChIP approach , it is possible that the mechanisms by which HPL-2 regulates mab-5 and egl-5 might be indirect , involve intermediate factors ( RNAi machinery , transcription factors ) and depend on an architectural level in the cellular context . In mammals , H1 regulates Hox gene activation by promoting DNA demethylation [13] . Although C . elegans does not possess methylated DNA , we speculate that H1 can still influence Hox gene regulation and , together with HPL-2 , regulate Hox gene expression as a part of the PcG silencing complex . The interaction of HPL-2 and HIS-24K14me1 with H3K27me3 can regulate the Hox gene in parallel pathway as MES-2 or MES-3 , and can be directed to specific parts of the genome . Notably C . elegans HP1/HPL-2 does not follow the H3K9me2/me3 code [37]–[41] but it is sufficient to recognize , and to bind H3K27me2/me3 . Remarkably , HIS-24 is required for optimal HPL-2 binding to H3K27me3 in vivo . Interestingly , some PcG proteins containing a chromodomain similar to that found in C . elegans HPL-2 and mammalian HP1s have been shown to bind H3K27me3 [30] , [42] . Overall , these and our previous results implicate that HPL and HIS-24 share some common functions even though there are differences among these proteins [16]–[17] , [26] . We conclude that a combination of the H3K27me3 methylation mark , HPL-2 and HIS-24K14me1 could be a major factor in the establishment of stable patterns of selected homeotic gene expression .
Nematodes strains were cultured and genetically manipulated as previously described [43] . The Bristol strain ( N2 ) was used as wild type . The following strains , obtained from the Caenorhabditis Genetics Center ( CGC ) , were used in this study: his-24 ( ok1024 ) X , hil-3 ( ok1556 ) X ( both strains outcrossed 1× ) , hpl-1 ( tm1624 ) X ( outcrossed 4× ) , hpl-2 ( tm1489 ) III ( outcrossed 4× ) . Transgenic strain ( transcriptional reporter ) expressing GFP under the control of the his-24 promoter was kindly provided by BC C . elegans Gene Expression Consortium , Canada . The double mutants hpl-1 ( tm1624 ) X his-24 ( ok1024 ) X , hpl-2 ( tm1489 ) III; his-24 ( ok1024 ) X , hpl-2 ( tm1489 ) III; hil-3 ( ok1556 ) X and the triple mutant strain hpl-2 ( tm1489 ) III; hpl-1 ( tm1624 ) X his-24 ( ok1024 ) X were generated by crossing . his-24::gfp ( stable integrated EC602 strain [26] ) and his-24K14A::gfp transgenic strains were crossed with the hpl-2 ( tm1489 ) III; his-24 ( ok1024 ) X . The generation of his-24K14A::gfp transgenic strain was previously described [16] . For the reporter gene analysis following transgenic strains: EM599 [egl-5::gfp; him-5 ( e1490 ) V; lin-15B ( n765 ) X; bxIs13] , OP27 [unc-119 ( ed3 ) III; wgIs27] , OP54 [unc-119 ( ed3 ) III; wgIs54] and HZ111 [mab-5::gfp; muIs16 II; sor-1 ( bp1 ) /qC1 dpy-19 ( e1259 ) glp-1 ( q339 ) III; him-5 ( e1490 ) V] , kindly provided by CCG , were used . The brood size was scored as previously described [14] . All C . elegans strains were maintained at 15°C or at 21°C , unless otherwise specified . C . elegans H1 extraction was performed as previously described [16] . Worms from wild type strain and the mutant worms were fixed and stained as previously described [26] . Gonads of worms were stained with fluorescent dye 4′ , 6′-diamidino-2-phenylindole ( DAPI ) diluted 1∶1000 . The slides were mounted with Vectashield Mounting Medium and analyzed by using Leica DMI 6000 microscope . Microarray analysis from two biological replicates was performed as previously described [16] , [44] . In brief , 80 to 100 animals in L4 larval stage raised at 21°C were used . The gene expression fold change was calculated from the duplicate microarray data . The fold change cut-off was 1 . 5 from 2 biological replicates . Abnormalities of rays were identified in single , double and triple mutant males in comparison to the wild type worms . Animals were transferred on agar pads ( 2% agarose ) and examined with differential interference contrast ( DIC ) , using Leica DMI 6000 microscope . The number of rays , their position in relation to the anterior-posterior body axis and their shape served as basics of the analysis . Rays which were found outside of their normal formation region were defined as ectopic . RNAi feeding experiment was performed in 50 mm NGM feeding plates ( NGM plates with 100 µg/ml ampicillin , 1 mM IPTG ) . him-14 ( RNAi ) ( CGC , USA ) , hpl-1 ( RNAi ) , mes-2 ( RNAi ) and mes-3 ( RNAi ) bacterial clones ( Sanger Institute , UK ) were grown overnight at 37°C in LB medium with 100 mg/ml ampicillin and were spotted onto 50 mm NGM plates . Mixed stage L3 and L4 mutant larval worms were transferred onto feeding plates and incubated at 21°C through several generations . Males were examined on the agar pads using Leica DMI 6000 microscope . Male progeny were scored for the presence of ectopic , under-developed rays and/or ray fusions . L3 stage and adult animals from each line were mounted on the agar pads and examined under Leica DMI 6000 microscope . Males were scored for the presence of ectopic EGL-5::GFP or MAB-5::GFP expression . ChIP was performed as previously described [45] with several modifications . Mixed stage L4 and adult worms were homogenized in ice-cold FA lysis buffer ( 50 mM HEPES/KOH pH 7 . 5 , 1 mM EDTA , 1% Triton X-100 , 0 . 1% sodium deoxycholate; 150 mM NaCl ) with complete protease inhibitor cocktail ( Roche ) and 1% Triton X-100 using liquid nitrogen . Worm lysate was sonicated with a Branson Digital Sonifier using following settings: 30% amplitude for 3 min total . Protein concentration of the extract was determined by the Coomassie Plus ( Bradford ) Protein Assay . Worm extract was incubated with the following antibodies: anti-H3 ( Abcam 1791 ) , anti-H3K27me2 ( Upstate 07-322 ) , anti-H3K27me3 ( kindly provided by T . Jenuwein ) , anti-H3K9me2 ( Abcam 1220 ) , anti-H3K4me3 ( Abcam 1012 ) , anti-GFP ( Roche ) and anti-HIS-24 . Proteins were immunoprecipitated using G-agarose beads ( Pierce ) . mab-5 and egl-5 genes were detected by qPCR using iCycler iQ™ Multi- Color real time PCR detection system ( Bio-Rad ) . Primer sequences are available on request . Peptide pull downs were performed as previously described [46] . 10 µg of each biotinylated peptide was coupled to streptavidin- agarose beads ( Pierce ) . For peptide binding experiments following peptides were used: H3 mono- , di- or trimethylated at K9 , H3 mono- , di- or trimethylated at K27 , H3 unmethylated at K27 , HIS-24 monomethylated at K14 and HIS-24 unmethylated at K14 . Peptides were generated by Squarix ( Germany ) . Worm extracts were incubated for 2 h at 4°C with the beads ( constant rotation ) . Beads were washed six times with PD 150 buffer ( 20 mM Hepes pH 7 . 9 , 150 mM KCl , 0 . 2% Triton-X 100 , complete protease inhibitor cocktail , 20% glycerol ) . Bounded proteins were separated on gradient NuPAGE SDS gel ( 4–12% ) . C . elegans lysates were prepared and analyzed by western blot as previously described [16] , [18] . Mixed populations of L4 worms carrying the hpl-2::gfp transgene or wild type worms were homogenized [47] . About 1 . 5 mg of total precleared protein was incubated with following antibodies GFP-TrapR –A beads ( Chromotek , Germany ) , anti- H3 ( Abcam 1791 ) , anti-H3K27me2 ( Upstate 07-322 ) , anti-H3K27me3 ( kindly provided by T . Jenuwein ) , anti-H3K9me2 ( Abcam 1220 ) , anti-H3K4me3 ( Abcam 1012 ) , anti-H3K9me3 ( Abcam 8898 ) or anti-H4K20me3 ( Abcam 9053 ) at 4°C overnight . Next , the complexes were washed six times with PD150 buffer for 5 minutes at 4°C ( 20 mM Hepes , pH 7 . 9; 150 mM KCl , 0 . 2% Triton X-100 , 1× Protease Inhibitor ( Roche ) , 20% glycerol ) . Finally , the immunoprecipitated proteins were resolved on NuPAGE SDS gradient gel ( 4–12% ) and western blotted with antibodies against H3K27me3 ( 1∶20 000 dilution ) , HPL-2 ( 1∶2000 dilution; kindly provided by F . Palladino ) , HIS-24K14me1 ( 1∶10000 dilution ) and GFP ( Roche; 1∶20000 dilution ) . The pGEX HPL-2a plasmid ( kindly provided by F . Palladino ) and HIS-24 pet3a plasmid were expressed in E . coli BL21 ( DE3 ) and the recombinant proteins were used for the peptide pull down assay . The microarray data can be found in the Gene Expression Omnibus ( GEO ) of NCBI through accession number GSE33339 .
|
Linker histone ( H1 ) and heterochromatin protein 1 ( HP1 ) play central roles in the formation of higher-order chromatin structure and gene expression . Recent studies have shown a physical interaction between H1 and HP1; however , the biological role of histone H1 and HP1 is not well understood . Additionally , the function of HP1 and H1 isoform interactions in any organism has not been addressed , mostly due to the lack of knockout alleles . Here , we investigate the role of HP1 and H1 in development using the nematode C . elegans as a model system . We focus on the underlying molecular mechanisms of gene co-regulation by H1 and HP1 . We show that the loss of both HP1 and H1 alters the expression of a small subset of genes . C . elegans HP1 and H1 have an overlapping function in the same or parallel pathways where they regulate a shared target , the Hox genes .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"developmental",
"biology",
"genetics",
"biology",
"molecular",
"cell",
"biology",
"genetics",
"and",
"genomics"
] |
2012
|
Transcriptional Repression of Hox Genes by C. elegans HP1/HPL and H1/HIS-24
|
Mycobacterium tuberculosis ( Mtb ) , the causative agent of tuberculosis ( TB ) , infects one third of the world's population . Among these infections , clinical isolates belonging to the W-Beijing appear to be emerging , representing about 50% of Mtb isolates in East Asia , and about 13% of all Mtb isolates worldwide . In animal models , infection with W-Beijing strain , Mtb HN878 , is considered “hypervirulent” as it results in increased mortality and causes exacerbated immunopathology in infected animals . We had previously shown the Interleukin ( IL ) -17 pathway is dispensable for primary immunity against infection with the lab adapted Mtb H37Rv strain . However , it is not known whether IL-17 has any role to play in protective immunity against infection with clinical Mtb isolates . We report here that lab adapted Mtb strains , such as H37Rv , or less virulent Mtb clinical isolates , such as Mtb CDC1551 , do not require IL-17 for protective immunity against infection while infection with Mtb HN878 requires IL-17 for early protective immunity . Unexpectedly , Mtb HN878 induces robust production of IL-1β through a TLR-2-dependent mechanism , which supports potent IL-17 responses . We also show that the role for IL-17 in mediating protective immunity against Mtb HN878 is through IL-17 Receptor signaling in non-hematopoietic cells , mediating the induction of the chemokine , CXCL-13 , which is required for localization of T cells within lung lymphoid follicles . Correct T cell localization within lymphoid follicles in the lung is required for maximal macrophage activation and Mtb control . Since IL-17 has a critical role in vaccine-induced immunity against TB , our results have far reaching implications for the design of vaccines and therapies to prevent and treat emerging Mtb strains . In addition , our data changes the existing paradigm that IL-17 is dispensable for primary immunity against Mtb infection , and instead suggests a differential role for IL-17 in early protective immunity against emerging Mtb strains .
Mycobacterium tuberculosis ( Mtb ) , the causative agent of tuberculosis ( TB ) , infects one third of the world's population . While most infected individuals develop latent TB , 5–10% of infected individuals develop active TB . In addition , although most infected people with latent TB remain asymptomatic , they have ∼10% lifetime risk of developing into active TB . Among these infections , clinical isolates being typed as belonging to the W-Beijing strain appear to be increasingly prevalent . In fact , recent reports show that W-Beijing family Mtb strains represent about 50% of Mtb isolates in East Asia , and are believed to account for at least 13% of all Mtb isolates worldwide [1]–[4] . More importantly , multiple studies have identified that W-Beijing strains are over-represented among drug resistant isolates [5] , [6] , and are significantly associated with human immunodeficiency virus ( HIV ) infection in humans [7] . In animal models , infection with Mtb HN878 isolate , the best studied of the W-Beijing isolates , is thought to be “hypervirulent” as it results in increased mortality and causes severe immunopathology in infected animals [8] , [9] . In addition , studies suggest that M . bovis Bacille Calmette-Guerin ( BCG ) vaccination may be less protective against W-Beijing genotype Mtb strains [4] , thus contributing to its successful recent worldwide emergence . The immune responses that mediate protective immunity against Mtb infection are through the production of proinflammatory cytokines such as Interferon gamma ( IFN-γ ) and Tumor necrosis factor alpha ( TNF-α ) , both cytokines that activate macrophages to mediate Mtb control . This is consistent with the finding that Mtb HN878 infection in mice induces a Type I Interferon response , which limits the generation of T helper type 1 cells ( Th1 ) , that produce IFN-γ and TNF-α [9] , [10] . In addition , Mtb HN878 infection also inhibits the production of TNF-α in macrophages [11] , suggesting that the increased virulence of Mtb HN878 infection may be due to the reduced generation of Th1 responses and impaired macrophage activation in the host . Interleukin-17 ( IL-17 ) is a pro-inflammatory cytokine , well described for its role in host defense against extracellular bacterial pathogens [12] . We had previously shown that the IL-17 pathway is not required for primary immunity against infection with the lab adapted strain , Mtb H37Rv [13]–[15] . However , it is not known whether IL-17 has any role to play in protective immunity against infection with clinical Mtb isolates . In the current study , we tested whether IL-17 is required for protective immunity following infection with the hypervirulent Mtb HN878 , and the less virulent Mtb CDC1551 clinical isolates . Surprisingly , we found that while lab adapted Mtb isolates such as Mtb H37Rv , or less virulent Mtb clinical isolates such as Mtb CDC1551 , did not require IL-17 for early protective immunity against infection , infection with Mtb HN878 required the production of IL-17 for protective immunity . Our data suggest that the dependence on IL-17 to drive protective immunity against Mtb HN878 is due to the differential ability of Mtb HN878 to induce high levels of IL-1β through a TLR-2-dependent mechanism , resulting in high IL-17 production . In addition , our data show that the role for IL-17 in mediating early protective immunity against Mtb HN878 is through IL-17 receptor ( IL-17R ) signaling in non-hematopoietic cells , to induce expression of the chemokine , CXCL-13 . CXCL-13 expression attracts cytokine-producing CXCR5+ T cells which localize near Mtb-infected macrophages , to form lung lymphoid follicles for optimal macrophage activation and Mtb control . Our novel results suggest that the protective immune requirements for emerging hypervirulent Mtb isolates are likely different from the requirements for lab adapted and less virulent Mtb isolates , and thus need to be studied independently as demonstrated here . As recent work has demonstrated a critical role for IL-17 in vaccine induced immunity against TB [16] , [17] , our results have far reaching implications for the design of vaccines and therapies to prevent and treat emerging Mtb strains such as W-Beijing strains .
B6 ( B6 ) , IL-17−/− [18] , IL17RA−/− ( IL-17R−/− ) [19] , CXCR5−/− [20] , IL-17GFP reporter , IL-1R−/− , TLR2−/− and ESAT-6 T-Cell Receptor ( TCR ) Transgenic ( Tg ) mice [21] were bred at the Children's Hospital of Pittsburgh or purchased from Jackson lab . ESAT-6 TCR Tg mice were also crossed to the CXCR5−/− mice to generate Tg mice which were deficient in CXCR5 . Experimental mice were age- and sex-matched and used between the ages of 6 to 8 wks . Mtb strains H37Rv , CDC1551 or HN878 were cultured in Proskauer Beck medium containing 0 . 05% Tween 80 to mid-log phase and frozen in 1 ml aliquots at −70°C . Mice were aerosol infected with ∼100 CFU of bacteria using a Glas-Col airborne infection system [16] . At given time points , organs were harvested , homogenized and serial dilutions of tissue homogenates plated on 7H11 plates and CFU determined . In some experiments , adenovirus over-expressing IL-17 or control adenovirus expressing luciferase vector was delivered once ( 5×108 pfu ) intratracheally . All mice were used in accordance following the National Institutes of Health guidelines for housing and care of laboratory animals and in accordance with University of Pittsburgh Institutional Animal Care and Use Committee guidelines and were approved under Protocol 0807913 . All efforts were made to minimize suffering and pain as described in this approved protocol . Lung cell suspensions were prepared as described [16] and single cell suspensions were stained with appropriate fluorochrome-labeled specific antibodies or isotype control antibodies . Cells were collected using a Becton Dickinson FACS Aria flow cytometer using FACS Diva software . Cells were gated based on their forward by side scatter characteristics and the frequency of specific cell types was calculated using FlowJo ( Tree Star Inc , CA ) . ESAT-61–20-specific IFN-γ and IL-17-producing IAb-restricted T cells from infected lungs or spleen were enumerated using peptide-driven ELISpot as described [22] . Briefly , 96 well ELISpot plates were coated with monoclonal anti-mouse IFN-γ or IL-17 , blocked with media containing 10% FBS . Cells from lungs and spleen were seeded at an initial concentration of 5×105 cells/well and subsequently diluted two fold . Irradiated B6 splenocytes were used as APCs at a concentration of 1×106 cells/well in the presence of ESAT-61–20 ( 10 µg/ml ) peptide and IL-2 ( 10 U/ml ) . After 24 hrs , plates were washed and probed with biotinylated anti-mouse IFN-γ or IL-17 . Spots were visualized and enumerated using a CTL-Immunospot S5 MicroAnalyzer . No spots were detected in cultures lacking antigen or when using cells from uninfected mice . BMDMs and BMDCs were generated from the bone marrow of B6 or gene deficient mice . Cells were extracted from femurs and 1×107 cells were plated with 10 ml of cDMEM supplemented with 20 ng/mL mouse recombinant GM-CSF ( Peprotech ) . Cells were cultured for 3 days at 37°C in 5% CO2 , after which an additional 10 ml of cDMEM containing 20 ng/ml rmGM-CSF was added . On day 7 , non-adherent cells were collected by centrifugation and counted as DCs , while adherent cells were collected and used as BMDMs . Lung or BMDCs were treated with Mtb HN878 or Mtb H37Rv components , such as whole cell lysate , cell wall or lipids preparations ( 20 µg/ml each , BEI Resources , obtained under National Institutes of Health [NIH] contract AI-75320 ) Single cell suspensions from DNAse/collagenase-treated lung tissue were prepared as previously described . CD11c+ cells were isolated using magnetic anti-CD11c beads ( Miltenyi Biotec Inc . ) , according to the manufacturer's instructions . BMDCs and lung dendritic cells prepared as previously described were plated at a density of 1×106 cells/ml and rested overnight . Cells were subsequently infected with different Mtb strains at a multiplicity of infection ( MOI ) of 5 for BMDCs or 0 . 1 for lung dendritic cells in antibiotic-free DMEM , for 24 hrs , following which total lung cell suspensions from B6 or gene deficient mice were added in a 1∶1 ratio to the infected dendritic cells , and co-cultured for 6 days . Naive CD4+ T cells were isolated from single cell suspensions generated from lymph nodes and spleens of ESAT-6 TCR Tg mice using a positive CD4 T cell isolation kit ( Miltenyi Biotech ) as described [23] . For generation of Th17 cells , cells were cultured in Complete Iscove's medium containing TGFβ ( 5 ng/ml ) , IL-6 ( 30 ng/ml ) , IL-23 ( 50 ng/ml ) , anti-IL-4 ( 10 µg/ml ) , anti-IFN-γ ( 10 µg/ml ) , and IL-2 ( 10 U/ml ) and APCs [24] . T cells were incubated for six days at 37°C and 5% CO2 and supplemented with an equivalent volume of media containing IL-2 ( 10 U/ml ) on day 3 . Cells were harvested on day 6 and washed twice with PBS . For adoptive transfer , 3–5×106 ESAT-6 TCR Tg Th17 cells were transferred intravenously into recipient mice , following which mice were rested for 24 hours and challenged with Mtb H37Rv by the aerosol route . To generate chimeric mice , mice were given a medicated Sulfa-Trim diet containing 1 . 2% sulfamethoxazole and 0 . 2% trimethoprim ( TestDiet ) two weeks prior to irradiation . Mice were sub-lethally irradiated with 1000 rads in two doses ( X-Rad 320 ) . Mice were subsequently reconstituted with 10×106 bone marrow cells from B6 or gene deficient mice via i . v . injection . Mice were allowed to reconstitute for 45 days while continuing to receive a Sulfa-Trim and acidified water diet following which they were used in experimental procedures . Lung lobes were instilled with 10% neutral buffered formalin and embedded in paraffin . Lung sections were stained with hematoxylin and eosin ( H&E ) and inflammatory features were evaluated by light microscopy ( Research Histology Core , University of Pittsburgh ) . For immunofluorescent staining , formalin-fixed lung sections were cut , immersed in xylene to remove paraffin and then hydrated in 96% alcohol and PBS . Antigens were unmasked with a DakoCytomation Target Retrieval Solution and non-specific binding was blocked with 5% ( v/v ) normal donkey serum and Fc block ( BD Pharmingen , San Diego , CA ) . Endogenous biotin ( Sigma Aldrich ) was neutralized by adding first avidin , followed by incubation with biotin . Sections were probed with anti-B220 to detect B cells ( Clone RA3-6B2 , BD Pharmingen , San Diego , CA ) , anti-CD3 to detect T cells ( Clone M-20 , Santa Cruz Biotechnology , Santa Cruz , CA ) , anti-CXCL13 ( Clone143614 , R & D Biosystems ) to detect CXCL-13 expression , and inducible NO synthase ( goat anti-mouse , M-19; Santa Cruz Biotechnology ) and F4/80 ( MCA497GA , Serotec ) to detect activated macrophages within inflammatory lesions . B cell lymphoid follicles were outlined with the automated tool of the Zeiss Axioplan 2 microscope ( Carl Zeiss ) and total area and average size was calculated in squared microns . F4/80 macrophages expressing iNOS in five random 20× fields were enumerated per lung ( n = 5 lungs ) and the average was calculated . 3–5 granulomas per lobe in each group were randomly chosen to quantify CXCL13 mRNA expression in 200× fields as described before [17] . Samples were analyzed in a blinded fashion . Paraffin embedded tissue sections were deparaffinized and washed in ethanol . Stringent in situ hybridization ( 50°C with 0 . 1M DTT in the hybridization mix ) was performed with 35S-labeled riboprobes as previously described [20] . Tissue sections were coated with NTB emulsion ( Carestream/Kodak ) , exposed at 10°C for 14 days , counterstained with hematoxylin ( Vector Laboratories ) and mounted with Permount ( Fisher ) . Images were visualized using an Olympus BX41 microscope ( Olympus ) and captured using a SPOT RT3 digital camera ( Diagnostics Instruments ) . Protein levels for cytokines and chemokines in culture supernatants , serum or lung homogenates were measured using a mouse Luminex assay ( Linco/Millipore ) . Differences between the means of groups were analyzed using the two tailed Student's t-test in GraphPad Prism 5 ( La Jolla , CA ) .
We had previously shown that IL-23 deficient mice ( IL-23−/− ) [13] and mice lacking IL-17 signaling ( IL-17RA−/− ) [14] , [15] are not more susceptible than C57BL/6J mice ( B6 ) mice , to low dose Mtb H37Rv aerosol infection . However , it is not known if the IL-17 pathway is required for protective immunity against infection with clinical isolates such as the hypervirulent Mtb HN878 strain . Thus , we aerosol infected B6 or mice lacking IL-17A ( IL-17−/− ) , with low doses of either Mtb H37Rv or Mtb HN878 and determined the effect of IL-17 deficiency on bacterial burden in the lungs . As previously published with IL-17RA−/− mice [14] , [15] , we found that the IL-17−/− mice are not more susceptible than B6 mice to low dose aerosol Mtb H37Rv infection , at either early or later time points ( Fig . 1A ) . Interestingly , when IL-17−/− mice were infected with low doses of hypervirulent Mtb HN878 infection , we found that IL-17−/− mice exhibited increased lung bacterial burden in the lungs at both early and later time points ( Fig . 1B ) . To further test if this requirement for IL-17 in mediating protective immunity against clinical Mtb isolates was limited to Mtb HN878 , or if IL-17 was required for protection against other clinical isolates , we infected B6 or IL-17−/− mice with the less virulent clinical isolate Mtb CDC1551 , and determined bacterial burden in the lungs . We found that similar to Mtb H37Rv infection , IL-17 was not required for protective immunity against Mtb CDC1551 infection , since IL-17−/− mice exhibited comparable lung bacterial burden to B6 infected mice at both early and late time points ( Fig . 1C ) . These data demonstrate for the first time that IL-17 is required for protective immunity against specific clinical Mtb isolates such as Mtb HN878 , but not Mtb lab strains such as Mtb H37Rv , or less virulent clinical isolates such as Mtb CDC1551 . Our data suggest that IL-17 is required for early protective immunity following infection with Mtb HN878 , but not Mtb H37Rv in mice . Thus , we next addressed if infection with Mtb H37Rv and HN878 induced different levels of IL-17 production in the lung . We found that lung IL-17 protein levels were significantly elevated in B6 mice infected with Mtb HN878 , when compared to levels of IL-17 in lungs of B6 mice infected with Mtb H37Rv ( Fig . 2A ) . In addition , when IL-17 reporter GFP mice were infected with similar low doses of either Mtb H37Rv or Mtb HN878 , we found that Mtb HN878 infection induced a higher frequency and total number of IL-17-producing cells in the infected lungs ( Fig . 2B–D ) . In addition , we found that the majority of the lung IL-17+ cells were CD3+ T cells ( Fig . 2E ) . Furthermore , we found that increased numbers of Mtb antigen-specific ESAT61–20 IL-17-producing cells were detected in Mtb HN878-infected lungs , when compared to Mtb H37Rv-infected lungs ( Fig . 2F ) . In contrast , no significant differences were observed in the number of ESAT-6–specific IFN-γ-producing cells in the lungs of B6 mice infected with either Mtb HN878 or Mtb H37Rv ( Fig . 2G ) . Differentiation of Th17 cells is dependent on instructive signals provided by APCs , thus we infected lung CD11c+ cells in vitro with the two strains of Mtb , and subsequently co-cultured them with naïve lung cells for 6 days . Increased levels of IL-17 were detected in Mtb HN878-infected co-culture supernatants , when compared to levels detected in supernatants from Mtb H37Rv-infected co-cultures ( Fig . 2H ) . These data together suggest that Mtb HN878 strain induces higher levels of IL-17 production in lung cells , when compared to Mtb H37Rv infection . In order to define the mechanism driving enhanced IL-17 production during Mtb HN878 infection , we analyzed induction of polarizing cytokines by lung DCs following infection with Mtb H37Rv and Mtb HN878 . We found increased IL-1β levels in supernatants of lung DCs infected with Mtb HN878 strain , but not Mtb H37Rv ( Fig . 3A ) . IL-1β has a critical role in IL-17 production [25] . Importantly , when lung DCs were infected with Mtb HN878 , and IL-1R−/− lung cells were co-cultured with infected DCs , IL-17 production was significantly impaired in culture supernatants ( Fig . 3B ) , demonstrating that IL-17 production is IL-1β pathway dependent . To confirm our in vitro findings in vivo during Mtb infection , we infected B6 or IL-1R−/− mice with low doses of aerosolized Mtb HN878 and found that IL-1R−/− mice had significantly higher bacterial burden ( Fig . 3C ) , and this coincided with significantly reduced induction of Mtb-specific IL-17-producing T cells ( Fig . 3D ) . Together , these data demonstrate that Mtb HN878-driven IL-17 production is IL-1β dependent , both in vitro and in vivo . To further define the mechanism underlying IL-1β induction by Mtb HN878 , we incubated bone marrow derived DCs ( BMDCs ) with different subcellular preparations from Mtb and found that treatment with Mtb HN878 whole cell lysates and cell wall extracts , but not lipids from Mtb HN878 , induced IL-1β in BMDCs ( Fig . 3E ) . Given that TLR-2 recognizes cell wall components from Mtb [26] , and that its activation has been linked to IL-1β production [27] , we next determined whether IL-1β and IL-17 production were dependent on activation through TLR-2 . We found that Mtb HN878-infection of TLR-2−/− lung DCs , had a dramatic impairment in IL-1β production when compared to infection of B6 DCs ( Fig . 3F ) , and coincided with low IL-17 production in DC: lung cell co-cultures ( Fig . 3G ) . These data together suggest that when compared to Mtb H37Rv infection , Mtb HN878 infection induces higher levels of IL-1β production in APCs through a TLR-2-dependent mechanism , and this drives increased IL-17 production . IL-17 is well documented to drive the production of molecules such as Keratinocyte chemoattractant ( KC ) , and Granulocyte Colony Stimulating Factor ( G-CSF ) to mediate neutrophil recruitment and accumulation [12] . However , we did not find any defects in total lung neutrophil numbers and lung neutrophil accumulation within granulomas of B6 or IL-17−/− Mtb HN878 infected mice ( data not shown ) . In addition , we did not find any differences in pulmonary levels of KC and G-CSF in B6 and IL-17−/− Mtb HN878 infected mice ( data not shown ) . These data together suggest the role for IL-17 in protective immunity against Mtb HN878 infection is not mediated through its well documented role in neutrophil recruitment . Since IL-17 can drive the induction of Th1 responses in some intracellular pulmonary bacterial infection models [28] , [29] , we next determined if there were any defects in generation of Th1 immune responses . Interestingly , we did not find any differences in the accumulation of IFN-γ ( Fig . S1A ) or TNF-α-producing ( Fig . S1B ) Mtb-specific T cells in the lungs of B6 and IL-17R−/− Mtb HN878-infected mice . In addition , we also found comparable numbers of activated CD4+ T cells accumulating in the Mtb HN878-infected B6 and IL-17R−/− lungs ( Fig . S1C ) , including total IFN-γ ( Fig . S1D ) and IL-2-producing T cells ( Fig . S1E ) . These data suggest that the generation and accumulation of activated Th1 responses are not defective in mice deficient in the IL-17 pathway . We have recently described a role for early vaccine-induced IL-17 in mediating CXCL-13 expression which resulted in localization of CXCR5+ cytokine producing T cells near Mtb-infected macrophages , an event crucial for optimal Mtb control [30] . The correct localization of T cells expressing CXCR5 within the lung parenchyma , results in formation of ectopic lymphoid structures , which is required for activation of infected macrophages for control of Mtb [30] . Thus , we next addressed whether the increased susceptibility in IL-17−/− mice infected with Mtb HN878 was due to defects in T cell localization near Mtb-infected macrophages within the lung . We found that IL-17−/− mice infected with Mtb HN878 exhibited increased lung perivascular T cuffing ( Fig . 4A ) , and this coincided with poorly formed lymphoid follicles within the lungs of IL-17−/− Mtb HN878-infected mice , when compared to B6 Mtb HN878-infected mice ( Fig . 4B ) . Importantly , this coincided with decreased accumulation of activated iNOS-producing macrophages within the IL-17−/− Mtb HN878-infected lung inflammatory lesions , when compared to the B6 Mtb HN878-infected lesions ( Fig . 4C ) . To exclude a direct effect of IL-17 on macrophage activation and Mtb control , bone marrow macrophages ( BMDMs ) from B6 mice were infected with Mtb HN878 and treated with either IFN-γ or IL-17 , or both cytokines . Although we found that IL-17 treatment induced some iNOS production in uninfected macrophages , it did not enhance iNOS production in Mtb HN878-infected macrophages or enhance control of Mtb in infected macrophages ( Fig . S2A–B ) . As expected , IFN-γ treatment resulted in increased iNOS production by macrophages and improved bacterial control , when compared to untreated macrophages or IL-17 treated Mtb-infected macrophages ( Fig . S2A–B ) . Instead , we found that coincident with reduced lymphoid follicle formation in IL-17−/− Mtb HN878-infected lungs , expression of CXCL-13 mRNA expression within lung lymphoid follicles was also reduced ( Fig . 4D ) . These data together suggest that following Mtb HN878 infection , enhanced IL-17 production driven by the infection , plays a role in induction of CXCL-13 , localization of cytokine producing T cells near Mtb-infected macrophages , and Mtb control . Our data show that IL-17 is required for protective immunity against Mtb HN878 infection by mediating correct T cell localization near Mtb-infected macrophages for optimal macrophage activation . We therefore tested if over expression of IL-17 by adenoviral vectors would rescue the increased susceptibility observed in IL-17−/− Mtb HN878 infected mice . Thus , IL-17−/− Mtb HN878 infected mice received a single intratracheal delivery of either adenovirus expressing a luciferase control vector or adenovirus expressing IL17 . We found that IL-17 overexpression reversed the increased bacterial burden seen in IL-17−/− Mtb HN878 infected mice ( Fig . 5A ) , resulted in decreased T cell perivascular cuffing ( Fig . 5B , C ) , improved formation of lung lymphoid follicles ( Fig . 5D ) , and coincident increase in CXCL-13 mRNA expression ( Fig . 5e ) and CXCL-13 protein ( Fig . 5F ) , within B cell lymphoid follicles in the lung . As a result , IL-17 overexpression resulted in increased accumulation of iNOS-expressing macrophages within the inflammatory lesions ( Fig . 5G ) . To further confirm that CXCL-13 expression is critical for localization of CXCR5-expressing Th1 cells within the lymphoid follicles and Mtb control , we infected CXCR5−/− mice with Mtb HN878 . We found that CXCR5−/− mice exhibited increased bacterial burden when compared to B6 Mtb HN878 infected mice ( Fig . 6A ) . Importantly , we found that CXCR5−/− Mtb infected mice demonstrated enhanced T cell perivascular cuffing ( Fig . 6B ) and this coincided with reduced formation of lymphoid follicles ( Fig . 6C ) , measured by determining the average area of B cell follicle within the lungs of B6 and CXCR5−/− mice . In addition , we found that the impaired localization of T cells within the lung parenchyma resulted in decreased accumulation of iNOS-producing macrophages ( Fig . 6D ) , within the lung inflammatory lesions , suggesting sub-optimal activation of macrophages for Mtb control . In order to mechanistically address if presence of Th17 cells could rescue the increased susceptibility seen in IL-17−/− mice , we adoptively transferred Mtb-specific Th17 cells generated from wild type ESAT-6 TCR Tg mice or from CXCR5−/− ESAT-6 TCR Tg mice , into IL-17−/− mice which were then infected with aerosolized Mtb HN878 . Consistent with a role for IL-17 in mediating protection against Mtb HN878 control , adoptive transfer of wild type Th17 cells significantly reduced the bacterial burden ( Fig . 6E ) . In contrast , adoptive transfer of Th17 cells derived from CXCR5−/− mice did not decrease lung bacterial burden ( Fig . 6E ) . These data together conclusively provide evidence that IL-17 expression during Mtb HN878 infection is required for effective induction of CXCL-13 , a chemokine that is key to facilitating productive interactions between cytokine producing CXCR5-expressing T cells and Mtb-infected macrophages for optimal Mtb HN878 control in the lung . IL-17R is primarily expressed on non-hematopoietic cells , but IL-17R can also be expressed on hematopoietic cells such as macrophages and dendritic cells [12] . Thus , we next addressed if IL-17R signaling on hematopoietic or non-hematopoietic cells was essential for inducing CXCL-13 expression and mediating protection against Mtb HN878 infection . Thus , we generated hematopoietic IL-17R−/− bone marrow chimeric ( BMC ) mice ( B6 host/−/− BM ) or non-hematopoietic IL-17R−/− BMC mice ( −/− host/B6 BM ) and infected with Mtb HN878 . As expected based on our data , complete IL-17R−/− BMC mice ( −/−host/−/− BM ) were more susceptible to Mtb HN878 infection than complete B6 BMC mice ( B6 host/B6 BM ) ( Fig . 7A ) , and demonstrated defects in T cell localization ( Fig . 7B ) , formation of lymphoid follicles and CXCL-13 mRNA expression ( Fig . 7C ) , and CXCL-13 protein expression within B cell lymphoid follicles ( Fig . 7D ) . Interestingly , non-hematopoietic , but not hematopoietic IL-17R−/− BMC mice had increased lung bacterial burden ( Fig . 7A ) , and this coincided with increased perivascular T cell cuffing ( Fig . 7B ) and reduced CXCL-13 mRNA and protein expression ( Fig . 7C , D ) . These data clearly demonstrate that IL-17 signaling in non-hematopoietic cells is required for CXCL-13 induction , to mediate correct T cell localization near Mtb infected macrophages to confer protective immunity against Mtb HN878 infection .
The Mtb isolate HN878 , belongs to the W-Beijing family of isolates , which have been associated with outbreaks throughout the world , and with clusters of drug-resistant disease in the United States [31] . W-Beijing strains have significant clinical relevance because they are over-represented among drug resistant Mtb isolates , and are associated with HIV infection in humans [7] . In animal models , infection is considered hypervirulent due to increased immunopathology and mortality [8] . In the current paper , we show that IL-17 is required for early protective immunity against Mtb HN878 infection , but not lab adapted Mtb isolates such as H37Rv , or less virulent Mtb clinical isolates such as CDC1551 . Our data also suggest that the dependence on IL-17 to drive early protective immunity against Mtb HN878 is due to the differential ability of Mtb HN878 to induce high levels of IL-17 production , through an IL-1β-TLR-2 dependent pathway . Thus , defining the differential immune requirements that mediate protective immunity against clinical isolates such as the widely spreading W-Beijing isolate , is critical for successful design of vaccines against emerging strains of Mtb . Thus , our novel results demonstrating a role for IL-17 in primary immunity to specific Mtb strains , have far reaching implications for the future design of vaccines and therapies to prevent and treat TB worldwide , especially emergent strains of clinical relevance for public health . Early studies using animal models showed that infection with Mtb HN878 induced Type I interferons , and this coincided with decreased induction of proinflammatory cytokines such as TNF-α , IFN-γ and IL-2 , reduced T cell activation and increased susceptibility to infection [8] , [10] , [11] . The hypervirulent phenotype of Mtb HN878 was initially linked to production of phenolic glycolipids ( PGL ) , one of the major lipid components of the mycobacterial cell wall , that could be mediating the inhibition of protective Th1 responses [8] , [10] , [11] . Interestingly , when PGL was expressed in Mtb H37Rv , a strain normally devoid of PGL synthesis , it did not lead to increased virulence in infected mice and rabbits , suggesting that PGLs in concert with other bacterial factors likely mediates the hypervirulence of W-Beijing Mtb strains [32] . Furthermore , it has been recently shown that PGL enhances infectivity through CCR2-mediated TLR-independent recruitment of permissive macrophages at the earliest stages of infection , suggesting the presence of PGL in Mtb HN878 as a likely factor contributing to the increased transmission of Mtb HN878 [33] . In contrast , in the current study , we show that DCs infected with Mtb HN878 are stimulated likely by a cell wall component that binds TLR2 , and triggers the secretion of IL-1β . In addition , the increased induction of IL-1β in infected DCs mediates the induction of IL-17 production , primarily in CD3+ T cells . Accordingly , our studies show that in the presence of low IL-17 induction following Mtb H37Rv infection , absence of IL-17 as shown here , or absence of IL-17R as published before [14] , [15] , does not impact protective immunity . In contrast , consistent with the increased induction of IL-17 seen in Mtb HN878 infected lungs , IL-17 is required for early protective immunity against infection with Mtb HN878 . Our data also show that a protective role for IL-17 is not observed for all clinical Mtb isolates , as IL-17−/− mice infected with the less virulent clinical isolate Mtb CDC1551 , does not show early increased susceptibility to infection . Our data described here has tested the early protective role for IL-17 in Mtb infection , and future studies determining whether IL-17 is required for maintenance of Mtb control during chronic stages of Mtb infection will be important . Without doubt , future studies focused on defining whether specific lineages of Mtb require IL-17 for protective immunity in acute and chronic Mtb infection will be critical for successful vaccine design for global TB control , by tailoring more specific strategies for prevalent emerging strains of Mtb in certain geographical areas . Our recent work has put forth the new “working hypothesis” that a protective TB granuloma in the lung is predominantly composed of lymphoid follicles , where T cells and B cells are strategically positioned near infected macrophages to form lymphoid follicles or optimal activation to control Mtb infection [17] , [20] . IL-17 induces expression of CXCL-13 in the lung and mediates generation of lymphoid follicles following inflammation [17] , [34] . Consistent with this newly described role for IL-17 in driving formation of lymphoid follicles in the lung , our data show that IL-17−/− mice which were more susceptible to Mtb HN878 infection have significant defects in T cell localization and formation of lung lymphoid follicles within TB granulomas . In addition , we found that the poorly formed lymphoid follicles with the lungs of IL-17−/− mice exhibited reduced induction of CXCL13 mRNA and protein . Our data show that IL-17−/− Mtb HN878-infected lungs accumulate similar numbers of proinflammatory T cells producing IFN-γ , IL-2 and TNF-α , suggest that the increased susceptibility to infection is not due to defects in generation , or accumulation of proinflammatory T cells in the lung , but due to defects in localization of cytokine-producing T cells within lymphoid follicles to mediate macrophage activation . The fact that overexpression of IL-17 using adenoviral vectors can reverse the increased susceptibility , CXCL-13 expression , T cell localization and lymphoid follicle formation within the IL-17−/− Mtb HN878-infected lung , further supports and validates this hypothesis . However , it is also possible that absence of IL-17 impacts the generation and accumulation of B cells or other T cells such as CD8+ T cells to mediate formation of B cell lymphoid follicles , and should be addressed in future studies . IL-17R−/− mice infected with Mtb H37Rv exhibit a transient early defect in lymphoid follicle formation , coinciding with absence of a role for IL-17 pathway in mediating protection [15] . These data together suggest that following infection with Mtb HN878 , IL-17 has a non-redundant and prominent role to play in induction of CXCL-13 expression , likely through its effects on non-hematopoietic cells such as epithelial cells and fibroblasts . This is consistent with the ability of IL-17 to drive CXCL-13 expression in fibroblasts in vitro [15] . Interestingly , both CXCL-13 and CXCR5−/− mice are equally susceptible to infection with either Mtb H37Rv [20] , [23] or Mtb HN878 described here , implicating IL-17-independent pathways in CXCL-13 induction during Mtb H37Rv infection . Further studies delineating and characterizing common versus differential protective correlates across different Mtb lineages will be necessary in designing vaccines , specific for clinical strains that are endemic in different geographical locations of the world . Our recent studies using CXCR5−/− mice in Mtb H37Rv infection model , have demonstrated that expression of CXCR5 is required for T cell localization within the lung and macrophage activation for Mtb control [20] . Importantly , we showed that adoptive transfer of CD4+ T cells expressing CXCR5 into Mtb H37Rv-infected CXCR5−/− mice , was sufficient to allow T cells to localize within the lung , reverse generation of lymphoid tissues and improve disease outcome [20] . Consistent with these recent findings , we report here that CXCR5−/− mice when infected with Mtb HN878 are also susceptible to Mtb HN878 infection , demonstrate defects in localization of T cells , formation of lymphoid follicles and activation of macrophages within TB granulomas . Importantly , IL-17−/− mice that receive Mtb-specific Th17 cells expressing CXCR5 , could reverse the increased susceptibility and associated disease phenotype , but adoptive transfer of Mtb-specific CXCR5−/− Th17 cells could not rescue the disease phenotype in IL-17−/− Mtb HN878-infected mice . These data together mechanistically provide evidence that following Mtb HN878 infection , cytokine producing CD4+ T cells express CXCR5 and respond to signals from IL-17-dependent CXCL-13 to localize within the lung parenchyma to form lymphoid follicles , and activate macrophages for Mtb control . IL-17 is generally thought to be required for protective immunity against extracellular pathogens , by inducing chemokines that drive neutrophil recruitment for pathogen control [12] . Thus , it was not surprising that early studies demonstrated that IL-17 was not required for protective immunity against intracellular pathogen such as Mycobacteria [13] , [14] , [35] , Listeria [14] and Salmonella infections [36] . However , more recent work by us and others has demonstrated that in some models of intracellular infections , IL-17 is required for protective immunity to drive the induction of IL-12 and generate Th1 responses [28] , [29] . In contrast , the more widely appreciated role for IL-17 in TB , is its role in mediating vaccine-induced protection against Mtb challenge [16] , [17] . Accordingly , targeting IL-17 to improve vaccine design for TB is an active avenue of research [37] . Thus , our new data demonstrating that the IL-17 pathway also plays a role in primary immunity following infection with Mtb HN878 , significantly changes the existing paradigm that IL-17 is not required for primary immunity against TB . Interestingly , some human studies report increased IL-17 in active TB patients [38] , [39] , while other studies report increased IL-17 in latent TB patients [40] , [41] and healthy controls [41] , [42] , suggesting IL-17 may be associated with either inflammation or protection , respectively . Thus , differential IL-17 expression by some emerging Mtb strains may define the final role for IL-17 in protection or pathology during TB , and needs to be carefully studied . In summary , data presented here demonstrates that IL-17 has an early protective role to play in immunity against Mtb HN878 infection , by mediating induction of chemokines , T cell localization within lymphoid follicle for macrophage activation and Mtb control . These results demonstrate a novel and previously undescribed role for IL-17 in primary immunity to TB , especially considering that W-Beijing strains such as Mtb HN878 are emerging as major drug resistant Mtb strains that are not protected by prior BCG vaccination . Thus , targeting IL-17 to improve vaccine stategies for protection against emerging Mtb strains such as W-Beijing Mtb strains , may prove critical in controlling global TB burdens .
|
Mycobacterium tuberculosis ( Mtb ) , the causative agent of tuberculosis ( TB ) , infects one third of the world's population . Among these infections , clinical isolates belonging to the W-Beijing are emerging , representing about 50% of Mtb isolates in East Asia , and about 13% of all Mtb isolates worldwide . In animal models , infection with W-Beijing strain , Mtb HN878 , is considered “hypervirulent” resulting in increased mortality . The proinflammatory cytokine Interleukin ( IL ) -17 is thought to be dispensable for primary immunity against Mtb infection . We report here that while IL-17 is dispensable for protection against infection with lab adapted Mtb strains such as H37Rv , or less virulent Mtb clinical isolates such as Mtb CDC1551 , IL-17 is required for early protective immunity against Mtb HN878 infection . The dependence on IL-17 to drive protective immunity against Mtb HN878 is due to the differential ability to induce high levels of IL-1β through a TLR-2-dependent mechanism , driving potent IL-17 responses , induction of the chemokine CXCL-13 and localization of T cells within lung lymphoid follicles for maximal macrophage activation and Mtb control . Together , our data change the existing paradigm that IL-17 is dispensable for primary immunity against Mtb infection , and suggests a differential requirement for IL-17 in protective immunity against some emerging Mtb strains .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"bacterial",
"diseases",
"infectious",
"diseases",
"medicine",
"and",
"health",
"sciences",
"tuberculosis",
"tropical",
"diseases"
] |
2014
|
Unexpected Role for IL-17 in Protective Immunity against Hypervirulent Mycobacterium tuberculosis HN878 Infection
|
Rickettsia ( R . ) helvetica is the most prevalent rickettsia found in Ixodes ricinus ticks in Germany . Several studies reported antibodies against R . helvetica up to 12 . 5% in humans investigated , however , fulminant clinical cases are rare indicating a rather low pathogenicity compared to other rickettsiae . We investigated growth characteristics of R . helvetica isolate AS819 in two different eukaryotic cell lines with focus on ultra-structural changes of host cells during infection determined by confocal laser scanning microscopy . Further investigations included partially sequencing of rickA , sca4 and sca2 genes , which have been reported to encode proteins involved in cell-to-cell spread and virulence in some rickettsiae . R . helvetica grew constantly but slowly in both cell lines used . Confocal laser scanning microscopy revealed that the dissemination of R . helvetica AS819 in both cell lines was rather mediated by cell break-down and bacterial release than cell-to-cell spread . The cytoskeleton of both investigated eukaryotic cell lines was not altered . R . helvetica possesses rickA , but its expression is not sufficient to promote actin-based motility as demonstrated by confocal laser scanning microscopy . Hypothetical Sca2 and Sca4 proteins were deduced from nucleotide gene sequences but the predicted amino acid sequences were disrupted or truncated compared to other rickettsiae most likely resulting in non-functional proteins . Taken together , these results might give a first hint to the underlying causes of the reduced virulence and pathogenicity of R . helvetica .
Rickettsia ( R . ) helvetica is the most prevalent rickettsia found in Ixodes ( I . ) ricinus ticks in Germany with varying prevalence up to 17% [1–4] . The organism has mainly been considered non-pathogenic and affected patients usually show a mild disease , manifesting in non-specific fever without erythema ( so-called uneruptive fever ) , headache , and myalgia [5] . Hence , only a few laboratories in Germany focus on R . helvetica infection as a differential diagnosis and infections might be underdiagnosed due to mild symptoms in most cases . The latter might be the reason that , to the authors’ best knowledge , no clinical case in humans has been reported from Germany so far . However , more severe clinical cases have been demonstrated in Sweden including septicemia [6] , myocarditis [7] and meningitis [8] . Only recently , complex phylogenetic studies showed that R . helvetica phylogenetically was misplaced in the spotted fever group ( SFG ) [9] . In contrast , R . conorii subsp . conorii , a typical representative of the SFG , causes Mediterranean spotted-fever , a disease that is characterized by fever , an eschar at the site of the tick bite , and a rash spreading to the palms and soles . The disease is endemic in southern Europe but single cases have also been reported from the central and northern European mainland [10] . As many obligate intracellular bacteria , rickettsiae proliferate in the cytoplasm as dispersed , individual bacteria but may also occasionally be found in clusters and in the nucleus [11] . The primary targets for rickettsiae during infection are endothelial cells of the middle and small vessels [12–13] . Disseminated infection of the endothelium and subsequent pathophysiological effects lead to most of the clinical characteristics described for rickettsial diseases [13] . Rapid spread within host tissues is a crucial step in many infectious diseases [14] . For some Rickettsia species it has been proven that rickettsiae harness the host cell actin cytoskeleton for intracellular movement and cell-to-cell spread . This phenomenon was first reported by Teysseire et al . [15] and Heinzen et al . [16] who observed associations between R . conorii and R . rickettsii and host F-actin . Most of the SFG rickettsiae ( SFGR ) as well as R . typhi assemble actin tails and undergo actin-based motility mediating cell-to-cell spread and enhancing virulence [17–18] . Two actin-polymerizing proteins have been identified in SFG rickettsiae: RickA , which activates the actin-related protein-2/3 ( Arp2/3 ) complex of the host [19–20] , and surface cell antigen 2 ( Sca2 ) which has been suggested to mimic eukaryotic formin proteins [18 , 21] . Cardwell and Martinez [22] identified the minimal domain within the Sca2-protein of R . conorii that is sufficient for stimulating actin polymerization . Most recently , Sca4 was identified as a secreted effector of spread independent from actin-based motility in the SFG R . parkeri [23] . So far—to the authors’ knowledge—only a single study on growth characteristic of R . helvetica in eukaryotic cells exists [24] besides the original description of R . helvetica in 1993 [25] . Based on host cell decomposition and intranuclear growth of R . helvetica , the study by Elfving et al . [24] underlined the pathogenic ability of R . helvetica . Here , we report growth characteristics of R . helvetica isolate AS819 in two different eukaryotic cell lines with focus on ultra-structural changes of host cells during infection as determined by confocal laser scanning microscopy ( LSM ) . Further investigations included sequencing of rickA , sca4 and sca2 genes , which have been reported to encode proteins involved in cell-to-cell spread in some rickettsiae .
Eukaryotic cell lines used in this study were obtained from LGC Standards , Wesel , Germany . L929 cells ( murine fibroblasts from connective tissue; up to 25 passages from the original LGC Standards culture ) and Vero E6 cells ( African green monkey kidney cells; up to 39 passages from the original LGC Standards culture ) were grown in Minimum Essential Medium ( MEM ) supplemented with Gibco GlutaMAX and 1x MEM non-essential amino acids ( NEAA ) solution ( Life Technologies GmbH , Darmstadt , Germany ) and 3% fetal calf serum ( FCS ) at 37°C , 5% CO2 . The R . helvetica AS819 isolate ( seventh passage from the original culture ) used for the growth studies was isolated from I . ricinus . Species identity was confirmed based on 100% ompB-sequence identity ( 4 , 848 nt , accession number MF163037 ) to R . helvetica type strain C9P9 ( accession number AF123725 ) . R . conorii ( Moroccan isolate VR141 ) and R . honei ( VR1472 ) were obtained from ATCC , Manassas , USA . All rickettsiae were cultured on Vero E6 monolayers at 32°C in MEM prepared as described above . Flasks were checked daily for detrimental alteration of the monolayer . Rickettsiae were grown until cell layers revealed plaque formation ( R . conorii , R . honei ) or detached from the flasks surface ( R . helvetica ) . The amount of R . helvetica in culture supernatants was determined by quantitative gltA real-time PCR as described below . A non-infected cell control ( MOCK ) was carried along with every infection for the evaluation of changes in the cell layers . L929 and Vero E6 cells ( 105 cells/well ) seeded in 4-well BD Falcon CultureSlides ( BD Biosciences , Heidelberg , Germany ) were grown to confluent monolayers overnight at 37°C , 5% CO2 . Prior to infection , monolayers were rinsed with FCS-free medium . Infection was performed with 100 μl of a seed culture containing 5 x 105 R . helvetica AS819-genome equivalents ( corresponding to 5 genome equivalents per cell ) determined by quantitative gltA real-time PCR as described below . After one hour of incubation at room temperature whilst constantly rocking , wells were filled up with 900 μl MEM supplemented with Gibco GlutaMAX NEAA and 3% FCS . Of every 4-well slide , three wells contained infected cells whilst the remaining served as a non-infected cell control ( MOCK ) . Culture slides were incubated at 32°C , 5% CO2 . Experiments were performed over a period of 20 days . A slide of each cell line was randomly selected on a daily basis and prepared for the quantification of rickettsia in the cell culture supernatant and the cell layer , respectively . Briefly , cell culture supernatants of each well were harvested individually and stored at -80°C . One ml of cell culture medium per well was added to the remaining cell layers and the slides were frozen at -80°C for one hour . Subsequently , freeze-thawed cells were scraped off , harvested and stored at -80°C until further processing . Nucleic acids were isolated from 200 μl of all samples applying the MagNA Pure LC Total Nucleic Acid Isolation Kit ( Roche , Mannheim , Germany ) and the MagNA Pure LC 2 . 0 system ( Roche ) according to the manufacturer’s instructions . Nucleic acids were eluted in a total volume of 50 μl . Quantification of rickettsiae was carried out by a real-time PCR targeting the single copy citrate synthase gene gltA in a Stratagene MX3000P Thermocycler ( Agilent Technologies , München , Germany ) as described earlier [26–27] . Five μl of nucleic acids were used as a template for the PCR . A commercially available DNA-standard ( AmpTech GmbH , Hamburg , Germany ) with a defined concentration ( 2 . 77 x 1010 copies/μl ) was used for the quantification of Rickettsia genome equivalents . The number of R . helvetica copies/μl of template was calculated using the Stratagene Software by comparing the samples to a serial dilution of the DNA-standard ( 2 . 77 x 105 to 2 . 77 x 101 copies/μl ) . Cells were seeded in μ-Slide 8 well ibiTreat chambers ( ibidi GmbH , Martinsried , Germany ) . Briefly , 2 . 5 x 105 cells/well were grown to near confluence overnight and were infected with 5 genome equivalents per cell as described above . After an initial incubation period of 1 h at room temperature , slides were transferred to 32°C , 5% CO2 . Fixation and staining procedures were performed at room temperature . At 24 hours-intervals , culture supernatants were discarded and cells were washed using phosphate-buffered saline ( PBS , pH 7 . 2 ) pre-heated to 37°C . Cells were fixed for 15 min in methanol-free formalin ( 3 . 7% ) . Following a wash with PBS , cells were permeabilized with 0 . 1% Triton X-100 in PBS for 4 min . To reduce nonspecific binding and background signal , blocking was performed in PBS with 1% bovine serum albumin for 60 min . Thereafter , cells were incubated with an anti-SFG Rickettsia antibody ( 1:2 , Fuller Laboratories , Fullerton , USA ) for 30 min . After washing with PBS , a mouse monoclonal anti-alpha-tubulin antibody ( 1:200 , Molecular Probes , Life Technologies ) was added for 30 min and F-actin was stained with Alexa Fluor 568 Phalloidin ( 1:100 , Molecular Probes , Life Technologies ) in parallel . Samples were washed four times and the secondary antibody for Rickettsia staining , an Alexa Fluor 647-labeled goat anti-human IgG ( 1:2000 , Molecular Probes , Life Technologies ) , was added for 30 minutes . In addition , Alexa Fluor 488-labeled goat anti-mouse IgG ( 1:2000 , Molecular Probes , Life Technologies ) was added for staining of microtubuli . Washing another four times was followed by nuclei counterstaining with DAPI ( 4' . 6-Diamidino-2-Phenylindole , Dilactate , 1:5000 , Molecular Probes , Life Technologies ) . Confocal images were obtained with a Zeiss LSM 710 using an EC Plan-Neofluar 40x/1 . 30 Oil DIC M27 objective and using the ZEN software ( Zeiss , Jena , Germany ) . For R . helvetica-infected cells , staining of slides was performed immediately after fixation each day during the incubation period of 20 d . R . conorii served as a control but the incubation period was limited to five days after infection due to the rapid cell spread of this rickettsia within cells . Images of L929 or Vero cells infected with R . helvetica were analyzed using the software Daime [28] . The detection of nuclei and rickettsiae by the software was optimized by comparing manual counts and software-based counts for three pictures each and the following parameters were chosen: Nuclei stained with DAPI were detected using threshold detection for objects larger than 500 pixels . Rickettsiae were detected using edge detection for objects larger than 20 pixels . As it was not possible to visually separate single rickettsiae in highly infected host cells at later time points , we decided to use the signal area instead of the number of objects in order to quantify the infection progress . The signal area corresponding to rickettsiae in at least 450 host cells was analyzed in images from five , ten , fifteen and twenty days post infectionem ( p . i . ) and the proportion of the area of rickettsiae to the area of cell nuclei was calculated for each time point . Confluent L929 and Vero E6 cells grown in 6-well plates were washed two times with FCS-free Medium . One ml of R . helvetica culture supernatant ( undiluted and serially diluted 10−1–10−4 ) was added per well . One well contained an uninfected control . Plates were incubated at room temperature for 1 h while constantly rocking . Double concentrated M199 ( Gibco Thermo Fisher Scientific , Waltham , USA ) containing 10% FCS was mixed with an equal volume of pre-heated 2% Agarose ( Agarose LE , Biozym , Hessisch Oldendorf , Germany ) and each well was then filled with 4 ml of agarose overlay . Plates were incubated for 21 d at 32°C , 5% CO2 . Formalin ( 3 . 7% ) fixation ( room temperature , 1 h ) was done on days 7 , 14 , and 21 p . i . . After removal of the agarose plugs the cells were stained using crystal violet ( 1% ) in 20% ethanol ( 30 min , room temperature ) . After staining , the plates were washed and examined . In parallel the BSL-2 Rickettsia R . honei was used as a positive ( i . e . plaque-forming ) control . Amplification of the partial rickA gene was conducted by PCR using primers described by Balraj et al . [29] . PCR conditions were established using DNA from R . conorii VR141 . Direct sequencing of PCR products was carried out by GATC Biotech AG sequencing service ( Konstanz , Germany ) . Determination of open reading frames and subsequent amino acid ( aa ) alignments with corresponding sequences retrieved from the GenBank database ( http://www . ncbi . nlm . nih . gov/ ) were performed using the software package BioEdit v . 7 . 2 . 5 [30] . The BioEdit Sequence Alignment Editor , Version 7 . 2 . 5 and the implemented ClustalW , Version 1 . 4 [31] were applied for sequence analyses . In addition , sequencing results of rickA were complemented by data obtained from whole genome pyrosequencing of R . helvetica AS819 using the Roche 454 GS-FLX platform ( data not yet published ) . 377 , 211 shotgun reads ( 135 , 929 , 920 bases ) were assembled using GS De Novo Assembler version 2 . 3 , GS Reference Mapper version 2 . 3 , and DNASTAR SeqMan Ngen version 10 . 1 . 0 . On average , the coverage of the R . helvetica AS819 plasmid was 146-fold and the coverage of the genome was 34-fold . We predicted CDSs using the RAST prokaryotic genome annotation server ( http://rast . nmpdr . org/rast ) . RNAs were identified using tRNAscan-SE v . 1 . 23 ( tRNA ) , aragorn v . 1 . 2 . 34 ( tRNA , tmRNA ) , and RNammer v . 2 . 1 ( rRNA ) . Database searches were done using BLAST and infernal . Hereby sca2 gene sequences were obtained .
The growth of R . helvetica AS819 in two eukaryotic cell lines is summarised in Fig 1A and 1B . Results were obtained from triplicates of infected cultures per cell line and time point . After two to four days of lag , a continuous increase of R . helvetica genome equivalents ( given as copies/μl ) was seen in both cell lines resulting in a maximum of 6 x 105 copies/μl at day 19 . Propagation in Vero E6 cells ( Fig 1A ) revealed a slight increase of intracellular R . helvetica DNA copies from day four to eleven after infection . A 100-fold increase was observed from day eleven to day twelve after infection . In L929 cells ( Fig 1B ) , DNA copies of intracellular R . helvetica tripled from day five to day six and a sharp increase of DNA was seen from day seven to day ten resulting in an approximately 1 , 000-fold rise . A maximum of 100-fold difference between intracellular and extracellular DNA copies/μl was measured in Vero E6 cells ( Fig 1A ) . In contrast , less than 10-fold difference was found using L929 cell lines ( Fig 1B ) . Rapid cell-to-cell spread was not seen in R . helvetica ( isolate AS819 ) -infected cell monolayers . LSM analyses revealed that the percentage of infected cells was rather constant over a period of several days indicating little cell-to-cell spread . Five to ten days p . i . , infected cells were scattered throughout the monolayers ( Figs 2 and 3 ) . At that time , up to ten rickettsiae were countable within the cytoplasm of a single cell . From day fifteen until the end of the experiment , the number of infected host cells increased . However , infected cells remained in clusters indicating that R . helvetica initiated mainly new infection of adjacent cells . R . helvetica build up in large numbers in the cytoplasm of single cells were visible . No intranuclear bacteria were seen . A quantification of rickettsiae per nucleus area over time revealed higher numbers of R . helvetica AS819 within L929 cells compared to Vero E6 cells during the whole infection experiment ( Fig 4 ) . In both cell lines the number of rickettsiae per cell nucleus doubled from time point to time point until 15 days after infection . The experiment could be pursued until day 20 p . i . using Vero E6 cells and Fig 4 shows an additional slight increase from day fifteen to day 20 after infection . The detachment of L929 cells increased after day fifteen p . i . , hence , due to the low number of attached cells LSM analyses were impossible after that point in time . In neither cell line actin polymerization due to R . helvetica AS819 was detected in contrast to the R . conorii–infected cell lines ( Fig 5 ) . Further , in the latter a prominent plaque formation was visible on day four p . i . indicating rapid cell-to-cell spread . The analysis of images of the MOCK-infected controls did not yield any signals corresponding to rickettsiae . No plaque formation was seen in R . helvetica-infected cell lines at 7 d p . i . ( Fig 6A and 6B ) , 14 p . i . and 21 d p . i . ( Fig 6C and 6D ) . As noticed before , the detachment of L929 cells increased with prolonged incubation time ( Fig 6D ) . In contrast , plaque formation was readily observed in both cell lines infected with R . honei at 7 d p . i . ( Fig 7A and 7B ) . The PCR targeting rickA and subsequent pyrosequencing resulted in a nucleotide sequence of 1 , 677 nt ( accession number MF163038 ) with highest similarity ( i . e . 90%; 1 , 528 nt/1 , 694 nt , 40 gaps ) to the complete coding sequence of the rickA gene of R . raoultii strain DnS14 ( accession number EU340900 ) . The open reading frame encoded a deduced 559 aa sequence . Using the RAST annotation server , this sequence was identified as hypothetical Wiscott-Aldrich Syndrome Protein ( WASP ) -like protein . A protein BLAST search revealed 100% identity ( 559/559 aa ) to a hypothetical protein of the R . helvetica type strain C9P9 ( accession number WP010421970 ) followed by 83% similarity ( 465/560 aa ) to the Arp2/3 complex-activating protein RickA of R . felis ( accession number WP039594871 ) . Hypothetical RickA in R . helvetica AS819 revealed an N-terminal domain for binding monomeric actin ( G-actin binding site ) and several proline ( P ) -rich repeats were counted within the protein sequence ( S1 Fig ) . In addition , the commonly called WCA region composed of a WASP-homology 2 ( WH2 ) region , a central ( C region ) , and an acidic domain was identified in R . helvetica AS819 . As described for partial RickA of different other Rickettsia species , human WASP and N-WASP , a conserved motif ( ФXXФXXФXXXRXXФ ) was found in the C region ( S1 Fig ) , with Ф representing an aliphatic amino acid , X any residue , and R an arginine [32] . In addition , R . helvetica AS819 possesses two WH2 regions ( S1 Fig ) which has also been described for several other rickettsiae . Analyses of nucleotide fragments obtained by pyrosequencing also resulted in 1 , 917 nt ( accession number MF163040 ) that revealed 97% ( 1 , 863 nt/1 , 918 nt , 13 gaps ) similarity to the partial protein PS 120 ( D ) gene sequence of R . helvetica type strain C9P9 ( accession number AF163009 ) but 99% ( 1 , 901/1 , 918 nt , 13 gaps ) to the partial sca4 nucleotide sequence of R . asiatica strain IO-1 ( accession number DQ110869 ) . Four partial R . helvetica sca4 gene sequences found in GenBank were identical ( 767/767 nt , accession number KR150775 ) or revealed 99% similarity ( accession numbers KT825971 , KT825970 , FJ358501 ) to the sequence of R . helvetica AS819 . The open reading frame resulted in a 639 aa sequence that shared 95% ( 599/632 aa , 4 gaps ) similarity to the partial protein PS 120 sequence of R . helvetica type strain C9P9 ( accession number AAL23857 ) and revealed 99% identity ( 630/637 aa , 4 gaps ) to the partial Sca 4 sequence of R . asiatica strain IO-1 ( accession number AAZ83584 ) . Within the aa sequence two stretches were recognized that resemble vinculin-binding sites ( Fig 8 ) . Furthermore , a 4 , 884-nucleic acid-stretch ( accession number MF163039 ) with highest identity ( 99% , 4 , 858/4 , 888 nt , 7 gaps ) to the R . helvetica C9P9 complete sca2 gene sequence ( accession number AY355375 ) was obtained . This sequence contains also multiple stop codons and therefore seems to be a pseudogene . Therefore , the predicted aa sequence is disrupted compared to other rickettsiae ( S2 Fig ) most likely resulting in a non-functional protein .
The pathogenicity of R . helvetica has not been investigated in depth to date . Its main vector , I . ricinus , is considered as a generalist with an extraordinarily broad host spectrum including mammals , birds , and reptiles [33] . Hence , potential transmission of R . helvetica to humans seems likely . In humans , seroprevalences up to 12 . 5% against R . helvetica have been demonstrated with forest workers being predisposed to infection [34–37] . However , fulminant clinical cases are rare indicating a rather low pathogenicity compared to other rickettsiae . There has been only one study describing the life cycle , growth characteristics and host cell response of R . helvetica in a Vero cell line [24] besides the first description of this species [25] . Concurrent to the results by Elfving et al . [24] we observed a short lag phase of up to four days after infection of the monolayers . However , in contrast to that study [24] R . helvetica AS819 grew rather slowly . One possible explanation might be that host cell fragments in our seed culture competed with intact host cells for rickettsial attachment thereby decreasing uptake efficiency as has been described for R . prowazeki seeds [38] . Elfving et al . [24] also reported a lag phase but used suspensions of lysed cells . Moreover , the culture supernatant of our seed culture may have contained an unknown amount of dead or late-growth-phase rickettsia that were no longer in an active growth state [38] . The latter would lead to a lag phase in the intracellular growth [38] as was noticed in our experiments . The percentage of viable R . helvetica AS819 was not assessed hence , the copy numbers calculated for the seed may have included a large amount of dead organisms . The differences between the Swedish and our study might also be attributed to the different culture techniques used in the respective study: conventional culture in the study at hand versus shell vial centrifugation technique used by the Swedish colleagues [24] . The latter technique has been described to increase infectivity [39] but does not resemble the host-pathogen-interaction during natural infection . No intranuclear R . helvetica AS819 was detected in the first description [25] and in our experiments , which is in addition in contrast to the study by Elfving et al . [24] and might also result from the centrifugation step used in their study . We showed that R . helvetica AS819 grew constantly but not rapidly after a phase of adaptation in both tested Vero E6 and L929cell lines , which was confirmed by the LSM investigations . LSM revealed that the dissemination of R . helvetica AS819 in both cell lines was rather mediated by cell break-down and bacterial release than cell-to-cell spread as has also been described for R . prowazekii [40] . This was indicated by the irregularly infected cell layers . R . helvetica AS819 did not produce cytopathogenic effects in Vero E6 and L929 cells which is in agreement with the first description of this species [25] . In addition , actin polymerization due to R . helvetica AS819 was not detected in both cell lines which has also been described for R . helvetica elsewhere [25] . This adds to the previous suggestion that R . helvetica lacks intracellular motility which makes it unlikely to invade nuclei [25 , 41] . Moreover , plaque formation in Vero E6 and L929 cells was absent after 21 d of infection suggesting that R . helvetica does not spread from cell to cell . Interestingly , this is in contrast to Rolain et al . [42] who reported formation of small plaques after up to 8 d of incubation using the same cell lines . This difference might be due to strain-specific variation in growth characteristics and virulence as has been described for different strains of R . rickettsii [43–45] . Moreover , the number of serial passages of cell lines and rickettsia can influence growth characteristics [42–43 , 46] . RickA-mediated nucleation of actin plays a role in the intracellular spread of some species [16 , 47] . Although R . helvetica AS819 possesses a gene encoding for RickA , a bacterial actin nucleator most closely related to WASP/N-WASP-family proteins [19] , no host actin polymerization ( HAP ) was observed . This has also been described for R . raoultii where rickA expression is not sufficient to promote actin-based motility [29] . Furthermore , R . felis and R . parkeri possess genes encoding full-length RickA but lack spread by HAP [48] . A disruption of the rickA coding sequence as has been shown in R . peacockii and REIS [49–50] was not confirmed in R . helvetica AS819 by our sequence analyses . In addition to RickA , Sca2 has been suggested to be necessary for the actin-based motility of rickettsiae [18] . Cardwell and Martinez [22] demonstrated that the first third of the Sca2 passenger domain is highly conserved among SFG rickettsia with the exception of a 39 amino acid deletion . They showed that the deletion of residues 309 to 347 resulted in complete abolishment of actin assembly in R . conorii [22] . Sca2 aa-sequence analysis from R . helvetica AS819 revealed a disrupted aa-sequence resulting in a non-functional protein . This is in concordance to the R . helvetica type strain C9P9 . In this strain , sca2 has been deemed a pseudogene with one or more fragments that don’t span the complete protein [48] . Most likely , this might be responsible for the lack of actin tails observed in R . helvetica AS819-infected cell lines . Sca2 also seems to play an important role in the initial bacterial-host interaction and Sca2 of R . conorii mediates both adhesion and invasion of mammalian cells in vitro [22] . However , our R . helvetica isolate AS819 did not exhibit an appreciable defect in adherence or invasion of Vero and L929 cells in vitro . This might be attributed to other highly conserved proteins which may compensate for the lack of functional Sca2 [22] . Only recently , Sca4 was identified as another protein from SFG rickettsia that promotes spread [23] . Specifically , Sca4 of R . parkeri binds to the cell-adhesion protein vinculin and inhibits its activity thereby reducing intercellular tension forces [23 , 51] . For R . helvetica isolate AS819 a 639 aa-stretch was identified that revealed 95% similarity to Sca4 of R . helvetica C9P9 . For the latter species Sca4 has been supposed to be a truncated protein [48] which most certainly might also be the case for the strain investigated in this study . Cell-to-cell spread is one crucial step in the intracellular life cycle of several pathogens including rickettsiae . Actin-based motility contributes to cell-to-cell spread and dissemination within the host . In contrast to other rickettsiae , R . helvetica isolate AS819 did not spread directly from cell to cell by actin-based motility presumably due to a deletion in the predicted Sca2 protein . As Sca2 is needed for virulence [14] our results suggest less virulence and pathogenicity of R . helvetica isolated from ixodid ticks in Germany .
|
The pathogenicity of Rickettsia helvetica has not been investigated in depth to date . In humans , seroprevalences up to 12 . 5% against R . helvetica have been demonstrated with forest workers being predisposed to infection . However , fulminant clinical cases are rare indicating a rather low pathogenicity compared to other Rickettsia species . We therefore investigated growth characteristics of a R . helvetica tick isolate ( AS819 ) in two different eukaryotic cell lines with focus on ultra-structural changes of host cells during infection as determined by confocal laser scanning microscopy . Further investigations included sequencing of rickA , sca4 and sca2 genes , which have been reported to encode proteins involved in cell-to-cell spread in some rickettsiae . In contrast to what is known from other rickettsiae , R . helvetica did not spread directly from cell to cell by actin-based motility presumably due to a deletion in the predicted Sca2 protein . As Sca2 is needed for virulence our results might indicate less virulence and pathogenicity of R . helvetica isolated from ixodid ticks in Germany .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"nuclear",
"staining",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"biological",
"cultures",
"microbiology",
"light",
"microscopy",
"rickettsia",
"eukaryotic",
"cells",
"confocal",
"laser",
"microscopy",
"microscopy",
"sequence",
"motif",
"analysis",
"confocal",
"microscopy",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"l929",
"cells",
"sequence",
"analysis",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"sequence",
"alignment",
"bioinformatics",
"medical",
"microbiology",
"microbial",
"pathogens",
"cell",
"lines",
"dapi",
"staining",
"host",
"cells",
"cell",
"biology",
"virology",
"database",
"and",
"informatics",
"methods",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] |
2018
|
In vitro studies of Rickettsia-host cell interactions: Confocal laser scanning microscopy of Rickettsia helvetica-infected eukaryotic cell lines
|
Human cytomegalovirus ( CMV ) is a herpes virus with poorly understood transmission dynamics . Person-to-person transmission is thought to occur primarily through transfer of saliva or urine , but no quantitative estimates are available for the contribution of different infection routes . Using data from a large population-based serological study ( n = 5 , 179 ) , we provide quantitative estimates of key epidemiological parameters , including the transmissibility of primary infection , reactivation , and re-infection . Mixture models are fitted to age- and sex-specific antibody response data from the Netherlands , showing that the data can be described by a model with three distributions of antibody measurements , i . e . uninfected , infected , and infected with increased antibody concentration . Estimates of seroprevalence increase gradually with age , such that at 80 years 73% ( 95%CrI: 64%-78% ) of females and 62% ( 95%CrI: 55%-68% ) of males are infected , while 57% ( 95%CrI: 47%-67% ) of females and 37% ( 95%CrI: 28%-46% ) of males have increased antibody concentration . Merging the statistical analyses with transmission models , we find that models with infectious reactivation ( i . e . reactivation that can lead to the virus being transmitted to a novel host ) fit the data significantly better than models without infectious reactivation . Estimated reactivation rates increase from low values in children to 2%-4% per year in women older than 50 years . The results advance a hypothesis in which transmission from adults after infectious reactivation is a key driver of transmission . We discuss the implications for control strategies aimed at reducing CMV infection in vulnerable groups .
Human cytomegalovirus ( CMV ) is a highly prevalent herpesvirus that infects between 30% and 100% of persons in populations throughout the world [1] . Usually thought to be a relatively benign persistent infection , CMV is able to cause serious disease in the immunocompromised and offspring of pregnant women with an active infection [2–5] . CMV also has been implicated in a variety of diseases in healthy persons [4 , 6–8] , and plays a role in aging of the immune system [9–12] , perhaps thereby reducing the effectiveness of vaccination in older persons [13–15] . Although the importance of CMV to public health is acknowledged , and even though the development and registration of a vaccine has been declared a priority [16 , 17] , little quantitative information is available on the transmission dynamics of CMV . At present , the only population-level data derive from serological studies , aiming to uncover which part of the population is infected at what age . These studies show that i ) a sizable fraction of infants is infected perinatally ( before 6 months of age ) , ii ) seroprevalence increases gradually with age and is usually higher in females than in males , and iii ) the probability of seropositivity is associated with both ethnicity and socioeconomic status , with non-western ethnicity and lower socioeconomic status being associated with higher rates of seropositivity [1 , 18–21] . CMV infection has a profound impact on the human immune system . Most prominently , it is able to mould the T cell immune repertoire , in particular by expansion of the CMV-specific CD8+ memory T cell pool , a phenomenon called memory inflation [12] . Similar result have been found for memory B cell immunity [22] . With regard to humoral immune responses , high levels of CMV-specific IgG antibodies are increasingly considered a biomarker for lack of control by the immune system of the host , and have been associated with high probability of reactivation ( [23 , 24] , see [12] and references therein ) . In view of this , it is not surprising that evidence is accumulating of an association between high levels of CMV-specific IgG antibodies , inflammation , vascular disease , and mortality [6 , 7] . Person-to-person transmission of CMV from an infected to an uninfected person can occur from a primary infected person , or from a person who is experiencing a reactivation episode or from a person who has been reinfected [4] . Here , we analyze data from a large-scale serological study to obtain quantitative estimates of the relative importance of these transmission routes [21] . We fit mixture models linked to age- and sex-specific transmission models to the data to study the ability of different hypotheses explaining the serological data . Specifically , we quantify the incidence and transmissibility of primary infection , re-infection , and reactivation . Throughout , our premise is that measurements of antibody concentrations provide information on whether or not a person has been infected , and whether or not re-infection or reactivation have occurred . Persons with low measurements are considered uninfected ( susceptible ) , while persons with intermediate and high antibody concentrations are infected with and without subsequent re-infection or reactivation , respectively . The analyses show that infectious reactivation in adults is necessary to explain the data , and is expected to be an important driver of transmission . The results have implications for control of CMV by vaccination , but also in the broader context of T cell immune memory inflation , vascular disease , and immunosenescence [12 , 25 , 26] .
The study was approved by the Medical Ethics Testing Committee of the foundation of therapeutic evaluation of medicines ( METC-STEG ) in Almere , the Netherlands ( clinical trial number: ISRCTN 20164309 ) . All participants or their legal representatives had given written informed consent . The analyses make use of sera from a cross-sectional population-based study carried out in the Netherlands in 2006-2007 . Details have been published elsewhere [21 , 27] . Briefly , 40 municipalities distributed over five geographic regions of the Netherlands were randomly selected with probabilities proportional to their population size , and an age-stratified sample was drawn from the population register . A total of 19 , 781 persons were invited to complete a questionnaire and donate a blood sample . Serum samples and questionnaires were obtained from 6 , 382 participants . To exclude the interference of maternal antibodies , we restrict analyses to sera from persons older than 6 months ( 6 , 215 samples ) . We further select Dutch persons and migrants of Western ethnicity to preclude confounding by ethnicity ( 5 , 179 samples ) and stratify the data by sex [21] , yielding 2 , 842 and 2 , 337 samples from female and male participants , respectively . The data are available at github . com/mvboven/cmv-serology . We use the ETI-CYTOK-G PLUS ( DiaSorin , Saluggia , Italy ) Elisa to detect CMV-specific IgG antibodies . The assay yields continuous measurements ( henceforth called ‘antibody concentration’ ) . A small number of samples is right-censored ( 140 persons ) . We perform a Box-Cox transformation of the data ( λ = 0 . 3 ) , yielding a distribution of low antibody concentrations ( -2 . 8< x ≤-0 . 5 ) that is approximately normal . According to the provider of the assay , samples with ( transformed ) measurement lower than -0 . 8 U/ml should be considered uninfected , while samples with measurement greater or equal than -0 . 8 U/ml should be classified as infected . Right-censoring is applied to the 140 samples above the upper limit of 3 . 41 U/ml . The data with model fit ( see below ) are shown in Fig 1 . The data are analyzed statistically using a mixture model with sex- and age-specific mixing functions . We distinguish three distributions , describing samples of low ( susceptible , S ) , intermediate ( latently infected , L ) , and high ( latently infected with increased antibodies , B ) antibody concentrations . The L and B distributions are modeled using normal distributions with means and standard deviations independent of age and sex . The S distribution is modeled by a mixture of a spike and a normal distribution ( an inflated normal distribution ) , as there appears a spike at -2 . 91 U/ml in the data ( 263 persons ) . In this way , samples with concentration at the spike belong to the susceptible component with probability 1 . We model the probability of each of the three outcomes in terms of log-odds , taking the probability of being in the S component as reference . This allows us to write the log-odds of being in component L or B as linear functions of age and sex . The design matrix of the resulting multinomial logistic model consists of natural cubic splines with interior knots at 20 , 40 and 60 years and boundary knots at 0 and 80 years . Hence , the mixing functions ( prevalences ) have flexible shape , which allows these to be optimally informed by the data . In the results , sex is put in the model as main effect , as analyses show no improvement in fit when including age by sex interaction . We estimate parameters in a Bayesian framework using R and JAGS [28 , 29] . Non-informative normal prior distributions are set on the means of the three component distributions ( N ( 0 , 0 . 001 ) ) ( mean and precision ) . Label switching is prevented by prior ordering of the means . The precisions of the components are given flat Gamma prior distributions ( Γ ( 0 . 5 , 0 . 005 ) ) . The spline parameters are also given non-informative normal prior distributions ( N ( 0 , 0 . 001 ) ) . We apply a QR-decomposition to the design matrix to improve mixing and run 10 MCMC chains in parallel , yielding a total of 10 , 000 samples . We apply an 1/10 thinning to give a well-mixed 1 , 000 samples from the posterior distribution . Next to the mixture model analyses , we estimate parameters of transmission models to investigate the ability of different transmission hypotheses explaining the data . To facilitate comparison between transmission models , take the medians of the estimated mixture distributions as input . In line with the above , we focus on a sex- and age-structured model in which persons are probabilistically classified as uninfected ( S ) , latently infected ( L ) , and latently infected after reactivation or re-infection ( B ) . As the infectious period is short relative to the lifespan of the host ( weeks versus decades ) , the infectious periods are modeled implicitly using the short-disease approximation [30] . Further , we focus on the endemic equilibrium of the transmission model so that all variables are time-independent [30 , 31] . Fig 2 shows a schematic of the model . For sexes i ∈ {♀ , ♂} , the differential equations for the age-specific relative frequencies S ( a ) , L ( a ) , and B ( a ) ( S ( a ) + L ( a ) + B ( a ) = 1 ) are given by d S i ( a ) d a = - λ i ( a ) S i ( a ) d L i ( a ) d a = λ i ( a ) S i ( a ) - ( ρ i ( a ) + z λ i ( a ) ) L i ( a ) d B i ( a ) d a = ( ρ i ( a ) + z λ i ( a ) ) L i ( a ) , ( 1 ) with forces of infection λ i ( a ) = ∑ j ∈ { ♀ , ♂ } ∫ 0 M c i j ( a , a ′ ) ( β 1 λ j ( a ′ ) S j ( a ′ ) + β 2 ( ρ j ( a ′ ) + z λ j ( a ′ ) ) L j ( a ′ ) ) d a ′ . ( 2 ) In Eqs ( 1 ) and ( 2 ) , zλj ( a ) and ρj ( a ) are the age-specific re-infection and reactivation rates , z is the susceptibility to re-infection of latently infected persons relative to the susceptibility of uninfected persons ( 0 ≤ z ≤ 1 ) , cij ( a , a′ ) represents the contact rate between persons of age a′ and sex j , and those of age a and sex i [32 , 33] , β1 and β2 are proportionality parameters determining the transmissibility of primary infection and reactivation/re-infection , and M is the maximum age . As the data do not extend beyond 80 years we take M = 80 years . Notice that λj ( a ) Sj ( a ) and ( ρj ( a ) + z λj ( a ) ) Lj ( a ) are the incidence of primary infection and the incidence of reactivation and re-infection , so that β1λj ( a ) Sj ( a ) and β2 ( ρj ( a ) + z λj ( a ) ) Lj ( a ) are the infectious output generated by primary infection and reactivation/re-infection , respectively [30] . As in earlier studies , contact rates are hard-wired into the model using data on social contact patterns , thereby adopting the social contact hypothesis [32–34] . Here we use the mixing matrix based on reported physical contacts [32] . The discretized contact function and demographic data are available at github . com/mvboven/cmv-serology . Below , we consider a suite of simplifications and variations of the full model specified by Eqs ( 1 ) and ( 2 ) . In the simplifications , we assume that ( i ) there is no re-infection ( z = 0 ) , ( ii ) there is no reactivation ( ρi ( 0 ) = 0 ) , or ( iii ) reactivation and re-infection are not infectious ( β2 = 0 ) . We also consider a variation of the model in which re-infection and reactivation do not only occur upon transition from L to B , but also in the B compartment . In these models the infectious output generated by reactivation and re-infection in Eq ( 2 ) ( β2 ( ρj ( a′ ) + zλj ( a′ ) ) Lj ( a′ ) ) is replaced by β2 ( ρj ( a′ ) + zλj ( a′ ) ) ( Lj ( a′ ) + Bj ( a′ ) ) . The differential equations can be solved in terms of the forces of infection using the variation of constants method . Here we assume , based on results of the mixture model , that a non-negligible fraction of infants is infected in the first six months of life and the fraction infected is equal in female and male infants [21] . Hence , we have S♀ ( 0 ) = S♂ ( 0 ) = S0 , L♂ ( 0 ) = L♀ ( 0 ) = 1 − S0 , and B♀ ( 0 ) = B♂ ( 0 ) = 0 as initial conditions , and the solution of ( 1 ) is given by S i ( a ) = S 0 i e − ∫ 0 a λ ( τ ) d τ L i ( a ) = ( 1 − S 0 i ) e − ∫ 0 a ρ ( τ ) + z λ ( τ ) d τ + S 0 i ∫ 0 a λ ( τ ) e − ∫ 0 τ λ ( τ ′ ) d τ ′ − ∫ τ a ρ ( τ ′ ) + z λ ( τ ′ ) d τ ′ d τ . ( 3 ) Insertion of Eq ( 3 ) in Eq ( 2 ) yields two integral equations for the age-specific forces of infection in females and males [34–37] . These equations cannot be solved explicitly in general . It is possible , however , to solve the equations for specific functions . Here , we assume that reactivation and contact rates are constant on certain predefined age-intervals . From Eq ( 2 ) , it then follows that the force of infection is piecewise constant as well . Throughout , we consider age intervals of fixed size Δa = 5 years , so that the limits of the n = M/Δa = 16 age classes are defined by the vector a = ( 0 , Δ a , 2Δ a , … , nΔ a ) . Hence , the j-th class ( j = 1 , … , n ) contains all persons with age in the interval [a[j] , a[j + 1] ) , where a[j] denotes the j-th element of a . Subsequently , the forces of infection λi ( a ) and reactivation rates ρi ( a ) are replaced by their counterparts λ j i and ρ j i . Similarly , Si ( a ) , Li ( a ) , and Bi ( a ) at the borders of the age-intervals are given by S j i , L j i , and B j i . Insertion in Eq ( 3 ) and integrating over the ( constant ) rates yields S j i = S 0 e - Δ a ∑ k = 1 j λ k i L j i = ( 1 - S 0 ) e - Δ a ∑ k = 1 j ρ k i + z λ k i + S 0 ∑ k = 1 j λ k i e - Δ a ( ρ k i + z λ k i ) - e - Δ a λ k i ( 1 - z ) λ k i - ρ k i e - Δ a ( ∑ ℓ = 1 k - 1 λ ℓ i - ∑ ℓ = k + 1 j ρ ℓ i + z λ ℓ i ) , ( 4 ) where i ∈ {♀ , ♂} and B j i = 1 - S j i - L j i . Insertion of Eq ( 4 ) in Eq ( 2 ) and making use of the fact that the cumulative incidences of infection and reactivation/re-infection in age class j are given by ∫ a [ j ] a [ j + 1 ] λ i ( a ) S i ( a ) d a = S i ( a [ j ] ) - S i ( a [ j + 1 ] ) and Bi ( a[j + 1] ) − Bi ( a[j] ) , yields 32 equations ( 16 per sex ) for the 32 forces of infection . As in the mixture model with spline mixing parameters , the log-likelihood of each observation is given by a mixture distribution , where the spline functions are replaced by Si ( a ) , Li ( a ) , and Bi ( a ) . For instance , the likelihood contribution of a sample with antibody measurement c in a person of sex i and age a is given by S i ( a ) f S ( c ) + L i ( a ) f L ( c ) + B i ( a ) f B ( c ) , where Si ( a ) , Li ( a ) , and Bi ( a ) are the age specific prevalences in sex i , and fS ( c ) , fL ( c ) , and fB ( c ) are the densities of the mixture distributions at antibody concentration c . In both sexes , reactivation rates are modeled by piecewise constant functions with steps at 20 and 50 years , i . e . with rates that are constant on the intervals [0 , 20 ) , [20 , 50 ) , and [50 , 80 ) years . Hence , the reactivation rates are characterized by three parameters in each sex , viz . ρ [ 0 , 20 ) i , ρ [ 20 , 50 ) i , and ρ [ 50 , 80 ) i ( i ∈ {♀ , ♂} ) . Bayesian parameter estimates are obtained using Markov chain Monte Carlo ( MCMC ) . Initially , results were obtained using tailored Mathematica code , using a single-component random walk metropolis algorithm while solving the consistency equations for the forces of infection using a Quasi-Newton ( secant ) method . As this became exceedingly slow for specific models , we recoded the models using Hamiltonian Monte Carlo with Stan ( mc-stan . org ) . Here , the discretized equations for the forces of infection ( 2 ) are solved by specifying that the differences between the left- and right-hand sides are small , and approximately N ( 0 , 10 - 4 ) ( mean and scale ) distributed . Cross-checking of the two methods yielded very similar results . All programs are available at github . com/mvboven/cmv-serology . Prior distributions of the parameters are as follows: β 1 ∼ N ( 0 . 1 , 10 ) ( mean and scale ) , β 2 ∼ N ( 0 . 1 , 10 ) , z ∼ U ( 0 , 1 ) , μ ρ ∼ N ( 0 , 10 ) , 1 / σ ρ ∼ N ( 0 , 10 ) , and ρ x i ∼ N ( μ ρ , σ ρ ) for all i and x . Whenever applicable , distributions are truncated to be positive . With these prior parameter distributions , the joint posterior distribution is strongly dominated by the data . Ten chains of 3 , 000 iterations are run in parallel , of which the first 500 iterations ( warmup ) are discarded . We apply 1/5 thinning , yielding a total of 5 , 000 samples per model scenario . For all parameters , effective sample sizes usually lie between 3 , 000 and 4 , 500 . Convergence of chains is assessed visually , and by assessment of the empirical variance within and between chains [38] . To prevent the occurrence of divergent transitions we set ADAPT_DELTA = 0 . 99 . Parameter estimates and bounds of credible intervals are represented by 2 . 5 , 50 , and 97 . 5 percentiles of the posterior samples . Results are usually obtained in 1-3 hours on a personal computer . Model selection is based on WAIC , a measure for predictive performance , and WBIC , a measure for identifying the most likely model generating the data [39–41] . WAIC is obtained directly from the posterior likelihood using the R-package loo ( cran . r-project . org ) . WBIC is calculated in a separate run as the average log likelihood over the posterior samples , using a sampling ‘temperature’ determined by the number of observations [39] .
Fig 1 presents the data stratified by sex and age , with fit of the statistical model . The data and model fit show peaks at low antibody measurements ( -2 . 9 U/ml and ≈-2 U/ml ) , corresponding to uninfected persons ( denoted by S ) . In both sexes , there is a third peak at higher measurements ( 1-3 U/ml ) that shifts to higher values with increasing age . This peak is composed of persons who are infected ( denoted by L ) and persons who are infected with high antibody concentrations ( denoted by B ) . Overall , the model appears to describe the data well . This is confirmed in Fig 3 , which shows the estimated components of the mixture distribution and diagnostic characteristics of the classification . The component distribution of uninfected persons hardly overlaps with the two component distributions for infected persons , while there is some overlap between the distributions of infected persons . This can be made more precise using detection theory . Specifically , in Fig 3 we graph the specificity Sp ( the probability of correctly classifying a negative subject ) and sensitivity Se ( the probability of correctly classifying a positive subject ) in a receiver operating characteristic ( ROC ) graph with antibody concentration specifying a cut-off for binary classification as parameter [42–44] . Subsequently , we use the maximal Youden index ( i . e . max ( Se + Sp − 1 ) ) to choose an optimal cut-off , and find that classification of persons as uninfected versus infected is near perfect ( Youden index: 0 . 97 , at cut-off -0 . 70 U/ml ) , while classification of persons with high antibody concentrations is good ( Youden index: 0 . 71 , at cut-off 1 . 81 U/ml ) . These results show that the classification is supported by the data ( i . e . has high probability yielding an informed decision ) . We further investigate whether mixture models with fewer or more components are able to provide an even better description of the data , and found that a model with two mixture components does not perform well ( ΔWAIC = 300 . 2 in favor of the three-component mixture distribution ) , while performance of models with four components depends sensitively on choice of prior distribution of the fourth distribution , and often yields broad posterior antibody distributions with small estimated prevalence that overlap with the other three component distributions . Hence , a mixture model with three components gives an optimal description of the data . Fig 4 shows the estimated prevalences in females and males as a function of age [42–44] . The prevalence of uninfected persons decreases gradually with age , from approximately 0 . 80 in infants ( females: 0 . 81 , 95%CrI: 0 . 77-0 . 85; males: 0 . 80 , 95%CrI: 0 . 76-0 . 84 ) to 0 . 27 ( 95%CrI: 0 . 22-0 . 34 ) and 0 . 38 ( 95%CrI: 0 . 32-0 . 45 ) at 80 years in females and males , respectively . In both females and males the latently infected prevalence remains approximately constant , ranging from 0 . 15 to 0 . 20 in females and from 0 . 18 to 0 . 28 in males . In contrast , the prevalence of persons with increased antibodies increases strongly with age , especially in females . In fact , the prevalence of persons with increased antibodies increases from 0 . 09 ( 95%CrI: 0 . 06-0 . 13 ) at 20 years to 0 . 57 ( 95%CrI: 0 . 47-0 . 67 ) at 80 years in females , and from 0 . 04 ( 95%CrI: 0 . 03-0 . 07 ) to 0 . 37 ( 95%CrI: 0 . 28-0 . 46 ) in males . Hence , in older persons the prevalence of persons with increased antibodies is 54% ( or 20 per cent points ) higher in females than in males . Of particular interest is the prevalence of infection in females of childbearing age , as this group is at risk of transmission to the fetus or newborn . Using the above analyses , we find that the prevalence of infection ( i . e . the combined prevalence in the L and B compartments ) is 0 . 30 ( 95%CrI: 0 . 27-0 . 33 ) in 20-year-old females and 0 . 42 ( 95%CrI: 0 . 39-0 . 46 ) in 40-year-old females . If we combine these figures with the observation that approximately 20% of children are infected at six months of age , and that less than 5% of children in the Netherlands in 2007 had a mother under 20 years or over 40 years , we deduce that the probability of perinatal transmission could be between 0 . 20/0 . 42 = 0 . 48 and 0 . 20/0 . 30 = 0 . 67 , with the exact figure depending on the distribution of ages at which mothers give birth . In addition , one could envisage that the highest risk of ( severe ) infection of the fetus or newborn is when mothers are infected or experience a reactivation episode . The estimated rates at which susceptible females of 20 and 40 years are infected are 0 . 0055 per year ( 95%CrI: 0 . 0036-0 . 0077 ) and 0 . 0092 per year ( 95%CrI: 0 . 0069-0 . 011 ) per year , respectively . The rates at which latently infected females of 20 and 40 years are re-infected or experience a reactivation episode are of similar magnitude , and are estimated at 0 . 0059 per year ( 95%CrI: 0 . 0038-0 . 0086 ) and 0 . 0093 per year ( 95%CrI: 0 . 0064-0 . 012 ) , respectively . The overall rates of infection , reactivation , and re-infection in 20 and 40 year-old females are given by the sum of the above estimates , and are approximately 1% and 2% per year , respectively . To evaluate the ability of different transmission hypotheses explaining the data , and to obtain parameter estimates that have a biological interpretation , we analyzed the data with transmission models . A comparison of model scenarios based on the information criteria WAIC and WBIC is given in Table 1 . Overall , the analyses show that models with the possibility of multiple infectious reactivations perform best ( Models E and F; lowest WAIC and WBIC ) , that models with at most one infectious reactivation perform worse ( Models A and B; ΔWAIC and ΔWBIC ≈10 − 15 ) , and that models without reactivation or with reactivation not being infectious have very low support ( Models C , D , and G ) . These results indicate that infectious reactivation is key to adequately explain the data with transmission models . This is true in our model with contact structure based on reported physical contacts [32] , and also in an alternative model formulation that assumes a uniform contact structure ( ΔWAIC = 151 . 9 in favor of the model with reactivation over the model without reactivation and no re-infection ) . Within the set of models with infectious reactivation there are only small differences between models that do and do not incorporate re-infection ( Model A versus Model B , and Model E versus Model F ) . This indicates that while infectious reactivation is essential to adequately describe the data , the analyses are inconclusive with respect to whether or not infectious re-infection should be included . Fig 5 and Table 2 show parameter estimates of the model with highest statistical support ( as judged by WBIC ) . The preferred model ( Model E ) includes multiple reactivations and re-infections , infectious reactivation , and infectious re-infection . In this model , the estimated transmissibility of primary infection ( β1 ) is much lower than the transmissibility of reactivation/re-infection ( β2 ) . In fact , the posterior median of β2 is more than an order of magnitude larger than the posterior median of β1 . Further , the relative susceptibility to re-infection ( i . e . the probability of re-infection in a contact that would lead to infection if the contacted person were uninfected ) has a broad posterior distribution , and cannot be estimated with meaningful precision from the data ( z ^ = 0 . 32 ; 95%CrI: 0 . 017-0 . 84 ) . Similar findings are obtained in other model scenarios , in particular Models A-B and E-F ( Table 1 ) . Estimates of the reactivation rates are quantitatively close in models with high support ( Models E-F ) . Reactivation rates generally increase with increasing age , and are substantially higher in females than in males . In the preferred model ( Model E ) , the estimated reactivation rate is 0 . 013 per year ( 95%CrI: 0 . 0042-0 . 021 ) in 0-20 year-old females , which increases to 0 . 021 per year ( 95%CrI: 0 . 013-0 . 029 ) in 20-50 year-old females , and then increases further to 0 . 028 per year ( 95%CrI: 0 . 017-0 . 040 ) in 50 + -year-old females ( Table 2 ) . The corresponding reactivation rates in males are 0 . 0054 per year ( 95%CrI: 0 . 0035-0 . 013 ) , 0 . 011 per year ( 95%CrI: 0 . 0035-0 . 018 ) , and 0 . 013 per year ( 95%CrI: 0 . 0043-0 . 021 ) . These estimates are slightly higher and slightly more precise in the model without re-infection ( Model F ) , and somewhat higher in models with a single reactivation/re-infection event ( Models A-B ) . In the two models with highest support ( Models E-F ) , estimates of the force of infection increase from approximately 0 . 012-0 . 013 per year in the youngest age group to 0 . 014-0 . 017 per year in 10-15 year-old girls ( Fig 6 ) . Owing to the slightly higher contact rates in females than in men , the estimated force of infection is usually slightly higher in females than in males in the age groups 10-25 years [32] . In older age groups , estimates of the forces of infection decrease to lower values ( ≈0 . 01 per year ) . Noteworthy , the extreme age-specific differences in the force of infection usually observed for directly transmitted infectious diseases , with high infection rates in children and much lower rates in adults , are much less pronounced here due to infectious reactivation in older age strata combined with age-assortative mixing [32 , 34 , 35] . In models with re-infection , estimates of re-infection rate ( zλi ( a ) ) are considerably smaller than estimates of the reactivation rates ( ρi ( a ) ) because the estimated forces of infection ( λi ( a ) ) are usually lower than the reactivation rates , especially in females ( Fig 6 ) . Hence , re-infection contributes little to boosting of the antibody concentrations in those age groups where most of the boosting occurs ( >20 years; Fig 4 ) . In fact , in adult females it is not uncommon that the reactivation rate is more than an order of magnitude higher than the estimated re-infection rate ( log10 ( ρ♀ ( a ) / ( zλ♀ ( a ) ) ) > 1 ) .
Our study of population-wide serological data shows that IgG antibody concentrations contain a wealth of information on the transmission dynamics of CMV . Specifically , the analyses reveal that ( i ) the prevalence of CMV increases gradually with age such that at old age the majority of persons in the Netherlands are infected; ( ii ) except for the very young , the prevalence of CMV is systematically higher in females than in males . This is mainly due to a higher incidence of infection in adult women than in adult men of similar age; ( iii ) antibody concentrations in seropositive ( i . e . infected ) persons increase monotonically with age , especially in women; ( iv ) the above findings ( i ) - ( iii ) cannot be explained by simple transmission models in which only primary infection is infectious . This is caused by the fact that transmissibility of primary infection determines the rate at which age-specific prevalence increases; if transmissibility of primary infection would be high then a high prevalence of infection is expected in children . In other words , the fact that seroprevalence increases gradually with age puts an upper bound on the force of infection , and this in turn constrains the transmissibility of primary infection to low values . While aforementioned findings ( i ) - ( iii ) have been noticed before in other settings ( [1] and references therein , [21] ) , our analyses are the first to provide precise estimates using a large population sample . Moreover , the results lead us to a new transmission hypothesis in which infectious reactivation is a key driver of transmission of CMV in the population . Since several other studies have found a gradual increase in seroprevalence [1] , this explanation may not be restricted to the Dutch situation , but hold in general . Underpinning this hypothesis , next to the well-known observations of shedding of CMV in breast milk and cervical material in the third trimester of pregnancy [45–47] , detectable virus also has been found in healthy adults in one study [24] , while in another study CMV DNA has been detected in urine of the majority of older persons [23] . The main implication is that the majority of CMV infections may not be caused by transmission among children after primary infection , even though levels of shedding can be high in infants [46 , 48] , but rather by older persons who go through one or more reactivation episodes . This contrasts with common childhood diseases such as measles , mumps , rubella , and pertussis . For these pathogens , infection in unvaccinated populations generally occurs at a young age , and children are the drivers of transmission . It also contrasts with other herpes viruses such as varicella zoster virus and Epstein-Bar virus for which well over 50% of the population is infected at the age of 10 years [34] . It may be comparable with other herpes viruses such as HSV1 and HSV2 , which show a slowly increasing age-specific seroprevalence [49] . A corollary is that persistence of CMV in the population is not possible with transmission from primary infected persons only , and is dependent on infectious reactivation . Currently , we are focusing on making this idea more precise by calculation of the basic reproduction number , and the reproduction numbers of perinatal transmission , primary infection , and reactivation [50] . This will help put bounds on the relative contribution of each of the transmission routes . With infectious reactivation and perinatal infection being putative drivers of transmission , it is to be expected that elimination by vaccination may prove more difficult than for directly transmitted pathogens , as it will require the pool of latently infected persons to dwindle to zero by demographic turnover . This can take up to the lifetime of one generation , and perhaps more if vaccination cannot prevent perinatal transmission to infants who are too young for vaccination . Thus , a question is whether vaccination formulations and strategies exist that minimize the probability of transmission to young infants . This is all the more of importance as a main source of morbidity is by congenital infection , and the timescale on which reductions in congenital disease are expected determines the projected health impact of vaccination [51] . In this context , next to the ability of a vaccine to prevent infection it may perhaps be equally important that a vaccine is able to reduce the probability of reactivation . Such reductions are likely mediated by T-cell responses of the host , and several ( but not all ) vaccines under development are expected to induce boosting of T-cell immune responses [52–54] . A number of limitations and assumptions deserve scrutiny . First , the transmission model analyses assume that the population is in endemic equilibrium . For a single cross-sectional data set such as the one considered in the present study this assumption is unavoidable if one does not want to introduce additional parameters that cannot be estimated by the data . Reassuringly , the patterns of infection present in the serological data have been found in several serological studies carried out in high-income countries over the past decades [1] . Also , no systematic patterns of increasing or decreasing seroprevalence over time have been found , and this is further reason to believe that there have not been major changes in the epidemiology of CMV over time [1] . Second , we assume that antibody measurements not only give information on CMV infection status , but also whether or not reactivation or re-infection have taken place . Unfortunately , there is no direct empirical evidence confirming or falsifying this assumption , and this is an area where in-depth comparison of the infection and immune status of persons with low and high antibody concentrations is urgently needed . Third , the analyses assume that person-to-person transmission is proportional to observed human contact patterns [32 , 33] . Although this assumption is commonly made and has met with considerable success ( e . g . , [33 , 44 , 55 , 56] ) , it is conceivable that transmission of CMV does not abide by the social contact hypothesis , and that a more complex contact structure would be able to explain the patterns of seroprevalence in a simple transmission model . To investigate the impact of the contact structure , we have analyzed transmission models with a uniform contact structure , and found that models with infectious reactivation still provide the best fit to the data ( ΔWAIC > 100; Results ) . As a final limitation we would like to add that , in principle , it is conceivable that the data can be explained alternatively by an intricate interplay between variation in the susceptibility to infection in conjunction with age-specific variations in the strength of the antibody response . Alas , evidence for or against this hypothesis is lacking . Our inferential analyses indicate that the transmissibility of primary infection is much lower than the transmissibility after reactivation . This seems to be at odds with the observation that prolonged and high-level virus shedding can occur in bodily fluids after primary infection in children [46 , 47] . However , it could be that transitions from the infected class to the infected class with increased antibodies are in effect not the result of a single reactivation or re-infection event , but rather the result of multiple underlying reactivations or re-infections . If this were true , as seems plausible , estimates of the reactivation and re-infection rates as well as the transmissibility of reactivation and re-infection should be interpreted as compound parameters that take into account multiple reactivations and re-infections occurring over the lifetime of an infected person .
|
Human cytomegalovirus ( CMV ) is a herpes virus causing lifelong infection . In high-income countries , the probability of infection increases gradually with age such that at old age up to 100% of the population is infected . CMV is thought to be transmitted mainly by transfer of saliva or urine , but little quantitative evidence is available about the transmission dynamics . We analyze serological data to estimate age- and sex-specific rates of infection , re-infection , and reactivation . The analyses show that infectious reactivation ( i . e . reactivation of the virus in an infected person that is sufficient for it to be transmitted to another person ) is essential to explain the data . We propose that infectious reactivation in adults is an important driver of transmission of CMV .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"immune",
"physiology",
"enzyme-linked",
"immunoassays",
"pathology",
"and",
"laboratory",
"medicine",
"infectious",
"disease",
"epidemiology",
"pathogens",
"immunology",
"microbiology",
"cytomegalovirus",
"infection",
"viruses",
"preventive",
"medicine",
"age",
"groups",
"dna",
"viruses",
"antibodies",
"immunologic",
"techniques",
"vaccination",
"and",
"immunization",
"herpesviruses",
"research",
"and",
"analysis",
"methods",
"human",
"cytomegalovirus",
"public",
"and",
"occupational",
"health",
"immune",
"system",
"proteins",
"infectious",
"diseases",
"serology",
"proteins",
"medical",
"microbiology",
"epidemiology",
"microbial",
"pathogens",
"immunoassays",
"people",
"and",
"places",
"biochemistry",
"physiology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"population",
"groupings",
"viral",
"diseases",
"organisms"
] |
2017
|
Infectious reactivation of cytomegalovirus explaining age- and sex-specific patterns of seroprevalence
|
Mutagenic translesion DNA polymerase V ( UmuD′2C ) is induced as part of the DNA damage-induced SOS response in Escherichia coli , and is subjected to multiple levels of regulation . The UmuC subunit is sequestered on the cell membrane ( spatial regulation ) and enters the cytosol after forming a UmuD′2C complex , ~ 45 min post-SOS induction ( temporal regulation ) . However , DNA binding and synthesis cannot occur until pol V interacts with a RecA nucleoprotein filament ( RecA* ) and ATP to form a mutasome complex , pol V Mut = UmuD′2C-RecA-ATP . The location of RecA relative to UmuC determines whether pol V Mut is catalytically on or off ( conformational regulation ) . Here , we present three interrelated experiments to address the biochemical basis of conformational regulation . We first investigate dynamic deactivation during DNA synthesis and static deactivation in the absence of DNA synthesis . Single-molecule ( sm ) TIRF-FRET microscopy is then used to explore multiple aspects of pol V Mut dynamics . Binding of ATP/ATPγS triggers a conformational switch that reorients RecA relative to UmuC to activate pol V Mut . This process is required for polymerase-DNA binding and synthesis . Both dynamic and static deactivation processes are governed by temperature and time , in which on → off switching is “rapid” at 37°C ( ~ 1 to 1 . 5 h ) , “slow” at 30°C ( ~ 3 to 4 h ) and does not require ATP hydrolysis . Pol V Mut retains RecA in activated and deactivated states , but binding to primer-template ( p/t ) DNA occurs only when activated . Studies are performed with two forms of the polymerase , pol V Mut-RecA wt , and the constitutively induced and hypermutagenic pol V Mut-RecA E38K/ΔC17 . We discuss conformational regulation of pol V Mut , determined from biochemical analysis in vitro , in relation to the properties of pol V Mut in RecA wild-type and SOS constitutive genetic backgrounds in vivo .
DNA polymerase V ( pol V ) is induced as part of the SOS regulon in Escherichia coli in response to DNA damage [1] . Pol V is assembled as a UmuD′2C heterotrimeric complex . This complex is activated extremely late in the induction process , at around 45 min after exposure to either UV light or to chemicals that damage DNA [2 , 3] . SOS-induced levels of pol V are about 60 molecules/cell [4] . In the absence of DNA damage , the constitutive level of pol V is barely detectable , ~ 2 molecules/cell observed by live-cell imaging [4] . Damage-induced SOS mutagenesis does not rise above spontaneous levels in the absence of pol V [5–7] . Therefore , pol V appears to be responsible for virtually all the increase in mutagenesis associated with damage-induced induction of the SOS response . This is true even though the two other SOS-induced pols II and IV are present in the cell at high constitutive levels , which increase further and rapidly ( < 1 min ) upon SOS induction [8 , 9] . Presumably to ensure both accurate transmission of genetic information and optimal cellular viability , E . coli takes great pains to restrict pol V access to undamaged DNA through low constitutive expression . Access to damaged DNA is limited by delayed induction , and rapid proteolysis of the Umu proteins [10] , thus affording ample time for the error-free repair of DNA templates to occur prior to calling upon error-prone pol V-catalyzed translesion synthesis ( TLS ) . The temporal control of pol V is just one facet of a highly complex scheme , which encompasses three additional regulatory processes , spatial [4] , conformational [11 , 12] , and internal [13] . Spatial regulation was recently revealed by live-cell imaging studies and entails the synthesis and sequestering of the UmuC subunit on the cell membrane [4] . Release of pol V into the cytosol requires binding to UmuD′2 [4] . However , pol V in the form of UmuD′2C is catalytically “dead” [11 , 13–15] . A subsequent 2-step activation process involving transfer of a RecA monomer from the 3′-proximal tip of a RecA nucleoprotein filament ( RecA* ) to form UmuD′2C-RecA , and then binding a molecule of ATP , is required to produce a catalytically active pol V “mutasome” , pol V Mut = UmuD′2C-RecA-ATP [11 , 13] . Conformational regulation entails serial conversions of a pol V Mut complex from ( i ) an initially catalytically inactive state that is unable to bind to primer/template ( p/t ) DNA to ( ii ) an activated state that copies DNA to ( iii ) a deactivated state that halts further synthesis . The RecA subunit of pol V Mut is retained in both activated and deactivated states [11] . Conformational regulation appears to be governed by reorientation of RecA relative to UmuC and UmuD′2 [12] . Internal regulation , on the other hand , is marked by the presence of a bound ATP and a DNA dependent ATPase activity unique to pol V Mut [13] . The intrinsic pol V Mut ATPase is distinct from the canonical DNA-dependent ATPase of RecA* [13] . The presence of a stably bound ATP , or slowly hydrolysable ATPγS molecule , is necessary for binding pol V Mut to primer-template ( p/t ) DNA [13] . The RecA subunit modulates conformational and internal regulation through its contact points with UmuC and UmuD′2 , and its interaction with ATP [12] . Pol V Mut can be assembled with wild-type RecA and with RecA mutants that exhibit widely different SOS mutagenic phenotypes , thus providing a unique opportunity to relate biochemical to genetic data . In 1982 , Witkin and colleagues identified strains of E . coli with mutations in recA that induced SOS constitutively and caused a ~100-fold increase in spontaneous SOS mutagenesis [16] . One particularly active allele is recA730 [17] , which was subsequently shown to harbor an E38K substitution ( RecA E38K ) [18] . Here , we explore conformational regulation mechanisms revealing how RecA and ATP function in the mutasome as a temperature-dependent on-off toggle switch . The experiments are performed using combinations of pol V Mut assembled with wild-type RecA ( pol V Mut wt ) and with an SOS constitutive RecA mutant containing an E38K amino acid substitution and a deletion of 17 amino acids at the C terminus ( pol V Mut E38K/ΔC17 ) .
DNA polymerases normally act in accord with sequential bisubstrate kinetics by binding first to p/t DNA and then to dNTP substrates . One , or multiple dNMP incorporations can occur during a single DNA binding event , which is followed by dissociation with the subsequent re-initiation of synthesis on another DNA molecule . Continuing pol-DNA cycling occurs until most of the DNA has been copied . However , unlike generic pols , pol V Mut behaves differently , depending on reaction temperature . Pol V Mut carries out just a single-round of DNA synthesis at 37°C , whereas multiple cycles occur at 30°C ( Fig 1 ) . The more active pol V Mut E38K/ΔC17 was exploited for many of these experiments as indicated . To measure the number of cycles of DNA synthesis , we incubated pol V Mut E38K/ΔC17 with either ATPγS ( Fig 1B , 1D and 1F ) , or ATP ( Fig 1C , 1E and 1G ) , using a 5-fold molar excess of p/t DNA over polymerase . Pol V Mut E38K/ΔC17 is limited to just one round of replication ( 20% p/t DNA elongation ) with either ATPγS or ATP at 37°C ( Fig 1F and 1G ) . Although pol V E38K/ΔC17 Mut is deactivated following one round of synthesis at 37°C , the enzyme is clearly not dead but is instead temporarily inactive , since substantial reactivation of the polymerase occurs , with ~ 75 to 85% p/t DNA subsequently extended following exposure to RecA* ( Fig 1F and 1G ) . In contrast , pol V Mut E38K/ΔC17 catalyzes multiple rounds of synthesis at 30°C ( Fig 1F and 1G ) . Pol V Mut E38K/ΔC17 performs four rounds of DNA synthesis at 30°C with ATPγS ( Fig 1F ) , with the extent of polymerase cycling remaining at 4 when the ratio of p/t DNA to polymerase is increased from 5-fold to 10-fold molar excess ( S1 Fig ) . When ATP is used instead of ATPγS , Pol V Mut E38K/ΔC17 performs 3 rather than 4 rounds of DNA synthesis at 30°C ( Fig 1G ) . The dynamic deactivation of pol V Mut is rapidly reversed by the addition of transRecA* at 37 and 30°C with either ATPγS or ATP . Reactivation is caused by a continual replenishment of activated pol V Mut [11 , 19] irrespective of the extent of cycling . There is a clear distinction to be made between polymerase deactivation and inactivation . Pol V Mut E38K/ΔC17 and pol V Mut wt are inactivated following incubation at 45°C for 15 min and cannot be reactivated by the addition of RecA* ( S2B and S2E Fig ) . In contrast , pol V ( S2C Fig ) and RecA ( S2D and S2F Fig ) when incubated alone at 45°C , remain functionally active since they retain the ability to assemble into an activated form of pol V Mut . Synthesis with ATPγS appears to be processive at 30°C , with most of the elongation gel bands extending either to the end of the template ( 12 nt ) or terminating one base prior to the end ( Fig 1D ) . Synthesis with ATPγS at 37°C is more limited ( Fig 1B ) , with only about 20% of the primers being extended under conditions where the p/t DNA substrate is in 5-fold excess to the enzyme . Thus , only one synthetic cycle is occurring . Product lengths vary , and the observed processive deoxynucleotide additions are taking place on the same p/t DNA molecule . For the same reason , i . e . , absence of cycling , processive synthesis appears to be occurring with ATP at 37°C , even though extension is limited to < 5 nt ( Fig 1C ) . However , with ATP at 30°C synthesis is essentially distributive as shown by the presence of a decreasing gradient of small to large primer elongation bands , along with far fewer primers that are extended to the end of the template ( Fig 1E ) . Pol V Mut wt performs three rounds of DNA synthesis at 30°C ( Figs 2B and 1C ) but is restricted to a single round at 37°C ( Fig 2A and 2C ) , as observed for pol V Mut E38K/ΔC17 ( Fig 1F ) . In a similar manner , synthesis appears to be processive in the presence of ATPγS , and RecA* reactivation occurs at both temperatures ( Fig 2A and 2B ) . However , a definitive difference in properties of the two forms of pol V Mut is that pol V Mut wt cannot synthesize DNA in the presence of ATP ( Fig 2D ) in vitro . Concomitantly , binding to p/t DNA is weak in the presence of ATP ( Fig 3A , ~20% increase in rotational anisotropy ) . Pol V Mut wt activity is robust with ATPγS ( Fig 2A and 2B ) , corresponding to a much stronger p/t DNA binding ( Fig 3A , ~2 . 5-fold increase in rotational anisotropy ) . Pol V Mut E38K/ΔC17 , binds much more strongly to DNA with either ATP or ATPγS ( Fig 3A , ~2 . 8-fold and ~3-fold increase in rotational anisotropy , respectively ) , and performs robust DNA synthesis ( Fig 1 ) . Neither form of pol V Mut binds to DNA in the absence of ATP/ATPγS ( Fig 3A ) , thus precluding DNA synthesis ( Fig 3B ) . The halt in DNA synthesis after a prescribed number of cycles of DNA synthesis defines dynamic deactivation . The dynamic deactivation profiles ( Fig 1F and 1G and Fig 2C ) can be analyzed using two parameters , a pol V Mut intrinsic DNA synthesis rate constant ( k ) , and a deactivation rate ( D ) ( S3 Fig ) . The synthesis rate constants for pol V Mut E38K/ΔC17 are k ~ 0 . 008 min-1 for ATPγS and ATP at 37°C , and about 1 . 5-fold faster at 30°C . Pol V Mut wt with ATPγS , behaves in a similar manner , k ~ 0 . 004 min-1 at 37°C and about 1 . 5-fold faster at 30°C . Using pol V Mut E38K/ΔC17-ATPγS , we also determined whether or not the presence of the β-sliding processivity clamp altered the dynamic deactivation profile , and found that it had no measurable effect ( S4 Fig ) . The deactivation rate ( D ) is the key parameter that characterizes the conformational regulation mechanism . At 37°C , the value of D is similar for pol V Mut E38K/ΔC17 ( D = 0 . 028 min-1 ) , pol V Mut wt ( D = 0 . 015 min-1 ) with ATPγS ( S3A and S3E Fig ) , and also for pol V Mut E38K/ΔC17 with ATP ( D = 0 . 026 min-1 ) ( S2C Fig ) . At 30°C , the deactivation rate constant ( D ) is reduced by about 3 to 5-fold compared to 37°C for pol V Mut E38K/ΔC17 ( D = 0 . 01 min-1 ) with ATPγS and ATP ( S3B and S3D Fig ) , and pol V Mut wt ( D = 0 . 003 ) with ATPγS ( S3F Fig ) . When assembled in an activated state , pol V Mut E38K/ΔC17 undergoes rapid dynamic deactivation during DNA synthesis at 37 and 30°C in the presence of either ATPγS or ATP ( Fig 1 ) . Deactivation of Pol V Mut in the absence of DNA synthesis is defined as static deactivation . Pol V Mut E38K/ΔC17 was incubated at either 37 or 30°C for varying lengths of time either alone , or + ATPγS , or ATPγS + DNA ( Fig 4A ) . Deoxynucleoside triphosphates were omitted in these incubations , precluding DNA synthesis . At each time point , remaining enzyme activity was assessed at 37°C by the addition of any components needed for DNA synthesis , ATPγS , p/t DNA , and dNTP substrates ( Fig 4A ) . Pol V Mut E38K/ΔC17 deactivates rapidly ( 0 . 015 min-1 ) at 37°C ( S5A Fig ) , losing ~ 80% of its activity in about 1 h , while retaining a residual level of activity at 4 h ( Fig 4B and 4C ) . In contrast , pol V Mut E38K/ΔC17 is considerably more stable at 30°C , losing less than half of its activity ( 0 . 008 min-1 ) ( S5A Fig ) , with about 60% activity retained after a 1 h incubation , decreasing to ~ 50% after 4 h ( Fig 4D and 4E ) . Pol V Mut E38K/ΔC17 is partially stabilized in its activated state at 37°C in the presence of ATPγS , whereas deactivation occurs more rapidly in the presence of ATPγS + p/t DNA ( Fig 4B and 4C ) . At 30°C , pol V Mut is fully stabilized by ATPγS , while a moderate rate of deactivation occurs with ATPγS + p/t DNA ( Fig 4D and 4E ) . Similarities and differences are observed in static deactivation rates in the presence of ATP ± p/t DNA ( S6A and S6B Fig ) compared to ATPγS ± p/t DNA ( Fig 4 ) . Stabilization of pol V Mut E38K/ΔC17 occurs with ATP , with greater stabilization at 30 vs 37°C ( S6A vs S6B Fig ) , which is qualitatively similar to ATPγS ( Fig 4 ) . However , the degree of stabilization is far weaker with ATP and the addition of DNA further destabilizes the ATPγS/ATP-stabilized deactivation rates at both temperatures ( Fig 4 ) . As observed for dynamic deactivation ( Fig 1 ) , substantial RecA* reactivation of pol V Mut E38K/ΔC17 occurs for all static deactivation conditions , pol V Mut E38K/ΔC17 , pol V Mut E38K/ΔC17 ± ATPγS/ATP ± p/t DNA ( Fig 4 , RecA* ) . The static deactivation properties of pol V Mut wt ( Fig 5 ) are similar to pol V Mut E38K/ΔC17 ( Fig 4 ) . Pol V Mut wt deactivates somewhat more slowly ( 0 . 017 min-1 ) at 30°C compared to 37°C ( 0 . 021 min-1 ) ( Fig 5 and S5B Fig ) . In contrast to pol V Mut E38K/ΔC17 ( Fig 4 ) , ATP/ATPγS weakly stabilizes pol V Mut wt ( Fig 5 , S6C Fig ) . Perhaps pol V Mut wt has more conformational flexibility than pol V Mut E38K/ΔC17 , thus allowing more rapid deactivation , and more efficient reactivation . To determine if ATPγS/ATP triggers conformational switching within pol V Mut , we measured UmuC-RecA crosslinking for each static deactivation condition ( see methods ) . We assembled pol V Mut with a crosslinkable variant of RecA wt or RecA E38K/ΔC17 , containing a p-benzoyl-l-phenylalanine ( pBpa ) residue at aa 113 , and performed UV crosslinking ± ATPγS/ATP ± p/t DNA . Two conformational changes were observed , the first upon addition of ATPγS or ATP , the second upon addition of ATPγS and p/t DNA ( Fig 6 ) . The UmuC subunit of pol V Mut E38K/ΔC17 shows no crosslinking with RecA E38K/ΔC17 ( Fig 6A and S7A Fig ) , and weak crosslinking with RecA wt of pol V Mut wt ( Fig 6B and S7B Fig ) . In contrast , strong UmuC-RecA crosslinking bands are observed when ATPγS or ATP was bound to either form of pol V Mut , which places N113 residue of RecA in close proximity with UmuC ( Fig 6 , S7 Fig ) . Therefore , binding of ATPγS/ATP induces a conformational change that reorients RecA relative to UmuC . After binding to ATPγS , binding to p/t DNA induces a second conformational change that eliminates UmuC-RecA crosslinking for both pol V Mut E38K/ΔC17 or pol V Mut wt ( Fig 6 , S7 Fig ) . When ATP is used , this second conformational change triggered by p/t DNA is not evident for pol V Mut wt , at least as measured by UmuC-RecA crosslinking ( Fig 6B ) . This result corresponds to the weak binding of pol V Mut wt to p/t DNA in the presence of ATP measured by the small increase in rotational anisotropy ( from about 0 . 045 to 0 . 05 , Fig 3A ) . With ATP and pol V Mut E38K/ΔC17 , the second conformation change is represented by a weak , yet detectable reduction in UmuC-RecA crosslinking upon p/t DNA addition ( Fig 6A and S7A Fig ) . The range of reduction in crosslinking of pol V Mut E38K/ΔC17 in the presence of ATP+DNA compared to ATP alone is 6–11% ( Fig 6A and S7A Fig ) , which was determined by repeating this experiment four times . Although , we would have expected to observe a stronger reduction in crosslinking , since ATP supports pol V Mut E38K/ΔC17 p/t DNA binding ( Fig 3A ) and DNA synthesis ( Fig 1B and Fig 3B ) , the binding and synthesis are moderately stronger in the presence of ATPγS ( Fig 3 ) . Pol V Mut E38K/ΔC17 and pol V Mut wt are not active with ADP [13] , AMP or AMPPNP ( S8A and S8C Fig ) . The addition of ADP to either form of pol V Mut induces UmuC-RecA crosslinking . However , since pol V Mut E38K/ΔC17 and pol V Mut wt are not active with ADP there is no change in crosslinking when p/t DNA is also added ( S8B and S8D Fig ) . No UmuC-RecA crosslinking is observed when AMP and AMPPNP is added to either form of pol V Mut ( S8B and S8D Fig ) . We have introduced another crosslinkable residue in pol V Mut at position F21 of RecA E38K/ΔC17 ( S9A and S9B Fig ) and RecA wt ( S9C and S9D Fig ) that forms a covalent linkage with UmuD′ ( S9A and S9C Fig ) , but not with UmuC ( S9B and S9D Fig ) . In this case , there are no discernible conformational shifts , since the antibody band intensities remain the same ±ATPγS , and in the presence of DNA ( S9A and S9C Fig ) . In contrast , the N113 residue of RecA does not crosslink with UmuD′ ( S8E Fig ) [12] . Rotational anisotropy measurements using fluorescently labeled p/t DNA have established that ATP or ATPγS must be present in a complex with pol V Mut to allow the polymerase to bind to DNA ( Fig 3 ) [13] . To investigate the molecular basis for the substantial differences in the effects of ATP/ATPγS , and wt and mutant RecA’s , on pol V Mut binding and catalysis ( Fig 1 , Fig 2 and Fig 3 ) , we measured time-resolved polymerase-p/t DNA binding rates and lifetimes . We have used smFRET to visualize the dynamics of pol V Mut ± ATP/ATPγS , diffusing free in aqueous solution and subsequently binding to and releasing from DNA . An Alexa Fluor 555 ( AF555 ) fluorescent donor , located internally on the DNA template strand , is annealed to a primer tethered to a glass coverslip , and an Alexa Fluor 647 ( AF647 ) fluorescent acceptor is placed on the RecA subunit of pol V Mut ( Fig 7A , sketch; see Methods ) . Donor-acceptor counter-correlated fluorescent signals ( Fig 7B; upper panel ) are detected when pol V Mut binds to DNA and are used to calculate FRET efficiencies ( Fig 7B , lower panel and Fig 7C , histogram ) . The distribution of FRET efficiencies ( 0 . 2–1 . 0 , peak at 0 . 7 , Fig 7C ) corresponds to a range of binding distances of 49 to 64 Å between AF555 on p/t DNA and AF647 on the RecA subunit of pol V Mut E38K/ΔC17; 51 Å is the Ro value of the FRET pair . This relatively broad distribution may reflect internal fluctuations in the location of RecA within the mutasome . To visualize the dynamics of pol V Mut-p/t DNA binding and release , three incubations were performed for pol V Mut containing either RecA E38K/ΔC17 or RecA wt subunits: Incubation 1 ) pol V Mut ( UmuD′2C-RecA ) added at t = 30 s; Incubation 2 ) pol V Mut ( UmuD′2C-RecA ) added at t = 30 s followed by addition of ATP or ATPγS at t = 60 s; Incubation 3 ) activated pol V Mut ( UmuD′2C-RecA-ATP/ATPγS ) added at t = 30 s . For incubation 1 , pol V Mut E38K/ΔC17 shows no detectable binding events in the absence of nucleotide cofactors , for the entire 3 min duration of data acquisition , a time period after which onset of pol V Mut photobleaching is observed ( Fig 7D ) . The locations of pol V Mut E38K/ΔC17 on the coverslip surface ( red dots ) do not colocalize with tethered p/t DNA ( Fig 7D and S1 Movie ) . For incubation 2 with pol V Mut E38K/ΔC17 , no pol V Mut binding is initially detected between t = 30s and t = 60s ( S2 Movie ) , as in incubation 1 . However , clear binding events , characterized by the colocalization of pol V Mut with p/t DNA and by smFRET signals , are observed 30 seconds after the addition of ATPγS at t = 60 s ( yellow dots , Fig 7E and S2 Movie ) . These data demonstrate that pol V Mut E38K/ΔC17 binds to p/t DNA only after binding ATP or ATPγS . The 30 s delay time for ATPγS pol V Mut E38K/ΔC17 binding to p/t DNA is attributable to a combination of the diffusion time required for ATPγS to bind to and activate pol V Mut E38K/ΔC17 , and for activated pol V Mut E38K/ΔC17 to then bind to p/t DNA . For incubation 3 , we preassembled activated pol V Mut E38K/ΔC17 containing bound ATPγS , in 1:1 stoichiometry , and detect numerous binding events ( yellow dots , Fig 7F and S3 Movie ) determined by the rapid appearance of smFRET signals at t = 5 s after enzyme addition at t = 30s ( Fig 7B , Fig 7F and S3 Movie ) . Therefore , a necessary and sufficient condition for mutasome binding to p/t DNA is that ATP/ATPγS must be present as part of an intact UmuD′2C-RecA-ATP/ATPγS complex in an activated conformational state . To establish that pol V Mut can bind to p/t DNA only when activated , we measured binding using smFRET for the deactivated form of pol V Mut E38K/ΔC17 ( Fig 8 ) . Pol V Mut was completely deactivated by incubation at 37°C for 4 h in the absence of ATPγS and DNA ( Fig 4A ) . Pol V Mut E38K/ΔC17 is active prior to incubation ( Fig 8A , lane 1 , t = 0 ) . Incubation of pol V Mut E38K/ΔC17 for 4 h results in enzyme deactivation ( Fig 8A , lane 2 ) . The deactivated enzyme fails to FRET with p/t DNA even after addition of ATPγS , indicative of no binding ( Fig 8B ) . When incubated for 4 h in the presence of ATPγS , pol V Mut E38K/ΔC17 retains substantial activity and its ability to bind DNA as observed by FRET ( Fig 8A , lane 4 and Fig 8C ) . However , when incubated for 4 h in the presence of ATPγS + p/t DNA ( in the absence of dNTPs ) , pol V Mut E38K/ΔC17 exhibits strong deactivation and minimal binding to p/t DNA ( Fig 8A , lane 6 and Fig 8D ) . All deactivated forms of the polymerase can be reactivated in the presence of RecA* ( Fig 8A , lanes 3 , 5 , and 7 ) . The important conclusion is that statically deactivated pol V Mut cannot bind to p/t DNA . From a statistical analysis of the smFRET data with activated pol V Mut , we extracted the characteristic dwell time of binding ( τbound ) of pol V Mut E38K/ΔC17 to DNA , with ATPγS or ATP ( Fig 9 ) . τbound is directly related to the unimolecular dissociation rate constant kdissoc = 1/τbound . Dissociation kinetics of Pol V Mut follow Poisson statistics with a typical exponential dwell time distribution . From a sample of size N = 298 pol V Mut E38K/ΔC17 with ATPγS , a maximum likelihood analysis [20] of the bound times yields τbound = 6 . 1 ± 0 . 7 s ( Fig 9A ) . With ATP , the pol V Mut E38K/ΔC17 residence times on DNA are shifted to shorter durations ( Fig 9B ) with a maximum likelihood estimate giving τbound = 3 . 0 ± 0 . 4 s ( N = 258 ) . Thus , pol V Mut E38K/ΔC17 stays bound to DNA about 2 times longer with ATPγS ( ~6s ) compared to ATP ( ~3s ) . Although this two-fold difference may seem relatively insignificant , it is important to note that the residence times follow an exponential decay . As seen in Fig 9A and Fig 9B , about 25% of the binding events persist for at least 10 s with ATPγS compared to only 1% for ATP . The longer binding times with ATPγS are reflected by increased processivities at 37°C and 30°C compared to ATP ( Fig 1A and 1B ) . The DNA synthesis rates are about 2-fold higher for ATPγS compared to ATP , and the rates are ~ 2-fold higher at 30°C compared to 37°C ( Fig 1A and 1B and S2A and S2B Fig ) . With ATPγS , the dwell time of pol V Mut wt on DNA is comparable to pol V Mut E38K/ΔC17 ( 5 . 6 s versus 6 . 1s , Fig 9C ) . In contrast , with ATP , pol V Mut wt shows only a small extent of steady-state binding to DNA ( Fig 3A ) , and there is no detectable DNA synthesis ( Fig 2D and Fig 3B ) [13] . The smFRET data show the presence of relatively infrequent binding events with a mean dwell time ~ < 2 s ( Fig 9D ) . A large reduction in the numbers of binding events per unit time and shorter dwell times may account for the absence of detectible DNA synthesis with ATP ( Fig 2D and Fig 3B ) . The affinity constant ( KD ) of pol V Mut towards DNA can be estimated from the ratio of its dissociation rate constant kdissoc and its association rate constant kassoc ( KD = kdissoc/kassoc ) where kassoc depends on the characteristic time during which DNA is unbound ( τunbound ) and the concentration of Pol V Mut ( kassoc = 1/ ( τunbound × [Pol V mut] ) . Unlike τbound , however , τunbound is more difficult to establish because unbound events do not register any FRET signature . We have used a distribution of the second binding times , τrebind , to try to estimate the relative values of τunbound in ATPγS versus ATP . Analysis of the smFRET data yields τrebind = 17 ± 2 s for 2 nM pol V Mut E38K/ΔC17 in ATPγS ( N = 218 ) , and 13 ± 2 s for 5 nM pol V Mut E38K/ΔC17 in ATP ( N = 122 ) . Together , these FRET statistics suggests that pol V Mut E38K/ΔC17 binding to DNA is approximately 3 . 8 times stronger in the presence of ATPγS compared to ATP .
When carrying out DNA synthesis at 37°C with ATPγS , pol V Mut E38K/ΔC17 ( Fig 1B ) and pol V Mut wt ( Fig 2A ) deactivate dynamically in about 1 . 5–2 h , which is sufficient to complete just one round of synthesis ( Fig 1F and Fig 2C ) . The dynamic deactivation rates are similar for the mutant and wild type polymerases , D ~ 0 . 02 and 0 . 015 min-1 for pol V Mut E38K/ΔC17 and pol V Mut wt , respectively ( S3 Fig ) . However , for synthesis at 30°C , both polymerase forms deactivate much more slowly , in about 3 h , which allows for multiple template cycling , 4 rounds of synthesis for pol V Mut E38K/ΔC17 ( D ~ 0 . 001 min-1 ) ( Fig 1F and S3B Fig ) and 3 rounds of synthesis for pol V Mut wt ( D ~ 0 . 003 min-1 ) ( Fig 2C , S3F Fig ) . How does internal regulation involving the roles of ATP and ATP hydrolysis contribute to the mechanisms governing pol V Mut DNA synthesis ? Clearly , pol V Mut’s intrinsic DNA-dependent ATPase [13] is not involved in on → off switching , since rapid dynamic deactivation occurs on a 1–3 h time scale with ATPγS ( Fig 1F and Fig 2C ) . However , ATP binding to pol V Mut appears to provide the initial on → off regulatory switch . The presence of bound ATP is an absolute requirement for pol V Mut activity . A previous bulk biochemical analysis using rotational anisotropy to detect pol V Mut binding to fluorescein-labeled p/t DNA showed that DNA binding to a 3′-primer OH required the presence of ATP in the polymerase ( ATP hydrolysis released pol V Mut from the DNA ) [13] . Although steady-state rotational anisotropy is an exceptionally useful way to measure protein-DNA binding interactions , it is somewhat of a blunt tool since it cannot detect transient binding events . The smFRET measurements visualize pol V Mut-p/t DNA binding events at sm resolution ( Figs 7–9 ) . The data demonstrate two key fundamental points . The presence of ATPγS or ATP in the pol V Mut complex is required for binding to DNA , and binding occurs only with Activated pol V Mut ( Fig 10 , State 2 ) , even though RecA is present in the mutasome complex in Inactive and Deactivated states ( Fig 10 , State 1 and State 2 , respectively ) . In this study , we have investigated RecA-ATP/ATPγS conformational regulatory mechanisms required for pol V Mut-p/t DNA binding and DNA synthesis ( activation ) , and cessation of synthesis ( deactivation ) . The regulation of activation would provide a cellular mechanism to call upon pol V Mut when needed to copy damaged DNA templates . The regulation of deactivation could provide a way to limit the extent of low fidelity DNA synthesis on undamaged DNA . It is clear that E . coli has gone to great lengths to limit the activity of error-prone pol V . This includes tight transcriptional control , inefficient post-translational processing of UmuD to UmuD’ , active proteolysis of UmuD , UmuD’ and UmuC , the requirement for physical protein-protein interactions ( RecA and β clamp ) , as well as intracellular spatial regulation ( all recently reviewed in [25] ) . The static and dynamic deactivation mechanisms reported here provide a way to limit the opportunity for pol V to introduce adventitious mutations on undamaged DNA templates following TLS ( dynamic deactivation ) or from binding to undamaged DNA when diffusing free in solution ( static deactivation ) . A longstanding question concerns the cellular location of a RecA* nucleoprotein filament required for assembly of pol V Mut . During TLS , RecA* could be located either in trans , away from a template lesion , or in cis , on the DNA template strand immediately downstream from a lesion . Pol V can be activated in trans [19] , as has generally been done in the experiments reported here . A cis arrangement has the inherent problem that a RecA* filament nucleated in a single strand DNA gap will quickly grow to encompass and thus block the adjacent 3′-primer terminus , preventing pol V binding . UmuD′2C can bind to an unblocked p/t DNA , but it cannot synthesize DNA until it is activated [13 , 15] . Once assembled in the cell , pol V Mut is presumably stable , i . e . , no longer subject to the relatively rapid proteolytic degradation of its individual UmuC and UmuD′ subunits [10] . In its activated state , containing ATP ( Fig 10 , State 2 ) , pol V Mut would be able to diffuse throughout the cell to catalyze TLS causing mutations at damaged template sites , or causing adventitious mutations at undamaged template site . We envision that the conformation regulation described in our manuscript is responsible for confining mutations to sites of TLS , and to minimize the possibility of causing mutations elsewhere . It would seemingly make sense for the polymerase to retain activity for 1 to 2 hours , giving it ample time to gain access to damaged template sites and then to catalyze TLS . We further envision that following deactivation , which can occur dynamically following TLS or statically in the absence of DNA synthesis ( Fig 10 , State 3 ) , that pol V remains in deactivated State 3 in the cell , unable to bind to DNA ( Fig 8 ) . Genetic data indicate that a trans-RecA* pathway functions under at least some situations in vivo . RecA E38K mutants are constitutively induced for SOS induction and mutagenesis in the absence of externally generated DNA template damage [17] . Our biochemical data show that pol V Mut assembled with RecA wt , RecA E38K and RecA E38K/ΔC17 exhibit similar conformational regulation behavior ( Figs 1 , 2 , 4 and 5 ) [11] . Recent live-cell imaging studies in a RecA E38K genetic background showed that pol V is present in high abundance in the cytosol in the absence of UV [4] . The ~ 100-fold increase in pol V-induced SOS mutations suggests that these are non-TLS mutations caused by error-prone pol V Mut E38K acting on undamaged DNA , perhaps by displacement of pol III core [26] . RecA E38K* might be formed on chromosomal DNA undergoing replication , or perhaps elsewhere in the cell . However , it is unlikely to assemble at a TLS site since few such sites would normally be available . Therefore , we suggest that for at least in the case of untargeted mutagenesis , pol V Mut E38K is likely to diffuse freely throughout the cell , as has been observed for all other DNA polymerases , and could also hold as well for damaged-induced TLS in a RecA wt genetic background . The dynamic deactivation kinetic profiles show closely similar mechanistic properties for pol V Mut E38K/ΔC17 and pol V Mut wt when activated with ATPγS . Pol V Mut E38K/ΔC17 behaves dynamically the same with ATPγS and ATP . However , that is not the case with pol V Mut wt , which shows no measurable DNA synthesis with ATP ( Fig 2B , Fig 3B ) . A likely explanation for the absence of detectible DNA synthesis with ATP is based on smFRET data showing a low frequency of pol V Mut-p/t DNA binding events combined with extremely short average binding times ( ~ 1 . 9 s ) ( Fig 8D ) . The presence of the β-sliding clamp is required to observe ATP-activated pol V Mut RecA wt DNA synthesis [13] . Pol V Mut contains an intrinsic DNA dependent ATPase [13] . A regulatory role for the ATPase remains an open question . Hydrolysis of a single ATP has been shown to release pol V Mut from a 3′-primer end [13] . Perhaps , the DNA-dependent ATPase acts to destabilize the pol V Mut-β clamp interaction to facilitate replication restart with pol III . Looking ahead , a more extensive smFRET study that models DNA repair gaps , particularly for the case of pol V Mut RecA wt could include RecA* nucleoprotein filament formation within lagging-strand gaps in the presence of RecFOR , and SSB . A role for SSB may be especially germane because of a long-standing observation of a UV-nonmutable SSB mutant [27] . Mutations occur with about equal frequencies on leading- and lagging DNA strands when TLS is targeted to UV damaged DNA template sites [28] . In contrast , the lion’s share of pol V Mut-induced SOS mutations ( ~ 85% ) occur on the lagging-strand during log-phase bacterial cell growth in the absence of external DNA damaging agents [29] . A high-resolution microscopy analysis of leading-strand continuous vs lagging-strand discontinuous gap synthesis could provide new mechanistic insights into how pol V-generated lesion-targeted vs untargeted mutations occur , which might finally reveal the mutagenic mechanism for the historically influential SOS constitutively induced RecA E38K mutant , which exhibits ~ 100-fold increase in pol V mutations in the absence of UV radiation [16] .
Pol V Mut was assembled following protocol by Erdem et al . 2014 [13] . Briefly , RecA nucleoprotein filament ( RecA* ) was assembled on Cyanogen Bromide Sepharose with covalently attached 45 nt ssDNA ( 3'-tip exposed ) . When RecA* formed , any access of unbound RecA and ATPγS was removed by extensive washes on small spin columns . His-tagged pol V ( purified according to [30] ) was incubated with RecA* and pol V Mut ( UmuC-UmuD'2-RecA ) complex was assembled . Pol V Mut is catalytically inactive and needs to be activated for DNA synthesis by adding ATP/ATPγS . The concentration of pol V Mut was determined by SDS-PAGE using purified pol V and RecA as protein concentration standards . The activity of pol V Mut E38K/ΔC17 or pol V Mut wt was measured at 37°C and 30°C on 12 nt oh ( overhang ) hairpin ( HP ) DNA ( 5′-CGA AAC AGG AAA GCA GTT AGC GCA TTC AGC TCA TAC TGC TGA ATG CGC TAA CTG C-3′ ) in the presence of ATP/ATPγS ( 500 μM ) and dNTP substrates ( mix of dTTP , dGTP , dCTP , 500 μM each ) in standard reaction buffer containing 20 mM Tris ( pH 7 . 5 ) , 8 mM MgCl2 , 5 mM DTT , 0 . 1 mM EDTA , 25 mM sodium glutamate and 4% ( v/v ) glycerol . To measure the number of DNA synthesis cycles , a 5-molar access of p/t DNA ( 1000 nM ) /pol V Mut ( 200 nM ) was used; the 12 nt overhang ( oh ) HP p/t DNA contained 100 nM 5'-32P labeled + 900 nM unlabeled DNA strands . The DNA synthesis reactions were initiated by adding pol V Mut with ATP or ATPγS , and the reactions were carried out for up to 300 min . The deactivated pol V Mut was reactivated by adding pre-assembled 200 nM RecA* . Aliquots were removed from reactions at given time points and the reactions were terminated with a stop solution containing 20 mM EDTA in 95% formamide . The p/t DNA product molecules were separated using 20% denaturing PAGE . Gel band intensities were measured by phosphorimaging with ImageQuant software , and the fraction of extended primer ( % PE ) was calculated by integrating the band intensities of extended hairpin DNA divided by the total integrated DNA band intensity . Each experiment was repeated 3 times , and the average % PE ( percent p/t DNA extended ) with the SD ( standard deviation ) was plotted for each reaction time point . The influence of the β sliding processivity clamp on pol V Mut E38K/ΔC17 deactivation was measured at 37°C on 32P –labeled 50 nt oh primer/template ( p/t ) DNA . The p/t DNA is: primer: 5'- ( Biotin ) dT CGA GGA TGG ATA TGG TTT AGT GGA TTT GGA TGA AGG TGA -3' , template: 5' A ( Biotin ) dTG ACA AGA CAA GAC AAG ACA AGA CAA GAC AAG ACA AGA CAA GAC AAG AAA TCA CCT TCA TCC AAA TCC ACT AAA CCA TA -3' . Streptavidin ( 400 nM ) was attached at both ends of the p/t DNA ( 25 nM ) to block the β clamp from sliding off . The γ clamp-loading complex ( 50 nM ) and ATPγS ( 1 mM ) was used to load β ( 200 nM ) onto the p/t DNA , and pol V Mut E38K/ΔC17 ( 100 nM ) activity was measured for 3 h in the presence of dNTP substrates containing dTTP , dGTP and dCTP ( 500 μM for each substrate ) . In parallel , pol V Mut E38K/ΔC17 activity was measured in the absence of β clamp . Experiments for pol V Mut E38K/ΔC17 + ATPγS ± β clamp were repeated 3 times and the average % of primer extension ( PE ) ± SD was plotted at each reaction time . The rate of pol V Mut primer extension follows first-order kinetics: dP/dt = k′P , where the pseudo-first-order rate constant k′ is proportional to the concentration of active enzyme k′ = k[E] . During the reaction , we assumed dynamic deactivation depletes the active enzyme concentration [E] by a Poisson process with deactivation rate D . Integrating the rate equation gives ln[1−P ( t ) ] = −kt+k ( e−Dt−Dt−1 ) /D , where P ( t ) is the fraction of primer extension as a function of time t . At short times , this equation reduces to ln[1−P ( t ) ] = −kt . Fitting the initial rate to a straight line therefore yields the intrinsic catalytic rate constant k , and then fitting the long-time data to the full equation produces the dynamic deactivation rate D ( S3 Fig ) . The parameters k and D ± SD were determined from an average of at least 3 independent measurements . Pol V Mut E38K/ΔC17 ( 300 nM ) , pol V Mut wt ( 300 nM ) , pol V ( 300 nM ) , RecA E38K/ΔC17 ( 9 μM ) and RecA wt ( 9 μM ) were incubated for 15 min at 37°C , 40°C , 42°C , and 45°C in standard reaction buffer ( see above ) followed by a measurement of DNA polymerase activity on 12 nt oh HP DNA ( 600 nM ) at 37°C for 1h . The activity of pol V Mut E38K/ΔC17 ( 300 nM ) and pol V Mut wt ( 300 nM ) was measured with ATPγS ( 500 μM ) and dNTP substrates ( dTTP , dGTP , dCTP , 500 μM each ) . Pol V activity was measured after transactivation with RecA* ( 300 nM ) . The effect of temperature on RecA monomers was measured by their ability to form RecA* nucleoprotein filaments on 30 nt long ssDNA ( 600 nM ) in the presence of ATPγS ( 500 uM ) followed by transactivation of pol V ( 300 nM ) . Pol V Mut E38K/ΔC17 ( 600 nM ) and pol V Mut wt ( 600 nM ) were incubated at 37°C or 30°C in standard reaction buffer ( see above ) , either alone or with 500 μM ATPγS/ATP , or with 500 μM ATPγS/ATP and 900 nM 12 nt oh HP DNA . Each static deactivation incubation was carried out for up to 240 min at either at 37°C or 30°C , and for each incubation time an aliquot of protein was removed and pol V Mut activity was measured at 37°C on 12 nt HP DNA ( final concentration in the reaction = 500 nM ) in the presence of ATPγS ( final concentration = 500 μM ) and dNTP substrates ( mix of dTTP , dGTP , dCTP , 500 μM each ) for 1 h . The DNA synthesis products were separated on a 20% denaturing polyacrylamide gel , and the bands intensities were calculated using ImageQuant . Static deactivation is expressed as the relative polymerase activity measured at each incubation time point divided by the polymerase activity measured at t = 0 . Each experiment was repeated 3 times and average relative activity along with the SD for each deactivation time point are shown in the graphs . An estimate of the initial static deactivation rate was obtained from the reduction in activity at 30 min ( S5 Fig ) , and the difference in static deactivation rates at 37°C and 30°C were used to determine a rough estimate of the difference in activation energies between pol V Mut E38K/ΔC17 and pol V Mut wt . The activation energy was extracted from an Arrhenius analysis of the initial rates of decay in enzyme activity at the two temperatures . To site specifically label RecA E38K/ΔC17 and RecA wt for smFRET experiments we substituted F21 amino acid with p-azido-L-phenyloalanine ( pAzpa ) . The RecA wtF21pAzpa and RecA E38K/ΔC17F21pAzpa was engineered , expressed , and purified as previously described [31] . To fluorescently label p-azido-phenylalanine modified RecA E38K/ΔC17 and RecA wt , 1 ml of 100 μM RecA wtF21pAzpa and RecA E38K/ΔC17F21pAzpa was incubated with Alexa Fluor 647 DIBO Alkyne ( AF647; ThermoFisher Scientific ) at a ratio of 1:5 in 20 mM Tris , 0 . 1 mM EDTA , 10% glycerol pH 7 . 5 buffer and rotated at 4°C overnight . Following incubation , dye and protein mixture were loaded onto a Ceramic Hydroxyapatite column ( Bio-Rad ) and the labeled RecA was eluted by running a phosphate gradient from 0 to 0 . 25 M potassium phosphate . Unbound dye eluted through the wash , AF647-labeled RecA eluted early during the potassium phosphate gradient and unlabeled RecA eluted last . Labeled RecA concentrations and labeling efficiency were determined via spectrophotometry at 280 nm and 650 nm respectively . Biotinylated 40 nt primer ( 5'- ( Biotin ) dT–CGA GGA TGG ATA TGG TTT AGT GGA TTT GGA TGA AGG TGA -3' ) and C6-amino-modified 54 nt long template ( 5′- TAG CAT GCG TCA GCT TCA CC AF555-T TCA TCC AAA TCC ACT AAA CCA TAT CCA TCC TCG -3′ ) were purchased from Integrated DNA Technologies . The ssDNA oligos were purified trough 20% denaturing polyacrylamide gel electrophoresis ( PAGE ) . To label C6-amino-modified template , 100 μl of 100 μM template was incubated with Alexa Fluor 555 NHS Ester ( AF555; ThermoFisher ) and rotated at 4°C overnight in 0 . 1 M sodium bicarbonate pH 8 . 3 . AF555-labeled template was purified trough PAGE . Prior to smFRET experiments , biotinylated DNA primer and AF555 template were annealed at a ratio of 1:1 . 2 by heating to 95°C and allowed to slowly cool to 16°C by decreasing 1°C per minute in a thermocycler . For single-molecule FRET experiments , high precision microscope glass coverslips ( Marienfeld , #1 . 5 , Ø25 mm ) were first cleaned by sonication in ddH20 for 1 min followed by 100 mM KOH for 20 min . Slides were then cleaned in a Piranha solution of 1:3 hydrogen peroxide to sulfuric acid for 5 min followed by further sonication in ddH20 for 10 min to clean off Piranha residue . Coverslips were then incubated in a solution containing 3 ml of 3-Aminopropyltriethoxysilane ( Sigma-Aldrich ) , 5 ml acetic acid , and 100 ml methanol for 20 min with 1-min sonication in the middle of the incubation . The coverslips were rinsed with methanol and dried prior to applying a solution containing 25 mM of a 5:1 polyethylene glycol ( PEG ) and biotin-PEG-Succinimidyl Valerate ( Laysan Bio ) in 0 . 1 M sodium bicarbonate pH 8 . 3 buffer and incubated overnight for surface functionalization . To reduce background fluorescence [32] , a second round of surface pegylation was done for 2 hr using 25 mM of 4-Methyl-PEG-NHS-Ester ( Thermo-Fisher ) , prior to washing with ddH20 and incubation with streptavidin ( Sigma ) for 10 min in buffer containing 20 mM Tris-HCl , 50 mM NaCl pH 7 . 5 . Coverslips were then washed and coated with 20 pM AF555-labeled primer template DNA for 5 min . Unbound p/t DNA was washed off in imaging buffer ( 1X reaction buffer + 50 ug/ml BSA , 2mM Trolox , 10 mM PCA , and 100 nM PCD [33] ) prior to the addition of pol V Mut labeled with Alexa 647 at F21 position of RecA wt or RecA E38K/ΔC17 . 1X reaction buffer contains 20mM Tris-HCl pH 7 . 5 , 25 mM sodium glutamate , 8 mM MgCl2 , 5 mM DTT , 4% glycerol , 0 . 1 mM EDTA . To assess pol V Mut binding on AF555- labeled p/t DNA , pol V Mut alone ( incubation 1 see Results and Fig 7D , right panel ) or ATPγS activated pol V Mut ( incubation 3 , see Results and Fig 7D , right panel ) were added at t = 30 s after start of image acquisition . For incubation 2 ( pol V Mut + ATPγS; see results and Fig 7D middle panel ) , pol V Mut was added at t = 30 s and 500 uM ATPγS was added at t = 60 s after start of image acquisition to assess the impact of ATPγS on pol V Mut DNA binding . Each experiment was carried out for 3 min , after which donor bleaching is observed . smFRET with deactivated pol V Mut was done as follow: 1 μM pol V Mut was incubated at 37°C in 1x reaction buffer +/- ATPγS and +/- DNA for 4 h prior to adding onto AF555-labeled p/t DNA covered slides . Pol V Mut was added in the presence of 500 μM ATPγS at t = 30 s after the start of image acquisition . Fluorescence imaging was performed on an inverted Nikon Eclipse Ti-E microscope equipped with total internal reflection optics , 561 nm and 647 nm fiber-coupled excitation lasers ( Agilent ) , a x100 1 . 49 NA objective , a two-camera imaging splitter ( Andor ) and two iXon EMCCD cameras ( Andor ) . A multi-band pass ZET405/488/561/647x excitation filter ( Chroma ) , a quad-band ZT405/488/561/647 dichroic mirror ( Chroma ) , an emission splitting FF-640-FDi01 dichroic mirror ( Semrock ) and two emission filters at 600/50 nm ( Chroma ) and 700/75 nm ( Chroma ) for AF555 p/t DNA and pol V Mut RecA wtAF647 respectively , were used . smFRET images were acquired using 561 nm laser excitation at an image acquisition rate of 100 ms/frame in each channel for ATP and 300 ms/frame for ATPγS . Channel alignment was performed using a few 40 nm TransFluoSphere beads ( 488/685 nm , Life Technologies ) as fiducials markers . The 8x8 pixel region of interests were drawn around individual p/t DNA molecules in overlaid two-color images , and signal intensity from both channels were extracted for each acquisition frame . Automated detection of countercorrelated smFRET signals on the slides were performed using in-house AI software . FRET efficiency was calculated using the formula: E=IA/ ( ID+IA ) where ID , is the signal intensity from an individual AF555-labeled p/t DNA molecule and IA is the signal intensity of a bound AF647-Pol V Mut in each frame . The smFRET efficiency histograms of Fig 9 were produced by binning the smFRET efficiency values for multiple individual smFRET pairs by 1 second . τ-on and τ-rebind were determined by measuring the residence time and time between two binding events , respectively , plotting them as a histogram and fitting the histogram with a single exponential decay function . For UV crosslinking experiment , RecA wt and RecA E38K/ΔC17 was modified with pBpa at the N113 and F21 positions and purified as previously described [12] . Pol V Mut was assembled with RecA E38K/ΔC17N113pBpA , RecA wtN113pBpa , RecA E38K/ΔC17F21pBpA and RecA wtF21pBpA were incubated for 30 min at 37°C either alone or with ATP/ATPγS ( 500μM ) in the presence of absence of 12 nt oh HP DNA . Following pre-incubation , the reactions were exposed to UV ( 365 nm ) for 60 min , with gentle mixing at 15 min intervals . Crosslinked products were boiled in protein loading dye and separated using 12% SDS-PAGE . UmuC-RecA crosslinked bands were detected using affinity-purified rabbit anti-UmuC ( 1:200 dilution ) or UmuD/UmuD′ ( 1:1000 dilution ) , using a standard Western Blot protocol . Each crosslinking experiment was repeated 3–4 times , and the % of crosslinked UmuC-RecA was calculated by measuring Western blot intensities with ImageQuant and ImageJ software .
|
Escherichia coli upregulates more than 40 genes as part of the DNA damage-induced SOS regulon , many of which are involved in DNA repair and cell division . However , three DNA polymerases , pols V , II , and IV , are also induced to rescue replication forks blocked at persisting template lesions . Pol V ( UmuD′2C ) , encoded by the UV mutagenesis genes ( umuDC ) , is primarily responsible for the increase in UV-induced chromosomal mutagenesis . However , pol V is catalytically inert . Interaction with a RecA nucleoprotein filament ( RecA* ) and ATP is required to convert pol V to an activated “mutasome” complex , pol V Mut = UmuD′2C-RecA-ATP . Here , we show that pol V Mut deactivates dynamically during DNA synthesis , and statically in the absence of synthesis . Activated and deactivated states are governed by a conformational switch that repositions RecA relative to UmuC . Switching rates are more rapid at 37 than at 30°C , and do not require ATP hydrolysis . ATP ( ATPγS ) binding plays two required regulatory roles: 1 ) it allows binding of pol V Mut to primer-template DNA; 2 ) it triggers the RecA-UmuC conformational switch that activates pol V Mut .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"chemical",
"bonding",
"nucleic",
"acid",
"synthesis",
"condensed",
"matter",
"physics",
"dna-binding",
"proteins",
"anisotropy",
"fluorophotometry",
"dna",
"damage",
"nucleoproteins",
"polymerases",
"materials",
"science",
"dna",
"dna",
"synthesis",
"chemical",
"synthesis",
"physical",
"chemistry",
"research",
"and",
"analysis",
"methods",
"fluorescence",
"resonance",
"energy",
"transfer",
"proteins",
"chemistry",
"cross-linking",
"biosynthetic",
"techniques",
"spectrophotometry",
"physics",
"biochemistry",
"dna",
"polymerase",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"material",
"properties",
"spectrum",
"analysis",
"techniques"
] |
2019
|
Conformational regulation of Escherichia coli DNA polymerase V by RecA and ATP
|
Despite the wide use of Caenorhabditis elegans as a model organism , the first virus naturally infecting this organism was not discovered until six years ago . The Orsay virus and its related nematode viruses have a positive-sense RNA genome , encoding three proteins: CP , RdRP , and a novel δ protein that shares no homology with any other proteins . δ can be expressed either as a free δ or a CP-δ fusion protein by ribosomal frameshift , but the structure and function of both δ and CP-δ remain unknown . Using a combination of electron microscopy , X-ray crystallography , computational and biophysical analyses , here we show that the Orsay δ protein forms a ~420-Å long , pentameric fiber with an N-terminal α-helical bundle , a β-stranded filament in the middle , and a C-terminal head domain . The pentameric nature of the δ fiber has been independently confirmed by both mass spectrometry and analytical ultracentrifugation . Recombinant Orsay capsid containing CP-δ shows protruding long fibers with globular heads at the distal end . Mutant viruses with disrupted CP-δ fibers were generated by organism-based reverse genetics . These viruses were found to be either non-viable or with poor infectivity according to phenotypic and qRT-PCR analyses . Furthermore , addition of purified δ proteins to worm culture greatly reduced Orsay infectivity in a sequence-specific manner . Based on the structure resemblance between the Orsay CP-δ fiber and the fibers from reovirus and adenovirus , we propose that CP-δ functions as a cell attachment protein to mediate Orsay entry into worm intestine cells .
For the past four decades , the nematode Caenorhabditis elegans ( C . elegans ) has been used as an important model organism for studying biological processes such as development , metabolism , aging , cell cycle and gene regulation [1] . However , it was not until the year 2011 that Orsay , the first virus that naturally infects C . elegans in the wild , was discovered [2] . Two other nematode viruses , Santeuil and Le Blanc , which both infect C . briggsae , have also been identified [2 , 3] . Infections by these newly identified viruses cause abnormal intestinal morphologies without obvious effects on longevity or brood size [2 , 4] . The Orsay virus has a bipartite , positive sense RNA genome of ~6 . 3 kb with three open reading frames ( ORFs ) , including the putative RNA-dependent RNA polymerase ( RdRP ) , the viral capsid protein ( CP ) , and a nonstructural protein δ [2] . A plasmid-based reverse genetics system directly using the worm host has been developed for the Orsay virus [5] . Considering the simplicity of Orsay and the many advantages of C . elegans as a widely used model organism , Orsay-C . elegans serves as an excellent model for dissecting important mechanisms related to virus replication , virus-host interaction and host innate antiviral responses . Phylogenetic analyses indicated that these nematode viruses are related to nodaviruses that infect primarily insects and fish [6] . Further molecular characterization revealed several fundamental differences compared to nodaviruses . For instance , translation of the Orsay CP utilizes an AUG-independent mechanism [7] . The nematode viruses also lack the subgenomic RNA3 found in nodaviruses that is used to express the nonstructural proteins B1/B2 . The nonstructural B1 and B2 proteins encoded by both alpha- and betanodaviruses [8–10] function as either a host RNA interference ( RNAi ) suppressor or an anti-necrotic death factor to modulate cell death during virus infection [9–15] . In particular , alphanodavirus B2 folds into an α-helical structure that dimerizes in solution and binds dsRNA in a sequence-independent manner [16–18] . However , a recent study showed that the Orsay nonstructural protein δ does not possess any RNAi suppression activity [19] . Recent studies of the Orsay infection indicate that the RNAi pathway plays an important role in the C . elegans antiviral response . While the wildtype N2 strain only supports limited infection , inactivation of genes in the RNAi pathway such as rde-1 [2] and drh-1 [20] can sensitize N2 worms to Orsay infection . In C . elegans , RNAi following Orsay infection is not systemic , and there is no transgenerational inheritance of Orsay virus-induced silencing [21] , different from the RNAi induced by exogenous dsRNA . In addition to RNAi , the ubiquitin-mediated defense also promotes Orsay resistance in N2 worms as revealed by a recent transcriptome study [22] . Recombinant Orsay CP is able to self-assemble into virus-like particles ( VLP ) in expression host . The crystal structure of an Orsay VLP has been determined , which displays a T = 3 icosahedral symmetry with 60 trimeric spikes [23] . Each Orsay CP can be divided into three linear regions , namely the N-terminal arm , the S domain forming the continuous capsid shell , and the P domain which forms trimeric surface protrusions . The Orsay CP is structurally distinct from alphanodaviruses ( e . g . Flock House virus ) [24–27] , but has a structural fold closely resembling that of the betanodavirus [28] . The function of the Orsay δ protein remains a mystery . It was recently demonstrated that during infection δ could be expressed as a CP-δ fusion protein , which is likely generated by ribosomal frameshifting at the end of the CP ORF [7] . The RNA structure motif mediating the ribosomal frameshifting is conserved in all three nematode-infecting viruses [7] . The CP-δ fusion protein was detected in both infected worms and purified virion samples [7] , but the expression of free δ has yet to be confirmed . Primary sequence alignment of the Orsay δ to those from Santeuil and Le Blanc gives 37% and 39% overall identity , respectively , suggesting an overall conserved structure and function . To better define the functional role of the Orsay δ and CP-δ during infection , here we report the structure of these two proteins using both X-ray crystallography and electron microscopy ( EM ) . Our results show that recombinant δ forms a fibrous molecule with a C-terminal globular domain . The N-terminal region of δ forms a pentameric α-helical bundle , but the rest of the protein is most likely β-stranded as suggested by sequence analysis and CD spectroscopy . The pentameric nature of the full-length δ fiber has been independently verified by both mass spectrometry and analytical ultracentrifugation . Coexpressing CP and CP-δ in insect cells produced Orsay VLPs with an enhanced amount of CP-δ compared to native virions . These VLPs were found to have multiple long fibers protruding from the capsid surface when observed under EM , indicating that the δ sequence adopts the same fibrous structure in both CP-δ and free δ . Considering its five-fold symmetry , the CP-δ fibers are expected to occupy five-fold vertices in the capsid . Furthermore , reverse genetics confirmed that the structural integrity of CP-δ is essential for Orsay infection , and competition assays showed that purified δ proteins , when added in trans , effectively inhibit Orsay infection . By analogy with other viral capsid-associated fiber proteins , the Orsay CP-δ likely functions in cell entry as a cell receptor binding protein .
To characterize the structure and functions of Orsay δ/CP-δ , the δ ORF was cloned into a pET vector for overexpression in E . coli . A 6×His SUMO tag was added at the N-terminus of the full-length δ to facilitate protein purification ( Fig 1A ) . Recombinant δ was expressed as a soluble protein and ~60% could be recovered in the cytoplasmic fraction . When subjected to a Superdex-200 gel filtration column , δ produced a sharp peak at ~58 ml , corresponding to an apparent molecular size of ~670 kDa , which is substantially higher than the calculated molecular mass of a monomer ( i . e . 38 . 4 kDa ) ( S1 Fig ) . Negative-staining transmission electron microscopy ( TEM ) was then performed to examine the molecular organization of δ . Surprisingly , TEM images showed that δ forms fibrous molecules of extended length ( Fig 1B ) , thus explaining its unusual elution profile from the size exclusion column . Based on the measurement of 136 molecules , the length of the δ fibers was determined to be ~419±52-Å ( Figs 1B and 2C ) . Close examination of individual δ molecules reveals several fine structural details: ( 1 ) at one end there is a globular head domain with a diameter of ~50-Å; ( 2 ) a second globule , which is slightly smaller , is found at roughly two-fifths of its length; ( 3 ) the thickness of the fiber is only ~20-Å in most parts; and ( 4 ) the other end of the fiber opposite to the large head domain often appears to be slightly enlarged in diameter ( Fig 1B ) . There was no particular bending point observed along the fiber , but the fibrous region near the globular head often exhibited more pronounced curvatures . Other than the ~420-Å δ fibers , longer or thicker filaments were not observed in EM , suggesting that further oligomerization of the δ fiber did not occur . Using bioinformatics tools PSIPRED [29] and Prof [30] , the secondary structure of δ was predicted based on its amino acid sequence ( S2 Fig ) . The N-terminal 65 residues were predicted to form an α-helical structure , whereas the rest of δ was predicted to be primarily β-stranded . The online server Motif Scan [31] also detected a valine-rich region in the middle of the Orsay δ sequence that is known to form regular arrays of β-strands in a number of exo/cytoskeleton-related proteins including putative insect cuticle proteins and a putative adhesin in Parabacteroides distasonis ( PDB ID: 3LJY ) . Circular dichroism ( CD ) was measured to experimentally determine the secondary structure content of δ ( Fig 1C ) . The CD spectra indeed showed a single trough near 220nm , which is highly characteristic for β-stranded structures , consistent with the results from secondary structure prediction . A quantitative analysis of the CD spectra was done using the program BeStSel [32] , which estimated that δ is composed of 3 . 3% α-helices and 42 . 6% β-strands ( Fig 1C ) . Considering its uniform length and non-repetitive structural features , each fiber molecule in the TEM images is likely a linear oligomer of δ that is arranged in either a parallel or an anti-parallel manner ( Fig 1B ) . To test this hypothesis , a number of δ truncates were designed with progressively larger amounts of N- and C-terminal sequences removed . While the N-terminally truncated mutants , including δ ( 167–346 ) , δ ( 195–346 ) , δ ( 220–346 ) , δ ( 241–346 ) and δ ( 256–346 ) , were poorly soluble , all C-terminally truncated mutants behaved similarly to the full-length δ during purification and were eluted as a major peak from the gel filtration column ( Fig 1A and S1 Fig ) . TEM further confirmed that both δ ( 1–162 ) and δ ( 1–241 ) formed fibrous molecules , however , their lengths were shorter than the full-length protein , with δ ( 1–162 ) and δ ( 1–241 ) fibers measured to be 215±30-Å ( n = 96 ) and 359±32-Å ( n = 35 ) long , respectively ( Fig 2B and 2C ) . Considering that the full-length δ protein is ~420-Å long , the length of these two mutants is roughly proportional to the size of their respective sequence . The large globular head domain was absent from both δ ( 1–162 ) and δ ( 1–241 ) , indicating that the head domain is formed by the C-terminal sequence . Overall , our results suggest that the δ protein fiber is a parallel oligomer with a C-terminal head domain , because these two δ protein truncates formed fibrous molecules resembling the left end of the δ protein fiber ( Fig 1D ) . It has been shown that the Orsay δ ORF could be translated as a CP-δ fusion protein by ribosomal frameshifting and the CP-δ was observed in infected cells as well as in purified virion samples [7] . In purified virions , the amount of CP-δ only counts for ~5% of the total CP [7] . While fusion proteins are frequently encoded by RNA viruses to regulate non-structural protein expression , fusion proteins as structural components are rarely observed in RNA viruses . To our knowledge , the only known exception is totiviruses ( e . g . yeast LA virus ) , a group of dsRNA viruses with non-segmented genomes , which express the viral RNA polymerases as gag-Pol fusion proteins that are incorporated into viral particles at low copy numbers [33] . To determine whether the δ sequence in CP-δ assumes a similar or different structure compared to the free δ , we subcloned the CP-δ sequence for recombinant protein expression . By inserting a single nucleotide “A” in front of the last nucleotide before the stop codon of the CP gene , we were able to position the δ ORF in the same coding frame as the preceding CP ORF ( Fig 1A ) . However , the Orsay CP-δ was found to be insoluble when expressed in either E . coli or insect cells . Because Orsay CP has a strong tendency to assemble into VLPs [23] , the most likely explanation for the solubility problem is that the different oligomerization behaviors of the CP component ( i . e . dimer , trimer , pentamer , and hexamer ) and the δ component ( i . e . pentamer–see below ) of the fusion protein resulted in an infinite molecular network and thus the formation of large aggregates . In an effort to resolve the solubility issue , we constructed a mini-fusion protein ( Fig 1A ) . The mini-fusion protein , also called CP-δ ( 215–485 ) , is comprised of the CP spike domain , a 29-aa linker , and the first 66 residues of the δ protein that were predicted to form α-helices ( Figs 1A and S2 ) . The CP spike domain forms trimeric surface protrusions [23] , but is not able to oligomerize any further in the absence of the rest of the CP polypeptide . When expressed in E . coli , the mini-fusion protein was soluble with an apparent MW of ~150 kDa based on gel filtration chromatogram , consistent with the theoretical calculation for a pentameric assembly ( Figs 2A and S3 ) . Considering that the mini-fusion protein is too short for EM observation , we next expressed and purified a midi-fusion protein , which contains the CP spike domain , the 29-aa linker , and the first 162 residues of δ ( Figs 1A and 2A ) . When the purified midi-fusion protein was subjected to negative-staining EM , fibrous molecules were again observed with a morphology similar to that of δ ( 1–162 ) ( Fig 2B ) . The length of the midi-fusion protein is about 293±29-Å ( n = 66 ) , which is slightly longer than the δ ( 1–162 ) fiber ( i . e . 215±30-Å ) ( Fig 2D ) . The presence of the CP spike domain , which is ~35-Å in height according to the structure of the VLP , likely counts for the length discrepancy between the midi-fusion protein and δ ( 1–162 ) . This finding led us to conclude that δ in the CP-δ fusion protein also adopts a fibrous structure similar to the free δ . It remains to be determined whether the 29-aa linker sequence adopts a particular conformation with a specific function or simply acts as a flexible linker . Considering the extended shape and flexible nature of the δ fiber , crystallization of the full-length δ protein would be highly challenging if not impossible . Therefore , several deletion mutants of δ , including δ ( 1–66 ) , δ ( 1–138 ) , δ ( 1–162 ) and δ ( 1–241 ) , were purified and subjected to crystallization . Among these four constructs , only δ ( 1–66 ) , which contains the first 66 residues of δ , produced crystals diffracted to better than 3-Å resolution . The structure of δ ( 1–66 ) was determined by single-wavelength anomalous dispersion ( SAD ) using SeMet-substituted crystals ( Table 1 ) . By subsequent molecular replacement and refinement against a native dataset , we were able to establish the final structure to 2 . 2-Å resolution ( Table 1 , Fig 3 ) . δ ( 1–66 ) assembles into a pentamer with the five subunits forming an α-helical bundle ( Fig 3A ) . This five-helical bundle is ~80-Å long and ~35-Å wide ( Fig 3A and 3B ) . Each δ ( 1–66 ) molecule folds into two α-helices that are connected by a 9-aa linker ( i . e . residues 21 to 29 ) ( Fig 3C ) . The longer α-helix , consisting of residues 30 to 63 , has a kink at around residue 40 ( Fig 3A ) . The helix after the kink contains three regular heptad repeats ( i . e . 40VSDKLDKISESLNTLVECVID60 , in which hydrophobic residues are highlighted in bold ) . Heptad repeats are frequently observed in coiled coil structures ( i . e . dimeric , trimeric , tetrameric and pentameric ) and they contain amino acid sequences arranged in the periodicity of ( a b c d e f g ) , with positions a and d predominantly occupied by hydrophobic residues . Hydrophobic side chains at the positions a and d make up a continuous hydrophobic surface on the α-helix so that multiple α-helices can wrap around each other to form a stable helical bundle [34 , 35] . Surface representation of the δ ( 1–66 ) pentamer shows a ~3 to 5-Å wide channel running through the entire molecule . A total of 18 residues are found to have their side chains pointing towards the interior of the channel , including Y6 , Y10 , T14 , L18 , A21 , Y23 , L25 , A26 , P28 , Y32 , W35 , F36 , V40 , L44 , I47 , L51 , L54 , and V58 ( Fig 3C and 3D ) . Therefore , the core of the entire helical bundle is mostly hydrophobic , except for a single location at T14 . In the electron density map , blobs of densities that are modeled as water molecules occupy the central channel ( Fig 3D ) . The large hydrophobic cavities at the center of the helical bundle may help to accommodate these water molecules [36] . It has been reported that the cartilage oligomeric matrix protein contains a five-stranded coiled-coil domain with a continuous axial pore with binding capacities for hydrophobic compounds , including prominent cell signaling molecules [37] . It remains to be found whether the Orsay N-terminal helical bundle has any specific ligand binding activity like the cartilage oligomeric matrix protein . The symmetry and the shape of the δ ( 1–66 ) structure also bear some resemblance to a class of pentameric viroporins , such as the small hydrophobic SH protein encoded by human respiratory syncytial virus [38] and the E protein found in SARS-CoV [39] . Ion channel activities , however , do not seem to apply to δ , as δ expression in E . coli is not associated with cytotoxicity and overexpressed δ is predominantly cytoplasmic instead of membrane-bound , which are markedly different from the reported behaviors of known viroporins [40] . To confirm that the full-length δ fiber is a pentamer as indicated by the δ ( 1–66 ) crystal structure , mass spectrometry was used to analyze the molecular weight of the molecule under non-denaturing conditions . Under even relatively energetic conditions a mass consistent with a homo-pentamer ( 192214 . 8 ±1 Da ) was the only major mass observed ( Fig 4A ) . This compares favorably to the average mass according to the sum of the relative atomic masses of 192210 Da and to the extremes of isotopic composition of 192187 . 4 and 192226 . 5 Da , as reported by IUPAC ( 2013 revision ) . It is important to note that neither the monomeric species , nor any other multimeric species , was observed even under the relatively energetic conditions shown ( and all conditions tried ) . This indicates that the homo-pentamer is exceptionally stable . Under sufficiently energetic conditions , the gas phase pentameric complex will dissociate upon collision with buffer gas to produce exclusively monomers . No intermediates were observed with minimal covalent bond cleavage ( S4A Fig ) . This resulted in a protein of observed mass 38441 . 1 Da which again compares well with the theoretical average mass of a monomer according to the sum of the relative atomic masses of 38442 . 2 Da and to the extremes of isotopic composition of 38437 . 5 to 38445 . 3 Da , as reported by IUPAC ( 2013 revision ) . Thus , the direct observation of pentamers and their dissociation to exclusive monomers upon sufficient activation energy confirm that the original complex is a homo-pentamer of high stability . The absence of alternate stoichiometries , either prior to dissociation or as a product or intermediate of the dissociation process , further supports the pentamer as the likely near exclusive stoichiometry . Resolution of the isotopic distribution of the monomer mass spectrum confirms the assigned charge state and gives another estimate of mass . This mass of 38440 . 12 Da is based on the most abundant isotope and fundamentally differs slightly from the average ( S4B Fig ) . Note that although greater accuracy is achieved here , the accuracy of all of the mass measurements presented are far more accurate than the inherent variance in mass due to the natural variation in isotopic composition found in different environments . Sedimentation velocity experiments ( SV ) were also used to study the Orsay virus full-length δ protein in solution . SV experiments characterize the solution behavior of macromolecules and observe the sedimentation and diffusion behavior of all species in a mixture , and report their partial concentrations , buoyant molecular weights , and anisotropies . Sedimentation coefficient distributions from the δ protein demonstrated the presence of a major species sedimenting with a fairly broad peak centered at 5 . 4 s with a frictional ratio of 2 . 1 , which indicates a high degree of anisotropy ( S5 Fig ) . This is consistent with a fibril-like conformation of the protein . A molar mass transformation of this peak resulted in a weight-average molar mass of 191 . 0 kDa , in excellent agreement with the calculated molecular weight of 192 . 2 kDa for the pentameric form of this protein ( Fig 4B , Table 2 ) . To analyze the structure of CP-δ in the context of a viral capsid , we co-expressed CP and CP-δ in insect cells by co-infection with two baculoviruses each expressing a different protein . The use of two baculoviruses would allow the control of the relative amount of CP and CP-δ to optimize particle assembly . Recombinant VLPs were purified by Ni-NTA affinity as both CP and CP-δ contained a C-terminal His-tag . Under negative-staining EM , we observed many spherically shaped particles associated with long fibers ( Fig 5A ) . The diameter of these particles is around 350-Å , closely matching that of the Orsay viron or VLP [2 , 23] . The length of particle-associated fibers , when measured from the surface of the capsid , is 387±42-Å ( n = 23 ) , which is similar to the length of free δ measured at 419±52-Å ( Fig 5B ) . Some of the fibers even show a head domain at their distal end , consistent with our assumption that the N-terminal coiled coil of the δ fiber is directly attached to the CP surface spike . Considering the 5-fold symmetry of the δ protein , in principle there can be up to 12 copies of the CP-δ fibers in each capsid , with one occupying each icosahedral vertex ( Fig 5D ) . Coomassie-stained SDS-PAGE gel of our capsid sample showed the mass ratio of CP-δ to CP is around 1:2 ( Fig 5C ) , which corresponds to roughly 1:4 in molar ratio , suggesting that on average there should be ~7 pentameric fibers in each particle . The most fibers we observed in a single VLP were seven . It is possible that some fibers were not visible due to staining artifacts , or that not all CP-δ was properly incorporated into capsids . Site-directed mutagenesis and reverse genetics were performed to confirm that the structural integrity of CP-δ is important for Orsay infectivity . Two residues K43 and L44 were targeted for mutation . The crystal structure of δ ( 1–66 ) shows that K43 and D45 form an intermolecular salt bridge on the surface of the α-helical bundle ( Fig 6A ) . L44 is located at the hydrophobic core of the pentameric coiled coil ( Fig 6B ) . Both mutations K43E and L44R were expected to disrupt the structure of δ/CP-δ . Indeed , our results showed that δ ( 1–66 ) constructs bearing either the K43E or the L44R mutation could no longer form regular pentamers , considering the substantial shifts in their peak positions in gel filtration profiles ( S6 Fig ) . Using transgenic C . elegans carrying virus cDNAs as previously described [5] , two recombinant viruses , one with the K43E mutation and the other with the L44R mutation , were generated . Recombinant viruses collected from transgenic worm lysate were applied on naïve worms sensitive to Orsay infection . These worms were then evaluated for viral infectivity by two measurements: infection symptoms and viral load . Orsay-infected worms often display a transparent intestine phenotype ( Fig 6C ) . The percentage of worms with such symptom was drastically reduced with the two mutant viruses ( Fig 6C ) . The L44R mutant virus appeared more defective than K43E , as worms infected with the L44R virus showed no difference than uninfected worms , while a small fraction of worms infected with the K43E virus showed the infection symptom ( Fig 6C ) . These observations were confirmed in multiple independent lines of transgenic worms ( Figs 6C and S7 , demonstrating that the differences were indeed caused by viral genotypes . In addition , the viral load in these worms was determined using qRT-PCR , and consistent results were obtained ( Fig 6D ) . Both mutations significantly reduced the viral load , with more severe defects with the L44R mutant virus . As the K43E and the L44R mutations could potentially affect both CP-δ and free δ , another recombinant Oray virus was generated to distinguish which protein led to the observed defects . In this recombinant virus ( i . e . δ-null ) , the start codon of free δ was mutated from ATG to CTG so that no free δ was produced . A high percentage of worms infected with this mutant virus showed the transparent intestine symptoms; the viral load was also similar to that of wild-type virus ( Fig 6C and 6D ) , suggesting that free δ was not required for infection based on our assay . Therefore , the K43E and the L44R mutant viruses lost their infectivity likely because of defective CP-δ . The lack of infectivity from δ mutants suggested that CP-δ is required for infection . We asked whether CP-δ functions in viral entry or at a later stage . We reasoned that if δ functions in viral entry steps such as receptor-binding at the cell surface , then adding purified δ in the culture medium would compete against the virus CP-δ for such binding sites , and would thus reduce the efficiency of Orsay infection . On the other hand , if CP-δ functions in steps post viral entry , such as intracellular viral replication , then adding proteins in the culture medium would have little impact on viral infectivity . To conduct the protein-competition assay , we first determined the viral titer and chose the lowest viral concentration with over 70% infectivity ( Fig 6E ) . At this viral concentration , adding 2μg/ml full-length δ to the culture medium significantly reduced the infectivity from 92% to 33% ( Fig 6F ) , demonstrating that δ functions at the viral entry step . In contrast , adding purified capsid or the N-terminal fragment δ ( 1–101 ) did not show this effect ( Fig 6F ) , consistent with our structural model that the δ C-terminal globular head functions in cell attachment .
Results from our study indicated that the CP-δ fusion protein plays a specific function in host cell entry during Orsay infection based on the following evidence: ( 1 ) Orsay δ forms pentameric fibers; ( 2 ) CP-δ is incorporated into viral capsid as a minor structural protein; ( 3 ) the δ portion of CP-δ forms a long projecting fibers with a globular head domain at the distal end; ( 4 ) disrupting the structural integrity of CP-δ results in non-viable virus mutants; and ( 5 ) the addition of recombinant δ to worm medium reduced Orsay infectivity . The use of recombinant VLPs enabled us to directly visualize the CP-δ fibers due to the enhanced amount of CP-δ in the VLP sample . In contrast , the native virion sample contains only ~5% CP-δ [7] , which corresponds to only one to two CP-δ fibers in average in each particle . It would be difficult to identify these long fibers by EM unless they lie flat on sample grids and interact evenly with heavy atom stains , thus explaining the difficulties we had trying to visualize such fibers using native virion samples . The Orsay δ/CP-δ fiber consists of several domains ( Fig 1D ) . The first ~60 residues at the N-terminus of δ form an α-helical bundle and play an important role in stabilizing the δ/CP-δ pentameric fiber . The rest of the sequence is largely β-stranded and likely forms β-barrels or β-spirals connected by non-structured loops . While β-barrels and β-spirals are frequently observed in viral structural proteins , pentameric β-fibers have not been previously reported [41 , 42] . The diameter of the β-fiber in the Orsay δ is only ~25-Å , smaller than that of the helical bundle at the tail end as shown by EM images . There is a large globular head at the C-terminal end of the δ/CP-δ fiber . Another globular domain , which is smaller in size , is found at the two-fifths position from the N-terminal end . The observation of a globular head at the distal end of the capsid-associated CP-δ fibers is consistent with our domain assignment . Although we have not experimentally verified that the CP-δ fibers are also pentamers , the consideration of stereochemical constraints in the context of an Orsay capsid suggests that pentameric fibers are energetically favored . The crystal structure of the Orsay capsid shows that the C-terminus of the Orsay CP is tucked underneath of a tightly bound trimeric spike [23] ( S8 Fig ) . Therefore , for a trimeric fiber to form , the polypeptide sequence would have to go around the timeric spike from outside , spanning a distance of at least 60-Å in order to reach the 3-fold axis . By comparison , the C-terminus of the CP points toward a depression around the 5-fold symmetry axis , with only a 25-Å traveling distance to the 5-fold , thus facilitating the formation of pentameric fibers . Our results from mass spectrometry also indicate that the pentameric δ fiber is very stable and does not dissociate unless under high energetic conditions , suggesting that it is unlikely for the δ sequence to adopt an alternative trimeric configuration in the form of the CP-δ fusion protein . While δ showed no sequence homology to any known proteins , the morphology of the CP-δ protein , as well as its localization in the capsid and its secondary structure content , is reminiscent of the fibers found in both reovirus and adenovirus [43 , 44] . Like CP-δ , reovirus σ1 fiber protein is organized into three modules: a coiled coil tail domain at the N-terminus , a β-filament body domain in the middle , and a C-terminal β-stranded head domain . The adenovirus fiber does not have a coiled coil region , but also has a head-and-tail morphology with a long shaft made of ~20 β-spiral repeats and a C-terminal head comprised of an eight-stranded β-barrel [45] . Both adenovirus and reovirus fibers are situated at five-fold symmetry axes with their N-terminal sequence interacting with the viral capsid and their C-terminal head at the distal end [46 , 47] , same as the Orsay CP-δ . The Orsay CP-δ forms pentameric fibers , however , while the adenovirus and reovirus fibers are both trimeric . In adenovirus and reovirus , the cell receptor binding sites are mapped to the globular head domain of their fibers , except that in some reoviruses a sialic acid binding site is found in the middle body domain of σ1 [48–51] . The overall lengths of the adenovirus ( i . e . Ad2 and Ad5 ) and reovirus fibers are ~325 and 385-Å , respectively [45] , slightly shorter than the Orsay CP-δ fiber . Results from the competition experiments using free δ ( Fig 6E and 6F ) and the close analogy between the CP-δ fiber and the fibers from reovirus and adenovirus suggest that the Orsay CP-δ likely functions as a cell receptor binding protein . The globular head of CP-δ likely hosts the cell receptor binding site as it does in reovirus and adenovirus . The binding of CP-δ to the host receptor should allow virus attachment to the host intestinal cells for the subsequent cell entry . The cell receptor molecule for Orsay has yet to be determined , but we expect that viral particles containing only CP but no CP-δ fibers would be non-infectious due to blocked cell entry . It is unclear whether having only 1 to 2 copies of the CP-δ fiber instead of a full complement of 12 would negatively impact Orsay’s infectivity , but dsDNA bacteriophages such as ϕ29 are highly infectious with only one tail structure in each viral particle [52 , 53] . For the bacteriophage T4 , it was demonstrated that three fibers per virion are sufficient for infectivity , and reducing the lipopolysaccharide receptor concentration on cell surface has the same effect as tail fiber limitation on phage infectivity [54] . Therefore , it is possible that not all 12 copies of the CP-δ fiber are needed for Orsay , especially if abundant receptor molecules exist on the C . elegans intestinal cell surface . Although our infectivity assays did not detect any obvious functional defects for the δ-null mutant , we cannot rule out the possibility that free δ may still play important roles during the virus life cycle that are distinct from the cell entry function mediated by CP-δ . Both of our infectivity assays in Fig 6 , one based on the body transparency of infected animals and the other measuring viral RNA in worm lysates , relied on one infection cycle and therefore mainly detected mutant defects in viral entry . Mutant defects downstream from viral RNA replication cannot be effectively measured using these assays . It remains possible that the free δ protein may interact with the host machinery to promote virus assembly and/or mediate the release of viral particles from the apical side of the worm intestine cells . Furthermore , many non-enveloped animal viruses are known to encode a lytic peptide or protein , but such function has not yet been reported for Orsay . Free δ may function as a lytic protein . Future cytological and biochemical analyses should help to identify interesting leads in this direction . By defining the structure and function of the Orsay CP-δ fibers , findings from our present study represent a major advance in our understanding of Orsay cell entry . Additionally , we expect our results to serve as a useful guide for future work related to Orsay host receptor identification as well as detailed characterization of the molecular interaction between Orsay and its host receptor .
The coding sequence of the full-length Orsay δ ( GenBank accession no . HM030971 . 2 ) was inserted into a modified pETDuet-1 vector that would add a 6xHis-SUMO tag to the recombinant protein at the N-terminus . Removal of the fusion tag using the SUMO protease Ulp should leave a dipeptide HM at the N-terminal end of the recombinant protein . C-terminal truncation mutants δ ( 1–66 ) , δ ( 1–101 ) , δ ( 1–162 ) , and δ ( 1–241 ) were made by introducing a termination codon at desired sites by PCR using a pair of complementary primers . To make CP-δ fusion protein constructs , a single nucleotide “A” was inserted in front of the last nucleotide before the stop codon of the CP ORF to shift the δ ORF to the same coding frame . The modified sequence would express CP-δ , the same as expected from ribosomal frameshifting . For the mini-fusion protein , the DNA sequence coding for residues 215–485 of CP-δ , which contains the protrusion domain of the CP , the 29-aa linker , and an N-terminal fragment of δ ( 1–66 ) , was cloned into the modified 6xHis-SUMO pETDuet-1 vector as mentioned above , For the midi-fusion protein , the DNA sequence coding for residues 215–581 of CP-δ , which contains the protrusion domain of the CP , the 29-aa linker , and an N-terminal fragment of δ ( 1–162 ) together with a C-terminal 6xHis tag , was cloned into pFastBac1 ( Thermo Fisher Scientific ) and the recombinant baculovirus was subsequently generated following the Bac-to-Bac Expression System manual . To produce recombinant Orsay capsid containing CP-δ , the DNA sequences coding for Orsay virus CP-δ fusion protein ( C-terminally 6xHis-tagged ) and CP ( N- and C-terminally 6xHis- tagged ) were each cloned into pFastBac1 . Two baculoviruses were generated as described above , one expressing CP and the other expressing CP-δ . For protein expression in E . coli , cells at the phase of exponential growth were induced using 1 mM Isopropyl β-D-1-thiogalactopyranoside ( IPTG ) when OD600nm reached 0 . 6–0 . 8 . After overnight shaking at 15°C , cells were harvested by centrifugation at 2 , 000xg for 20 min and sonicated in lysis buffer containing 50 mM Tris pH8 . 0 , 300 mM NaCl , 10% glycerol ( v/v ) , 5 mM 2-Mercaptoethanol ( 2-ME ) , 1 mM NaN3 and 1 mM phenylmethylsulfonyl fluoride ( PMSF ) . 6xHis-SUMO-tagged proteins were first purified by affinity chromatography using the Ni-NTA resin ( Thermo Fisher Scientific ) . After Ni-NTA affinity , the eluates were collected and incubated with a SUMO protease ( Ulp ) at a mass ratio of 1: 10 [Ulp: His6—SUMO - δ ( 1–66 ) ] overnight at 4°C for affinity tag removal . Afterward , the mixture was brought to 25 mM imidazole and re-applied to Ni-NTA resin and the flow-through containing the δ ( 1–66 ) from the second Ni-NTA was collected . The sample was next purified by size exclusion using a Superdex 200 gel filtration column that with an elution buffer containing 50 mM Tris-HCl ( pH 7 . 5 ) , 250 mM NaCl , 350 μl 2-ME , and 1 mM NaN3 . Peak fractions containing δ ( 1–66 ) were loaded onto a 2-ml HisTrap HP column ( GE Healthcare Life Sciences ) for a final cleanup . The flow-through containing purified proteins was concentrated to 5 mg/ml and stored at 4°C . SeMet-substituted δ ( 1–66 ) was expressed in M9 minimal medium supplemented with SeMet [55] . Expression was induced with 1 mM IPTG for 24 h at 15°C . For the midi-fusion protein , ~2X108 ( or 200 ml ) Spodoptera frugiperda 21 ( Sf21 ) insect cells grown in supplemented Grace’s insect medium ( Life Technologies ) were infected with baculovirus and harvested 60 h post-infection . The cell pellets were washed with cold phosphate-buffered saline ( PBS ) and sonicated in a cold lysis buffer containing 50 mM Tris-HCl ( pH 8 . 0 ) , 300 mM NaCl , 1 mM NaN3 , 1 mM PMSF , 10% ( v/v ) glycerol , 0 . 5% ( v/v ) Triton X-100 , 10 μg/ml DNase , and 15 μg/ml RNase . The midi-fusion protein was purified by Ni-NTA affinity followed by gel filtration chromatography . To produce recombinant Orsay capsids containing CP-δ , ~2X109 ( or 2 liters ) Sf21 insect cells grown in supplemented Grace’s insect medium ( Life Technologies ) were co-infected with 100 ml of the recombinant baculovirus expressing CP-δ and 100 ml of the recombinant baculovirus expressing CP . Cells were harvested 60 h post-infection . The cell pellets were washed with cold PBS and sonicated in a cold lysis buffer containing 50 mM Tris-HCl ( pH 8 . 0 ) , 300 mM NaCl , 1 mM NaN3 , 1 mM PMSF , 10% ( v/v ) glycerol , 0 . 5% ( v/v ) Triton X-100 , 10 μg/ml DNase , and 15 μg/ml RNase . The clarified lysate was loaded onto a Ni-NTA column . Eluted fractions were collected and further purified by a 2-ml HisTrap HP column ( GE Healthcare Life Sciences ) . For EM sample preparation , FCF400-Cu grids ( Electron Microscopy Sciences ) were pretreated by glow-discharge at 5 mA for 1 min as previously described [56] . 5 μl of the protein solution was then added onto the grid and sat for 30 s to allow absorption . To optimize particle spread , a number of different protein concentrations ranging from 1 mg/ml to 0 . 01 mg/ml were prepared simultaneously . The protein solution was removed from the grids by filter paper blotting . The grids were then rinsed twice with distilled water and stained with freshly prepared 0 . 75% Uranyl formate solution for 60 s . After air-drying overnight , the grids were examined using a JEOL 1230 High Contrast transmission electron microscope at 80 kV . Images were recorded on a Gatan CCD detector . Both native δ ( 1–66 ) and SeMet-substituted δ ( 1–66 ) were crystallized at 20°C by hanging-drop vapor diffusion . For native crystals , drops containing 1 μl of native δ ( 1–66 ) at 5 mg/ml concentration were combined with 1 μl of mother liquor containing 1 . 6 M ammonium sulfate , 0 . 1 M MES monohydrate ( pH 6 . 5 ) , and 12% ( v/v ) dioxane . Plate-shaped crystals appeared within 4 to 5 days . For SeMet-substituted crystals , drops containing 1 μl of SeMet-substituted δ ( 1–66 ) ( 5 mg/ml ) were combined with 1 μl of mother liquor containing 0 . 1 M citric acid ( pH 5 . 0 ) and 14% polyethylene glycol ( PEG ) 6000 . Rod-shaped crystals appeared in two weeks . Crystals were cryo-protected using 25% ( v/v ) glycerol and flash-frozen in liquid nitrogen . Diffraction data were collected from single crystals at the Life Sciences Collaborative Access Team ( LS-CAT ) at the Advanced Photon Source ( APS ) . Data were processed using HKL2000 [57] . The structure of δ ( 1–66 ) was determined by single-wavelength anomalous dispersion ( SAD ) . The SeMet sites and experimental phases were calculated by the AutoSol Wizard in the PHENIX software suite [58] . The protein model was built with PHENIX Autobuild and COOT [59] and refined with phenix . refine . The structure was finally refined against a native dataset at 2 . 22-Å resolution . The final structure , which contains 284 residues and 197 waters , has a final Rwork of 20 . 64% and Rfree of 23 . 14% . The coordinates have been deposited in the RCSB Protein Data Bank ( PDB ID: 5JIE ) . All structure figures were prepared using the program PyMOL unless otherwise specified ( The PyMOL Molecular Graphics System , Version 1 . 2r3pre , Schrödinger , LLC ) . The protein sample was dialyzed into 20 mM Potassium phosphate pH 7 . 4 , and the concentration was adjusted to 0 . 5 mg/mL . Circular dichroism signal was measured using a J-815 Circular Dichroism Spectropolarimeter ( Jasco Analytical Instruments ) . The wavelength range was set from 200 nm to 280 nm with a step size of 0 . 2 nm . The results were analyzed using BeStSel [32] . A solution of 10ug/mL of protein complex in 5% ACN , 0 . 1% FA was directly infused at 3 uL/min with a nano-spray source into an Orbitrap Fusion Lumos ( Thermo ) . An ionization voltage of 2200 V was used in combination with a 70V source fragmentation voltage . The intact complex was best observed with MS2 ETD 0 . 1 ms reaction time , Quadrupole isolation of 3300 m/z with a 500 m/z window and 15 , 000 resolution in an Orbitrap analyzer . Note that these conditions produce effective intact protein complex observation and do not result in substantial dissociation of the non-covalent complex or fragmentation of covalent peptide bonds . For the dissociation of the pentameric complex and observation of the intact monomer , the following conditions were used: MS2 with HCD 12% energy , 70V Source Fragmentation , 60000 Resolution Orbitrap with Quadrupole isolation of 3300 m/z with a 500 m/z window in high mass range , 47 scans . Complex dissociation was efficiently achieved with HCD of 9% collision energy and complete complex dissociation was observed around 12%-13% HCD collision energy . CID required much higher energies: up to 90% to see similar complex dissociation . Note the relatively high collisional energies used to achieved dissociation of the pentamer , with essentially no cleavage of covalent bonds . Orbitrap high resolution analysis ( 500 , 000 resolving power setting ) allows isotopic separation of the monomer with both HCD and CID fragmentation; however , isotopic resolution of the intact 192kDa was not possible , nor expected . For isotopic resolution of the monomer the following conditions were used: MS2 with CID 50% Collision energy , 70V Source Fragmentation , 500 , 000 resolving power setting on the Orbitrap with Quadrupole isolation of 3300 m/z with a 500 m/z window in high mass range mode . A solution of the Orsay virus full-length δ protein at 0 . 91 OD 280 nm ( 27 . 6 μM ) was sedimented at 20°C and 30 , 000 rpm , and measured by UV intensity detection in a Beckman Optima XLI analytical ultracentrifuge at the Center for Analytical Ultracentrifugation of Macromolecular Assemblies at the University of Texas Health Science Center at San Antonio , using an An60Ti rotor and standard 2-channel epon centerpieces ( Beckman-Coulter ) . All data were analyzed with UltraScan-III ver . 3 . 5 , release 2174 ( http://www . ultrascan3 . uthscsa . edu ) . All samples were measured in a 50 mM TRIS buffer , pH 7 . 5 , containing 250 mM NaCl . Hydrodynamic corrections for buffer density and viscosity were estimated by UltraScan to be 1 . 010 g/ml and 1 . 036 cP . The partial specific volume of the delta protein ( 0 . 7399 ml/g ) was estimated by UltraScan from protein sequence analogous to methods outlined in Laue et al [60] . SV data were analyzed according to the approach described in [61] . Optimization was performed by 2-dimensional spectrum analysis ( 2DSA ) [62] with simultaneous removal of time- and radially-invariant noise contributions [63] and meniscus fitting . After noise subtraction and meniscus fitting , the data were analyzed by the 2DSA Monte Carlo analysis to identify particle distributions in the frictional ratio–sedimentation coefficient domain [64] . The distribution suggested that a decreasing sigmoid parameterization is suitable for fitting the data with the parametrically constrained spectrum analysis ( PCSA-DS ) , using Tikhonov regularization [65] . The calculations are computationally intensive and are carried out on high-performance computing platforms [66] . All calculations were performed on the Lonestar cluster at the Texas Advanced Computing Center at the University of Texas at Austin and on Comet and Gordon at San Diego Supercomputing Center . The resulting fit produced random residuals and is shown in S5 Fig in the Supplemental Information . The plasmids pHIP_RNA1 and pHIP_RNA2 were obtained as a gift from Dr . David Wang [5] . Site mutations K43E , L44R , and ATG→CTG were introduced to the δ ORF in pHIP_RNA . 50 ng/μl mutant pHIP_RNA2 , 50 ng/μl pHIP_RNA1 , and 100 ng/μl pRF4 [67] were mixed and microinjected into N2 day-1 adults . Animals were cultured at 15°C on NGM plate seeded with OP50 bacteria following standard culture conditions [68] . F1 worms with the roller phenotype were picked individually to a new plate , and screened for subsequent generations of rollers to obtain stable transgenic lines . The stable transgenic line for wild-type recombinant Orsay virus was a gift from Dr . David Wang [7] . 6-well RNAi plates ( NGM with 1mM IPTG and 50 ng/μl Carbenicillin ) seeded with rde-1 RNAi bacteria ( a clone from the Ahringer library ) were prepared as described [69] . 30 L4 rollers from each stable transgenic line were picked onto a 6-well plate ( 5 worms/well ) , and cultured for 5 days at 20°C . The worms were heat induced at 33°C for 2 hours and then at 25°C for 2 days [5] . Worms were fed with IPTG-induced rde-1 RNAi bacteria throughout the course to prevent starvation . Worms were then washed off the plates using S Basal [68] . Excess liquid was aspirated so that the volume of worm pallet and liquid was about 1:1 . The mixture was then sonicated to obtain worm lysate . The crude lysate was centrifuged at 10 , 000xg for 10 min at 4°C . The supernatant was filtered through a 0 . 22 μm syringe filter and kept at 4°C till used for infection . glp-4 ( bn2 ) ; rde-1 ( RNAi ) worms were used as naïve worms . The transparent intestine symptom was best observed on day-3 adults . glp-4 ( bn2 ) worms were used because they are sterile at high temperatures [70] , and can easily grow to day-3 adults without getting starved due to progeny . rde-1 ( RNAi ) were used to make the worms sensitive to Orsay infection . Synchronized L1 naïve worms were obtained by bleaching [68] . 150 L1 worms and 200 μl of viral filtrate were added to each well of a 6-well RNAi plate . The infected worms were cultured at 20°C for five days till they were day-3 adults . Three independent trials of infection ( biological replicates ) were performed . In each trial , three wells ( of a 6-well plate ) of worms were infected by the viral filtration from each transgenic C . elegans line . Worms from one well were counted under a stereoscope for the transparent intestine phenotype . The other two wells were used for qRT-PCR test of viral load . For qRT-PCR , worms were washed off plates and rinsed four times with S Basal . RNA was extracted using Trizol ( Invitrogen ) . cDNA was generated using the RETRO script Kit ( Ambion ) . qRT-PCR was performed using PerfeCTa SYBR Green SuperMix ( Quanta Biosciences ) , with the primers GW194 , GW195 for viral RNA1 fragment and AMA-1F , AMA-1R for the internal reference gene ama-1 [2] . The viral product was first normalized to ama-1 , and then normalized to the values of the wild-type recombinant virus . Three technical replicates were performed for qRT-PCR , and their average was used as one data point for Fig 6D . ~100 synchronized L1 naïve worms were added to each well of a 96-well plate that contained 100 μl of S medium [68] with 1mM IPTG , 50 ng/μl Carbenicillin , rde-1 RNAi bacteria , Orsay virus , and 2μg/ml Capsid , 2μg/ml δ , or 0 . 57 μg/ml N-terminus δ fragment δ ( 1–101 ) . Theses protein concentrations were used so that the molar concentrations of the three proteins were similar . To determine the virus concentration , 2-fold serial dilution of viral filtration was conducted to determine a titration curve . The lowest viral concentration that can infect ≥70% plates was used for protein-competition . Animals were cultured on a 20°C incubator shaker till they were day-3 adults . Animals were then transferred from each well to an unseeded NGM plate to count worms with the transparency symptom . A plate with over 50% transparent worms was scored as an infected plate . The percentage of infected plates was calculated to measure viral infectivity .
|
Orsay , the only virus known to naturally infect the nematode Caenorhabditis elegans , expresses a capsid-δ fusion protein . We have demonstrated that the Orsay fusion protein is incorporated into Orsay capsid and forms a ~420-Å long fiber protruding from the capsid surface . Crystal structure of an N-terminal fragment of the δ sequence shows a pentameric coiled coil , suggesting that the Orsay fiber is incorporated into viral capsid at five-fold vertices . Our model predicts that δ consists of three domains: an N-terminal helical domain that is covalently linked to the capsid , a middle shaft that is mostly β-stranded , and a C-terminal head that likely hosts the cell receptor binding site for host attachment . Our findings have substantially advanced our understanding of the infection mechanism of this new virus . The Orsay δ protein represents the first pentameric viral fibers reported up-to-date with potential bioengineering applications .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"rna",
"interference",
"crystal",
"structure",
"caenorhabditis",
"condensed",
"matter",
"physics",
"microbiology",
"viral",
"structure",
"parasitic",
"diseases",
"animals",
"nematode",
"infections",
"animal",
"models",
"caenorhabditis",
"elegans",
"model",
"organisms",
"experimental",
"organism",
"systems",
"protein",
"structure",
"epigenetics",
"crystallography",
"research",
"and",
"analysis",
"methods",
"solid",
"state",
"physics",
"genetic",
"interference",
"proteins",
"gene",
"expression",
"viral",
"packaging",
"recombinant",
"proteins",
"viral",
"replication",
"molecular",
"biology",
"physics",
"biochemistry",
"rna",
"nucleic",
"acids",
"virology",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms",
"macromolecular",
"structure",
"analysis"
] |
2017
|
Structure of a pentameric virion-associated fiber with a potential role in Orsay virus entry to host cells
|
Despite decades of research , the question of how the mRNA splicing machinery precisely identifies short exonic islands within the vast intronic oceans remains to a large extent obscure . In this study , we analyzed Alu exonization events , aiming to understand the requirements for correct selection of exons . Comparison of exonizing Alus to their non-exonizing counterparts is informative because Alus in these two groups have retained high sequence similarity but are perceived differently by the splicing machinery . We identified and characterized numerous features used by the splicing machinery to discriminate between Alu exons and their non-exonizing counterparts . Of these , the most novel is secondary structure: Alu exons in general and their 5′ splice sites ( 5′ss ) in particular are characterized by decreased stability of local secondary structures with respect to their non-exonizing counterparts . We detected numerous further differences between Alu exons and their non-exonizing counterparts , among others in terms of exon–intron architecture and strength of splicing signals , enhancers , and silencers . Support vector machine analysis revealed that these features allow a high level of discrimination ( AUC = 0 . 91 ) between exonizing and non-exonizing Alus . Moreover , the computationally derived probabilities of exonization significantly correlated with the biological inclusion level of the Alu exons , and the model could also be extended to general datasets of constitutive and alternative exons . This indicates that the features detected and explored in this study provide the basis not only for precise exon selection but also for the fine-tuned regulation thereof , manifested in cases of alternative splicing .
How are short exons , embedded within vast intronic sequences , precisely recognized and processed by the splicing machinery ? Despite decades of molecular and bioinformatic research , the features that allow recognition of exons remain poorly understood . Various factors are thought to be of importance . These include the splicing signals flanking the exon at both ends , known as the 5′ and 3′ splice sites ( 5′ss and 3′ss , respectively ) , auxiliary cis-elements known as exonic and intronic splicing enhancers and silencers ( ESE/Ss and ISE/S ) that promote or repress splice-site selection , respectively [1] , [2] , and exon [3] and intron length [4] . There is an increasing body of evidence that secondary structure is a powerful modifier of splicing events [5]–[12] . Secondary structure is thought to present binding sites for auxiliary splicing factors , correctly juxtapose widely separated cis-elements , and directly affect the accessibility of the splice sites . However , only very few studies have used bioinformatic approaches to broadly study the effects of secondary structure on splicing [13]–[15] . Many of the above-listed factors have been subjected to analysis in the context of comparison between constitutively and alternatively spliced exons . It has been found , for example , that constitutively spliced exons are flanked by stronger splicing signals , that they contain more ESEs but fewer ESSs , and are longer but flanked by shorter introns with respect to their alternatively spliced counterparts ( reviewed in [16] ) . However , to what extent do these features contribute to the selection of exons and allow discrimination between true exons and “non-exons” , i . e . sequences resembling exons but not recognized by the splicing machinery ? This question is fundamental for understanding the process of exon selection by the spliceosome , and yet has not been subjected to much analysis . This is presumably because unlike alternatively and constitutively spliced exons , both of which are relatively easy to define computationally , defining a non-exon or a pseudo-exon is more of a challenge . One approach is to compare exons to sequences of up to a certain length which are flanked by splicing signals exceeding a certain threshold [17] , [18] . Although this approach is powerful and has contributed to the discovery of the “vocabulary” of exons , it is also limited . The primary limitation is that it is circular: For the mere definition of pseudo-exons , we are forced to fix various features—such as minimal splice site strength and exon length—that we would prefer to infer . To circumvent these obstacles , we have studied Alu exonization events . Alu elements are primate-specific retroelements present at about 1 . 1 million copies in the human genome . A large portion of Alu elements reside within introns [19] . Alus are dimeric , with two homologous but distinct monomers , termed left and right arms [20]–[22] . During evolution , some intronic Alus accumulated mutations that led the splicing machinery to select them as internal exons , a process termed exonization [23]–[25] . Such exonization events may occur either from the right or the left arm of the Alu sequence , but are observed predominantly in the antisense orientation relative to the mRNA precursor . Almost invariably , such events give birth to an alternatively spliced exon , as a constitutively spliced exon would compromise the original transcriptomic repertoire and hence probably be deleterious [19] , [24] , [26] , [27] . The fact that exonizing and non-exonizing Alus have retained high sequence similarity but are perceived as different by the splicing machinery makes them excellent candidates for studying the factors required for precise recognition of exons by the spliceosome . The natural control group of non-exonizing Alus obviates the need to fix different parameters in the control set , and the high degree of sequence similarity shared by all Alus , regardless of whether they do or do not undergo exonization , enables direct comparison of a wide array of features . Based on the comparison between Alu exons and their non-exonizing counterparts , we were able to identify several key features that characterize Alu exons and to determine the relative importance of these features in the process of Alu exonization . A novel result of this comparison was the importance of pre-mRNA secondary structure: More thermodynamically stable predicted secondary structure in an Alu arm harboring a potential Alu exon decreases the probability of an exonization event originating from this Alu . Thus , this study is among the first to provide wide-scale statistical proof of the importance of secondary structure in the context of exon selection . We identified numerous further factors differentiating between Alu exons and non-exons , and integrated them in a machine learning classification model . This model displayed a high performance in classifying Alu exons and non-exons . Moreover , the strength of predictions by this model correlated with biological inclusion levels , and higher probabilities of exonization were given by the model to constitutive exons than to alternative ones . These findings indicate that the features identified in this study may form the basis for precise exon selection , and make the difference between a non-selected element , an alternatively-selected element , and a constitutively selected one .
We set out to determine the features underlying the recognition of Alu exons by the splicing machinery . We therefore required datasets of Alus that undergo and that do not undergo exonization . We took advantage of the fact that Alu elements may exonize either from the right or from the left arm , and composed three core datasets ( Figure 1A ) : ( 1 ) A dataset of 313 Alu exons ( AEx ) that are exonized from the right Alu arm , termed AEx-R; ( 2 ) A dataset of 77 Alus that undergo exonization in the left arm , termed AEx-L; ( 3 ) A dataset of 74 , 470 intronic Alus lacking any evidence of exonization , called No AEx . In all these datasets , Alus had to be embedded in the antisense orientation within genes , since most exonization events of Alus occur in this orientation [19] , [23] , [28] . Finally , to allow direct comparison between parallel positions in different Alus , we used pairwise alignments to align each Alu in each of the datasets against an Alu consensus sequence . We next computationally searched for the optimal borders , or splice sites , of non-exons within both the right arm and the left arm of the sequences in the No AEx dataset . This was done in two steps: ( 1 ) We first empirically determined the positional windows in which the selected 3′ss and 5′ss appeared within exonizing Alus; ( 2 ) We next searched the above-determined positional windows for the highest scoring splicing signals ( see Materials and Methods ) . We found that computational selection of the highest scoring splicing signal yielded a high extent of congruence ( ranging between 74%–96% , depending on the arm and on the signal ) with the “true” splicing signal based on EST data . Since the congruence was not perfect , we created two control datasets based on the AEx-R and AEx-L group , termed AEx-R ( c ) and AEx-L ( c ) , respectively , in which exon borders were searched for computationally as in the No AEx dataset . These two subsets were used to verify that differences between the exonizing and non-exonizing datasets were not due to the manner in which exons and non-exons were derived ( ESTs versus computational predictions ) . To complete the picture , we computationally searched for non-exons within the right arm of the AEx-L group and in the left arm of the AEx-R group . Notably , we demanded that all exons within all datasets have a minimal potential 3′ss ( AG ) and 5′ss ( GT/GC ) , because lacking such minimal conditions Alus cannot undergo exonization at all . Thus , our analyses are based on three core and two control sets of Alus with two sets of start and end coordinates mapped for each Alu—one in the right arm and one in the left ( see Materials and Methods for further details ) . Previous studies , based on much smaller datasets , implicated the 3′ss [24] and the 5′ss [26] splicing signals as major factors determining exonization events . To assess whether this held for our dataset as well , we calculated the strength of the 5′ss and 3′ss of the exons/non-exons in the right and in the left arms in each of the five datasets . Indeed , we found that in the right arms the 3′ss and the 5′ss scores were highest among those Alus that underwent exonization ( Figure 1B and 1C , respectively ) . Similarly , in the left arms , the scores of the 3′ss and the 5′ss are highest among the exonizing Alus ( Figure 1D and 1 E , respectively ) . These results were highly statistically significant ( see Text S1 ) . Moreover , these differences are even more pronounced when comparing the two control datasets to their non-exonizing counterparts ( compare the results for AEx-R and AEx-L to AEx-R ( c ) and AEx-L ( c ) , respectively , in Figure 1B–E ) . Thus , these analyses fit in with previous analyses emphasizing the role of the two major splicing signals . We were interested in assessing the role of secondary structure in the context of Alu exonization events . We therefore began by computing the thermodynamic stabilities of the secondary structures predicted for the Alus in each of the core datasets . We used RNAfold [29] to calculate the secondary structure partition function; but rather than use this metric directly , we used a dinucleotide randomization approach to yield a Z-score that is not sensitive to sequence length or nucleotide composition ( see Materials and Methods ) . We found that Alus that gave rise to exonization events , regardless of whether from the left or from the right arm , were characterized by weaker secondary structures than Alus that do not undergo exonization ( Figure 2A ) . This was highly significant in the case of exonizations originating from the right arm ( AEx-R vs . No AEx p = 9 . 8E−12 ) and of borderline significance for the left arm exonizations ( AEx-L vs . No AEx p = 0 . 07 ) . This provided the first indication that strong secondary structures might prevent Alu exonizations . To pinpoint the subsequences to which the differences in strength of secondary structure could be attributed , we next calculated secondary structure Z-scores for each of the two Alu arms separately . We found that the secondary structures of right and the left arms were weakest in cases in which these arms undergo exonization ( Figure 2B and 2C , respectively ) . These changes relative to the No AEx group were highly significant ( p = 2E−15 and p = 1 . 08E−5 , respectively ) . Interestingly , the non-exonizing arm tended to have weaker secondary structure in those cases in which the opposite arm underwent exonization ( p = 0 . 001 when comparing the left arm of the AEx-R to the No AEx dataset , and p = 0 . 055 when comparing the right arm of the AEx-L to the No AEx dataset ) . These observations suggested that secondary structures have a detrimental effect on the recognition of Alu exons primarily when the structure incorporates sequence from the exon itself , but also when stable structures are located in relative proximity to the exon . Secondary structure has been shown to impair exon recognition by affecting the accessibility of splice sites [8] , [9] , [11] , [12] , [30] . To examine whether sequestration of splice sites within secondary structures plays a role in the context of Alu exonizations , we used a measure indicating the probability that all bases in a motif are unpaired ( denoted probability unpaired or PU value ) [31] . Briefly , this measure indicates the probability that a motif , located within a longer sequence , is participating in a secondary structure . Higher values indicate that the motif is more likely to be single stranded and lower values indicate a greater likelihood of participating in a secondary structure ( see Materials and Methods ) . We assessed the single strandedness of the two most frequently selected 5′ss in the right arm located at positions 156 and 176 relative to the consensus ( also termed sites B and C [28] ) and the most frequently selected 5′ss of the left arm , located at position 291 ( see Figure 2A ) . We found that 5′ss selected in exonization events are characterized by significantly higher PU values than their non-exonizing counterparts , indicating that selected 5′ss have a lower tendency to participate in secondary structures ( see Figure 2E–G ) . We repeated this analysis for the two most frequently selected 3′ss in the right arm and the most frequently selected 3′ss in the left arm , but did not observe higher single-strandedness in the selected 3′ss with respect to their non-selected counterparts ( data not shown ) . However , this finding may also be attributed to the fact that all Alus , regardless of whether they undergo exonization or not , are characterized by relatively strong 3′ss , due to the poly-T stretch characterizing them ( see Discussion ) . See Text S1 for description of a control analysis . Intron-exon architecture has well-documented effects on splicing . Therefore , we compared the lengths of the Alu exons to their counterpart non-exons ( diagram in Figure 3A ) . We found that exons were ∼10 nt longer than their non-exonizing counterparts ( Figure 3C and 3D ) . Exons in the right arm of the AEx-R dataset were 112 nt long , on average , whereas non-exons were only 102 nt long in the No AEx dataset . The same trend was observed in the AEx-L dataset: Exons in the left arm of the AEx-L dataset were 88 nt long , whereas the non-exons in the No AEx group were 78 nt long . In both cases , the differences were highly statistically significant ( see Text S1 ) . This indicates that increased exon length is an advantage in terms of exonization of Alu elements . Analyzing the lengths of the flanking introns , we found that introns flanking Alu exons were almost 50% shorter than those flanking their non-exonizing counterparts . Introns upstream of Alu exons in the AEx-R or AEx-L dataset were 7 , 216 and 9 , 497 nt long , respectively , on average ( Figure 3B ) , but 14 , 458 nt long upstream of the non-exons in the No AEx group . These differences were highly significant ( No AEx vs . AEx-R p = 1 . 38E−13 , No AEx vs . AEx-L p = 0 . 0047 ) . Highly significant findings were observed in the downstream intron as well . These introns were 7 , 844 and 9 , 210 nt long for exons in the AEx-R and AEx-L dataset , respectively , but 14 , 808 nt long for Alus in the No AEx dataset ( Figure 3E ) . Taken together , these results indicate that recognition of exons by the splicing machinery correlates positively with exon length but negatively with intron length , yielding insight on the constraints and the mechanism of the splicing machinery ( see Discussion ) . Based on both biologic and bioinformatic methodologies , datasets of exonic splicing enhancers ( ESEs ) and silencers ( ESSs ) have been compiled; these sequences are believed to increase or decrease , respectively , the spliceosome's ability to recognize exons . Indeed , exons were found to be enriched in ESRs with respect to pseudo-exons or exons [32]–[34] . Thus , our next step was to determine the densities of ESEs and ESSs in exons and non-exons . We made use of four datasets of exonic splicing regulators ( ESRs ) : the groups of SR-protein binding sites in ESEfinder [35] , the dataset of ESEs from Fairbrother et al . [36] , the exonic splicing regulatory sequences compiled by Goren et al . that consists mostly of ESEs [37] , and the ESS dataset compiled by Wang et al . [38] . For each exon ( or non-exon ) in the two Alu arms ( Figure 4A ) in the three core and two control datasets , we calculated the ESR density for the four groups of ESRs . The ESR density was calculated as the total number of nucleotides within an exon that overlap with motifs from a given dataset divided by the length of the exon . We found that Alu exons showed a marked tendency for enrichment in ESEs and depletion in ESSs with respect to their non-exonizing counterparts . Right arm Alu exons had significantly higher densities of ESEfinder ESEs than their counterparts in the No AEx group ( Figure 4B , p = 0 . 00007 ) and higher densities of ESEs from Fairbrother et al . ( Figure 4C , p = 0 . 00009 ) . Higher densities were also observed in terms of ESEs found in Goren et al . ( Figure 4D ) , whereas slightly lower densities were observed for the ESSs of Wang et al ( Figure 4E ) ; However , the trends for the latter two datasets were not statistically significant . In the left arms , similar tendencies were observed: Exons originating from this arm were highly enriched in ESEs of Goren et al . ( Figure 4H , p = 0 . 0001 ) and depleted in ESSs of Wang et al . ( Figure 4I , p = 0 . 0003 ) . They also tended to be enriched in ESEs of Fairbrother et al . ( Figure 4G ) , although this was not significant ( p = 0 . 12 ) ; and in this arm no differences were found in terms of ESEs of ESEfinder ( Figure 4F , p = 0 . 72 ) . To summarize , in all cases in which significant differences were observed , these differences reflect an increase in ESE densities in parallel with a decrease in ESS densities in exons relative to non-exons . Since the splicing machinery is able to differentiate between exonizing and non-exonizing Alus , we were interested in discovering whether the features identified here can give rise to such precise classification . Toward these aims , we used Support Vector Machine ( SVM ) machine learning , which has shown excellent empirical performance in a wide range of applications in science , medicine , engineering , and bioinformatics [39] . We created two classifiers: One discriminating between non-exonizing Alus and Alus exonizing from the right arm and one discriminating between non-exonizing Alus and Alus exonizing from the left arm . Receiver-operator curves ( ROC curves ) were used to test performance . Briefly , ROC curves measure the tradeoff between sensitivity and specificity of a given classification . A perfect classification with 100% sensitivity and 100% specificity will yield an area under the curve ( AUC ) of 1 , whereas a random classification will yield an AUC of 0 . 5 ( see Materials and Methods for complete details of the SVM protocol used ) . 14 features were selected for the machine learning . These were divided into 5 clusters: 5′ss strength ( 1 feature: 5′ss score ) , 3′ss strength ( 1 feature: 3′ss score ) , secondary structure ( 5 features: z-scores for the stability of secondary structure of the entire Alu and of each of the two Alu arms , PU values of the 5′ss , and PU values of the 3′ss ) , exon-intron architecture ( 3 features: lengths of upstream intron , of Alu exon , and of downstream intron ) , and ESRs ( 4 features: density in terms of each of the 4 groups of ESRs ) . Based on the above-described features , we were able to achieve a high degree of classification between exonizing and non-exonizing Alus . Figure 5A presents the ROC curves and AUC values for the classification between Alus exonizing from the right arm and non-exonizing Alus and Figure 5B presents these values for the classification between the Alus exonizing from the left arm and the non-exonizing ones . The AUC values of ∼0 . 91 , demonstrate that our features achieve a high degree of accuracy in discriminating between true exons and non-exons , thus mimicking the role of the splicing machinery . If selection of an Alu exon is indeed determined by this set of features , then this same set of features may well also determine the inclusion level of an Alu exon . A “strong” set of features will lead to a high selection rate by the spliceosome , and hence to high inclusion levels , whereas “weaker” features may lead to a more reduced selection rate by the spliceosome and to lower inclusion levels . Indeed , we found a positive , highly significant correlation between probabilities of exonization based on the SVM model and between inclusion levels of exons based on EST data in the case of right arm Alu exons ( Pearson , r = 0 . 28 , p = 6 . 35e−07 ) . For the sake of comparison , the correlation between 5′ss scores and inclusion levels is considerably lower and less significant ( r = 0 . 15 , p = 0 . 007 ) . Thus , although the computational model was explicitly trained on the basis of a dichotomous input ( Alus were labeled either as exonizing or as non-exonizing ) , the model managed to capture the more stochastic nature of the spliceosomal recognition of exons . A positive correlation existed in the left arm as well , but this correlation was not significant presumably due to the fewer number of Alus in the AEx-L dataset . Although our model was trained on Alus , and specifically on comparing non-exonizing Alus to mostly alternatively recognized Alus , we reasoned that the same set of features which make the difference between a non-recognized and an alternatively-recognized Alu exon might also make the difference between an alternatively recognized exon and a constitutively recognized one . We therefore applied the SVM model to datasets of constitutive and cassette exons . For this purpose , we generated a dataset of 55 , 037 constitutive and 3 , 040 cassette exons based on EST-data ( see Materials and Methods ) . For each of these exons , we first extracted all above-described features , and then applied the SVM model to them . Our model classified constitutive and alternative exons as different in a highly statistically significant manner . The mean probability of undergoing exonization , provided by the logistic regression transformed SVM model , was 73% for the constitutive exons , but only 60% for the alternative ones ( Mann-Whitney , p<2 . 2e−16 ) . In addition , 82% of the constitutive exons were classified as “exonizing” , in comparison to only 63% of the alternative exons . These results demonstrate that the features learned by the SVM model are relevant for exonization in general , and control not only the shift of non-exons to alternative ones , but also of alternative exons to constitutive ones . Finally , we were interested in assessing the importance of different features in allowing correct discrimination between exonizing and non-exonizing elements . For this purpose , we used ΔAUC to measure the contribution of each feature cluster . This measure compares the performance of the classification with and without each cluster of features , with greater differences indicating greater contribution of a given cluster of features to precise classification . The feature with the highest contribution , both in the right arm ( Figure 5C ) and in the left arm ( Figure 5D ) , was the strength of the 5′ss , in concordance with previous bioinformatic findings [26] . However , much information is included in the other features as well . The second most important feature both in the left and in the right arm was exon-intron architecture . Secondary structure and the 3′ss had a comparable contribution in the right and left arm . Despite the differences in terms of ESR densities between the different datasets , this feature cluster had a negligent contribution to classification in the right arm , and a slightly higher one in the left arm . Using a mutual information based metric to measure the contribution of the different features , yielded similar , consistent results ( see Text S1 ) .
In this study , we sought to determine how the splicing machinery distinguishes true exons from non-exons . Alu exonization provided a powerful model for approaching this question . Exonizing Alus have retained high sequence similarity to their non-exonizing counterparts but are perceived differently by the splicing machinery . Past studies have emphasized mainly the splice sites , but our results indicate the importance additional features that lead to exonization . These features , which include splicing signals ( splice sites and ESRs ) , exon-intron architecture , and secondary structural features , achieved a high degree of classification between true Alu exons and non-exons , demonstrating the biological relevance of these layers in determining and controlling exonization events . Perhaps the most interesting result to emerge from this study is that secondary structure is critical for exon recognition . It has been assumed that pre-RNA is coated in vivo by proteins [10] and that these RNA-protein interactions either prevent pre-mRNAs from folding into stable secondary structures [40] or provide pre-mRNAs with a limited time span for folding [41] . However , an increasing number of studies are finding that secondary structure plays a crucial role in the regulation of splicing . Secondary structures involving entire exons ( e . g . , [5]–[7] ) , the splice sites only ( e . g . , [8] , [11] , [12] ) , or specific regulatory elements [42] , [43] were shown to be involved in the regulation of alternative splicing . Hiller et al . [14] recently found that regulatory elements within their natural pre-mRNA context were significantly more single stranded than controls . Our current study puts these findings into a broad context , and provides bioinformatic evidence for the notion that the structural context of splicing motifs is part of the splicing code . Such a structure , as we have shown , is detrimental for exonization in general , and specifically if it overlaps the 5′ss . Several intriguing observations can be made when merging our results based on the exonizing and non-exonizing Alus with those of the alternative and constitutive datasets . In terms of inclusion level , these four groups form a continuum , with non-exonizing Alus having a 0% inclusion level , exonizing Alus having a mean inclusion level of 10% , cassette exons having a mean inclusion level of 25% , and constitutive exons being included in 100% of the cases . Gradual changes when moving from non-exonizing Alus , to exonizing Alus , to alternative exons , to constitutive ones are observed in several additional features: The strength of the 5′ss gradually increases from non-exonizing Alus to constitutive exons , the strength of the secondary structure gradually decreases , lengths of the upstream and downstream introns gradually decrease while length of the exons gradually increase ( see Figure 6 for detailed values ) . These gradual changes are all coherent in biological terms: Stronger 5′ splice sites allow higher affinity of binding between the spliceosomal snRNAs and the 5′ss , and have well documented effects in increasing exon selection [28] , [44]; stronger secondary structure can sequester binding sites of spliceosomal components; And it has been previously shown that longer flanking introns profoundly increase the likelihood that an exon is alternatively spliced [4] , and that alternative exons tend to be shorter than their constitutive counterparts ( reviewed by [16] ) , presumably due to spliceosomal constraints . In addition , our finding that selective constraints are simultaneously applied both on the lengths of the exons and of their flanking introns suggests that the exon and its flanking introns are recognized , to some extent , as a unit . This challenges the more traditional exon-definition and intron-definition models [3] , [45] , according to which either the exon , or its flanking introns , but not both , are recognized by the splicing machinery . Notably , in our search for features differentiating between exonizing and non-exonizing Alus , we focused only on features which can potentially be mechanistically employed by the splicing machinery to differentiate between exons and introns . For this reason , we did not use phylogenetic conservation , nor the age of the Alu exons , nor the location of the exonization event ( CDS vs . UTR ) as features . Although these features are informative as well ( see Text S1 , and [32] ) , and thus may potentially boost the performance of our classifier , these cannot be directly sensed by the spliceosome . Rather , these elements reflect the evolutionary pressures to which an exonizing Alu element is subjected . In our study we found that introns flanking exonizing Alus are dramatically shorter than the introns flanking their non-exonizing counterparts . These results appear to contradict recent results [46] according to which there is a tendency for new exons to form within longer introns . However , two points must be borne in mind in this context: First , the introns flanking exonizing Alus are longer than average introns , and thus our results are consistent with the above study in that exonizations occur in longer introns . Second , our findings may reflect an upper bound in terms of intron length within which exonization optimally occurs , and introns longer than a certain threshold may cease to be good candidates for exonization . Our results indicate that the Alu-trained model could be applied to a more general context of alternative and constitutive exons , where it yielded coherent results . This does not , however , imply that all findings made in the context of Alus can be directly extrapolated to exons in general . For Alu sequences , we found the 5′ss to be the most informative feature for correctly predicting exonization events , in agreement with previous findings [26] , [28] . We found , however , that the 3′ss , which was also found to play a major role in exonization [24] , is less critical . This finding may not necessarily hold for all exons . The relatively low contribution of the 3′ss to Alu exonization may reflect the general tendency of Alus to have relatively strong splice signals at their 3′ end , regardless of whether they undergo exonization or not . This is since the poly-T track , present in all Alus in the antisense orientation , serves as a strong polypyrimidine tract [24] , [47] . On the other hand , our results regarding the importance of ESRs are consistent with several previous studies that have found exons to be enriched in ESRs with respect to pseudo-exons , more poorly recognized exons , and introns [32]–[34] . Thus , while the importance of different features may vary from one exon to another , our results provide a general understanding of the features impacting on exon recognition . It is noteworthy , that the majority of Alu exonization events in our two exonizing datasets presumably reflect either errors of the splicing machinery or newly born exons , which presumably do not give rise to functional proteins ( see also [48] ) . This is indicated by the low inclusion level of the Alu exons , averaging 13% and 10% in the AEx-R and AEx-L groups , respectively . In addition , the symmetry of the Alu exons ( i . e . , divisibility-by-three ) , at least in the AEx-R dataset , is very low: Only 23% of the exons are symmetric ( in the AEx-L dataset 55% of the Alus are symmetric ) . Thus , the majority of Alus in this dataset insert a frame-shift mutation . These numbers contrast with the 73% symmetry found in alternative events conserved between human and mouse [49] . However , since our objective in this research was to understand the requirements of the spliceosome , the potential function of the transcript is irrelevant . Moreover , newly born alternatively spliced Alu exons are the raw materials for future evolution: Given the right conditions and time , further mutations might generate a functional reading frame . The features identified here provided good , but not perfect , classification using machine learning . A number of factors underlie the non-perfect classification: For example , EST data is very noisy and far from providing a comprehensive coverage of all genes in all tissues [50] . Therefore , many Alus categorized as non-exonizing may , in fact , undergo exonization in certain tissues . Moreover , the features uncovered here may well not be exhaustive . Finally , as suggested by the correlation between the strength of predictions and the inclusion level , the alternative splicing pattern of the Alu exons may imply that the spliceosome itself does not perfectly recognize the exons . In this sense , the non-perfect classification of the machine learning model may reflect , to some extent , the non-perfect selection of the splicing machinery , giving rise to alternative events .
We compiled a dataset of intronic Alus in the antisense orientation that do not undergo exonization and datasets of Alus that are exonized in their right arms and left arms . We used a similar , but improved , procedure to the one described in [28] . We retrieved all human intronic Alus in the antisense orientation by querying the TranspoGene database [51] . Using the needle application [52] , we next performed pairwise , global alignments between the Alu sequences and the Alu-Jo consensus sequence which was downloaded from RepBase [53] ( http://www . girinst . org/ ) . Since we desired only Alus sharing a ‘reasonable’ degree of similarity which would ensure a common basis for comparison , we next filtered out all Alus with over 40 indels relative to the Alu consensus sequence; this cutoff was set empirically . Finally , we filtered out all redundant entries based on overlapping genomic coordinates . We next identified all cases in which EST evidence ( based on the TranspoGene query ) supported exonization from the right arms and from the left arms . These Alus formed the initial AEx-R and AEx-L datasets . To form the No AEx group , we began with all Alus lacking any evidence of exonization and retained only those Alus overlapped by ≥20 ESTs , based on the hg17 ‘Spliced ESTs’ table downloaded from the UCSC website ( http://genome . ucsc . edu/ ) . To identify optimal exon boundaries ( i . e . flanking 3′ and 5′ splice sites ) in the left arms of the AEx-R dataset , in the right arm of the AEx-L dataset , and in both arms of the No AEx dataset , we first characterized the position windows in which 5′ and 3′ splice sites tended to be located , in the right and left arms of sequences in the AEx-R and AEx-L datasets , respectively , as in [28] . The 3′ss was defined as the 15-nucleotide ( nt ) sequence covering the 14 last intronic nucleotides and the first exonic nucleotide and the 5′ss was defined as a 9-nt sequence covering the 3 terminal exonic nucleotides and the first 6 intronic nucleotides . We found that 97% of the 3′ss in the right arm of the AEx-R dataset were located upstream of position 58 ( relative to the consensus ) and that >98% of the 5′ss in the right arm of the AEx-R were located between positions 105 and 181 . Similarly , >95% of the 3′ss were located between position 181 and 204 and all 5′ splice sites in the left arm of the AEx-L group were downstream of position 249 . We next searched for the highest scoring splicing signals within the relevant positional windows of the left arm of the AEx-R group , the right arm of the AEx-L group , and both arms of the No AEx group . Alus lacking a minimal potential splice site in either arm , defined as an ‘AG’ for the 3′ss and a GT/C for the 5′ss , were filtered out . Splice site scores were determined by first calculating log-odd scores based on position specific scoring matrices ( PSSMs ) of the relevant splicing signal and subsequently rescaling them to lie between 0 and 100 , as described in [54] . The 5′ss PSSM , spanning 3 exonic and 6 exonic positions , was derived from the Analyzer Splice Tool webserver ( http://ast . bioinfo . tau . ac . il/SpliceSiteFrame . htm ) , and the 3′ss PSSM , spanning 14 intronic and 1 exonic position , was derived from [55] . The two control datasets , AEx-R ( c ) and AEx-L ( c ) were created based on the same set of Alus as the AEx-R and AEx-L groups , respectively , but by defining exon borders in both arms based on the computational prediction rather than on EST evidence . The predicted free energy of the ensemble of all secondary structures of a sequence was obtained via the RNAfold application in the Vienna RNA Package [29] , [56] . However , these measures are highly sensitive to sequence length and to dinucleotide composition [57] . To overcome these biases , we used DiShuffle [58] to generate 50 random sequences from each original sequence sharing its length and dinucleotide composition . The Z-scores were calculated as the difference between the free energy of the original sequence and the mean partition function of the 50 randomized sequences , divided by the standard deviation of the partition functions of the randomized sequences . PU values represent the probability that all bases in a motif are unpaired ( denoted as probability unpaired or PU value ) [14] . These motifs were calculated as in [31] . 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 folding temperature . Eall and Eunpaired were computed using the partition function version of RNAfold [29] . For Eunpaired , we assured that the region a to b was unpaired by applying additional constraints ( RNAfold parameter-C ) . To reduce the dependency on a single fixed context length , we considered all symmetrical context lengths from 11 up to 31 nt upstream and downstream of the splicing motif in increments of 5 , similar to [14] . Thus , for a 5′ss motif of length 9 nt , we considered sequences with a total length of 31 nt ( for context length of 11 , 2×11+9 = 31 ) , 41 nt ( for context length 16 ) , 51 nt , 61 nt , and 71 nt ( for context length of 31 ) . We computed the PU value of the splicing motif for each of these context lengths and averaged them . To determine the length of the introns flanking the Alu exons/non-exons , we downloaded the UCSC hg18 Known Genes track , creating a separate record for each exon . We used the LiftOver application ( available in http://hgdownload . cse . ucsc . edu/downloads . html ) to convert the Alu start and end coordinates from hg17 ( used by TranspoGene ) to hg18 . For each Alu in each of the datasets , we identified the most proximal exon upstream and downstream of the Alu , based on which we calculated lengths of the introns flanking the Alu elements . Our aim was to build two classifiers discriminating between Alus undergoing and not undergoing exonization: one between the Alus undergoing exonization in the right arm and the non-exonizing ones and one between Alus undergoing exonization in the left arm and the non-exonizing ones . Since we had two datasets representative of Alus exonizing from the right ( AEx-R and AEx-R ( p ) ) and two from the left ( AEx-L and AEx-L ( p ) ) , in practice we built four classifiers , with each classifier distinguishing between the No AEx group and one of the above four groups . The machine learning approach we decided to use was support vector machine ( SVM ) . We made use of the e1071 package [59] , which provides an interface to LIBSVM [60] in R statistical package [61] . All variables were normalized prior to the machine learning process , to zero mean and unit variance . The variables of intron length and PU values were first log-transformed , as well . The difference in several orders of magnitude between the size of the datasets of Alus undergoing and not undergoing exonization causes the instance of the former to ‘drown’ within the latter . In our machine learning , we therefore maintained a 3∶1 ratio between non-exonizing and exonizing Alus by randomly selecting three Alus from within the non-exonizing dataset for each Alu in the exonizing dataset . SVM training involves fixing several hyper-parameters , which have a crucial effect on the performance of the trained classifier [39] . To identify an optimal hyper-parameter set , we used 10-fold cross-validation on the training set and performed a grid search with linear , polynomial , and Gaussian kernels and with a range of cost and gamma values . For this purpose , we used the tune ( ) function in the e1071 package . We found that in the majority of cases the SVM performed best with a linear kernel and a cost factor of 1 . Evaluation of the SVM prediction was achieved by implementing a 10-fold stratified cross-validation procedure ( i . e . , in each run maintaining the 3∶1 ratio between the training and test sets ) using the area under the ROC curve ( AUC ) as a global performance measure . We performed 10 cross-validation runs ( in each such run using a different set of randomly selected non-exonizing Alus ) and the 100 ROC curves from these runs were averaged and displayed in Figure 6 via the ROCR package [62] . The mean AUC in these runs was calculated as well , as an overall performance measure . Logistic regression fitted to the decision values of the SVM classifier was applied using the probability = TRUE option in the svm ( ) function . Unless explicitly stated otherwise , the hypothesis that a factor distributed equally across different groups was tested using the non-parametric Kruskal-Wallis one way analysis of variance test . As post-hoc tests , we then performed Mann-Whitney tests between each pair of groups . The results for these tests are all presented in Text S1 . Coordinates of human ( hg18 ) exons based on the Refseq track and coordinates of spliced EST alignments were downloaded from the UCSC genome browser ( http://genome . ucsc . edu/ ) . For an EST to support exon inclusion , we demanded that the exon be either fully included in the alignment of the EST sequence , or that at least 50 nt of the exon and either of its two splicing signals form part of the alignment . Alignment gaps of less than 8 nt were ignored , as in the UCSC visualization defaults . An EST was defined as supporting exon skipping if no alignment between them was observed , and if the EST was defined as supporting the two flanking exons . For the constitutive exons , we selected all exons whose inclusion was supported by at least 20 ESTs and lacking any ESTs supporting exon skipping , whereas for the alternative dataset we selected all exons with at least 5 ESTs supporting inclusion and 5 ESTs supporting skipping . As a final step , the features described in the manuscript were extracted for each of the exons in the two datasets , with the exception of secondary structure of the opposite Alu arm and within the entire Alu , since these features cannot be applied to exons in general .
|
A typical human gene consists of 9 exons around 150 nucleotides in length , separated by introns that are ∼3 , 000 nucleotides long . The challenge of the splicing machinery is to precisely identify and ligate the exons , while removing the introns . We aimed to understand how the splicing machinery meets this momentous challenge , based on Alu exonization events . Alus are transposable elements , of which approximately one million copies exist in the human genome , a large portion of which within introns . Throughout evolution , some intronic Alus accumulated mutations and became recognized by the splicing machinery as exons , a process termed exonization . Such Alus remain highly similar to their non-exonizing counterparts but are perceived as different by the splicing machinery . By comparing exonizing Alus to their non-exonizing counterparts , we were able to identify numerous features in which they differ and which presumably lead to the recognition only of the former by the splicing machinery . Our findings reveal insights regarding the role of local RNA secondary structures , exon–intron architecture constraints , and splicing regulatory signals . We integrated these features in a computational model , which was able to successfully mimic the function of the splicing machinery and discriminate between true Alu exons and their intronic counterparts , highlighting the functional importance of these features .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"computational",
"biology/alternative",
"splicing"
] |
2009
|
Alu Exonization Events Reveal Features Required for Precise Recognition of Exons by the Splicing Machinery
|
Ribosomal proteins are essential to life . While the functions of ribosomal protein-encoding genes ( RPGs ) are highly conserved , the evolution of their regulatory mechanisms is remarkably dynamic . In Saccharomyces cerevisiae , RPGs are unusual in that they are commonly present as two highly similar gene copies and in that they are over-represented among intron-containing genes . To investigate the role of introns in the regulation of RPG expression , we constructed 16 S . cerevisiae strains with precise deletions of RPG introns . We found that several yeast introns function to repress rather than to increase steady-state mRNA levels . Among these , the RPS9A and RPS9B introns were required for cross-regulation of the two paralogous gene copies , which is consistent with the duplication of an autoregulatory circuit . To test for similar intron function in animals , we performed an experimental test and comparative analyses for autoregulation among distantly related animal RPS9 orthologs . Overexpression of an exogenous RpS9 copy in Drosophila melanogaster S2 cells induced alternative splicing and degradation of the endogenous copy by nonsense-mediated decay ( NMD ) . Also , analysis of expressed sequence tag data from distantly related animals , including Homo sapiens and Ciona intestinalis , revealed diverse alternatively-spliced RPS9 isoforms predicted to elicit NMD . We propose that multiple forms of splicing regulation among RPS9 orthologs from various eukaryotes operate analogously to translational repression of the alpha operon by S4 , the distant prokaryotic ortholog . Thus , RPS9 orthologs appear to have independently evolved variations on a fundamental autoregulatory circuit .
The evolution and function of spliceosomal introns are among the largest unsolved mysteries of eukaryotic genomes . Pronounced differences in intron evolution between lineages and between introns within the same lineage provide insight into 1 ) the selective and mutational forces governing intron evolution and 2 ) the potential roles of introns in gene function . Here we study the case of ribosomal protein genes ( RPGs ) in the model yeast Saccharomyces cerevisiae . RPGs are highly over-represented among intron-containing genes ( 69% of RPGs contain introns compared to ∼5% of non-RPGs ) , which has been suggested to reflect ongoing selection for introns that provide one or more functions in gene expression [1] , [2] . However , two major facets of this hypothesis — the action of selection ( intron evolution ) , and the source of this selection ( intron function ) — remain unknown . First , biased intron loss has not been specifically tested within hemiascomycetous yeasts ( S . cerevisiae and relatives ) . And second , the effect of intron loss on RPG expression remains uncertain . RPG expression is remarkable both in terms of synthesis rate and control [3]; thus , RPG introns may function to promote these aspects of gene expression . One proposal predicts that RPG introns function to promote high levels of expression . Consistent with this view , intron-containing genes , including RPGs , produce some of the highest transcript and protein abundances in S . cerevisiae [4] . However , the requirement for introns to enhance RPG expression has not been directly tested . In addition to the above , two other proposals predict that RPG introns function by providing an opportunity for splicing regulation . One possibility is that introns provide rapid regulation in response to environmental stress , as suggested by splicing inhibition of RPG pre-mRNAs in response to amino acid starvation [5] . Another possibility is that introns provide an opportunity to fine-tune gene expression through autoregulation . For example , negative feedback control of RPL30 and RPS14B expression is achieved through the binding of their respective protein products to RNA structures within their own unspliced transcripts , thereby regulating splicing [6] , [7] . Interestingly , nearly all the ribosomal proteins of Escherichia coli are regulated by key ribosomal proteins in an analogous manner; for example , bacterial S4 directly binds its own mRNA to repress the translation of itself and three other RPGs [8] , [9] . Given that the majority of S . cerevisiae RPGs contain introns , intron-dependent autoregulation may be more common than previously appreciated . We report the first direct tests of both the action and the source of selection on RPG introns . First , we used comparative genomics to show that RPG introns have been preferentially retained following whole genome duplication ( WGD ) , indicating ongoing selection for retention of RPG introns . Second , we generated S . cerevisiae strains harboring precise deletions of 16 RPG introns to distinguish between selective hypotheses . We found that RPG introns generally reduce gene expression , suggesting that RPG introns allow for splicing regulation rather than promoting high levels of expression . In particular , we identified intron-dependent cross-regulation between the RPS9A and RPS9B genes , which both encode ribosomal protein S9 ( S9 ) . Finally , overexpression of RpS9 in D . melanogaster S2 cells , and analysis of available EST sequences , suggest that autoregulation of RPS9 orthologs may involve different forms of splicing regulation between species , but also appears to be widespread across disparate lineages .
Introns are over-represented in the RPGs of both Candida albicans and S . cerevisiae [1] , [2] . While this shared over-representation may reflect selection pressure to maintain RPG introns prior to the divergence of these two species from a common ancestor , it may also reflect the action of selection in more recent history , since their divergence from a common ancestor . This distinction is important , since selection pressure to maintain RPG introns in more recent history is more likely to be relevant to the biology of S . cerevisiae . To determine if RPGs have resisted intron loss compared to other genes since the divergence of C . albicans and S . cerevisiae ( ∼200–800 million years ago [10] ) , we assessed the fates of S . cerevisiae introns in paralogs ( a . k . a . gene pairs ) that were duplicated ∼100 million years ago by whole-genome duplication ( WGD ) [11] . To determine the fates of introns after genome duplication , we took advantage of the well-annotated genome of S . cerevisiae , which has been exhaustively searched for introns [12] , [13] . With these annotations , we identified 121 intron-containing genes among 554 WGD-derived gene pairs obtained from Yeast Gene Order Browser [14] . Assuming that intron loss has largely dominated intron evolution in hemiascomycetous yeast species [15] , we inferred intron loss if one of the WGD-derived gene copies had fewer introns than the other . Using this criterion , we calculated the number of apparent intron losses in RPG pairs compared to all other gene pairs . Strikingly , this simple accounting revealed that 16 of 23 non-RPG pairs have a gene with fewer introns than its copy , whereas none of the 46 RPG pairs did . Nonetheless , this analysis ignores intron losses that occurred independently in both gene copies and assumes that intron gain did not occur . To better assess whether WGD-derived RPG pairs have been biased for either intron gain or loss ( including losses in both gene copies ) , we reconstructed the hypothetical intron distribution of the pre-WGD ancestor that existed prior to the WGD event . For each of the 554 S . cerevisiae duplicated gene pairs , we assigned the presence or absence of an intron in the hypothetical pre-WGD ancestral ortholog based on intron annotations and predictions from the genomes of the pre-WGD ( so-called protoploid ) species ( C . albicans , Lachancea waltii , L . thermotolerans , L . kluyveri , Eremothecium gossypii , Kluyveromyces lactis , and Zygosaccharomyces rouxii ) and the genomes of the post-WGD species ( Vanderwaltozyma polyspora , Naumovia castellii , C . glabrata , and S . bayanus ) . A complete list of intron predictions and annotations can be found in Table S1 . Our analysis revealed 73 intron-containing genes that were likely present in the pre-WGD ancestor from which the duplicated gene pairs in S . cerevisiae were descended ( Figure 1A ) . Based on this hypothetical intron distribution of the pre-WGD ancestor , we inferred the number of S . cerevisiae WGD-derived gene pairs that have gained or lost an intron for each post-WGD gene pair ( Figure 1B ) . From this improved analysis , we identified 5 S . cerevisiae non-RPG pairs that appear to have independently lost introns from both gene copies after gene duplication . This was in addition to 14 non-RPG pairs in which one of two introns were lost ( Figure 1B , right and middle columns , respectively ) . Once again , we inferred no intron losses in S . cerevisiae RPG pairs ( Figure 1B , left column ) . Thus , RPG introns appear to have been biased against loss in the lineage leading to S . cerevisiae during the last ∼100 million years . Next , we asked whether intron gains contributed to the bias for introns in S . cerevisiae RPGs . For a given S . cerevisiae gene , we inferred that an intron was gained if introns were absent in both the pre-WGD ancestor and the majority of post-WGD orthologous gene pairs . Using this criterion , we did not infer intron gains in any of the S . cerevisiae RPGs . On the other hand , two introns in non-RPGs ( i . e . USV1 and BMH2 ) have possibly been gained in the S . cerevisiae lineage ( Table S1 ) ; however , since both of these introns are located in the 5′ UTR and are not well annotated in other species , it is therefore difficult to be confident of this conclusion . Taken together , the bias for introns in S . cerevisiae RPG pairs appears to have been dominated not by intron gains in RPGs , but by intron losses in non-RPGs . Having found a bias against RPG intron loss , we sought to determine if RPG introns have a function in gene expression . To mimic the effect of RPG intron loss , we created S . cerevisiae mutant intron deletion strains ( henceforth denoted as Δi ) . Each Δi mutant was created with a precise deletion of a single RPG intron , such that only an intronless copy of the gene remained at the endogenous locus ( See Methods ) . Because RPGs are among the most highly expressed genes in the genome , we tested the model that introns are required in cis for high levels of gene expression by assessing the expression profiles of 16 Δi mutants compared to a wild-type strain . We also considered the possibility that Δi mutations may affect other genes in trans , in particular , the WGD-derived gene copies of RPG pairs . To measure changes in expression of the gene from which an intron was deleted ( in addition to 124 RPG and 911 non-RPG features ) we used custom splicing-sensitive microarrays designed to detect pre- , mature , and total mRNA species ( using intron , junction , and exon probes , respectively [16] ) . To assess the effect of Δi mutation on gene expression , we plotted the expression change for the intronless gene ( Figure 2A , red lines ) compared to all the other genes on the microarray ( Figure 2A , boxplots ) . Thus , the most significant expression changes lie outside the whiskers of the boxplot and are , by definition , statistical outliers . Intron deletion mutations , as assessed by microarray , typically had only modest effects on gene expression ( Figure 2A , compare red lines to boxplots ) . Nonetheless , these effects were biased toward increased expression of the intronless gene ( 14 out of 16 ) , rather than decreased expression ( Figure 2A “up” and “down , ” respectively ) . Moreover , the four most substantial expression changes increased the expression of the intronless gene ( Figure 2A “outlier” ) . These data suggest that yeast introns are generally not required for the high expression levels of RPGs . Further , only a few genes showed substantial increases in expression , which suggests that splicing may be more inefficient for these genes than most other RPGs . We also sought to determine if any of the deleted introns were required for splicing regulation . As controls , we deleted the introns of RPS14A and RPS14B , as it has been known for some time that S14 binds to the RPS14B intron ( but not the RPS14A intron ) to inhibit splicing and to cause rapid degradation [7] , [17] . As expected , deletion of the RPS14B intron led to a substantial increase in its expression compared to the other genes on the microarray ( Figure 2A “outlier” ) , whereas deletion of the RPS14A intron had little effect on expression ( Figure 2A “down” ) . Thus , our microarrays have the sensitivity required to detect the derepression of RPS14B expression . An unexpected and novel finding is the substantial effect that Δi mutations have on the expression of the two gene copies encoding ribosomal protein S9 ( hereafter referred to as S9 ) . Our microarray experiments revealed that RPS9A and RPS9B Δi mutations increased the expression of the intronless genes ( Figure 2A “outlier” ) and also decreased the expression of the wild-type gene copies ( Figure 2B ) . We hypothesized that the decreased expression of the wild-type RPS9A and RPS9B genes was caused by decreased splicing efficiency due to negative feedback . Therefore , we tested whether Δi mutations caused an increase in the ratio of pre-mRNA to total mRNA of the wild-type gene copies by calculating the Intron Accumulation Index of these genes , which is a measure of inefficient splicing [18] . Of all the mutants tested by microarray , only RPS9A and RPS9B showed substantial increases in the Intron Accumulation Index compared to the other intron containing genes on the array ( Figure 2C , compare blue lines to boxplots ) . Taken together , these data suggest that the RPS9A and RPS9B genes require introns to repress their own expression . Further , derepression of RPS9A resulted in increased repression of RPS9B through splicing inhibition ( and vice versa ) , suggesting that these genes cross-regulate . Our custom microarray platform is precise; however , it lacks control probe sets needed for highly accurate quantification . As such , our microarrays “compress” fold-changes compared to equivalent determination by qPCR . To validate our most surprising observations , we assessed RPS9A and RPS9B expression by RT-qPCR . Importantly , we designed at least one qPCR primer to the 3′UTR in an effort to maximize specificity and to minimize artifacts caused by primer cross-hybridization to the other gene copy . As expected , qPCR measurements validated our microarray results for both RPS9A and RPS9B genes in the rps9bΔi and rps9bΔi mutants ( Figure 2D , second and third columns ) . In the case of the rps9aΔi mutant , Δi mutation was associated with a substantial increase ( >4-fold of wild-type ) in RPS9A expression and a modest decrease ( <2-fold of wild-type ) in RPS9B expression ( Figure 2D , second column ) . Conversely , in the rps9bΔi mutant , Δi mutation was associated with a modest increase ( <2-fold of wild-type ) in RPS9B expression and a substantial decrease ( >8-fold of wild-type ) in RPS9A expression ( Figure 2D , third column ) . Having validated the surprising effects of deleting the RPS9A and RPS9B introns , we hypothesized that the genes reciprocally cross-regulate through a shared negative feedback circuit . We made two strong predictions from this hypothesis: 1 ) deletion of both the RPS9A and RPS9B introns should eliminate cross-regulation , and therefore , derepress both gene copies and 2 ) the wild-type gene copy should compensate for a derepressed copy by an equal and opposite number of transcripts . First , to determine if repression of RPS9A expression in the rps9bΔi mutant required the RPS9A intron ( and vice versa ) , we created a double rps9a/bΔi mutant and tested the effect on expression by RT-qPCR . As predicted , both RPS9A and RPS9B were derepressed in the rps9a/bΔi mutant ( Figure 2D , fourth column ) . Second , we sought to determine if changes in the number of RPS9A transcripts were compensated by a nearly equal and opposite change in number of RPS9B transcripts . We first estimated the percent of transcripts encoding S9 contributed by the RPS9A and RPS9B genes ( 6% and 94% , respectively ) from a published RNA-seq data set from a wild-type strain [19] . In order to calculate the number of transcripts in each Δi mutant , we then simply multiplied the percent of transcripts encoding S9 ( as determined by RNA-seq ) by the relative change in expression ( as determined by qPCR ) for each Δi mutant . As predicted for the rps9aΔi mutant , a substantial relative increase in RPS9A expression mutant was nearly equally compensated by a modest relative decrease in RPS9B expression , such that the total number of transcripts encoding S9 was nearly unchanged ( Figure 2D and 2E , second column ) . In the rps9bΔi mutant , however , a modest relative increase in RPS9B expression mutant was only partially compensated at the expense of nearly all RPS9A transcripts ( Figure 2D and 2E , second column ) . In this case , it appears that RPS9A defied our prediction and presumably because its contribution to the total number of S9 transcripts was limiting . Lastly , deletion of both introns increased the total number of transcripts encoding S9 to 170% of wild-type levels ( Figure 2E , fourth column ) . Taken together , these data suggest that the RPS9A and RPS9B genes reciprocally cross-regulate by a common intron-dependent mechanism . Further , the large relative effects detected for RPS9A compared to RPS9B may simply reflect the large difference in expression level between the two gene copies . Reminiscent of the cross-regulation between S . cerevisiae RPS9A and RPS9B genes , several metazoan RPGs have been shown to autoregulate through alternative splicing coupled to NMD ( so-called “Regulated Unproductive Splicing and Translation” or RUST ) : a process in which the synthesis of productively-spliced mRNA is repressed in favor of unproductive mRNA isoforms encoding premature termination codons ( PTC+ ) [20]–[23] ( reviewed in [24] ) . While this process is conserved between distantly related eukaryotes , there is no known overlap between the genes regulated by RUST in yeast and metazoans to facilitate mechanistic comparisons . Intriguingly , an alternatively-spliced RpS9 PTC+ mRNA isoform was recently identified in Drosophila melanogaster [25] . Thus , we considered the possibility that other RPS9 orthologs autoregulate in a manner analogous to RPS9A and RPS9B cross-regulation . We hypothesized that D . melanogaster RpS9 expression is regulated in response to excess protein production by alternative splicing coupled to NMD . Therefore , we predicted that increased RpS9 expression would result in increased abundance of the PTC+ mRNA isoform . To test this hypothesis , we measured the affect of exogenous RpS9 overexpression and NMD inhibition on alternative splicing of RpS9 messages using RT-qPCR primer sets specific to endogenous RpS9 mRNA isoforms ( Figure 3A ) . We first verified that the previously identified RpS9 PTC+ isoform in S2 cells was degraded by NMD through RT-PCR amplification of RpS9 transcripts from S2 cells incubated with either of two dsRNAs targeting Upf1 ( Figure 3B ) . To then test the effect of increased RpS9 expression on the abundance of the PTC+ mRNA isoform , we exogenously overexpressed a cDNA copy of RpS9 ( Figure 3C ) . In S2 cells overexpressing RpS9 , we detected an increase in the abundance of the PTC-containing mRNA isoform ( Figure 3D , top panels , compare red and blue points ) and a decrease in the total RpS9 expression as compared to the empty vector control ( Figure 3D , bottom left panel , compare red and blue points ) . As expected , we observed a UPF1-dependent decrease in total endogenous RpS9 abundance in response to increased RpS9 expression ( Figure 3D , compare bottom left and right panels , blue points ) . Taken together , these results suggests that Drosophila RpS9 autoregulates by RUST , in which excess expression shifts the balance of alternative splicing from the synthesis of productively spliced messages towards the synthesis of unproductive RpS9 PTC+ messages that are selectively degraded by NMD . We hypothesized that RpS9 autoregulation had an important function and would thus be conserved in other animals . Further , we hypothesized that conserved RNA structures were involved in the cross-regulation of RPS9A and RPS9B in S . cerevisiae and the autoregulation of RpS9 in D . melanogaster , because E . coli S4 ( the bacterial ortholog ) , requires an RNA structure to autoregulate by translational repression . Therefore , we predicted that RPS9 orthologs would be associated with alternatively-spliced mRNA isoforms , conserved RNA structures , and PTCs . To identify such messages , we summarized expressed sequence tags ( ESTs ) data from diverse animals . Indeed , EST coverage extends outside exons and into introns , which support the existence of rare unspliced or alternatively-spliced transcripts ( <5% maximum coverage ) ( Figure 4 , gray bars ) . To identify ESTs that specifically support alternative splice site usage or cassette exon inclusion , we mapped putative EST exon-exon junctions that spanned both 5′ GT and 3′ AG splice sites ( Figure 4 , blue and red bars , respectively ) . With the exception of Petromyzon marinus , ESTs from various vertebrates ( e . g . H . sapiens , Rattus norvegicus , Xenopus tropicalis , Danio rerio , and Oryzias latipes ) reveal cassette exons that introduce PTCs from the last canonical intron ( Figure 4 and Figure S1 ) . P . marinus and D . melanogaster ESTs , on the other hand , reveal alternative 5′ splice sites that also introduce PTCs from a homologous intron ( Figure 4 and Figure S1 ) . Most intriguingly , Ciona intestinalis ESTs also support alternative 5′ splice site usage , but in a non-homologous intron compared to those of other animals ( Figure 4 ) . Thus , our surveys of animal ESTs suggest that animal RPS9 orthologs are often alternatively-spliced to utilize RUST . Further , the conservation of alternatively-spliced cassette exons within the last intron among distantly related vertebrates ( e . g . ∼400 million years between humans and fish [10] ) suggest that these isoforms are functional . Also consistent with function , PTC positions in RPS9 orthologs were associated with high nucleotide conservation ( Figure 4 ) . To determine if RPS9 orthologs were also associated with thermodynamically-stable and structurally-conserved RNA structures , we screened the gene bodies of RPS9 orthologs for statistically significant RNA structures using RNAz [26] on alignments obtained from the UCSC Genome Browser [27] . In order to examine both intronic and exonic sequences , we obtained sets of nucleotide alignments from closely-related groups of organisms: mammals , drosophilids , teleosts , and hemiascomycetous yeasts . Scanning RPS9 ortholog alignments in 400 bp windows , we identified predicted RNA structures ( P>0 . 9 ) , specifically within the last intron of mammalian , drosophilid , and teleost RPS9 orthologs , each overlapping with PTC positions ( Figure 4 , green lines and red octagons ) . Similarly , sequence alignments of RPS9 orthologs from hemiascomycetous yeasts also revealed predicted RNA structures specifically within the single yeast intron , which if unspliced , would introduce a PTC ( Figure 4 ) . Due to the lack of sequences similar to the C . intestinalis RPS9 gene corresponding to the PTC in its third intron , we did not test this region for conserved elements and predicted RNA structures . In any case , these data indicate the potential for autoregulation among distantly related RPS9 orthologs through the use of different forms of alternative splicing , perhaps through structured RNA elements .
The genes of S . cerevisiae , and hemiascomycetous yeasts in general , contain very few introns compared to other eukaryotes [28] , which is generally attributed to uncommonly high rates of intron loss within this lineage [15] . Previous observations that S . cerevisiae introns are biased for RPGs [1] , [2] , [29] , [30] and other highly expressed genes [4] have been cited as evidence that many S . cerevisiae RPG introns have one or more functions . Intriguingly , similar biases are also observed in the intron-poor genomes of Encephalitozoon cuniculi [31] , [32] and the nucleomorph of Guillardia theta [33] , suggesting that the bias against RPG intron loss is not limited to yeasts . By measuring the rates of intron loss among recently-duplicated genes , we confirm that an ongoing bias against RPG intron loss is apparent in the lineage leading to S . cerevisiae ( Figure 1 ) . Thus , the few remaining introns in S . cerevisiae may reflect biases in 1 ) the mechanisms of intron loss and/or 2 ) selection to keep important introns . In addition to previously-proposed functions of RPG introns ( see below ) , several lines of evidence suggest that the conservation of RPG introns is not merely a function of mutation rates . Reverse transcription-mediated intron loss is expected to preferentially remove 3′ end biased introns from highly-expressed genes [34] . First , intron biases for RPGs run counter to the expectation for intron loss among highly-expressed genes , since these transcripts would be more likely to be reverse transcribed ( as discussed in [32] ) . Second , the majority of S . cerevisiae intron losses observed here are not 3′ end biased; in fact , several introns were lost from the 5′ UTR ( e . g . GBP1 , NHP6A and ARF1 ) . Lastly , at least 21 RPG introns that are present in both the Lachantea and Saccharomyces clades appear to have been lost from Z . rouxii , indicating that species-specific RPG intron losses can occur , but have not done so in the lineage leading to S . cerevisiae ( Table S1 ) . Biased intron loss , therefore , may reflect species-specific selective pressure to retain functional introns . In many eukaryotes , the presence of large numbers of introns permit alternative splicing , which can be used to increase protein diversity [35] . However , the simple gene architectures of S . cerevisiae provide limited opportunity for the generation of multiple protein isoforms through alternative splicing ( although a few instances have been described [36] , [37] ) . Instead , S . cerevisiae RPG introns have been proposed to confer other functions , such as transcriptional enhancement [4] and splicing regulation . We tested these two hypotheses directly by deleting introns from 16 S . cerevisiae genes and assessing the effect on gene expression by microarray . Unlike intronless copies of some mammalian genes [38] , the expression of many RPGs were unaffected or even increased by deleting introns ( Figure 2A ) . Thus , the persistence of these introns may be due to selection for other intron functions , such as splicing regulation , perhaps in response to amino acid starvation [16] . Alternatively , our splicing microarray platform may not provide the sensitivity needed to confidently identify subtle , but potentially important , changes in expression levels . Nonetheless , we did observe large increases in gene expression for RPS14B , which is known to autoregulate through splicing inhibition [7] . Thus , it seems likely that this intron and the RPS9A and RPS9B introns are under additional selection pressure to maintain homeostasis of protein levels . Consistent with this view , regions within the RPS9A and RPS9B introns are highly conserved ( Figure 4 and Figure S2 ) , which strongly suggests that mutations within these introns have been detrimental to fitness during natural history . Therefore , the strong bias against RPG intron loss ( see above ) may reflect ongoing selection for splicing regulation . The propensity for RNA-binding proteins to utilize alternative splicing for the purpose of autoregulation has long been noted [39] and , in the case of RNA-binding proteins , is remarkably common [40]–[42] . To our knowledge , however , regulation at the level of splicing between organisms as evolutionarily distant as S . cerevisiae and humans is exceedingly rare . While autoregulation of RPGs by alternative splicing is common and can be conserved as distantly as worms and humans [20] , [21] , we find no evidence that other yeast RPGs ( i . e . RPL30 and RPS14 ) are regulated by splicing in both yeast and mammals ( Figure S3 ) . Interestingly , S9 orthologs in bacteria ( and possibly archaea ) are among a small class of RPGs that autoregulate by translational repression [43] , [44] . Thus , an intriguing notion is that S9 autoregulation is of particular importance to life or particularly likely to evolve . Presumably , autoregulation of S9 production would benefit the cell by reducing waste [3] and by preventing potentially harmful interactions with low-affinity targets [45] . Cross-regulation , such as between the RPS9A and RPS9B genes , has also been observed between multiple sets of paralogous splicing regulators , including hnRNPL and hnRNPLL [46] , as well as PTB , nPTB and ROD1 [47] . We speculate that these genes exemplify a straightforward principle of gene duplication and evolution: upon gene duplication , autoregulation would inherently become cross-regulation . As the paralogs diverge in abundance and/or protein function , this cross-regulation could become asymmetric ( Figure 5A ) . In theory , such asymmetric cross-regulation among RPG pairs may allow differential expression of functionally-distinct ribosomal proteins to produce a “ribosome code” [48] . What distinct functions are provided by RPS9A and RPS9B gene products remain to be seen . Interestingly , the RPS9A and RPS9B genes encode S9 proteins that differ primarily within a small C-terminal acidic patch that may be required for proper ribosomal disassociation [49] . Mutational analyses of these three differing amino acids are needed to definitively test whether S . cerevisiae utilizes differential expression of RPS9A and RPS9B genes to exploit functional differences in the proteins they encode . How does excess S9 regulate the splicing of the RPS9 orthologs ? One possibility is that S9 binds its own mRNA like bacterial S4 . A strong paradigm has been set by S14 and L30 in yeast and S26 and S13 in animals , in which these ribosomal proteins bind RNA structures present in their introns [6] , [7] , [22] , [23] . It seems likely that S9 might operate under the same paradigm . Intuitively , conserved RNA structures within the introns of RPS9 orthologs make for likely targets for S9 binding ( Figure 4 ) . However , we do not observe obvious similarities between these predicted structures and the E . coli S4 regulatory site , which forms a double pseudoknot [43] . E . coli S4 can also bind and regulate a Bacillis subtilis mRNA that contains a dissimilar pseudoknot structure [50] . Intriguingly , the conserved RNA structures in RPS9A and RPS9B also appear to have the potential to form a pseudoknot ( Figure S4 ) . Thus , it seems plausible that the putative RNA structures within yeast and animal introns may yet be binding sites for S9 despite considerable structural divergence . This , however , is mere speculation and in vitro binding assays are needed to determine if ribosomal protein S9 directly regulates its own expression in S . cerevisiae and other eukaryotes . If auto- and cross-regulation were indeed directly mediated by ribosomal protein S9 binding , then comparative biochemical studies using proteins and RNA sequences from different species could provide mechanistic detail to describe how S9 mediates the different forms of alternative splicing . Why are there so many forms of splicing regulation among RPS9 orthologs ? One possibility is that particular aspects of these forms are ancient and conserved , while others have evolved independently in different lineages . For example , the genetic circuits that specify the development of diverse animal forms ( e . g . eyes and limbs ) exemplify deep homology , where recent evolutionary innovations overlay a shared “genetic toolkit” [51] . By analogy , genetic circuits themselves ( in this case , autoregulation ) may share a common “biochemical toolkit” comprised of highly conserved biochemical processes ( e . g . RNA∶protein interactions ) , while independently evolving elaborations on these basic circuits . Thus , translational inhibition of the alpha-operon by S4-binding may represent just one of many possible forms of regulation accessible to the highly conserved S4 RNA-binding domain proteins found throughout cellular life . Alternative splicing in animals and regulated splicing in S . cerevisiae may be different elaborations on this autoregulatory circuit , perhaps mediated by different RNA structures within introns ( Figure 5B ) . Thus , we propose that the highly-conserved function of ribosomal protein S9 ( and RNA-binding proteins in general ) is one part of a biochemical toolkit that is frequently used and reused , as the fundamental autoregulatory circuit is maintained , elaborated and reinvented .
To estimate the propensity for intron loss among RPGs and non-RPGs , we compared annotated S . cerevisiae intron-containing genes and WGD-derived gene pairs . S . cerevisiae intron annotations were obtained from the Saccharomyces Genome Database ( http://www . yeastgenome . org/ ) on 7/20/2011 . WGD-derived gene pairs ( as inferred from genomic synteny; a . k . a . “Ohnologs” ) were obtained from the Yeast Gene Order Browser ( http://wolfe . gen . tcd . ie/ygob/ ) [14] . Because introns are commonly identified by gaps in BLAST-based homology searches , intron-containing genes with short first exons are commonly misannotated . To identify annotated introns upstream of an annotated gene , custom scripts written in R ( http://www . r-project . org/ ) were used to scan 800 bp upstream and 100 bp downstream of the ORF start site with a regular expression that recognizes >90% of S . cerevisiae introns by identifying the most common splice sites to minimize false positive matches . The regular expression matches sequences meeting the following criteria in order: 1 ) any one of the 4 most common of 5′ SSs , 2 ) an S1 length of at least 30 , 3 ) any one of the 5 most common branchsites , 4 ) an S2 length between 1 and 50 , and 5 ) any one of the 3 most common 3′ SS trinucleotides , which was formalized as: “ ( gtatgt|gtacgt|gtaagt|gtatga ) . {30 , } ? ( tactaac|gactaac|aactaac|tgctaac|cactaac ) . {1 , 50} ? [tca]ag” . The pre-WGD ancestor was inferred to contain an intron if the majority of available outgroup pre-WGD species orthologs ( C . albicans , L . waltii , L . thermotolerans , L . kluyveri , E . gossypii , K . lactis ) and 1 ) the Z . rouxii ortholog had an intron or 2 ) the majority of post-WGD species intron ( S . cerevisiae , S . bayanus , C . glabrata , N . castellii and V . polyspora ) gene pairs had at least one intron . In this manner , we distinguished independent intron gains and losses in Z . rouxii from intron gains and losses immediately after the WGD event . Intron deletion mutants were generated by a replacement strategy similar to a previously-described method for intron deletion [52] . Briefly , a PCR product amplified from the plasmid pJPS1232 ( generously provided by J . Staley , University of Chicago ) , which contains the CORE construct [53] fused to the I-SceI endonuclease site , using gene-specific primers containing exon 1 and exon 2 sequences that allow integration and subsequent intron deletion via homologous recombination . Transformed diploid cells ( yAP047 ) were incubated for 4 h at 30°C in the presence of 2% galactose to induce I-SceI endonuclease expression and precise deletion of the CORE cassette . Sporulated haploid cells were confirmed to harbor intron deletions by PCR . To ensure that a precise intron deletion was obtained without any additional mutations , the region surrounding the newly-created exon-exon junction ( at least 100 bp ) was PCR amplified and sequenced . Strains described in Figure 2 were also confirmed for Δi mutation by decreased microarray intron probe signal . Gene specific primers used for mutagenesis are detailed in Table S2 . Routine passaging of S2 cell cultures and RNAi depletion was performed as described [54] with the following modifications . Briefly , 1 µg/ml of dsRNA was incubated with 3 . 5E5 cells in 350 µl of media in 24-well plates . After 48 h incubation with dsRNA , cells were transfected with 0 . 2 µg plasmid with Effectene ( Quigen ) according to manufacturer's instructions . Cells were harvested with 1 ml TRIzol ( Invitrogen ) for analysis by RT-qPCR ( see below ) . Primers used to generate PCR products used for dsRNA synthesis ( Promega RiboMAX ) are described in Table S2 . To analyze the expression of genes in S . cerevisiae intron deletion strains , 15 ml cultures of mutant and wild-type yeast were grown in parallel at 30°C in rich medium supplemented with 2% glucose to an optical density between A600 = 0 . 5 and 0 . 7 . For microarray hybridization , RNA was isolated by acid-phenol extraction and converted to cDNA as described [16] . A similar protocol was performed for qPCR applications with the following modifications . After RNA isolation , 2 µg of DNase-treated RNA was random primed in a 40 µl reaction containing 1 µg dN9 primer , 50 mM TrisHCl ( pH 8 . 4 ) , 75 mM KCl , 3 mM MgCl2 , 10 mM DTT , 0 . 5 mM dNTPs , and 5 ng murine Moloney leukemia virus ( M-MLV ) RT . Primers were hybridized at 60°C for 7 min prior to the addition of enzyme , and then incubated with enzyme at 42°C for at least 2 h . Prepared cDNA was diluted at least 10-fold before use in qPCR . Similarly , to analyze D . melanogaster S2 cell , 725 µl cultures of S2 cells ( UCSF cell culture facility ) were grown in 24-well plates at 25°C in Schneider's Drosophila Medium ( Gibco ) supplemented with 10% fetal bovine serum ( UCSF ) to a count of ∼5E6 cells . RNA was extracted with 1 ml TriZol ( Invitrogen ) according to manufacturer's instructions . After RNA isolation , cDNA was prepared with SuperScript III ( Invitogen ) and random priming according to manufacturer's instructions . Prepared cDNA was diluted at least 10-fold before use in qPCR . Splicing-sensitive microarrays were constructed and performed as described [16] . In each experiment , a wild-type strain derived from the same parent as the intron deletion mutant strain was used as a reference . Data was analyzed using the R Bioconductor packages marray ( ) and limma ( ) [55] in a custom pipeline based on the Goulphar program [56] . Microarray data used in this study are available in the Gene Expression Omnibus at NCBI ( GSE35541 ) . Quantitative PCR primers ( Table S2 ) were designed using Primer3 [57] and S . cerevisiae or D . melanogaster genomic sequence obtained from the UCSC Genome Browser ( SacCer1 or dm3 , respectively ) [58] , [59] . Serial dilutions of DNA ranging from 100 to 0 . 16 ng of the genomic DNA were used to obtain calibration curves , measure primer efficiencies , and ensure that quantification was in a linear dynamic range . Primer sets yielding multiple amplification products or calibration curves with R-squared values of <0 . 96 were excluded . For each qPCR sample , diluted cDNA was amplified in 25 µl volume reactions containing 250 µM dNTPs , 1× ( NH4 ) 2SO4 buffer ( Fermentas ) , 0 . 5 µM primer , 1 . 5 mM MgCl2 , 1 . 25 Units Dynazyme II ( Finnzymes ) , and Sybr Green I fluorescent dye ( Sigma ) . Fluorescence was measured on a BioRad Opticon machine using standard cycling conditions ( 3 min at 95°C , 40 cycles of 15 s at 95°C , 30 s at 55°C , and 15 s at 72°C ) . Biological replicate qPCR values were determined as the median of technical replicates . For each of 3 biological replicates , target gene values ( e . g . RPS9A ) were divided by reference gene values ( e . g . SCR1 ) before log transformation . Plots were generated using the R package ggplot2 ( ) [60] . To assess EST coverage and splicing , we obtained genomic coordinates corresponding to GenBank ESTs from the UCSC Genome Browser ‘spliced EST’ track , which span at least one canonical intron of at least 32 bases [61] . Custom R scripts were used to calculate EST coverage per genomic nucleotide position , and identify all exon-exon junctions that span putative GT/AG splice site [62] . The following genome assemblies were used in the analysis: Branchiostoma floridae ( braFlo1 ) ; Ciona intestinalis ( Ci2 ) ; Danio rerio ( danRer7 ) ; Drosophila melanogaster ( dm3 ) , Oryzias latipes ( oryLat2 ) ; Petromyzon marinus ( petMar1 ) ; Xenopus tropicalis ( xenTro2 ) ; Rattus norvegicus ( rn4 ) ; Mus musculus ( mm9 ) ; Homo sapiens ( hg19 ) .
|
Eukaryotic genes are littered with non-coding intervening sequences , or introns , that must be precisely excised from a messenger RNA before it can be properly translated into protein . Despite their ubiquity , the evolution and function of introns remain poorly understood . Consequently , we cannot accurately predict the functions of individual introns in any organism . In this manuscript , we used a combination of comparative genomics and experimental tests to identify functional introns . First , we looked for signatures of selection to identify important introns in the model yeast Saccharomyces cerevisiae , which focused our attention on the introns of ribosomal protein genes . We then genetically deleted these introns to assess their function . Unlike mammalian introns , we found that yeast introns were not required for high levels of gene expression . Instead , particular introns ( we focus on those within genes encoding ribosomal protein S9 ) were required to fine-tune gene expression through autoregulation . Surprisingly , animal orthologs of these genes also use introns to autoregulate through multiple forms of alternative splicing . We speculate that the introns of ribosomal protein genes , in particular , readily evolve means for autoregulation to meet the demanding requirements of ribosomal protein genes to maintain tight control of gene expression .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"animal",
"models",
"forms",
"of",
"evolution",
"genomics",
"model",
"organisms",
"gene",
"expression",
"genetics",
"molecular",
"genetics",
"biology",
"evolutionary",
"biology",
"yeast",
"and",
"fungal",
"models",
"microbiology",
"evolutionary",
"processes",
"genetics",
"and",
"genomics"
] |
2012
|
Diverse Forms of RPS9 Splicing Are Part of an Evolving Autoregulatory Circuit
|
The acylphloroglucinol rhodomyrtone is a promising new antibiotic isolated from the rose myrtle Rhodomyrtus tomentosa , a plant used in Asian traditional medicine . While many studies have demonstrated its antibacterial potential in a variety of clinical applications , very little is known about the mechanism of action of rhodomyrtone . Preceding studies have been focused on intracellular targets , but no specific intracellular protein could be confirmed as main target . Using live cell , high-resolution , and electron microscopy we demonstrate that rhodomyrtone causes large membrane invaginations with a dramatic increase in fluidity , which attract a broad range of membrane proteins . Invaginations then form intracellular vesicles , thereby trapping these proteins . Aberrant protein localization impairs several cellular functions , including the respiratory chain and the ATP synthase complex . Being uncharged and devoid of a particular amphipathic structure , rhodomyrtone did not seem to be a typical membrane-inserting molecule . In fact , molecular dynamics simulations showed that instead of inserting into the bilayer , rhodomyrtone transiently binds to phospholipid head groups and causes distortion of lipid packing , providing explanations for membrane fluidization and induction of membrane curvature . Both its transient binding mode and its ability to form protein-trapping membrane vesicles are unique , making it an attractive new antibiotic candidate with a novel mechanism of action .
The vast majority of antibiotics currently used in the clinic are derived from microbial sources [1] . However , plants represent an enormous source of potent bioactive molecules and several plant-derived compounds show promising antibacterial activities [2] . Considering the alarming rise in antibiotic-resistant pathogens , it is crucial to explore the potential of antimicrobials from herbal sources for new antibiotic development [3 , 4] . One such promising compound is the acylphloroglucinol rhodomyrtone , isolated from the leaves of the rose myrtle Rhodomyrtus tomentosa [5] . Rhodomyrtone is highly active against a broad range of Gram-positive bacteria , among which Bacillus , Enterococcus , Staphylococcus , and Streptococcus species , including clinical isolates and multi-resistant strains [5 , 6] , and eradicates mature biofilms of Propionibacterium acnes [7] , Staphylococcus aureus , and Staphylococcus epidermidis [8] . Rhodomyrtone is bactericidal and its minimal inhibitory concentrations are comparable to that of the last-resort antibiotics vancomycin and daptomycin [8] . Attempts to cultivate resistant mutants in the laboratory have not been successful in multiple passaging experiments [6] . Importantly , no cytotoxic effects have been observed on human fibroblasts and erythrocytes [6 , 9] . Rhodomyrtone did also not cause skin irritation upon topical application in rabbits [10] and acute toxicity tests did not show adverse effects in mice , when injected [11] . Furthermore , it has been proven effective against Propionibacterium acne biofilms [10 , 12] , showed excellent results in preventing staphylococcal adhesion and invasion in a tissue model of bovine mastitis [13] , and prevented adhesion of dental pathogens to plastic surfaces and human buccal cells [14] . Despite being such a promising new antibiotic candidate , the mechanism by which rhodomyrtone kills bacteria is not yet understood . Early studies with Streptococcus mutans , Streptococcus pyogenes , and S . aureus reported no distinct cell lysis or leakage of intracellular content , but upregulation of core metabolic pathways [5 , 6 , 15] . These results prompted the conclusion that rhodomyrtone likely has an intracellular target . A computational docking approach identified the dihydrofolate reductase DfrA as potential target but this could not be experimentally confirmed [16] . The same study also found a possible interaction of rhodomyrtone with the essential cell division protein FtsZ , and an earlier proteomic study showed reduced FtsZ levels and changes in cell size , shape , and septum formation in rhodomyrtone-treated S . aureus [17] . However , a recent study demonstrated that rhodomyrtone is unable to specifically inhibit B . subtilis FtsZ polymerization in vivo but rather affects several different cell division proteins [18] . Because of these conflicting observations , we investigated the effect of rhodomyrtone on the bacterial cell envelope more closely using a recently established cell biology-based approach to study the mode of action of antibacterial compounds [19 , 20] . Using fluorescence light and high-resolution microscopy , together with specialized membrane dyes , in vitro techniques , and molecular modeling , we found that rhodomyrtone primarily acts on the cell membrane , but in an unexpected manner . Rhodomyrtone induces dramatic membrane invaginations resulting in intracellular vesicles that attract and trap a variety of membrane proteins . These structures also attract flexible membrane lipids leading to highly fluid ( liquid-disordered ) membrane domains and subsequent rigidification ( liquid-ordered ) of the rest of the membrane , resulting in further delocalization of peripheral membrane proteins . These membrane distortions impair multiple essential membrane-associated processes , including respiration and ATP synthesis . Molecular dynamics simulations suggested that rhodomyrtone accomplishes membrane remodeling by transiently interacting with phospholipid head groups , thus without integrating into the lipid bilayer . Such a molecular mechanism has not been observed for any membrane-targeting antimicrobial before and explains why rhodomyrtone has long been thought not to target the cell envelope .
Previous studies have suggested that rhodomyrtone could inhibit intracellular processes [5 , 16 , 21] . To investigate this further , we employed a recently developed assay that uses delocalization of marker proteins to identify the cellular process that is targeted by an antimicrobial compound [19] . This technique is also known as bacterial cytological profiling [22 , 23] . Firstly , we examined whether rhodomyrtone causes DNA damage , or inhibits DNA replication , RNA , or protein synthesis . To this end , the relevant enzymes were labelled with GFP and their cellular localization was monitored using fluorescence light microscopy ( Fig 1 , see S1 Table for strains list ) . A change in the localization of RecA ( DNA-repairing enzyme ) , DnaN ( DNA polymerase subunit ) , RpoC ( RNA polymerase subunit ) , and RpsB ( ribosomal subunit ) is indicative of DNA damage , inhibition of DNA replication , RNA synthesis , and protein synthesis , respectively ( S1 Fig ) . As shown in Fig 1 , the cellular localization of none of these proteins was affected by rhodomyrtone , indicating that the compound does not inhibit any of these intracellular processes . Previous studies have shown that rhodomyrtone induces cell shape deformations and cell lysis [17 , 18] , suggesting that the compound could target the bacterial cell envelope . To examine this more closely , we monitored the localization of a representative set of peripheral membrane proteins that have been used to study the effects of membrane-active antibiotics [19 , 24 , 25]: ( i ) the cell division proteins FtsA , DivIVA , and MinD , ( ii ) the phospholipid synthase PlsX , ( iii ) the cell shape-determining protein MreB , ( iv ) the cell wall synthesis protein MurG , and ( v ) the succinate dehydrogenase SdhA , which is part of the respiratory chain . All proteins showed aberrant localization patterns after 10 min of treatment , which aggravated when treatment was continued for 30 min ( Fig 2A ) . DivIVA , MinD , PlsX , MreB , and MurG accumulated in large fluorescent foci while FtsA and SdhA were completely detached from the membrane . These changes were observed in nearly all cells ( see S2 and S3 Figs for overview pictures of GFP-MreB ) . To test how fast these changes happen , we selected DivIVA for a time lapse experiment . As shown in Fig 2B and S1 Movie , delocalization of DivIVA was already visible after 2 min . Thus , rhodomyrtone rapidly affects multiple membrane-bound processes . Several studies have shown that rhodomyrtone treatment results in cell shape deformations and some degree of cell lysis [17 , 18] , which fits well with our observation that the localization of MurG , an enzyme involved in the synthesis of the cell wall precursor lipid II , was affected by rhodomyrtone ( Fig 2A ) . However , at 1x MIC we did not observe any lysis based on optical density ( OD ) measurements , and 2x MIC only led to a gradual OD reduction ( Fig 3A ) . To examine whether rhodomyrtone affects the integrity of the cell wall , we employed a microscopic fixation method to visualize cell wall damage [26] . As shown in Fig 3B , rhodomyrtone did not have an immediate effect on cell wall integrity , which was in sharp contrast to the antimicrobial peptides daptomycin , gramicidin S , and MP196 , all of which interfere with the synthesis of lipid II [19 , 27] . Thus , cell wall synthesis does not seem to be the primary target of rhodomyrtone . Previously , we have shown that the localization of several peripheral membrane proteins , including MinD and FtsA , depends on the membrane potential , as the attachment of their membrane-targeting amphipathic helices to the cell membrane is strongly stimulated by the electric potential difference over the membrane [24] . Since we observed delocalization of a number of peripheral membrane proteins ( Fig 2 ) , among which MinD and FtsA , we examined whether rhodomyrtone dissipates the membrane potential using the membrane-potentiometric probe DiSC ( 3 ) 5 [28] ( Fig 3C ) . Indeed , cells were depolarized within approximately 3 min . This was not as fast and strong as the depolarization that occurs with the antibiotic peptide gramicidin , which forms a K+/Na+ channel in the membrane [29] ( Fig 3C ) , suggesting that rhodomyrtone neither forms an ion channel nor a membrane pore . The latter was confirmed using propidium iodide ( PI ) , a fluorescent indicator that enters cells through pores or severe membrane breaches [30] . As shown in Fig 3D , no significant PI influx was observed , not even after 30 min treatment with 4x MIC . To test whether rhodomyrtone enhances the ion permeability of the membrane , we measured the intracellular concentrations of potassium with the potassium-selective fluorescent probe APG-2 for 20 min ( Fig 3E ) . Indeed , rhodomyrtone caused a continuous , concentration-dependent leakage of potassium . However , this leakage occurred too slowly ( 10–30 min ) to explain the much quicker ( 3 min ) membrane depolarization ( Fig 3C ) . Furthermore , at 1x MIC cells recovered their potassium levels after approximately 20 min ( Fig 3E ) but not their membrane potential ( S4 Fig ) . Depolarization can not only be caused by permeabilization of the membrane but also by impairing the electron transport chain , which generates the proton gradient [27] . In fact , Fig 2A shows that succinate dehydrogenase ( SdhA ) , which channels protons from the TCA cycle into the electron transport chain [31] , fully detaches from the membrane when cells are incubated with rhodomyrtone . Importantly , we have shown in previous work that this protein does not delocalize when the proton motive force is dissipated [25] . Thus , delocalization of SdhA is not a general effect of membrane depolarization but specific to rhodomyrtone treatment . To assess whether the activity of the respiratory chain was indeed affected by rhodomyrtone , we measured the reduction rates of resazurin , which is commonly used to measure the reductive capacity of cells [19 , 32] . Indeed , incubation with rhodomyrtone considerably diminished the reductive capacity of B . subtilis cells , which is indicative of a strong inhibition of the electron transport chain ( Fig 3F ) . This effect was comparable to that of the proton ionophore carbonyl cyanide m-chlorphenylhydrazone ( CCCP ) , a de-coupler of the electron transport chain , and much stronger than that of sodium azide , which inhibits complex IV of the respiratory chain ( Fig 3F ) . Thus , rhodomyrtone affects both maintenance and generation of the proton motive force . While our results showed that rhodomyrtone affects different membrane functions , it is still unclear how the compound achieves these membrane effects . In order to shed light on its molecular mechanism , we first investigated the effect of rhodomyrtone on membrane organization using the fluorescent membrane dye FM5-95 . Addition of the compound caused highly fluorescent membrane foci in almost 90% of the cells ( Fig 4A and 4B , see S5 and S6 Figs for overview pictures ) , while it did not affect the nucleoid ( Fig 4A , see S7 and S8 Figs for overview pictures ) . Such aberrant membrane stains have also been observed when the membrane potential was dissipated with CCCP , which has been attributed to the accumulation of highly flexible lipids [25] . A high concentration of such flexible lipids , i . e . lipids with short , branched and/or unsaturated fatty acid chains , increases local membrane fluidity ( liquid-disordered state ) and thus facilitates the insertion of fluorescent membrane probes and/or increases their fluorescence quantum yield , resulting in strongly enhanced fluorescence signals [19 , 25 , 33] . In logarithmically growing B . subtilis cells such flexible lipids will form microscopically visible fluid microdomains ( S9 Fig ) stimulated by the actin homologue MreB [24] . These regions of increased fluidity ( RIFs ) can be stained with the fluorescent lipid-mimicking dye DiIC12 , which has a strong preference for fluid membrane domains due to its short acyl chain [25] . When DiIC12-stained cells were treated with rhodomyrtone , the regularly distributed RIFs quickly collapsed into large foci ( Fig 4C ) , indicating that the strongly fluorescent FM5-95 patches are indeed enriched in flexible membrane lipids . Quantification of the number of DiIC12 foci per cell showed a clear reduction of multiple RIFs to one or few large DiIC12-stained fluid lipid foci per cell ( Fig 4D , see S10 and S11 Figs for overview pictures ) , suggesting that RIFs might fuse together to generate these domains , a notion that was supported by time lapse microscopy ( Fig 4E ) . RIFs are formed by membrane-attached MreB polymers by a yet unknown mechanism [25 , 34] . Since MreB was delocalized by the compound ( Fig 2A ) , we tested whether the rhodomyrtone-induced clustering of RIFs requires MreB or one of its homologues ( Mbl , MreBH ) using a ΔmreB Δmbl ΔmreBH deletion mutant . This mutant forms round cells since it lacks the longitudinal organization of peptidoglycan synthesis [35] . As shown in Fig 4F , rhodomyrtone was still able to cause fluorescent membrane foci , demonstrating that MreB is not required for this . To determine whether the clustering of flexible ( = fluidizing ) lipids affects overall membrane fluidity , we employed the fluorescence polarization probe laurdan , which changes its fluorescence properties depending on head group spreading and fatty acid chain flexibility [36] . Addition of rhodomyrtone resulted in a rapid ( 2 min ) , concentration-dependent reduction of laurdan generalized polarization ( GP ) , which is indicative of an increase in membrane fluidity ( Fig 5A ) , as can be seen from the positive control , the known membrane fluidizer benzyl alcohol [37] . In order to examine how fluidity changes at the single cell level , we employed laurdan microscopy ( Fig 5B–5D ) . In line with the focal enrichment of flexible lipids , membrane foci of highly increased fluidity were apparent , and the rest of the cell membrane showed a clear reduction in fluidity ( liquid-ordered state ) . Thus , the overall increase in membrane fluidity observed in Fig 5A is caused by the strongly fluid membrane foci in the cell . This effect is enhanced by the accumulation of laurdan in these fluid membrane areas ( Fig 5B ) , resulting in higher signal intensities in these areas ( S12 Fig ) . Interestingly , when we compared the fluid membrane patches formed by rhodomyrtone with the fluid lipid patches formed by CCCP , we found that rhodomyrtone had a substantially stronger local fluidizing effect ( Fig 5D ) , supporting the notion that rhodomyrtone-induced patches are fundamentally different from fluid lipid clusters formed by CCCP through MreB delocalization [25] . In order to show that the effects of rhodomyrtone on membrane fluidity are not simply an effect of growth inhibition , we treated cells with ciprofloxacin , which does not target the bacterial membrane but kills bacteria by inhibiting DNA synthesis . Incubation of B . subtilis cells for 10 min with lethal concentrations of ciprofloxacin did not affect overall membrane fluidity and did not result in clear fluid membrane foci ( S13 Fig ) . If focal membrane fluidization is indeed a key feature of rhodomyrtone activity , it is likely that bacteria will attempt to adjust their fatty acid composition to compensate for these changes . Indeed , when cells were incubated with sub-inhibitory concentrations of rhodomyrtone , cells showed a significant decrease of short chain fatty acids in favor of long chain fatty acids , which will increase membrane rigidity ( Fig 5E and S2 and S3 Tables ) . It should be noted that we did not observe changes in the ratios of iso- to anteiso-branched chain fatty acids or saturated to unsaturated fatty acids , which are the major adaptation strategy of B . subtilis in response to physical changes of membrane fluidity [38 , 39] . Since B . subtilis adapts its fatty acid composition towards higher membrane rigidity , it seems that the strong focal fluidization of the membrane is the major problem for cells rather than the rigidification of the rest of the cell membrane . In previous studies we have shown that the highly fluorescent membrane spots that are formed when cells are treated with membrane potential-dissipating drugs , are not a consequence of the accumulation of extra membrane material , e . g . due to membrane invagination [25] . This was proven by showing that the localization of the transmembrane FoF1 ATP synthase remained undisturbed and that there was no accumulation of this protein at areas that showed strong fluorescence of membrane dyes . It was therefore remarkable that when we repeated this experiment with rhodomyrtone , AtpA , the subunit of the FoF1 ATP synthase complex , became delocalized and accumulated in foci when cells were treated with rhodomyrtone ( Fig 6A , S14 Fig ) , suggesting that the compound does actually lead to membrane invaginations . To examine this , we used super-resolution Structured Illumination Microscopy ( SIM ) . Cell membranes were stained with mitotracker green , which provides excellent SIM contrast and optimal resolution of membrane structures . After 10 min incubation with rhodomyrtone , membrane invaginations and large intracellular vesicles became clearly visible ( Fig 6B ) . Since SIM image reconstruction can sometimes produce visual artefacts [40] , we confirmed the presence of membrane invaginations using a B . subtilis strain expressing cytosolic GFP from the strong ribosomal PrpsD promoter ( Fig 6C ) [40] . Indeed , the large vesicle-like structures ( yellow arrows ) lacked GFP , indicating that they originated from cell membrane invaginations . Some of the smaller membrane invaginations ( blue arrows ) did not clearly show a displaced GFP signal , which can be the case when their internal volume is below SIM resolution . SIM microscopy of DiIC12-stained cells confirmed that the invaginations accumulated the fluid membrane probe DiIC12 , as expected ( Fig 6D ) . To confirm the generation of internal membrane vesicles by rhodomyrtone with an independent technique , we performed transmission electron microscopy ( TEM ) ( Fig 6E ) . In line with the SIM data , we observed large intracellular membrane structures , some of which were filled ( Fig 6Eiii and 6Ev ) and others devoid of cytosolic material ( Fig 6Eiv and 6Evi ) , as well as smaller membrane vesicles ( Fig 6Eiii and 6Eiv ) . Similar membrane invaginations have been observed after overexpressing the acetyl-CoA carboxylase AccABCD , the enzyme catalyzing the rate-limiting step of fatty acid synthesis [41 , 42] . However , it takes several hours of AccABCD overexpression to form invaginations of significant size [42] , whereas rhodomyrtone already causes large membrane invaginations after only 2–10 min of treatment . Moreover , inhibition of protein and lipid synthesis by chloramphenicol and triclosan , respectively , did not prevent the formation of invaginations by rhodomyrtone ( S15 Fig ) , further indicating that the compound is directly responsible for membrane invaginations . Since AtpA , our reporter for membrane invagination , accumulated into foci ( Fig 6A ) , and the rhodomyrtone-induced membrane patches ( = invaginations ) were enriched in fluidizing lipids ( Fig 4C–4E ) , we wondered whether these fluid membrane structures could attract membrane proteins in general , resulting in the patchy GFP localization patterns observed in Fig 2 . To examine this , we stained cells expressing GFP-tagged peripheral membrane proteins , and cells expressing GFP-tagged integral membrane proteins , together with the fluid lipid domain dye DiIC12 . As shown in Fig 7A , all tested peripheral membrane proteins showed a clear accumulation at bright DiIC12 spots . Importantly , Fig 7B shows that also the transmembrane proteins MraY and PBP2B , involved in cell wall biosynthesis , and the phospholipid synthase PgsA accumulated at the rhodomyrtone-induced fluid lipid clusters . We selected MreB as exemplary protein for a time lapse experiment to follow protein localization during the generation of fluid lipid clusters ( Fig 7C ) . The accumulation of MreB clearly correlated with the formation of DiIC12-stained domains ( arrows in Fig 7A ) and was retained there . Together , these results indicate that both peripheral and integral membrane proteins become attracted by the fluid membrane domains and are then trapped in the resulting membrane vesicles caused by rhodomyrtone ( see S16 Fig for the transmembrane protein MraY ) . These membrane domains were stable . Cells were neither able to remove these membrane structures and re-establish a normal membrane morphology ( Fig 4C ) , nor to recover normal protein localization patterns ( example MraY in S16 Fig ) for the time spans observed in this study ( 60–90 min ) . Clearly , this trapping of membrane proteins in vesicles will strongly impair their function and explains the strong antimicrobial effect of rhodomyrtone . Structurally , rhodomyrtone does not resemble a typical membrane-integrating agent . While it does contain both polar and non-polar groups , these are spread out over the molecule and do not define distinct amphipathicity ( Fig 8A ) , as is typically the case for membrane-integrating antimicrobials [43] . Thus , it seemed rather unlikely that the compound intercalates in between phospholipids . To gain insight into how rhodomyrtone could interact with and affect phospholipids , we performed molecular dynamics simulations using a 3:1 mixture of 1-palmitoyl-2-oleoyl-posphatidylglycerol ( POPG ) and 1-palmitoyl-2-oleoylphosphatidylethanolamine ( POPE ) , which mimics the Gram-positive cell membrane . As expected , rhodomyrtone was unable to penetrate into the lipid bilayer . Instead , it temporarily attached to phospholipid head groups ( Fig 8B and 8C ) resulting in repeated events of binding and release ( S2 Movie ) . This binding involved both the polar and non-polar residues of the antibiotic but did not involve any selectivity for either PG or PE ( Fig 8C , S4 Table ) . Binding of rhodomyrtone to phospholipid head groups affected the vertical arrangement of the phospholipid bilayer by temporarily dragging lipids out of the membrane ( Fig 8C ) , resulting in tighter head group packing ( Fig 8D ) and rearrangement of the bound lipids in the outer membrane leaflet ( Fig 8E and 8F ) . These changes in the outer leaflet also affected lipid packing in the inner leaflet in terms of spreading of fatty acid chains , promoting locally increased membrane disorder ( Fig 8D–8F ) , which would explain the fluidizing effect of rhodomyrtone . To investigate whether increased concentrations of rhodomyrtone molecules showed a different effect , we also performed molecular dynamics simulations with two , four , and eight molecules of rhodomyrtone . Despite the fact that rhodomyrtone molecules tended to form clusters , the effects on lipid packing were similar ( S17 Fig ) . Together with our in vivo data , the molecular dynamics simulations indicated that rhodomyrtone is the first membrane-active antibiotic molecule that does not integrate into the lipid bilayer and causes severe lipid packing defects solely by interaction with phospholipid head groups . Molecular dynamics simulations showed a transient interaction between rhodomyrtone and phospholipid head groups , suggesting that it should be possible to remove the compound by washing . Indeed , when we treated B . subtilis with 1x MIC of rhodomyrtone for 2 min , cells recovered completely from membrane deformations , when they were washed and further grown in fresh medium ( S18 Fig ) . This is in contrast to the membrane-intercalating antibiotic daptomycin , which cannot be washed out [19] . Cells treated with higher concentrations of rhodomyrtone or incubated for longer than 2 min were unable to recover , probably due to irreparable membrane damage , which is well in line with the formation of non-reversible membrane invaginations . The molecular dynamics simulations also indicated that rhodomyrtone does not display any selectivity for PG or PE lipids . To test this in vivo , we tested whether the lipid head group composition plays a role in the activity of rhodomyrtone , using B . subtilis mutants devoid of either cardiolipin , phosphatidylethanolamine ( PE ) , lysyl-phospatidylglycerol ( L-PG ) , or depleted for phosphatidylglycerol ( PG ) , the four main phospholipid species in bacteria . If rhodomyrtone preferentially binds to one of these phospholipid species , it should be less active against a deficient strain . As a positive control we tested daptomycin , which is less effective against strains depleted for its docking molecule PG [44–46] . Whereas the MIC of daptomycin indeed doubled for PG-depleted cells , none of the tested mutations reduced the activity of rhodomyrtone ( Fig 9A ) . This supports the molecular dynamics results suggesting that rhodomyrtone has no preference for a specific lipid head group type . It should be noted that a slightly enhanced rhodomyrtone activity was observed for the PG-depleted strain , which could be interpreted as a hampering effect on the action of rhodomyrtone by PG . However , it has been shown before that PG depletion leads to severe growth defects and morphological changes [25] , and the lower MIC is likely a consequence of the reduced general fitness of PG-depleted cells . To prove this , we determined the MIC of several other antibiotics with different targets and indeed observed that PG-depleted cells were more sensitive to all of them ( S5 Table ) . Our molecular dynamics simulations suggested that membrane lipids are the direct target of rhodomyrtone . To confirm this , we performed an in vitro laurdan GP experiment using liposomes prepared from Escherichia coli polar lipid extract . Benzyl alcohol , which has as direct fluidizing effect on lipid membranes [47–49] , was used as positive control ( Fig 9B ) . As shown in Fig 9C , the addition of rhodomyrtone caused a clear reduction in laurdan GP at concentrations that roughly correspond to the estimated compound to lipid ratio under our in vivo experiments ( 1:20 , further commented on in the Method section ) . Thus , membrane fluidization is a direct effect of rhodomyrtone on the phospholipid bilayer and does not require a protein target . Molecular dynamics simulations also showed that rhodomyrtone induced tighter head group packing in the outer membrane leaflet ( Fig 8D ) , which could lead to bending of the membrane and induction of membrane invaginations . To test whether rhodomyrtone is able to bend membranes , we analyzed the shape of liposomes treated with rhodomyrtone using SIM microscopy . At low rhodomyrtone concentrations liposomes became much smaller than in an untreated control sample ( Fig 9D and 9E ) , indicating membrane remodeling and vesiculation [50 , 51] . Higher concentrations led to severe deformation of liposomes resulting in strongly bend lipid structures ( Fig 9D ) . Membrane deformation and vesiculation are in line with our SIM and TEM data ( Fig 6B–6E ) . Clearly , rhodomyrtone itself is capable of bending the membrane and forming small membrane vesicles . To determine whether the results obtained with B . subtilis can be transferred to pathogenic microorganisms , we tested the formation of DiIC12-stained fluid lipid accumulations and the formation of membrane vesicles in S . aureus and S . pneumoniae . Indeed , we observed clear membrane invaginations in both organisms after treatment with 1x MIC rhodomyrtone ( 1xMIC , 1 μg/ml ) for 10 min , using the Nile red membrane dye ( Fig 10A ) . When S . aureus and S . pneumoniae were stained with the fluid lipid domain dye DiIC12 , bright fluorescent patches became visible , indicating the accumulation of fluid lipid domains ( Fig 10B ) . High resolution SIM microscopy of the same cells , confirmed that these fluid lipid-enriched domains are membrane invaginations ( Fig 10B ) , showing that rhodomyrtone has the same principal mechanism of action against clinically relevant pathogens . Although the rhodomyrtone-producing plant R . tomentosa is used in traditional Asian medicine [52] and several in vitro studies have shown promising results in terms of antibacterial activity and safety of rhodomyrtone [5–15] , there is so far no information available on its effectiveness in an animal infection model . Therefore , we performed S . pneumoniae infection experiments with one day old zebra fish embryos , a well-established bacterial infection model [53–55] . To monitor the effect on S . pneumoniae cells in fish , we used a GFP-expressing strain [53 , 56] . Using fluorescence microscopy , we could clearly see that rhodomyrtone significantly reduced the bacterial load in the fish ( Fig 10C , S19 Fig ) . Infection of zebra fish embryos with S . pneumoniae leads to characteristic damage to the heart region of the fish and bloating , which can easily be observed under the microscope ( Fig 10C , S19 Fig ) . Injection of rhodomyrtone notably reduced this symptom ( Fig 10C , S19 Fig ) . These findings show that the compound is effective against S . pneumoniae infection in vivo . Addition of rhodomyrtone to the water did not reduce bacterial load or bloating of the heart region ( Fig 10C ) . We did not observe signs of toxicity at the highest rhodomyrtone concentration injected ( Fig 10C , S19 Fig ) , indicating that there is a therapeutic window for systemic applications . In line , we did not observe any fluid lipid accumulations up to 200x MIC , when we stained human erythrocytes with the fluid lipid domain dye DiIC12 ( S20 Fig ) , indicating that the induction of membrane invaginations by rhodomyrtone is unique to bacterial membranes . So far , it is unclear which mechanism underlies the selectivity of rhodomyrtone for bacterial cells . Therefore , we performed laurdan spectroscopy of liposomes consisting of either POPG or 1-palmitoyl-2-oleoylphosphatidylcholine ( POPC ) . Interestingly , no membrane fluidization was observed in POPC liposomes ( S21 Fig ) . In contrast to PG , which is a typical bacterial membrane lipid , PC is abundant in mammalian cell membranes but not present in most bacteria . This might provide a first indication why rhodomyrtone is selective for bacterial over mammalian cells .
Since rhodomyrtone does not form large membrane pores and does not strongly affect the cell wall , it has long been thought not to target the bacterial cell envelope [5 , 6 , 15] . Here , we show that rhodomyrtone acts on the cytoplasmic membrane via a novel molecular mechanism , resulting in large membrane invaginations and intracellular vesicles that trap membrane proteins . The results of this study are summarized in a working model schematically depicted in Fig 11 . Contrary to other membrane-active antimicrobials [43] , rhodomyrtone does not have a clear amphipathic membrane-interacting structure or domain and is very unlikely to insert into the cell membrane . Molecular dynamics simulations indicated that multiple repeats of binding and release cause tighter packing of the phospholipid head groups in the outer leaflet , resulting in wider fatty acid spreading in the inner membrane leaflet and bending of the membrane ( Fig 11A ) . In addition , fluidizing lipids , i . e . lipids with short , branched , or unsaturated fatty acids , with a higher degree of flexibility , will locate to these sites since they better fit into this disordered lipid environment than less flexible lipid molecules ( Fig 11B ) . The formation of these highly fluid domains in the membrane has three effects . Firstly , the local accumulation of fluidizing lipids leads to rigidification of the rest of the cell membrane resulting in dissociation of peripheral membrane proteins ( Fig 11C ) , which require a certain degree of fluidity to insert their membrane-binding domains [57] . Secondly , the formation of membrane domains of different thickness causes hydrophobic mismatches that are likely to compromise the membrane barrier function and allow leakage of ions leading to a gradual dissipation of the membrane potential [19 , 58–62] ( Fig 11D ) . And thirdly , these highly fluid lipid domains attract both peripheral and integral membrane proteins ( Fig 11E ) , since a more fluid lipid environment accommodates most membrane proteins better than a rigid environment [57] . Finally , the combination of local membrane curvature and a high concentration of fluidizing lipids stimulates the formation of membrane invaginations leading to vesicles , effectively trapping the accumulated membrane proteins ( Fig 11F ) . Trapping of membrane proteins in vesicles by rhodomyrtone disturbs a wide variety of membrane processes . The delocalization of proteins involved in cell shape ( MreB ) , cell wall synthesis ( MurG , MraY , PBP2B ) , and cell division ( FtsA , MinD , DivIVA ) explains the effects on the cell size and shape of B . subtilis , S . aureus , and S . pyogenes that were observed in earlier studies [17 , 18 , 21] . In contrast to the antimicrobial peptides daptomycin , MP196 , and gramicidin S , all of which affect the function of the cell wall synthesis protein MurG [19 , 27] , rhodomyrtone did not have an immediate effect on cell wall integrity ( Fig 3B ) , which was surprising considering the fact that the compound delocalizes MurG , MraY , and PBP2B . However , this is well in line with the stress response of S . aureus to treatment with rhodomyrtone , which did not show upregulation of typical cell envelope stress proteins , such as the liaRS two-component system [17 , 63] . Moreover , cell shape defects caused by rhodomyrtone become only visible after prolonged treatment times of one to four hours [18] , suggesting that inhibition of cell wall synthesis is not the main mechanism of growth inhibition . While rhodomyrtone had no immediate effect on cell wall synthesis , it had a rapid and strong effect on cellular respiration ( Fig 3F ) , probably due to the displacement of respiratory chain proteins such as SdhA and dissociation of the FoF1 ATP synthase complex ( Figs 2A and 6A and S17 Fig ) . This is in agreement with the observed upregulation of metabolic pathways in earlier rhodomyrtone studies [21] , since energy depletion ( ATP limitation ) typically leads to an unspecific inhibition of the synthesis of the main cellular macromolecules [19 , 27] . Such a strong inhibitory effect on cellular respiration has not yet been observed with other antibiotics that cause membrane protein delocalization , including daptomycin , MP196 , and gramicidin S [19 , 27] . In fact , no other antibiotic tested so far has been able to promote delocalization of the ATP synthase [19 , 24 , 27] . The common view of membrane-targeting antibiotics is that they form pores or ion channels [64] . However , over the last few years a number of membrane-targeting compounds have been discovered that do not form membrane pores , including the antimicrobial peptides daptomycin , cWFW , MP196 , and gramicidin S [19 , 27 , 65] . While daptomycin specifically inserts into specialized membrane microdomains [19] , the other compounds act according to the interfacial activity model , i . e . they do not penetrate deeply enough into the membrane bilayer to form membrane-spanning pores , but rather act at the interface between phospholipid head groups and fatty acid chains of membrane lipids leading to membrane deformation [66] . None of these models applies to rhodomyrtone , which only binds to phospholipid head groups and does not intercalate in between membrane lipids . Thus , rhodomyrtone is the first antibiotic to achieve substantial changes in membrane architecture solely by binding to phospholipid head groups in the outer membrane leaflet . Daptomycin , MP196 , and gramicidin S also affect localization of certain membrane proteins ( MurG and PlsX , or MurG and cytochrome c , respectively ) [19 , 27] and the antimicrobial peptide cWFW spatially separates peripheral and integral membrane proteins into domains of different fluidity [67] . However , rhodomyrtone is unique in that it strongly affects a very broad range of membrane proteins and traps these proteins in intracellular membrane vesicles . R . tomentosa leave decoctions are traditionally used in Asia to treat gastrointestinal infections and prevent postpartum infections , and crushed leaves are applied as wound dressings [52] . Rhodomyrtone , isolated from these leaf extracts [8] , is not only bactericidal and very potent against a number of important bacterial pathogens , it also belongs to a new compound class and possesses a novel mechanism of action . Although it remained unclear which mechanism underlies its selectivity for bacterial membranes , it has been shown that the compound does not display toxicity against mammalian cells [6 , 8 , 11] . Here we could provide a first clue to the mechanism underlying the selectivity of rhodomyrtone , which appears to have a higher affinity for bacterial over mammalian membrane lipids ( S20 and S21 Figs ) . We could also show for the first time that rhodomyrtone is effective against bacteria in an infection model , suggesting a therapeutic window for systemic treatment . Since rhodomyrtone affects a number of different cellular processes and not a single protein target , a rapid development of resistance against the compound is not expected . In fact , no stable rhodomyrtone-resistant S . aureus mutant could be isolated in multiple passaging experiments [6] . Bacteria have developed several mechanisms of resistance against membrane-targeting antibiotics . For example , reduction of the PG content ( in favor of L-PG or cardiolipin ) in the cytoplasmic membrane and cell wall charge modifications are the most effective resistance mechanism against daptomycin and other membrane-targeting compounds [43 , 44] . In contrast to such compounds , rhodomyrtone does not have a strong preference for either negatively charged ( PG ) or neutral phospholipid ( PE ) head groups , and its activity is not affected by changes in the concentration of one of the main phospholipid species in bacteria ( Fig 9A ) . In addition , rhodomyrtone delocalizes important membrane-bound lipid synthases ( PlsX , PgsA ) , thereby complicating possible attempts by the bacterial cell to adapt the membrane lipid composition . Moreover , rhodomyrtone is not charged and therefore is also unlikely to be affected by D-alanylation of lipoteichoic acids in the cell wall , a common adaptation mechanism to positively charged antimicrobial molecules [68] . Therefore , the common resistance mechanisms against other membrane-targeting compounds are likely not effective against rhodomyrtone , minimizing the risk of cross-resistance with other antibiotics . In fact , rhodomyrtone is active against vancomycin-resistant strains of S . aureus [6] . Furthermore , it is effective against non-growing antibiotic-tolerant bacterial cultures ( S22 Fig ) , suggesting that it could also eradicate persister cells that cause dormant and recurring infections . In conclusion , rhodomyrtone is a potent new antibiotic candidate that belongs to a novel compound class and possesses a unique mechanism of action .
Rhodomyrtone was extracted from leaves of Rhodomyrtus tomentosa with 95% ethanol as described previously [21 , 69] . MP196 was synthesized as described before [70] . Daptomycin was purchased from Novartis . All other antibiotics were purchased from Sigma Aldrich in the highest possible purity . Rhodomyrtone , gramicidin , gramicidin S , triclosan , CCCP , MP196 , benzyl alcohol , and mitomycin C were dissolved in sterile DMSO . Daptomycin , ciprofloxacin , and sodium azide were dissolved in sterile water . Chloramphenicol , erythromycin , and rifampicin were dissolved in ethanol . The bactericidal concentration of ciprofloxacin was determined by plating cultures treated with the antibiotic for 10 min on unselective agar plates . Treatment with 1 μg/ml ciprofloxacin resulted in no CFU after overnight incubation . B . subtilis strains used in this study are listed in S1 Table . Unless otherwise noted , all strains were grown in Luria Bertani ( LB ) broth at 37°C or 30°C ( for microscopy ) under steady agitation in the presence of inducer where appropriate ( see S1 Table ) . Strain 4277 was grown in the presence of 20 mM MgCl2 . Strain HM1365 was grown in the presence of selection marker . S . aureus was grown in LB at 37°C . S . pneumoniae was grown in Todd-Hewitt broth supplemented with 0 . 5% yeast extract at 37°C . All experiments were performed in early exponential growth phase ( OD600 = 0 . 3–0 . 35 ) . Unless otherwise stated , experiments were carried out in biological triplicates . Strain UG-10 ( amyE::spc Pxyl-recA-mgfp ) was constructed by restriction cloning of the recA gene into the pSG1729 plasmid [71] using Escherichia coli Top 10 as cloning host . RecA was amplified with the primers UG03a ( 5’GCGCGCCTCGAGATGAGTGATCGTCAGGCAGCC3’ ) and UG04a ( 5’CGCGCGGAATTCGGATCCTGAGCCGCTTCCTGAGCCTTCTTCAAATTCGAGTTC3’ ) and cloned into the vector using the XhoI and EcoRI restriction sites . The resulting plasmid ( TNVS-recA-4GS-mgfp ) was transformed in B . subtilis 168 using a standard starvation protocol [72] . TNVS30D ( amyE::spc-Pxyl-mgfp-pgsA ) was constructed by restriction cloning into the pSG1729 plasmid ( carrying monomeric GFP ) using XhoI/EagI restriction sites . The pgsA gene was amplified using the primers pgsA-Fw ( gggCTCGAGggctcaggaagcggctcaggatccTTTAACTTACCAAATAAAATCACACTAGCT ) and pgsA-Re ( cccCGGCCGttaGTTAGATGTTTTTAACGCTTCCCA ) . The resulting plasmid ( pTNV13 ) was transformed to B . subtilis 168 resulting in strain TNVS30D . MICs were determined following a modified broth microdilution method recommended in the Clinical Laboratory Standardization Institute ( CLSI ) guidelines . Two-fold serial dilutions of rhodomyrtone were prepared in LB in 96-well microtiter plate format . Exponentially growing B . subtilis 168 was inoculated to each well to a final colony forming unit ( CFU ) count of 5x105 CFU/ml and incubated at 37°C for 16 h . The MIC was defined as the lowest concentration inhibiting visible growth . Growth experiments were performed with a Biotek Synergy MX plate reader in 96-well format under continuous shaking in a final volume of 150 μl per well . B . subtilis 168 was grown in LB at 37°C until an OD600 of 0 . 3 and subsequently treated with 0 . 25 , 0 . 5 , and 1 μg/ml rhodomyrtone ( 0 . 5x , 1x , and 2x MIC ) , or left untreated as control . Strains expressing fluorescent protein fusions were grown at 30°C in the presence of appropriate inducer concentrations ( S1 Table ) until an OD600 of 0 . 3 . Cells were subsequently treated with 0 . 5 μg/ml ( 1x MIC ) rhodomyrtone or 1% DMSO ( negative control ) . Unless otherwise stated , samples for microscopy were withdrawn after 10 and 30 min . Cells were immobilized on a thin film of 1% agarose and immediately observed using an Olympus BX 50 microscope equipped with a Photometrics CoolSNAP fx digital camera . Images were analyzed with Image J . B . subtilis HS63 ( divIVA-msfgfp ) was grown as described above . Cells were mounted on 1% agarose patches containing 10% LB and were placed into a pre-warmed flow chamber ( Ibidi sticky-Slide VI 0 . 4 ) . Cells were kept at 30°C in a constant medium flow ( 10% LB , 20 μl/minute ) and allowed to adjust for 10 min . After taking a picture of the untreated cells , rhodomyrtone , diluted in 10% LB to a final concentration of 0 . 5 μg/ml ( 1x MIC ) , was fed into the chamber with a continuous flow of 20 μl/min . Images were acquired every 2 min over the course of 30 min . Images were taken with a Nikon Eclipse Ti microscope equipped with a CFI Plan Apochromat DM 100x oil objective , an Intensilight HG 130 W lamp , a C11440-22CU Hamamatsu ORCA camera , and NIS elements software , version 4 . 20 . 01 . Images were analyzed with Image J . Exponentially growing ( 37°C ) B . subtilis 168 cells were treated with 0 . 5 μg/ml ( 1x MIC ) or 1 μg/ml ( 2x MIC ) rhodomyrtone , 1 μg/ml gramicidin S , 1 μg/ml daptomycin ( in the presence of 1 . 25 mM CaCl2 ) , or 10 μg/ml MP196 for 10 min . 200 μl of sample were withdrawn and mixed with 1 ml 1:3 acetic acid methanol . This organic fixation leads to extrusion of the protoplast through cell wall holes , which occur when lipid II synthesis is inhibited while cell wall autolysins are still active [26] . Cells were mounted on 1% agarose films and observed with phase contrast microscopy using a Nikon Eclipse Ti microscope as specified above . The membrane potential was determined with the potentiometric fluorescent probe 3 , 3′-dipropylthiadicarbocyanine iodide ( DiSC3 ( 5 ) ) using a Biotek Synergy MX plate reader as described before [28] . In short , 1 μM DiSC3 ( 5 ) was added to exponentially growing ( 37°C ) B . subtilis 168 cultures and the baseline was recorded for 5 min ( 651 nm excitation , 675 nm emission ) . Compounds were added ( 0 . 5 μg/ml ( 1x MIC ) , 1 μg/ml ( 2x MIC ) rhodomyrtone , 1 μg/ml gramicidin ( 1x MIC ) , 1% DMSO ) and samples were measured for another 25 min ( 37°C , shaking ) . Exponentially growing ( 37°C ) B . subtilis 168 cultures were treated with 0 . 5 μg/ml ( 1x MIC ) , 1 μg/ml ( 2x MIC ) , 2 μg/ml ( 4x MIC ) rhodomyrtone , 0 . 5% SDS ( positive control ) , or 1% DMSO ( negative control ) . Samples were withdrawn after 5 , 10 , 15 , or 30 min and stained with 10 μg/ml PI for 5 min in the dark ( 37°C , shaking ) . Cells were washed twice with pre-warmed phosphate buffered saline ( PBS ) and resuspended in the same buffer . PI fluorescence was measured using a Biotek Synergy MX plate reader ( 535 nm excitation , 617 nm emission ) . B . subtilis 168 was grown in Belitzky minimal medium ( BMM ) at 37°C under steady agitation until early log phase . Cells were stained with 5 μM Asante Potassium Green-2 ( APG-2 ) for 60 min in the dark ( 37°C , shaking ) . Samples were washed twice with pre-warmed BMM and resuspended in the same medium . Cells were transferred to black polystyrene 96-well plates and the baseline was recorded for 10 min ( 37°C , shaking ) using a Biotek Synergy MX plate reader ( 488 nm excitation , 540 nm emission ) . Subsequently , antibiotics were added ( 0 . 5 μg/ml ( 1x MIC ) , 1 μg/ml ( 2x MIC ) , and 2 μg/ml ( 4x MIC ) rhodomyrtone , 1 μg/ml ( 1x MIC ) and 2 μg/ml ( 2x MIC ) gramicidin ( positive control ) , 1% DMSO ( negative control ) ) and fluorescence was measured for additional 20 min ( 37°C , shaking ) . The reductive capacity of B . subtilis 168 cells , indicative of respiratory chain activity , was measured with resazurin as described before [19] . Exponentially growing cells ( 37°C ) were treated with 0 . 25 μg/ml ( 0 . 5x MIC ) , 0 . 5 μg/ml ( 1x MIC ) , and 1 μg/ml ( 2x MIC ) of rhodomyrtone , 100 μM CCCP , or 150 μM sodium azide . Aliquots were withdrawn after 10 and 30 min and adjusted to an OD600 of 0 . 15 with pre-warmed LB and 100 μg/ml resazurin was added . After 5 min incubation ( 37°C , shaking ) the optical density was measured at 540 and 630 nm using a Biotek Synergy MX plate reader . All membrane-specific experiments were performed at 30°C . Unless otherwise stated in the figure legends , exponentially growing cells were treated with 0 . 5 μg/ml rhodomyrtone ( 1x MIC ) for 10 min . Membranes were stained with 2 μg/ml FM5-95 ( Molecular Probes ) for 10 min immediately prior to microscopy . Nucleoids were stained with 1 μg/ml DAPI ( Thermo Scientific ) for 2 min immediately prior to microscopy . For RIF staining with DiIC12 ( Anaspec ) overnight cultures were diluted 1:200 in LB containing 1% DMSO , 2 μg/ml DiIC12 . Cells were washed four times and resuspended in pre-warmed LB containing 1% DMSO , prior to antibiotic treatment . Laurdan ( Sigma Aldrich ) microscopy was performed as described previously [19] . Analysis of laurdan microscopy images was performed with Image J using the ‘calculate GP’ plugin . GP calculation from microscopy pictures requires the following steps: correction of the image size , brightness and contrast enhancement , background subtraction , and GP calculation . Since the brightness and contrast adjustments must be the same for both pictures , cells are typically surrounded by a certain fluorescence background . Similarly , intracellular background fluorescence results in GP values in the cytosol of the cells , which are not representative of membrane fluidity . To clearly identify the membrane in these images , black and grey GP images were overlaid with the 460 nm fluorescence image ( red ) . All microscopy was performed using the Nikon Eclipse Ti as specified above . Liposomes were either prepared from Escherichia coli polar lipid extract , POPG , or POPC ( Avanti Polar Lipids ) using a detergent-dialysis method described before [24] and extruded 20 times through 0 . 4 μm ( spectroscopy ) or 0 . 8 μm ( microscopy ) filters . Concentrated ( 10 mg/ml ) liposome stock solutions were diluted to 1 mg/ml working solutions in 5 mM Tris , pH 7 . 4 ( spectroscopy ) or 50 mM Tris , pH 7 . 4 ( microscopy ) . Liposomes were stained with 10 μM laurdan for 30 min or with 1 μg/ml DiIC12 for 2 min . Liposome samples were kept at 30°C during all experiments . Compound to lipid ratios in vivo were calculated based on lipid per cell data published for E . coli ( 2 . 2x107 lipids per cell ) [73] , using our standard treatment conditions ( OD600 of 0 . 3 , 0 . 5 μg/ml rhodomyrtone ) considering that a B . subtilis cell is twice as long as an E . coli cell and assuming that all rhodomyrtone would be bound to the membrane . This calculation resulted in an in vivo compound to lipid ratio of 1:20 . However , since it is unknown how much rhodomyrtone is bound to the membrane , free in the medium , or sequestered in the B . subtilis cell wall , this number can only be considered a rough estimate of the magnitude of the in vivo compound to lipid ratio . Membrane fluidity was determined in batch culture with laurdan as described before [19] . Cells were grown until mid-log phase , stained with 10 μM laurdan , washed four times in PBS supplemented with 2% glucose and 1% DMF , and resuspended in the same buffer to give an OD600 of 0 . 3 . Liposomes were stained with 10 μM laurdan for 30 min . Laurdan fluorescence was measured in a BioTek Synergy MX plate reader using 350 nm excitation and 460 and 500 nm emission wavelengths . Readings were taken every 2 min . Laurdan generalized polarization ( GP ) values were calculated with the following formula: ( I460-I500 ) / ( I460+I500 ) . Antibiotics were added to cells or liposomes after 4 min of measurements and fluorescence was further measured over 30 min . Overnight cultures were diluted 1:100 and grown at 30°C under steady agitation . At an OD600 of 0 . 3 , cultures were split and 500 ml were treated with 0 . 25 μg/ml ( 0 . 5x MIC ) rhodomyrtone or 1% DMSO as negative control . Lower concentrations of rhodomyrtone were necessary for these experiments because of different growth rates of B . subtilis in large cell culture flasks compared to microtiter plates due to different aeration ( S23 Fig ) . Cells were further grown to an OD600 of 0 . 6 and subsequently chilled on slush ice for 10 min . Cells were harvested by centrifugation and washed once in ice-cold 100 mM NaCl . Washed cell pellets were immediately flash-frozen in liquid nitrogen and lyophilized at -50°C overnight . Freeze-dried pellets were sonicated and dissolved in hexane prior to analysis by gas chromatography-mass spectrometry ( GC-MS ) as described by Medema et al . [74] . In short , samples were taken up in 1 ml transmethylation reagent ( 3 M HCl in methanol ) and incubated in the presence of 10 nmol internal standard ( the methyl ester of 18-methylnonadecanoic acid ) at 90°C for 4 hours . After cooling , the aqueous layer was extracted with 2 ml hexane , dried under nitrogen flow and resuspended in 80 μl hexane . One microliter of this solution was injected into a gas chromatograph ( Hewlett Packard GC 5890 ) equipped with an Agilent J&W HP-FFAP , 25m , 0 . 20mm , 0 . 33μm GC Column . Eluting fatty acid methyl esters were detected by flame ionization . Fatty acid concentrations were calculated using the known amount of internal standard and expressed as percentage of the total amount of lipids . Experiments were performed in biological duplicates . 3D SIM was performed using a Nikon Eclipse Ti N-SIM E microscope setup equipped with a CFI SR Apochromat TIRF 100x oil objective ( NA1 . 49 ) , a LU-N3-SIM laser unit , an Orca-Flash 4 . 0 sCMOS camera ( Hamamatsu Photonics K . K . ) , and NIS elements Ar software . Cultures were grown and mounted on microscopy slides as described above . Cell membranes were stained with 1 μg/ml mitotracker green or 1 μg/ml nile red for 2 min . Liposomes were stained with 1 μg/ml DiIC12 . For observation of samples stained with nile red or DiIC12 coverslips were coated with poly-dopamine as described earlier [28] . B . subtilis 168 was aerobically grown in LB until an OD600 of 0 . 3 and subsequently treated with 0 . 5 μg/ml ( 1x MIC ) rhodomyrtone or left untreated as control . After 10 min of antibiotic treatment , 200 μl of sample were withdrawn , pelleted by centrifugation ( 16 , 000x g , 1 min ) , and resuspended in 10 μl fresh LB . Concentrated cell suspensions were spotted on 0 . 25 mm thick ( Gene Frame AB0576 , ThermoFischer Scientific ) 1 . 5% agarose patches , and allowed to dry . Agarose patches were transferred to aluminum dishes and cells were fixed with 5% glutaraldehyde in 0 . 1 M cacodylate buffer ( pH 7 . 4 ) for 20 min . After fixation , samples were washed three times with 0 . 1 M cacodylate , pH 7 . 4 for 5 min each , followed by incubation in 1% OsO4 ( EMS ) / 1% KRu ( III ) ( CN ) 6 ( Sigma Aldrich ) for 30 min and three times washing with ultrapure water for 5 min each . Dehydration was performed in a series of incubation steps with rising concentrations of ultrapure ethanol as follows: 5 min 30% ethanol , 5 min 50% ethanol , twice 15 min 70% ethanol , 1 h 80% ethanol , 15 min 90% ethanol , 15 min 96% ethanol , 15 min 100% ethanol , 30 min 100% ethanol , water-free . Cells were then incubated for 30 min in a 1:1 mixture of EPON and propylene oxide , followed by 30 min incubation in 2:1 EPON / propylene oxide . Agarose patches were then transferred to fresh aluminum dishes , covered with fresh EPON , and incubated at 65°C for 48 h . After embedding , a region of interest was selected by observing the EPON-embedded bacterial layer under a light microscope and mounted for thin sectioning . Ultrathin sections ( ~80 nm ) were cut parallel to the bacterial layer , collected on single-slot , Formvar-coated copper grids , and subsequently counterstained with uranyl acetate ( Ultrostain I , laurylab ) and lead citrate ( Reynolds ) in a Leica EM AC20 ultrastainer . Bacteria were imaged at 6000x magnification using a JEOL 1010 transmission electron microscope at an electron voltage of 60 kV using a side-mounted CCD camera ( Modera , EMSIS ) . To mimic the B . subtilis membrane , a phospholipid bilayer composed of 3:1 1-palmitoyl-2-oleoyl-phosphatidylglycerol ( POPG ) and 1-palmitoyl-2-oleoylphosphatidylethanolamine ( POPE ) was used [75 , 76] . Two membrane systems were created , one with ( S ) - and one with ( R ) -rhodomyrtone . The bilayer system contained a total of 60 lipids in each leaflet and about 4000 water molecules were built . Sodium ions were added to neutralize the negatively charged head groups of each POPG molecule . Either ( R ) - or ( S ) -rhodomyrtone was added randomly . All system preparations were simultaneously performed using Packmol software [77] . The dimension size was in a rectangular box of 60 Å × 60 Å × 120 Å . All simulations were subjected to 1000 steps using the steepest descent algorithm , followed by 1000 steps of conjugate gradient minimization . To equilibrate the system , 0 . 5 ns NVT and 5 . 5 ns NPT simulations were performed at 310 K and 1 atm . Production simulations ( NPT ) for each lipid system were then conducted in the NPT ensemble at 310 K and 1 atm for 220 ns . All simulations used periodic boundary conditions , particle mesh Ewald with a 10 A° cut-off , and the SHAKE algorithm to constrain the bonds containing hydrogen atoms . All MD simulations were carried out with PMEMD implemented in AMBER16 package . The first 120 ns simulation was omitted and the last 1000 equidistant snapshots from 100 ns simulation were taken for an average and analysis . The analysis was carried out using the CPPTRAJ module in AMBER16 package and custom Fortran 95 written programs . Visualization was operated using Visual Molecular Dynamics ( VMD 1 . 9 . 3 ) suite [78] . Probability per lipid of lipid type interacting with rhodomyrtone was calculated as follows: Probabilityperlipid=NumberofframesoflipidinteractingwithrhodomyrtoneNumberoftotalsimulatedframesNumberoflipidsinthemembranestructure All methods were carried out in accordance with relevant guidelines and regulations . Danio rerio ( zebrafish ) were handled in compliance with the local animal welfare regulations and maintained according to standard protocols ( zfin . org ) . The breeding of zebrafish in authorized institutions such as the Amsterdam Animal Research Center of the VU University Amsterdam is in full compliance with the Dutch law on animal research . All animal experiments are supervised by the local Animal Welfare Body ( Instantie voor Dierenwelzijn , IvD ) of the VU University and the VU University Medical Center ( IvD VU/VUmc ) . All used research protocols adhere to the international guidelines on the protection of animals used for scientific purposes , the EU Animal Protection Directive 2010/63/EU , which allows zebrafish embryos to be used up to the moment that they are able to independently take up external food ( 5 days after fertilization ) without additional approval by the Central Committee for Animal Experiments in the Netherlands ( Centrale Commissie Dierproeven , CCD ) . Because the zebrafish embryos used in this study meet these criteria , this specific study was therefore approved by the IvD VU/VUmc . Casper zebrafish embryos were infected at 1 day post fertilization in the caudal vein by microinjection with green fluorescent Streptococcus pneumoniae D39 strain , in which the superfolder green fluorescent protein is fused to HlpA , as previously described [56 , 79] . The embryos were treated 45 min after infection either by microinjection of rhodomyrtone into the caudal vein or by addition of the compound to the water at the indicated concentrations . Infected embryos were monitored at specific time-points with a Leica MZ16FA fluorescence microscope attached with a Leica DFC420C camera . All experiments were performed in triplicate . B . subtilis 168 was grown over night at 37°C . Stationary phase cultures were then incubated with 4 or 40 μg/ml of rhodomyrtone or gramicidin S , respectively , for 9 h at 37°C under continuous shaking . CFU counts were determined by plating dilution series in LB plates without antibiotics .
|
Bacterial antibiotic resistance constitutes a major public healthcare issue and deaths caused by antimicrobial resistance are expected to soon exceed the number of cancer-related fatalities . In order to fight resistance , new antibiotics have to be developed that are not affected by existing microbial resistance strategies . Thus , antibiotics with novel or multiple targets are urgently needed . Rhodomyrtone displays excellent antibacterial activity , has been safely used in traditional Asian medicine for a long time , and resistance against this promising antibiotic candidate could not be detected in multiple passaging experiments . Here we demonstrate that rhodomyrtone possesses a completely novel mechanism of action , which is opposed to that of existing cell envelope-targeting drugs , minimizing the risk of cross-resistance , and in fact rhodomyrtone is highly active against e . g . vancomycin-resistant Staphylococcus aureus . Thus , rhodomyrtone is an extremely interesting compound for further antibacterial drug development .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Material",
"and",
"methods"
] |
[
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"drugs",
"membrane",
"staining",
"bacillus",
"microbiology",
"membrane",
"proteins",
"prokaryotic",
"models",
"antibiotics",
"experimental",
"organism",
"systems",
"pharmacology",
"cellular",
"structures",
"and",
"organelles",
"bacteria",
"bacterial",
"pathogens",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"lipids",
"medical",
"microbiology",
"microbial",
"pathogens",
"cell",
"membranes",
"biochemistry",
"membrane",
"characteristics",
"cell",
"biology",
"bacillus",
"subtilis",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"peripheral",
"membrane",
"proteins",
"organisms"
] |
2018
|
The novel antibiotic rhodomyrtone traps membrane proteins in vesicles with increased fluidity
|
Marijuana and its main psychotropic ingredient Δ9-tetrahydrocannabinol ( THC ) exert a plethora of psychoactive effects through the activation of the neuronal cannabinoid receptor type 1 ( CB1 ) , which is expressed by different neuronal subpopulations in the central nervous system . The exact neuroanatomical substrates underlying each effect of THC are , however , not known . We tested locomotor , hypothermic , analgesic , and cataleptic effects of THC in conditional knockout mouse lines , which lack the expression of CB1 in different neuronal subpopulations , including principal brain neurons , GABAergic neurons ( those that release γ aminobutyric acid ) , cortical glutamatergic neurons , and neurons expressing the dopamine receptor D1 , respectively . Surprisingly , mice lacking CB1 in GABAergic neurons responded to THC similarly as wild-type littermates did , whereas deletion of the receptor in all principal neurons abolished or strongly reduced the behavioural and autonomic responses to the drug . Moreover , locomotor and hypothermic effects of THC depend on cortical glutamatergic neurons , whereas the deletion of CB1 from the majority of striatal neurons and a subpopulation of cortical glutamatergic neurons blocked the cataleptic effect of the drug . These data show that several important pharmacological actions of THC do not depend on functional expression of CB1 on GABAergic interneurons , but on other neuronal populations , and pave the way to a refined interpretation of the pharmacological effects of cannabinoids on neuronal functions .
Cannabinoids are a class of pharmacological compounds that comprise derivatives of the plant Cannabis sativa ( marijuana ) and represent one of the oldest known sources of psychotropic drugs [1 , 2] . Δ9-tetrahydrocannabinol ( THC ) is the prototypical plant-derived psychoactive cannabinoid and the main cause of the psychotropic effects of marijuana , which is the most widespread illicit drug in the world . Administration of cannabinoids to animals and humans induces a complex pattern of behavioural effects , which can be analyzed in laboratory settings [3] . In particular , in mice , cannabinoids produce a specific array of effects in the same dose range and within the same time frame . These effects , consisting of hypolocomotion , hypothermia , antinociception , and catalepsy ( impaired ability to initiate movements ) , represent the so-called “tetrad model” of cannabimimetic activity [3–5] . Despite the fact that the “tetrad” of effects does not exhaustively represent the myriad cannabinoid behavioural and autonomic actions , it is one of the best available measures of cannabimimetic activity of drugs and has been extensively used to identify and classify cannabinoid compounds [3 , 4] . The psychoactive effects of cannabinoids are mediated by the cannabinoid receptor type 1 ( CB1 ) and , in particular , the tetrad effects of THC are abolished in mutant mice lacking the expression of CB1 [6 , 7] and are blocked by CB1 antagonists [5 , 8 , 9] . Moreover , psychotropic effects of marijuana were shown to be attenuated by blockade of CB1 receptors in humans , confirming the central importance of these receptors in the pharmacology of psychotropic cannabinoids [10] . CB1 is a seven-transmembrane G protein-coupled receptor expressed at very high levels in the central nervous system and at lower levels in peripheral tissues [11 , 12] . Together with CB2 cannabinoid receptors , CB1 is the molecular target of specific endogenous lipid signalling molecules , the endocannabinoids [13–15] . Cannabinoid receptors , endocannabinoids and the enzymatic machinery for endocannabinoid synthesis and degradation constitute the endocannabinoid system , which is involved in several physiological and pathophysiological processes such as , among many others , retrograde signalling at neuronal synapses [16 , 17] , memory processing [15 , 17] , pain perception [15 , 18 , 19] , regulation of locomotion [15 , 18] , and inflammation [15 , 20 , 21] . In the brain , the expression of CB1 correlates with the psychotropic effects of cannabinoids [11 , 22] . However , the neuronal mechanisms and the neuronal circuitries responsible for these effects have not yet been clarified . In the brain , where THC and other cannabinoids exert most of their behavioural effects , CB1 receptors are expressed at different levels in different neuronal subpopulations . In particular , CB1 protein and mRNA are present at very high levels in cortical GABAergic interneurons ( those that release γ aminobutyric acid [GABA] ) , where they mediate cannabinoid-dependent inhibition of GABA release [17 , 23 , 24] . However , CB1 receptors are also expressed in other neuronal subpopulations , including , among others , glutamatergic cortical principal neurons [24–30] . Given the extraordinary high expression of CB1 on GABAergic interneurons , the modulation of the activity of these neurons is generally believed to mediate most of the effects of exogenously administered and endogenously released cannabinoids [23] . However , due to the lack of suitable experimental tools , this concept has not yet been investigated in vivo . With the advent of conditional mutagenesis techniques , which are aimed also at obtaining specific deletion of genes in particular cell types [31 , 32] , it is now possible to address directly the involvement of different neuronal populations in the pharmacological effects of cannabinoids . We used recently generated conditional mutant mice for the CB1 receptor , bearing a deletion of the CB1 gene in principal neurons ( defined as projecting neurons as opposed to interneurons , independently of their neurochemical characteristics , normally expressing Ca2+/calmodulin-dependent kinase IIα [CaMKIIα] [33 , 34] ) , in cortical glutamatergic neurons , and in GABAergic neurons [26 , 27] . Furthermore , we produced a new mouse mutant line where the Cre-mediated deletion of the CB1 gene is driven by the regulatory sequences of the D1 dopamine receptor . All of these conditional CB1 mouse mutants were tested in the “tetrad” battery of THC effects . The results indicate that the typical pharmacological effects of cannabinoids rely on complex anatomical substrates and that , at odds with previous concepts , GABAergic interneurons appear not to be involved in these effects .
The “tetrad” effects of THC were shown to depend on the expression of CB1 receptors , because these effects are abolished in CB1-null mutant mice and are blocked by CB1 antagonists [5 , 7–9] . However , discrepancies in the phenotype of CB1-null mutant lines were observed in different experimental setups , likely due to differences of test conditions and genetic background of the mice [6 , 7] . To verify that the “tetrad” effects of THC fully depend on CB1 expression under our experimental conditions and in our strain of CB1-null mutants , we first performed a dose-response ( 0 , 1 , 3 , and 10 mg THC/kg mouse body weight [mg/kg] ) study of these effects of THC in mice and then used the dose of 10 mg/kg on CB1−/− mice and their littermates CB1+/+ [35] . The results clearly showed that THC dose-dependently exerts the “tetrad” effects in our experimental conditions ( Figure 1A–1D ) and that the deletion of CB1 does not alter per se the observed basal conditions of vehicle-treated animals , but fully abolished the effects of THC ( Figure 1E–1H; interaction genotype × treatment , F1 , 34 > 9 . 6 , p < 0 . 004 ) . To dissect the neuronal circuits involved in the effects of THC , we used available conditional mutants lacking CB1 in specific neuronal populations [26 , 27] . However , these lines present overlapping patterns of deletion , which might limit the interpretation of the results . For instance , GABA-CB1−/− mice and CaMK-CB1−/− mice both lack CB1 expression in striatal medium spiny neurons ( MSNs ) [27] . As a consequence , possible differences in the effects of THC in these two mouse lines ( see below ) could be difficult to interpret . Therefore , to obtain a further specific deletion of CB1 in striatal neurons , we crossed CB1f/f mice with a transgenic mouse line where Cre recombinase is expressed under the control of the regulatory sequences of dopamine receptor D1 ( D1-Cre ) [36 , 37] to generate D1-CB1−/− mice ( Figure 2 ) . Mutant mice were fertile and did not show any obvious phenotypic alteration . In situ hybridization ( ISH ) analysis of CB1 expression revealed the expected pattern of recombination , with CB1 mRNA absent in the majority of striatal neurons but still expressed in other brain regions ( Figure 2A ) . Overall , in D1-CB1−/− mice , the great majority of MSNs of caudate putamen do not show CB1 expression , leaving only 27 . 9% of CB1 expression in the caudate putamen ( total CB1-positive cells: D1-CB1+/+ , 13 . 9 ± 1 . 3 cells/field versus D1-CB1−/− , 3 . 9 ± 0 . 2 cells/field; Figure 2A , 2B , and 2B′ ) , which is in agreement with the expected recombination pattern induced by Cre recombinase under the control of D1 regulatory sequences [36] . In the ventral striatum , owing to the very low levels of expression of CB1 mRNA [24 , 25] , it is difficult to quantify the loss of CB1 in mutant mice ( Figure 2A ) . An evaluation based on dark-field images of radioactive ISH for CB1 mRNA ( Figure 2C and 2C′ ) indicate that about 50% of neurons still express CB1 in the mutant mice as compared with wild-type littermates . Dopamine D1 receptors are expressed not only in striatal neurons , but also in other brain regions . In particular , Cre-mediated recombination under the control of D1 regulatory sequences occurs also in layer VI of the neocortex [36 , 37] . In this region , CB1 mRNA is present both at high levels in GABAergic interneurons [24] and at low levels in principal glutamatergic neurons , coexpressed with the glutamatergic neuronal marker vesicular glutamate transporter 1 [27] ( and unpublished observations ) . Due to the low levels of CB1 mRNA in these latter neurons and the fact that layer VI has a relatively small dimension , it is difficult to assess the possible deletion of CB1 by an observation of low-magnification pictures ( Figure 2A ) . Therefore , we performed a semi-quantitative counting of the CB1-expressing neurons in layer VI of the neocortex from images captured at high magnification ( see Materials and Methods ) . The results show that the number of high CB1-expressing neurons is not changed in D1-CB1−/− mice ( D1-CB1+/+ , 2 . 6 ± 0 . 4 cells/field versus D1-CB1−/− , 2 . 4 ± 0 . 4 cells/field , p > 0 . 05 , n = 5 sections per genotype; Figure 2D and D′ ) , whereas a significant reduction was observed in low CB1-expressing neurons ( D1-CB1+/+ , 17 . 3 ± 1 . 4 cells/field versus D1-CB1−/− , 7 . 7 ± 1 . 3 cells/field , p < 0 . 01 , n = 5 sections; Figure 2D and 2D′ ) . In other regions of the brain , including forebrain cortical region ( Figure 2A , 2E , and 2E′ ) , forebrain subcortical regions ( Figure 2A ) , midbrain , and hindbrain regions ( Figure 2A ) , no evident alteration was observed in the expression of CB1 in D1-CB1−/− as compared to wild-type D1-CB1+/+ . More detailed analysis of CB1 expression in the caudate putamen by double ISH revealed that CB1 is absent in the great majority of non-D2 expressing neurons of mutant mice , indicating the general deletion of the gene in neurons expressing D1 ( Figure 2F and 2G ) . However , in agreement with the expected recombination pattern of D1-Cre mice [36] , a certain number of D2-expressing neurons appear to be affected by the Cre-mediated recombination , because the number of double CB1/D2-expressing neurons was decreased in mutant mice . In fact , whereas CB1 is present in 8 . 3 ± 0 . 9 cells/field ( corresponding to 36 . 6% ± 2 . 1% of total D2-expressing neurons ) in wild-type D1-CB1+/+ , this number shows a 5-fold reduction in mutant D1-CB1−/− ( to 1 . 6 ± 0 . 2 cells/field , corresponding to 7 . 2% ± 0 . 6% of total D2-expressing neurons; p < 0 . 0001 [Figure 2F and 2G] ) . Conversely , not all the cells of caudate putamen that maintained expression of CB1 in D1-CB1−/− belong to the D2-positive population . In fact , in wild-type D1-CB1+/+ mice , single CB1-expressing neurons ( i . e . , non–D2-positive neurons ) were counted as 5 . 7 ± 0 . 5 cells/field , whereas their presence is not abolished in D1-CB1−/− ( 2 . 5 ± 0 . 1 cells/field ) . These results indicate that non-D1 , non-D2 neurons in the caudate putamen keep their expression of CB1 in the mutant mice . These neurons very likely belong to the population of striatal GABAergic interneurons , which were shown to contain CB1 mRNA [38] . Altogether , these data show that in D1-CB1−/− mice , CB1 expression is strongly reduced in the striatum , with less than 30% of neurons still containing mRNA of the receptor in the caudate putamen and approximately 50% of loss of CB1 expression in the ventral striatum . Of the remaining CB1-positive neurons , one part belongs to the D2-positive subpopulation of MSNs and the other part likely to striatal GABAergic interneurons . Moreover , a significant subgroup of presumably glutamatergic projecting neurons of layer VI of the neocortex shows deletion of CB1 in D1-CB1−/− mice . The recombination pattern of CB1 in D1-CB1−/− mice is similar to the known expression of D1 receptors . However , other cell types might be affected too , such as dopamine receptor D2-expressing MSNs , which are normally considered not to coexpress D1 receptors [39 , 40] . However , other studies report that the overlap between D1- and D2-expressing neurons in the striatum might be higher than generally believed [41] . It is not the aim of the present study to address this controversy . However , for simplicity , in the following , we will refer to D1-CB1−/− mice as mice with a deletion of CB1 in “dopamine receptor D1-expressing neurons” . This is a given definition and does not imply that Cre-mediated recombination of the CB1 gene has occurred exclusively in D1-expressing neurons . The different mutant mouse lines used in this study , their assigned nomenclature , and a general description of their pattern of expression of CB1 receptor are listed in Table 1 . The different mutant lines present specific brain regions and cell types carrying the Cre-mediated deletion of the CB1 gene [27] . However , overlapping deletions are also present between different lines . In Figure 3 , a schematic representation of the patterns of expression of the different lines is shown . Deletion of CB1 in GABAergic neurons did not alter the analgesic effect of THC . Neither baseline hot-plate escape latencies after vehicle injection ( Figure 6A , p > 0 . 05 ) nor THC-induced analgesia showed any significant difference between wild-type GABA-CB1+/+ and mutant GABA-CB1−/− littermates ( interaction genotype × treatment F1 , 57 = 1 . 7 , p = 0 . 204; Figure 6A ) . Identical results were obtained with a lower dose of THC ( 3 mg/kg ) , which was also unable to induce significantly different effects between genotypes ( two-way ANOVA; Figure S1B and Text S1 ) . Therefore , CB1 receptors that are expressed in GABAergic neurons do not appear to play a significant role in THC-induced analgesia . Vehicle-treated CaMK-CB1−/− did not show any difference in escape latency as compared to vehicle-treated control wild-type littermates ( Figure 6B ) . Conversely , wild-type CaMK-CB1+/+ mice showed normal sensitivity to the analgesic effect of THC ( p < 0 . 001; Figure 6B ) , whereas the mutant littermates CaMK-CB1−/− did not respond to the injection of the cannabinoid drug ( p > 0 . 05; Figure 6B ) , as shown by a highly significant genotype × treatment interaction ( F1 , 36 = 17 . 9 , p < 0 . 001; Figure 6B ) . These data indicate that principal neurons expressing CB1 receptors play a central role in the analgesic effect of THC . However , at odds with locomotor and hypothermic effects of THC , cortical glutamatergic neurons do not appear to contribute to the analgesic effect of THC . In fact , the normal effect of the drug observed in wild-type Glu-CB1+/+ was present also in mutant Glu-CB1−/− ( Figure 6C ) with no difference between genotypes ( interaction genotype × treatment , F1 , 33 = 0 . 1 , p = 0 . 772 ) , indicating that CB1-postive principal neuronal populations outside of the neocortex are likely involved in these effects of THC . The deletion of CB1 receptors in dopamine receptor D1-expressing neurons also did not alter the analgesic effects of THC . The drug exerted analgesic effects in both D1-CB1+/+ and D1-CB1−/− littermates ( p < 0 . 01; Figure 6D ) , without any significant difference between genotypes ( interaction genotype × treatment , F1 , 27 = 0 . 5 , p = 0 . 483 ) , indicating that CB1 expressed in the majority of striatal neurons and in layer VI of neocortex are unlikely to mediate this effect of THC . THC-induced catalepsy was normally present in both GABA-CB1+/+ and GABA-CB1−/− mice ( p < 0 . 01 and p < 0 . 001 , respectively , Figure 7A ) , without any difference between genotypes ( statistics not shown ) . A dose-response study confirmed the lack of statistically significant difference between the cataleptic effects of THC in the two groups of mice ( two-way ANOVA , Figure S1E and Text S1 ) . Conversely , the cataleptic effect of 10 mg/kg THC could be observed in CaMK-CB1+/+ ( p < 0 . 001; Figure 7B ) , but not in CaMK-CB1−/− mice ( p > 0 . 05; Figure 7B ) . Two-way ANOVA revealed a highly significant difference in the effect of THC between the two genotypes ( interaction genotype × treatment , F1 , 36 = 44 . 6 , p < 0 . 0001 ) , indicating the dependency of this effect of THC on principal neurons . However , deletion of CB1 from glutamatergic cortical neurons is not sufficient to impair the cataleptic effect of THC . In fact , Figure 7C shows that the drug was able to induce the effect both in Glu-CB1+/+ and in Glu-CB1−/− without any significant interaction between treatment and genotype ( statistics not shown ) . Interestingly , 10 mg/kg THC did induce a strong cataleptic effect in D1-CB1+/+ mice ( p < 0 . 001; Figure 7D ) , whereas D1-CB1−/− mice were insensitive to this action ( p > 0 . 05; interaction genotype × treatment , F1 , 31 = 55 . 9 , p < 0 . 0001; Figure 7D ) . Thus , GABAergic neurons and cortical glutamatergic neurons do not appear to play a central role in this effect , whereas principal neurons expressing D1 receptors appear to mediate THC-induced catalepsy .
The present study addressed the neuronal circuits mediating some of the most common effects of cannabinoids in mice , the so-called “tetrad” battery of effects [3 , 4] . To this aim , we used a combined genetic and pharmacological approach and analysed the effects of THC in previously described CB1 conditional mutant mice [26 , 27] and in newly generated mice , which lacked CB1 expression in the majority of striatal MSNs and a subset of glutamatergic neurons in layer VI of the neocortex . Taken together ( Table 2 and Figure 8 ) , our genetic and pharmacological results show that ( 1 ) various pharmacological effects of cannabinoids are mediated by different neuronal circuits , which can be dissected by genetic approaches; ( 2 ) GABAergic interneurons do not appear to mediate these effects of THC; ( 3 ) cortical glutamatergic neurons mediate a large portion of hypolocomotor effects of cannabinoids; ( 4 ) similarly , CB1 expression in cortical glutamatergic neurons plays a prominent role in mediating the hypothermic effects of THC , with a possible partial involvement of a subpopulation of MSNs; ( 5 ) the simultaneous activation of CB1 receptors located on striatal neurons and glutamatergic neocortical neurons is likely to be necessary to exert the cataleptic effect of THC; and ( 6 ) analgesic effects depend on principal neurons of the central nervous system , but their precise identity is not yet clearly identifiable . Methodological aspects have to be considered in these experiments . CB1 cannabinoid receptors are widely expressed in the central nervous system and , importantly , are present in different neuronal subpopulations [11 , 17 , 24 , 25 , 50] . This study was undertaken with the aim to dissect important pharmacological effects of THC with respect to their cellular mechanisms and , in particular , to the differential involvement of distinct neuronal subpopulations in these effects . To address this issue , we used conditional mutagenesis , and our results were able to identify several likely sites of action of THC in the brain . However , genetic manipulations are not free of possible confounding issues [51] . Indeed , deletion of a gene by gene targeting ( although in a conditional manner ) could lead to compensatory mechanisms , which might cause misleading interpretations . Moreover , the use of specific regulatory sequences to drive the expression of Cre recombinase is not devoid of caveats , because the Cre-induced recombination pattern might be different from the expected one , based on the known cell types where the regulatory sequences are supposed to drive Cre expression . For instance , the Nex-Cre mice used to generate Glu-CB1−/− mice , besides showing the expected recombination in cortical glutamatergic neurons , also showed a small degree of recombination in other neurons . The other neurons , however , likely contain very low , if any , expression of CB1 receptors [45] . Nevertheless , conditional mutagenesis is the only tool available to date to dissect with a high degree of precision the role of a given gene in different neuronal populations , and our results , though surprising to a certain extent , fit with other data and concepts present in the literature and , importantly , are confirmed by the parallel analysis of several complementary conditional CB1 mutant mouse lines . The persistence of the “tetrad” effects of THC in mice lacking CB1 receptors in GABAergic neurons ( Table 2 , Figure 8 ) is surprising . Due to the extremely high levels of CB1 receptors present on cortical GABAergic interneurons , the current hypothesis has been that this neuronal population might mediate most of the effects of exogenously applied or endogenously released ( endo ) cannabinoids [23] . The lack of THC effects is unlikely to be caused by developmental or compensatory effects of the lack of CB1 in GABAergic neurons in GABA-CB1−/− mice [52] for the following reasons . First , the same effects of THC are abolished in CaMK-CB1−/− and partially reduced in Glu-CB1−/− mice , which still express CB1 in GABAergic interneurons . Second , another important function of CB1 receptors , i . e . , the physiological protection against excitotoxic seizures induced by kainic acid , was recently shown to be preserved in GABA-CB1−/− mice and strongly impaired in CaMK-CB1−/− and Glu-CB1−/− mice [27] . Third and importantly , our previous data show that physiological retrograde release of endocannabinoids that act at CB1 receptors expressed in hippocampal GABAergic terminals , mediating short-term forms of synaptic plasticity ( depolarization-induced suppression of inhibition , DSI ) , is abolished in GABA-CB1−/− mice [27] , thereby indicating that these mice indeed lack functional expression of CB1 receptors . The overall normal pharmacological effects of THC in GABA-CB1−/− mice , accompanied by the absence of DSI , and the strong reduction of the “tetrad” effects in CaMK-CB1−/− and Glu-CB1−/− mice , accompanied by normal expression of DSI [27] , show that these particular endocannabinoid-dependent electrophysiological phenomena are not involved in the “tetrad” pharmacological effects of cannabinoids . These results then suggest that other effects of CB1 activation , such as depression of glutamatergic transmission , might play a more relevant role in this context . Cannabinoids can control glutamatergic transmission via CB1 receptors in several brain regions [28 , 53–56] . Recently , using GABA-CB1−/− and CaMK-CB1−/− , we could confirm that exogenous application of cannabinoids controls glutamatergic transmission through CB1 receptors present on glutamatergic neurons of the amygdala , the hippocampus , and the neocortex [27 , 29] . Therefore , it is possible that control of glutamatergic transmission represents the most important neuronal mechanism underlying “classical” effects of cannabinoids in mice . In particular , hypolocomotor and hypothermic effects of THC appear to depend to a large extent on the functional expression of CB1 on cortical glutamatergic neurons , because they are strongly reduced both in CaMK-CB1−/− and in Glu-CB1−/− mice ( Table 2 ) . In particular , corticostriatal glutamatergic projection neurons might indeed represent the main site of action of THC to induce hypolocomotor effects ( Figure 8A ) by reducing the excitatory input onto basal ganglia , as previously shown by electrophysiological recordings [43] . Striatal neurons , being GABAergic principal neurons , are depleted of CB1 expression both in GABA-CB1−/− mice ( where THC has a normal effect on locomotion ) and in CaMK-CB1−/− mice ( where the hypolocomotor effect is strongly reduced ) . Given the strong impairment of this effect in Glu-CB1−/− mice ( which express normal levels of CB1 mRNA in the striatum ) [27] , the presence of CB1 protein in corticostriatal glutamatergic neurons appears as a plausible candidate to mediate the hypolocomotor effects of THC in mice . To verify this hypothesis , we used mice lacking CB1 expression in the majority of striatal neurons ( D1-CB1−/− mice ) . In these mice , the effect of THC was only very slightly reduced . It is , therefore , possible that a subpopulation of striatal neurons partially contribute to the hypolocomotor effect of THC . However , D1-CB1−/− mice lack CB1 also in a subgroup of pyramidal neurons in the layer VI of neocortex . Given the normal effect of THC in GABA-CB1−/− mice , which lack CB1 in the totality of striatal neurons , but still express normal levels in pyramidal cortical neurons , the most parsimonious interpretation of the data taken as a whole is that the absence of CB1 in layer VI pyramidal neurons of the neocortex likely accounts for the slight reduction of hypolocomotor effect of THC in D1-CB1−/− mice ( Figure 8A ) . However , although striatal projections have been described [57] , layer VI pyramidal neocortical neurons project mainly to thalamic regions and their possible function in control of locomotion has been proposed but is still not fully elucidated [58 , 59] . It is important to note that the hypolocomotor effect of THC in CaMK-CB1−/− , Glu-CB1−/− , and D1-CB1−/− is not completely abolished in any of the three lines . This might be due to residual expression of CB1 in other brain regions . For instance , cerebellar neurons are not affected by Cre-mediated recombination in either of the lines [27] ( Figure 2 ) . Given the importance of this brain region in controlling locomotion , it is very likely that the conserved expression of CB1 in the cerebellum accounts for the portion of maintained effect of THC . Concerning hypothermic effects , a pharmacological synergy between N-methyl-D-aspartic acid ( NMDA ) glutamate receptor antagonists and cannabinoids was recently shown in rats [46] . Therefore , inhibition of glutamatergic transmission is a likely mechanism of THC-dependent hypothermia . Moreover , the preoptic anterior hypothalamic nucleus has been proposed to mediate the hypothermic effects of cannabinoids , where they likely regulate glutamatergic transmission [60] ( and references inside ) . Therefore , our data , by showing that hypothermic effects of THC are reduced in CaMK-CB1−/− and Glu-CB1−/− ( Table 2 and Figure 8B ) , are in agreement with the notion that a CB1-dependent reduction of glutamatergic transmission ( possibly at the level of cortico-hypothalamic projections ) is the main mechanism of THC-induced hypothermic effects . The reduced and shorter lasting effect of THC in D1-CB1−/− mice is more difficult to interpret . On the one hand , the absence of CB1 expression from a subpopulation of glutamatergic neurons of neocortical layer VI might account for the phenotype of these mutant mice . However , a very recent publication suggested that the hypothermic effect of THC and its neuroprotective consequences during cerebral infarction might rely on CB1 expressed in cortical and striatal regions [61] . This suggests that specific striatal neurons might participate in the hypothermic effects of THC . Hypoalgesic effects of THC , although likely independent from GABAergic interneurons , do not appear to depend on CB1 expressed on cortical glutamatergic neurons nor on striatal neurons , because they are normally present in Glu-CB1−/− and D1-CB1−/− mice , respectively ( Table 2 , Figure 8C ) . The exact determination of the neuronal circuitries and the brain regions responsible for each single effect of cannabinoids will require further studies using even more sophisticated experimental approaches , such as viral-induced cell-type–specific deletion of the CB1 gene . However , the present data already allow some speculation . For analgesic effects , regions such as the periaqueductal grey might play an important role [19 , 62–64] . However , given the relatively low levels of CB1 expression in these regions of the central nervous system , it is difficult to evaluate the possible lack of expression of CB1 receptors in CaMK-CB1−/− mice , which do not show THC-induced hypoalgesia . Nevertheless , our data clearly show that GABAergic neurons are not implicated in this effect . Moreover , it is possible that hypoalgesic effects of cannabinoids depend on the expression of CB1 at spinal sites and/or even in peripheral neurons [19 , 64 , 65] . Indeed , recent data clearly showed that conditional deletion of the CB1 gene in peripheral neurons strongly reduces analgesic effects of cannabinoid drugs [65] . In addition , interaction with noradrenergic neurons in the spinal cord might be a mechanism of THC-induced hypoalgesia [66] . Therefore , given the wide expression of CaMKIIα at several sites ( likely including also spinal and peripheral neurons ) , the lack of hypoalgesic effect of THC in CaMK-CB1−/− cannot , at the moment , be ascribed to a precise location in the complex pathways mediating pain perception . Cataleptic effects of THC are impaired both in CaMK-CB1−/− and in D1-CB1−/− mice ( Table 2 , Figure 8D ) . The basal ganglia are involved in the adjustment and fine tuning of voluntary movements through two major pathways: the D1-type dopamine receptor-containing direct and the D2-type dopamine receptor-containing indirect pathways [42] . Interestingly , catalepsy can be observed by pharmacological manipulation of both pathways , and its exact mechanisms are not yet understood in detail [67–69] . Moreover , catalepsy is one of the hallmark effects of treatments , such as 6-OHDA and reserpine , that are able to induce parkinsonian symptoms in animals [56] . In CaMK-CB1−/− mice , CB1 receptors are lost in all projecting neurons of the brain , thus making it difficult to define the exact site of action of THC to induce catalepsy . However , noradrenergic and serotonergic pathways ( involving 5HT1a receptors ) were proposed to play a role in THC-induced catalepsy [70 , 71] . Noradrenergic and serotonergic neurons likely express very low , but significant levels of CB1 [72 , 73] . As these neurons are principal projecting neurons , it is possible that they lack CB1 expression in CaMK-CB1−/− . Thus , lack of THC-induced control of serotonergic and/or noradrenergic transmission might participate in the phenotype of CaMK-CB1−/− . In the D1-CB1−/− mice , on the other hand , most incoming striatal afferents from cortex and thalamus still express CB1 normally ( apart from some neurons of neocortical layer VI ) . However , MSNs of the direct pathway ( traditionally considered as D1-positive ) [57] lack CB1 receptors , together with a certain amount of putative neurons belonging to the D2-positive subpopulation ( believed to constitute the indirect pathway ) . Therefore , cannabinoid treatment cannot affect GABA release at the output sites in the globus pallidus and the substantia nigra . Yet , GABA-CB1−/− mice , lacking CB1 from all GABAergic neurons ( including all striatal principal neurons ) express catalepsy normally after THC treatment . Cannabinoid-induced catalepsy , therefore , is produced most probably by a simultaneous disturbance at more than one site in the basal ganglia motor pathway . In this regard , the lack of CB1 from a subgroup of pyramidal neurons in the neocortical layer VI of D1-CB1−/− mice might be particularly interesting . Indeed , it is possible to speculate that the simultaneous action of THC at CB1 receptors expressed both in layer VI cortical neurons and in D1-expressing MSNs is necessary to exert the typical cataleptic effect of the drug . In this frame , the presence of CB1 in cortical glutamatergic neurons of GABA-CB1−/− mice would be sufficient to exert the normal cataleptic effect of THC . Interestingly , MSNs belonging to the direct and indirect pathways were recently shown to be differentially regulated by endocannabinoids , and this phenomenon might have a particular importance in the pathophysiology of Parkinson disease [56] . Therefore , it will be very interesting to explore the phenotype of D1-CB1−/− in models of this disease , in order to start dissecting the loci where cannabinoid-based therapy might be useful in the treatment of this important neurological disorder [15 , 18 , 42] . The “tetrad” battery of effects does not represent the myriad pharmacological functions of cannabinoids , ranging from effects on learning [74 , 75] , stress responses [12 , 75] , neuroendocrine and energy balance [12] , reward [76] , and many others [15] . However , the fact that it is now possible to dissect each of these effects will pave the way to a novel concept of cannabinoid pharmacology and to new insights into its mechanisms . This might also include the future possibility for therapeutic targeting of specific cannabinoid effects exerted in distinct neuronal subpopulations . It is becoming more and more evident that CB1 receptors physically interact with specific proteins in different neuronal populations and that these interactions are able to modify their pharmacological profile [77–80] . Therefore , it will be possible in the future to identify cannabinoid ligands that are able to interact with CB1 receptors specifically coupled or uncoupled with other proteins and , thereby , expressed in different populations . The use of these drugs and the precise determination of the sites of action of cannabinoids for different effects might provide the opportunity to exploit better the therapeutic potentials of cannabinoids , avoiding possible undesirable side effects .
Male mice , aged 2–5 mo , were used in all experiments , maintained in standard conditions with food and water ad libitum . All experimental procedures were approved by the Committee on Animal Health and Care of the local government . Conditional CB1 mutant mice were obtained by using the Cre/loxP system [31] . The respective Cre-expressing mouse line was crossed with CB1f/f mice [26] , using a three-step breeding protocol . CaMK-CB1−/− , Glu-CB1−/− , and GABA-CB1−/− mice were obtained as described [26 , 27] . Genotyping was performed by PCR as described for CaMK-CB1−/− and for CB1f/f [26] . For the GABA-CB1−/− line , genotyping for the Cre transgene was performed by PCR using the following primers: forward 5′-GAT CGC TGC CAG GAT ATA CG; reverse: 5′ - CAT CGC CAT CTT CCA GCA G , whereas genotyping for the CB1f/f locus was performed as described [26] . CB1+/+ and CB1−/− mice were generated and genotyped as described [35] . All lines were in a mixed genetic background , with a predominant C57BL/6NCrl contribution . All animals used in single experiments were littermates . Experimenters were always blind to genotype and treatment . The abbreviations used to identify the different mutants and their wild-type littermates are summarized in Table 1 . To generate the D1-CB1−/− line ( Figure 1 ) , CB1f/f mice [26] were crossed with dopamine receptor D1-Cre line [36 , 37] , in which the Cre recombinase was placed under the control of the dopamine receptor D1A gene ( Drd1a ) regulatory sequences using transgenesis with modified bacterial artificial chromosomes . The pattern of Cre expression recapitulated the expression pattern of the endogenous Drd1a [36 , 37] . Genotyping for the Cre transgene was performed by PCR using the following primers: forward 5′-GAT CGC TGC CAG GAT ATA CG; reverse: 5′ - CAT CGC CAT CTT CCA GCA G , whereas genotyping for the CB1f/f locus was performed as described [26] . Δ9-tetrahydrocannabinol ( THC , Sigma-Aldrich; http://www . sigmaaldrich . com ) was purchased as a 10 mg/ml ( w/v ) solution in 100% ethanol . This solution was concentrated to 100 mg/ml using a SpeedVac and , immediately before injection , mixed with Tween 80 , and then diluted with 0 . 9% saline and shaken for 10 min at 37 °C . Vehicle control contained all ingredients ( 1 drop / 3 ml of Tween 80 and ethanol diluted 1:40 with saline ) except Δ9-THC . Drug/vehicle was administered intraperitoneally with an injection volume of 10 ml/kg body weight . For behavioural tests , mice of each genotype received different doses of THC or vehicle , as previously described [4 , 81] . To determine the effects of THC the following number of animals were used in the experiments ( Veh indicates vehicle ) : GABA-CB1+/+-Veh , 10; GABA-CB1+/+-THC , 19; GABA-CB1−/−-Veh , 10; GABA-CB1−/−-THC , 20; CaMK-CB1+/+-Veh , 9; CaMK-CB1+/+-THC , 10; CaMK-CB1−/−-Veh , 10; CaMK-CB1−/−-THC , 11; Glu-CB1+/+-Veh , 9; Glu-CB1+/+-THC , 10; Glu-CB1−/−-Veh , 8; Glu-CB1−/−-THC , 9; D1-CB1+/+-Veh , 9; D1-CB1+/+-THC , 9; D1-CB1−/−-Veh , 10; D1-CB1−/−-THC , 11 . Animals were placed in a reversed dark/light cycle ( light off 8 a . m . , light on 8 p . m . ) for at least 15 d before the experiments . All behavioural tests were performed in the dark phase ( between 10 a . m . and 3 p . m . ) under dim red light illumination in the following order . Basal body temperature was measured in all animals before injection . One hour after the injection of THC or vehicle , temperature was again measured and animals were placed in the open field to assess locomotion for 15 min . Immediately afterwards , the mice were placed on the hot-plate test and analgesia was measured for maximum of 60 s ( see below ) . Mice returned to their home cage for about 14 min and then catalepsy was tested . After returning to home cage for another 13–14 min , the temperature was measured again ( 2 h after drug injection ) . Locomotion was measured 60 min after injection of THC or vehicle by an automated open field system ( box size 32 × 32 cm; illumination of 0–10 lux , MOTION , TSE GmbH; http://www . tse-systems . com ) . Animals were individually tested for 15 min . The cumulative horizontal distance the animals moved within the box was recorded . THC-induced analgesia was measured using a hot plate analgesia meter ( type 12801 , Bachofer Laboratoriumsgeräte , Reutlingen , Germany ) 75 min after injection of the drug or the vehicle . The plate was heated to 55 ± 0 . 5 °C and the time until mice showed the first sign of discomfort ( licking or flinching of the paws or jumping on the plate , here defined as escape latency ) was recorded . A cut-off time of 60 s was set to prevent tissue damage . Body temperature was measured immediately prior , as well as 60 and 120 min after injection of drug or vehicle using a C-1600 infrared thermometer ( Linear Laboratories; http://www . linearlabs . com ) , which was placed between the forepaws at a distance of approximately 3 cm from the abdomen . THC-induced catalepsy was measured by the bar catalepsy test 90 min after drug or vehicle injection . The forepaws of mice were placed on a 1-cm-diameter bar fixed horizontally at 3 . 5 cm from the bench surface . The descent latency was recorded for an observation period of 20 s . Single and double ISH was carried out as previously described [24 , 82] using radioactive and nonradioactive CB1- and radioactive dopamine receptor D2-specific riboprobes [24 , 82] , respectively . Counting of cells expressing CB1 in the VI layer of the neocortex of D1-CB1−/− and control D1-CB1+/+ littermates was carried out on single radioactive ISH experiments . Random regions corresponding to the VI layer of neocortex ( sensory-motor part , right above the corpus callosum ) were captured at 40× magnification . Numbers were randomly assigned to the images , and an observer ( blind of the genotype of each single image ) counted high CB1-expressing neurons ( corresponding to GABAergic interneurons and defined as in [24] ) and low CB1-expressing neurons ( belonging to pyramidal glutamatergic cortical neurons ) [27] . Numbers of low and high CB1-expressing neurons per analyzed field were calculated . For CB1/D2 coexpressing neurons in the striatum , a similar approach was used in double ISH slides . Numbers of single CB1- , single D2- and double-expressing neurons were calculated . Data are presented as mean ± standard error of the mean ( SEM ) of individual data points . Vehicle-mediated effects were compared between genotypes in absolute values . For open field , data are expressed as percentages of vehicle-treated wild-type animals ( absolute data of vehicle-injected mice are reported in the text ) . Body temperature data are expressed as differences from vehicle-treated animals of the same genotype ( defined as ΔT in the figures ) , with absolute values of vehicle-treated mice reported in the text . Data were analysed using two-way ANOVA , using genotype and treatment as variables , and Newman-Keuls post-hoc test . In some cases , to appreciate differential effects of THC between different strains , percentage “relative effects” of THC were calculated for each genotype and analyzed with Student's t-test . Graphs and statistics were generated by GraphPad Prism 4 . 03 ( GraphPad Software; http://www . graphpad . com ) and Statistica 5 . 0 ( StatSoft; http://www . statsoft . com ) , respectively .
|
Marijuana and its main psychoactive component , THC , exert a plethora of behavioural and autonomic effects on humans and animals . Some of these effects are the cause of the widespread illicit use of marijuana , while others might be involved in the potential therapeutic use of this drug for the treatment of several neuronal disorders . The great majority of these effects of THC are mediated by cannabinoid receptor type 1 ( CB1 ) , which is abundantly expressed in the central nervous system . The exact anatomical and neuronal substrates of each action are , however , not clearly known at the moment . We addressed this issue by using an advanced genetic approach . Control and conditional mutant mice , lacking CB1 expression in defined neuronal subpopulations but not in others , were treated with THC , and typical effects of the drug on motor behaviour , pain , and thermal sensation were scored . Our results show that different neuronal subpopulations mediate different effects of THC and could lead to a refined interpretation of the pharmacological actions of cannabinoids . Moreover , these data might provide the rationale for the development of drugs capable of selectively activating CB1 in specific neuronal subpopulations , thereby better exploiting cannabinoids' potential therapeutic properties .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"anesthesiology",
"and",
"pain",
"management",
"pharmacology",
"neuroscience"
] |
2007
|
Genetic Dissection of Behavioural and Autonomic Effects of Δ9-Tetrahydrocannabinol in Mice
|
The pathogenic fungus Cryptococcus neoformans uses the Bwc1-Bwc2 photoreceptor complex to regulate mating in response to light , virulence and ultraviolet radiation tolerance . How the complex controls these functions is unclear . Here , we identify and characterize a gene in Cryptococcus , UVE1 , whose mutation leads to a UV hypersensitive phenotype . The homologous gene in fission yeast Schizosaccharomyces pombe encodes an apurinic/apyrimidinic endonuclease acting in the UVDE-dependent excision repair ( UVER ) pathway . C . neoformans UVE1 complements a S . pombe uvde knockout strain . UVE1 is photoregulated in a Bwc1-dependent manner in Cryptococcus , and in Neurospora crassa and Phycomyces blakesleeanus that are species that represent two other major lineages in the fungi . Overexpression of UVE1 in bwc1 mutants rescues their UV sensitivity phenotype and gel mobility shift experiments show binding of Bwc2 to the UVE1 promoter , indicating that UVE1 is a direct downstream target for the Bwc1-Bwc2 complex . Uve1-GFP fusions localize to the mitochondria . Repair of UV-induced damage to the mitochondria is delayed in the uve1 mutant strain . Thus , in C . neoformans UVE1 is a key gene regulated in response to light that is responsible for tolerance to UV stress for protection of the mitochondrial genome .
The ability to sense light provides well-known advantages to organisms , such as adapting to photosynthetic light sources in plants and for vision in animals , yet the benefits of light-sensing in non-photosynthetic and non-motile organisms are less established . The fungi contain a suite of potential photoreceptor proteins , with the White Collar complex ( WCC ) being found throughout the fungal kingdom , except for species in which the two genes encoding the complex were lost . Light influences different responses in different fungi , including phototropism , induction of pigmentation , asexual and sexual sporulation , changes is primary and secondary metabolism , and regulating the circadian clock: most of which are controlled , where established , by the WCC [1]–[3] . A major question is what advantage is provided in using light as an environmental signal to regulate these processes . One compelling hypothesis is that protecting DNA from damage provides a selective pressure , and that wavelengths in the visible spectrum are sensed to indicate the presence of deleterious ultraviolet radiation . However , a DNA repair system common to light-sensing fungi and that acts directly downstream of the WCC is unknown to date . The White collar-1 and White collar-2 proteins interact to form a complex ( WCC ) capable of sensing blue and near UV light . Both proteins were originally characterized in the ascomycete fungus Neurospora crassa [4]–[7] where WC-1 acts as a photoreceptor . In N . crassa , WCC has roles to play both in light and dark environments . In the dark the WCC has a major function as a circadian clock component , via regulation of FRQ gene expression [8]–[10] . Light-dependent functions of WCC include conidiation , carotenoid production and mating [11] , [12] . The photons and the signal are transduced via the chromophore flavin adenine dinucleotide ( FAD ) bound within an N-terminal specialized type of PAS domain ( from the Per , Arnt , Sim proteins ) , named the LOV ( light , oxygen and voltage ) domain [13] . The two proteins interact using other PAS domains , to form the transcription factor complex that regulates transcription of target genes via their GATA-type zinc finger DNA-binding domains [4] , [5] , [7] , [14] . The human pathogen Cryptococcus neoformans is a member of the phylum Basidiomycota , a distant relative to N . crassa . The fungus grows vegetatively as a budding yeast and during mating in a dikaryotic filamentous form . C . neoformans is divided into two varieties: var . grubii is the most prevalent in the clinic and var . neoformans is less common but was the most amenable to experimental methods until the discovery of an opposite mating partner for var . grubii about a decade ago [15] . C . neoformans primarily causes disease in immunocompromised people . The closest relative to C . neoformans is C . gattii , which is also a human pathogen but with a greater tendency to infect immunocompetent individuals [16] . Homologs of N . crassa wc-1 and wc-2 are present in Cryptococcus species , designated BWC1/CWC1 and BWC2/CWC2 , and have been characterized in strains of both var . grubii and var . neoformans [17] , [18] . The Bwc1-Bwc2 complex has three known functions in C . neoformans: it represses mating in the light , promotes virulence , and provides protection against UV light . The downstream targets of Bwc1-Bwc2 that control these functions remain to be elucidated . Several genes have been identified from C . neoformans that are regulated by light . These include SXI1α and MFα1 , found within the mating type locus and encoding a transcription factor and pre-pheromone protein required for sexual reproduction [17] . Their expression could explain the repression of mating by light . However , the regulation of these transcripts was compared after a long time exposure of 24 h in the light or constant darkness , such that this time point likely reflects indirect regulation by Bwc1-Bwc2 . More recently a microarray experiment was carried out in C . neoformans to detect light-regulated transcripts with an hour of exposure to light . The HEM15 gene , encoding ferrochetalase that is the last step in the heme biosynthetic pathway , was identified as a gene under control of Bwc1-Bwc2 [19] . However , the phenotype of the knockout hem15 strain differs significantly from mutation of BWC1 or BWC2 . For example , HEM15 is essential for viability , and yet the bwc1 and bwc2 mutants do not show any growth defects in the light or dark . Another light-regulated gene is CFT1 , required for iron uptake and virulence [20] , yet no iron-dependent phenotype of bwc1 and bwc2 mutants is known . Thus , while CFT1 and HEM15 may be targets of the Bwc1-Bwc2 complex , they likely play minimal roles in the physiological response of C . neoformans to light . Moreover these two genes are also under the control of other transcription factors in addition to WCC . As a transcription factor complex , it was puzzling that more light-regulated genes were not identified by microarray analysis that could potentially explain the phenotypes of deleting the WCC from C . neoformans . The microarray results of C . neoformans contrast to ascomycete species . For instance , in N . crassa a microarray study in wild type , wc-1Δ , and wc-2Δ strains at different time periods identified 314 light-regulated genes , constituting 5 . 6% of the total detectable transcripts in the genome [11] . These genes were grouped into two broad categories , the early or late light responsive genes ( ELRGs and LLRGs ) . Some of these ELRGs are involved in the synthesis of vitamins , photo-protective pigments , prosthetic groups and cofactors , cellular signaling , DNA processing , circadian rhythm and secondary metabolism . Many of the LLRGs are implicated in carbohydrate metabolism , fatty acid oxidation and free radical detoxification . In N . crassa there is also a link between the WCC and DNA repair , for example through the clock gene prd-4 , which is a cell cycle checkpoint kinase 2 [21]–[23] . As the White Collar complex acts as an UV/blue light photoreceptor and mutation of the complex causes an increase in sensitivity to UV light , potential targets are hypothesized to be genes involved in repairing DNA damage for survival under UV stress . The UVE1 gene of C . neoformans was previously identified in a UV sensitive strain in a collection of insertional mutants [24] . The product of UVE1 is a homolog of an apurinic/apyrimidinic endonuclease that is best characterized in fission yeast Schizosaccharomyces pombe . In S . pombe the gene was called UVDE for UV damage endonuclease , and renamed uve1 for consistency with nomenclature [25] , [26] . The mus-18/UVE-1 gene is the homolog characterized from N . crassa [27] . The protein removes UV-induced cyclobutane pyrimidine dimers and 6-4 photoproducts , acting in its own pathway termed the UVDE-dependent excision repair ( UVER ) pathway . UVDE recognizes single-stranded DNA nicks , apurinic/apyrimidinic sites , and nucleotide mismatches [26] , [28]–[30] , a suite of DNA lesions that also extends a possible role for UVDE in repairing the equivalent types of DNA damage caused by reactive oxygen species [31] . In S . pombe , UVDE is localized and functional in both the nucleus and mitochondria , and was suggested to act as a reserve mechanism for repairing UV-induced DNA damage in the mitochondria [32] . Homologs of UVE1 are present in a subset of species in the Archaea , Bacteria and Eukaryotes [25] , [33]–[37] , with one exception being humans where there is no homolog . A preliminary northern blot experiment suggested that UVE1 in C . neoformans var . neoformans is a light-regulated gene with two isoforms , triggering this investigation . We hypothesized that UVE1 is a downstream target of the WCC that functions in repairing UV-induced damage , and tested this hypothesis in the experiments described below .
A T-DNA insertion mutant within the promoter of the UVE1 gene was identified previously [14] . Under UV stress conditions the mutant showed negligible survival as compared to the wild type ( KN99α ) and the unexposed strains ( Figure 1A ) . Transformations derived from Agrobacterium T-DNA delivery can have phenotypes that are not due to the insertion of the T-DNA into the host genome . The UV sensitivity phenotype of the original mutant was verified by constructing UVE1 gene replacement strains for both C . n . var . neoformans and C . n . var . grubii . The knockout strains in both varieties had reduced survival after exposure to UV ( Figure 1A , B ) . To confirm that the UV sensitivity phenotype was because of the absence of UVE1 , an uve1Δ strain was complemented with a wild type copy of UVE1 . The UVE1 complemented strain completely rescued the UV sensitivity phenotype ( Figure 1 ) . These results show that in C . neoformans the UVE1 gene is required for survival under UV stress . The responses of the uve1 mutant to stresses other than UV light were tested , showing no other phenotypes ( Figure S1 ) . The gene also plays no major role in the formation of mating filaments ( Figure S2A , B ) . A prior analysis , using comparative growth in a pool of 48 strains in the mouse lung , suggested the UVE1 gene has no role in virulence [38] . To examine the uve1Δ strain in isolation , its virulence was tested in an insect model . While the bwc1Δ strain is less pathogenic than wild type in this model , the uve1Δ strain is equally virulent as wild type ( Figure S2C ) . Thus , the only established function of Uve1 in C . neoformans is in response to UV stress . Since UVE1 is required for surviving exposure to UV light , regulation of UVE1 was tested in response to light . Wild type strains of C . n . var . neoformans , C . n . var . grubii and C . gattii were grown in the dark , and one replicate provided a one hour exposure to white light . Expression of the UVE1 gene was examined by northern blot analysis on polyA RNA purified from these cultures . The UVE1 transcript levels were higher in light grown conditions for all wild type strains ( Figure 2 ) . Two transcripts were observed for var . neoformans strain JEC21 , one longer ( L , for light ) induced specifically under light conditions and another shorter ( D , dark ) expressed in the dark . For var . grubii and C . gattii one isoform of UVE1 was expressed in response to light with negligible expression in dark grown conditions . The role for the Bwc1 photoreceptor in UVE1 induction by light was examined in the bwc1Δ deletion strains of C . neoformans . In the var . neoformans deletion there was loss of induction of the longer isoform by light , and residual expression of this isoform in both light and darkness . There was no observed effect of bwc1Δ on the shorter isoform in dark grown conditions . Dark isoforms expressed equally well in both light and dark conditions in some replicates , possibly reflecting the age of the culture or shading by other cells . In the var . grubii bwc1Δ mutant , the UVE1 transcript was barely detectable under either illumination regime ( Figure 2 ) . These analyses in the bwc1 mutant backgrounds indicate that UVE1 expression is controlled by the Bwc1-Bwc2 complex . Rapid amplification of cDNA ends ( RACE ) was used in the var . neoformans strain to define the 5′ and 3′ ends of the two transcripts that were produced in the light and the dark . The UVE1 dark isoform is part of the UVE1 light isoform , with the dark isoform starting in the middle of the light isoform and both sharing a common 3′ end ( GenBank accessions KF234405 and KF234406; Figure 3 ) . Alignment of the Uve1 homologs from various fungi indicated that the dark isoform of C . neoformans is truncated and missing key residues in the active site of this protein ( Figure 3; Figure S3 ) . The light-dependent regulation of UVE1 in wild type strains suggested that induction of the gene prior to UV exposure would correlate with increased UV resistance . The wild type , bwc1Δ and the complemented strains were grown in complete darkness overnight . One set was kept in the dark and the other set exposed to light for 2 hours , prior to treatment of both sets with UV light ( Figure S4 ) . All three strains grown in darkness showed similar levels of sensitivity to UV light . Exposure of the strains with the wild type copy of BWC1 to light before UV stress increased their resistance to UV , a property not seen for the bwc1Δ strain . Hence , light signaled by Bwc1 promotes UV resistance . Two isoforms of UVE1 were observed in the var . neoformans strain , raising the possibility that differential transcript sizes may be a common feature for DNA repair genes in C . neoformans . To identify other genes involved in protecting the fungus against UV damage , a collection of 1200 defined knock out mutants in the var . grubii background [38] was screened for those sensitive to UV light . 13 strains were identified , including the uve1Δ strain in the collection ( Figure S5A ) . For the corresponding genes , northern analysis were performed for var . neoformans and var . grubii cultured under light and dark conditions ( Figure S5B ) . No altered size or induction in response to light was observed , as had been for the UVE1 gene . To examine if UVE1 photoregulation is common among fungi , northern blot analysis of UVE1 was performed in two fungi , Neurospora crassa and Phycomyces blakesleeanus ( Figure 4 ) . The species are light-sensing models in the phylum Ascomycota and subphylum Mucoromycotina , respectively . In N . crassa the expression of UVE1 has already been reported in a microarray study of the light-induced genes [11] . For the N . crassa wild type strain we observed by northern blot analysis the induction of one transcript in light grown conditions and minimal expression of the UVE1 transcript in dark grown mycelia , confirming the previous microarray data . For the P . blakesleeanus wild type strain two transcripts were present in light grown conditions that were both absent in samples grown in the dark . Based on the expressed sequence tag information in the genome database , the 5′ end of the UVE1 homolog is shared between the two transcripts and the 3′ end differs , the reverse of the situation in C . neoformans var . neoformans . The longer isoform in P . blakesleeanus has a 3′ extension due to transcriptional read-through into the 3′ neighboring gene . We also examined the transcript profile of UVE1 in a white collar-1 mutant ( wc-1 ) of N . crassa and a madA-madB mutant of P . blakesleeanus ( madA and madB are functional wc-1 and wc-2 homologs in this species [39] ) . We observed complete loss of light-dependent induction of UVE1 in these mutant strains of N . crassa and P . blakesleeanus . As an additional control to show that UVE1 induction was not an indirect effect of light on the media , UVE1 expression was examined in S . pombe , a “blind” species because it encodes no homologs of the WCC . We observed equal transcript levels of the S . pombe UVE1 homolog in cultures grown under light and dark conditions ( Figure 4 ) . These studies demonstrate that UVE1 is photoregulated among highly-diverged fungal species , and substantiates the WC-1 dependent light-induction of UVE1 in the fungal kingdom . The protein sequences predicted for the two isoforms in var . neoformans from RACE were examined by bioinformatic approaches for their subcellular localization . PSORT II and MitoProt analysis of UVE1 light and dark isoforms predicts the longer form to be most likely mitochondrial and no specific localization pattern for the dark isoform . To confirm these predictions , light and dark isoforms of UVE1 were fused to the N-terminal end of GFP and expressed in the uve1Δ strain AI191 . To assess mitochondrial localization we used MitoTracker red , which specifically stains respiring mitochondria . Confocal fluorescence microscopy for the GFP-fused light form of Uve1 showed co-localization of Uve1-GFP with MitoTracker red , giving a yellow fluorescence in merged images ( Figure 5A ) . No expression of Uve1-GFP light form was observed in the nucleus , confirmed by co-staining with Hoechst ( Figure S6 ) . For the dark isoform of Uve1-GFP , GFP localization was throughout the cell , but clearly excluded from the mitochondria ( Figure 5B; Figure S6 ) . Transformation of the UVE1-GFP constructs into a uve1Δ genetic background enabled a test of their functionality in complementing the UV sensitive phenotype . Expression of the light form of UVE1 rescued in part the UV sensitive phenotype of strain AI191; however , the dark isoform fused to GFP did not . These experiments suggest that the light isoform of UVE1 is localized solely to the mitochondria , and as such it protects the mitochondrial genome from lethal effects of UV-induced DNA damage in C . neoformans . The function of Uve1 in the fungi at a biochemical level is best characterized in S . pombe [25] , [29] , [40] , [41] . Alignment of the two homologs suggests that they are similar , sharing residues within the active site of the enzyme ( Figure S3 ) . To infer functional similarity , a cross-species complementation test was performed . An uve1Δ knockout was generated in S . pombe by replacing the gene via homologous recombination with the KanMX cassette that confers resistance to G-418 . We then expressed in this S . pombe knockout strain the cDNA clones of light or dark isoforms of UVE1 from C . neoformans var . neoformans . The UV sensitivity of the S . pombe strains was tested ( Figure 6A , B ) . The uve1 deletion strain was highly sensitive to UV irradiation , as was the control strain transformed with the empty vector . In the strain expressing the light isoform of UVE1 , UV sensitivity was rescued as the strain survived UV doses equivalent to the wild type strain . For the uve1::KanMX+C . neoformans dark isoform , no rescue in the UV sensitive phenotype was observed . These observations suggest that the C . neoformans light isoform of UVE1 is functionally active and has the equivalent biochemical functions of Uvde from S . pombe required for repair of DNA damaged by UV exposure . It also suggests that the dark isoform of C . neoformans Uve1 may not have any function in conferring protection against UV . C . n . var neoformans Uve1 was localized in S . pombe as a GFP fusion to see if the rescue of UV sensitivity phenotype in S . pombe is due to complementing mitochondrial or nuclear genome repair by Uve1 ( Figure 6C , D ) . The Uve1-GFP construct was functional , complementing the UV sensitive phenotype of the S . pombe uve1 mutation ( Figure 6A , B ) . The localization of Uve1 ( L ) -GFP is in part nuclear , as confirmed by co-localization with the nuclear Hoechst stain ( Figure 6C ) . No localization in mitochondria was observed ( Figure 6D ) . These results suggest that the Cryptococcus Uve1 protein , which seems to be important for mitochondrial DNA repair in C . neoformans , also plays a role in nuclear DNA repair in S . pombe . This observed localization pattern conforms to previous reports where Uve1 in S . pombe repairs nuclear DNA after UV stress , rather than mitochondrial DNA [32] . If Uve1 localizes to mitochondria in C . neoformans , it is expected to play a role in mitochondrial DNA repair in consequence of DNA damage due to UV stress . As no nuclear localization was observed for Uve1 , a negligible role of this endonuclease is expected in nuclear DNA damage repair . We performed a PCR-based DNA damage assay to assess the role of Uve1 in mitochondrial and nuclear DNA repair post-UV stress . The assay is based on the principle that damaged DNA impedes the progression of Taq polymerase on the template DNA in a PCR reaction [42] . Hence , there is an inverse relationship between the amount of DNA damage and PCR amplification products . Long template size increases the sensitivity of assay , as the longer the DNA the more chances of encountering damage ( dimers ) . Small template amplification serves as a control , minimizing chances of encountering a damaged DNA strand , and is used for normalization of the amount of starting DNA or mitochondrial DNA copy number . We compared DNA damage between the nuclear genome and mitochondrial genome of UV treated samples for both wild type and uve1Δ ( Figure 7 ) . After one hour there is delayed repair for both nuclear and mitochondrial genomes in the uve1Δ strain compared to wild type , probably reflecting retrograde signaling between the mitochondria and nucleus or that repair pathways of the nuclear genome can be ATP dependent [43]–[48] . The key observation is that in the uve1Δ strain there is lag in the mitochondrial repair . For later time points of 4 hour and 6 hours , the uve1Δ strain repair of mitochondrial genome is delayed as compared to wild type strain , which returned to normal ( Figure 7 ) . At 6 hours recovery , amplification of the mitochondrial genome in the wild type is as it was prior to DNA damage , but for uve1Δ the lesion damage still persists . However , the uve1Δ strain shows more efficient repair of lesions in the nuclear genome . For the 4 hour and the 6 hour time points , as the repair process of the mitochondrial genome initiates in uve1Δ by some unknown mitochondrial DNA repair enzymes , the repair of the nuclear genome is faster and comparable to wild type levels . These data implicate Uve1 in the efficient repair of UV-damaged mitochondrial DNA , with evidence for a complex interplay between mitochondrial and nuclear repair and the contributions of other repair pathways . No induction of the UVE1 transcript encoding the functional form of the protein was observed under light conditions in bwc1Δ mutants ( Figure 2 ) . We examined if UVE1 is a direct target of the Bwc1-Bwc2 complex through two approaches . First , we tested if overexpression of Uve1 could rescue the UV hypersensitive phenotype of the bwc1Δ mutant in C . neoformans . Uve1 from var . neoformans was expressed under a galactose-inducible promoter ( PGAL7 ) in the bwc1 deletion mutants of var . neoformans and var . grubii . The strains were grown overnight in media with glucose or galactose as the primary carbon source , serial diluted and plated , and UV sensitivity tests performed . Results from the var . neoformans strains are illustrated in Figure 8 , and var . grubii in Figure S7 . Uve1 overexpression rescued the UV sensitive phenotype of bwc1Δ as comparable to wild type when induced by galactose . In contrast , there was only slight rescue in strains grown in non-inducing glucose ( Figure 8 ) . For the bwc1Δ+PGAL7-UVE1 overexpression strains and wild type strains we performed northern analysis to check the levels of UVE1 induction ( Figure S8 ) . The levels of UVE1 transcripts were comparable between the galactose-induced PGAL7-UVE1 and the light-induced wild type strains . These results provide one piece of evidence for the Bwc1-Bwc2 complex directly controlling UV resistance in C . neoformans through regulation of the effector protein Uve1 . The second piece of evidence that UVE1 is a direct target of Bwc1-Bwc2 comes from gel mobility shift assays . Bwc2 has a C-terminal GATA-type zinc finger whose binding target sites are not known in C . neoformans . We searched the promoter of UVE1 for putative Bwc2 binding sites based on those used by the WCC of N . crassa [11] , [14] , [49] . For instance , one light regulated element ( LRE ) found in the frq promoter is TCGATCCGCTCGATCCCCT , with the underlined nucleotides similar to a TCGATCTTCATCTCGATCTCCA sequence found in the promoter of C . neoformans UVE1 . We amplified and radiolabeled the UVE1 promoter region with this site and performed gel mobility shift assays with recombinant Bwc2 ( amino acids 26 to 383 ) expressed and purified from Escherichia coli ( Figure 9A ) . A retardation in gel migration of the UVE1 promoter DNA was observed , that increased with higher Bwc2 concentration or by adding zinc which is expected for a zinc finger protein ( Figure 9B ) . The nature of the higher mobility forms is unknown , but may represent aggregation of Bwc2 monomers . Control interactions using a non-specific DNA fragment confirmed the specificity of Bwc2 for the UVE1 promoter . These observations indicate that the UVE1 promoter is a direct target for Bwc2 binding .
Light influences diverse aspects of fungal biology , presumably by acting on unique pathways in specific species . However , the potential for conserved regulation also exists , and this is predicted to reflect the original selective pressure ( s ) and current maintenance of light-sensing in extant fungi . Some responses to light in ascomycete fungi relate to protection against damage caused by light . For instance , expression of the DNA repair enzyme photolyase and genes for biosynthesis of carotenoid pigments are often induced by light . There are previous reports of links between light via its input in circadian rhythms with DNA repair in fungi , as well as in mice and humans . For instance PRD-4 is a checkpoint kinase 2 homolog in N . crassa that is regulated by the WCC and contributes to DNA repair [21] . Another DNA repair protein , XPA , has been shown to have circadian rhythm dependent oscillations in mouse brain [50] , [51] . However , between 2 . 8–6% of genes are regulated at the transcript level in response to a light exposure in ascomycete species [11] , [52] , [53] , yielding a long list of candidate genes for further analysis . In contrast , the basidiomycete C . neoformans may serve as a simpler model for understanding the evolution of light-sensing in fungi , because ( a ) it does not encode a photolyase gene and pigmentation is not induced by light , ( b ) few genes are induced in response to light at the transcript level [19] , ( c ) there is no evidence of photoadaptation [17] , a trait that influences the intensity of the response to light , and ( d ) the White collar complex contains only one protein with a zinc finger DNA binding domain [17] , [18] . Here , we identify the UVE1 gene as a downstream target of the WCC in C . neoformans , and show that homologs are also light regulated in species that represent two other major branches in the fungal kingdom . We suggest that UVE1 acts as the key factor controlling the UV sensitive phenotype caused by mutating BWC1 or BWC2 in C . neoformans ( Figure 1 , Figure 10 , Figure S4 ) , and likely plays similar roles in other fungi to survive the deleterious effects of sunlight , which has UV wavelengths as an inevitable DNA damaging component . Northern blot analysis and characterization of 5′ and 3′ ends of C . neoformans var . neoformans UVE1 showed two transcripts of differing size for the UVE1 gene . The functional complementation experiments ( Figure 6 ) in S . pombe uve1Δ by the homologous UVE1 long isoform implies that the C . neoformans protein has similar DNA repair activities; such as against cyclobutane pyrimidine dimers , 6-4 photoproducts , apurinic/apyrimidinic sites , and stretches of single-stranded DNA nicks or gaps . Moreover the localization of C . neoformans Uve1-GFP in S . pombe is in part nuclear rather than mitochondrial ( Figure 6 ) . This explains the functional complementation of the UV stress tolerance phenotype in S . pombe uve1Δ strains , as in S . pombe the nuclear UVER pathway is attributed for survival under UV stress [32] . The UVE1 short isoform did not show any functional complementation . Bioinformatic analysis to identify the active domain of UVE1 ( pfam03851 ) provides a possible explanation for the inactivity of the short isoform , because it does not encode the complete conserved region ( Figure 3 , Figure S3 ) . Another possibility may be the absence of subcellular localization signals on the dark form , such that the protein is rendered inactive due to improper compartmentalization . Correct subcellular localization of proteins involved in DNA repair has been implicated in countering genotoxic stress [54] , as improper localization can result in loss-of-function and may even lead to disease development in humans [55] , [56] . The light isoform of Uve1 in Cryptococcus localizes to the mitochondria , as shown by Uve1-GFP fusion studies . C . neoformans is an obligate aerobe: it cannot survive loss of mitochondrial function from something like unrepaired DNA damage . Based on the following evidence: ( a ) no observed localization of Uve1 in nucleus ( Figure S6 ) ; ( b ) localization of Uve1 to mitochondria ( Figure 5 ) ; ( c ) strains with the UVE1 gene deleted exhibited reduced survival under UV stress and Uve1-GFP partially complements the UV sensitive phenotype of the uve1Δ strain; and ( d ) reduced mitochondrial DNA damage repair in uve1Δ strain comparatively to the wild type strain ( Figure 7 ) suggest that Uve1 in Cryptococcus is required for protection of mitochondrial DNA for survival under UV stress . Overexpression of UVE1 in bwc1 mutants of either var . grubii or var . neoformans restores UV sensitivity to the wild type level ( Figure 8 ) , and recombinant Bwc2 physically binds to the promoter of UVE1 to cause a gel mobility shift ( Figure 9B ) . These data further corroborate the hypothesis that UVE1 is a direct downstream target of Bwc1 . The light-induced genes in C . neoformans were previously sought by a whole genome microarray expression analysis in var . neoformans [19] . The UVE1 gene was not identified in that study although UVE1 transcript data are present for five of the six biological replicates , with an average 1 . 15 fold difference between dark and light treatment . By contrast , quantification by ImageJ of our northern blot data normalized to actin levels indicates that the longer isoform is upregulated about 16 fold in the light ( Dataset S1 ) . UVE1 remained undetected in microarray experiments because the 70-mer probe on the array is common to both light and dark transcript isoforms ( Figure 3 ) . The dark isoform must be under control of another transcription factor , using elements within the light isoform to drive transcription . These findings demonstrate one limitation of the microarray technique in comparison to more comprehensive transcript analysis techniques like tiling arrays or RNA-seq , or to conventional hybridization techniques like northern blotting that can detect alternative transcripts . Two other phenotypes associated with mutation of BWC1 or BWC2 in C . neoformans are loss of the inhibition of mating by light and reduced virulence . To further verify the role of Uve1 in other BWC related phenotypes , we examined the role of Uve1 in C . neoformans mating by crossing uve1Δ strains of both mating types under light and dark conditions . We did not find any contribution of this gene in mating , as the uve1Δ strains behaved like wild type for repression of mating by light ( Figure S2 ) . Similarly , the phenotype of the uve1Δ mutant in animal studies does not phenocopy that of bwc1Δ or bwc2Δ . A large-scale analysis of virulence has been undertaken in C . neoformans , measuring competitive survival of strains in mouse lungs [38] . Both bwc1Δ and bwc2Δ strains showed reduced proliferation , consistent with their reported role in virulence [17] . In contrast , the uve1Δ mutant had no defect in this virulence assay . We corroborated these results in a wax moth larvae model of virulence ( Figure S2 ) . We also examined a possible role of UVE1 under oxidative stress based on the mild phenotype observed in S . pombe [31] , but did not find any phenotypic difference in the uve1Δ strains compared to wild type . Thus , we postulate that Bwc1-Bwc2 modulates its function via more than one downstream target ( Figure 10A ) ; of which those for virulence and mating remain to be discovered in future studies . We propose a model for the relationship between light-sensing via the WCC and Uve1 function in the mitochondria in C . neoformans ( Figure 10B ) . It is possible that this model applies also to other fungal species for the protection of mitochondrial , nuclear or both genomes under UV stress . The White Collar complex is conserved across the fungal kingdom , with homologs present in the chytrids , Mucoromycotina , Glomeromycotina , Ascomycota and Basidiomycota . However the complex has been lost in some fungal lineages , like its absence from the Saccharomycotina [1] , [2] , [57] . One important question is whether any WCC downstream targets are conserved . Transcript comparisons by northern blot analysis in N . crassa and P . blakesleeanus , members of the Ascomycota and Mucoromycotina , demonstrate that UVE1 homologs are photo-regulated in at least one species in each of these fungal groups . The UVE1 homolog is also induced by light , as measured by microarray studies , in Aspergillus nidulans and N . crassa [11] , [53] . Absence of regulation in bwc1 and madA-madB mutants in N . crassa and P . blakesleeanus further implicate White Collar-dependent regulation of UVE1 . This suggests that in fungi that have White Collar , UVE1 is regulated in a light-dependent manner , and this regulation is lost or alternative regulation evolved in fungal species missing White Collar proteins . The presence of LOV domain containing flavin-binding photoreceptor proteins and UVE1 homologs in bacteria , like Bacillus subtilis [25] , [58] , warrant further examination if through convergent evolution the LOV domain type photoreceptors might be involved in regulation of UVE1 expression even more widely . The repair of photo-damage by Uve1 is conserved in many fungi and important under UV stress , irrespective of the presence of base excision ( BER ) or nucleotide excision repair ( NER ) pathways . Repair of DNA damage from UV by Uve1 is faster in comparison to NER [40]; under ancient environments in which ultraviolet levels were higher than today Uve1 could have provided a selective advantage . We estimate from genome sequencing projects that 95% of the fungal genomes encoding WC-1 also encode a copy of UVE1 . Many of the exceptions have homologs of photolyase present in their genome , which may play an equivalent role as Uve1 , and can also repair mitochondrial DNA [59] , [60] . In summary , light triggers a number of physiological and morphological changes in fungi . The advantages of using light as a signal that are conserved have remained unclear although there is increasing evidence for a role in protecting cells from damage . Here , we demonstrate that protection of DNA , including the mitochondrial genome , through photo-regulation of Uve1 provides a benefit that is present in fungi that are able to sense light through the White Collar Complex .
Gene knockout cassettes were constructed by fusion of around 1000 bp flanks 5′ and 3′ of the UVE1 gene with nourseothricin acetyltransferase ( NAT ) coding sequence for strain JEC21 ( var . neoformans , serotype D ) and neomycin phosphotransferase ( NEO ) coding sequence for strain KN99α ( var . grubii , serotype A ) . Oligonucleotide primer sequences are listed in Table S1 . To make JEC21 uve1Δ , 5′ and 3′ gene flanks were amplified by primer set AISV030/AISV034 and AISV032/AISV035 , respectively , using JEC21 genomic DNA . To make KN99α uve1Δ , 5′ and 3′ gene flanks were amplified by primer set ai830/ai831 and ai832/ai833 , respectively , using KN99α genomic DNA . The NAT and NEO ORFs were amplified by primer set ai290/ai006 . Overlap PCR was performed to obtain 5′-UVE1-NAT-UVE1-3′ and 5′-UVE1-NEO-UVE1-3′ cassettes by mixing equimolar ratio of 5′-UVE1 , NAT , UVE1-3′ for JEC21 and 5′-UVE1 , NEO , UVE1-3′ for KN99α . Primers used to perform overlap PCR were AISV030/AISV032 and ai830/ai833 for JEC21 and KN99α , respectively . About 2 µg of 5′-UVE1-NAT-UVE1-3′ or 5′-UVE1-NEO-UVE1-3′ cassette was transformed into strains JEC21 and KN99α using biolistic delivery with a PDS 1000/He particle delivery system ( Bio-Rad , Hercules , CA ) [61] . Gene replacement was confirmed by PCR and Southern blots for the correct integration of the gene cassette . Strains and genotypes are provided in Table S2 . For complementation of UVE1 in serotype A , a wild type copy of UVE1 was amplified with primers ALID0001 and ALID0002 and cloned into the pCR2 . 1 TOPO plasmid . The insert was excised with BamHI-XhoI and subcloned into the BamHI-SalI site of pPZP-NATcc . The plasmid was transformed into strain AI191 ( uve1::NEO ) by biolistics , with positive transformants selected for growth on yeast extract-peptone-dextrose ( YPD ) +nourseothricin ( 100 µg/ml ) plates . The plasmids used or constructed in this study are listed in Table S3 . The UV sensitivities of strains were tested by applying UV stress in an XL-1500 UV cross linker ( Spectronics Corporation , Lincoln , NE ) . Unless otherwise stated , strains were exposed to the laboratory ambient light ( 400–800 LUX ) during experiments . Cultures were prepared for C . neoformans strains KN99α , AI81 ( bwc1Δ ) , JEC21 , AI5 ( bwc1Δ ) , C . gattii ( R265 ) , S . pombe L972 , N . crassa wild-type FGSC 4200 , N . crassa ( wc-1 ) FGSC 4398 , P . blakesleeanus wild-type NRRL1555 , and P . blakesleeanus ( madA madB ) mutant L51 . All strains of each species were plated with equal optical density or numbers of spores in duplicates on 15 cm diameter petri dishes containing YPD , and kept in darkness . For N . crassa only , 50 ml liquid cultures were grown in 50 ml YPD medium . Cultures were grown for 23 h or 47 h depending on the growth kinetics of the species . On completion of 23 h or 47 h , one of the sets was exposed to cool white light ( dual Sylvania 4100 K 32W bulbs ) of 1600–2600 Lux for 1 h and the other set left in the dark . On completion of the 24 h or 48 h time periods both the light and dark cultures were scraped in the light or under safe red light ( GBX LED safelight , Kodak , Rochester , NY ) . N . crassa cultures were harvested directly from the liquid medium . All cultures were pelleted , frozen using dry ice+ethanol , lyophilized and stored at −80°C . Total RNA was isolated using Trizol reagent ( Invitrogen , Grand Island , NY ) . To address the low transcript abundance of UVE1 , total RNA was further purified for polyA mRNA isolation starting with 1 mg total RNA , using the PolyATract Kit ( Promega , Madison , WI ) , except for P . blakesleeanus and C . gattii where 40 µg of total RNA were used . RNA samples were resolved on 1 . 4% agarose denaturing formaldehyde gels and blotted on to Zeta Probe membrane ( Bio-Rad ) . Probes for northern analysis were amplified using specific primer sets ( Table S4 ) and radiolabeled with [α-32P] dCTP ( PerkinElmer , Waltham , MA ) using the RediPrime II labeling kit ( Amersham , Pittsburg , PA ) . The blots were stripped and re-probed with fragments of actin homologs as loading controls . Autoradiograms were scanned , and transcript levels were compared by ImageJ analysis ( Dataset S1 ) . The localization of the two isoforms of Uve1 within the cell was assessed by fusions to green fluorescent protein ( GFP ) . C . n . var . neoformans strain JEC21 genomic DNA was used as the template to amplify the UVE1 light ( L ) isoform using primers AISV001/AISV003 and UVE1 dark ( D ) isoforms using primers AISV002/AISV003 . The histone 3 promoter for Cryptococcus ( PH3 ) and GFP-NAT were amplified from the pPZP-GFP-NATcc plasmid using ai255/AISV005 ( overlap primer for L ) or ai255/AISV006 ( overlap primer for D ) for PH3 , and AISV004/ai256 for GFP-NAT . The overlap construct was amplified by mixing equimolar ratios of the three amplicons for the light and dark isoforms using primers M13F and M13R . About 2 µg of gel purified UVE1 L and D overlap constructs were transformed into strain AI191 by biolistics . Positives clones were selected by their growth on YPD+100 µg/ml nourseothricin plates and confirmed by PCR , DNA sequencing , western blotting for GFP , and fluorescence signal . Strains AISVCN28 and AISVCN02 were used for localization of the Uve1 light and dark isoforms fused to GFP . Strains were stained with MitoTracker Red CMXRos ( Invitrogen ) at 3 nM , kept in the dark for 20 min , washed and suspended in phosphate buffered saline ( PBS ) and used for microscopy . Cells were imaged using Olympus confocal microscopes FLUOVIEW FV10i or FV300 . Cultures for C . neoformans strains KN99α ( wild type ) and AI191 ( uve1Δ ) were grown overnight in YPD and washed with distilled water . Cells were suspended in phosphate buffered saline to 4×104 cells/ml . For each strain totals of 150 ml cells were distributed in 30 ml aliquots for time points 0 min ( after stress ) , 1 h , 4 h , 6 h and control ( no stress ) . For each aliquot the cells were placed in 15 cm petri dishes and exposed to UV light ( 50 J/m2 ) using a UV cross linker . Immediately after the UV stress , cells were transferred to 50 ml tubes and kept on ice . Control and 0 min cells were pelleted and frozen in liquid nitrogen . For 1 h , 4 h and 6 h time points , cells were re-suspended in YPD and incubated at 30°C for these respective times , then centrifuged to pellet and snap frozen . All samples were lyophilized , and DNA was extracted by the CTAB buffer method [62] . The relative DNA damage to the mitochondrial and nuclear genomes were assessed using a PCR assay based on established methods [42] . Concentrations of DNA samples from each treatment were standardized by measuring them by spectrophotometry and making appropriate dilutions . Primers used for amplification of fragments of the mitochondrial genome were AISV87/AISV91 ( 11 Kb ) . PCR conditions for long mitochondrial PCR were 94°C 4 min , 23 cycles for 98°C 10 s , 68°C 15 min , and a final extension of 72°C 10 min using Ex Taq ( Takara , Kyoto , Japan ) . For nuclear long amplification ( 8 kb ) primers were AISV85/AISV95 . Conditions for long nuclear PCR were 94°C 4 min , 23 cycles for 94°C 20 s , 58°C 20 s , 72°C 6 min and a final extension of 72°C 7 min . Short amplification primers for mitochondrial genome were AISV89/AISV99 and for the nuclear genome AISV85/AISV97 amplifying about 250 bp . PCR conditions for small mitochondrial and nuclear amplicons were 94°C 4 min , 23 cycles for 94°C 20 s , 55°C 20 s , 72°C 1 min , and a 72°C 7 min final extension . All PCR amplicons were resolved on agarose gels , and intensities were quantified using ImageJ software . DNA damage was compared by calculating relative amplification of large PCR fragments of the UV treated samples to that of the respective untreated controls using the method reported in reference [42] , and adjusting for differences between nuclear and mitochondrial amplicon sizes . The Galleria mellonella virulence assay followed methods that were previously described [63] . Overnight cultures in YPD medium for strains KN99α , AI191 and AI181 were washed three times with PBS . Cells were suspended in PBS to 2×107 cells/ml . For each strain 11–12 larvae were injected with 5 µl of the cells , as well as the control PBS . Wax moth were incubated at 37°C and survival monitored daily . An S . pombe uve1 knockout strain was constructed to serve for the functional analysis of UVE1 isoforms from C . neoformans . For the construction of the gene knockout cassette , genomic DNA from S . pombe strain L972 was used to amplify around 320 bp 5′-uve1 and 300 bp 3′-uve1 fragments using primer pairs AISV007a/AISV009 and AISV010/AISV011 , respectively . The KanMX fragment was amplified using primer set AISV007/AISV008 from plasmid pFA6a-GFP ( S65T ) -kanMX6 . Overlap PCR was performed to generate the gene knockout cassette using primer set AISV007a/AISV011 . Around 2 µg of the PCR construct were transformed into S . pombe ( strain MM72-4A ura4-D18 h− ) by lithium acetate transformation and cells plated on to YPD+100 µg/ml G-418 . Gene knockouts were confirmed by PCR and Southern blotting . S . pombe uve1 knockout strain AISVSP1 was selected for C . neoformans UVE1 complementation studies . UVE1 cDNA was reverse transcribed from RNA of C . neoformans strain JEC20 using Superscript III First strand Synthesis System ( Invitrogen ) , as per company instructions ( JEC20 is isogenic to JEC21 , with a different MAT allele; [64] ) . The synthesized cDNA was amplified by site directed mutagenesis to abolish an NdeI site inconvenient for subcloning while conserving the encoded amino acid residue , and to introduce NdeI and BamHI restriction sites at the start and end of the UVE1 gene . Primers used for amplification of fragment 1 of the L and D form were AISV014/AISV013 and AISV015/AISV013 , respectively . Primers used for amplification of fragment 2 were AISV012/AISV016 . Overlap PCR was performed to amplify full L and D genes from fragment 1 and 2 , using primers AISV014/AISV016 for UVE1 L form and AISV015/AISV016 for UVE1 D form . The NdeI and BamHI digested cassettes were ligated into the NdeI-BamHI site in the pREP42 vector enabling expression from an nmt promoter [39] . Positive clones were confirmed by sequencing . Plasmids containing UVE1 L and D isoforms were transformed into S . pombe strain AISVSP1 ( ura4-D18 uve1::kanMX ) by the lithium acetate method . Empty vector pREP42 was transformed into strain AISVP1 as a control . Positive S . pombe transformants were selected on minimal medium without uracil and were confirmed by PCR . C . neoformans var . neoformans Uve1 ( L ) C- terminal GFP localization in S . pombe was done by fusion of UVE1 ( L ) to GFP by overlap PCR . For fragment 1 , UVE1 ( L ) was amplified from pREP42-UVE1 ( L ) using primers AISV014/AISV003 and the fragment 2 , GFP , was derived from pPZP-GFP-NATcc using primers AISV004/AISV066 . Overlap PCR joining fragments 1 and 2 was performed by primer set AISV014/AISV066 . The overlap PCR product was cloned into pCR 2 . 1 TOPO , and transformed by heat shock into E . coli DH5α . Positive clones were selected and sequenced . A plasmid containing the desired Nde1-UVE1-GFP-BamHI overlap was digested with NdeI and BamHI . The Nde1-UVE1-GFP-BamHI digest was ligated with NdeI and BamHI digested pTN157 . Plasmid pTN157-UVE1-GFP was transformed into S . pombe strain AISVSP1 ( genotype ura4-D18 uve1::kanMX ) by the lithium acetate method . Positive S . pombe transformants were selected on minimal medium without uracil and were confirmed by PCR . Transformants were examined for fluorescence signal and their UV resistant phenotype . Confocal microscopy was performed for strain AISVSP15 . For both mitochondrial and nuclear staining , cells were grown in Edinburgh Minimal Medium ( EMM ) . Mitochondrial staining was performed with MitoTracker Red CMXRos ( Invitrogen ) at a final concentration of 3 nM in water , kept in the dark for 20 min , washed and suspended in PBS and used for microscopy . Hoechst 33342 was used to stain the nucleus . Cells grown in EMM were washed and suspended in Hoechst ( 1 µg/ml in water ) for 10 min . Cells were washed and suspended in PBS , and microscopy was performed . Overnight cultures for strains L972 , AISVSP1 , AISVSP2 , AISVSP3 , AISVSP4 and AISVSP15 in YES media were subcultured the following day . Strains in exponential phase were dotted on YES medium in ten-fold serial dilutions . One set of plates was exposed to UV light ( 120 J/m2 ) using the UV cross linker , and another plate kept unexposed . Both UV treated and unexposed sets were incubated at 30°C for 2 days . For UV dose response experiments exponentially growing cells for strains at the same optical density were ten-fold serially diluted and equal volumes for all dilutions of the cells were plated on YES media . The control was kept unexposed to UV and others were exposed at UV doses of 60 , 120 and 180 J/m2 . All plates were incubated at 30°C for 3 days , and colony forming unit data analyzed for percentage survival . The UVE1 gene was over-expressed in bwc1 knockout backgrounds to assess if UVE1 can rescue the UV sensitive phenotype of bwc1 mutation . A fusion cassette of the JEC21 GAL7 promoter ( PGAL7 ) and UVE1 gene was made by overlap PCR . The GAL7 promoter was amplified by primer set AISV025/AISV028 and UVE1 was amplified using primer set AISV027/AISV026 on JEC21 genomic DNA . The PGAL7-UVE1 fusion cassette was amplified by mixing GAL7 promoter and UVE1 PCRs in equimolar ratios by primer set AISV025/AISV026 . The PGAL7-UVE1 fusion cassette was cloned in a TA vector ( pCR 2 . 1 TOPO; Invitrogen ) , and transformed by heat shock in E . coli Top10 cells . Positive clones were selected and sequenced . A plasmid containing PGAL7 -UVE1 was digested with BamHI and XhoI . The fragment was ligated with pPZP-NEO Agrobacterium vector digested with the same enzymes , and transformed into Top10 cells . Positive clones were selected by their growth on kanamycin and verified by PCR and restriction digestion . The pPZP-NEO-PGAL7-UVE1 vector was transformed by electroporation into A . tumefaciens strain EHA105 . Selected positive Agrobacterium transformants were co-cultured with C . neoformans strains AI5 and AI81 . Positive transformants were selected for growth on YPD+G-418+cefotaxime plates . Transformants were cultured overnight in yeast nitrogen base ( YNB ) medium +2% galactose or 2% glucose , and 10-fold serial dilutions placed on YPD medium plates . Dotting was performed in duplicate and one set was exposed to UV radiation at 120 J/m2 . Both sets were incubated at 30°C for 2 days . JEC21 RNA was reverse transcribed to make cDNA using the Superscript III First strand Synthesis System ( Invitrogen ) . A fragment of BWC2 cDNA was amplified using primers AISV040/AISV041 , containing BamHI and EcoRI restriction sites . The amplicon was digested with BamHI and EcoRI , and ligated into the pRSETA vector ( Invitrogen ) digested with the same enzymes . Top10 cells were transformed with the ligation product . An error free clone was identified by sequencing . The plasmid containing BWC2 was transformed into BL21 ( DE3 ) pLysS cells and selected on ampicillin+chloramphenicol SOB medium plates . Protein induction and expression using 1 mM IPTG was performed as described for the pRSET expression system by Invitrogen . The ( Histidine ) 6-tagged Bwc2 protein was semi-purified using Pure Proteome Nickel Magnetic beads ( Millipore Corporation , Billerica , MA ) as per the manufacturer's instructions . For electrophoretic mobility shift assays ( EMSA ) , a 244 bp ( P1 ) region 5′ of the start codon of UVE1 containing putative Bwc2 binding sites was amplified using primer set AISV019/AISV020 . A control nonspecific DNA fragment ( NS ) of 278 bp was amplified from pRS426 vector using primers ALID1229/ALID1230 . About 600 ng of the amplified fragment from UVE1 promoter region and nonspecific probe was radiolabeled with [γ-32P] dATP ( PerkinElmer ) using T4 polynucleotide kinase ( New England Biolabs , Ipswich , MA ) . Labeling conditions were 1X T4 polynucleotide buffer , 5 µl 6000 Ci/mmol γ-32P ATP , 10 units T4 polynucleotide kinase in a 50 µl reaction mixture at 37°C for 30 min . The radiolabeled probes were purified using PCR purification columns ( Qiagen , Germantown , MD ) . The composition of 5X EMSA binding buffer used was 20% w/v glycerol , 5 mM MgCl2 , 250 mM NaCl , 2 . 5 mM EDTA pH 8 , 10 mM DTT with or without 12 µM ZnSO4 ( reference [65] with slight modifications ) . 20 to 50 µg of purified Bwc2 protein were pre-incubated with 1X EMSA binding buffer for 20 min at room temperature for all reactions . Total reaction mixture of 20 µl consisted of 1X EMSA binding buffer , 20 to 50 µg of purified protein , cold probe ( if added ) , 1 µl of radiolabeled probe and 1X phosphate buffer , followed by 50 min incubation at room temperature . For competition reactions , 600 ng and 900 ng of the cold probes were included in the reaction mixture prior to addition of radiolabeled probes . At the end of 50 min incubation , samples were transferred to ice followed by addition of EMSA loading dye . Samples were loaded on 6% polyacrylamide Tris-borate-EDTA ( TBE ) gels ( Invitrogen ) , and run at 130 V for 2 h in 1X TBE at 4°C . Autoradiography was performed by exposure to Gene Mate Blue ultra autoradiography films .
|
The majority of fungi sense light using the White Collar complex ( WCC ) , a two-protein combination of a photoreceptor and a transcription factor . The WCC regulates circadian rhythms , sexual development , sporulation , metabolism , and virulence . As such , the exposure to light controls properties of fungi that are beneficial and detrimental to people , depending on the species and its interaction with humans . Despite the importance of light on fungal biology , the underlying evolutionary benefit of light-sensing in fungi has remained a mystery . Here we identify a DNA damage repair endonuclease , Uve1 , required for UV stress tolerance in the human pathogen Cryptococcus neoformans . UVE1 is a direct target of the WCC in C . neoformans , and UVE1 homologs are also regulated by WCC in two other major lineages of fungi , the Ascomycota and Mucoromycotina . The divergence of the three groups indicates that for about a billion years the same transcription factor complex has regulated a common gene to protect fungal genomes from deleterious effects of light . Curiously , in C . neoformans Uve1 localizes to mitochondria and contributes to mitochondrial DNA repair , implicating its importance in genome repair of this organelle . Thus , light-sensing in fungi exists to protect them against harmful light , and likely all other responses to light relate to or are a secondary consequence of this selective pressure .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
The Uve1 Endonuclease Is Regulated by the White Collar Complex to Protect Cryptococcus neoformans from UV Damage
|
Correct repair of damaged DNA is critical for genomic integrity . Deficiencies in DNA repair are linked with human cancer . Here we report a novel mechanism by which a virus manipulates DNA damage responses . Infection with murine polyomavirus sensitizes cells to DNA damage by UV and etoposide . Polyomavirus large T antigen ( LT ) alone is sufficient to sensitize cells 100 fold to UV and other kinds of DNA damage . This results in activated stress responses and apoptosis . Genetic analysis shows that LT sensitizes via the binding of its origin-binding domain ( OBD ) to the single-stranded DNA binding protein replication protein A ( RPA ) . Overexpression of RPA protects cells expressing OBD from damage , and knockdown of RPA mimics the LT phenotype . LT prevents recruitment of RPA to nuclear foci after DNA damage . This leads to failure to recruit repair proteins such as Rad51 or Rad9 , explaining why LT prevents repair of double strand DNA breaks by homologous recombination . A targeted intervention directed at RPA based on this viral mechanism could be useful in circumventing the resistance of cancer cells to therapy .
Because genomes are subject to different kinds of insults , cells have evolved a variety of mechanisms to repair damage [1] . Homologous recombination ( HR ) , non-homologous end joining ( NHEJ ) , base excision repair ( BER ) , nucleotide excision repair ( NER ) , and mismatch repair ( MMR ) are repair systems designed to counter different kinds of damage . Inability to correct nascent mutations is an important issue in cancer . Estimates suggest that there are from 1 , 000 up to 100 , 000 somatic mutations in common adult cancers [2] . DNA viruses have discovered the value of manipulating DNA repair pathways [3] . ATM , which is activated at double-strand breaks ( DSBs ) [4] , is associated with replication of viruses like SV40 , murine polyomavirus , herpes simplex virus ( HSV ) , human cytomegalovirus ( HCMV ) , and Epstein Barr virus ( EBV ) [3] . For murine polyoma , replication is tenfold less efficient in ATM ( −/− ) fibroblasts than in wild type cells [5] . The DNA damage response contributes to SV40 DNA replication [5] , [6] , [7] . ATM phosphorylation of SV40 LT antigen is important for viral DNA synthesis [3] . A decrease in ATM function reduces SV40 DNA synthesis postponing both formation of viral replication centers and recruitment of DNA repair proteins at these sites [3] . Activation of ATM and the MRN ( MRE11/Rad50/NBS1 ) complex regulates HSV-1 replication . However , adenovirus ( Ad ) specifically inactivates the MRN complex by either mislocalization or degradation at the infection onset to promote Ad DNA replication [8] . SV40 LT deregulates multiple DNA damage pathways [4] . SV40 LT forms a tight complex with NBS1 , one member of the MRN complex [9] . Levels of MRN subunits decline during SV40 infection [10] . SV40LT expression induces promyelocytic leukemia protein interaction with RAD51 [4] . Although different kinds of repair mechanisms , each constituting a complex network of signaling components , coordinate responses to different kinds of DNA damage , a common molecular component that responds to most genotoxic insult is RPA [11] . RPA has been shown to be involved in both repair of UV damage [12] and MRN complex recruitment to DSBs induced by etoposide [13] . RPA acts as a sensor for UV induced DNA damage that recognizes cyclobutane thymine dimers and regulates the efficient removal of the lesion [14] . In addition , it participates in the formation of repair foci in response to etoposide induced DSBs [13] . Furthermore , depletion of RPA has been shown to cause spontaneous DNA damage and apoptosis in HeLa cells [15] . ATM can phosphorylate RPA [16] , [17] . This is an example of cross talk among the repair proteins and underscores the complexity of the DNA damage response ( DDR ) . Polyoma LT plays critical roles in the viral life cycle . Broadly , these can be divided into issues related to DNA replication or to control of cell phenotype . In productive infection , LT initiates viral DNA replication [18] , has helicase [19] and ATPase activities [20] and associates with pol α-primase [21] , as well as promotes integration of the viral genome into the host [22] or promotes recombination [23] . It has numerous effects on cell phenotype , many of which are dependent on its association with members of the retinoblastoma tumor suppressor family . For example , it immortalizes primary cells [24] , blocks differentiation [25] and promotes apoptosis [26] . This work describes a new connection between DNA viruses and DNA repair pathways . Binding of RPA by LT sensitizes host cells to DNA damage by as much as 100-fold . Since the same result is obtained with UV irradiation or etoposide exposure , agents that cause different kinds of lesions , multiple repair systems are being affected . Mapping indicates that binding of the origin-binding domain ( OBD ) of LT to RPA is sufficient to sensitize cells . Confirming this connection , cells overexpressing RPA are protected from LT , while knockdown of RPA triggers sensitization of cells when exposed to DNA damage even in the absence of LT . LT prevents the recruitment of RPA to DNA damage repair foci , suggesting why repair fails .
Infection with murine polyomavirus sensitized cells to DNA damage . Treatment of virus-infected secondary mouse embryo fibroblasts with 4 J/m2 dose of UV or 100 µM of etoposide at eighteen hours after infection led to rapid cell death as seen in the phase microscope at 24 hours after infection ( Fig . 1A ) . By contrast , uninfected cells were not obviously affected by UV at 40 J/m2 or 100 µM of etoposide , presumably because they could repair the DNA damage . Killing of controls comparable to that seen in the infected cells was observed only at a much higher dose of UV ( 400 J/m2 ) . This raised the possibility that polyomavirus was interfering with DNA repair . Since previous work indicated that SV40LT could interact with DNA repair proteins such as NBS1 [9] or RPA , we hypothesized that PyLT might be involved . To study whether polyoma LT affects cellular responses to DNA damage , immortalized mouse embryonic fibroblasts ( MEFs ) were prepared that conditionally expressed full-length LT using doxycycline in a tet-off system . Uninduced cells and cells expressing LT were treated with UV irradiation ( 40 J/m2 ) or with 100 µM etoposide . While UV light primarily causes photoproducts , etoposide induces strand breaks in DNA by inhibiting topoisomerase II . By 16 hours after DNA damage , MEFs expressing LT showed a dramatic change in phenotype ( Fig . 1B ) . LT expressing cells exposed to UV-irradiation or etoposide looked rounded , refractile and displayed a loss of cell-to-cell contact . Uninduced mouse embryo fibroblasts exposed to these levels of damaging agents or LT-expressing cells not exposed to DNA damaging agents did not show these morphological changes . The expression of LT in infected cells and after induction in the inducible cell line was similar ( Fig . 1C and Fig . S1 ) . Immunofluorescence showed that in each case virtually all cells expressed LT , while western blotting of cell extracts showed that the levels of LT expression were similar . Because the LT origin-binding domain ( OBD , LT residues 264–420 ) interacts with DNA , its role in sensitivity to damage was tested in cells conditionally expressing it ( Supplemental Fig . S1 ) . OBD induces dramatic changes in phenotype similar to full-length LT following UV-irradiation or etoposide treatment ( Fig . 1D ) . In general , the effects on DDRs described here for full-length LT can be demonstrated with the OBD alone . Several lines of evidence suggested that the cells were showing enhanced stress from DNA damage and were dying from apoptosis . Fig . 1E shows that LT-expressing cells have enhanced activation of JNK1 and 2 as well as p38 as determined by activation-specific phosphoantibodies after as little as 4 J/m2 UV treatments . In control cells , 400 J/m2 UV was required to produce the same activation as LT-expressing cells treated with 1/100 the dose . DAPI-staining of nuclear chromatin showed a large number of condensed and fractured nuclei in OBD- or LT- expressing cells following UV at 40 J/m2 and etoposide ( 100 µM ) ( Fig . 1F ) . DNA fragmentation , a characteristic marker of apoptosis , was seen by DNA laddering in the cells that express OBD post-UV irradiation or etoposide treatment ( Fig . 1G ) . Another marker for apoptosis is the activation and cleavage of PARP ( Poly ADP Ribose polymerase-1 ) [27] . LT enhanced the activation of poly ADP ribose polymerase ( PARP ) as seen by its cleavage ( Fig . 1H ) . Again it took 100 times as much UV to generate the same amount of PARP cleavage in control cells as in LT expressing cells . Inhibition of PARP by pretreating cells overnight with a PARP inhibitor ( 30 µM of TiQA ) prior to UV exposure had no effect on the early stress responses of Jnk and p38 activation ( not shown ) . However , the activation of PARP was important for the apoptosis , because cells were protected against either UV or etoposide when pretreated overnight with 30 µM of TiQA ( Fig . 1F , Panel 6 and 10 ) . Apoptotic cell death in response to UV [28] or etoposide [29] has been recognized for a long time . Changes in death proteins are expected in cells undergoing apoptosis . The pro-death protein BAD is upregulated in cells expressing LT treated with 4 J/m2 UV , but only in control cells when treated with 100 times the dose of UV ( Fig . 2A ) . In parallel , the pro-survival protein BclXL is down regulated in LT expressing cells and uninduced cells treated with high levels of UV . There is one difference between LT expressing cells and controls . BIM , a proapoptotic BH3 protein of the Bcl2 family disappears after UV treatment even at low UV dose in controls while it is shifted in mobility , but only slightly decreased , in LT cells . Moreover , after UV treatment in OBD-cells , Bim unexpectedly translocates to the nucleus ( Fig . 2B ) . The significance of this effect is unclear , because efficient knockdown of Bim did not protect cells from enhanced damage caused by OBD ( Fig . 2C ) . Although LT effects on survival might arise by modulating survival pathways , it seemed more likely that LT was enhancing DNA damage . Comet assays can be used to detect DNA breaks in single cells [30] . Damage is seen as a comet that can be quantified by calculating tail moments that reflect the relative amount and distribution of DNA in the tail . MEF controls or cells induced to express OBD were exposed to UV light ( 40 J/m2 ) or etoposide ( 100 µM ) . Comet tails were observed for OBD ( or LT ) expressing cells that had been exposed to DNA damage ( Fig . 3A ) . Cells that did not express LT or OBD displayed nuclear DNA without the characteristic streaming that is observed in the presence of DNA damage . Average tail moments can be calculated giving a quantitative estimate of damage [31] . In the experiment of Fig . 3B , the tail moment for LT went from 4 to 55 after UV treatment . Uninduced cells could be treated with UV to produce comets , but it again required much higher doses of UV ( 400 J/m2 ) to produce the same effect as LT at 4 J/m2 . Etoposide treatment also resulted in more DNA breaks in OBD expressing cells than in uninduced cells . A question might be whether DNA breakage seen in comet assays reflects apoptosis triggered by DNA damage treatments of cells expressing LT/OBD . Two kinds of observations argue against this . First , comet tails were observed even when cells were processed immediately after UV treatment . More convincingly , treatment with PARP inhibitor TiQA blocked death and nuclear fragmentation ( Fig . 1F ) , but had no effect on the generation of comets immediately after UV treatment ( Fig . 3C ) . Both results indicate that breakage is part of the DNA damage/repair process and not apoptosis . The next question is whether the cells expressing LT are more sensitive to the initial DNA insult , perhaps from a change in chromatin structure , or whether the effect is more downstream , at the level of DNA repair . This is most easily tested after UV irradiation . Cyclobutane pyrimidine dimer ( CPD ) and pyrimidine-pyrimidone ( 6-4 ) photoproduct ( 64PP ) , the major DNA lesions directly induced by UV irradiation , are recognizable by antibodies against the altered bases [32] . FACS analysis shows that CPD formation increases as the dose of UV increases , but expression of LT has no effect ( Fig . 3D ) . The same result is seen in Fig . 3E for 6-4 photoproducts . These experiments suggest that LT affects the repair process and not initial formation of damaged DNA . The OBD of LT is multifunctional . It binds DNA specifically at GAGGC pentanucleotides and also binds DNA in a non-site-specific manner [33] , [34] . The OBD activates transcription through CREB sites , in part by binding CREB [34] . Mutants defective in DNA-binding and activation of transcription sensitize cells to DNA damage just like wild type . Stable MEF cell lines that express mutant S306P defective for sequence specific recognition and the double mutant S306P/V358A defective even for non-specific DNA binding ( Fig . S1 ) still caused the same sensitization to DNA damage seen by morphology and comet assay as wild type ( Fig . 4A & 4B ) . Additionally , stable MEF cell lines that express mutant P402R/G403D ( PGRD ) and E343K/E344K ( 343KK ) defective in transcriptional activation ( Fig . S1 ) nonetheless sensitized cells . Mutant PGRD near the end of the OBD showed reduced transactivation of CREB responsive promoters ( Fig . 4C ) . Comet assays confirmed a significant increase in DNA damage in MEFs expressing the mutant forms of LT ( Fig . 4D ) . Neither DNA binding nor ability to activate transcription are therefore important for sensitization to damage . Since LT sensitizes cells to different kinds of DNA damage , it is plausible that some element common to repair of different kinds of damage is targeted by OBD . RPA , a heterotrimeric , single-stranded DNA binding protein is such a protein [35] . Furthermore , RPA is an indispensable component of polyomavirus DNA replication [36] , [37] . A physical interaction between SV40 LT and the RPA high-affinity ssDNA-binding domains was mapped to the SV40 OBD [38] . First , the interaction of full-length polyoma LT with RPA was demonstrated . LT was immunoprecipitated using antibody to RPA70 , and RPA was brought down by antibody to LT ( Fig . 5A ) . The RPA heterotrimer has subunits of 70 ( RPA70 ) , 32 ( RPA32 ) and 14 kDa ( RPA14 ) [35] . The small 14 kDa subunit was not found in the LT complex . This result is surprising , since RPA14 and RPA32 form a subcomplex . Most tellingly , LT mutant P402R/G403D ( PGRD ) , defective in transcriptional activation and LT mutant S306P/V358A , which is defective in both specific and non-specific DNA binding showed wild type RPA binding ( Fig . 5B ) . Sequence comparison showed that R154 , an SV40 residue critical for RPA binding [38] was conserved between SV40 and polyoma . The comparable polyoma residue , K308 , was converted to glutamate . Mutant K308E failed to bind RPA ( Fig . 5B ) . Cell lines expressing K308E were not sensitive to DNA damage . They did not show drastic morphological changes upon UV treatment ( Fig . 5C ) . A second mutant defective in RPA binding ( E320A ) was identified ( Fig . 5B ) ; it also did not cause increased DNA damage ( Fig . 5C ) . Comet assay results confirmed that the RPA binding mutant K308E fails to enhance the DNA damage response ( Fig . 5D ) , suggesting that abrogation of the interaction of LT with RPA might be able to disrupt LT's ability to increase the DDR in the host cell . To confirm that DNA repair processes requiring RPA were disrupted by LT , repair of double-strand breaks by homology directed repair was tested [39] . DR-U2OS cells were transfected with I-SceI to generate a double-strand break , and repair was measured by the recovery of intact GFP from two non-functional molecules . By flow cytometry 3 . 9% of the control cells showed recombination resulting in expression of GFP ( Fig . 5E ) . Only 1 . 3% of cells cotransfected with WT LT showed recombination , while 3 . 6% of cells cotransfected with K308E were GFP positive . This shows that LT interfered with homology-directed repair in an RPA-dependent manner . To confirm the role of RPA in sensitization of cells expressing polyoma LT following exposure to DNA damaging agents , we generated stable MEF cell lines that inducibly expressed wild type OBD and simultaneously overexpressed GFP-tagged RPA . Overexpression of RPA about three times higher than the endogenous level protected cells against DNA damage triggered by UV ( Fig . 6A & B ) . Unlike cells that express OBD alone , cells that also overexpress RPA did not show the characteristic increase in comet tail moments in their DNA ( Fig . 6C ) . A final test of the hypothesis that effects on RPA were central to LT sensitization was made by transient RPA70 knockdown . Transient knockdown of RPA70 , like LT expression , is accompanied by sensitization to DNA damage ( Fig . 6D ) and activation of stress responses ( Fig . 6E ) . Examination of LT effects on RPA localization provided clue to the problems in DNA repair . After DNA damage RPA is recruited to nuclear sites of damage repair seen as foci [40] . In Fig . 7 it is clear that when LT is expressed , RPA is diffusely nuclear , rather than localizing to the damage foci . Rad51 is critical for homologous recombination [41] . As a result of LT expression , Rad51 is also not recruited to foci after damage , explaining the defect in homologous recombination . The Rad9/Rad1/Hus1 ( 9-1-1 ) complex is a sliding clamp important for DNA repair [42] . Like Rad51 , Rad9 is prevented from reaching damage foci by LT . The RPA binding mutant K308E had no effect on localization of either Rad51 or Rad9 .
These results point to a novel connection between DNA viruses and DNA damage regulation . LT sensitizes cells as much as one-hundred fold to DNA damage from UV irradiation or etoposide . The effect of LT is somewhat reminiscent of past reports of SV40 LT and bleomycin-induced spontaneous DNA damage [29] . LT does not modulate initial DNA damage as measured by the formation of photoproducts after UV , but rather interferes with repair . The result is excessive DNA damage revealed by the comet assays leading to apoptosis . The only unusual feature of LT induced death is the stabilization of Bim and its translocation to the nucleus . This has been seen before with Human Herpes Virus-8 , which uses nuclear translocation of Bim to inhibit its activity [43] . The importance of this observation is unclear , because knockdown of Bim had no effect on phenotype . However , it remains possible that more than one member of the BH3 family is perturbed to cause the phenotype . Genetic studies and biochemical analysis identified the single-stranded DNA binding protein RPA as the target bound by LT to produce sensitivity . Here we have shown that LT through its OBD binds RPA . LT-RPA complexes differ from the endogenous complex in that the 14 kDa RPA3 subunit is lacking . This is somewhat reminiscent of the PyST/PyMT interactions with heterotrimeric protein phosphatase 2A ( PP2A ) , where the PP2A A and C subunits are found in the T antigen complexes , but the B subunit is missing [44] . SV40LT is reported to bind RPA70 constructs that contain DNA binding domains A and B [38] . PyLT also binds an RPA70 A/B construct expressed in E . coli ( not shown ) . The simplest interpretation is that there is an LT-RPA70-RPA32 complex . There may be additional interactions or steric hindrance with the heterotrimer in addition to the A/B interaction that prevent RPA14 association . However , given that RPA14 seems to form a structural core with the DNA binding domain C of RPA70 and DNA binding domain D of RPA32 , it is very surprising that RPA14 is missing . It raises a possibility that there are separate LT-RPA70 and LT-RPA32 dimeric complexes . In any case , it is hardly surprising that complexes lacking RPA14 seems to be non-functional in DNA repair . The interaction of RPA with SV40 LT has also been shown to perturb processivity of DNA polymerase α , so it may have effects in replication as well [45] . We have identified two LT mutants , K308E and E320A , which fail to bind RPA and fail to sensitize cells . Both K308E and E320A can activate E2F-containing promoters ( not shown ) , indicating that they retain LT function towards the Rb family . Other functions of OBD , including DNA binding and transcriptional activation , were not required for RPA binding or sensitization . That RPA is the relevant target was confirmed by the demonstration that overexpression of RPA protected from LT and knockdown of RPA mimicked the LT phenotype . RPA is a protein important for DNA replication and DNA repair [11] , [35] , [46] . It is required for SV40 DNA replication [38] , [47] , [48] and it functions for polyoma as well [21] . It is also a common molecular component of most repair mechanisms ( see [11] for a recent review ) . In particular , RPA is a sensor of UV induced DNA damage that is required for repair of the lesions [14] , [16] . In addition , RPA participates in the formation of repair foci in response to etoposide-induced double-stranded DNA breaks ( DSBs ) [13] . LT binding prevents RPA from localizing to sites of DNA damage . This means that DNA damage sites that would normally be occupied by RPA after the DNA insult lack RPA required to trigger efficient removal of the DNA lesions . Thus Rad51 and Rad9 are not recruited to damage foci when LT is expressed . This would account for the observed failure to repair double-strand breaks by homologous recombination when LT was expressed . All of these observations suggest that titration of RPA by LT , pushing in the direction of its replicative functions and away from its repair functions , is the basis for our effect . In summary , our results demonstrate that interaction of LT with RPA is the pivotal contributing factor to sensitization to DNA damaging agents . It suggests that targeting RPA function might be a useful way to regulate survival . LT-mediated inhibition of RPA can provide a vital strategy in overcoming chemotherapeutic drug resistance and therefore for the treatment of cancer . Inhibition of RPA has in fact been considered as a therapeutic [49] . In the polyomavirus field , Merkel Cell Polyomavirus ( MCV ) is thought to be responsible for a class of human skin cancers [50] . Although the DNA-binding domain is eventually deleted in MCV tumors , it could easily be imagined that a pro-mutagenic phenotype promoted by MCV LT might contribute to the early progression in such cancers .
NIH 3T3 cells were grown in Dulbecco's modified Eagle's medium ( DMEM ) supplemented with 10% calf serum . Tet-off regulated mouse embryonic fibroblasts ( MEFs ) that contain the pBI-G Tet-off vector ( Clontech ) expressing LT antigen , its mutants and the OBD were obtained by selection in 5 µg/ml puromycin after cotransfection with a vector for puromycin resistance and the relevant LT construct . To exclude clonal variation we have analyzed at least six each of full-length and OBD expressing clones . HEK 293T cells were grown in DMEM with 10% fetal calf serum . pCMVLT , HA-tagged Origin binding domain of PyLT ( residues 264 to 420 ) were previously described [34] . All LT mutations were introduced into pBI-G LT or pCMV LT using site-directed mutagenesis and verified by sequencing . MISSION shRNA clones from Sigma-Aldrich that are sequence-verified shRNA lentiviral plasmids were tested for maximum gene silencing effects . The target sequences for Bim were selected and synthesized by Sigma-Aldrich ( NM_009754 ) . Self-inactivating replication incompetent viral particles were produced in packaging cells ( HEK293T ) by co-transfection with compatible packaging plasmids . The targeting sequence used for Bim that achieved maximum knockdown as measured by immunoblot analysis was purchased from Sigma Aldrich: TRCN0000009694 . For siRNA mediated transient knockdown of RPA70 , RNA duplexes used for targeting mouse RPA70 were purchased from Qiagen ( Gaithersburg , MD; GS68275 ) . The small interfering RNAs ( siRNAs ) were introduced in cells using Lipofectamine RNAiMAX reagent ( Invitrogen ) by reverse transfection according to the manufacturer's protocol at a final total concentration of 20 nM . Non-targeting negative control siRNA used was from Qiagen ( 1027310 ) . After 48 hours , medium was collected and whole cell extracts prepared that were subjected to immunoblotting analysis as described below . Cells were washed with cold PBS , harvested and resuspended in lysis buffer ( 20 mM Tris , pH 7 . 5; 150 mM NaCl , 1 mM EGTA , 1 mM EDTA , 1% Triton X-100 , 2 . 5 mM sodium pyrophosphate , 1 mM β-glycerolphosphate , 1 mM Na3VO4 , in the presence of protease inhibitors ( 1 µg/ml leupeptin , pepstatin , and aprotinin ) and phosphatase inhibitors I and II ( 1∶100; Sigma ) for 30 min . Cleared extracts were incubated with specific antibody and protein G Sepharose beads ( Amersham ) for 4 hours , with rocking , at 4°C . Cell extracts were boiled directly in SDS dissociation buffer . After electrophoresis , samples were blotted onto nitrocellulose and analyzed by immunoblotting [34] . Cells on glass coverslips were fixed with cold methanol ( −20°C ) for 20 min at room temperature and washed again three times for 5 min each with TBS ( 50 mM Tris-HCl , pH 7 . 4 , 150 mM NaCl ) . Cells were quenched in fresh 0 . 1% sodium borohydride in TBS for 5 min , washed three times for 5 min each , and blocked with blocking buffer ( 10% goat serum , 1% BSA , 0 . 02% NaN3 , in TBS ) for 1 hour at room temperature . Fixed cells were then incubated with primary antibody in blocking buffer overnight at 4°C ( 1∶100 ) . Cells were again washed three times for 5 minutes each with TBS , followed by incubation with Trit-C or Fit-C labeled secondary antibody at 1∶800 in 1% TBS for one hour at room temperature . Cells were washed again three times for 5 min each and mounted on glass slides with Vectashield ( Vector Laboratories ) that stains the nuclei . Cells were observed by fluorescence microscopy . Images were captured using a Spot advanced imaging system . MEF cells were grown in 100 mm plates , chilled on ice for 15 minutes , collected by scraping and centrifugation , washed once with cold PBS , and lysed in 0 . 4 ml of lysis buffer ( 10 mM Tris , pH 7 . 4 , 25 mM EDTA , PEG 5000 2 . 5% , 1M NaCl and 0 . 25% Triton X-100 ) on ice for 30 minutes . This was followed by centrifugation at 13 , 800×g for 15 minutes , and the supernatant was treated with RNase A ( 200 µg/ml ) at 37°C for 1–2 h , followed by incubation with Proteinase K ( 100 µg/ml ) at 56°C overnight . The mixture was then purified sequentially with phenol-chloroform and chloroform and then precipitated with 0 . 1 volume of 5M NaCl and 2 volumes of ethanol at −20°C overnight . After resuspension , equal amounts of the DNA ( determined by spectrometry at 260/280 nm ) were loaded on a 2% agarose gel ( 50 volts for 2 hours ) , stained with ethidium bromide ( 1 µg/ml ) , and observed by UV illuminator . DNA damage in mouse embryonic fibroblast lines was determined under alkaline conditions using the Comet Assay kit from Trevigen ( Gaithersburg , MD ) . Briefly , the cells were trypsinized , washed in ice-cold PBS , combined with molten agarose , and pipetted onto a comet slide . After solidification at 4°C for 20 min , the slides were immersed in lysis solution . For single cell electrophoresis ( detects single and double strand DNA breaks , DNA cross-links , and base damage ) , the slides were placed in alkaline buffer and electrophoresed at 20 volts for 20 minutes at 4°C . Slides were then washed 2 times consecutively for 10 minutes each with H2O followed by 70% ethanol for 5 minutes . Air-dried slides were then stained with SYBR green I and analyzed using a fluorescence microscope . Cells with damaged DNA display streaming of DNA fragments from nucleus in the form of a comet tail , whereas undamaged DNA appears in the form of a nucleus ) . Comet images were analyzed using CASP software ( Comet Assay Software Project 1 . 2 . 2 ) . At least 100 comets were analyzed for each sample . Comet assays were performed three times , each time in duplicate . NG59RA viral suspension was sonicated and then incubated at 37°C for 20 minutes . Secondary mouse embryo fibroblasts in 100-mm dishes were infected with 2 ml of the viral suspension after washing . Following adsorption for 2 hours at 37°C , cultures were further incubated in fresh medium containing 15% fetal calf serum . Control cultures were mock-infected under identical conditions , but without virus . Stable MEF cells expressing LT or OBD under inducible conditions for 48 h after splitting were allowed to grow to 95% confluence until the day of harvest prior to UV treatment . Cell monolayers were washed twice with 2 ml PBS and irradiated with 40 J/m2 UV using a UV Stratalinker 2400 ( Stratagene ) . At various times post-UV , cells were washed with PBS plus 50 mM EDTA , trypsinized , resuspended in 1 ml of PBS plus 50 mM EDTA , and fixed by the addition of 3 ml of ice-cold 100% ethanol added dropwise . 1×106 fixed cells were then washed with PBS plus 50 mM EDTA , resuspended in either 0 . 5% Triton X-100 plus 0 . 1N HCl ( for 6-4 photoproduct ( 6-4PP ) detection ) or 0 . 5% Triton-X 100 plus 2N HCl ( for CPD detection ) , and incubated for 20 minutes at 22°C . Cells were washed with 0 . 1M Na2B4O7 ( pH 9 . 0 ) and then with PBS and resuspended in 300 µl of RNase ( 100 µg/ml in PBS ) for 1 h at 37°C followed by washing with PBS-TB ( 1% bovine serum albumin plus 0 . 25% Tween 20 in PBS ) . Cells were resuspended in PBS-TB containing a primary monoclonal antibody against either CPD or 6-4PP ( Kamiya Biomedical Company ) for 1 hour at room temperature . Pellets were washed twice with PBS-TB and resuspended in 300 µl of fluorescein isothiocyanate-conjugated rabbit anti-mouse secondary antibody for 45 minutes at room temperature . Pellets were washed twice with PBS-TB . Samples were then subjected to flow cytometry and analyzed by WinList 3D . The efficiency of homology-directed recombination repair was evaluated using the DR-GFP recombination reporter construct that contains two mutated , non-functional copies of a GFP gene with an 18 base pair I-SceI recognition site . Double-strand breaks induced in the chromosomally integrated GFP gene with the expression of the I-SceI endonuclease was repaired by homologous recombination restoring the expression of the intact functional GFP gene . DR-U2OS cells were transfected with control , LT or K308E with or without cotransfection of I-SceI expression vector for 48 h . Cells were processed for flow cytometry . GFP expression in gated live populations was analyzed using Summit 4 . 3 Software .
|
DNA repair protects genome integrity and unrepaired DNA damage can cause cancer . We have identified a new mechanism by which a tumor virus makes cells hypersensitive to DNA damage . The Large T Antigen ( LT ) of polyoma virus blocks DNA repair pathways , making cells 100 fold more sensitive to DNA damage . LT does this by targeting replication protein A ( RPA ) . RPA is central to both DNA replication and repair . Ordinarily RPA and then other DNA repair proteins are recruited to sites of DNA damage . LT blocks recruitment of these proteins to damage foci . Current cancer treatment strategies like radiation therapy and chemotherapeutics cause DNA damage to block the growth and spread of cancer . This work suggests a target that might increase the efficacy of such treatment .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2013
|
Viral Interference with DNA Repair by Targeting of the Single-Stranded DNA Binding Protein RPA
|
Somatic mutations affecting ETV6 often occur in acute lymphoblastic leukemia ( ALL ) , the most common childhood malignancy . The genetic factors that predispose to ALL remain poorly understood . Here we identify a novel germline ETV6 p . L349P mutation in a kindred affected by thrombocytopenia and ALL . A second ETV6 p . N385fs mutation was identified in an unrelated kindred characterized by thrombocytopenia , ALL and secondary myelodysplasia/acute myeloid leukemia . Leukemic cells from the proband in the second kindred showed deletion of wild type ETV6 with retention of the ETV6 p . N385fs . Enforced expression of the ETV6 mutants revealed normal transcript and protein levels , but impaired nuclear localization . Accordingly , these mutants exhibited significantly reduced ability to regulate the transcription of ETV6 target genes . Our findings highlight a novel role for ETV6 in leukemia predisposition .
Acute leukemias comprise the most common form of pediatric cancer , among which acute lymphoblastic leukemia ( ALL ) makes up 80–85% of the cases[1 , 2] . It is well recognized that a proportion of affected children develop the disease due to an underlying predisposition . The currently recognized genes responsible for autosomal dominant transmission of childhood leukemia include TP53 , CEBPA , PAX5 and GATA-2[3–8] . Occasionally , acute leukemia presents in the context of thrombocytopenia . Consistent with this feature , several heritable thrombocytopenia syndromes are known to exist , some of which are associated with an increased incidence of leukemia . Genes associated with these syndromes include RUNX1 , ANKRD26 , GATA1 , MPL , HOXA11 and RMB8A[9–16] . Despite the identification of these genes , there remain many cases for which the underlying mechanism remains unexplained . In this study , we analyzed one large kindred and one parent-child trio , both affected by ALL and thrombocytopenia . By exome sequencing and also sequencing plausible candidate genes such as those involved in B-lymphocyte development and differentiation , we identified germline mutations in the transcription factor ETV6 that co-segregated with disease in each kindred . Functional studies support a pathogenic role for the observed mutations , both of which affect the DNA binding domain . These findings are consistent with independent observations describing additional kindreds characterized by thrombocytopenia and predisposition to hematopoietic malignancy[17 , 18] and provide insights into the mechanisms of leukemia susceptibility and clinical phenotypes associated with germline ETV6 mutations[19] .
As part of a collaborative study focusing on pedigree analysis and gene discovery in childhood leukemia , we identified a Polish/Moroccan kindred in which 10 individuals developed thrombocytopenia and 4 individuals developed thrombocytopenia and ALL ( Kindred 1 in Fig 1A ) . In the 3 ALL cases in Kindred 1 for whom flow-cytometric data were available , all were of the pre-B-ALL subtype . In 3 cases with thrombocytopenia and no evidence of ALL , the mean corpuscular volume ( MCV ) was decreased in 1 case and normal in 2 others ( Table 1 ) . In 2 individuals with no evidence of hematologic abnormalities , there was a history of renal cell cancer and duodenal adenocarcinoma . A second unrelated Western European/Native American family was identified in which a child developed ALL followed by myelodysplastic syndrome and acute myeloid leukemia ( AML ) . This child’s mother , maternal aunt and maternal grandfather exhibited thrombocytopenia ( Kindred 2 in Fig 1B ) . This patient was evaluated by a geneticist due to subtle dysmorphic features; however , clinical assessment did not suggest a known genetic syndrome , and microarray and karyotype did not reveal any large deletions , rearrangements or other structural chromosomal abnormalities . DNA from 16 individuals in Kindred 1 ( 9 individuals with thrombocytopenia and/or ALL and 7 unaffected individuals ) was subjected to Sanger sequencing for all exons of a targeted panel of leukemia-associated genes ( Methods ) . Co-segregation of identified variants was tested using an autosomal dominant mode of inheritance . Published demographic data and medical literature were manually reviewed for all variants observed . Only one variant chr12:12 , 037 , 415 T>C satisfied the criteria of segregation as well as rarity , as evidenced by its absence in public genomic databases such as dbSNP[20] , 1000 genomes[21] , Exome Sequencing Project[22] and Exome Aggregation Consortium ( http://exac . broadinstitute . org ) . This variant , identified in 9 out of 9 ( 100% ) affected family members tested , represents a heterozygous missense c . T1046C mutation in ETV6 ( NM_001987 ) . One individual ( generation 3 , individual 13 ) with thrombocytopenia and leukemia was not tested . This nucleotide change is predicted to result in the substitution of proline for leucine at codon 349 ( L349P; Fig 1A and Table 1 ) . Seven out of 7 ( 100% ) unaffected family members tested exhibited a wild type ( WT ) ETV6 sequence . Fibroblast and lymphocyte DNA from the proband with ALL and parents in Kindred 2 were analyzed by clinical whole exome sequencing ( Ambry Genetics , Aliso Viejo , CA , USA ) . The proband and his mother harbored a heterozygous deletion of 5 nucleotides ( c . 1153-5_1153_1delAACAG ) within ETV6 . This deletion is predicted to lead to a frameshift at codon 385 and truncation of the ETV6 protein at codon 389 ( N385fs , Fig 1B and Table 1 ) . Genome-wide DNA copy alteration analysis using single nucleotide polymorphism microarrays of the diagnostic ALL sample from the proband in Kindred 2 revealed deletion of the wild type and retention of the mutant ETV6 allele , as well as deletions of IKZF1 , PAX5 , BTG1 , and RB1 . Other than the 2 mutations in ETV6 , there were no pathologic genetic mutations associated with ALL or thrombocytopenia that co-segregated with disease in either kindred . Both ETV6 variants were absent in the National Heart Lung Blood Institute ( NHLBI ) Exome Sequencing Project ( ESP ) ( http://evs . gs . washington . edu/EVS/ ) , Exome Aggregation Consortium ( ExAC ) ( http://exac . broadinstitute . org/ ) , or St . Jude Children’s Research Hospital–Washington University Pediatric Cancer Genome Project ( PCGP ) databases[23] . SIFT[24] and Polyphen prediction tools suggest the mutations to be deleterious and probably damaging to protein function . To understand how these two mutations might influence protein function , we modeled their effect on the ETV6 protein structure . Both the L349P and the N385fs mutation are located in the ETS domain of ETV6 ( Fig 2A ) . The L349P mutation is predicted to cause significant conformational changes in areas adjacent to the ETS domain by introducing a kink in the H2 α-helix , resulting in possible ETV6 protein misfolding . The N385fs mutation affects the ETS domain and is predicted to truncate ETV6 at a region involved in DNA interaction ( Fig 2B ) . To evaluate the functional consequences of these mutations , we first assessed whether L349P and N385fs might impair transcriptional repression by ETV6 . HeLa cells were transiently co-transfected with constructs encoding the WT or mutant ETV6 , as well as constructs containing the PF4 or MMP3 promoters , which harbor multiple ETS binding sites and are natural ETV6 targets . We compared the results to those obtained using other recently described germline ETV6 variants , P214L , R369Q , R399C [17] . As expected , WT ETV6 repressed expression of both reporters ( Fig 3A ) , while each of the ETV6 mutants exhibited significantly decreased repression . To further explore the effects of the ETV6 mutations , we analyzed the expression of EGR1 and TRAF1 , genes that are normally upregulated by WT ETV6 [17] . Consistent with published reports , EGR1 and TRAF1 were upregulated 3-fold in cells transfected with WT ETV6 . In contrast , the mutants induced minimal to no upregulation for both of these target genes ( Fig 3C ) . Indeed , the levels were significantly reduced compared to WT ETV6 . In each of these assays , we observed comparable levels of WT and ETV6 mutant mRNA transcripts ( Supporting Information 1 ) . Thus , transcript stability appears to be unaffected by the ETV6 mutations . To examine whether the L349P and N385fs mutations negatively impact translation or alter subcellular localization of the ETV6 protein , we performed cell fractionation assays and western blotting of HeLa cells transiently transfected to express WT or mutant ETV6 . Both proteins were detectable by Western blotting , with a smaller product observed for the N385fs mutation . Both mutants were undetectable in the nucleus ( Fig 4A ) , but detected within the cytoplasmic fraction ( Fig 4B ) , This is in contrast to the described mutants P214L , R369Q and R399C , which were detected in cytoplasmic as well as nuclear fractions . These patterns were quantitated and confirmed by measuring the nuclear to cytoplasmic ratio ( Fig 4C ) . Fusions involving ETV6 in leukemia have long been recognized [25–27] . Other mutation types , including single nucleotide variations , insertions , deletions , frame-shifts and non-sense alterations are also becoming increasingly evident in hematologic malignancies[17 , 18 , 28] . We performed additional sequence analysis on exons 5–8 of ETV6 in unrelated probands from 27 unrelated kindreds with a family history of ALL , but identified no mutations in this region of ETV6 . To further characterize the spectrum of germline and somatic ETV6 mutations that contribute to childhood leukemia , we screened a cohort of 588 leukemia patients evaluated through the PCGP , a genomic sequencing effort involving pediatric cancers [4 , 28–42] ( accession# EGAS00001000348 , EGAS00001000654 , EGAS00001000380 , EGAS00001000253 , EGAS00001000246 , EGAS00001000447 ) . Seventeen distinct somatic ETV6 variants and two rare germline variants were identified ( V37M , R181H; Fig 2A ) . Both rare variants occurred in patients with B-ALL , but with no evidence for loss or mutation of the WT ETV6 allele within the leukemia samples . In one of these cases , there was a secondary vulvar squamous cell carcinoma . There was nofamily history of leukemia or thrombocytopenia in either of these cases . Luciferase assays performed on these variants showed no significant changes in transcriptional repression activity when compared to WT ETV6 ( Fig 3B ) . We queried several public variant databases for the presence of these two variants . The 1000 genomes project has a total of 2 , 819 samples from the world’s major populations . The current version of the Exome sequencing project ( EVS/ESP ) has a set of 2 , 203 African-American and 4300 European-American unrelated individuals , totaling 6 , 503 samples . The Exome aggregation consortium ( ExAC ) has 60 , 706 unrelated individuals sequenced as part of various disease-specific and population genetic studies . In total , we have queried over 140 , 000 chromosomes . The V37M variant ( chr12:11905459G>A ) was seen only in the ExAC data at an allele frequency of 1 . 649x10-05 and the R181H variant ( chr12:12022436 G>A ) was found at an allele frequency of 1 . 071x10-04 in ExAC . It was also found 2 times in NHLBI-ESP ( AF = 1 . 162x10-04 ) and assigned as rs150089916 . While V37M was predicted in silico as benign by SIFT , R181H was classified as deleterious . Based on these preliminary findings the clinical significance of these two additional rare germline variants remains to be determined and at this time is classified as variants of unknown significance .
ETV6 encodes an ETS family transcription factor that is frequently rearranged or fused with other genes in human leukemias of myeloid or lymphoid origin[28] . Also known as the TEL oncogene , ETV6 is a sequence specific transcriptional repressor , regulated by auto-inhibition and self-association[43 , 44] . Descriptions of ETV6 largely focus on the ETV6/RUNX1 fusion , which is a product of a t ( 12;21 ) chromosomal translocation , the most common genetic abnormality in pediatric ALL[25] . While somatic deletions or mutations in ETV6 are increasingly recognized in ALL , nothing is known regarding the impact of germline ETV6 mutations[17 , 28] . Here we extend the description of the clinical phenotype and functional effects associated with novel germline ETV6 L349P and ETV6 N385fs mutations , both of which reside in the highly conserved ETS DNA binding domain and co-segregate with disease in 2 unrelated kindreds affected by thrombocytopenia and ALL . In both kindreds ETV6 mutations were inherited in an autosomal dominant manner with variable expression of thrombocytopenia and/or ALL . There was no evidence for parent of origin or sex-delimited expression , as males and females equally transmitted the putative predisposing alleles with associated phenotypes manifesting in daughters as well as sons . Interestingly , in addition to his leukemia , the proband in Kindred 2 exhibited craniofacial and musculoskeletal anomalies ( anterior placement of the right ear , downward shaped mouth , joint hypermobility and CNS heterotopias seen on magnetic resonance imaging ) . No other obvious pathogenic variants were identified in this individual by whole exome sequencing . In addition to atypical physical features , the proband in Kindred 2 developed grade 3 myelosuppression following exposure to anti-metabolite therapy; this feature of chemotherapy hypersensitivity was shared by another patient with T-/myeloid mixed phenotype leukemia and a germline ETV6 mutation ( P214L ) [17] . In addition , two of the three individuals affected with ALL and harboring ETV6 mutations in the kindreds reported here required bone marrow transplantation , and 1 of the 3 expired from disease , in contrast to the 90% rate of cure with chemotherapy alone in more typical ALL . Whether germline ETV6 mutations might serve as markers for toxicity and outcome will require larger studies controlling for other prognostic variables . In vitro studies revealed impaired function of the ETV6 mutants identified in both kindreds . While ETV6 L349P and N385fs exhibited normal mRNA levels , both mutations were associated with decreased transcriptional regulation ( repression and activation ) . Structural modeling suggests that both ETV6 mutations would impair transcriptional activity by altering the conformation of the ETV6 protein or truncating it within the DNA binding domain . Interestingly , neither mutant localized to the nucleus . Although the precise mechanism for this behavior remains unclear , it seems likely that these two mutations may affect intracellular transport . Consistent with its putative role as a tumor suppressor , examination of the diagnostic leukemia sample in the proband from Kindred 2 revealed retention of the mutant and deletion of the WT ETV6 allele . Our findings are in agreement with 2 recent reports describing additional ETV6 mutations , including R399C , R369Q[17] and R418G[18] in the ETS DNA binding domain and P214L[17 , 18] , located in a serine-proline phosphorylation motif present in the internal linker domain . In the 3 reports of germline ETV6 mutations to date ( including the current series ) , a mixed phenotype of thrombocytopenia and ALL is observed . An association with elevated MCV was not observed in 3 cases included here , which is in contrast to one of the other recent reports[18] . The 2 additional germline variants reported here in patients with ALL ( V37M and R181H ) did not impair transcriptional repression of ETV6 . While this was expected given that these mutations are not located in or close by the ETS DNA binding domain , we cannot exclude that these variants impair ETV6 function on another functional level . The discovery of mutations in ANKRD26 , RUNX1 , and the ETS family transcription factors has led to an increased understanding of the genetic basis of hereditary syndromes involving thrombocytopenia , red cell macrocytosis and leukemia [9 , 10 , 17 , 18] and of the pathways regulated by these genes [17 , 45] . Constitutional alterations in RUNX1 predispose individuals to thrombocytopenia and hematological malignancies , mainly myelodysplastic syndrome and AML , but also T-ALL [3 , 9 , 10 , 46 , 47] . Mutations in RUNX1 have been shown to result in either haploinsufficiency or can act in a dominant-negative manner , the latter resulting in an increased risk of hematological malignancies [48 , 49] . Inherited mutations in ANKRD26 [10 , 45] , which is transcriptionally regulated by RUNX1 lead to a similar clinical phenotype , in which thrombocytopenia is often associated with AML and in some cases , with chronic myelogenous leukemia , chronic lymphocytic leukemia and myelodysplastic syndrome [38] . However , there remain additional kindreds affected by thrombocytopenia and/or leukemia that do not demonstrate germline mutations of RUNX1 or ANKRD26 . Our data suggest that at least a proportion of these cases result from ETV6 mutations . To date , it is not known whether the ETV6 pathway contributes to non-leukemic cancer phenotypes . We observed no pathogenic germline ETV6 mutations in children with cancers other than ALL in the PCGP . Therefore , the contribution of ETV6 mutations to solid tumor predisposition remains to be determined . Improved understanding of the heritable nature of childhood cancers has important clinical implications pertaining to genetic counseling and testing of other family members , therapeutic decisions , donor selection for hematopoietic transplantation , and long-term monitoring for therapy-associated or second primary neoplasms [17 , 50 , 51] . Evaluation for germline alterations of ETV6 is therefore warranted in families with acute lymphoblastic leukemia , particularly when there is preceding evidence of thrombocytopenia .
All individuals analyzed for purposes of our research were formally consented to Memorial Sloan Kettering Cancer Center’s IRB approved research Protocol , Protocol #00–069 , “Ascertainment of Families for Genetic Studies of Familial Lymphoproliferative Disorders” , or St . Jude Children’s Research Hospital IRB approved research Protocols NR14-132 , “Case report of child with novel ETV6 mutation associated with development of leukemia” and/or NR14-162 , “ETV6 germline variants in children with acute lymphoblastic leukemia” , respectively . For Kindred 1 , we included 9 affected and 7 unaffected individuals for sequencing . For Kindred 2 , we included the proband with ALL , his mother with thrombocytopenia and his unaffected father . For both kindreds , the presence and subtype of leukemia were confirmed by review of pathology reports , while thrombocytopenia was confirmed by medical history .
|
Inherited mutations of transcription factors have recently been associated with susceptibility to acute leukemia . Here we report two unrelated kindreds with inherited mutations in ETV6 , the gene encoding the transcription factor ETS variant 6 . These families were characterized by a low platelet count ( thrombocytopenia ) and acute lymphoblastic leukemia ( ALL ) . Sequencing a panel of genes identified germline ETV6 mutations associated with leukemia and thrombocytopenia in multiple individuals tested . In one family , there was a substitution within the DNA binding domain of ETV6 , termed L349P , and in the second there were five base pairs missing in ETV6 ( N385fs ) , causing an abnormally truncated protein . We overexpressed the ETV6 mutants in the HeLa cell line and measured protein levels and localization within the cells . Instead of localizing to the nucleus , as expected for a transcription factor , the mutant proteins were found in the cytoplasm . The mutant proteins also showed decreased ability to regulate the expression of other genes typically suppressed by ETV6 . These findings suggest that germline ETV6 mutations cause a new type of heritable leukemia . This discovery makes possible the pre-symptomatic diagnosis of leukemia susceptibility in families with germline ETV6 mutations , and also provides new information on the causes of leukemia .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Discussion",
"Materials",
"&",
"Methods"
] |
[] |
2015
|
Germline ETV6 Mutations Confer Susceptibility to Acute Lymphoblastic Leukemia and Thrombocytopenia
|
Plasmode is a term coined several years ago to describe data sets that are derived from real data but for which some truth is known . Omic techniques , most especially microarray and genomewide association studies , have catalyzed a new zeitgeist of data sharing that is making data and data sets publicly available on an unprecedented scale . Coupling such data resources with a science of plasmode use would allow statistical methodologists to vet proposed techniques empirically ( as opposed to only theoretically ) and with data that are by definition realistic and representative . We illustrate the technique of empirical statistics by consideration of a common task when analyzing high dimensional data: the simultaneous testing of hundreds or thousands of hypotheses to determine which , if any , show statistical significance warranting follow-on research . The now-common practice of multiple testing in high dimensional experiment ( HDE ) settings has generated new methods for detecting statistically significant results . Although such methods have heretofore been subject to comparative performance analysis using simulated data , simulating data that realistically reflect data from an actual HDE remains a challenge . We describe a simulation procedure using actual data from an HDE where some truth regarding parameters of interest is known . We use the procedure to compare estimates for the proportion of true null hypotheses , the false discovery rate ( FDR ) , and a local version of FDR obtained from 15 different statistical methods .
“Omic” technologies ( genomic , proteomic , etc . ) have led to high dimensional experiments ( HDEs ) that simultaneously test thousands of hypotheses . Often these omic experiments are exploratory , and promising discoveries demand follow-up laboratory research . Data from such experiments require new ways of thinking about statistical inference and present new challenges . For example , in microarray experiments an investigator may test thousands of genes aiming to produce a list of promising candidates for differential genetic expression across two or more treatment conditions . The larger the list , the more likely some genes will prove to be false discoveries , i . e . genes not actually affected by the treatment . Statistical methods often estimate both the proportion of tested genes that are differentially expressed due to a treatment condition and the proportion of false discoveries in a list of genes selected for follow-up research . Because keeping the proportion of false discoveries small ensures that costly follow-on research will yield more fruitful results , investigators should use some statistical method to estimate or control this proportion . However , there is no consensus on which of the many available methods to use [1] . How should an investigator choose ? Although the performance of some statistical methods for analyzing HDE data has been evaluated analytically , many methods are commonly evaluated using computer simulations . An analytical evaluation ( i . e . , one using mathematical derivations to assess the accuracy of estimates ) may require either difficult-to-verify assumptions about a statistical model that generated the data or a resort to asymptotic properties of a method . Moreover , for some methods an analytical evaluation may be mathematically intractable . Although evaluations using computer simulations may overcome the challenge of intractability , most simulation methods still rely on the assumptions inherent in the statistical models that generated the data . Whether these models accurately reflect reality is an open question , as is how to determine appropriate parameters for the model , what realistic “effect sizes” to incorporate in selected tests , as well as if and how to incorporate correlation structure among the many thousands of observations per unit [2] . Plasmode data sets may help overcome the methodological challenges inherent in generating realistic simulated data sets . Catell and Jaspers [3] made early use of the term when they defined a plasmode as “a set of numerical values fitting a mathematico-theoretical model . That it fits the model may be known either because simulated data is produced mathematically to fit the functions , or because we have a real—usually mechanical—situation which we know with certainty must produce data of that kind . ” Mehta et al . ( p . 946 ) [2] more concisely refer to a plasmode as “a real data set whose true structure is known . ” The plasmodes can accommodate unknown correlation structures among genes , unknown distributions of effects among differentially expressed genes , an unknown null distribution of gene expression data , and other aspects that are difficult to model using theoretical distributions . Not surprisingly , the use of plasmode data sets is gaining traction as a technique of simulating reality-based data from HDEs [4] . A plasmode data set can be constructed by spiking specific mRNAs into a real microarray data set [5] . Evaluating whether a particular method correctly detects the spiked mRNAs provides information about the method's ability to detect gene expression . A plasmode data set can also be constructed by using a current data set as a template for simulating new data sets for which some truth is known . Although in early microarray experiments , sample sizes were too small ( often only 2 or 3 arrays per treatment condition ) to use as a basis for a population model for simulating data sets , larger HDE data sets have recently become publicly available , making their use feasible for simulation experiments . In this paper , we propose a technique to simulate plasmode data sets from previously produced data . The source-data experiment was conducted at the Center for Nutrient–Gene Interaction ( CNGI , www . uab . edu/cngi ) , at the University of Alabama at Birmingham . We use a data set from this experiment as a template for producing a plasmode null data set , and we use the distribution of effect sizes from the experiment to select expression levels for differentially expressed genes . The technique is intuitively appealing , relatively straightforward to implement , and can be adapted to HDEs in contexts other than microarray experiments . We illustrate the value of plasmodes by comparing 15 different statistical methods for estimating quantities of interest in a microarray experiment , namely the proportion of true nulls ( hereafter denoted π0 ) , the false discovery rate ( FDR ) [6] and a local version of FDR ( LFDR ) [7] . This type of analysis enables us , for the first time , to compare key omics research tools according to their performance in data that , by definition , are realistic exemplars of the types of data biologists will encounter . The illustrations given here provide some insight into the relative performance characteristics of the 15 methods in some circumstances , but definitive claims regarding uniform superiority of one method over another would require more extensive evaluations over multiple types of data sets .
Steps for plasmode creation that are described herein are relatively straightforward . First , an HDE data set is obtained that reflects the type of experiment for which statistical methods will be used to estimate quantities of interest . Data from a rat microarray experiment at CNGI were used here . Other organisms might produce data with different structural characteristics and methods may perform differently on such data . The CNGI data were obtained from an experiment that used rats to test the pathways and mechanisms of action of certain phytoestrogens [8] , [9] . In brief , rats were divided into two large groups , the first sacrificed at day 21 ( typically the day of weaning for rats ) , the second sacrificed at day 50 ( the day , corresponding to late human puberty , when rats are most susceptible to chemically induced breast cancer ) . Each of these groups was subdivided into smaller groups according to diet . At 21 and 50 days , respectively , the relevant tissues from these rat groups were appropriately processed , and gene expression levels were extracted using GCOS ( GeneChip Operating Software ) . We exported the microarray image ( * . CEL ) files from GCOS and analyzed them with the Affymetrix Package of Bioconductor/R to extract the MAS 5 . 0 processed expression intensities . The arrays and data were investigated for outliers using Pearson's correlation , spatial artifacts [10] and a deleted residuals approach [11] . It is important to note that only one normalization method was considered , but the methods could be compared on RMA normalized data as well . In fact , comparisons of methods' performances on data from different normalization techniques could be done using the plasmode technique . Second , an HDE data set that compares effect of a treatment ( s ) is analyzed and the vector of effect sizes is saved . The effect size used here was a simple standardized mean difference ( i . e . , a two sample t-statistics ) but any meaningful metric could be used . Plasmodes , in fact , could be used to compare the performance of statistical methods when different statistical tests were used to produce the P-values . We chose two sets of HDE data as templates to represent two distributions of effect sizes and two different null distributions . We refer to the 21-day experiment using the control group ( 8 arrays ) and the treatment group ( EGCG supplementation , 10 arrays ) as data set 1 , and the 50-day experiment using the control group ( 10 arrays ) and the treatment group ( Resveratrol supplementation , 10 arrays ) as data set 2 . There were 31 , 042 genes on each array , and two sample pooled variance t-tests for differential expression were used to create a distribution of P-values . Histograms of the distributions for both data sets are shown in Figure 1 . The distribution of P-values for data set 1 shows a stronger signal ( i . e . , a larger collection of very small P-values ) than that for data set 2 , suggesting either that more genes are differentially expressed or that those that are expressed have a larger magnitude treatment effect . This second step provided a distribution of effects sizes from each data set . Next , create the plasmode null data set . For each of the HDE data sets , we created a random division of the control group of microarrays into two sets of equal size . One consideration in doing so is that if some arrays in the control group are ‘different’ from others due to some artifact in the experiment , then the null data set can be sensitive to how the arrays are divided into two sets . Such artifacts can be present in data from actual HDEs , so this issue is not a limitation of plasmode use but rather an attribute of it , that is , plasmodes are designed to reflect actual structure ( including artifacts ) in a real data set . We obtained the plasmode null data set from data set 1 by dividing the day 21 control group of 8 arrays into two sets of 4 , and for data set 2 by dividing the control group of 10 arrays into two sets of 5 arrays . Figure 2 shows the two null distributions of P-values obtained using the two sample t-test on the plasmode null data sets . Both null distributions are , as expected , approximately uniform , but sampling variability allows for some deviation from uniformity . A proportion 1−π0 of effect sizes were then sampled from their respective distributions using a weighted probability sampling technique described in the Methods section . What sampling probabilities are chosen can be a tuning parameter in the plasmode creation procedure . The selected effects were incorporated into the associated null distribution for a randomly selected proportion 1−π0 of genes in a manner also described in the Methods section . What proportion of genes is selected may depend upon how many genes in an HDE are expected to be differentially expressed . This may determine whether a proportion equal to 0 . 01 or 0 . 5 is chosen to construct a plasmode . Proportions between 0 . 05 and 0 . 2 were used here as they are in the range of estimated proportions of differentially expressed genes that we have seen from the many data sets we have analyzed . Finally , the plasmode data set was analyzed using a selected statistical method . We used two sample t-tests to obtain a plasmode distribution of P-values for each plasmode data set because the methods compared herein all analyze a distribution of P-values from an HDE . P-values were declared statistically significant if smaller than a threshold τ . Box 1 summarizes symbol definitions . When comparing the 15 statistical methods , we used three values of π0 ( 0 . 8 , 0 . 9 , and 0 . 95 ) and two thresholds ( τ = 0 . 01 and 0 . 001 ) . For each choice of π0 and threshold τ , we ran B = 100 simulations . All 15 methods provided estimates of π0 , 14 provided estimates of FDR , and 7 provided estimates of LFDR . Because the true values of π0 and FDR are known for each plasmode data set , we can compare the accuracy of estimates from the different methods . There are two basic strategies for estimating FDR , both predicated on an estimated value for π0 , the first using equation ( 1 ) below , the second using a mixture model approach . Let PK = M/K be the proportion of tests that were declared significant at a given threshold , where M and K were defined with respect to quantities in Table 1 . Then one estimate for FDR at this threshold is , ( 1 ) The mixture model ( usually a two-component mixture ) approach uses a model of the form , ( 2 ) where f is a density , p represents a P-value , f0 a density of a P-value under the null hypothesis , f1 a density of a P-value under the alternative hypothesis , π0 is interpreted as before , and θ a ( possibly vector ) parameter of the distribution . Since valid P-values are assumed , f0 is a uniform density . LFDR is defined with respect to this mixture model as , ( 3 ) FDR is defined similarly except that the densities in ( 3 ) are replaced by the corresponding cumulative distribution functions ( CDF ) , that is , ( 4 ) where F1 ( τ ) is the CDF under the alternative hypothesis , evaluated at a chosen threshold τ . ( There are different definitions of FDR and the definition in ( 4 ) is , under some conditions , the definition of a positive false discovery rate [12] . However , in cases with a large number of genes many of the variants of FDR are very close [13] ) . The methods are listed for quick reference in Table 2 . Methods 1–8 use different estimates for π0 and , as implemented herein , proceed to estimate FDR using equation ( 1 ) . Method 9 uses a unique algorithm to estimate LFDR and does not supply an estimate of FDR . Methods 10–15 are based on a mixture model framework and estimate FDR and LFDR using equations ( 3 ) and ( 4 ) where the model components are estimated using different techniques . All methods were implemented using tuning parameter settings from the respective paper or ones supplied as default values with the code in cases where the code was published online . First , to compare their differences , we used the 15 methods to analyze the original two data sets , with data set 1 having a “stronger signal” ( i . e . , lower estimates of π0 and FDR ) . Estimates of π0 from methods 3 through 15 ranged from 0 . 742 to 0 . 837 for data set 1 and 0 . 852 to 0 . 933 for data set 2 . ( Methods 1 and 2 are designed to control for rather than estimate FDR and are designed to be conservative; hence , their estimates were much closer to 1 . ) Results of these analyses can be seen in the Supplementary Tables S1 and S2 . Next , using the two template data sets we constructed plasmode data sets in order to compare the performance of the 15 methods for estimating π0 ( all methods ) , FDR ( all methods except method 9 ) , and LFDR ( methods 9–15 ) . Figures 3 and 4 show some results based on data set 2 . More results are available in the Figures S1 , S2 , S3 , S4 , S5 , and S6 . Figure 3 shows the distribution of 100 estimates for π0 using data set 2 when the true value of π0 is equal to 0 . 8 and 0 . 9 . Methods 1 and 2 are designed to be conservative ( i . e . , true values are overestimated ) . With a few exceptions , the other methods tend to be conservative when π0 = 0 . 8 and liberal ( the true value is underestimated ) when π0 = 0 . 9 . The variability of estimates for π0 is similar across methods , but some plots show a slightly larger variability for methods 12 and 15 when π0 = 0 . 9 . Figure 4 shows the distribution of estimates for FDR and LFDR at the two thresholds . The horizontal lines in the plots show the mean ( solid line ) and the minimum and maximum ( dashed lines ) of the true FDR value for the 100 simulations . A true value for LFDR is not known in the simulation procedure . The methods tend to be conservative ( overestimate FDR ) when the threshold τ = 0 . 01 and are more accurate at the lower threshold . Estimates of FDR are more variable for methods 11 , 13 , and 14 and estimates for LFDR more variable for methods 13 and 14 , with the exception of a few unusual estimates obtained from method 9 . The high variability of FDR estimates from method 11 may be due to a “less than optimal” choice of the spanning parameter in a numerical smoother ( see also Pounds and Cheng [27] ) . We did not attempt to tune any of the methods for enhanced performance .
Researchers have been evaluating the performance of the burgeoning number of statistical methods for the analysis of high dimensional omic data , relying on a mixture of mathematical derivations , computer simulations , and sadly , often single dataset illustrations or mere ipse dixit assertions . Recognizing that the latter two approaches are simply unacceptable approaches to method validation [2] and that the first two suffer from limitations described earlier , an increasing number of investigators are turning to plasmode datasets for method evaluation [28] . An excellent example is the Affycomp website ( http://affycomp . biostat . jhsph . edu/ ) that allows investigators to compare different microarray normalization methods on datasets of known structure . Other investigators have also recently used plasmode-like approaches which they refer to as ‘data perturbation’ [29] , [30] , yet it is not clear that these ‘perturbed datasets’ can distinguish true from false positives , suggesting greater need for articulation of principles or standards of plasmode generation . As more high dimensional experiments with larger sample sizes become available , researchers can use a new kind of simulation experiment to evaluate the performance of statistical analysis methods , relying on actual data from previous experiments as a template for generating new data sets , referred to herein as plasmodes . In theory , the plasmode method outlined here will enable investigators to choose on an empirical basis the most appropriate statistical method for their HDEs . Our results also suggest that large , searchable databases of plasmode data sets would help investigators find existing data sets relevant to their planned experiments . ( We have already implemented a similar idea for planning sample size requirements in HDEs [31] , [32] . ) Investigators could then use those data sets to compare and evaluate several analytical methods to determine which best identifies genes affected by the treatment condition . Or , investigators could use the plasmode approach on their own data sets to glean some understanding of how well a statistical method works on their type of data . Our results compare the performance of 15 statistical methods as they process the specific plasmode data sets constructed from the CNGI data . Although identifying one uniformly superior method ( if there is one ) is difficult within the limitations of this one comparison , our results suggest that certain methods could be sensitive to tuning parameters or different types of data sets . A comparison over multiple types of source data sets with different distributions of effects sizes could add the detail necessary to clearly recommend certain methods over others [1] . Other papers have used simulation studies to compare the performance of methods for estimating π0 and FDR ( e . g . , Hsueh et al . [33]; Nguyen [34]; Nettleton et al . [35] ) . We compared methods that use the distribution of P-values as was done in Broberg [36] and Yang and Yang [37] . Unlike our plasmode approach , most earlier comparison studies used normal distributions to simulate gene expression data and incorporated dependence using a block diagonal correlation structure as in Allison et al [26] . A key implication and recommendation of our paper is that , as data from the growing number of HDEs is made publicly available , researchers may identify a previous HDE similar to one they are planning or have recently conducted and use data from these experiments to construct plasmode data sets with which to evaluate candidate statistical methods . This will enable investigators to choose the most appropriate method ( s ) for analyzing their own data and thus to increase the reliability of their research results . In this manner , statistical science ( as a discipline that studies the methods of statistics ) becomes as much an empirical science as a theoretical one .
The quantities in Table 1 are those for a typical microarray experiment . Let N = A+B and M = C+D and note that both N and M will be known and K = N+M . However , the number of false discoveries is equal to an unknown number C . The proportion of false discoveries for this experiment is C/M . Benjamini and Hochberg [6] defined FDR as , P ( M>0 ) where I{M>0} is an indicator function equal to 1 if M>0 and zero otherwise . Storey [12] defined the positive FDR as . Since P ( M>0 ) ≥1− ( 1−τ ) K , and since K is usually very large , FDR≈pFDR , so we do not distinguish between FDR and pFDR as the parameter being estimated and simply refer to it as FDR with estimates denoted ( and ) . Suppose we identify a template data set corresponding to a two treatment comparison for differential gene expression for K genes . Obtain a vector , δ , of effect sizes . One suggestion is the usual t-statistic , where the ith component of δ , is given by ( 5 ) where ntrt , nctrl are number of biological replicates in the treatment and control group , respectively , X̅i , trt , X̅i , ctrl are the mean gene expression levels for gene i in treatment and control groups , and , is the usual pooled sample variance for the ith gene , where the two sample variances are given by , . In what follows , we will use this choice for δi since it allows for effects to be described by a unitless quantity , i . e . , it is scaled by the standard error of the observed mean difference X̅i , trt−X̅i , ctrl for each gene . For convenience , assume that nctrl is an even number and divide the control group into two sets of equal size . Requiring nctrl≥4 allows for at least two arrays in each set , thus allowing estimates of variance within each of the two sets . This will be the basis for the plasmode “null” data set . There are ways of making this division . Without loss of generality , assume that the first nctrl/2 arrays after the division are the plasmode control group and the second nctrl/2 are the plasmode treatment group . Specify a value of π0 and specify a threshold , τ , such that a P-value ≤τ is declared evidence of differential expression . Execute the following steps . One can then obtain another data set and repeat the entire process to evaluate a method on a different type of data , perhaps from a different organism having a different null distribution , or a different treatment type giving a different distribution of effect sizes , δ . Alternatively , one might choose to randomly divide the control group again and repeat the entire process . This would help assess how differences in arrays within a group or possible correlation structure might affect results from a method . If some of the arrays in the control group have systematic differences among them ( e . g . , differences arising from variations in experimental conditions—day , operator , technology , etc . ) , then the null distribution can be sensitive to the random division of the original control group into the two plasmode groups , particularly if nctrl is small .
|
Plasmode is a term used to describe a data set that has been derived from real data but for which some truth is known . Statistical methods that analyze data from high dimensional experiments ( HDEs ) seek to estimate quantities that are of interest to scientists , such as mean differences in gene expression levels and false discovery rates . The ability of statistical methods to accurately estimate these quantities depends on theoretical derivations or computer simulations . In computer simulations , data for which the true value of a quantity is known are often simulated from statistical models , and the ability of a statistical method to estimate this quantity is evaluated on the simulated data . However , in HDEs there are many possible statistical models to use , and which models appropriately produce data that reflect properties of real data is an open question . We propose the use of plasmodes as one answer to this question . If done carefully , plasmodes can produce data that reflect reality while maintaining the benefits of simulated data . We show one method of generating plasmodes and illustrate their use by comparing the performance of 15 statistical methods for estimating the false discovery rate in data from an HDE .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"biotechnology",
"mathematics",
"science",
"policy",
"computational",
"biology",
"molecular",
"biology",
"genetics",
"and",
"genomics"
] |
2008
|
Evaluating Statistical Methods Using Plasmode Data Sets in the Age of Massive Public Databases: An Illustration Using False Discovery Rates
|
The intracellular pathogenic bacterium Brucella generates a replicative vacuole ( rBCV ) derived from the endoplasmic reticulum via subversion of the host cell secretory pathway . rBCV biogenesis requires the expression of the Type IV secretion system ( T4SS ) VirB , which is thought to translocate effector proteins that modulate membrane trafficking along the endocytic and secretory pathways . To date , only a few T4SS substrates have been identified , whose molecular functions remain unknown . Here , we used an in silico screen to identify putative T4SS effector candidate proteins using criteria such as limited homology in other bacterial genera , the presence of features similar to known VirB T4SS effectors , GC content and presence of eukaryotic-like motifs . Using β-lactamase and CyaA adenylate cyclase reporter assays , we identified eleven proteins translocated into host cells by Brucella , five in a VirB T4SS-dependent manner , namely BAB1_0678 ( BspA ) , BAB1_0712 ( BspB ) , BAB1_0847 ( BspC ) , BAB1_1671 ( BspE ) and BAB1_1948 ( BspF ) . A subset of the translocated proteins targeted secretory pathway compartments when ectopically expressed in HeLa cells , and the VirB effectors BspA , BspB and BspF inhibited protein secretion . Brucella infection also impaired host protein secretion in a process requiring BspA , BspB and BspF . Single or combined deletions of bspA , bspB and bspF affected Brucella ability to replicate in macrophages and persist in the liver of infected mice . Taken together , these findings demonstrate that Brucella modulates secretory trafficking via multiple T4SS effector proteins that likely act coordinately to promote Brucella pathogenesis .
Intracellular parasites have evolved specialized mechanisms that exploit a variety of host cellular pathways to generate idiosyncratic niches of replication or persistence . Among these , various bacterial pathogens including Brucella spp . , Legionella pneumophila , Chlamydia spp . and Salmonella enterica serovar Typhimurium target several compartments of the secretory pathway to promote their replication [1] . The secretory pathway orchestrates the synthesis , modification and transport of proteins and lipids [2] . It is organized into successive membrane-bound compartments including the endoplasmic reticulum ( ER ) , ER-to-Golgi intermediate compartment ( ERGIC ) , Golgi apparatus , trans-Golgi network ( TGN ) , and the plasma membrane [3] . Secretory cargo is selected and transported from ER exit sites ( ERES ) to the Golgi apparatus via the sequential action of COPII and COPI coat complexes , the activities of which are regulated by Rab-family and ARF-family small GTPases [4] , [5] , which are targets of bacterial modulation [1] . Brucella spp . are Gram-negative intracellular pathogens of various mammals that cause the worldwide zoonotic disease known as brucellosis or Malta fever [6] . Key to the pathogenesis of these bacteria is their ability to infect both phagocytic and non-phagocytic cells ranging from macrophages and dendritic cells to epithelial cells [7] , [8] , [9] , [10] . Upon entry into host cells , Brucella reside within a membrane-bound compartment known as the Brucella-containing vacuole ( BCV ) , whose trafficking along the endocytic and secretory pathways is controlled by the bacterium [11] , [12] , [13] . During maturation along the endocytic pathway , the BCV is acidified and acquires late endosomal markers such as Rab7 and LAMP1 [8] , [9] , [13] , before being redirected towards the early secretory pathway through intimate interactions with ERES , eventually fusing with the ER in a process that depends upon the small GTPase Sar1 and thus on the formation of COPII-dependent transport vesicles [14] . During this process , the BCV recruits the small GTPase Rab2 and GAPDH [15] that regulate membrane traffic between the ER and ERGIC , and are required for Brucella replication . Biogenesis of the rBCV depends upon the VirB Type IV secretion apparatus [8] , [13] , [16] , [17] , [18] , a crucial virulence factor of Brucella that delivers effector molecules into the host cell [19] , [20] that are thought to modulate BCV trafficking . Importantly , the VirB Type IV secretion apparatus is essential for Brucella pathogenesis , since virB mutants are incapable of survival and replication in host cells and attenuated in a mouse model of infection [8] , [14] , [16] , [17] , [18] . Recently , a number of Brucella VirB-dependent effector proteins have been identified [19] , [20] . The first VirB substrates , VceA and VceC , were uncovered by screening for genes co-regulated with the virB operon [19] , of which VceC induces inflammation through the induction of ER stress [21] . Since then , various strategies have been implemented to identify Brucella effectors , including the use of in silico screening for proteins with distinct features [20] , a strategy that has proven to be successful in identifying T4SS effectors of other intracellular pathogens such as L . pneumophila and Coxiella burnetti [22] , [23] . Using this approach , four proteins , BPE123 , BPE043 , BPE005 and BPE275 , were identified as VirB substrates [20] , yet their molecular functions and roles in Brucella pathogenesis remain unknown . A high throughput yeast two-hybrid screening approach for potential host interactors also recently identified RicA , a protein translocated in a VirB-dependent manner that interacts with Rab2 [24] . Despite our understanding of the VirB T4SS roles in the Brucella intracellular cycle and the identification of several effector proteins , VirB-associated molecular functions and the cellular pathways that Brucella effectors modulate to control the bacterium's intracellular trafficking still remain unknown . In particular , given Brucella's reliance on cellular processes associated with the secretory pathway , determining how the VirB T4SS-mediated functions interfere with this compartment is key to a molecular understanding of Brucella pathogenesis . In an effort to identify novel Brucella VirB T4SS effector proteins , here we have used a genome-wide in silico analysis and protein translocation reporter assays . We report the identification of 11 Brucella proteins translocated into host cells , among which 10 are novel and at least 5 are translocated in a VirB T4SS-dependent manner . Importantly , several of these proteins target compartments of the secretory pathway and contribute to Brucella interference with host cellular protein secretion , intracellular proliferation and persistence in vivo . This study demonstrates Brucella modulation of host secretion via novel effector proteins and provides the first evidence of a role of VirB T4SS effector proteins on intracellular membrane trafficking important to Brucella pathogenesis .
To identify Brucella putative VirB T4SS effector proteins , we chose to focus on Brucella predicted proteins annotated as hypothetical proteins of unknown functions , based on the assumption that such effectors fulfill functions specific to Brucella intracellular pathogenesis and should consequently be either restricted to the Brucella genus or conserved mostly among closely related α2- proteobacteria . Additionally , we reasoned that they may possess a positive net charge in their C-terminal end and possibly Arginine-rich motifs , which are features of known VirB effector proteins in the closely related organisms Agrobacterium tumefaciens and Bartonella henselae [25] , [26] . Out of the 3 , 085 predicted proteins from the Brucella abortus strain 9-941 genome sequence annotated in the ERGO Genome Analysis Suite ( http://integratedgenomics . com/ergo . html ) , 823 proteins without assigned functions were examined using the Blastp algorithm ( http://blast . ncbi . nlm . nih . gov/Blast . cgi ) for limited similarities with proteins in other bacteria , and the net charge of their C-terminal 20 amino-acids residues calculated using an EMBOSS bioinformatics tool called “charge” ( http://emboss . bioinformatics . nl/ ) . Using this analysis , 93 proteins that were either Brucella-specific or highly represented within the α2- proteobacteria and contained a positively-charged C-terminus were selected for further screening ( Table S1 ) . An additional criterion for selection was the deviation in GC content from that of the entire genome ( Table S1 ) , based on the possibility that VirB T4SS effectors may have been acquired by horizontal transfer . Further analyses used an array of bioinformatics tools including SMART ( http://smart . embl-heidelberg . de/ ) , TMpred ( http://www . ch . embnet . org/software/TMPRED_form . html ) , Coils ( http://www . ch . embnet . org/software/COILS_form . html ) , and SignalP v4 . 1 ( http://www . cbs . dtu . dk/services/SignalP/ ) for the presence of functional or structural domains and presence or absence of Sec-dependent secretion signals , following the assumption that VirB T4SS effector proteins may possess domains or motifs either present in eukaryotic proteins or consistent with interactions with , and modulation of host factors . This combinatorial analysis led to the selection of 20 predicted proteins as candidate VirB T4SS substrates that fulfilled most or parts of the selection criteria ( Table S2 ) . Additionally , a predicted protein encoded by the locus BAB1_1671 was independently selected based on the presence of a coiled-coil domain and a C-terminal transmembrane domain in an arrangement reminiscent of eukaryotic membrane fusion soluble NSF attachment protein receptor ( SNARE ) proteins ( Table S2 ) . To test whether the Brucella putative VirB T4SS effector proteins are translocated into host cells during infection , we first used the TEM1 β-lactamase protein translocation reporter assay [27] previously employed to identify Brucella VirB T4SS substrates [19] , [24] . Based on a shift in fluorescence emission by the membrane-permeant β-lactamase substrate CCF2/AM upon β-lactamase translocation into host cells , this assay allows for the detection and quantification of protein translocation in cells infected with Brucella strains expressing TEM1 fusions with putative translocated effectors [27] . Given the lack of knowledge of VirB-mediated translocation signals in Brucella , translational fusions of the TEM1 protein with either the N-terminus or the C-terminus of each putative effector protein were generated and expressed under the control of the isopropyl-ß-D-thiogalactopyranoside ( IPTG ) -inducible Ptrc promoter in either pJC120 or pJC121 , respectively ( Table S2 ) , in wild-type B . abortus 2308 . With the exception of BAB1_0227-TEM1 , BAB1_1386-TEM1 and BAB1_1495-TEM1 , the expression of all fusions was detected by Western blot analysis using anti-β-lactamase and anti-Brucella outer membrane lipoprotein Omp19 antibody as loading control , albeit to varying levels ( Fig . 1E ) . To test translocation of these TEM1 fusions , J774 . A1 macrophage-like cells were infected with TEM1 fusion-expressing wild type Brucella strains at an MOI of 1000 for 16 h and processed for fluorescence microscopy analysis . TEM1-GST and GST-TEM1 fusion proteins were used as negative controls of translocation and a TEM1-VceC fusion [19] and TEM1 fusions to the BPE123 effector protein [20] were used as positive controls . Expression of the TEM1-VceC fusion triggered CCF2/AM conversion in infected J774 . A1 cells , indicating protein translocation , while almost all cells infected with Brucella strains expressing either TEM1-GST or GST-TEM1 fusions did not show any translocation ( Fig . 1A–D ) . Additionally , translocation of BPE123-TEM1 was observed ( Fig . 1A–B and D ) , despite undetectable expression ( Fig . 1E ) , but not that of TEM1-BPE123 , consistent with the previous demonstration of this effector's translocation [20] . Under these validated experimental conditions , 11 of the 21 Brucella putative effector proteins were repeatedly translocated into host cells ( Fig . 1 , S1 and S2 ) . Ten of these Brucella proteins , BAB1_0227 , BAB1_0678 , BAB1_0712 , BAB1_0847 , BAB1_1611 , BAB1_1671 , BAB1_1864 , BAB1_1948 , BAB2_0119 and BAB2_0541 were newly identified as translocated by Brucella into host cells , while translocation of BAB1_1865 ( also named BPE865 ) was previously reported [20] . BAB1_0678 , BAB1_1864 , BAB1_1865 , BAB1_1948 and BAB2_0119 were translocated into host cells regardless of the N- or C-terminal tagging with TEM1 ( Fig . 1A–D ) . However , only the N-terminal TEM1 fusions of BAB2_0541 and BAB1_0227 showed translocation ( Fig . 1A–D ) . While BAB2_0541 fusion constructs were expressed to comparable levels , the BAB1_0227-TEM1 fusion protein was not detected by Western blot analysis ( Fig . 1E ) , potentially explaining its lack of translocation . Yet , the inability to detect protein expression in Brucella did not hinder the translocation of BAB2_0123-TEM1 fusion ( Fig . 1B , D and E ) , suggesting that very low levels are sufficient to detect translocation . Conversely , only the C-terminal TEM1 fusions to BAB1_0712 , BAB1_0847 , BAB1_1611 and BAB1_1671 were translocated , even-though both N- and C-terminal constructs were expressed in Brucella ( Fig . 1E ) . Taken together , these results demonstrate that the majority ( 11 out of 21 ) of our putative Brucella effector proteins identified in silico are translocated by Brucella during infection and the position of the TEM1 tag at either the N- or C-terminus of the protein can influence its translocation in this assay . To determine whether the newly identified Brucella effector proteins were translocated in a VirB T4SS-dependent manner , we needed to examine their translocation by a VirB-deficient strain . Given that virB mutants do not survive in macrophages and are progressively killed [8] , [14] , a translocation assay in which similar numbers of intracellular viable wild type and VirB-deficient bacteria are present was required to generate conclusive data . Consistent with previous reports [8] , [14] , [16] , infection of J774A . 1 cells with both the wild type 2308 strain and an isogenic ΔvirB9 mutant showed a progressive decrease in recovery of viable ΔvirB9 bacteria , reaching a 2 Log difference with 2308 by 16 h pi ( Fig . S3A ) , a time point required by our β-lactamase reporter system . We therefore chose to use the B . pertussis calmodulin-dependent adenylate cyclase ( CyaA ) reporter system [28] , which has previously identified substrates of the VirB T4SS of Brucella [20] , at an infection time point ( 6 h pi ) where MOI adjustments yielded similar numbers of recoverable CFUs for both wild type and VirB-deficient bacteria ( Fig . S3B ) . CyaA translational fusions with the identified proteins translocated by Brucella ( Fig . 1A–B ) were generated in plasmids pJC125 and pJC126 , by positioning the CyaA tags as the TEM1 moieties ( Table S2 ) , introduced into either wild type 2308 or the ΔvirB9 mutant strain and their expression was verified by Western blot analysis using a anti-CyaA antibody ( data not shown ) . Cyclic AMP ( cAMP ) levels were measured in J774 . A1 cells infected for 6 h with either wild type or ΔvirB9 strains expressing the CyaA fusions and were normalized to intracellular CFUs to account for variations in infection levels and viability between strains . Compared to pJC125 and pJC126 empty vector controls , expression of the VirB-dependent effector protein BPE123 [20] fused to CyaA generated significant cAMP production in cells infected with the wild type , but not with the ΔvirB9 mutant ( Fig . 2A ) , validating our experimental conditions . Similarly , 10 of the 11 Brucella protein fusions showed a significant increase in cAMP levels compared to negative controls ( Fig . 2A ) , confirming our β-lactamase assay results ( Fig . 1A–B ) . Notably , the translocation of BAB1_0678 , BAB1_0712 , BAB1_0847 , BAB1_1671 and BAB1_1948 was VirB T4SS-dependent , as the cAMP levels were significantly lower ( P<0 . 05 ) after infection with the ΔvirB9 mutant strain ( Fig . 2A ) . Cells infected with ΔvirB9 mutant strains expressing either BAB1_0227 or BAB2_0541 fusion proteins did not show any significant decrease of cAMP levels compared to the wild type strain ( Fig . 3 ) , indicating that translocation of these fusion proteins is not VirB-dependent . Intriguingly , VirB-dependency of translocation of the BAB1_1864 , BAB1_1865 and BAB2_0119 fusion proteins was inconclusive since the location of the CyaA tag influenced the results: translocation of these fusion proteins was VirB-dependent when the CyaA tag was C-terminal , but VirB-independent when N-terminal ( Fig . 2A ) . Additionally , VirB-dependent translocation of BAB1_1611 could not be assessed , since the CyaA fusion led to protein instability in Brucella ( data not shown ) . Taken together , our results demonstrate that we have identified 11 Brucella proteins translocated into host cells during infection , at least 5 of which are Brucella VirB T4SS effector proteins . We therefore named these proteins Brucella secreted proteins ( Bsp ) BspA ( BAB1_0678 ) , BspB ( BAB1_0712 ) , BspC ( BAB1_0847 ) , BspD , ( BAB1_1611 ) , BspE ( BAB1_1671 ) , BspF ( BAB1_1948 ) , BspG ( BAB1_0227 ) , BspH ( BAB1_1864 ) , BspI ( BAB1_1865 ) , BspJ ( BAB2_0119 ) and BspK ( BAB2_0541 ) , of which BspA , BspB , BspC , BspE and BspF are VirB T4SS substrates ( Fig . 2B ) . Interestingly , BspA contains the Pfam domain DUF2062; BspB , a SCOP structural domain ( d2gsaa ) present in pyridoxal phosphate ( PLP ) -dependent transferases flanked by 2 predicted transmembrane ( TM ) domains; BspD and BspE contain Coiled-coil ( CC ) and TM domains; BspF , a GNAT-family acetyltransferase domain; BspG , internal repeat sequences ( RPT ) in its C-terminal end; BspH , Armadillo ( ARM ) repeats; BspI , a GTPase Activating Protein ( GAP ) domain . BspC is the only translocated protein containing a predicted Sec-dependent signal peptide ( Fig . 2B and Table S1 ) . To gain insight into the functions of the Bsp proteins , we first tested whether they target a particular intracellular compartment when expressed ectopically in mammalian cells . HeLa cells were transfected with expression plasmids encoding either Hemagglutinin ( HA ) - or Green Fluorescent Protein ( GFP ) -tagged Bsp proteins for 12 h , and processed for fluorescence confocal microscopy analysis for representative expression patterns in low-expressing cells . In most cases , HA- or GFP-tagged Bsp proteins displayed similar expression patterns . While HA-BspH , and HA-BspI displayed a cytosolic localization , HA-BspG partitioned between the cytosol and the nucleus ( Fig . 3 ) , with cytosol and nuclear localizations in some cells ( data not shown ) . Similarly , HA-BspJ also partitioned between the cytosol and punctate structures in the nucleus reminiscent of nuclear speckles ( Fig . 3 ) . HA-BspE formed discrete vesicles predominantly in the perinuclear area , while HA-BspF localized to both the cytosol and plasma membrane protrusions ( Fig . 3 ) . Interestingly , GFP-BspA , HA-BspB , BspD-GFP and HA-BspK displayed distinct reticular expression pattern consistent with a localization to the ER ( Fig . 3 ) . Colocalization of GFP-BspA , HA-BspB , BspD-GFP and HA-BspK with the ER-resident chaperone protein , Calnexin ( Fig . 4 ) confirmed that these proteins accumulate in the ER . In addition , GFP-BspC localized to a discrete perinuclear compartment consistent with the Golgi apparatus ( Fig . 3 ) , which was confirmed by colocalization of GFP-BspC-positive vesicles with the Golgi marker GM130 ( Fig . 4 ) . Hence , ectopically expressed Bsp proteins localize to distinct intracellular compartments in HeLa cells , with 5 out of 11 ( BspA , BspB , BspC , BspD and BspK ) targeting the secretory pathway . Given the propensity of the identified Bsp proteins to target the secretory pathway and considering that Brucella manipulates functions of this particular compartment to generate the rBCV , we sought to determine whether expression of Bsp proteins affects functions of the secretory pathway . To achieve this , we first examined secretion of a reporter protein , the secreted embryonic alkaline phosphatase ( SEAP ) [29] , in HEK293T cells co-expressing SEAP and HA- or GFP-tagged Bsp proteins . The GDP-locked allele of the small GTPase ARF1[T31N] was used as a control for inhibition of secretion , since its expression disrupts the early secretory pathway [30] . With the exception of HA-BspC , all HA- or GFP-tagged Bsp proteins were expressed in HEK293T cells when analyzed by Western blot using anti-HA or anti-GFP antibodies , although to varying levels ( Fig . S4A ) . After 24 h of co-transfection , HEK293T cells were processed for SEAP secretion analysis and the secretion index calculated as the extracellular SEAP activity versus intracellular SEAP activity normalized to transfections with the SEAP plasmid only . While cotransfections of empty vectors with the SEAP plasmid did not alter the levels of secreted SEAP , expression of HA- or GFP-tagged ARF1[T31N] led to a ∼80% reduction in SEAP secretion ( Fig . 5A and B ) . Interestingly , expression of HA-BspA , HA-BspB , or HA-BspF led to a significant decrease in SEAP secretion , with BspF inhibiting secretion by ∼50% , while BspA and B showed a stronger inhibition in secretion ( >80% ) to levels comparable to the effect of ARF1[T31N] ( Fig . 5A ) . The levels of secretion inhibition did not correlate with Bsp expression levels , since BspA and BspB showed lower expression than BspF ( Fig . S4A ) . In comparison , expression of either BspC , BspD , BspE , BspG or BspJ did not affect SEAP secretion levels ( Fig . 5A–B ) , regardless of their expression levels ( Fig . S4A ) . Since expression of BspC , BspD or BspK did not alter SEAP secretion despite their targeting of secretory compartments ( Fig . 3 and 4 ) and BspF does not accumulate in a secretory compartment yet impairs SEAP secretion ( Fig . 3–5 ) , BspA- , BspB- or BspF-mediated inhibition of SEAP secretion is not simply due to their accumulation along the secretory compartment . Rather , inhibition of protein secretion may result from specific activities of these effectors . The decrease of processing in the secretory pathway is a hallmark of ER stress [31] , which can quickly down regulate SEAP secretion [32] . To examine whether the decrease in SEAP secretion observed upon expression of either BspA , BspB or BspF was due to induction of ER-stress , we monitored this response in cells expressing these proteins using the ERSE reporter ( luc ) dual-luciferase assay and expression of downstream protein targets of ER stress , CHOP and BiP/GRP78 [33] . Tunicamycin , an ER stress stimulus that inhibits NH2-linked glycosylation and protein folding in the ER , and expression of HA-VceC , a Brucella VirB T4SS effector that induces ER stress and proinflammatory cytokine expression upon ER stress induction [21] were used as positive controls in these assays ( Fig . S4B ) . Compared to Tunicamycin treatment or VceC expression , which induced ER stress to varying degrees , expression of either HA-BspA , HA-BspB or HA-BspF did not induce ER stress ( Fig . S4B–C ) , nor did expression of HA-BspD , HA-BspE , HA-BspH or HA-BspI , although HA-BspF increased BiP levels ( Fig . S4C ) . In comparison , GFP-BspC , HA-BspG or HA-BspK expression significantly induced ER stress to levels similar or higher than VceC and upregulation of BiP , but not CHOP ( Fig . S4B–C ) . These results verify that inhibition of SEAP secretion by BspA and BspB is not due to induction of ER stress , while this remains unclear for BspF due to its ability to increase levels of BiP expression , and also demonstrates that BspC , BspG and BspK induce significant ER stress when ectopically expressed in mammalian cells . Since inhibition of secretory transport can also result from the physical disruption of secretory compartments [34] , [35] , [36] , we also examined the effect of HA-BspA , HA-BspB or HA-BspF expression on the morphology of the secretory pathway , using immunofluorescence confocal microscopy . In HeLa cells expressing either GFP-BspA , HA-BspB or HA-BspF proteins , no significant difference was observed in the morphology of the ER , ERES , ERGIC or Golgi apparatus ( Fig . S5 ) , indicating that the inhibitory effect these proteins have on SEAP secretion does not result from overt physical disruption of the secretory compartment . To independently confirm that BspA , B and F inhibit host protein secretion , we then examined the anterograde transport of VSV-Gts045 , a temperature sensitive mutant reporter protein from the vesicular stomatitis virus that is retained in the ER at a non-permissive temperature of 40°C due to a temperature sensitive misfolding , but can be folded normally and transported to the Golgi and the plasma membrane ( PM ) at 32°C [37] , [38] . HeLa cells were co-transfected with plasmids expressing VSVGts045-GFP and either HA-BspA , HA-BspB or HA- BspF for 20 h at 40°C and then shifted to 32°C to monitor VSV-Gts045-GFP trafficking by fluorescence microscopy . In control cells expressing VSVGts045-GFP alone , the majority of the protein was transported from the ER to the Golgi apparatus within 20 min after temperature shift , and was located at the PM by 60 and 120 min ( Fig . 5C ) . VSVGts045-GFP was transported with apparent similar kinetics in cells expressing BspF ( Fig . 5C ) , but not in cells expressing BspA , where it was strongly retained in the ERGIC and Golgi apparatus even after 120 min ( Fig . 5C ) . Expression of BspB also altered VSVGts045-GFP transport specifically from the Golgi to the PM , since the majority of the mutant protein remained in Golgi compartments for up to 60 min and did not reach the PM by 120 min ( Fig . 5C ) . Hence , BspA and BspB expression in HeLa cells impairs secretory trafficking , consistent with their inhibitory effects on protein secretion ( Fig . 5A ) . Given the effect of ectopically-expressed BspA , BspB and BspF on host protein secretion , we hypothesized that Brucella interferes with secretory transport to the plasma membrane during infection . To test this hypothesis , we monitored host protein secretion during Brucella infection utilizing both SEAP and VSV-G transport assays . First , HeLa cells were transfected with the SEAP plasmid for 16 h , then mock-infected or infected with wild type Brucella strain 2308 . At 24 h pi , SEAP secretion was synchronized by treating cells for 30 min with brefeldin A ( BFA ) to reversibly disrupt the early secretory pathway and block transport from the ER [39] , [40] and cycloheximide to inhibit new SEAP biosynthesis and allow for monitoring secretion of a given pool of SEAP . BFA was then washed out , which allowed Golgi apparatus full reassembly within 2 h [41] with similar kinetics in mock- and Brucella-infected cells ( Fig . S7 ) , and SEAP secretion was monitored at various time points post washout . Compared to mock-infected cells , Brucella infection caused a strong delay in SEAP secretion ( Fig . 6A ) , which was only restored to control levels by 10 h post washout . It is also worth noting that a maxiumum of 30% of cells were both transfected and infected with Brucella in these experiments ( data not shown ) , likely causing an underestimation of the inhibitory effect that Brucella exerts on cellular secretion . These results demonstrate that Brucella trafficking to and replication in the ER affects secretion along the secretory pathway . Brucella-induced delayed transport of a given pool of secretory cargo suggests that this bacterium affects a subcompartment of the secretory pathway that can eventually support trafficking . To investigate further Brucella inhibitory effects on secretion , we examined the trafficking kinetics of VSV-Gts045-GFP in Brucella-infected cells . HeLa cells were mock-infected or infected with Brucella for 24 h and then transfected with VSV-Gts045-GFP for 20 h at 40°C . Confocal micrographs were collected at various time points after a temperature shift to 32°C to quantify the localization of VSV-Gts045-GFP to different subcompartments ( ER , ERGIC , Golgi , Golgi to PM and PM ) of the secretory pathway . Upon temperature shift , VSV-Gts045-GFP rapidly trafficked out of the ER in mock-infected cells ( Fig . 6B ) to accumulate in the Golgi apparatus by 20 min ( Fig . 6D ) , undergo post-Golgi trafficking by 40 min ( Fig . 6E ) and reach the PM by 60 min ( Fig . 6F ) . While VSV-Gts045-GFP trafficking out of the ER was only slightly delayed in infected cells ( Fig . 6B ) , it accumulated and was retained in the ERGIC from 20 to 40 min ( Fig . 6C ) , after which it underwent post-Golgi trafficking ( Fig . 6E ) but inefficiently reached the PM by 120 min ( Fig . 6F ) . Hence , consistent with the SEAP secretion data ( Fig . 6A ) , these results demonstrate that Brucella disrupts host protein secretion by impairing cargo trafficking between the ER and the Golgi apparatus and decreasing transport to the PM . Since Brucella disrupts host cell secretion during infection ( Fig . 6 ) and the VirB T4SS effectors BspA , BspB and BspF also affect this process when ectopically expressed ( Fig . 5 ) , we next investigated a direct role of these effectors on the host secretory pathway during infection . We therefore generated single or multiple in-frame deletions of bsp genes in B . abortus strain 2308 ( Table S3 and Fig . S6 ) . HeLa cells transfected with the SEAP plasmid were infected with either the wild type 2308 strain or various single bsp mutants . At 24 h pi , SEAP secretion was synchronized as described above using BFA and cycloheximide , and measured 6 h post BFA washout , a time point when Brucella infection caused significant inhibition of SEAP secretion ( Fig . 6A ) . As a positive control for secretion inhibition , BFA treatment strongly inhibited SEAP secretion when maintained throughout the experiment ( Fig . 7A–B ) . While infection with the wild type 2308 strain significantly reduced SEAP secretion by ∼50% , the ΔbspB and ΔbspF failed to inhibit SEAP secretion ( Fig . 7A–B ) . Complementation of either bspB or bspF mutants with a single chromosomal copy of the respective gene restored the inhibitory phenotype ( Fig . 7B ) , demonstrating that BspB and BspF specifically contribute to Brucella-mediated inhibition of secretion during infection . By contrast , the ΔbspA mutant strain inhibited SEAP secretion similar to the wild type strain , despite strongly decreasing protein secretion when overexpressed in HEK293T or HeLa cells ( Fig . 5 ) , suggesting it does not contribute to inhibition of secretion in the context of an infection . As controls , deletion of either bspC , bspD , or bspK , whose products all target the secretory pathway , did not restore host protein secretion , consistent with their lack of effect on this pathway when ectopically expressed ( Fig . 5 ) , and deletion of either bspE or bspG , which encode two effector proteins unrelated to the secretory pathway ( Fig . 3 and 5 ) , did not relieve the secretory inhibition phenotype ( Fig . 7A ) . Taken together , these results demonstrate that BspB and BspF specifically interfere with cellular secretion . Interestingly , the combinatorial deletion of bspA , bspB and bspF rescued the SEAP secretion to levels comparable to uninfected cells expressing SEAP protein alone ( control; Fig . 7A ) and more completely than upon deletion of single genes , suggesting that at least BspB and BspF coordinately act to impair protein secretion . Since overexpression of BspA in HeLa cells inhibits host secretion ( Fig . 5 ) , yet deletion of bspA does not relieve Brucella-mediated inhibition of secretion ( Fig . 7A ) , we hypothesized that BspA effect on the host secretory pathway may be counteracted by ( an ) other unidentified Brucella effector ( s ) that would be absent in an ectopic expression context , and that BspA overexpression in Brucella may enhance and reveal its inhibitory effect on secretion . To test this hypothesis , BspA , BspB , BspF , BspD and BspE fused to a 3×FLAG tag according to their translocation requirements ( Fig . 1 and 2 ) were expressed under IPTG-inducible Ptrc promoters on multicopy plasmids pJC123 or pJC124 in the wild-type 2308 strain ( Fig . S8 ) and SEAP secretion was measured in infected HeLa cells . Compared to the secretory inhibition caused by the wild-type strain or strains carrying empty vector controls and under experimental conditions where both the ΔbspB and ΔbspF mutants failed to inhibit protein secretion , overexpression of BspA-3×FLAG and BspB-3×FLAG significantly enhanced inhibition of SEAP secretion , while that of 3×FLAG-BspF did not ( Fig . 7D ) . This increased inhibition of host secretion was specific to these effectors , since it was not observed upon overexpression of either BspD-3×FLAG , which localizes to the ER when expressed ectopically in HeLa cells without affecting host secretion , or BspE-3×FLAG , a VirB-dependent effector that does not target the secretory pathway ( Fig . 7D ) . Taken together and consistent with our ectopic expression data ( Fig . 5 ) , these results indicate that bacterially-expressed BspA also affects the host secretory pathway in infected cells . While the above results demonstrate that Brucella inhibits host secretion at 30 h pi , i . e . when the bacteria replicate within ER-derived rBCVs , we next sought to examine whether they also exert an inhibitory effect on host secretion prior to reaching the ER , i . e . when VirB-associated functions control BCV trafficking [8] , [14] , [16] . When SEAP secretion was measured at 8 h pi , i . e . before bacteria generate rBCVs and replicate , infection with the wild-type strain 2308 inhibited SEAP secretion to levels comparable to those seen at 30 h pi ( Fig . 7C ) . Deletion of either bspB or bspF , but not bspA , relieved secretion inhibition , which was restored upon genetic complementation of the ΔbspB or ΔbspF mutants ( Fig . 7C ) , as observed at 30 h pi ( Fig . 7A–B ) . Additionally , infection with a ΔvirB9 mutant , which was interpretable at this time point due to its similar intracellular viable numbers to 2308 ( Fig . S3 ) , also failed to inhibit SEAP secretion ( Fig . 7C ) , consistent with the roles of BspB and BspF in this process . Altogether , these results demonstrate that Brucella-mediated inhibition of host secretion is a VirB-dependent process that requires BspB and BspF and occurs prior to rBCV biogenesis . Given the effect of BspB , BspF and BspA on the host secretory pathway and the importance of this compartment for Brucella intracellular trafficking and replication , we sought to determine whether these proteins play a role in Brucella pathogenesis . First , intracellular growth of the single and triple deletion mutants was examined in murine BMMs by CFU enumeration over 24 h . No significant difference was observed in Brucella replication between the wild type strain and the ΔbspA or ΔbspB mutants , as similar numbers of CFUs were recovered during the time course of infection , while the ΔbspF mutant showed decreased survival by 8 h pi but underwent subsequent replication similar to the wild type strain ( Fig . 8A–C ) . In contrast , a stronger decrease in recoverable bacteria was observed at 8 h pi and afterwards for the triple ΔbspABF mutant compared to the wild type strain ( Fig . 8D ) , indicating that the combined deletion of several Bsp effectors affects Brucella intracellular growth . Since CFU enumeration examines global intracellular growth within a population , it may lack the ability to uncover minor , yet significant , phenotypic defects at the cellular level . To evaluate the intracellular growth of the bsp mutants by single cell analysis , we examined bacterial replication in BMMs at 24 h pi using immunofluorescence microscopy . Under infection conditions where 1–4 bacteria were initially phagocytozed by BMMs ( data not shown ) , the percentage of infected cells supporting bacterial replication was scored as those containing at least 10 bacteria and showed that the ΔbspB mutant strain displayed a significant replication defect when compared to the wild type strain , which was genetically complemented ( Fig . 8E ) . Consistent with the CFU enumeration ( Fig . 8D ) , the ΔbspABF mutant was also significantly impaired in replication inside BMMs ( Fig . 8E ) , confirming its intracellular growth defect . Hence , both BspB and the coordinated action of BspA , BspB and BspF promote Brucella intracellular replication . To extend these findings and examine whether the intracellular defects of these mutants translate into virulence defects , we monitored establishment and persistence of these strains in a BALB/c mouse model of chronic brucellosis . Following intraperitoneal inoculation with either the wild type 2308 strain , the ΔbspA , ΔbspB , ΔbspF or ΔbspABF deletion mutants , splenomegaly was assessed as a readout for inflammation , and bacterial loads enumerated in the spleen and liver at day 3 , 7 and 42 pi , to assess initial infection ( day 3 ) , establishment of the chronic stage ( day 7 ) and persistence ( day 42 ) . In this infection model , a ΔvirB9 mutant did not induce any splenomegaly and failed to establish a chronic infection and persist in both spleen and liver ( Fig . S9 ) . Although all strains induced splenomegaly similarly and colonized the spleen equally at all times points , the numbers of recoverable ΔbspB and ΔbspABF mutants from the liver at day 42 pi were significantly lower than the wild type strain ( Fig . S10 ) , indicating a role of BspB and the cumulative functions of BspA , BspB and BspF in bacterial persistence in this organ .
Many Gram-negative bacterial pathogens including Brucella spp . , Legionella pneumophila , Coxiella burnetii , Bartonella spp . , Helicobacter pylori , Bordetella pertussis , and Rickettsia prowazekii use T4SS to translocate effector proteins into host cells as part of their pathogenic process [42] . The identification and characterization of these proteins is a crucial step in our molecular understanding of many human diseases and has been the subject of several bioinformatics-based efforts with recent successful outcomes in the case of Legionella , Coxiella , Bartonella and Brucella [19] , [20] , [22] , [23] , [26] , [43] . Here we have used similar approaches and protein translocation reporter systems to identify in Brucella abortus 11 translocated proteins ( BspA to K ) , 10 of which are novel since BspI ( BPE865 ) has previously been reported [20] , and at least 5 are translocated in a VirB T4SS-dependent manner . These findings expand the number of known Brucella effector proteins to 19 , including VceA and VceC [19] , RicA [24] , and 6 BPEs [20] and suggest that this pathogen translocates into host cells an array of proteins with possibly redundant functions . The initial demonstration of protein translocation of the Bsp effectors used a TEM1 β-lactamase reporter system , in which we systematically tested both N- and C-terminal TEM1 fusions . This was a necessary precaution due to a lack of consensus for Brucella VirB-mediated translocation signals , as exemplified by the findings that the N-terminal region of BPE123 is necessary for its translocation into host cells [20] while the C-terminal region of VceC is required for efficient secretion [19] . Interestingly , only the C-terminal TEM1 fusion proteins to BspB , BspC , BspD and BspE and only the N-terminal TEM1 fusions to BspG and BspK showed translocation , while those to BspA , BspF , BspH , BspI and BspJ showed translocation regardless of the TEM1 tag location . This suggests that the position of the reporter tag may interfere with translocation signals and protein translocation . Consequently , one cannot exclude that TEM1 tagging at either end of the negative candidate effector proteins also interfered with their translocation or led to fusion proteins in which TEM1 activity was impaired . These observations highlight the necessity to characterize the molecular mechanisms by which the Brucella VirB T4SS translocates proteins into host cells and the caveats associated with using reporter translocation systems . A retrospective analysis of the criteria used to identify the Bsp proteins did not reveal consensus features shared by these effectors , including some that would typify VirB substrates . While BspA , BspB , BspC , BspD , BspH and BspK all possessed a RXR ( or RXK in the case of BspC ) motif in their C-terminus ( Table S2 ) , reminiscent of the A . tumefaciens Arginine-rich motifs in VirB effectors [25] , this could not be strictly correlated with VirB-dependency or a requirement for a free C-terminus for translocation in our assays , nor was the net charge of their C-termini . Except for BspC , none of the Bsp proteins contained a predicted Sec-dependent signal peptide sequence , suggesting that their translocation does not require Sec-dependent translocation into the periplasm . Most bsp genes nonetheless showed a significant deviation in their GC content from that of the total genome ( Table S2 ) . Despite the intracellular lifestyle of Brucella spp . , which intuitively limits genetic exchange with other bacteria , and their overall genetic homogeneity , this suggests that these bsp genes have been acquired via horizontal transfer during evolution of an ancestral genome , perhaps in extracellular environments , in agreement with evidence for discrete horizontal transfer events gained from genomic comparisons between Brucella spp . [44] . Consistently , the colinear genetic organization of bspH and bspI adjacent to an IS2020 transposase-encoding open reading frame ( BAB1_1863 ) argues for the presence of a mini genomic island carrying these effectors . A CyaA translocation reporter assay with sufficient sensitivity to detect Brucella effector translocation at 6 h pi identified BspA , BspB , BspC , BspE and BspF as VirB T4SS substrates , since their translocation was significantly reduced in the ΔvirB9 mutant background . In contrast , the translocation of BspG and BspK was independent of a functional T4SS , suggesting the presence of another unknown translocation system in Brucella . This is in line with the VirB-independent translocation of BPE865 ( BspI ) and BPE159 [20] . Of note , translocation of the BspI-CyaA ( BPE865 ) was VirB-dependent under our experimental conditions , unlike a previous report by Marchesini et al . [20] . Although the same CyaA-based translocation assays were used in both studies , we found that normalization of cAMP levels to recoverable CFUs was essential to uncover VirB dependency in these assays , a technical variation that could explain the contrasting results obtained . Surprisingly , the translocation of the BspH , BspI and BspJ was VirB-dependent when tagged at the C-terminus , indicating these proteins are VirB T4SS substrates , but VirB-independent when tagged on their N-terminus . These results suggest that these proteins may be promiscuously translocated by more than one translocation system , the VirB T4SS and another yet unknown secretion system , through translocation signals that confer pathway specificity and are differentially affected by the CyaA tag location . Translocation of bacterial effectors via multiple type three secretion systems ( T3SS ) and the flagella export system has been shown for YplA in Yersinia [45] and SptP in Salmonella [5] . Brucella does possess all the flagellar genes needed for the assembly of a functional flagellum , but not those encoding for chemotaxis [46] , [47] . Considering that flagellar and virB genes are co-regulated in Brucella [48] , some of these Brucella secreted proteins could also be targeted to and translocated by the flagella export pathway , a hypothesis that remains to be tested . Additionally , the secretion of RicA in broth culture in a VirB T4SS- and flagella-independent manner supports the possibility of yet another unidentified translocation system in Brucella [20] , [24] . With the exception of VceC , shown to induce ER stress and secretion of proinflammatory cytokines [21] , the recently discovered Brucella effectors remain uncharacterized and virulence phenotypes associated with individual effector proteins have not been observed , presumably due to some functional redundancy [20] . Additionally , the role of Brucella effectors in the biogenesis and trafficking of the BCV along the endocytic and secretory pathway remains poorly understood . RicA was shown to interact with Rab2 , a small GTPase recruited on BCVs and required for Brucella intracellular replication [15] , potentially assigning a direct role for RicA in the intracellular trafficking of Brucella [24] . Yet , how RicA interaction with Rab2 contributes to the intracellular lifestyle of Brucella remains speculative . In our attempt to characterize the functions of the newly identified Brucella effectors , a high proportion of the Bsp proteins localized to compartments of the secretory pathway when ectopically expressed in HeLa cells . BspA , BspB , BspD and BspK localized to the ER and BspC localized to the Golgi apparatus . Considering the importance of the secretory pathway in Brucella intracellular pathogenesis , where BCVs engage early secretory compartments by initiating interactions with ERES [14] and recruit Rab2 and GAPDH [15] , we pursued these initial observations to demonstrate that BspA , BspB and BspF specifically inhibit host protein secretion and membrane trafficking along the secretory pathway . That BspF affects host protein secretion despite a cytosolic and plasma membrane localization is not inconsistent , as i ) the bacterially-translocated form may localize differently via interactions with other bacterial effectors or BCV-associated host factors , and ii ) various host cell proteins that mediate membrane trafficking events localize to the cytosol and are targeted to vesicular compartments via regulatory proteins [49] . A number of T4SS and T3SS bacterial effectors in other pathogens also affect functions of the secretory pathway . When overexpressed in eukaryotic cells , the L . pneumophila Dot/Icm effectors DrrA/SidM [50] , [51] , and LidA [52] disrupt the Golgi apparatus , and RalF [53] , AnkX [23] , the Coxiella Dot/Icm effector CBU0635 [54] , and the Enteropathogenic Escherichia coli ( EPEC ) T3SS effector NleA [55] inhibit host protein secretion , similar to the effect of BspA , BspB or BspF . Unlike the L . pneumophila Dot/Icm effector proteins that cause Golgi apparatus fragmentation [50] , [52] , the overexpression of BspA , BspB or BspF did not alter the morphology of secretory compartments including the ERES , ERGIC , and Golgi apparatus , suggesting subtle effects on this compartment . We therefore speculate that the effect these effectors exert on protein secretion results from discrete interference with specific molecular processes along the secretory pathway and may reflect a specific inhibitory role in protein secretion related to Brucella pathogenesis , and/or a collateral consequence of BCV trafficking-related modulation of membrane transport . Importantly , inhibition of host protein secretion and secretory membrane trafficking was also revealed during Brucella infection and required the functions of BspB and BspF , functionally correlating effector ectopic expression and infection , as previously reported only in the case of EPEC NleA [55] . Contrary to the effect of ectopically-expressed BspA and unlike BspB and BspF , deletion of bspA did not restore secretion in Brucella-infected cells , suggesting some functional redundancy between BspA and other translocated Brucella proteins , otherwise absent in the ectopic expression model used . Yet , BspA overexpression in Brucella enhanced inhibition of host secretion , arguing for a role of this effector in also impairing secretory functions in the context of an infection , as seen with ectopically-expressed BspA . While our results clearly assign modulatory functions along the secretory pathway to the VirB T4SS effectors BspA , BspB and BspF , we also provide evidence that they contribute to biogenesis of rBCVs and bacterial replication in macrophages , and long-term persistence in the liver of chronically infected mice . Given that combining deletions in bspA , bspB and bspF resulted in additive effects on restoration of inhibition of secretion by Brucella and intracellular replication defects , we also speculate that these effectors coordinately act on modulating the secretory pathway to promote Brucella pathogenesis . Collectively , our results potentially link VirB T4SS effector-mediated modulation of the secretory pathway with BCV trafficking , although we cannot yet conclude whether BspA- , BspB- and BspF-mediated inhibition of host protein secretion is a requirement for rBCV biogenesis and bacterial replication , or is unrelated to their requirement for optimal intracellular growth . Additionally , Brucella-mediated inhibition of host protein secretion may affect secretion or surface exposure of immunity-related molecules to the pathogen's benefit . Consistent with our results , Brucella induces retention of MHC Class I molecules in the Golgi apparatus in human macrophages , resulting in downmodulation of CD8+ cytotoxic T-cell responses [56] , possibly via BspA , BspB and/or BspF . Similarly , it will be most interesting to examine whether these effectors downmodulate secretion of pro-inflammatory cytokines , an effect that may explain the persistence defects of the ΔbspB and ΔbspABF mutants observed in the liver . By identifying novel VirB T4SS effectors and characterizing their activities and contribution to pathogenesis , this study expands our knowledge at the molecular level of how Brucella manipulates the host secretory pathway to promote its intracellular survival . Future studies focusing on the mechanisms of BspA , BspB and BspF actions and on the characterization of the additional Bsp proteins identified should reveal new facets of Brucella molecular interactions with host cells , generating much-needed knowledge about its pathogenic mechanisms .
Antibodies were from the following sources: mouse anti-Beta-lactamase ( QED Bioscience , Inc . ) , mouse monoclonal anti-Omp19 SC10 and anti-p17 ( gifts from Dr Axel Cloeckaert , INRA , France ) , mouse anti-CyaA 3D1 ( Santa Cruz Biotechnology , Inc . ) , mouse monoclonal anti-HA , mouse monoclonal anti-FLAG M2 and mouse anti-SEAP ( Sigma-Aldrich Co . LLC . ) , rabbit polyclonal anti-GFP ( Life Technologies ) , rabbit anti-Calnexin ( Enzo Life Sciences ) , mouse anti-GM130 ( BD Transduction Laboratories ) , rabbit anti-COPII ( Thermo Scientific , Pierce Antibodies ) , mouse anti-ERGIC-53 ( Enzo Life Sciences ) , goat anti-Brucella LPS ( gift from Dr Renée Tsolis , UC Davis , CA ) , rat anti-mouse LAMP-1 ( clone 1D4B , developed by J . T . August and obtained from the Developmental Studies Hybridoma Bank ( DSHB ) developed under the auspices of the NICHD and maintained by the University of Iowa , Department of Biological Sciences , Iowa City , IA 52242 ) , rabbit polyclonal anti-Actin ( Bethyl Laboratories , Inc . ) , rabbit anti-BiP and mouse monoclonal anti-CHOP ( Cell Signaling Technology ) ; Alexa Fluor 488-donkey anti-mouse , anti-rat antibodies ( Invitrogen , Life Technologies ) , or Cyanin 5-conjugated goat anti-rabbit and anti-mouse ( Jackson ImmunoResearch Laboratories , Inc . ) . HRP-conjugated anti-rabbit IgG or anti-mouse IgG ( 1∶10 , 000 , Cell Signaling Technology ) and the ECL western blotting substrate ( Thermo Scientific , Pierce Protein Biology Products ) were used for Western blotting . DAPI ( Invitrogen , Life Technologies ) was used for DNA staining . Dulbecco's modified Eagle medium ( DMEM ) , DMEM without phenol red , phosphate saline buffer ( PBS ) , Hank's Balanced Salt Solution ( HBSS ) , fetal bovine serum ( FBS ) and gentamicin were from Life Technologies . FuGENE HD and X-tremeGENE 9 transfection reagents were used to transfect cell lines according to the manufacturer's instructions . SEAP secretion assays were performed with the SEAP reporter gene assay , chemiluminescent kit ( Roche Applied Science ) . The levels of intracellular cAMP were measured with the cAMP Enzyme Immunoassay kit from Sigma-Aldrich Corporation LLC while the beta-lactamase activity was assayed with the Beta-lactamase loading solution and Probenecid from Life Technologies ( Invitrogen ) . ER stress measurements were implemented with the Cignal ERSE Reporter Luciferase assay kit from SABiosciences . Luciferase measurements were performed with the Passive Lysis Buffer and Dual-Luciferase Reporter Assay System from Promega Corporation . Tunicamycin , Cycloheximide and Brefeldin A were purchased from Sigma-Aldrich . Brucella abortus 2308 strains and derivatives were grown on tryptic soy agar ( TSA ) ( Difco ) for 72 h at 37°C or in tryptic soy broth ( TSB ) at 37°C with shaking overnight to late logarithmic phase . B . abortus strains 2308ΔvirB9 [14] and DsRedm-expressing 2308 ( pJC44 ) [13] have been described previously . Brucella strains harboring pJC120 , pJC121 , pJC125 and pJC126 plasmids described in Table S2 were cultured in TSA or TSB supplemented with kanamycin ( 50 µg/ml ) . All manipulations of B . abortus strains were performed in a Biosafety Level 3 facility according to standard operating procedures approved by the Rocky Mountain Laboratories Institutional Biosafety Committee and in compliance with the CDC Division of Select Agents and Toxins regulations . Escherichia coli strains were grown in Luria-Bertani ( LB ) broth at 37°C , supplemented with 50 µg/ml of kanamycin and 100 µg/ml of ampicillin when necessary . Primers and plasmids used in this study are described in Table S3 . The pSEAP-2 control vector , pCMV-HA , pEGFP-C1 and pEGFP-N1 were purchased from Clontech . Plasmids pHA-ARF1[T31N] and pGFP-ARF1[T31N] have been described [30] and pEGFP-VSVGts045 was a gift from Dr Jennifer Lippincott-Schwartz ( NIH , Bethesda , MD ) . The β-lactamase TEM-1 ( pJC120 and pJC121 ) and CyaA ( pJC125 and pJC126 ) fusion plasmids were constructed from pBBR1MCS-2 [57] . First , the lac promoter region of pBBR1MCS-2 was replaced with a lacIq-Ptrc fragment by ligating XhoI-digested pBBR1MCS-2 to a PCR product from plasmid pTrc99A [58] obtained with primers TW331 ( CGGGCCCCCCCTCGAGCCGCCAACACCCGCTGAC ) and TW332 ( TACCGTCGACCTCGAGCATTATTACCACCTCCTCTG ) using the In-Fusion PCR Cloning System ( Clontech ) . The modification was confirmed by sequencing . Vectors pJC120 and pJC121 were then constructed by cloning PCR products containing the blaM gene from pCX340 [27] with primers TW333 ( GCTTGATATCGAATTCCACCCAGAAACGCTGGTGAAAG ) or TW340 ( CGGGCTGCAGGAATTCCCAATGCTTAATCAGTGAGGC ) , and TW333 and TW334 ( CGGGCTGCAGGAATTCTCACCAATGCTTAATCAGTGAGGC ) respectively , into the EcoRV-digested pBBR1-lacIq-Ptrc derivative using In-Fusion , and were confirmed by sequencing . The resulting plasmids pJC120 and pJC121 encode the mature form of TEM-1 under the control of the lacIq gene and the IPTG-inducible Ptrc promoter , allowing for cloning of potential effector protein genes with blaM to generate effector-TEM-1 fusion proteins . The stop codon of the blaM gene was removed in pJC120 to allow generation of in-frame N-terminal fusions , while it was left intact in pJC121 to allow generation of in-frame C-terminal fusion proteins . The plasmids pJC123 and pJC124 were constructed by PCR amplifying the 3×FLAG sequence from a synthesized template with primers TW337 ( GCTTGATATCGAATTCGATTATAAAGATGATGATG ) , TW338 ( CGGGCTGCAGGAATTCCTATTTATCATCATCATCT ) , and TW337 , TW343 ( CGGGCTGCAGGAATTCTTTATCATCATCATCT ) , respectively , into the EcoRI-digested pBBR1-lacIq-Ptrc plasmid using In-Fusion . Inserts were confirmed by sequencing . The plasmids pJC125 and pJC126 were constructed by PCR amplification of the catalytic portion of the cyaA gene ( first 399 codons ) from plasmid pJB2581 [59] with primers TW682 ( CGGTATCGATAAGCTTCAGCAATCGCATCAGGCTGGT ) or TW684 ( ATTCGATATCAAGCTTCTGGCGTTCCACTGCGCCCAGCGA ) , and TW682 and TW683 ( ATTCGATATCAAGCTTTCACTGGCGTTCCACTGCGCCCAGCGA ) , respectively , and cloning into the HindIII-digested pBBR1 lacIq-Ptrc derivative using In-Fusion , and were confirmed by sequencing . The stop codon in the cyaA open reading frame in plasmid pJC125 was removed to allow for generation of N-terminal in-frame fusions of potential effector-CyaA fusions , while it was left intact in pJC126 to allow for generation of in-frame C-terminal fusions . The resulting plasmids: pJC120 , pJC121 , pJC123 , pJC124 , pJC125 and pJC126 were used to construct either TEM-1 , 3×FLAG or CyaA Brucella effector fusions ( Table S3 ) . Full-length genes encoding putative Brucella effectors to be fused to either the blaM or cyaA fragments were amplified by PCR using genomic DNA from B . abortus 2308 strain as a template with sequence specific primers containing restriction sites for cloning into either pJC120 ( XhoI ) , pJC121 ( SalI ) , pJC125 ( EcoRI ) or pJC126 ( SalI ) ) vectors ( Table S3 ) using the In-Fusion PCR Cloning System ( Clontech , 639650 ) . To tag putative Brucella effectors with either GFP or HA , open reading frames of the Brucella target genes were amplified by PCR using genomic DNA from B . abortus 2308 strain as template , Platinum Pfx DNA Polymerase ( Invitrogen ) , and specific primers ( Table S3 ) . The PCR-amplified DNA fragments were cloned into either BglII-digested pEGFP-C1 ( Clontech ) , PstI-digested pEGFP-N1 ( Clontech ) , or EcoRI-digested pCMV-HA ( Clontech ) using the In-Fusion PCR Cloning System ( Clontech ) . The identity and orientation of all constructs generated by PCR were confirmed by restriction digest analysis , sequencing and protein expression where indicated . In-frame deletion mutants of B abortus strain 2308 were generated using the SacB-assisted allelic replacement suicide vector pJC80 , as described previously [14] , and were designed to preserve the integrity of the downstream genes and avoid any polar effect of the deletion . Recombinant in-frame deletions of either BAB1_0678 ( bspA ) , BAB1_0712 ( bspB ) , BAB1_0847 ( bspC ) , BAB1_1611 ( bspD ) , BAB1_1671 ( bspE ) , BAB1_1948 ( bspF ) , BAB_10227 ( bspG ) and BAB2_0541 ( bspK ) were constructed by fusing 5′ and 3′ fragments flanking the open reading frame of the gene using PCR amplification and the relevant primers ( Table S4 ) . The 5′ fragments contained the upstream region , the start codon and the first 1 to 3 codons of each gene . The 3′ fragments contained the downstream region , the last 1 to 3 codons of each gene , and the stop codon . 5′ and 3′ fragments were fused and cloned into the BamHI and SalI sites of pJC80 using the In-Fusion PCR Cloning System ( Clontech ) and fully sequenced . To perform allelic replacement in the chromosome of B abortus 2308 strain , deletion plasmids were introduced by electroporation , carbenicillin-resistant , sucrose-sensitive electroporants were selected and sub-cultured further in TSB . Cultures were subjected to sucrose counter-selection by plating onto TSA plates supplemented with 5% sucrose ( wt/vol ) in order to isolate clones that had undergone allelic replacement . The presence of the deleted allele within the correct chromosomal locus was verified by PCR using primers listed in Table S3 . Independent clones carrying the correct in-frame deletion were isolated and used for further studies . The multiple deletion mutant ΔbspABF was generated by sequential allelic replacement . The B . abortus 2308 ΔbspB and ΔbspF in-frame deletion mutants were complemented using a derivative of the mini-Tn7 system . First , the plasmid pmTn7K was constructed by amplifying a 937 bp fragment from pBBR1MCS-2 containing the aphA3 gene and its own promoter region using primers RC7 and RC8 ( Table S3 ) and cloning into pUC18Tmini-Tn7 [60] digested with EcoRV using the In-Fusion PCR Cloning System , in order to generate a mini-Tn7 derivative expressing kanamycin resistance . The bspB and bspF genes and their respective promoter regions were subsequently cloned into pmTn7K using SpeI and BamHI . Briefly , the 561 bp bspB gene was amplified using primers RC389 and RC390 ( Table S3 ) , and a 150 bp fragment upstream of BAB1_0710 was amplified using primers RC360 and RC388 ( Table S2 ) . The two fragments were fused by overlap PCR and cloned into pmTn7K to generate pmTn7K-bspB using the In-Fusion PCR Cloning System . A fragment containing the 1283 bp bspF gene and 340 bp upstream of its start codon were PCR amplified using primers RC395 and RC396 ( Table S3 ) and inserted into pmTn7K to create pmTn7K-bspF . All inserts were confirmed by sequencing . The resulting pmTn7K-bspB and pmTn7K-bspF plasmids were introduced into the ΔbspB and ΔbspF mutants , respectively , together with the helper plasmid pTNS2 [60] by electroporation . Electroporants were selected on TSA plates containing 50 µg/ml of kanamycin , and clones carrying the mTn7K derivatives inserted within the attTn7 site downstream of the BAB2_0658 glmS gene were confirmed using PCR primers RC603 ( ATCATCCTCATCACCGACAA ) and RC604 ( GCTATATTCTGGCGAGCGAT ) . Murine bone marrow-derived macrophages ( BMMs ) from 6–12 week-old female C57BL/6J mice ( Jackson Laboratories , Bar Harbor , ME ) were generated and cultured as described before [61] . HeLa cells ( ATCC clone CCL-2 ) were cultured as described [13] . J774 . A1 macrophage-like cells ( ATCC , TIB-67 ) were maintained in Dulbecco's high glucose DMEM supplemented with 10% FBS and 4 mM L-glutamine at 37°C in 7% CO2 . Human Embryonic Kidney 293T ( HEK293T; ATCC CRL-11268 ) were cultured in Dulbecco's Modified Eagle's Medium ( DMEM ) supplemented with 10% FBS at 37°C in 7% CO2 . HeLa and HEK293T cells were transfected with either FuGENE HD or X-tremeGENE 9 transfection reagents ( Roche Applied Science ) according to the manufacturer's instructions at a ratio of 3∶1 ( reagent∶DNA; v/w ) . HeLa cells and BMMs were infected with Brucella strains at the appropriate multiplicity of infection ( MOI ) as previously described [13] . For experiments that required incubation at 40°C or 32°C , pre-warmed media was employed at each step . Temperature shifts from 40°C directly to 32°C were performed by removing the 40°C medium and replacing it with medium at 32°C . For infections of J774 . A1 cells , two days before infection , cells were seeded at either 8×103 ( 96-well plates ) or 5×104 ( 24-well plates ) cells/well in complete medium , then infected with Brucella strains in triplicate wells at the desired MOI . Following 30 min of incubation at 37°C , cells were washed five times with pre-warmed DMEM and incubated for another 30 min before a 1 h treatment with 100 µg/ml gentamicin to kill residual extracellular Brucella . Intracellular bacterial growth was evaluated in triplicates by lysing infected cells with 0 . 1% Triton X-100 in H2O at the indicated times after infection and plating a series of 1∶10 dilutions on TSA plates for colony-forming unit ( CFU ) enumeration . The translocation of translational fusions between TEM1 and the Brucella candidate proteins described in Table S1 was evaluated by detecting β-lactamase activity in infected J774 . A1 cells as previously described [19] . TEM1 fusions described in Table S3 were transformed in Brucella by electroporation and the expression of the fusions was verified by Western blot analysis with an anti-β-lactamase antibody ( 1∶1000 , QED Bioscience Inc , San Diego , CA ) . Bacteria were grown in TSB supplemented with 50 µg/ml of kanamycin overnight , then treated with 1 mM IPTG for 2 h to induce expression of fusion proteins before infection . J774 . A1 cells were then infected with Brucella strains harboring the TEM1 fusions at an MOI of 1000 in presence of 0 . 1 mM of IPTG throughout the infection . This MOI was necessary to achieve high levels of infection within the population ( >95% infected cells , data not shown ) and within individual cells and detect TEM fusion translocation . Infected cells were centrifuged at 400× g for 10 min to initiate bacterial-cell contact followed by incubation at 37°C for 30 min after which the cells were washed 5 times and incubated for another 30 min . At 1 h pi , cells were treated with gentamicin for 1 h to kill extracellular bacteria . At 16 h pi , cells were washed two times in DMEM and loaded with the fluorescent substrate CCF2/AM ( LiveBLAzer-FRET B/G loading kit; Invitrogen ) in the β-lactamase loading solution supplemented with 15 mM Probenecid ( Invitrogen ) . Cells were incubated in the dark for 90 min at room temperature and then observed under epifluorescence using a Carl Zeiss Axiovert 200 M fitted with a β-lactamase Blue/Aqua 41031 filterset ( Chroma Technology Corp . ) . At least 300 cells were counted in triplicate wells to determine the percentage of cells emitting a blue fluorescence ( TEM1-positive ) . The presented data are mean values ± SD from three independent experiments performed in triplicate . In each experiment , cells were then fixed and stained for Brucella ( see below ) to verify that >95% cells were infected , ensuring that the CCF2/AM conversion events occurred in infected cells . CyaA adenylate cyclase protein translocation assays were performed as previously described [20] , [28] with the following modifications . CyaA fusion constructs described in Table S3 were introduced into either the wild type B . abortus 2308 strain or its isogenic ΔvirB9 mutant [14] by electroporation , and their expression was detected by Western blot analysis using monoclonal anti-CyaA 3D1 antibody ( Santa Cruz Biotechnology , Inc . ) . Bacteria were grown in TSB supplemented with 50 µg/ml of kanamycin overnight and treated with 1 mM IPTG for 2 h before infection . J774 . A1 cells were infected at an MOI of either 1000 ( wild type strains ) or 5000 ( ΔvirB9 strains ) . These MOIs were necessary to achieve high levels of infection that allowed for detection of cAMP production . Infected cells were centrifuged at 400× g for 10 min followed by incubation at 37°C for 30 min after which the cells were washed 5 times and incubated for another 30 min . At 1 h pi , cells were treated with gentamicin for 1 h to kill extracellular bacteria . At 6 h post infection , cells were lysed and processed for cAMP levels using a colorimetric direct cAMP Enzyme Immunoassay Kit ( Sigma , CA200 ) according to the manufacturer's instructions . Each independent experiment was performed in triplicates and the levels of cAMP were normalized to the number of intracellular bacteria ( CFUs ) . Data are means ± SD from three independent experiments . For immunofluorescence staining , cells cultured on 12 mm glass coverslips in 24-well plates or in 96 well plates were fixed in 3% paraformaldehyde in PBS , pH 7 . 4 for 20 min at 37°C and processed for immunostaining as described previously [13] . Samples were observed with a Carl Zeiss MicroImaging AxioImager epifluorescence microscope for quantitative analysis or a LSM710 confocal laser scanning microscope for image acquisition ( Carl Zeiss Micro Imaging , Thornwood , NY ) . Representative confocal micrographs of 1024×1024 pixels were acquired and assembled for presentation using Adobe Photoshop CS3 . SEAP secretion assays were performed as previously reported [62] with some modifications . Briefly , HEK293T cells were seeded in 24-well plates to ∼60% confluency and co-transfected with plasmids expressing the indicated Brucella secreted proteins ( 300 ng DNA ) and the secreted embryonic alkaline phosphatase ( SEAP ) ( 200 ng DNA ) . At 16 h post transfection , cells were washed and fresh tissue culture medium was added . Eight hours later , media containing culture supernatant ( extracellular SEAP ) was removed to collect secreted SEAP and the cell associated ( intracellular ) SEAP was harvested by washing cells once with PBS and lysing them with equal volume of culture medium containing 0 . 5% Triton X-100 . SEAP activities were measured in triplicate wells using SEAP reporter gene assay , chemiluminescent kit ( Roche Applied Science ) . Data are presented as the SEAP secretion index , which is a ratio of extracelluar SEAP activity to intracellular SEAP activity normalized to values obtained from cells co-transfected with empty vector controls . Each experiment was carried out in triplicate and at least three independent experiments were performed . For SEAP secretion measurements in infected cells , HeLa cells were plated in 24-well plates at 5×104 cells/well . Sixteen hours after transfection , cells were infected with the indicated Brucella strains as previously described [13] . At 24 h post infection , all wells were treated with 5 µg/ml Brefeldin A ( BFA ) and 10 µg/ml of cycloheximide for 30 min to reversibly block secretion and novel SEAP biosynthesis and synchronize SEAP secretion . BFA was then removed with 5 washes using DMEM and subsequent incubation in complete medium containing 10 µg/ml cycloheximide to allow reconstitution of the secretory pathway and secretion of the SEAP pool . SEAP activity was measured in quadruplicate at indicated times post BFA washout and independent experiments were performed at least three times . Staining of the Golgi apparatus of cells grown on glass coverslips using a mouse anti-GM130 antibody was performed to evaluate the kinetics of secretory pathway reconstitution after BFA washout in uninfected and Brucella-infected cells . HeLa cells were co-transfected with equal amounts ( 0 . 5 µg ) of VSV-Gts045-GFP [63] , [64] and plasmids expressing HA-tagged Brucella effectors and incubated at 37°C for 4 h to allow transfection to occur . Cells were then moved to 40°C and incubated for 16–20 h at this non-permissive temperature to allow for VSV-Gts045-GFP accumulation in the ER . Cells were replenished with fresh medium containing 25 mM HEPES pH 7 . 4 and 10 µg/ml cycloheximide ( Sigma ) to stop further protein synthesis , and incubated for an additional hour at 40°C . Transfected cells were then moved to 32°C ( permissive temperature ) to allow VSV-Gts045-GFP to fold properly and traffic along the secretory pathway and then fixed and analyzed by immunofluorescence microscopy . In the case of Brucella infections , HeLa cells were first infected with Brucella for 24 h at 37°C , then transfected to express VSV-Gts045-GFP for 20 h at 40°C and further processed as above . Quantitative data of VSV-Gts045-GFP transport are means ± SD from three independent experiments . To evaluate ER stress , the transcriptional activity of the ER stress element ( ERSE ) was measured using the Cignal ERSE Reporter Luciferase Assay Kit ( SA Biosciences , Qiagen , Frederick , MD ) . The ERSE reporter assay is a mixture of an ERSE-responsive luciferase construct and a constitutively expressed Renilla luciferase construct ( 40∶1 ) . Following the manufacturer's instructions , HEK293T cells expressing HA- or GFP-tagged Brucella effectors were transfected in 96-well plates with either a ERSE reporter plasmid , a negative control , or a positive control ( a mixture of constitutively expressing GFP , constitutively expressing firefly luciferase , and constitutively expressing Renilla luciferase constructs ( 40∶1∶1 ) ) for 24 h according to the manufacturer's instructions . The ER stress inducer Tunicamycin ( 5 µg/ml ) was used as a positive control . At 32 h post-transfection , cells were washed with Dulbecco's PBS and harvested in 100 µl of Passive Lysis Buffer ( Promega ) and used to measure luciferase activity . The luciferase assay was developed with the Dual-Luciferase Reporter Assay System ( Promega ) according to the manufacturer's instructions . Bioluminescence was detected using a Tecan Infinite M1000 luminometer ( Tecan Group LTD ) . Data was processed by dividing the luminescence intensity of firefly luciferase by that of Renilla luciferase , to calculate the “relative luciferase activity” as the luminescence ratio of the test plasmid to that of the reporter-control . Experiments were performed in triplicates and repeated at least three times and the standard deviation is indicated . Expression levels of BiP and CHOP were determined by Western blot analysis using either an anti-BiP monoclonal antibody or a mouse anti-CHOP monoclonal antibody ( 1∶1 , 000 ) . The membranes were stripped and reprobed using an anti-Actin antibody ( 1∶25 , 000 ) to verify equal loading . Bacterial or mammalian cell lysates were generated using cell lysis buffer 10× ( Cell Signaling Technology , Inc ) . Samples were normalized according to colony forming units ( CFUs ) or protein concentrations where indicated , resolved on SDS-PAGE and transferred onto Amersham Hybond-ECL nitrocellulose membranes ( GE Healthcare ) . Western blots were probed using relevant primary antibodies , HRP-conjugated secondary anti-goat , mouse or rabbit antibodies , all diluted in in TBS-Tween 20 with 5% skim milk and developed using the ECL western blotting substrate ( Thermo Scientific , Pierce Protein Biology Products ) . Signals were acquired using a Kodak Image Station 4000MM Pro and assembled for presentation using Adobe Photoshop CS3 . Six to eight weeks old BALB/c female mice purchased from Jackson Labs ( Bar Harbor , ME ) were acclimated for a minimum of 1 week prior to infection and used in experimental groups of 5 animals . Mice were infected intraperitoneally ( i . p . ) with a total dose of approximately 105 CFUs of B . abortus strain wild type strain 2308 or its isogenic bsp mutants suspended in 100 µl of PBS . Infectious doses were confirmed by plating serial dilutions of inocula on TSA plates . Groups of five mice per strain were euthanized at day 3 , 7 and 42 post inoculation by isoflurane inhalation overdose followed by cervical dislocation . Spleens and livers were collected aseptically . Spleens were weighed to evaluate splenomegaly and bacteria were enumerated from livers and spleens through homogenization in PBS , plating of serial 10-fold dilutions on TSA plates and growth at 37°C for three days . All animal rearing , handling and experimental methods were conducted under protocols ( #2010-043 and #2012-031 ) approved by the RML Institutional Animal Care and Use Committee ( IACUC; USDA Permit Number: 51-F-0016 , PHS number: A4149-01 ) in strict accordance with the recommendations of the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health . All infections were performed in an Animal Biosafety Level 3 ( ABSL3 ) facility according to protocols reviewed and approved by the Rocky Mountain Laboratories Institutional Biosafety Committee and the RML Institutional Animal Care and Use Committee ( IACUC ) , in compliance with the CDC Division of Select Agents and Toxins regulations . Statistical analysis was performed using the GraphPad Prism software . All results are presented as means ± SD from at least three independent experiments , unless otherwise stated . Statistical significance was determined by either an unpaired , two-tailed Student t test or in the case of groups , a one-way ANOVA followed by the Tukey's test . A P value<0 . 05 indicates a statistically significant difference .
|
Many intracellular parasites ensure their survival and proliferation within host cells by secreting an array of effector molecules that modulate various cellular functions . Among these , Brucella abortus , the causative agent of the worldwide zoonosis brucellosis , controls the intracellular trafficking of its vacuole , the Brucella-containing vacuole ( BCV ) , towards compartments of the secretory pathway via the expression of a Type IV secretion system ( T4SS ) , VirB , which is thought to translocate effector proteins . Here , we have used bioinformatic algorithms and protein translocation reporter assays to identify novel Brucella proteins translocated into host cells , some of which are VirB T4SS substrates and targeted secretory pathway compartments when ectopically expressed in mammalian cells . Three VirB effectors , BspA , BspB and BspF , inhibited protein secretion and contributed to varying degrees to bacterial inhibition of host protein secretion , pathogen intracellular growth and persistence in the liver of infected mice . These findings demonstrate that Brucella modulates secretory trafficking via multiple T4SS effector proteins to promote Brucella pathogenesis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"genome",
"analysis",
"tools",
"genomics",
"molecular",
"cell",
"biology",
"emerging",
"infectious",
"diseases",
"microbial",
"pathogens",
"membranes",
"and",
"sorting",
"host-pathogen",
"interaction",
"microbiology",
"biology",
"pathogenesis",
"bacterial",
"pathogens"
] |
2013
|
Brucella Modulates Secretory Trafficking via Multiple Type IV Secretion Effector Proteins
|
The diagnosis of cystic echinococcosis ( CE ) is based primarily on imaging , in particular with ultrasound for abdominal CE , complemented by serology when imaging results are unclear . In rural endemic areas , where expertise in ultrasound may be scant and conventional serology techniques are unavailable due to lack of laboratory equipment , Rapid Diagnostic Tests ( RDTs ) are appealing . We evaluated the diagnostic accuracy of 3 commercial RDTs for the diagnosis of hepatic CE . Sera from 59 patients with single hepatic CE cysts in well-defined ultrasound stages ( gold standard ) and 25 patients with non-parasitic cysts were analyzed by RDTs VIRapid HYDATIDOSIS ( Vircell , Spain ) , Echinococcus DIGFA ( Unibiotest , China ) , ADAMU-CE ( ICST , Japan ) , and by RIDASCREEN Echinococcus IgG ELISA ( R-Biopharm , Germany ) . Sensitivity , specificity and ROC curves were compared with McNemar and t-test . For VIRapid and DIGFA , correlation between semiquantitative results and ELISA OD values were evaluated by Spearman’s coefficient . Reproducibility was assessed on 16 randomly selected sera with Cohen’s Kappa coefficient . Sensitivity and Specificity of VIRapid ( 74% , 96% ) and ADAMU-CE ( 57% , 100% ) did not differ from ELISA ( 69% , 96% ) while DIGFA ( 72% , 72% ) did ( p = 0 . 045 ) . ADAMU-CE was significantly less sensitive in the diagnosis of active cysts ( p = 0 . 019 ) while DIGFA was significantly less specific ( p = 0 . 014 ) compared to ELISA . All tests were poorly sensitive in diagnosing inactive cysts ( 33 . 3% ELISA and ADAMU-CE , 42 . 8% DIGFA , 47 . 6% VIRapid ) . The reproducibility of all RDTs was good-very good . Band intensity of VIRapid and DIGFA correlated with ELISA OD values ( r = 0 . 76 and r = 0 . 79 respectively , p<0 . 001 ) . RDTs may be useful in resource-poor settings to complement ultrasound diagnosis of CE in uncertain cases . VIRapid test appears to perform best among the examined kits , but all tests are poorly sensitive in the presence of inactive cysts , which may pose problems with accurate diagnosis .
Cystic echinococcosis ( CE ) is a parasitic zoonosis caused by the larval stage of the dog tapeworm Echinococcus granulosus complex . The parasite is transmitted between canids ( definitive hosts harboring in the intestine the adult stage of the tapeworm ) , and livestock , particularly sheep ( intermediate hosts becoming infected by fecal-oral route with eggs shed with dog feces ) . In the intermediate host , the larval stage develops as an expanding fluid-filled cyst , which can infect the definitive host eating infected organs . Humans behave as accidental intermediate hosts , where CE cysts develop mostly in the liver , followed by lungs . The infection is prevalent worldwide especially in rural livestock-raising areas such as the Mediterranean , Eastern Europe , North and East Africa , South America , Central Asia , China and Australia . The most recent estimates indicate 1 . 2 million people affected worldwide with 3 . 6 million Disability Adjusted Life Years lost due to human disease and over 2 , 190 million USD lost yearly in animal production [1] . Human CE is a chronic , clinically complex and neglected disease [2] . The spectrum of clinical manifestations range from asymptomatic to serious , even life-threatening conditions . Most cases remain a- or pauci-symptomatic for years or even decades and maybe diagnosed accidentally . The diagnosis of human CE is mainly based on imaging . Ultrasound ( US ) is the imaging technique of choice for the diagnosis of abdominal CE [3] . The current international WHO-IWGE ( Informal Working Group on Echinococcosis ) classification of CE cyst stages is based on the pathognomonic features of cysts on US , and guides their clinical management [4 , 5] . Serology should complement imaging-based diagnosis when imaging features are unclear , although currently available serology tests are burdened by the lack of standardization and by unsatisfactory sensitivity and specificity [6 , 7] . In underserved rural endemic areas , the diagnosis of CE poses important problems as expertise in US diagnosis and management of CE may be scant and/or difficult to access , and conventional serology techniques are unavailable or unreliable due to the lack of laboratory equipment . These conditions may not only cause under-diagnosis of CE in patients requiring therapy , but also result in poor differential diagnosis and unnecessary or inappropriate treatments . This is particularly true when serology is used alone without visualization of a compatible lesion by imaging , as the positive predictive value of CE serodiagnosis is very low [8] , and when lesions do not show pathognomonic signs of a parasitic origin , such as young CE1 cysts or inactive CE4–CE5 cysts . Unfortunately , these stages are also those with the broader differential diagnosis ( e . g . simple cysts , neoplastic lesions ) whose serology results are also difficult to interpret and often negative [9] . The use of Rapid Diagnostic Tests ( RDTs ) is particularly useful is resource-poor settings , and in the context of CE they may be suitable to complement imaging where diagnosis is uncertain . Several reports described the performance of commercial and experimental RTDs in the diagnosis of CE [10–15]; however , no study so far compared the performance of commercially available RDTs . Here we performed a comparison of the diagnostic performance and reproducibility of three commercially available RDTs for the diagnosis of CE and compared them with those of a commercial ELISA test routinely used in the parasitology diagnostic laboratory of San Matteo Hospital Foundation , Pavia , Italy . Our results show that the evaluated RDTs have an overall comparable performance to the ELISA test in the diagnosis of hepatic CE in well-defined stages , although significant differences exist among them . If confirmed in a bigger cohort , these results would support the use of RDTs instead of conventional techniques to complement imaging in the diagnosis of CE .
Sera included in the analysis were frozen ( -80°C ) stored samples from patients with hepatic CE and non-parasitic hepatic cysts seen between 2010 and 2015 in the Ultrasound Diagnostic Service of the Division of Infectious and Tropical Diseases , San Matteo Hospital Foundation , Pavia , Italy , where the WHO Collaborating Centre for Clinical Management of Cystic Echinococcosis is based . Clinical information related to each patient and sample was retrieved retrospectively in March 2015 from the electronic database of patients visited in the Centre . Patients included in the study formed a convenience series . Selection criteria were presence of a single cyst , located in the liver , of non-parasitic nature ( controls ) or with a well-defined CE stage according to the WHO-IWGE classification , as assessed by abdominal US by an experienced sonographer ( EB ) ( gold standard ) . When possible , sera were collected from people who had never received treatment for CE or whose treatment ended > 12 months before serum collection . Patients with non-parasitic hepatic cysts were used as controls because non-parasitic cysts represent the most common differential diagnosis of hepatic CE cysts . All patients signed the informed consent for storage and scientific use of the leftover serum at the moment of blood sampling for routine serology . Ethics approval was granted by the Ethics Committee of San Matteo Hospital Foundation , Pavia ( approval n . 20150004877 ) . Cysts were classified according to the WHO-IWGE classification . For the analysis , CE cysts were grouped into active ( CE1 , CE2 , CE3a and CE3b ) and inactive ( CE4 and CE5 ) . Experimental and clinical data prove that CE3b are biologically active ( i . e . viable ) cysts , while CE3a cysts can be both biologically active or not [16–18] . However , in our analysis , we grouped CE3a cysts with the other active stages as disruption of the integrity of the cyst wall , irrespective of the viability of the cyst , allows parasite antigens to stimulate antibody production . Therefore , it can be speculated that cyst wall integrity is likely a more important condition than the biological viability per se to influence serological responses . Patients with small CE1 cysts are often seronegative , although cysts in this stage are unequivocally active [19]; thus this stage should likely be grouped independently in serology analysis . However , not enough samples were present to carry out this sub-analysis . CE4 cysts that reached inactivation spontaneously but recently ( or only temporarily inactivated after unsuccessful treatment ) should likely also be grouped with “active cysts” , while stably inactive CE4 and CE5 cysts constitute the real “inactive cysts” group [20 , 21] . However , this more precise classification at present is not possible in the absence of either long-term follow-up of active cysts without therapy or performing invasive sampling for the assessment of biological activity of cysts , both options burdened by practical and ethical constraints . Therefore , CE4 and CE5 cysts are grouped here in the inactive group . These considerations are at the basis of the choice of cyst grouping used in this study . Selected sera were analyzed using the following three commercially available immunochromatographic rapid diagnostic tests: VIRapid HYDATIDOSIS ( based on purified antigen B and antigen 5; Vircell , Salamanca , Spain ) , Echinococcus Dot Immunogold Filtration Assay ( DIGFA , based on purified cyst fluid , protoscolex antigen , antigen B and antigen Em2 of E . multilocularis; Unibiotest , Wuhan , China ) , and ADAMU-CE ( based on recombinant antigen B; ICST , Saitama , Japan ) , following the manufacturer’s instructions . The sera were also tested in double with RIDASCREEN Echinococcus IgG ELISA ( R-Biopharm , Darmstadt , Germany ) , routinely used in the parasitology diagnostic laboratory of San Matteo Hospital Foundation , Pavia , Italy , following manufacturer’s instructions . For the ELISA test , Optical Density ( OD ) results were used to calculate and interpret a Sample Index ( SI ) , as per manufacturer’s instructions . ELISA results were considered positive for SI ≥1 . 1 , negative for SI <0 . 9 , and border line for 0 . 9 ≤SI<1 . 1 . Borderline results were considered negative for the analysis of results . In this work , “OD” will always refer to Sample Indexes , not to raw OD values; the terminology “OD” was preferred due to more immediate understanding . All tests were performed in parallel in a single session in April 2015 . Each test was read by a single operator experienced in laboratory procedures ( FT for ELISA and VIRApid , MM for DIGFA , IC for ADAMU-CE ) . Readers were blind to cyst stage and to results of other tests at the time of reading . Results were recorded as positive or negative and the semiquantitative colorimetric reading of tests was also recorded for VIRapid HYDATIDOSIS and DIGFA tests , as well as OD values for the ELISA test . For DIGFA test , positivity was considered when either “Echinococcus spp” or”E . granuolosus or E . multilocularis” or “E . granulosus” indicators were present and the semiquantitative reading was based on the color intensity of the least intense spot . Examples of RDTs results are shown in Fig 1 . The sample size was constrained by the procurement of tests . With a sample of 84 subjects , the study had 80% power for the pairwise comparison of the Area Under the ROC Curve ( AUC ) of the RDTs , calculated according to the method of Obuchowski [22] and based on the diagnostic performances of the ELISA test , as assessed in a previous work [9] . Difference in AUC was set at 15% and the correlation between two tests at 0 . 3 . Alpha value was set at 0 . 01 to account for multiple comparisons . The Shapiro-Wilk test was used to test the normal distribution of quantitative variables . When quantitative variables were normally distributed , the results were expressed as the mean value and standard deviation ( SD ) , otherwise median and interquartile range ( IQR; 25th -75th percentile ) were reported . Qualitative variables were summarized as counts and percentages and differences were analysed with Chi-square test or Fisher exact test , as appropriate . For each test , overall and group-specific ( active vs inactive ) Sensitivity and Specificity values , as well as AUC , were calculated together with their 95% Confidence Interval ( CI ) . US classification of cysts was considered the gold standard . The performance of the RDTs was compared to those of the ELISA test using McNemar and t-test as appropriate . The semiquantitative reading values of VIRapid HYDATIDOSIS and DIGFA tests were correlated with the ELISA OD values using Spearman’s rank correlation coefficient . Sixteen sera were randomly selected using an electronic random numbers generator and re-analyzed with the three RDTs for assessment of result reproducibility using Cohen’s Kappa coefficient . P<0 . 05 was considered statistically significant . A Bonferroni-Holm correction was applied for multiple test . All tests were two-sided . The data analysis was been performed with the STATA statistical package ( release 13 . 1 , 2014 , Stata Corporation , College Station , Texas , USA ) .
Eighty-four sera from 84 patients fulfilling inclusion criteria were available for the study . Of these , 59 were patients with single CE cysts of the liver , while 25 had single non-parasitic hepatic cysts . Of the 59 CE patients , 38 had active and 21 inactive cysts , according to the WHO-IWGE classification . Eleven ( 18 . 6% ) CE patients had received medical treatment with albendazole before sample collection ( median 19 . 4 months before; IQR 10 . 6–51 . 6; range 3 . 1–113 . 0 ) . The size of the cyst ( larger diameter ) was not significantly different between active and inactive CE cysts ( p = 0 . 82 ) , while non-parasitic cysts were significantly smaller than CE cysts ( p<0 . 001 ) . Clinical and demographic characteristics of included patients and sera are summarized in Table 1 . In one case VIRapid HYDATIDOSIS gave an invalid result ( absence of the control band ) and was therefore excluded from the analysis . In no case did the DIGFA test give a univocal “E . multilocularis” result . In 19 ( 38% ) cases ( 13 out of the 43 [30 . 23%] CE cysts with positive serology and in 6 out of the 7 [85 . 71%] non-parasitic cysts with positive serology ) DIGFA test failed to individuate E . granulosus , but provided an “Echinococcus spp” result or an “E . granulosus or E . multilocularis” result . General test sensitivity and specificity and comparison with the results of the ELISA test are shown in Table 2 . The performance of VIRapid HYDATIDOSIS was not statistically different from those of the ELISA test , while those of DIGFA ( p = 0 . 045 ) and ADAMU-CE ( p = 0 . 074 ) showed a borderline significant difference . When we analyzed Sensitivity and Specificity of the tests within groups ( active , inactive , and non-parasitic ) , we found that ADAMU-CE was significantly less sensitive in the diagnosis of active cysts ( p = 0 . 019 ) , and DIGFA was significantly less specific when applied on samples from patients with non-parasitic cysts ( p = 0 . 014 ) , compared to ELISA ( Table 2 ) . Although a statistical analysis by individual CE stage was not possible due to the limited number of samples , results are indicated in Table 2 . To explore the discrepancies between ELISA and RDTs results , we analyzed the percentage of positive and negative RDTs results of sera from CE patients stratified by ELISA OD groups , set as follows: negative OD < 1 . 1; low-positive 1 . 1 ≤ OD ≤ 5 . 0; high-positive OD > 5 . 0 . The threshold between low-positive and high-positive OD values was set arbitrarily . As shown in Table 3 , we found that for all RDTs the percentage of positive results increased passing from negative to low-positive to high-positive ELISA OD groups , with discrepancies between ELISA and RDTs tests being most frequent in the low-positive ELISA OD group . ROC AUC characteristics and results of comparison between ROC curves are shown in Table 4 and Fig 2 . In this analysis a statistically borderline significant difference was seen only between VIRapid HYDATIDOSIS and DIGFA ( p = 0 . 042 ) . The reproducibility of the RDTs was good ( DIGFA , k = 0 . 71; ADAMU-CE , k = 0 . 62 ) to excellent ( VIRapid HYDATIDOSIS , k = 1 . 00 ) . When we examined the correlation between ELISA OD values and the visual semiquantitative reading of band/dots color intensity of VIRapid HYDATIDOSIS and DIGFA , respectively , we found a significant positive correlation in both cases ( p < 0 . 001 ) , as shown in Fig 3 .
In rural underserved areas , where CE is most prevalent and health systems are basic and/or difficult to access , the availability of RDTs to help in the differential diagnosis of suggestive US lesions would be very useful . Although several reports described the performance of experimental RDTs in the diagnosis of CE , studies assessing and comparing the diagnostic accuracy of commercially available tests are very scant [10–15] . Feng and colleagues [12] , using DIGFA with sera from China , reported a sensitivity of 83 . 4% for hepatic CE and a specificity of 93 . 4% when sera came from hospitalized patients . In our centre , the DIGFA test gave clearly inferior results; while our results were comparable with those found by the authors when sera from US screening campaigns were used ( Se 71 . 8% for abdominal CE; Sp 78 . 1% ) . Feng and colleagues attributed these differences to the presence , in the field setting , of subjects exposed to the parasite without developing detectable lesions , or the presence of old lesions not accompanied by positive serology . However , sera from hospitalized patients were collected less than 2 years after surgical treatment for CE . Moreover , the authors did not mention the distribution of CE stages in the two patient cohorts . It is therefore likely that the different performance between the two cohorts , and with our results , are at least in part due to the difference in these variables , known to affect serology results . Santivanez and colleagues [23] using a previous form of the ADAMU-CE test on a panel of sera from surgically confirmed CE patients , found a better Se ( 80% on sera from liver cysts ) and same Sp ( 100% if sera from patients with alveolar echinococcosis were excluded—89 . 8% if included ) compared to our results . In their work , however , they do not provide details of the cyst stages . Therefore , the differences with our results may be at least in part due to differences in these conditions , although different performances between the two “versions” of the kit may not be excluded . Similarly , Tamer and colleagues [14] , evaluating the performances of the VIRapid test , reported a better Se ( 96 . 8% ) and same Sp ( 96% if sera from patients with other parasitoses were excluded – 87 . 5% if included ) compared to our results , but they did not provide data on cyst characteristics; all CE patients included in their cohort where surgically confirmed , suggesting that predominantly active CE stages were included . Finally , Chen and colleagues [11] , using sera from hospital cases , reported that the use of recombinant antigens in the DIGFA test might improve the specificity of the test , but at the expense of sensitivity . In our centre , the VIRapid HYDATIDOSIS test showed the overall best diagnostic accuracy among the three RDTs , although it did not result statistically significant better than the ELISA test . On the contrary , in comparison with the ELISA test , the ADAMU-CE test was significantly less sensitive in the diagnosis of active cysts , while the DIGFA test was significantly less specific . These results are in line with the literature , reporting overall better sensitivity for tests based on native antigens and better specificity for those based on recombinant antigens [6 , 7 , 19] . Not surprisingly , all RDTs were as poorly sensitive as the ELISA test in the diagnosis of inactive cysts . These results confirm the limits of serology in the diagnosis of CE and in supporting the differential diagnosis of CE1 and CE4-CE5 cysts from other hepatic lesions . Evidence exist that patients with CE have both common and stage-specific serology profiles , indicating that the development of both infection- and stage-specific immunoassays is possible [24 , 25] . Ahn and coworkers [24] showed that antigen 5 seems to be immunoreactive in every stage , as opposed to antigen B , whose proteoforms revealed a reduced antibody capture in CE1 , CE4 and CE5 stages . Unfortunately , so far , conventional methods used for antigen discovery such as 2D gel electrophoresis of cyst fluid and immunoblot using sera from infected patients did not allow the identification of stage-specific antigens to be used , alone or in a cocktail , for a more sensitive and stage-specific diagnosis and follow-up of patients . Clearly this should be the focus of high-priority work in the filed . As mentioned previously , many variables are known to influence CE serology results [9 , 19] . In this study , only sera from subjects with a single cyst located in the liver have been included , excluding number and location of the cyst influencing results . Similarly , the size of the cyst was not significantly different between active and inactive cysts; therefore this variable should not have significantly influenced the results . Finally , only 3 out of the 11 CE patients who were treated with albendazole before serum collection ended drug intake less than 12 months prior to sampling . Therefore , the influence of this variable should have only marginally influenced our results given that previous research has shown that treatment that has ended more than 1 year before sampling does not have a significant impact of ELISA test results [9] . The samples size of this study was constrained by the strictness of the inclusion criteria and by the procurement of tests . However , it is pivotal that the first evaluation of diagnostic tests is performed on well-characterized and homogeneous samples . This , unfortunately , is very rarely done , with consequent problems in the interpretation and reliability of the results . The limitation of the number of samples that could be included in this work to comply with this principle was therefore weighted against the quality of baseline data on the evaluation of the RDTs that such an approach could provide . In this work , we included sera from patients with non-parasitic cysts as controls because non-parasitic cysts represent the most common differential diagnosis of hepatic CE cysts . Surely further work should thoroughly evaluate the specificity of the tests with sera from patients with other parasitoses , in particular alveolar echinococcosis . However , it must be stressed that serology for CE should be performed only after lesions compatible with echinococcosis are found by imaging , to increase the pre-test probability of the presence of infection . Indeed , due to the low prevalence of infection ( and consequent very low Positive Predictive Value of any test ) , the generally low specificity of serodiagnostic tests ( especially an issue in areas where contact with the parasite without cyst development and number of other diseases affecting the population may be significant ) , and the very low sensitivity of serodiagnosis in extra-hepatic CE ( limiting the use of serology to diagnose CE in organs not explorable by ultrasound ) , the value of serological screenings is limited and should be conducted only after careful evaluation of the scientific question such studies want to answer . To conclude , our results show that RDTs have overall comparable performance to the routine ELISA test in the diagnosis of hepatic CE in well-defined stages , although significant differences in diagnostic accuracy exist among them . These results support their use in resource-poor settings to complement ultrasound diagnosis of CE in doubtful cases . However , all tests are poorly sensitive in the presence of inactive and CE1 cysts , which are cyst stages that may pose considerable problems of differential diagnosis . Furthermore , further studies are warranted to explore the performance of RDTs in the follow-up of CE patients , often extremely difficult to perform with regular US examinations in endemic areas . VIRapid HYDATIDOSIS appeared to perform best among the examined kits and deserves further testing with a larger cohort including other control groups ( e . g . with other parasitoses ) and sera from patients with extra-hepatic CE cysts and with CE cysts of different parasite genotypes . The test also deserves further evaluation in the field and with the use of whole blood from fingerprick sampling . Finally , benefit studies on the use of RDTs , and serology tests in general , are lacking in the field of CE and deserve future efforts .
|
Cystic echinococcosis ( CE ) is a parasitic zoonosis prevalent worldwide , especially in economically poor livestock raising areas . Parasitic cysts develop most commonly in the liver and are diagnosed primarily by ultrasound . Serology helps with diagnosis , particularly when ultrasound features are unclear . Unfortunately , in underserved endemic rural areas , expertise in ultrasound diagnosis of CE may be scant , and conventional serology techniques are unavailable due to the lack of laboratory equipment . In these circumstances , Rapid Diagnostic Tests ( RDTs ) may be very useful . In this work , we evaluated the diagnostic performance of three RDTs and compared them with a commercial ELISA test routinely used in our diagnostic laboratory . Our results show that RDTs have overall comparable performances to ELISA in the diagnosis of hepatic CE in well-defined stages , although significant differences exist among them . If confirmed and expanded on a bigger cohort , these results would support the use of RDTs instead of conventional techniques to complement imaging in the diagnosis of CE .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"enzyme-linked",
"immunoassays",
"diagnostic",
"radiology",
"pathology",
"and",
"laboratory",
"medicine",
"ultrasound",
"imaging",
"tropical",
"diseases",
"parasitic",
"diseases",
"animals",
"echinococcus",
"veterinary",
"diagnostics",
"neglected",
"tropical",
"diseases",
"immunologic",
"techniques",
"veterinary",
"science",
"veterinary",
"medicine",
"research",
"and",
"analysis",
"methods",
"echinococcosis",
"serology",
"imaging",
"techniques",
"serodiagnosis",
"immunoassays",
"flatworms",
"helminth",
"infections",
"radiology",
"and",
"imaging",
"diagnostic",
"medicine",
"biology",
"and",
"life",
"sciences",
"organisms"
] |
2016
|
Comparison of the Diagnostic Accuracy of Three Rapid Tests for the Serodiagnosis of Hepatic Cystic Echinococcosis in Humans
|
The high level of functional diversity and plasticity in monocytes/macrophages has been defined within in vitro systems as M1 ( classically activated ) , M2 ( alternatively activated ) and deactivated macrophages , of which the latter two subtypes are associated with suppression of cell mediated immunity , that confers susceptibility to intracellular infection . Although the Leishmania parasite modulates macrophage functions to ensure its survival , what remains an unanswered yet pertinent question is whether these macrophages are deactivated or alternatively activated . This study aimed to characterize the functional plasticity and polarization of monocytes/macrophages and delineate their importance in the immunopathogenesis of Post kala-azar dermal leishmaniasis ( PKDL ) , a chronic dermatosis of human leishmaniasis . Monocytes from PKDL patients showed a decreased expression of TLR-2/4 , along with an attenuated generation of reactive oxidative/nitrosative species . At disease presentation , an increased mRNA expression of classical M2 markers CD206 , ARG1 and PPARG in monocytes and lesional macrophages indicated M2 polarization of macrophages which was corroborated by increased expression of CD206 and arginase-1 . Furthermore , altered vitamin D signaling was a key feature in PKDL , as disease presentation was associated with raised plasma levels of monohydroxylated vitamin D3 and vitamin D3- associated genes , features of M2 polarization . Taken together , in PKDL , monocyte/macrophage subsets appear to be alternatively activated , a phenotype that might sustain disease chronicity . Importantly , repolarization of these monocytes to M1 by antileishmanial drugs suggests that switching from M2 to M1 phenotype might represent a therapeutic opportunity , worthy of future pharmacological consideration .
Leishmaniases comprise a group of heterogeneous parasitic diseases caused by the protozoan parasite Leishmania with its spectrum ranging from a self-healing cutaneous variant to the often fatal visceral leishmaniasis ( VL ) . Post kala-azar dermal leishmaniasis ( PKDL ) is the dermal sequel of VL , wherein Leishmania parasites remain restricted to the skin and manifest as nodular , papular , hypopigmented macular lesions , erythematous plaques and/or a mixed phenotype , termed as polymorphic [1] . The etiopathogenesis of PKDL is still unclear and a consensus is yet to emerge regarding possible causes for the generally viscerotropic L . donovani parasite to generate PKDL . In PKDL , similar to other leishmaniasis , Leishmania have developed several strategies to outmanoeuvre host immunity via subverting and/or suppressing macrophage microbicidal activities [2] . It is universally accepted that monocytes-macrophages have a range of biological roles being inducers , regulators and effectors of innate and acquired immunity . They also actively participate in physiological processes associated with wound healing and tissue repair [3] . Upon stimulation with Th1-associated cytokines , notably IFNγ , they acquire a heightened effector function against intracellular pathogens , referred to as a classically activated or M1 phenotype . Conversely , in the milieu of Th2 associated cytokines e . g . IL-4 , IL-13 , IL-33 , TGF-β and IL-10 [4] or via microbial triggers [5] , there is M2 polarization or alternative activation . The differentiation of M1 and M2 monocytes/macrophages is regulated by cardinal genes that include inducible nitric oxide synthase ( iNOS ) , arginase 1 ( ARG1 ) , mannose receptor ( CD206 ) and Fizz1 among others [4 , 6] . M2 macrophages can impede protective immunity to protozoan infection . In an animal model of cutaneous leishmaniasis ( CL ) , Holscher et al . , [7] demonstrated that alternative activation favoured disease progression , whereas the impairment of M2 macrophages significantly delayed disease progression . In studies regarding human leishmaniasis , raised levels of arginase have been demonstrated in neutrophils and low density granulocytes [8 , 9] . Similarly , in diffuse cutaneous leishmaniasis ( DCL ) , a more chronic form of leishmaniasis more akin to PKDL , there was an elevation of the arginase pathway ( arginase-1 , ornithine decarboxylase and polyamine ) [10] . Furthermore , this study validated that inhibition of arginase-1 or ornithine decarboxylase abrogated parasite replication within human macrophages [10] . Our understanding of the phenotypic and functional complexity of M2 monocytes-macrophages is limited by discordance between data derived from murine vs . human systems [11] . Unlike classically activated macrophages , where human and murine cells respond similarly , the molecular phenotype of alternatively activated macrophages in mice and humans have to date shown a limited overlap [12] . Additionally , as the dichotomy of M1 vs . M2 is still not clearly defined [4] , it emphasized the importance of undertaking human studies , especially with regard to infectious diseases . Accordingly , this study aimed to delineate in patients with PKDL the activation status of monocytes in peripheral blood and dermal macrophages , thus providing the first characterization of M2 polarized macrophages in human dermal leishmaniasis . Furthermore , we demonstrate the repolarization of monocytes after antileishmanial chemotherapy suggesting that therapeutic options designed to restore the M1:M2 balance may be effective in disease elimination .
The study received approval from School of Tropical Medicine Kolkata ( STM ) and Institute of Postgraduate Medical Education and Research , Kolkata . Studies on tissue biopsies were approved by the UK National Research Ethics Service . Written informed consent was obtained and for a minor , their legally accepted representative provided the same . From 2010–11 , patients suspected with PKDL ( n = 34; Table 1 ) were recruited from the Dermatology Outpatient Department , STM , based on clinical features and a prior history of VL or were resident in a VL endemic area . Diagnosis was confirmed by the rK39 strip test ( In Bios International , Seattle , USA ) and ITS-1 PCR from skin biopsies . Patients received Miltefosine ( 100 mg/day p . o . 4 months , n = 14 ) or sodium antimony gluconate ( 20 mg/kg b . w . , i . m . 4 months , n = 20 ) . Age and sex-matched healthy volunteers ( n = 15 ) recruited from non-endemic areas were seronegative for anti-leishmanial antibodies . Samples were collected at disease presentation and upon completion of treatment ( Table 1 ) , which varied from 3–4 months , based on patient compliance . The analysis plan is shown in Fig 1 . Peripheral blood diluted with phosphate buffered saline ( 0 . 01M pH 7 . 2 , PBS ) was layered over a monocyte isolation medium ( 3:1; HiSep LSM-1073 ) and centrifuged ( 400g , 30 minutes ) . The monocyte rich interface was washed twice in PBS and then resuspended in RPMI-1640 , supplemented with penicillin ( 100 U/mL ) , streptomycin ( 100 μg/mL ) and 10% heat-inactivated fetal bovine serum . To confirm purity , monocytes were initially gated on their forward vs . side scatter characteristics followed by CD14 positivity . The absence of PMNs was checked by CD15 negativity and cells were used for immunophenotyping and/or mRNA expression studies . After isolation of monocytes , they were surface stained with anti human CD14 FITC ( Biolegend , San Diego , CA , USA ) and incubated for 30 minutes at room temperature ( RT ) . The cells were then washed twice with PBS followed by fixation and permeabilization by incubating with a fix-perm buffer ( 2% paraformaldehyde + 0 . 05% saponin + 3% FBS in PBS ) for 20 minutes at RT . Cells were then stained with anti human TLR-2 PE and TLR-4 FITC ( BD Biosciences , San Jose , CA , USA ) for 15 minutes . Cells were then washed twice and resuspended in PBS-2% FBS for acquisition in a flow cytometer ( BD FACS Calibur , BD Biosciences , San Jose , CA , USA ) . Plasma levels of circulating cytokines , IL-4 , IL-10 and IL-13 were measured in patients with PKDL and healthy controls by sandwich ELISA according to the manufacturer’s instructions ( Immunotools , Friesoythe , Germany ) . Whole blood ( 100 μL ) was stained for 20 minutes with either CD14 FITC and CD16 PE or antihuman CD14-PerCP and CD206-Alexafluor 488 with appropriate isotype controls [Biolegend , San Diego , USA , 13] . Cells were then washed twice with PBS , resuspended in 400 μL of PBS and acquired in a Flow Cytometer . For intracellular staining , monocytes ( 1x106 cells/well/mL ) were cultured overnight , followed by Brefeldin A ( 1 μg/mL , 4h ) and surface stained with CD14-FITC [Biolegend , San Diego , USA . They were then stained for IL-6-PE , IL-1β-PE ( eBioscience , San Diego , USA ) ] , IL-8-APC , IL-12p40-PE , Latency associated peptide ( LAP ) -TGF-β1-APC ( Biolegend , San Diego , USA ) along with isotype controls and acquired in a flow cytometer . 5000 monocytes were acquired and data analyzed by Cell-Quest pro software . The frequency of cells with a particular phenotype was expressed as % of monocytes and was calculated by dividing percentages of the upper right quadrant ( CD14+ marker+ ) by the sum of the upper and lower right quadrant ( CD14+ marker- ) . Monocytes ( 5x105 ) after centrifugation ( 400g , 5 minutes ) were resuspended in PBS and stained with 4 , 5-diaminofluorescein diacetate ( DAF-2DA , 2 μM , 30 minutes , 37°C , Cayman Chemicals , Ann Arbor , Michigan , USA ) . The fluorescence of DAF-2T was acquired in a flow cytometer in the FL-1 channel [14] . The generation of ROS and levels of non protein thiols was measured in monocytes ( 5 x 105/mL ) stained with 5- ( and-6 ) -carboxy-2' , 7'-dichloro dihydrofluorescein diacetate , acetyl ester ( CMH2DCFDA , 2 . 5 μM ) and 5-chloromethylfluorescein diacetate ( CMFDA , Molecular Probes , Carlsbad , CA , USA ) respectively [15]; the fluorescence of DCF and CMF was acquired in a flow cytometer . Superoxide production was measured using the cytochrome c reduction assay [16] . For analysis , monocytes were gated on their forward vs . side scatter characteristics as previously shown [14]; 5000 monocytes were acquired and data analyzed by Cell-Quest pro software . The expression of markers was indicated as geometric mean fluorescence channel or GMFC . To minimise day to day experimental variation and auto-fluorescence , an unstained control was included for each sample . Patient samples were analyzed alongside a healthy control , to minimise the effects of any temporal changes in experimental setup Using total RNA extracted from monocyte enriched PBMCs ( 1x106 cells , Ambion , Life Technologies , Carlsbad , CA , USA ) , reverse transcriptase-PCR was performed on RNA ( 50 ng ) with a one-step reverse transcriptase-PCR kit ( Qiagen , Hilden , Germany ) using gene-specific primers for IL-12p40 , ARG1 , CD206 , Peroxisome proliferator activated receptor gamma ( PPARG ) , Vitamin D receptor ( VDR ) , 25-Hydroxyvitamin D3 1-alpha-hydroxylase ( CYP27B1 ) , LL-37 ( cathelicidin ) and β-actin ( S1 Table ) . Primers were designed using NCBI gene bank reference sequences of human genes and their specificity for humans were confirmed by Basic Local Alignment Test ( BLAST ) in NCBI . The amplification cycle comprised 35 cycles of denaturing ( 94°C for 30 seconds ) , annealing for 30 seconds ( varying temperature for each primer set; S1 Table ) , extension ( 72°C for 60 seconds ) , and a final extension at 72°C ( 10 minutes ) . Products were resolved on agarose gels ( 2% ) containing ethidium bromide ( 0 . 5 mg/mL ) , observed and analyzed in G-BOX gel doc [Syngene , Cambridge , UK] using Gene Tools ( Version 4 . 01 . 04 ) software . The values were normalized to β-actin , which was considered as 100% for each individual . Immunofluorescent staining was performed on paraffin embedded skin biopsies , mounted on glass slides , deparaffinised in xylene and then rehydrated in graded alcohol . For antigen retrieval , slides were placed in a pre-warmed antigen retrieval solution ( S1699 DAKO citrate buffer pH 6 . 0 , diluted 1:10 , Cambridge , UK ) and incubated in a water bath for 30 minutes at 95°C . The slides were then brought to room temperature ( 20 minutes ) and washed with PBS . After blocking the non specific binding sites with PBS + 5% goat serum for 30 minutes , they were stained overnight at 4°C with anti human CD68 ( Abcam , Cambridge , UK ) , 1: 500 dilution in PBS and/or rabbit anti human arginase-1 ( Protein tech , Manchester , UK ) , 1: 50 in PBS , rabbit anti human CD206 ( Protein tech , Manchester , UK ) , 1:100 in PBS along with appropriate isotype matched controls . After three washings with PBS + 0 . 05% BSA , binding was detected using secondary antibodies , anti mouse Alexa fluor 594 and anti mouse Alexa fluor 488 ( diluted 1:200 in PBS for CD68 Invitrogen , Life Technologies Ltd . , Paisley , UK ) , anti rabbit Alexa fluor 488 ( diluted 1:200 in PBS for arginase-1 , Invitrogen , Life Technologies Ltd . , Paisley , UK ) and anti rabbit Alexa fluor 647 ( diluted 1:200 in PBS for CD206 , Invitrogen , Life Technologies Ltd . , Paisley , UK ) . All incubations were for 1 h at room temperature in the dark and followed by three washings . The slides were then incubated with DAPI ( 1 μg/mL , 200 μL , 10 minutes ) and finally mounted overnight at 4°C with Pro-long Gold anti-fade ( Invitrogen , Life Technologies Ltd . , Paisley , UK ) . The images were captured in an inverted LSM 710 Confocal microscope ( Carl Zeiss Microimaging , Cambridge , UK ) and analyzed via LSM 7500 software and image J software . Plasma levels of 25 ( OH ) Vitamin D3 were measured using a 25 hydroxyvitamin D radio immunoassay kit ( DiaSorin , Stillwater , Minnesota , USA ) , range being 9 . 0–37 . 6 ng/mL and sensitivity was 1 . 5 ng/mL . Data was analyzed between groups by Kruskal-Wallis test followed by Dunn’s multiple comparison tests for non-parametric data for analysis of variance . For non-parametric paired data , Wilcoxon signed rank test was performed using Graph Pad Prism software ( version 5 . 0 ) , p<0 . 05 being significant . All data are expressed as mean ± SEM and horizontal bars in graphs indicate SEM .
Patients with PKDL ( n = 34; Table 1 ) showed a male preponderance , the male: female ratio being 7 . 5:1 . Majority had polymorphic lesions i . e . hypopigmented macules and nodules/papules , while a minority showed exclusively hypopigmented macules . The rk39 test was positive in 33/34 ( the rk39 negative patient was confirmed by ITS-1 PCR ) and in polymorphic lesions , the presence of Leishman Donovan bodies was identified by Giemsa staining . Irrespective of treatment , the assessment of cure was clinical and parasitological ( ITS-1 PCR negative ) . At presentation , their hemoglobin levels and leukocyte counts were comparable with controls i . e . no anemia or pancytopenia which are consistent features of VL . In patients , a maximum of 10 ml blood was provided which yielded 1–2 x 107 cells . As 5 x 105–2 x 106 cells were required per analysis , all markers could not be evaluated in each patient , and patients were randomly selected for individual assays . In PKDL , as compared to healthy controls , the frequency of CD14+ monocytes expressing TLR-2+ was significantly reduced ( 54 . 70 ± 7 . 52% vs . 86 . 32 ± 2 . 78%; p<0 . 01; Fig 2A ) , as was the frequency of TLR-4+ monocytes ( 41 . 26 ± 8 . 90% vs . 75 . 95 ± 3 . 56%; p<0 . 05; Fig 2B ) . Treatment significantly increased TLR-2+ monocytes ( 79 . 67 ± 3 . 23 , p<0 . 05 , Fig 2A ) and became comparable with healthy controls . A similar scenario was demonstrated with TLR-4+ ( 66 . 07 ± 3 . 21; Fig 2B ) . In terms of expression ( GMFC ) , TLR-2 was significantly downregulated in patients vs . healthy controls ( 7 . 30 ± 1 . 56 vs . 29 . 44 ± 4 . 03 ) and reverted post-treatment ( 30 . 52 ± 4 . 88 , p<0 . 05 ) . However , the expression of TLR-4 was comparable during active disease with healthy controls ( 16 . 33 ± 3 . 59 vs . 9 . 01 ± 1 . 24 ) and post treatment ( 10 . 24 ± 0 . 84 ) . At presentation , the ex-vivo levels of NO in monocytes was significantly diminished as compared to controls ( GMFC , 125 . 40 ± 32 . 74 vs . 306 . 30 ± 20 . 78 respectively; p<0 . 01; Fig 2C ) . With treatment , monocytes regained their ability to generate NO ( 371 . 50 ± 76 . 51; p<0 . 01 vs . pre-treatment; Fig 2C ) . Similarly , the generation of ROS was significantly attenuated at presentation vis-a-vis controls ( 858 . 70 ± 171 . 70 vs . 2132 . 00 ± 259 . 90 respectively; p<0 . 01; Fig 2D ) . Treatment increased fluorescence , but remained lower than controls ( 1396 . 00 ± 158 . 20; Fig 2D ) . Alongside , the production of superoxide too was significantly lowered at disease presentation ( 1 . 98 ± 0 . 39 vs . 4 . 40 ± 0 . 12 nM , p<0 . 01 ) but changed minimally with treatment ( 3 . 00 ± 0 . 79 nM ) . Variations in the anti-oxidant status impact on the redox balance and thereby on macrophage host defence functions . The intramonocytic levels of non protein thiols which primarily comprise glutathione were examined in terms of CMF derived fluorescence , wherein increased fluorescence indicated enhanced intracellular levels of non protein thiols [15] . At disease presentation , the fluorescence of CMF was significantly higher than controls ( 2931 . 00 ± 445 . 20 vs . 926 . 00 ± 160 . 60; p<0 . 01; Fig 2E ) . Importantly , it negatively correlated with decreased levels of ROS ( r = -0 . 57 ) , corroborating the presence of a robust anti-inflammatory milieu . Treatment significantly attenuated fluorescence vis a vis active disease ( 1541 . 00 ± 415 . 90 , p<0 . 05 , Fig 2E ) . Monocytes can differentiate into inflammatory or anti-inflammatory subsets , but their classification in relation to functional phenotypes has not been precisely defined . Three subsets of blood monocytes , namely classical ( CD14++CD16− ) , intermediate ( CD14++CD16+ ) , and non-classical ( CD14+CD16++ ) have been described and attributed with discrete functions [17] . At disease presentation , there was a minimal decrease in the proportion of classical ( CD14++CD16- ) monocytes as compared to healthy controls ( 81 . 20 ± 4 . 83% vs . 89 . 60 ± 1 . 20% , S1 Fig ) . There was a nominal increase in the intermediate variant ( CD14++CD16+ ) being 5 . 46 ± 1 . 19% vs . 3 . 46 ± 0 . 79% , S1 Fig ) and the non-classical phenotype ( CD14+CD16++; 14 . 29 ± 3 . 46% vs . 6 . 90 ± 1 . 56% , S1 Fig ) . With treatment , the frequency of classical ( 86 . 20 ± 2 . 19% , S1 Fig ) , intermediate ( 3 . 91 ± 1 . 28% , S1 Fig ) or non-classical monocytes ( 9 . 74 ± 2 . 06% , S1 Fig ) was comparable with healthy controls . As decreased generation of reactive oxygen and nitrogen radicals are suggestive of an alternative activation [4 , 18] , monocytes from PKDL patients were examined for a M2 phenotype . In circulating monocytes from controls and treated patients , the mRNA expression of nuclear receptor PPARG which regulates oxidative metabolism in macrophages [19 , 20] was minimal , but increased ~50-fold during active disease ( Fig 3A ) . The arginase activity , being downstream of PPARγ signaling [4 , 20] was also enhanced in PKDL [13] . It was supported by a 5 . 29 fold increase in mRNA accumulation of ARG1 in circulating monocytes ( p<0 . 01; Fig 3B ) , which with treatment significantly decreased by 2 . 6 fold ( p<0 . 01; Fig 3B ) . In circulating monocytes from controls , mRNA accumulation of mannose receptor ( CD206/MR ) was negligible ( Fig 3C ) , but increased 14 fold at disease presentation , and returned to baseline after treatment ( p<0 . 01 , Fig 3C ) . This translated into an enhanced surface expression of CD206 on monocytes during active disease and decreased with treatment ( p<0 . 05 , Fig 3D ) . Collectively , blood monocytes from PKDL patients showed strong evidence of M2 polarization , which with disease resolution repolarized to a M1 phenotype . In PKDL , a mixed cytokine profile with a Th2 bias has been reported [13 , 21] , but the contribution of monocytes remains unclear . We analyzed the intracellular cytokine producing ability of CD14+ monocytes by gating monocytes initially on their morphology , followed by CD14 expression and then calculated the % of CD14+-cytokine positive cells . At disease presentation , the frequency of CD14+IL-6+ monocytes was significantly attenuated vis a vis controls ( 37 . 31 ± 8 . 67% vs . 92 . 56 ± 3 . 59% , p<0 . 05 , Fig 4 and S2A Fig ) , but increased significantly post-treatment ( 76 . 36 ± 8 . 42% , p<0 . 05 , Fig 4 and S2A Fig ) . Similarly , IL-1β showed a significant 1 . 92 fold decrease vis-a-vis controls ( 44 . 17 ± 11 . 99% vs . 84 . 88 ± 7 . 22% , p<0 . 01 , Fig 4 and S2B Fig ) and reverted with treatment ( 82 . 90 ± 7 . 29% , p<0 . 05 , Fig 4 and S2B Fig ) . Like IL-6 and IL-1β , the frequency of CD14+IL-8+ monocytes was decreased at presentation ( 52 . 31 ± 9 . 13% vs . 84 . 02 ± 2 . 68% , p<0 . 05 , Fig 4 and S2C Fig ) , but was restored post-treatment ( 82 . 02 ± 3 . 64% , p<0 . 05 , Fig 4 and S2C Fig ) . Contrary to expectation , the frequency of IL-12p40 significantly increased in PKDL vis a vis controls ( 2 . 37 ± 0 . 44% vs . 0 . 55 ± 0 . 11% , p<0 . 05 , Fig 4 and S2D Fig ) . It remained elevated post-treatment ( 2 . 46 ± 0 . 38% , p<0 . 01 , Fig 4 and S2D Fig ) , and was substantiated at the mRNA level ( S3 Fig ) . The intramonocytic levels of TNF remained unchanged during active disease ( 2 . 49 ± 0 . 59% vs . 1 . 13 ± 0 . 36% ) and with treatment ( 1 . 56 ± 0 . 40% ) . A key anti-inflammatory cytokine secreted by M2 monocytes is TGF-β , a homodimer comprising LAP and TGF-β1 . During cellular activation , cleavage of TGF-β1 from LAP results in TGF-β1 secretion . Therefore , the intramonocytic LAP-TGF-β1 complex indirectly reflects the status of TGF-β1 [22] . In PKDL , the frequency of LAP-TGF-β1 vs . controls was significantly reduced ( 4 . 26 ± 1 . 27% vs . 33 . 66 ± 10 . 44% , p<0 . 05 Fig 4 and S2E Fig ) , indicative of raised functional levels of TGF-β1 , which increased with treatment ( 31 . 06 ± 8 . 65% , p<0 . 05 , Fig 4 and S2E Fig ) . To confirm that this decrease in LAP-TGF-β1 was not attributable to decreased synthesis , mRNA expression of the LAP-TGF-β1 complex was measured . A 10 fold increase was evident at presentation ( 31 . 05 ± 5 . 74 vs . 3 . 00 ± 0 . 44 , p<0 . 05 , S3 Fig ) which decreased post-treatment ( 16 . 36 ± 4 . 84 , S3 Fig ) . As M2 polarization requires a milieu comprising IL-4 , IL-10 and IL-13 [19] , levels of these cytokines were estimated in patients with PKDL . The levels of IL-4 were significantly raised as compared to healthy individuals ( 126 . 80 ± 12 . 57 vs . 61 . 35 ± 6 . 36 pg/ml , p<0 . 05 ) , as was IL-10 ( 37 . 06 ± 4 . 14 vs . 12 . 68 ± 2 . 05 pg/ml , p<0 . 01 ) and IL-13 ( 184 . 10 ± 45 . 22 vs . 7 . 77 ± 1 . 38 pg/ml , p<0 . 001 ) , thus corroborating with previous reports [13] . Treatment caused a significant decrease in IL-10 ( 21 . 33 ± 3 . 66 pg/ml , p<0 . 01 ) , reiterating its importance in leishmaniasis . In patients with PKDL , an increased proportion of CD68+ macrophages was reported at the lesional sites , that regressed with treatment [23] , and was corroborated in this study . An increased accumulation of PPARγ mRNA as compared to controls and post-treatment ( Fig 5A ) was accompanied by a significant increase in the mRNA of ARG1 ( Fig 5B ) . Confocal immunofluorescence confirmed localisation of arginase-1 within CD68+ macrophages . Post-treatment , the decrease in CD68+ was associated with a concomitant decrease in arginase-1 ( Fig 5C ) . The lesional 13 . 9 fold increase in CD206 mRNA as compared to controls and post-treatment reinforced the M2 polarized status ( 13 . 86 ± 0 . 92; p<0 . 01 and p<0 . 001 respectively; Fig 5D ) . In addition , this was mirrored by raised protein expression evident via confocal microscopy ( Fig 5E ) . With treatment , the decreased proportion of CD68+ macrophages resulted in a decreased expression of CD206 ( Fig 5E ) . An expression of CD206 was also identified on CD68- cells whose number also decreased with treatment , but their identity remains to be established . H&E staining confirmed the absence of polymorphonuclear cells ( PMNs ) . Collectively , in PKDL , the M2 polarization that was evident in peripheral blood is also a consistent feature of monocytes/macrophages within dermal lesions . As Vitamin D receptor signaling has been linked to M2 polarization and generation of antimicrobial peptides [24] , this pathway was examined as it may underlie the systemic and local M2 polarization in PKDL . Plasma 1α , 25-dihydroxyvitamin D3 ( 1 , 25D3 ) was significantly raised during PKDL , compared to controls and post-treatment ( 20 . 50 ± 3 . 32 vs . 8 . 05 ± 2 . 99 vs . 13 . 68 ± 2 . 92 ng/mL respectively , Fig 6A ) . The bioactive 1 , 25D3 is generated from its inactive prohormone by vitamin D-1α-hydroxylase , encoded by CYP27B1 . In keeping with the elevated plasma levels of 1 , 25D3 , an increase in CYP27B1 mRNA accumulation was evident ( Fig 6B and 6C ) . VDR is responsible for nuclear signaling of 1 , 25D3 , and its accumulation in patient monocytes and skin biopsies was accompanied with an 11-fold increase in the downstream antimicrobial effector peptide cathelecidin ( hCAP18/LL-37 , Fig 6B and 6C ) . Treatment reduced most components of the Vitamin D signalling pathway , though levels did not always return to levels comparable with controls ( Fig 6B and 6C ) .
Papulo-nodules in PKDL being parasite-rich have fuelled speculation of its pivotal role in the transmission of VL especially in South Asia , where VL is anthroponotic making patients with PKDL the strongest contenders to be the disease reservoir . Accordingly , its eradication should be an essential component of the VL elimination programme [http://www . who . int/tdr/publications/documents/kala_azar_indicators . pdf; last accessed on 1st March 2015] , emphasising the importance of establishing a greater understanding of the immunopathogenesis of PKDL . In the absence of an experimental model , the challenge of delineating the immunoclinical determinants of PKDL lies squarely on the shoulders of clinical researchers which constituted the focus of this study ( Fig 1 ) . In experimental and human VL , attenuation of the oxidative burst , secondary to reduced phosphorylation of MAPKs occurred through the TLR-2 pathway [25 , 26] or the CD40 signalosome [27] . Similarly in PKDL , the decreased expression of TLR-2/4 might also translate into impaired MAPK signalling , resulting in the intramonocytic redox imbalance tilting towards an anti-inflammatory milieu ( Fig 2 ) . Importantly , with chemotherapy , the generation of NO , but not ROS was significantly increased ( Fig 2 ) . This was concordant with studies in VL wherein NO played a pivotal role and was associated with TLR-4 [26 , 15] . However , in patients with CL ( L . braziliensis and/or L . guyanenesis ) , NO played an insignificant role , ROS being more important [28] , suggesting that alterations in the redox status are attributable to parasite species and disease manifestations . Taken together , monocyte-macrophage dysfunction is a hallmark of PKDL and is facilitated by the immunosuppressive microenvironment of IL-4 , IL-13 and IL-10 [13] , which are essential stimuli for driving monocytes towards innate deactivation or alternate activation . In mouse monocytes/macrophages , the intricate network of signalling molecules , associated transcription factors along with post transcriptional regulators mediating the different forms of activation are well delineated [29] . IL-4 and IL-13 via STAT6 activation are known to skew the macrophage function towards the M2 phenotype leading to transcription of genes typical of M2 polarization , notably Mannose receptor ( Mrc1 ) , Arginase ( Arg1 ) , PPARγ ( PPARG ) and Fizz1 among others [12] . However , efforts to simulate the murine scenario by pulsing human monocyte derived macrophages with IL-4 or IL-13 failed to demonstrate arginase-1 expression , based on which conclusions were inferred that Arg1 is strictly restricted to mouse M2 cells [11] . However , this ex-vivo situation may well be a cytokine induced artefact , as in vivo , human monocytes require multiple signals to establish the entire spectrum of alternative activation [30] . Indeed , in patients with filariasis , an increased expression of arginase-1 in M2 monocytes was demonstrated [31] . This was mirrored in our study and endorsed our proposition that in parasitic diseases , an enhanced expression of arginase-1 is a feature of M2 polarized human monocytes/macrophages ( Fig 3B ) . Studies in human VL and CL have also demonstrated increased arginase activity sourced from low density granulocytes [8 , 9] . Importantly , this increased presence of arginase supports parasite survival by increasing the availability of polyamines [10 , 32] . Additionally , the decreased availability of the microbicidal nitric oxide would also support disease progression . In a hamster model of VL , L . donovani directly induced STAT-6 phosphorylation and increased arginase-1 leading to disease progression [33 , 34] . The process was augmented by exogenous IL-4 and other factors like insulin like growth factor 1 ( IGF-1 ) and fibroblast growth factor [FGF , 33 , 34] . Furthermore , the raised plasma arginase [13] translated into depletion of L-arginine and led to an impairment of T cell functions [32] . Taken together , induction of arginase-1 drives disease progression and its blockade might support disease resolution . Although during PKDL , monocytes/macrophages parasites reside primarily in dermal lesions , changes were observed in systemic monocytes ( Figs 3 and 4 ) suggesting that M2 polarization is not a direct consequence of intracellular subversion strategies employed by the parasite [35] . However , we cannot formally rule out the possibility that parasite-derived immune modulators e . g . in exosomes [36] mediated this effect . As plasma IL-4 , IL-10 , IL-13 and TGF-β was elevated in PKDL [13] , it is more likely that polarization occured following bystander responses to inflammation and immune activation [37] . As monocytes are yet to be segregated into M1 and M2 , it would be prudent to endorse the increased expression of PPARγ , Arg1 and CD206 ( Fig 3A–3D ) by complimentary studies wherein biomarker expression of infected monocytes should be compared with uninfected monocytes after being pulsed with IL-4 and/or IL-13 . A key factor for development of the M2 phenotype is activation of PPARγ , a transcription factor of the nuclear hormone receptor family that acts downstream of STAT6 signalling to regulate macrophage metabolism . In experimental models of leishmaniasis [38] , upregulation of PPARγ by IL-4 was demonstrated to instill monocytes/macrophages with potent anti-inflammatory and Th2 functionalities [38] , essential for disease progression . PPARγ through its transrepressive action blocks expression of iNOS and NF-κB mediated transcription of pro-inflammatory mediators [39] . In addition , cytosolic PPARγ by interfering with activation of PKC-α can suppress NADPH oxidase and impair generation of superoxide and NO . Furthermore , as PPARγ is responsible for the enhanced expression of CD206-mannose receptor [40] , their increased expression at disease presentation in lesional macrophages and circulating monocytes , provided strong endorsement of M2 monocyte/macrophage polarization ( Figs 3 and 5 ) . The presence of CD68-CD206+ cells in dermal lesions could be inflammatory dermal dendritic cells , which although absent in normal human skin have been demonstrated in the epidermis of patients with psoriasis and atopic dermatitis [41 , 42] . Ideally , to specifically address the relevance of CD68+ macrophages in disease outcome , the features of M2 polarization should be demonstrated in isolated CD68+ macrophages by comparing arginase expression in macrophages vs . arginase expressed by other cells e . g . low density neutrophils . However , owing to limited availability of clinical material , this was not feasible and importantly , PMNs were absent in H&E stained sections at disease presentation and post treatment . The M1 monocytes/macrophages produce primarily pro-inflammatory cytokines ( IL-6 , IL-8 , IL-1β , TNF-α and IL-12 ) , whereas M2 monocytes sustain their immunoregulatory and immunosuppressive phenotype via IL-10 and TGF-β [4] . In agreement , monocytes in PKDL patients generated lower amounts of IL-6 , IL-1β and IL-8 ( Fig 4 and S2A–S2C Fig ) . Interestingly , a small but significant population of monocytes expressed the pro-inflammatory IL-12p40 , and their frequency remained high post treatment ( Fig 4 and S2D Fig ) , similar to a study by Gupta et al [43] . In PKDL , IL-10 and TGF-β , signature cytokines of M2 monocytes play a vital role in disease progression [44] . This was confirmed by our study wherein circulating monocytes were established to be a rich source of TGF-β ( Fig 4E ) . With treatment , the decrease of TGF-β in monocytes strengthened the notion that in PKDL , monocytes are alternatively activated , with treatment repolarizing them towards a M1 phenotype . As lesions in PKDL tend to mirror clothing habits and consistently appear in sun exposed areas [1 , 2] , the contribution of Vitamin D3 , whose synthesis increases following UV or sunlight exposure was explored . VitD3 is a potent immunosuppressant of macrophages/monocytes that downregulates TLR-2/4 and monocyte co-stimulatory molecules [45] . Additionally , VitD3 inhibits production of intramonocytic pro-inflammatory cytokines by modulating the MAPK phosphatase-1 [46] and induces M2 polarization [47] . In Behcet’s disease , a chronic inflammatory disorder , TLR 2/4 expression negatively correlated with vitamin D3 and importantly , the dose dependent treatment of vitamin D3 decreased inflammation , as also decreased the expression of TLR-2/4 [48] . Our data in PKDL suggests a similar scenario as decreased expression of TLR 2/4 was concomitant with increased vitamin D3 ( Figs 2 and 6 ) . This was in concordance with a L . major model wherein VDR knockout mice showed resistance to infection . Alongside , addition of 1 , 25 ( OH ) 2D3 to L . major-infected macrophages translated into induction of arginase-1 , down regulation of iNOS and parasite persistence [49] . Similarly , in PKDL , raised serum 25 ( OH ) D3 was accompanied by an enhanced mRNA expression of CYP27B1 , VDR and LL-37 , indicating that infection upregulated the molecular switch needed for monocyte polarization towards a M2 phenotype ( Fig 6 ) . Differential polarization of macrophages in diverse disease conditions confirms the plasticity of macrophages , with M1 polarization evident in inflammatory and autoimmune diseases such as diabetes , atherosclerosis and sepsis , while a strong M2 or M2-like polarization has been proposed in cancers , chronic parasitic , viral or bacterial diseases [4 , 19] . Conversely , the inability to switch to an M2 phenotype may underlie the failure to resolve inflammation e . g . chronic venous ulcers [50] . Like in tumors , M2 polarized macrophages and dendritic cells , have been proposed to contribute towards subversion of adaptive immunity thus promoting tumor growth and progression [51] , it may be envisaged that the Leishmania parasite ensures its survival by creating an immunosuppressive milieu via M2 polarization of macrophages and a decrease in dendritic cells [23] , thus collectively causing impairment of antigen presentation . Ideally , studies confirming the functional phenotype are recommended and best addressed in an ex vivo assay , the limiting factor being the availability of lesional material . Accordingly , reorientation of these polarized macrophages is now an integral component of macrophage targeted therapy [19] . It can be envisaged that in diseases like leishmaniasis , the strategy of reshaping and reorientation of macrophage polarization could be a promising therapeutic modality worthy of future consideration .
|
Monocyte/macrophage subsets following their polarization by the microenvironement serve as important immune sentinels that play a vital role in host defense and homeostasis . The polarization of macrophage function has been broadly classified as M1 ( classical ) and M2 ( alternate ) activation , wherein M1 polarised cells display a strong pro-inflammatory microbicidal response , while M2 polarization is linked to production of an anti-inflammatory milieu leading to tissue regeneration and wound healing . Data pertaining to macrophage polarization are primarily derived from murine models , but increasing evidence is highlighting the inadequacy of direct inter-species translation . In leishmaniasis , a protozoan infection caused by the genus Leishmania , manipulation of host macrophage function is central to pathogenesis . In this study we report that monocyte/macrophage subsets in Post kala-azar dermal leishmaniasis are polarized to an M2 phenotype . This study provides insights into systemic and local regulation of macrophage/ monocyte functions in this important human disease and highlights the influence of immunomodulatory anti-leishmanial chemotherapy on macrophage/monocyte polarization .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[] |
2015
|
M2 Polarization of Monocytes-Macrophages Is a Hallmark of Indian Post Kala-Azar Dermal Leishmaniasis
|
Necrotrophic plant pathogens acquire nutrients from dead plant cells , which requires the disintegration of the plant cell wall and tissue structures by the pathogen . Infected plants lose tissue integrity and functional immunity as a result , exposing the nutrient rich , decayed tissues to the environment . One challenge for the necrotrophs to successfully cause secondary infection ( infection spread from an initially infected plant to the nearby uninfected plants ) is to effectively utilize nutrients released from hosts towards building up a large population before other saprophytes come . In this study , we observed that the necrotrophic pathogen Dickeya dadantii exhibited heterogeneity in bacterial cell length in an isogenic population during infection of potato tuber . While some cells were regular rod-shape ( <10μm ) , the rest elongated into filamentous cells ( >10μm ) . Short cells tended to occur at the interface of healthy and diseased tissues , during the early stage of infection when active attacking and killing is occurring , while filamentous cells tended to form at a later stage of infection . Short cells expressed all necessary virulence factors and motility , whereas filamentous cells did not engage in virulence , were non-mobile and more sensitive to environmental stress . However , compared to the short cells , the filamentous cells displayed upregulated metabolic genes and increased growth , which may benefit the pathogens to build up a large population necessary for the secondary infection . The segregation of the two subpopulations was dependent on differential production of the alarmone guanosine tetraphosphate ( ppGpp ) . When exposed to fresh tuber tissues or freestanding water , filamentous cells quickly transformed to short virulent cells . The pathogen adaptation of cell length heterogeneity identified in this study presents a model for how some necrotrophs balance virulence and vegetative growth to maximize fitness during infection .
Based on the modes of nutrition acquisition , plant pathogens can be categorized into biotrophs , necrotrophs , and hemibiotrophs [1 , 2] . Obligate biotrophs acquire nutrients from living plant cells , and thus have to maintain host viability . Consequently , biotrophs generally do not produce lytic enzymes and toxins , but rely on sophisticated immune suppression systems to escape host surveillance . Obligate necrotrophs , in contrast , feed on nutrients released from dead or dying cells , and thus have to actively kill host cells and break down host tissues through the production of plant cell wall degrading enzymes ( PCWDEs ) and toxins . Many pathogens display both biotrophic and necrotrophic phases of nutrient acquisition and are categorized as hemibiotrophs . Dickeya dadantii is the causal agent of bacterial soft rot disease . Although D . dadantii can survive without producing symptoms during a long time in latent infections , it has a typical necrotrophic phase during the active infection of plants . Infections caused by D . dadanntii mostly occur on nutrient-rich plant organs , such as tubers , rhizomes , bulbs , and succulent stems and leaves [3] . During infection , D . dadantii secretes a series of PCWDEs , mainly pectate lyases , to disintegrate the plant cell walls . The type III secretion system ( T3SS ) , a secretion apparatus that injects protein effectors into the host cell , is another necessary virulence factor to elicit cell death in D . dadantii [4 , 5] . Pathogens consume the nutrients released from dead plant cells , and spread to adjoining tubers to incite secondary infections as liquid from the rotting tubers percolates onto others [6] . Under favorable conditions , complete decay of a potato tuber or storage roots can occur within two to three days [3] . One practical challenge for D . dadantii and many other obligate necrotrophs is how to efficiently consume the large amount of nutrients released from the nutrient-rich organs and use them towards building up the pathogen populations for secondary infection [2 , 6] . Efficient nutrient consumption and energy conversion seem to be more important for necrotrophs than for biotrophs and hemibiotrophs , as breaking down of the host epidermis and tissue structures results in the immediate release of nutrients that are available to essentially any microbes in the surrounding environment . Vegetative growth and virulence expression are two distinct physiological states for bacterial pathogens [7 , 8] . Virulence is induced under nutrient limited or stress conditions , whereas vegetative growth occurs under nutrient rich , growth favorable conditions [9 , 10] . For example , genes encoding the T3SS are induced when bacteria are cultured under a nutrient limited hrp-inducing minimal medium condition but were repressed in nutrient rich media such as Lysogeny Broth [9] . In addition to their different environmental cues , a clear tradeoff has also been observed between virulence and growth in bacteria , as the production of virulence factors compromised the rate of metabolic processes associated with growth [7 , 11 , 12] . Pathogens have developed complex regulatory systems to sense environmental signals and shift between the two lifestyles upon environmental changes [13] . However , the co-occurrence of both lifestyles in a homogeneous population has not been clearly documented and well understood . Many bacterial pathogens have been observed to exhibit variable phenotypes in an isogenic population in the host . Heterogeneity is defined as a diversity of phenotypes displayed by isogenic bacteria in similar environments that allow new functionality . Phenotypic heterogeneity increases the flexibility and versatility of bacteria as a population . Phenotypic heterogeneity has been observed in many bacteria in various functions such as antibiotic persistence [14] , virulence [10 , 15–18] , biofilm formation [19] , motility [20] , and sporulation [21 , 22] . In this study , we report that the soft rot pathogen D . dadantii displays heterogeneity in cell length during infection on potato tubers . Within a clonal population , a subpopulation of D . dadantii cells formed elongated filamentous cells while the rest of the cells remained short . We show that the proportion of filamentous cells and short cells was affected by infection organs , environmental conditions , stages of infection , and presence of freestanding water . We further demonstrated that filamentous cells and short cells have distinct properties: while short cells were virulent , motile , and relatively tolerant to stress , the filamentous cells were non-virulent , non-motile , and highly sensitive to stress . However , filamentous cells display upregulated metabolic genes and increased growth compared to the short cells . When environmental conditions change , filamentous cells began reverting to short cells . D . dadantii differentiation into heterogeneous cell types was dependent on the differential expression of the alarmone ppGpp . These findings provide insight into the role of phenotypic heterogeneity in the overall fitness of a bacterial population during host-pathogen interactions .
In our preliminary study of gene expression in single cells of D . dadantii , we noticed that some D . dadantii cells presented an elongated cell morphology when inoculated on potato tubers . To determine if D . dadantii displayed heterogeneity in cell morphology during the infection of potato tuber , tubers inoculated with a clonal culture of wild type D . dadantii strain 3937 at 48 hrs post inoculation ( hpi ) were examined under scanning electron microscopy . A heterogeneity in cell length was indeed observed within the isogenic population of D . dadantii: while some cells were rod-shaped with length between 1 to 10 μm , other cells were highly elongated to lengths greater than 10 μm ( Fig 1 ) . The filamentous cells were plated and re-identified to rule out contamination of other non-Dickeya species . Hereafter , we define cells shorter than 10 μm as short cells , and cells longer than 10 μm as filamentous cells . Ratio of filamentous cells to short cells range between 1:0 . 91 to 1:3 . 83 at 48 hpi . Next , we determined whether the formation of filamentous cells observed in potato tubers also occurs in other organ types ( leaves and stems ) . To visualize bacterial cells in vivo , we introduced a green fluorescence protein ( gfp ) reporter constitutively expressed under the control of an nptII promoter , into the wild type D . dadantii . We confirmed that expression of gfp did not affect cell length ( S1 Fig ) and used this gfp-expressing strain in the characterization hereafter . No obvious cell length differentiation was observed in either nutrient-limited organs such as leaves or stems ( filamentous cell percentage <0 . 3% of total cells ) inoculated with D . dadantii at 24 and 48 hpi , as observed in the nutrient rich organ potato tubers ( filamentous cell percentage >17 . 8% of total cells , Fig 2A ) . We next determined whether the filamentous cell formation could be affected by bacterial position on a potato tuber . Cell lengths of D . dadantii at the surface edge of infection , at the surface center of infection , and at the internal edge of infection of infected potatoes ( P1 , P2 , and P3 respectively , as indicated in Fig 2B ) were quantified at 48 hpi . A significantly higher ratio of filamentous to short cells was observed at the surface center position ( P2 , 1:1 . 33 ) , where disease was well established , compared to that observed at the interfaces of healthy and diseased tissues ( P1 and P3 , 1:399 and 1:37 . 5 ) ( Fig 2B ) . Echoing this observation , we also observed a steady increase in the proportion of filamentous cells from an early time point after inoculation ( 0% of the total population at 6 hpi ) to a later time point ( 42 . 9% at 48 hpi; Fig 3 , right panel ) . Water plays an important role in soft rot disease initiation [6] . We observed that the formation of filamentous cells was repressed by the presence of freestanding water on top of the inoculated tubers , with only 3 . 0% of total cells filamented at 48 hpi , compared with the aforementioned 42 . 9% of cells filamented without the presence of water ( Fig 3 ) . Taken together , these results demonstrate that D . dadantii forms filamentous cells on potato tubers , at later stages of infection , and at well-established infection positions on the surface ( P2 ) . In contrast , the formation of filamentous cells was repressed on leafy tissues , at early infection time points , at disease interfaces ( P1 and P3 ) , and on water-covered tissues . The observation that most filamentous cells were formed only after infection became well-established i . e . , after host cell lysis was underway , suggested that this subpopulation may not need virulence capabilities . We first determined the level of expression of genes encoding the T3SS , a critical virulence factor in D . dadantii . We introduced a dual fluorescence promoter reporter PnptII-gfp-PhrpN-mCherry into the wild type D . dadantii and inoculated it onto potato tuber . The reporter contains two promoter-fluorescence gene fusions , PnptII-gfp and PhrpN-mCherry , which enables visualization of the total population of D . dadantii ( both filamentous cells and short cells ) in green fluorescence , while monitoring the expression of hrpN in red fluorescence in these two subpopulations . At 48 hpi , a clear differentiation of filamentous cells and short cells was observed at surface center position P2 in the absence of freestanding water ( Fig 4A , left panel ) . Interestingly , a strong negative correlation was observed between cell length and the expression of hrpN at the P2 position ( Fig 4A , correlation co-efficiency R2 = 0 . 9789 ) . A total of 88 . 2% of the hrpN-expressing cells had a cell length of less than 10 μm and 11 . 8% of the hrpN expressing cells had a cell length between 10 μm and 20 μm . No hrpN expression was detected in cells with lengths greater than 20 μm ( Fig 4A indicated by arrows , and 4B ) . At the P1 position , only short cells were observed , and 41 . 7% of them expressed hrpN ( Fig 4A and 4B ) . The expression of a non-T3SS gene , rsmB , was evenly expressed in both filamentous and short cells ( Fig 4A ) . As we determined that the T3SS genes were not expressed in the filamentous cells , and the proportion of filamentous cells increased as the infection progressed from early to later stage at P2 position ( Fig 3 without freestanding water ) , next , we determined whether the increase of the filamentous cell subpopulation would result in an overall reduction of the T3SS expression at P2 position . As shown in Fig 4C , filamentous cell subpopulation increased from 4 . 0% at 18 hpi to 18 . 6% at 48 hpi at P2 position . At the same time , a decrease in T3SS expression was also indeed observed ( from 33 . 0% of total cells at 18 hpi to 6 . 8% of the total cells at 48 hpi , Fig 4C ) . No increase in filamentous cell proportion nor decrease of the T3SS expressing cells in the total population was observed at the disease interface P1 ( Fig 4B ) . To further confirm the differential expression of T3SS genes in filamentous cells and short cells , we developed a sucrose-gradient centrifugation method to separate enriched pools of filamentous cells and short cells from position P2 of the infected potato tubers ( S2 Fig ) . Expression of two T3SS marker genes , hrpN and hrpA , was analyzed in the two subpopulations by qRT-PCR . As shown in Fig 5A , greater than 75% reduction of the expression of both genes was observed in filamentous cells compared to short cells , further supporting the hypothesis that filamentous cells express T3SS genes at a reduced level . We next compared the mRNA levels of another virulence factor , the pectate lyase gene pelD , in the two subpopulations . Compared to the short cells , the mRNA of pelD is more than fivefold reduced in the filamentous cells . No significant difference in the expression of gyrA , a gene encoding DNA gyrase , was detected between the two subpopulations ( Fig 5A ) . Together , our results suggest that filamentous cells , mostly formed when disease is well-established , may not participate in virulence and host invasion . Previous studies have suggested a role of alarmone ppGpp in determining bacterial cell length; mutation of ppGpp biosynthesis genes resulted in elongated cell morphology in Erwinia and Pseudomonas plant pathogens [23 , 24] . However , it is unclear whether the formation of naturally occurring filamentous cells could be caused by the differential expression of the ppGpp biosynthesis genes . To determine the role of ppGpp in filamentous cell formation in D . dadantii , we measured the mRNA levels of two putative ppGpp biosynthesis genes , relA and spoT , in the separated short cells and enriched filamentous cells collected from infected potato tubers . Expression of both genes was significantly reduced in the filamentous cells compared to the short cells ( Fig 5A ) . To validate this differential expression , a Western blot was performed to quantify the RelA and SpoT proteins in the two subpopulations . Similar to the qRT-PCR results , higher levels of RelA and SpoT proteins were detected in the short cells compared to that in the filamentous cells ( Fig 5B ) . The above results conclude that the two putative ppGpp biosynthesis genes are differentially expressed in filamentous cells and short cells of D . dadantii . The reduction in relA and spoT expression in filamentous cells led us to hypothesize that relA and spoT-mediated ppGpp production is essential for D . dadantii filamentation on potato tubers . First , we confirmed whether relA and spoT are required for ppGpp biosynthesis and degradation in D . dadantii as in other bacteria . Deletion mutants of relA and spoT were generated and intracellular ppGpp levels were measured . Compared to the wild type , decreased and increased ppGpp levels were observed in ΔrelA and ΔspoT respectively , and could be restored through complementation ( Table 1 ) . In the ΔrelAΔspoT double mutant , drastic reduction of ppGpp levels were observed ( Table 1 ) . These observations suggest that as in model species of bacteria , D . dadantii RelA is responsible for ppGpp synthesis , while SpoT has a major function of ppGpp hydrolysis as well as a minor function of ppGpp synthesis . We observed that filamentous cells formed under natural infection conditions exhibited reduced expression of both relA and spoT . To determine whether these genes are necessary for the filamentous cell formation in D . dadantii , we compared the cell length of wild type , ΔrelAΔspoT , single deletion mutants and complementation strains in culture ( LB broth ) and on potato tuber . Cell length was significantly increased when both ppGpp biosynthesis genes were mutated ( ΔrelAΔspoT in Fig 6A ) . We also observed a strong negative correlation between a strain’s intracellular ppGpp levels ( Table 1 ) and cell length; cell length was increased in deletion mutants with reduced intracellular ppGpp levels ( ΔrelA and ΔrelAΔspoT , Fig 6A ) and was decreased in a deletion mutant with increased ppGpp level ( ΔspoT , Fig 6A ) . On potato tuber , a similar correlation was also observed ( Fig 6B ) , although the cell length of ΔrelAΔspoT could not be determined , as the double mutant was non-virulent . These observations further confirmed that intracellular ppGpp strongly affects bacterial cell length , and abolishing ppGpp synthesis through double mutation of relA and spoT results in filamentous cell formation similar to that observed in potato tuber . To confirm that ppGpp itself and not another function of RelA or SpoT is a signal for cell elongation , we next tested the effect of adding chemically synthesized ppGpp on filamentous cell formation . We hypothesized that if ppGpp causes elongation and has limited permeability [25] , adding chemically synthesized exogenous ppGpp at a concentration much higher than the intracellular ppGpp concentration to D . dadantii would repress filamentous cell formation . To test this hypothesis , ppGpp at two different concentrations was added on top of the potato tubers inoculated with wild-type D . dadantii . A significant reduction in the proportion of filamentous cells was observed upon addition of ppGpp ( S3 Fig ) , which further supported the hypothesis that low ppGpp triggers filamentous cell formation during D . dadantii infection . Having confirmed that cell length heterogeneity is dependent on the differential expression of ppGpp biosynthesis genes , and that filamentous cells can be artificially induced through deletion of both relA and spoT , we used the double deletion ΔrelAΔspoT mutant to uncover putative biological functions of the naturally occurring filamentous cells . Naturally occurring filamentous cells exhibited significantly reduced expression of the T3SS and pectate lyase genes , which indicates that these cells may not participate in pathogenicity during the host-microbe interaction . To test this hypothesis , we compared the bacterial virulence of wild type and ΔrelAΔspoT , along with single mutants and complementation strains , on potato tubers . Compared to the wild type which caused necrotic lesions on potato tuber at 18 hpi , ΔrelAΔspoT lost its virulence capability ( Fig 7A ) . To determine whether the reduction in T3SS and pectate lyase gene expression in the naturally occurring filamentous cells would also result in reduced activities of T3SS and of pectate lyases , we compared the hypersensitive response ( HR ) and pectate lyase production in wild type D . dadantii and the ΔrelAΔspoT double mutant , as well as single mutants and complementation strains . Compared to the wild type , ΔrelAΔspoT lost its ability to elicit the HR on tobacco leaves ( Fig 7B ) . Consistent with phenotypic observations , mRNA levels of the T3SS genes hrpA , hrpN , dspE , and hrpL were also significantly reduced in ΔrelAΔspoT as observed in naturally occurring filamentous cells under the in vitro hrp-inducing condition ( Fig 7B ) . A significant reduction in pectate lyase production and in expression of pectate lyase-encoding genes pelD and pelE was also observed in ΔrelAΔspoT compared with the wild type under in vitro Pel producing conditions ( Fig 7C ) . Taken together with the results from Fig 4 , these results suggest that whether environmentally or mutationally induced , filamentous cells of D . dadantii do not produce two essential virulence factors and do not contribute to pathogenicity . Motility is essential for D . dadantii to spread and cause secondary infections , especially under wet conditions [26] . Our observation that D . dadantii filamentous cell formation only occurred in the absence of freestanding water suggests that filamentous cells may be non-motile . To test this hypothesis , a mixture of filamentous and short cells collected at the P2 position of potato tubers at 48 hpi were resuspended in sterile distilled water and observed under the microscope for their motility ( S1 Video ) . As expected , filamentous cells showed significantly reduced motility ( 0 . 723 ± 0 . 6 microns per second ) compared to short cells ( 5 . 508 ± 2 . 621 microns per second ) . We also compared the swimming and swarming motility of wild type D . dadantii with deletion mutants of the ppGpp biosynthesis genes . A complete abolishment of both swimming and swarming motility was observed in ΔrelAΔspoT ( Fig 8A and S4 Fig ) . These results demonstrate that filamentous cells are non-motile . It also indicates that ppGpp biosynthesis is required for bacterial motility in D . dadantii . Previous studies documented that some bacteria form filamentous cells as a response to stress e . i . antibiotic persistence or host immune response [27] . To determine whether filamentous cells of D . dadantii serve the same purposes , we compared the survival of double and single deletion mutants of ppGpp biosynthesis genes with the wild type upon two stress conditions hydrogen peroxide ( H2O2 ) and antibiotic ampicillin treatment . Compared to the wild type , ΔrelAΔspoT displayed increased susceptibility to sub-lethal dosage of both hydrogen peroxide and ampicillin ( Fig 8B ) . These results suggest that , in contrast to previous examples , mutationally induced filamentous cells of D . dadantii not only do not grant the bacteria any advantage , but rather compromised their ability , in tolerating H2O2 and antibiotic ampicillin . The observation that filamentous cells are more sensitive to oxidative stress is also in agreement with our observation that filamentous cells are more abundantly present at later stage of infection , when the host cells are well decayed and host defense is significantly compromised . Despite finding that filamentous cells do not participate in virulence , are non-motile , and more sensitive to stress , we sought clues as to the fitness advantages of these filamentous cells during the host-microbe interactions . To further elucidate the biological functions , a transcriptomic analysis was performed in the separated short cells and enriched filamentous cells collected from potato tubers at 48 hpi ( S2 Fig ) . Compared to the short cells , 27 and 38 genes were found to be down-regulated and up-regulated in the filamentous cells , respectively ( Fig 9 and Table 2 ) . Among the down-regulated genes ( p < 0 . 05 ) , genes encoding the T3SS ( hrpA ) , motility ( motB ) and stress tolerance ( RS18760 and dprA ) were identified . Down regulation of ppGpp biosynthesis genes relA ( relative mRNA expression filamentous cells/short cells = 0 . 43 ) and spoT ( 0 . 85 ) , as well as a few pectate lyase and T3SS genes ( pelE , hrpY and hrcV; relative mRNA expression filamentous cells/short cells = 0 . 23 , 0 . 46 , and 0 . 09 respectively ) were also identified , although the difference was not statistically significant ( p >0 . 05 ) , possibly due to the fact that the enriched filamentous cell samples still contained mixed-in short cells ( S2 Fig ) . Among the up-regulated genes , surprisingly , over 60% ( 23/38 ) of them are related to bacterial growth . Specific function categories included substrate transport ( 5 genes ) , respiration ( 8 genes ) , and metabolism ( 10 genes ) ( Fig 9 and Table 2 ) . To determine if the higher rate of metabolism in filamentous cells would result in faster growth on potato tuber , we compared the number of progenies produced by the wild type , which formed both filamentous cells and short cells , and by ΔspoT , which only formed short cells ( Fig 6A ) . Forty-eight hours after inoculation , 2 . 9×109 CFU of wild type were recovered per 0 . 03 gram of potato tissue after incubation in water to induce the short cells ( Fig 10 , 48 hpi ) . Although ΔspoT showed no attenuation in virulence compared to the wild type , it only produced 1 . 7×108 CFU of cells from the same amount of tissue within the same period of time , or approximately 6% of the wild type ( Fig 10 , 48 hpi ) . This reduction in growth in ΔspoT could be partially restored to the wild-type level through complementation ( 7 . 8×108 CFU ) . Interestingly , such growth difference between wild type and ΔspoT was not observed at earlier stage of infection when filamentous cells were not formed ( Fig 10 , 14 hpi ) . These data suggest that filamentous cells formed at later stage of infection granted D . dadantii faster growth on potato tuber . Filamentous cells contain large amount of biomass and may have the potential to split into short cells . To test this hypothesis , a mixture of filamentous cells and short cells collected from the P2 position of a decayed potato after 48 hpi were transferred to a fresh potato tuber . The percentage of filamentous cells and short cells were monitored after the transfer . As expected , exposure to fresh potato tissue caused a drastic reduction in filamentous cell percentage in the total population , from 67 . 9% at 0 hpi , to 18 . 0% at 1 hpi , and only 3 . 9% at 4 hpi ( Fig 11A ) . To visualize the transformation process , a time-lapse movie was captured documenting long , filamentous cells dividing to form multiple shorter cells upon exposure to freestanding water ( S2 Video ) . As additional evidence , we used DAPI to stain the nucleoids of the filamentous cells . Multiple segmented nucleoids were observed in some of the filamentous cells , suggesting that filamentous cells have the potential to divide into short cells ( Fig 11B ) . All together , these data support that the non-virulent filamentous cells may be transformed to virulent short cells when environmental conditions are suitable for infection and/or spread .
In this study , we report heterogeneity in the morphology and physiology of a genetically homogeneous population of D . dadantii during host infection . We determined that D . dadantii adopts a bistable state in the host in which subpopulations adopt a suite of multiple coordinated phenotypes: short cells were associated with virulence gene expression , slower growth , and active motility , while filamentous cells were associated with lack of virulence gene expression , faster growth , reduced stress tolerance , and lack of motility . Furthermore , we found that the alarmone signal molecule ppGpp is critical for the differentiation of cells into the two subpopulations . The phenotypic states we observed were most obviously characterized by differentiation in cell morphology and in virulence . Bacterial elongation and filamentation under natural conditions have been previously reported in a few bacteria [27–30] , with proposed biological roles mainly related to stress tolerance [27] . For example , upon detection of host immune activation , subpopulations of uropathogenic Escherichia coli ( UPEC ) become filamentous during acute infection in the oligotrophic bladder environment [29] . Filamentous cells are more resistant to neutrophil phagocytosis , and are thought to play a role in the survival and growth of UPEC during maturation of the intracellular bacterial communities ( IBC ) [30] . In other organisms , bacterial filamentation was shown to confer tolerance to protist gazing [31] and antibiotic exposure [32 , 33] . Different from the previous studies , here we showed that filamentous cells of D . dadantii are not involved in tolerance of the two stress conditions tested . Instead , they were induced in a nutrient-rich host tissue when infection was well established , possibly as a mechanism to maximize pathogen growth and nutrient utilization . While we observed that filamentation may be associated with increased growth capacity and expression of metabolic genes , the adaptive benefit to cell elongation itself remains to be seen . Heterogeneity may provide fitness advantages for a bacterial population as a whole [34 , 35] . Heterogeneity in virulence expression has been observed in other bacterial pathogens [10 , 11 , 17] , where it was observed that expression of virulence genes imposes a penalty on bacterial growth rate [11] . However , the best-studied example of physiological heterogeneity is the phenomenon of antibiotic persistence . Persistence is a strategy in which a subpopulation of cells ( persisters ) exists in a dormant state that confers increased tolerance to environmental stress [36] . Here heterogeneity is thought to work as a “bet-hedging” strategy , benefiting the population as a whole by increasing its adaptability to changing environments . It was also observed that persister cell formation is enhanced by nutrient limitation and other sources of cellular stress [37 , 38] . In contrast to persistence , we found evidence that a subpopulation of D . dadantii forms filaments and increases its growth capacity in high-nutrient tissues , but this comes at a cost of virulence gene expression and tolerance to certain stresses . We hypothesize that in D . dadantii , phenotypic heterogeneity may allow the pathogen to simultaneously accomplish the two conflicting goals of virulence and vegetative growth . This is has been called a “division of labor” strategy of phenotypic heterogeneity , in which a population benefits from two physiological states with different tradeoffs . A proposed model for the occurrence and function of phenotypic heterogeneity during the D . dadantii infection cycle is described in Fig 12 . Interestingly , the alarmone signal ppGpp was implicated in a positive role in the induction of the persister state [39] . Evidence points to stochastic noise in ppGpp fluctuation as the most likely driver in the transition between persister and nonpersister states in individuals of a bistable bacterial population [40] . Here we observed a strong negative relationship between ppGpp and the filamentation of D . dadantii in tubers , finding that ppGpp biosynthesis genes are differentially expressed in filamentous and short cells , that exogenous application of ppGpp shifts the proportion of filamentous cells , and that filamentation phenotypes can be recapitulated through the deletion of ppGpp biosynthesis genes . In light of these findings , we propose that ppGpp variation also mediates the multi-phenotype bistable state observed in D . dadantii during host infection . ppGpp regulates diverse phenotypes including virulence , growth and motility [39 , 41 , 42] , thus cell-to-cell variation in ppGpp concentration could drive the tandem expression of multiple phenotypes in single cells of D . dadantii . In linking ppGpp to a second form of bistable state marked by coordinated cell length , virulence , and metabolic traits , this study suggests that ppGpp-mediated pathogen population heterogeneity may be more diverse and significant under native infection conditions than previously recognized . This bistable state we observed was specific to certain conditions of host tissues , infection sites , and water availability levels; the environmental factors and additional mechanisms that trigger conditional bistability is unknown . However , these findings raise the possibility that similar phenomena are occurring in other pathogens that do not undergo such easily observed morphological changes during infection .
Bacterial strains , plasmids , and oligonucleotide primers used in this study are listed in S1 Table . Strains were stored at -80°C in 20% glycerol . Dickeya dadantii strains were cultured in Lysogeny Broth ( LB ) medium , mannitol-glutamic acid ( MG ) medium ( 1% mannitol , 0 . 2% glutamic acid , 0 . 05% potassium phosphate monobasic , 0 . 02% NaCl and 0 . 02% MgSO4 ) , pel-inducing minimal medium ( pel-MM ) ( 0 . 5% of polyglacturonic acid , 0 . 1% of yeast extract , 0 . 1% of ( NH4 ) 2SO4 , 1 mM MgSO4 and 0 . 5% glycerol ) or hrp-inducing minimal medium ( hrp-MM ) at 28°C [43–45] . Escherichia coli strains were cultured in LB medium at 37°C . Antibiotics were supplemented to the media at the following concentrations as needed: ampicillin ( 100 μg ml-1 ) , kanamycin ( 50 μg ml-1 ) and spectinomycin ( 100 μg ml-1 ) . For inoculations on potato tubers , mature tubers ( Solanum tuberosum ‘Russet Burbank’ ) were dissected from the middle , and half of a tuber was placed in a 500 ml glass beaker with the dissection facing up . Three milliliters of water was added to the bottom of the beaker . Two hundred microliters of bacterial suspension ( 1 × 108 CFU ml-1 ) in 0 . 5 × phosphate saline buffer ( PBS ) was added on top of the tuber . Beakers containing inoculated tubers were sealed with plastic wrap and incubated at 28°C . For samples with freestanding water , 500 μl of sterile distilled water was added on top of the tuber every 4 h during the incubation . Tissues from the inoculated potato tubers were collected at different time points and were used directly for microscopy , staining , or cell separation procedures . Each experiment consisted of 3–5 biological replicates , and experiments were repeated at least three times . Leaf and stem inoculations were performed using a previously described method [46] . For scanning electron microscopy ( SEM ) , potato tuber tissues with typical soft rot symptoms were collected and fixed in paraformaldehyde/ glutaraldehyde ( 2 . 5% of each compound in 0 . 1 M sodium cacodylate buffer ) ( Electron Microscopy Sciences , Hatfield , PA ) at 25°C overnight . Fixed tissues were dehydrated in 25 , 50 , 75 , and 90% ethanol for 1 hour each and in 100% ethanol three times for 30 minutes . Dehydrated samples were air dried at room temperature and mounted on aluminum mounting stubs . Images were captured using a Zeiss Sigma VP FESEM ( Carl Zeiss Inc . , Oberkochen , Germany ) . For epifluorescence microscopy , symptomatic tissues were mounted on glass microscope slides with 1–2 μl of distilled water and covered with a coverslip . If it was necessary for cells to be fixed onto the slides to produce a still image , slides coated with agarose ( 5 μl of melted agarose on a glass slide to air dry ) were used . Images were observed using a Zeiss Axioplan 2 fluorescence microscope ( Carl Zeiss Inc ) outfitted with a Zeiss AxioCam digital camera . For bacterial cell length measurement , ImageJ software [47] was used to measure at least 1 , 000 individual cells from at least 5 different microscopic views for each sample . For confocal microscopy , samples were prepared using the same way as the epifluorescence microscopy . Observation was made using a Leica SP5 confocal microscope ( Leica , Wetzlar , Germany ) , equipped with four lasers , 405 nm , multi-line Argon , 561 nm and 633 nm , and two HyD detectors . Potato tuber was inoculated with D . dadantii in the absence of freestanding water for 48 h . Infected potato tissues were collected from P2 position and was re-suspended in a RNA Protect ( Qiagen , Hilden , Germany ) -water solution ( ratio 1:2 ) . The tubes were vortexed for 30 s to break up potato tissues and the intertwined bacterial cells . The resulting suspension was filtered through a 70-μm laboratory sifting mesh ( Carolina Biological Supply Company , Burlington , NC , U . S . A . ) to remove large plant debris . The flow-through containing bacterial cells was loaded on top of a prepared sucrose gradient solution ( 20% , 30% , 40% , 45% and 50% ) and was centrifuged at 1 , 000 × g for 30 min . After centrifugation , short cells were collected from the 20% and 30% sucrose layers while the mixture cells containing enriched filamentous cells were collected from 45% and 50% sucrose layers . The enriched filamentous cells were further purified by repeating this sucrose gradient centrifugation process . Filamentous cells collected from the 45% and 50% sucrose layers of the second round purification and the short cells from both the first and second rounds of purification were used for RNA extraction and Western blot . Cell length prior to and after the sucrose gradient centrifugation procedure was visualized by microscopy . This procedure was illustrated in S2 Fig . To generate the relA-HA and spoT-HA in D . dadantii , we used double crossover mutagenesis to insert a HA tag in frame at the 3’ end of the open reading frame ( ORF ) of relA and spoT right before the stop codon in the D . dadantii chromosome . The 3’ region of each gene ORF without the terminator was PCR-amplified with an HA Tag and terminator , and the fragment downstream the gene ORF were also amplified . The kanamycin cassette was amplified from pKD4 [48] , and was cloned between two flanking regions using three-way cross-over PCR . The PCR construct was inserted into the suicide plasmid pWM91 , and the resulting plasmid ( pWM-relA-HA or pWM-spoT-HA ) was transformed into D . dadantii 3937 by conjugation using E . coli strain S17-1 λ-pir . To select strains with HA tag , the conjugated bacteria were first grown on MG medium amended with kanamycin , and then plated on MG plates amended with kanamycin and 5% sucrose . Sequences of target genes linked with HA tag were confirmed by Sanger sequencing . D . dadantii cells with respective HA tags were separated by sucrose gradient centrifugation and suspended in 0 . 5 × PBS buffer . Cells were lysed by sonication . Proteins in crude lysates were boiled and loaded onto a 10% SDS/PAGE gel . Proteins were then transferred onto a polyvinylidene fluoride membrane ( Millipore , Burlington , MA , USA ) . Blots were washed with PBS containing 0 . 05% Tween-20 and probed with an anti-HA antibody ( Thermo Fisher Scientific , Waltham , MA , USA ) . The GAPDH protein stained with Coomassie blue was used as a loading control . The blots were incubated for 5 min in enhanced chemiluminescence reagent ( Bio-Rad , Hercules , CA , USA ) and detected using Kodak X-ray film ( Kodak , Rochester , NY , USA ) . Total RNA collected from symptomatic potato tuber tissue was isolated using an RNeasy Plant Mini Kit ( Qiagen ) . Total RNA harvested from pel-MM or hrp-MM was isolated using an RNeasy mini kit ( Qiagen ) . Extracted RNA was treated with Turbo DNase I ( Ambion , Austin , TX , USA ) and cDNA was synthesized from 1 μg of treated total RNA with iScript cDNA synthesis kit ( Bio-Rad , Hercules , CA , USA ) . The complementary cDNA level of target genes was quantified by qRT-PCR using a SsoAdvanced universal SYBR Green supermix ( Bio-Rad ) , as described previously [46] . Data were analyzed using a relative expression software tool [49] . The expression level of rplU was used as an endogenous control . The experiment was repeated twice with similar results . Total RNA quality and integrity was determined by a nanodrop ( Thermo Fisher Scientific ) and by an Agilent Bioanalyzer ( Agilent Technologies , Santa Clara , CA , USA ) respectively . RNA transcripts were enriched by selectively depleting ribosomal RNA molecules using the Ribo-Zero Bacteria Kit ( Epicentre , Madison , WI , USA ) and then sheared by incubation at 94°C . Following first-strand synthesis with random primers , second strand synthesis was performed with dUTP for generating strand-specific sequencing libraries . The cDNA library was then end-repaired , and A-tailed , adapters are ligated and second-strand digestion was performed by Uricil-DNA-Glycosylase . Sample concentrations were normalized to 10 nM and loaded onto Illumina Rapid or High-output flow cells at a concentration that yields 150–250 million passing filter clusters per lane . Samples were sequenced using paired-end sequencing on an Illumina HiSeq 2500 according to Illumina protocols . D . dadantii 3937 genome ( NC_014500 . 1 ) from NCBI was indexed using HISAT2 [50] and was used as the reference genome in the RNA-seq analysis . The reads obtained from the machine were trimmed for quality and length using customs scripts . The trimmed reads were then aligned with the indexed D . dadantii genome using HiSAT2 . Annotation for the reference genome was obtained from the NCBI and used in the analysis . The transcript abundances were estimated using ballgown [51] . The differential gene expression was calculated using DESeq2 [52] , and the results were visualized using R . The relA and spoT single and double deletion mutations were generated by marker exchange mutagenesis . Briefly , two fragments flanking each target gene , as well as a kanamycin cassette were PCR amplified . The three fragments were linked together using a three-way cross-over PCR with kanamycin cassette flanked by the upstream and downstream sequences of the target gene . The fusion PCR was inserted into a suicide plasmid pWM91 [5] . The resulting plasmid ( pWM-relA or pWM-spoT ) was transformed into D . dadantii 3937 by conjugation using E . coli strain S17-1 λ-pir . To select strains with chromosomal deletions , recombinants , grown on MG medium amended with kanamycin , were plated on MG agar plates with 5% sucrose . To generate the ΔrelAΔspoT double deletion mutant , pWM-spoT was transformed into a relA marker-less mutant via conjugation . The marker-less ΔrelA was generated by transferring plasmid pFLP-1 [5] into the ΔrelA strain through electroporation . Transformants were plated on LB plates containing ampicillin . Colonies that lost the kanamycin resistance were selected . pFLP-1 was cured from the mutants by replica plating on MG sucrose plates . To generate complementation strains , open reading frame regions of relA and spoT were PCR amplified and cloned into a low copy number plasmid pCL1920 . Marker-less deletions were confirmed by PCR . All mutants and complementation strains were confirmed by sequencing . Bacterial strains were grown in 500 ml LB to OD600 = 1 ( 109 CFU ml-1 ) . Cell pellet was harvested and immediately suspended in 2 ml of 100% cold methanol , vortexed , frozen in liquid nitrogen and thawed on ice . Cell suspensions were then centrifuged at 4 , 000 rpm for 10 min at 4°C . The bacterial supernatants were collected and stored on ice . Same extraction procedure was repeated using the remaining pellet . Methanol extracts from the two extractions were combined , freeze-dried and re-suspended in 200 μl distilled water for LC/MS/MS analysis . Pure ppGpp ( TriLink Biotech , CA , USA ) was used as a positive control . Mass spectrometric quantification was performed using a XEVO_TQS ( Waters Corporation , MA , USA ) using ultra performance liquid chromatography and tandem mass spectrometry ( UPLC-MS/MS ) . Samples were run through an Acquity UPLC BEH C18 1 . 7μm column ( Waters Corporation , MA , USA ) maintained at 40°C . Samples were processed under the electrospray -ve ( ES- ) ion mode with a capillary voltage of 3 kV and a desolvation temperature of 400°C with a sample injection volume of 10 μL . Methanol ( A ) and 8mM DMHA+2 . 8nM acetic acid ( B ) in water were used as the two solvents under a flow rate of 0 . 3 ml/min . The linear gradient for solvent B was as follows: 0–10 mins , 60%; 10–15 mins , 100% . MassLynx was used as the main interface software ( Waters Corporation , MA , USA ) . Area of the corresponding peak in each sample was quantified using ImageJ . Pathogenicity assay was performed on potato tubers . Briefly , overnight bacterial cultures were harvested and re-suspended in 0 . 5x PBS to a final concentration of 1×108 CFU ml-1 . Potato tubers were dissected from the middle and a half potato was placed in a beaker with the dissected section facing up . Two hundred microliters of bacterial suspension was added on top of dissected tuber . Inoculated potato tubers were kept at 28°C with 100% relative humidity for 18 h . Weight of decayed tuber tissue was measured using an analytical balance . Hypersensitive response assay was conducted using four week old tobacco ( Nicotiana tabacum L . cv . Xanthi ) as previously described [53] . Briefly , overnight bacterial cultures were harvested and re-suspended in 0 . 5x PBS to a final concentration of 1×108 CFU ml-1 , and were infiltrated into leaves using needle-less syringe . Infiltrated plants were maintained in greenhouse , and HR symptoms were recorded at 20 hpi . Plate assay for Pel production was performed as described by Matsumoto et al . [54] . Briefly , D . dadantii strains were cultured in pel-MM [55] . Twenty microliters of bacterial supernatant was acquired by centrifuging 1 ml of the overnight bacterial culture , and was applied to each well made in the Pel plate ( 1% of PGA , 1% of yeast extract , 50 μM CaCl2 , 50 mM Tris-HCl pH 8 . 5 , 0 . 8% agarose and 0 . 2% sodium azide ) with a No . 2 cork borer ( ɸ5 mm ) . The bottom of each well was sealed with 0 . 8% of molten agarose . The plates were incubated at 28°C for 20 h , and Pel production indicated by halo rings was developed by incubating the plates with 5 N of H2SO4 for 5 min . For quantification of bacterial motility from decayed potato , potato tissues inoculated with D . dadantii ( pnptII-gfp ) at 48 hpi were collected and resuspended in sterile distilled water . Cells were viewed and imaged at 400× magnification on a fluorescence microscope . Videos were captured for at least 30 seconds using screen capture tool CamStudio ( https://camstudio . org/ ) . The speed of bacterial swimming was calculated by measuring the mobile distance between two time points by using Adobe Photoshop ( Adobe , San Jose , CA , USA ) . At least ten cells of each sample were used for speed calculation . For swimming and swarming motility plate assays , cells were cultured in LB broth overnight , harvested and adjusted to a final concentration of 1×108 CFU ml-1 in 0 . 5×PBS . Bacterial suspension of 10 μl was spotted onto the center of swimming plates ( 0 . 2% MG agar ) and swarming plates ( 0 . 4% LB agar ) . The plates were incubated at 28°C for 19 h , and the diameter of the radial growth was measured [56] . Each experiment consisted of three biological replicates and was been repeated at least two times . For oxidative stress assay , overnight bacterial cultures were inoculated 1:100 into fresh LB broth . After 18 h , H2O2 were added to the bacterial suspension to reach a final concentration of 10 mM . The cell suspensions were incubated for an additional 2 h at 28°C . The number of colony forming unites ( CFU ) was determined by serial dilution and plating at time 0 h and then every 30 min over a 2-h incubation period . For ampicillin persistence assay , the overnight bacterial cultures were inoculated 1:100 into fresh LB broth . After 18 h , the cell suspensions were serial-diluted and plated onto LB plates with and without ampicillin ( 1 μg ml-1 or none ) to determine the number of CFU . The wild type , ΔspoT mutant , and complementation strain of D . dadantii ( with nptII-gfp ) were cultured in LB broth and adjusted to the same concentration ( 108 CFU ml-1 ) . Two hundred microliters of cell suspension were added on top of dissected potato tuber , respectively . Forty-eight hours post inoculation , 30 mg of macerated potato tuber tissue was collected from the surface center of infected potato tuber respectively . The tissues were re-suspended and incubated in 10 ml of sterile distilled H2O in a 100 ml grass flask with shaking for 1 h at 28°C . After 1 h incubation , the bacterial suspension was serial-diluted and plated on LB Amp plates to determine the number of progeny . Bacteria were inoculated onto a potato tuber slide in the absence of freestanding water to induce filamentous cell formation for 24 hrs . At the end of the incubation , 2 ml of sterile distilled H2O was added onto the tuber surface to induce the filamentous cell division . Twenty four hours later , the remaining filamentous cells collected from the surface center of the infected tuber slide was seeded and grown on the agarose pads on microscope slides and the cell division was documented using time-lapse microscopy with the protocol described by Young et al . [57] . DAPI staining was performed as described by Markus et al [58] . Briefly , a mixture of filamentous cells and short cells was collected from the surface center of the infected potato tuber , resuspended in 0 . 1 ml PBS buffer , and then fixed in 75% ethanol for 10 min . The fixed cell sample was collected by centrifugation and re-suspended in 20 μl of 10 mM Tris-MgCl2 buffer containing 1 μg ml-1 DAPI ( 10 mg ml-1stock ) , and was incubated for 15 min at 23°C . Stained cells were then imaged using excitation/emission filters at 360 nm ( BW 40 nm ) using a Zeiss Axio Scope ( Carl Zeiss Inc ) outfitted with a SPOT RT3 digital camera . Means and standard deviations of experimental results were calculated using Excel ( Microsoft , Redmond , WA ) , and statistical analyses were performed using the one-way analysis of variance ( ANOVA ) model in the ‘stats’ package in R unless specified . All experiment included at least three biological replicates and was repeated at least three times with similar results observed .
|
Virulence and vegetative growth are two distinct lifestyles in pathogenic bacteria . Although virulence factors are critical for pathogens to successfully cause infections , producing these factors is costly and imposes growth penalty to the pathogen . Although each single bacterial cell exists in one lifestyle or the other at any moment , we demonstrated in this study that a bacterial population could accomplish the two functions simultaneously by maintaining subpopulations of cells in each of the two lifestyles . During the invasion of potato tuber , the soft rot pathogen Dickeya dadantii formed two distinct subpopulations characterized by their cell morphology . The population consisting of short cells actively produced virulence factors to break down host tissues , whereas the other population , consisting of filamentous cells , was only engaged in vegetative growth and was non-virulent . We hypothesize that this phenotypic heterogeneity allows D . dadantii to break down plant tissues and release nutrients , while efficiently utilizing nutrients needed to build up a large pathogen population at the same time . Our study provides insights into how phenotypic heterogeneity could grant bacteria abilities to “multi-task” distinct functions as a population .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"cell",
"physiology",
"plant",
"anatomy",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"pathogens",
"dna-binding",
"proteins",
"cell",
"metabolism",
"plant",
"science",
"potato",
"plants",
"vegetables",
"proteins",
"gene",
"expression",
"pathogen",
"motility",
"biochemistry",
"tubers",
"eukaryota",
"cell",
"biology",
"virulence",
"factors",
"genetics",
"biology",
"and",
"life",
"sciences",
"biosynthesis",
"organisms",
"solanum"
] |
2019
|
Cell-length heterogeneity: a population-level solution to growth/virulence trade-offs in the plant pathogen Dickeya dadantii
|
Optimality principles have been proposed as a general framework for understanding motor control in animals and humans largely based on their ability to predict general features movement in idealized motor tasks . However , generalizing these concepts past proof-of-principle to understand the neuromechanical transformation from task-level control to detailed execution-level muscle activity and forces during behaviorally-relevant motor tasks has proved difficult . In an unrestrained balance task in cats , we demonstrate that achieving task-level constraints center of mass forces and moments while minimizing control effort predicts detailed patterns of muscle activity and ground reaction forces in an anatomically-realistic musculoskeletal model . Whereas optimization is typically used to resolve redundancy at a single level of the motor hierarchy , we simultaneously resolved redundancy across both muscles and limbs and directly compared predictions to experimental measures across multiple perturbation directions that elicit different intra- and interlimb coordination patterns . Further , although some candidate task-level variables and cost functions generated indistinguishable predictions in a single biomechanical context , we identified a common optimization framework that could predict up to 48 experimental conditions per animal ( n = 3 ) across both perturbation directions and different biomechanical contexts created by altering animals' postural configuration . Predictions were further improved by imposing experimentally-derived muscle synergy constraints , suggesting additional task variables or costs that may be relevant to the neural control of balance . These results suggested that reduced-dimension neural control mechanisms such as muscle synergies can achieve similar kinetics to the optimal solution , but with increased control effort ( ≈2× ) compared to individual muscle control . Our results are consistent with the idea that hierarchical , task-level neural control mechanisms previously associated with voluntary tasks may also be used in automatic brainstem-mediated pathways for balance .
Although optimality principles have been presented as a general framework for understanding motor control in animals and in humans [1] , the ability of optimization to explain experimental data using high-dimensional musculoskeletal models remains largely unknown . Studies using optimization approaches have demonstrated an impressive ability to predict qualitative features of motor behaviors , such as the presence of low-dimensional muscle patterns [2] , [3] , and the presence of high levels of noise in some redundant degrees of freedom and low levels of noise in others [4] . Further , studies using approaches based on optimal feedback control have even predicted features such as countermovements [1] , [5] . However , much of this evidence relies on biomechanical models that are abstract [1] , that lack muscles [6] , [7] or that have reduced degrees of freedom for computational efficiency [3] , [8] , [9] , [10] . When complex musculoskeletal models are used to predict experimental data , the greatly increased complexity often precludes investigation of more than a single experimental condition [2] , [11] , which may be insufficient to discriminate different candidate control strategies or cost functions [12] , [13] . Here , our goal was to test optimization as a predictive tool for understanding motor control by predicting detailed changes in experimentally-measured quantities across multiple biomechanical conditions . The postural response to perturbations during standing balance is a motor paradigm in which consistent patterns of motor outputs are elicited across different biomechanical contexts , but the degree to which these patterns reflect neural control or biomechanical mechanisms is unknown . To maintain balance , the center of mass ( CoM ) , a task-level variable , must be maintained above the base of support of the feet . Robust patterns of muscle activity referred to as the automatic postural response ( APR ) occur about 40 ms after horizontal translations of the support surface [14] and are consistently tuned to the direction of CoM motion across different perturbation types [15] , suggesting that these long-latency responses reflect task-level control of the CoM by the nervous system . This robustness is surprising given that in a quadruped , the net force acting on the CoM can be produced by many combinations of individual limb forces . Further , each limb force can be produced by many patterns of muscle activity due to muscular redundancy [16] , [17] . Active ground reaction forces during the postural response ( ∼100 ms latency ) tend to be directed along a diagonal axis either towards or away from the CoM across perturbation directions [18] . The distribution of individual limb force direction and magnitude in the horizontal plane is consistently altered by varying the distance between the fore- and hind-feet , yet surprisingly , the directional tuning of muscle activity remains intact ( Figure 1 ) [19] , suggesting that the limb force variation may be due largely to differences in biomechanical context across postural configurations . However , we demonstrated that biomechanical constraints alone are insufficient to determine the active production of limb forces during perturbations to standing balance . Measured postural forces are nearly ten times smaller than the absolute force production capability of a detailed musculoskeletal model of the isolated cat hindlimb in all directions [20] . The diagonal orientation of the forces is not predicted from anisotropies in the force-generating capability of the limb , which is greatest in the anterior-posterior direction . Further , changes in postural force directions across biomechanical contexts cannot be attributed to alterations in the force-generating capability of the limb , as peak force directions do not change appreciably across postural configurations [21] . Therefore , here we sought to improve our predictions of experimental measures through the addition of a model of a neural control mechanism that could achieve appropriate task-level forces and moments at the CoM while coordinating redundancy across both multiple muscles and across multiple limbs . The optimal feedback control of CoM dynamics predicts the timecourse of activity in single muscles during balance control in both quadrupeds and bipeds [22] , [23] , [24]; however , it remains unknown whether task-level constraints at the CoM are sufficient to predict execution-level motor patterns across multiple muscles and limbs in a complex and redundant musculoskeletal system . Such redundancy has previously been resolved by minimizing neural control effort , assumed to be equivalent to the sum squared muscle activation or sum squared motor commands [2] , [13] , [25] , [26] . Such optimizations have been applied to predict muscle tuning curves across conditions in relatively simple or quasi-static motor tasks [2] , [26] , or to deduce complex muscle activation patterns from detailed kinetic and kinematic measures [27] . Moreover , effort minimization is also sometimes treated as equivalent to energy minimization , which can predict aspects of gait in simple models of locomotion in humans and other animals [28] , [29] , [30] . However , predicting muscle coordination in detailed musculoskeletal models by minimizing quantities like effort remains challenging [11] . Further , it has been argued that low-dimensional muscle patterns emerge from optimization of the activation of individual muscles , without explicit neural constraints on muscle activation [1] , [2] . While low-dimensional patterns in the form of muscle synergy groupings have been observed experimentally [19] , [31] , [32] , [33] , [34] , [35] , [36] , studies using planar musculoskeletal models have noted similarities in motor behaviors predicted by optimally controlling individual muscles or muscle synergies [3] , [37] . Such predictions have been based on relatively simple or abstract musculoskeletal models , and thus it is not clear whether such emergent low-dimensional patterns are competent to predict forces and muscle activation patterns in more behaviorally-relevant motor tasks . It has also been argued that muscle synergies may allow for near-optimal performance with simplified computations based on a reduced number of controlled variables [37] , [38] , but may increase control effort due to additional coactivation [39] . However , direct comparisons of the energetic cost associated with controlling individual muscles or muscle synergies in a 3D model of a natural behavior have not been performed . Here , we sought to identify a task-level optimization framework that could predict execution-level limb forces and muscle tuning measured in an unrestrained balance task across different biomechanical contexts . We hypothesized that features of execution-level patterns of limb forces and muscle activity reflect the minimum-effort solution for achieving appropriate forces and moments at the CoM . We compared predictions using a static quadrupedal musculoskeletal model of the cat to data from experiments . Specifically , we predicted that limb forces would be directed along the diagonal for long stance distances , and more evenly distributed in direction at short stance distance . Further , we predicted that muscle activity would be low-dimensional , and that muscle tuning to perturbation direction would scale , but not shift as postural configuration varied . By varying cost functions and task-level variables we demonstrated that the predicted outputs depended on the optimization formulation , and not simply the biomechanical constraints . Finally we compared results from optimal control of individual muscles to those based on controlling experimentally-derived muscle synergies . Our work suggests that the neural control of this natural behavior can be well described by a cost function that minimizes effort expended in the muscles in order to achieve appropriate forces and moments to stabilize the CoM . Further , our results are consistent with the idea that the computation may be implemented in a hierarchical control framework that allows for approximately-optimal motor patterns with a reduced number of controlled variables .
To test the hypothesis that execution-level variables reflect optimal control of task-level variables , we predicted patterns of limb forces and muscle activity in response to multidirectional postural perturbations in cats based on achieving task-level mechanics while minimizing different formulations of control effort ( Table 1 ) . Using a detailed static quadrupedal musculoskeletal model of standing balance , we first identified patterns of muscle activity that produced forces and moments at the CoM necessary to maintain balance in response to postural perturbations in twelve different perturbation directions while minimizing neural control effort ( model MMe ) . We considered multiple postural configurations with altered stance distance between the fore- and hind-feet . We compared identified muscle activation patterns and the resulting ground reaction forces to mean values measured experimentally during the initial response . In order to demonstrate that biomechanical constraints alone could not account for the identified solutions , we demonstrated that alternate cost functions and task goals produced qualitatively different results . We compared predictions from minimum effort control of CoM force and moment to predictions from minimizing an alternative cost function designed to be a better representation of the metabolic energy used in the muscles ( model MMm ) . Additionally , we compared predictions of controlling an alternate task-level variable , the position of the center of pressure ( CoP; model MPe ) . Finally , to investigate whether task-level control of the CoM could be accomplished with a small number of muscle synergies , rather than with individual muscles , we constrained the muscles in the model to activate in muscle synergies adapted from previously-observed experimental data ( models SMe and SMc ) . We estimated and compared the energetic cost , the computational cost , match to experimental data , and the dimensionality of the muscle activation patterns predicted by controlling individual muscles or postural muscle synergies . We parameterized the musculoskeletal model and assessed predicted limb forces and muscle activation patterns using previously-collected data of three cats during quiet standing and postural perturbations in multiple postural configurations [19] . The cats ( bi , 2 . 7 kg; ru , 4 . 2 kg; ni , 3 . 5 kg ) were trained to stand unrestrained with weight evenly distributed on four 8 cm-square force plates mounted on a moveable perturbation platform that could translate in any of 12 directions in the horizontal plane ( Figure 2 ) . Translations were 15 cm/s velocity and 5 cm amplitude . Data were collected in a self-selected postural configuration ( preferred configuration ) , and in postural configurations in which the stance distance between the fore- and hind- force plates was altered . The following stance distances were examined in each of the animals: bi , 30 cm , 27 cm ( preferred ) , 20 cm , and 13 cm; ru , 40 cm , 29 cm ( preferred ) , 24 cm , and 18 cm; ni , 29 cm ( preferred ) , 24 cm , and 18 cm . Stance width between the left and right force plates was 8 cm in all conditions . We modeled muscle activity and limb forces associated with the initial period of the automatic postural response ( APR ) to perturbation , which can be studied as a quasi-static process . Multiple experimental and modeling studies have demonstrated that the forces during the initial portion of the APR can be attributed primarily to muscular forces [19] , [40] , [41] . During this period the acceleration- and velocity-dependent terms in the equations of motion are negligible so that the influence of dynamic terms on ground reaction forces is minimal [42] and the task can be approximated as quasi-static . This feature is due to the fact that there are distinct delays between the perturbation onset , the evoked muscular activity , and the subsequent active force . EMG activity due to the initial perturbation acceleration occur approximately 60 ms after the onset of the perturbation and only produces active forces at the ground after an additional 60 ms delay . Thus , there is no interaction between the perturbation acceleration and the active forces which occur during the constant-velocity , e . g . quasi-static phase of the perturbation [15] . Similarly , the acceleration of the body segments is largest while the acceleration of the platform is transmitted across all body segments [43] , whereas after this period , the CoM has approximately constant horizontal-plane velocity ( note the approximately constant slope of the CoM displacement during the active period indicated by the gray bar , Figure 2 ) . Therefore , inertial forces associated with segment accelerations are not appreciable during the active response . Second , due to the relatively short latency of the active response compared to the overall motion , the posture of the animal has not changed appreciably from quiet standing at the onset of the active response . The posture of the animal affects gravitational forces , as well as torque generation via the muscle moment arm matrix . However , at the onset of the active force , the total displacement of the CoM is typically less than 1 cm and the effective tilt angle of the CoM is 1–2° [15] . Therefore the posture can be considered to be static , with no appreciable changes in gravitational forces or muscle moment arms . Therefore , our model assumes that all of the ground-reaction forces during the initial period of the APR are due to muscular activation , rather than dynamic terms . We created the quadrupedal musculoskeletal model by modifying and assembling four instances of an existing static , 3-D musculoskeletal model of the cat right hindlimb [20] , [21] . The hindlimb model relates 31-element muscle excitation vectors to the six-element force and moment system produced at the hindlimb endpoint: ( 1 ) where the vector is comprised of the model's seven kinematic degrees of freedom: three at the hip , and two each at the knee and ankle , designates the Moore-Penrose pseudoinverse of the transpose of the geometric system Jacobian ( pinv . m ) , designates the moment-arm matrix , and and are diagonal matrices of maximum isometric forces and scaling factors based on muscle force-length properties [44] . Hindlimb model parameters are provided for each animal and experimental condition in Dataset S1 . The muscles included in the hindlimb model and recorded in experimental data are summarized in Table 2 . A closed-form expression for the Jacobian was identified with AutoLev software ( Online Dynamics , Inc . , Sunnyvale , CA , USA; currently being developed as MotionGenesis Kane ) and implemented in Matlab ( Mathworks , Natick , MA ) . The model of the left hindlimb was created by duplicating the right hindlimb model and reversing the sign of the lateral force component . Prior analyses demonstrated that the hindlimb model is insensitive to the pseudoinverse operation , although the choice of pseudoinverse can be particularly important in robotics applications [45] , [46] . Two previous studies demonstrated that the overall hindlimb force production capability is unchanged whether one degree of freedom ( hip rotation ) is locked , making the Jacobian 6×6 and exactly invertible [20] , or whether the pseudoinverse is used [21] , because the majority of the muscles in the model have hip rotation moment arms that are small in comparison to other degrees of freedom at the hip . Further , very similar endpoint force directions are produced by the muscles in the model in these two conditions . Across muscles , animals , and experimental conditions , the average difference in predicted endpoint force direction between the hip-locked and pseudoinverse conditions was only a few degrees ( 2 . 8±5 . 0° , dorsal plane; 4 . 4±11 . 3° , sagittal plane ) . These results are consistent with recent experimental results in which similar mappings between muscle forces and endpoint forces and torques were identified when mechanical degrees of freedom were locked or freed [46] . Because a detailed musculoskeletal model of the forelimb was unavailable , we approximated the forelimb by modifying the hindlimb model into a vertical strut that transformed muscle activation to vertical force . Although the forelimbs do not always contribute to horizontal-plane forces during the postural response [18] , they contribute non-negligible vertical forces , of magnitudes several times larger than their horizontal force magnitudes . There also may be less potential for horizontal-plane forces to be produced by the extensor muscles in the cat forelimb because the morphology is more columnar than that of the hindlimb . Therefore , we approximated the forelimb as a transformation from muscle activity to vertical force by eliminating all rows of Equation 1 except for the row corresponding to vertical force . The transformation from muscle activation to CoM force and moment in the quadrupedal musculoskeletal model was found using the forces from each limb and the approximate location of the CoM . Resultant CoM force was calculated as the sum of the individual limb forces . Resultant CoM moment was calculated as the sum of the vector cross products between the vectors from the CoM to the limb endpoints and the limb forces . Limb endpoint moments were assumed to make negligible contributions to the net moment at the CoM . The net force and moment at the CoM due to individual limb forces is thus: ( 2 ) Where designates the vector from the CoM to the endpoint of limb . The transformation from muscle activation to force and moment at the CoM was formulated as a 6×124 matrix equation for each postural configuration and animal relating muscle activation levels ( 31 muscles in each limb , for 124 total ) to the 6D CoM force and moment . We identified joint angles in the musculoskeletal model that best approximated the recorded kinematics of each cat during quiet standing in each postural configuration ( Figure 3A ) . Positions of kinematic markers located on the platform and the left sides of the body were collected at 100 Hz during each trial for each cat . Locations of joint centers were estimated from marker positions by subtracting off joint radii , skin widths , and marker widths . The joint angles that minimized the squared error between the sagittal- and frontal-plane angles of the femur , shank , and foot in the model and in the background-period kinematics of each trial of each cat were identified using numerical optimization ( fmincon . m ) [20] . All residual segment angle errors were ≤10−4° . Joint angles were averaged across like trials . Muscle moment arm values and fiber lengths were determined with SIMM software ( Musculographics , Inc . , Santa Rosa , CA ) and averaged across like trials . We approximated the location of the CoM with respect to the feet in the musculoskeletal model separately for each cat in each postural configuration based on kinematic data and morphological parameters . For all conditions , the CoM was assumed to be located midway between the limb endpoints in both the anterior-posterior and medial-lateral directions . The height of the CoM above the plane of the feet was estimated from kinematic data and morphological parameters separately for each cat in each postural configuration . Across postural configurations , average CoM heights for each cat were ( mean ± SD ) : bi , 12 . 6±0 . 4 cm; ru , 15 . 2±0 . 4 cm; ni , 12 . 7±0 . 8 cm . To determine whether minimum-effort task-level control of the CoM could predict execution-level limb forces and muscle activity , we first identified patterns of muscle activity in the musculoskeletal model that produced forces and moments at the CoM similar to observed values while minimizing squared muscle activation ( model MMe ) . Task-level constraints on CoM force and moment were based on average values from experimental data ( Figure 3B ) . Average limb forces and CoM positions during the active period 120–200 ms after perturbation onset [19] were combined to estimate the average forces and moments at the CoM for each perturbation direction and postural configuration of each animal . Moments generated at the limb endpoints were assumed to make negligible contributions to the net CoM moment . Because values were similar across animals and postural configurations , a single set of average CoM forces and moments that was considered representative for all animals was then created and used as the optimization constraint: net horizontal-plane forces directed in the perturbation direction of 2 . 5 N magnitude , net vertical forces of 30 N , and net pitch-roll moments of 0 . 75 N-m magnitude directed perpendicular to the perturbation direction . CoM yaw moment was left unconstrained . Muscle activation patterns that satisfied task-level constraints could not be identified analytically without violating physiological bounds on muscle activation [44] . Therefore , optimizations were formulated as quadratic programming problems ( quadprog . m ) to identify muscle activation patterns that satisfied task-level constraints while minimizing total squared muscle activation: ( 3 ) Where designates a vector containing the activity levels of all muscles in the model ( 124 ) . Additional constraints ensured that the activation levels of each muscle were in the interval ( 0 , 1 ) and that vertical ground reaction forces were ≥0 . Separate optimizations were performed for each animal , postural configuration , and perturbation direction . To investigate whether similar force predictions could arise from optimization criteria other than the minimum effort criterion used in model MMe , we next altered the cost function to better approximate metabolic energy consumption in the muscles , in terms of Joules/second , than minimizing Equation 3 , but without the added complexity of Hill-type muscle models [47] . In single muscle fibers , metabolic energy usage ( Joules/sec ) is proportional to stress [48] , equivalent to muscle activation in the model used here [44] . We assumed that the number of fibers in a muscle , and therefore its energy consumption , is proportional to its mass . Therefore , we performed additional optimizations with constraints and methods identical to the first model formulation , but minimizing total squared muscle activation weighted by muscle mass: ( 4 ) Where is a diagonal matrix of muscle masses . Masses for each muscle are included in Dataset S1 . The majority of muscle masses ( 23/31 hindlimb muscles ) were taken from the literature [49] . Because muscles for which no data were available were typically small , these masses were all set to a common low-midrange value . To investigate whether the minimum-effort control of an alternate task-variable could predict similar limb forces , we tested a formulation similar to model MMe , except constrained to match displacements of the CoP in each perturbation direction , leaving the net force at the CoM unconstrained . Some studies of sagittal-plane balance in humans have suggested that the location of the CoP is the task-level variable controlled during balance [50] . Task-level constraints on CoP displacement were based on average values from experimental data ( Figure 3C ) . The average displacement of the CoP at the midpoint of the active period in each perturbation direction for each postural configuration of each animal was calculated from the four vertical forces [15] . Similar to the first model formulation , a single set of corrections in CoP location ( 3 . 3 cm in magnitude and directed opposite the direction of the perturbation ) was created and used as task-level constraints in the optimizations . Next , to determine whether task-level control of the CoM could be accomplished with a small number of muscle synergies , rather than individual muscles , we constrained the muscles in each limb of the musculoskeletal model to activate in 5 muscle synergies based on muscle synergy force vectors previously observed in the same animals during the balance task [19] . The model [21] , assumes that the activation of each muscle results from the additive combination of a few muscle synergies , recruited by scaling coefficients . The activation level of the muscles in the model is therefore: ( 5 ) where each element of represents the activation of the muscle by the muscle synergy , restricted to be within the interval ( 0 , 1 ) , and the elements of scaling coefficients are restricted to be greater than zero . Five muscle synergies and related ground reaction force vectors were previously extracted from experimental data of each animal using nonnegative matrix factorization [19] . The muscle synergy patterns used in the model were subsequently derived by identifying patterns of muscle activation in the hindlimb model that could produce each ground reaction force vector while minimizing squared muscle activation ( Equation 3 ) [21] . Identical muscle synergies were used in each limb and in all postural configurations . The constraints and solution method in this formulation were very similar to model MMe , with the exception that muscle synergy activation levels were identified rather than muscle activation levels . Synergy activation levels were constrained to be positive with respect to a level that created a background net vertical force . We considered two different cost functions in optimizations of muscle synergy control . Optimizations were performed that minimized muscle effort , ( model SMe ) , as in Equation 3 , but with the addition of muscle synergy constraints: ( 6 ) Further , to determine whether optimal solutions could be identified entirely in reduced-dimension space , optimizations were also performed ( model SMc ) that satisfied task-level constraints on CoM force and moment while minimizing sum squared muscle synergy activation: ( 7 ) We calculated goodness-of-fit between predicted left hindlimb and right forelimb forces and experimental data from each animal across experimental conditions . Because vertical force ( VF ) magnitudes are several times larger than horizontal force ( HF ) magnitudes , they were analyzed separately . We compared predicted left hindlimb ( LH ) HF direction , LH HF magnitude , LH VF magnitude , and right forelimb ( RF ) VF magnitude with experimental data . R2 values for each force component were calculated across perturbation directions for each postural configuration for each animal and subjected to two-way ANOVAs ( postural configuration×animal ) evaluated with a significance level of α = 0 . 05 adjusted with a Bonferroni correction for multiple comparisons ( α = 0 . 0125 ) to determine whether the predictive ability of each formulation depended on the experimental condition . Left hindlimb HF magnitudes in perturbation directions that loaded the hindlimb ( 0° through 90° ) were also subjected to two-way ANOVA ( postural configuration×animal ) evaluated at α = 0 . 05 to determine whether magnitudes decreased as postural configuration was varied . R2 values predicted by the different model formulations were subjected to three-way ANOVAs ( postural configuration×animal×formulation ) and evaluated at the Bonferroni-corrected level of α = 0 . 0125 . We compared predicted muscle tuning curves to mean values from each animal and experimental condition . Mean values of EMG were calculated during the initial burst of muscle activity 60–140 ms after perturbation onset , and averaged across like trials . We compared the scaling and shifting in predicted tuning curve peak values across postural configurations to changes observed in data . Muscle tuning curves were normalized to maximum values observed in the preferred postural configuration of each cat . The peak magnitude and perturbation direction of each muscle tuning curve in each postural configuration of each animal was identified and expressed as a change from the preferred configuration value , either as a magnitude change , or as a direction change in degrees . In tuning curves with more than one peak , we tracked the peak value that was dominant in the preferred postural configuration . Tuning curve scaling was assessed by regressing peak values onto postural configuration ( L , P , S , SS ) and comparing the resulting regression coefficients for each cat and model . Tuning curve shifting was assessed by calculating the maximum change in peak direction across postural configurations . These values were then subjected to one-way ANOVA evaluated at α = 0 . 05 to determine whether shifts predicted by each model were comparable to observed values . We assessed the dimensionality of muscle activation patterns predicted by models MMe , SMe , and SMc using a simple criterion based on principal components analysis ( PCA ) . As we were primarily interested in comparing muscle activity pattern dimension predicted by controlling individual muscles ( MMe ) versus that predicted by controlling postural muscle synergies ( SMe , SMc ) , we used a simple criterion that excludes components that contribute less variance than any individual variable in the original dataset [51] , [52] . Vectors of predicted left hindlimb muscle activation were assembled into matrices arranged with perturbation directions along the rows and muscles along the columns . Separate matrices were assembled for each postural configuration and animal . The dimensionality of each matrix was then estimated as the number of eigenvalues of the data correlation matrix ≥1 . 0 . Dimensionality estimates were pooled across animals and postural configurations and subjected to one-way ANOVA at a significance level of α = 0 . 05 to determine whether the formulations predicted similar muscle activity dimensionality . Dimensionality estimates from each model were compared to 5 , the previously reported value [19] . Comparisons were performed with t-tests at a significance level of α = 0 . 05 , adjusted with a Bonferroni correction for multiple comparisons to α = 0 . 0167 . We compared the total control effort required for controlling individual muscles ( MMe ) versus that required for controlling postural muscle synergies ( SMe , SMc ) . The control effort required for the muscle activity predicted by each model formulation was calculated with Equation 3 . Values were normalized to 100% of the value predicted by optimal muscle control in the preferred postural configuration of each cat . We then performed one-way ANOVA on the resulting values , at a significance level of α = 0 . 05 , to determine whether the three formulations predicted similar sum-squared muscle activity . We estimated the computational cost predicted by the three formulations by measuring and comparing the time required for each formulation to identify muscle activity patterns in all perturbation directions in each experimental condition . Resulting values were subjected to one-way ANOVA at a significance level of α = 0 . 05 , to determine whether the three formulations required similar computation time .
Task-level constraints on CoM force and moment or CoP location were satisfied by all of the models considered , but each predicted different patterns of muscle activity and limb forces , demonstrating the high level of redundancy of the quadrupedal musculoskeletal system . Experimentally-measured horizontal plane limb forces at preferred stance distance were predicted by task-level control of CoM forces and moments using either the minimum-effort or the minimum-energy cost functions ( models MMe and MMm ) , whereas solutions predicted by control of CoP control ( MPe ) differed substantially . However , differences between forces and moments predicted by models MMe and MMm were revealed when limb forces were examined across postural configurations; although MMe solutions varied in magnitude across stance distances in a similar fashion to experimental measures , MMm solutions did not predict any qualitative differences in limb forces across stance distances . Limb forces similar to MMe predictions were found when a muscle synergy constraint was enforced ( models SMe and SMc ) . In all three models that matched experimental limb forces across postural configurations ( MMe , SMe , SMc ) , muscle tuning directions were found to be invariant across postural configurations , similar to experimental data , resulting in low-dimensional overall muscle activity patterns . However , using muscle synergies derived from experimental data ( SMe , SMc ) allowed better predictions of activity in flexors , some of which were not activated in the independent muscle coordination conditions ( MMe ) . Finally , control effort increased by several times , but the time required for the quadratic programming search was decreased , when muscle synergies were controlled ( SMe , SMc ) rather than individual muscles ( MMe ) . Although we did not explicitly try to match experimentally-measured limb forces with the model , task-level control of CoM force and moment using either the minimum effort ( model MMe ) or minimum energy ( MMm ) cost functions nonetheless predicted horizontal plane forces directed towards and away from the CoM characteristic of the force constraint strategy described previously [18] in the preferred postural configuration ( Figure 4A , B , 27 cm; Figure 5A , 27 cm ) . Across all perturbation directions , predicted left hindlimb HF directions were similar to data ( MMe: mean R2 = 0 . 89±0 . 08 , P<1e-3; MMm: 0 . 84±0 . 08 , P<1e-3 ) . In perturbation directions that loaded the left hindlimb ( 0° to 90° ) , predicted HF forces were directed towards the CoM , similar to data ( data: mean direction 56±28°; MMe: 67±19°; MMm: 76±8° ) . In perturbation directions that unloaded the left hindlimb ( 180° to 270° ) , horizontal-plane forces were directed away from the CoM , again similar to data ( data: mean direction 263±9°; MMe: 254±13°; MMm: 258±6° ) . Left hindlimb HF magnitudes predicted by both cost functions varied as bimodal functions of perturbation direction similar to experimental data , particularly in loaded perturbation directions ( MMe: mean R2 = 0 . 94±0 . 09; MMm: 0 . 91±0 . 05 ) . Fits of left hindlimb HF magnitudes across all perturbation directions were reduced somewhat because of the small recorded force magnitudes in the unloaded perturbation directions ( MMe: mean R2 = 0 . 77±0 . 29; MMm: 0 . 48±0 . 09 ) . Maximal left hindlimb HF magnitudes were observed near 30° perturbations that loaded the hindlimb and minimal values for perturbations towards 120° , near the opposite diagonal axis . Average hindlimb HF magnitudes in perturbation directions where the left hindlimb was loaded ( 0° to 90° ) were 1 . 2±0 . 4 N in data vs . 2 . 4±0 . 9 N and 3 . 3±0 . 3 N , in the MMe and MMm models , respectively . Absolute predicted HF magnitudes were larger than recorded values , which was necessary in order to account for the absent contributions of the forelimbs . VF magnitudes predicted by both cost functions exhibited a realistic exchange between the forelimbs and hindlimbs as a function of the perturbation direction ( R2>0 . 98 ) . For perturbations diagonally to the right ( near 30° ) , left hindlimb vertical forces were maximal ( data: 11 . 4±3 . 4 N; MMe: 12 . 4±1 . 9 N; MMm: 12 . 3±1 . 8 N ) , whereas recorded right forelimb vertical forces were near minimal ( data: 3 . 8±2 . 5 N; MMe: 2 . 1±2 . 3 N; MMm: 2 . 0±2 . 4 N ) . Both cost functions predicted complete unloading ( 0 N ) of the left hindlimb and right forelimb in some cases , whereas the minimum vertical reaction forces observed in data were 1 . 0 N in the hindlimb and 0 . 6 N in the forelimb . Differences between the predictions of models MMe and MMm became apparent when other postural configurations were considered . Variations in left hindlimb HF direction and magnitude were observed across stance distances similar to data [53] in model MMe , but not in model MMm . As stance distance was decreased , a wider range of HF directions was observed in MMe but not MMm ( e . g . , compare changes between 27 cm and 13 cm in Figure 4A versus Figure 5A ) . Similarly , HF magnitude from longest to shortest stance had a greater decreasing trend in MMe ( −9±22% , P<0 . 25 ) than MMm ( −4±8%; P<0 . 55 ) in unloaded directions ( 180° to 270° ) . However , neither reached the degree of HF magnitude change observed experimentally ( −59±31%; P≪0 . 001 ) in unloaded directions . Over all directions , HF magnitude fits to data were significantly higher ( P<0 . 001 ) in MMe compared to MMm ( Table 3 ) . Moreover , HF magnitude fits were similar across postural configurations in MMe , but were significantly decreased at shorter stance distances in MMm ( P<0 . 0002 ) . Differences in forces across postural configurations were due to the fact that MMe favored recruitment of large muscles whereas MMm favored recruitment of small muscles . Large muscles that produce downward and backward endpoint forces relative to the limb axis were preferentially activated in MMe . When stance distance is shortened , the force rotates to have a more vertical orientation , thus reducing the component of force in the horizontal plane [19] , [21] . In contrast , smaller muscles produce forces with relatively small elevations in the horizontal plane , so that horizontal plane force components are relatively constant as stance distance is shortened . Compared to MMe , model MMm reduced the activation of large antigravity muscles by several times ( LG , mass 12 . 4 g , 1/3×; VL , 19 . 6 g , 1/4× ) and increased the activation of small muscles by 5–1000 times ( PSOAS , 4 . 0 g , 4×; SOL , 4 . 03 g , 20×; VI , 4 . 39 g , 5×; PT , 1 . 06 g , 1000× ) . Unlike experimental data , model MPe predicted HF directions near the strongest axis of force production in the isolated hindlimb [20] in all perturbation directions and postural configurations ( Figure 5B ) to achieve task-level constraints on CoP location . Because CoP location is measured about the projection of the CoM on the ground , predicted CoM forces and moments deviated significantly from experimental measures ( peak deviations: anterior force , 18 . 7±2 . 0 N; rightwards force , 1 . 9±0 . 4 N , roll-right moment , 0 . 2±0 . 1 N-m; pitch-up moment , 2 . 5±0 . 3 N-m ) . Although VF magnitudes predicted by model MPe were similar to data ( R2>0 . 86 ) , HF direction fits were poor ( R2 = 0 . 36±0 . 15 ) , and CoM-directed horizontal-plane forces were never observed . Instead , average left hindlimb HF directions were 90±3° and 98±3° for perturbation directions that loaded , and unloaded the left hindlimb respectively , near the direction of maximum force production of the hindlimb [20] . CoP control requires only modulation of VF magnitude across all four legs; large horizontal forces result from the fact that the minimum-effort muscle activation pattern to produce a vertical force component also has a very large horizontal component . These predictions were similar when the minimum energy cost function ( Equation 3 ) was used ( not shown ) . Adding muscle synergy constraints ( models SMe and SMc ) resulted in limb forces that were similar overall to predictions of model MMe ( Table 1 ) ; however , SMc additionally predicted a reduction in HF magnitude at shorter postural configurations that was comparable to the data ( see arrows in Figure 6 ) . As in MMe predictions , muscle synergy control models predicted characteristic HF directions towards ( SMe , 83±100°; SMc , 68±80° ) and away ( SMe , 254±45°; SMc , 255±42° ) from the CoM; however , visual inspection suggested that HF directions were more dispersed compared to MMe . Superior to MMe predictions , both muscle synergy control models predicted statistically-significant decreases in HF magnitudes in unloaded perturbation directions as stance distance decreased from preferred to shortest ( SMc , −31±39% , P≪0 . 0001; SMe , −4±44% , P<0 . 04 ) , although decreases were still less than those observed experimentally ( −59±31% ) . VF magnitudes were predicted well in both the left hindlimb and right forelimb in SMe and SMc ( R2 = 0 . 93±0 . 05 ) , although MMe predictions remained superior ( P<0 . 001 ) . As in MMe predictions , both the left hindlimb and right forelimb completely unloaded in some cases for SMe and SMc ( Figure 7 ) . In some perturbation directions of the shortest postural configuration of cat bi ( SMe , 5/132 total; SMc , 6/132 ) VF magnitude constraints were relaxed to allow CoM constraints to be achieved; these were excluded from further analysis . All models that predicted realistic limb forces across postural configurations ( MMe , SMe , SMc ) predicted smooth cosine muscle tuning to perturbation direction similar to experimental data , particularly in morphologically simple extensors ( Figure 8 ) . Experimentally-observed tuning curves from left hindlimb extensors were typically cosine-shaped and centered around rightwards perturbations ( 0° ) with approximate widths of 90°–120° at half-maximum ( e . g . , VM , GMED ) . Models MMe , SMe , and SMc made similar predictions for several extensors , including GMAX , GMED , VI , VM , VL , and SOL . Recruitment was not identical across models; for example , hip extensor BFA was recruited with similar tuning in SMe and SMc , but only in 1/3 cats in MMe . Some multifunctional extensors were more difficult to predict; for example , hip flexor/knee extensor RF was recruited in posterior/rightwards perturbations towards 330° experimentally , but predicted tuning curves ( MMe , SMe , SMc ) were centered about 0° . Ankle extensor/knee flexor MG was recruited with tuning curves centered near 180° by all models , unlike experimental results [15]; this tuning was similar to that observed in flexors , suggesting that the function at the knee might be dominating , with ankle extension being provided by extensor-tuned SOL . Ankle extensor/knee flexor LG was also recruited with tuning near 180° ( 1/3 cats , MMe ) or with bimodal tuning to leftwards and rightwards perturbations ( 3/3 cats , SMe , SMc ) . In some cases , the activation of flexor muscles was predicted by models SMe and SMc , but not by model MMe . Although some flexors were recruited with realistic cosine tuning about 180° in MMe , including PSOAS and SART ( Figure 8 ) , others were recruited in SMe and SMc but were never recruited in MMe . Ankle flexor TA was recruited with realistic cosine tuning to leftwards perturbations only in SMe , and only in cat bi . Some bifunctional muscles with flexor contributions were recruited in SMe and SMc but not in MMe . For example , hip extensor/knee flexor BFP was never recruited in MMe , but was recruited in 3/3 cats in SMe and SMc . Hip extensor/knee flexor GRAC was similar ( 2/3 cats , SMe , SMc; 0/3 cats , MMe ) , although predicted tuning curves were phase shifted somewhat from the anterior/leftwards tuning observed experimentally . Although hip extensor/knee flexor STEN was never recruited in MMe , it was recruited in SMe and SMc , but with either a bimodal ( 2/3 ) or extensor pattern ( 1/3 ) . Models MMe , SMe , and SMc all predicted muscle tuning curves that scaled in magnitude and shifted as stance distance was decreased comparable to experimental data ( Figure 9 ) . EMG peak magnitude increased as stance distance was shortened both in experimental data ( regression slopes of 0 . 25 , P<0 . 0001 , bi; 0 . 10 , ni; 0 . 10 , ru ) and in model predictions ( MMe , 0 . 19±0 . 01; SMe , 0 . 26±0 . 27; SMc , 0 . 22±0 . 29; all P<0 . 022 ) . Tuning curves predicted by all three models exhibited shifting with postural configuration that was not significantly different ( P>0 . 05 ) from recorded values ( average variation in peak tuning direction in data , 24±24°; MMe , 18±24°; SMe , 23±21°; SMc , 29±28° ) , although model SMc predicted increased tuning curve shifting compared to predictions of model MMe . Models MMe , SMe , and SMc all predicted low dimensional muscle activity patterns , with muscle synergy control predicting lower dimensional EMG than individual muscle control . Patterns of left hindlimb muscle activity predicted in MMe were characterized by 4 . 3±0 . 5 principal components across cats and postural configurations , significantly higher ( P<0 . 0001 ) muscle synergy control predictions ( SMe: 3 . 2±0 . 6; SMc: 3 . 1±0 . 7 ) . Dimensionality estimates from models MMe , SMe , and SMc were all significantly lower ( P<0 . 0001 ) than 5 , the number of muscle synergies previously identified in the balance task [19] . Models of muscle synergy control required more control effort , but less computation time during the quadratic programming search , than model MMe ( Figure 10B ) . Using muscle synergy control reduced the computation time by a factor of 8 compared to MMe ( P≪0 . 001 ) whereas control effort increased 2–4 times ( P<0 . 0005 ) . Post hoc analyses revealed a significant contrast between the control effort required for the MMe and SMc models ( P<0 . 05 ) . To test whether model MMe might be predicting unrealistic endpoint moments , MMe optimizations in the preferred postural configuration of each animal were repeated with additional constraints such that the moments at each limb endpoint were limited to zero . This formulation predicted fits to experimentally-observed left hindlimb HF directions that were similar to those of the MMe model ( P<0 . 83 , paired t-test ) . Due to the additional constraints , 5/12 optimizations of cat Ni failed to converge and were excluded . Convergence failures occurred in the same conditions in ten repetitions of these optimizations .
Although prior studies demonstrated that temporal patterns of activation of individual muscles during balance could be predicted from task-level optimal control of CoM dynamics and control effort , they did not address the partitioning of control effort across redundant muscles or limbs . Temporal patterns of individual muscle activity during balance can be predicted from an optimal tradeoff between minimizing CoM excursion and control effort in both humans and cats [23] , [24] . However , previous models of CoM control during balance have eliminated redundancy by examining single-plane movements , as well as by controlling the joints with torques [7] , [54] , [55] , [56] , [57] , [58] , [59] or single muscles [23] , [24] . In contrast , we focused on predicting spatial patterns of activity at the initial timepoint of the CoM feedback response in order to understand the coordination of multiple muscles and limbs across multiple perturbation directions spanning the horizontal plane . Here , we found that detailed patterns of muscle activity and limb forces across biomechanical contexts were predicted from interactions between a common optimization framework – achieving task-level constraints while minimizing effort – and the changing properties of the musculoskeletal system . Prior studies demonstrated that the properties of single-limb biomechanics [20] , [21] were insufficient to predict the force directions observed across multiple postural configurations [14] , [18] , [53] , leaving the role of biomechanics in determining this behavior unclear . These results suggest that control effort costs influence the way that the nervous system distributes effort across the redundant musculature when different combinations of muscles can realize the constraints of the task , and that the characteristic changes in forces observed during the balance task emerge as optimal patterns of distribution are applied in different biomechanical configurations . Moreover , constraints on net CoM mechanics allowed both muscle and limb force redundancy to be simultaneously resolved by minimizing control effort [2] , [13] , [26] , eliminating the need to explicitly minimize limb force [25] , [60] . Our results also demonstrate the feasibility of muscle synergies to produce approximately optimal motor patterns in the context of a detailed model in a realistic motor task . Multiple studies have demonstrated that muscle synergies might be a feasible and effective way for the nervous system to produce movement [61] , [62] , [63] , and that the control of muscle synergies can closely approximate the optimal control of individual muscles , particularly in planar or idealized tasks [37] , [64] , [65] . We found that muscle synergy control was sufficient to achieve the task constraints , in some cases recreating the activation of flexors that was not well-predicted by minimizing the activation of individual muscles . However , in general , solutions from optimal muscle control and muscle synergy control were broadly similar , consistent with the results of other studies [2] , [66] . For example , extensor muscle activity and the limb forces in perturbations for which the hindlimb was loaded were well-predicted whether the activity of individual muscles or of muscle synergies was optimized . Although our study does not resolve the debate over whether low-dimensional muscle activation patterns reflect optimal patterns of individual muscle control or explicit muscle synergy constraints , these results demonstrate the feasibility of muscle synergies for the implementation of near-optimal motor solutions in a realistic motor task . Taken together , the results of this and previous studies are consistent with the idea that the temporal and spatial patterning of muscle activity during the automatic postural response can be well-described by a hierarchical optimal control framework . Hierarchical optimal control is based on the idea that higher levels of the nervous system operate on increasingly abstract variables , such as CoM kinematics , while relying on lower-level controllers to locally control high-dimensional musculoskeletal dynamics [67] , [68] . We hypothesize that the high-level representation is critical because multiple studies have demonstrated that lower-level kinematic variables such as joint angles are insufficient to predict the activation of individual muscles during balance control , whereas CoM kinematics robustly predicts which muscles will be activated [15] , [54] , [69] , [70] , [71] , [72] . Such a hierarchical structure may be required in neural control structures due to neural conduction and computation delays . One idea proposed for the low-level control architectures is that they might implement local feedback control to linearize the nonlinear , fast dynamics of the musculoskeletal system , or implement other regulatory functions [68] , [73] . Our concept of a muscle synergy is proposed as a transformation between high-level task goals and low-level dynamics , that may be parameterized to optimally actuate musculoskeletal mechanics [64] or to provide stability [74] , but not necessarily to function as a controller per se . We speculate that CoM feedback may be used to recruit muscle synergies , and in support of this , a recent study in human balance control demonstrated that CoM kinematics are sufficient to describe the temporal recruitment of postural muscle synergies throughout complex perturbations [75] . Despite the various differences , the similarity between solutions arising from optimization of the activity of individual muscles and optimization of the activity of muscle synergies are consistent with the idea that muscle synergies may reflect mid- or low-level control structures within a general hierarchical optimal control scheme for movement . While control of the CoP was sufficient to explain the results of previous studies that considered a limited range of biomechanical conditions , we were able to compare CoP and CoM as task-level variables by examining their ability to predict individual limb forces across multiple directions of perturbation . Both the CoM and CoP have been proposed as controlled variables for balance control [50] , [70] , but control of CoP involves fewer constraints and is based on the control of vertical and not horizontal limb forces . These candidate control variables have typically been investigated in models of only a single plane of movement [7] , [23] , [24] , [54] , [55] , [56] , [57] , [58] , [59] , where they may make indistinguishable predictions . Here , forces predicted by the two candidate task-variables were similar for the direction of primary limb loading in which lateral forces were small . Predictions of sagittal-plane limb forces were also similar across both models ( MMe vs . MPe ) in the directions across all directions in which the limb was loaded . Given the anisotropic force generation characteristics of the hindlimb [20] , it seemed plausible that the control of vertical forces could be sufficient to determine shear forces as well . However , the models produced qualitatively different horizontal plane forces , suggesting that additional constraints on CoM moment and force were necessary to predict the observed force patterns in a quadruped . It is possible that CoM and CoP control are indistinguishable in sagittal plane balance control in humans where force generation is primarily in the vertical direction [76] , [77] . However , the predictions of CoP control are likely to break down when significant horizontal place forces are required such as in our quadrupedal model , or in medial-lateral human balance control . Further , CoP control in human and robot walking has been limited to quasi-static conditions [57] , [78] , [79] , whereas more dynamic conditions suggest that angular momentum about the CoM due to CoM moments is an important control variable [80] , [81] , [82] , [83] , [84] , [85] . Importantly , these results demonstrate that the observed muscle activity patterns and forces could result from an optimization framework in which task-level goals are specified , independent of individual limb forces . We noted that different cost functions produced qualitatively different patterns of limb forces , demonstrating that the experimentally measured patterns are not simply due to musculoskeletal constraints , but indeed depend upon the nature of the optimization framework . Prior studies have found that multiple cost functions could produce similar results [13] , [86] , suggesting that solutions may be qualitatively determined by biomechanical constraints , independent of any optimization framework or control policy . In contrast , our study and other recent studies demonstrate that some cost functions can be eliminated based on their robustness across a wider range of experimental conditions [12] , [25] . Here , minimization of muscle effort ( MMe ) versus energy ( MMm ) predicted similar horizontal plane forces in the preferred postural configuration , but not in short or long stance configurations . In order to more precisely determine a physiological cost function inverse optimization approaches could be used [25] , [87] , [88] . However , it is unlikely that composite cost functions based on weightings between MMe and MMm [25] , [89] would improve fits to recorded muscle activity ( e . g . absent flexors , SOL recruited rather than MG ) , as both cost functions strongly penalize muscle coactivation . Neither are these differences likely to be resolved using alternative cost functions such as minimization of signal dependent noise , which predicts muscle activity patterns similar to minimization of control effort [90] . To further investigate either the task-level variable or the cost function would require implementation of task-level control within a dynamic musculoskeletal model . Although balance control is a dynamic task , we were able to use a static musculoskeletal model to examine the force-sharing problem at a specific instant in time during the postural response that is most amenable to description by a quasi-static model ( see Methods ) . Here we sought only to reproduce the net CoM forces and moments observed in the initial postural response , which in turn can be predicted by an optimal feedback control model in a low-dimensional biomechanical model [22] , [23] , [24] . Integrating an optimal controller with a realistic musculoskeletal model would allow us to test various optimal control models for dynamic balance control , which might implicate criteria relevant to the balance task beyond the control cost formulations presented here . Specifically , considering the longer time constants required to deactivate versus activate muscle [91] would likely improve model predictions by encouraging activation of the flexors . Similarly , rewarding recruitment of muscles with fast fiber types would likely encourage the ankle extensor function of MG ( primarily fast muscle fibers ) , over that of SOL ( primarily slow muscle fibers; [92] ) . Other criteria such as those related to mechanical stability might also be used to explain the absent coactivation [16] . For example , arm impedance is increased in unstable environments , likely requiring additional coactivation [93] . It is possible that these costs could be incorporated within an optimal control formulation penalizing response time in a tradeoff with costs such as control effort , as optimal control models without fixed terminal time have recently been developed for motor tasks [94] , [95] , [96] . A dynamic model would also allow for further refinement of the task variable . Although we were able to differentiate between CoM and CoP , the current model cannot differentiate between CoM and some other candidate task-level variables – for example , translations of the CoM along the anterior-posterior axis – since a static model ignores inertial contributions such that an equivalent moment can be computed about any point . We consider it unlikely that adding additional detail to either the hindlimb or the forelimb models would appreciably influence the forces predicted here . Based on the high level of similarity in the force production capability between the static hindlimb model used here and previous dynamic models , it is unlikely that including a linearized dynamic model with the mass matrix would appreciably influence the results . Previous linearized and fully dynamic versions of the hindlimb model that include the mass matrix have demonstrated nearly identical force production capability to the static model used here [20] , with force production capability biased along the anterior-posterior axis [74] , [97] . Based on earlier versions of the present model and experimental results , it is also unlikely that including a detailed forelimb model would appreciably influence the predicted forces . A previous model that included forelimbs as hindlimbs with reflected anterior-posterior force production capability did not fundamentally change the forces predicted by model MMe [98] in the preferred and long postures . However , as the stance distance shortens , the geometry of the forelimbs in a real animal becomes increasingly like that of a vertical strut , whereas the hindlimbs remain flexed , breaking the symmetry of the forces between the fore- and hind-limbs Although the fore-hind force asymmetry in the shorter postures was not very pronounced in these particular animals modeled here , in some cases the forelimb forces are not elongated at all [14] , [18] , suggesting that the forelimbs can be very well approximated as vertical struts in these conditions . Significant electrophysiological evidence exists for the neuroanatomical substrates required for the hierarchical , task-level neural control mechanisms investigated by this and other studies . While we and others have demonstrated that muscle activity and movements can be described by mathematical tools like optimization , these techniques do not explain how such relationships and computations are achieved within the nervous system [99] . Importantly , electrophysiological evidence from both cortically-mediated as well as brainstem-mediated motor tasks exists to support the idea that the hierarchical , task-level control frameworks suggested here may describe aspects of the organization of the neural substrates for motor control . For example , electrophysiological evidence demonstrates that task-level variables such as the direction of the limb endpoint are represented in motor cortex during reaching [100] , [101] . Although lesion studies demonstrate that the balance task considered here does not require the cortices [102] , [103] , similar task-level representations are found in brainstem , where neurons in the pontomedullary reticular formation respond equivalently to perturbations of different limbs [104] , [105] . Electrophysiological evidence also demonstrates that increasingly abstract representations of the motor periphery are assembled in increasingly higher levels of the nervous system . For example , higher-level representations of limb length and orientation , rather than individual joint angles , are encoded in the dorsal root ganglia and dorsal spinocerebellar tract [106] , [107] . Muscle synergies may describe how task-level representations are mapped to execution-level activity of motoneurons , via the divergent projections to multiple muscles that have been identified at various levels of the nervous system [108] , [109] , [110] , [111] . For example , both cortical and brainstem neurons project to multiple motoneurons , or to spinal interneurons [112] whose activity has been shown to reflect the patterns of muscle synergies rather than individual muscles [113] . These results support the hypothesis that muscle synergies may be important physiological mechanisms for the implementation of near-optimal motor solutions with a reduced number of controlled variables . The original concept of the muscle synergy hypothesis was that it would offer computational “simplification” due to the large numbers of independent variables that must be simultaneously controlled by the nervous system [114] . In our study , using muscle synergies significantly decreased the search time the optimization algorithm required to identify a motor solution , similar to a previous report [64] . This search time decrease illustrates the possible benefits of a reduced dimension solution space during gradient-based searches , although the computational mechanisms in the nervous system are certainly different than a computer . Stochastic search approaches , for example , might realize less benefit from reducing the dimension of the solution space . Moreover , the results do not imply that the nervous system is re-optimizing the cost function de novo every time the motor task is presented [25] , but instead are consistent with the idea that optimal motor solutions could be refined over the course of motor learning and adaptation . Such refined solutions could be encoded within the nervous system in sparse representations that use small number of neurons at any given time . Sparse representations have been hypothesized to increased storage capacity in associative memories and increased energy efficiency [115] as well as accelerate motor learning . For example , a neural-network model demonstrated accelerated motor learning with decreases in the number of independent neural commands [38] . However , this interpretation may be somewhat controversial , as other evidence demonstrates that sparse motor representations based on muscle synergies may slow the learning of motor tasks for which the library of available muscle synergies is inappropriate [116] . We speculate that muscle synergies implement a transformation from task-level goals to muscle activation patterns that is computationally similar to a lookup table that is assembled over motor learning , the structure of which likely reflects the statistics of the behavioral repertoire as well as the motor system [117] . Similar to the arguments advanced for sparse coding of sensory inputs , we speculate that muscle synergies are reinforced over the course of motor learning through biologically-plausible local learning rules ( e . g . “cells that fire together wire together” ) . Through such learning rules , simple model neurons can learn the principal components of their synaptic input weightings [118] . We speculate that groups of muscles would be reinforced , rather than individual muscles , because the function of individual muscles ( in this case , the output force ) may vary depending on the activity of the other muscles in the limb [119] . We speculate that the increased control effort required when using experimentally-derived muscle synergies versus individual muscles may be physiologically reasonable , particularly if considerations beyond energy efficiency are important in balance control . Whereas prior work demonstrated that similar efficiency could be found by controlling individual muscles or muscle synergies developed from optimality criteria [64] , [65] , we show that controlling experimentally-derived muscle synergies requires additional control effort . Although minimizing energetic cost may be critical in some contexts , particularly in ongoing movement tasks like locomotion over evolutionary timescales [28] , [29] , [30] , we speculate that in discrete tasks like the balance responses presented here strictly effort-minimal solutions may not be necessary . For example , in discrete arm posture tasks , subjects can be cued to maintain high levels of coactivation out of habit even at levels of muscle activation that are considerable proportions of maximal voluntary contraction [120] . The forces observed during balance are well within the boundaries of the absolute musculoskeletal capabilities [21] , and the magnitudes of the individual muscle activations predicted by model MMe were moderate , as proportions of MVC ( notice that the scale maxima in Figure 8 vary between 0 . 002 and 0 . 4 ) . Thus the additional effort cost predicted by muscle synergy control may be physiologically plausible . The fact that experimentally measured co-activation is absent in the MMe model predictions further suggests that the physiological state does not necessarily correspond to the minimum effort solution . We speculate that muscle synergies may be organized to implicitly account for criteria related to the dynamic response described above ( e . g . fiber type , etc . ) . Particularly in balance control , using more than the absolute minimum amount of muscle activation required to achieve stability may be advantageous .
|
The nervous system has the ability to rapidly and flexibly coordinate many muscles and limbs to produce movements . This neuromechanical transformation must robustly achieve motor goals under the changing mechanics of the body and environment , and select one solution amongst many alternatives . What computational principles govern such decisions ? Although optimality principles have predicted features of biological movement in simple models , here we show that this computational principle can robustly predict detailed experimental measures in an unrestrained , whole-body balance task . Detailed patterns of muscle activity and forces across multiple movement directions and body configurations were predicted based on interactions between musculoskeletal mechanics of the limbs , and task-level neural strategy of controlling the CoM mechanics while minimizing control effort . Moreover , similar muscle activity and forces were generated when muscles were coupled together in groups called muscle synergies , reducing the number of independent variables that are controlled . Our work is consistent with the idea that the nervous system may learn to coordinate muscles and limbs by minimizing effort in producing natural movements , and may use approximate solutions based on muscle synergies . Understanding such neural mechanisms may allow us to predict the effects of neural injury and disease on motor function .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"anatomy",
"and",
"physiology",
"neuroscience",
"biomechanics",
"motor",
"systems",
"control",
"engineering",
"computational",
"neuroscience",
"neurological",
"system",
"nervous",
"system",
"physiology",
"musculoskeletal",
"system",
"biology",
"control",
"systems",
"bone",
"and",
"joint",
"mechanics",
"physiology",
"neurophysiology",
"engineering"
] |
2012
|
Optimization of Muscle Activity for Task-Level Goals Predicts Complex Changes in Limb Forces across Biomechanical Contexts
|
Heritable differences in gene expression between individuals are an important source of phenotypic variation . The question of how closely the effects of genetic variation on protein levels mirror those on mRNA levels remains open . Here , we addressed this question by using ribosome profiling to examine how genetic differences between two strains of the yeast S . cerevisiae affect translation . Strain differences in translation were observed for hundreds of genes . Allele specific measurements in the diploid hybrid between the two strains revealed roughly half as many cis-acting effects on translation as were observed for mRNA levels . In both the parents and the hybrid , most effects on translation were of small magnitude , such that the direction of an mRNA difference was typically reflected in a concordant footprint difference . The relative importance of cis and trans acting variation on footprint levels was similar to that for mRNA levels . There was a tendency for translation to cause larger footprint differences than expected given the respective mRNA differences . This is in contrast to translational differences between yeast species that have been reported to more often oppose than reinforce mRNA differences . Finally , we catalogued instances of premature translation termination in the two yeast strains and also found several instances where erroneous reference gene annotations lead to apparent nonsense mutations that in fact reside outside of the translated gene body . Overall , genetic influences on translation subtly modulate gene expression differences , and translation does not create strong discrepancies between genetic influences on mRNA and protein levels .
Many genetic differences among individuals influence gene expression levels . Such regulatory variants are responsible for a large fraction of the variation in disease risk among humans and are also thought to be important for the evolution of phenotypes [1]–[3] . Regulatory variants can be mapped as expression quantitative trait loci ( eQTL ) . Due to the relative ease and low cost of mRNA quantification , most eQTL studies have used levels of mRNA , rather than protein , as a measure of gene expression . The few studies that have used mass-spectrometry to examine genetic influences on protein levels reported surprisingly different genetic architectures for protein and mRNA levels [4]–[6] . For a given gene , many eQTL did not correspond to a protein QTL ( “pQTL” [7] ) and vice versa . Some analyses even suggested that eQTL and pQTL for certain groups of genes have significantly less overlap than expected by chance [5] . While more recent work [8]–[10] has found that eQTL and pQTL are more concordant than seen in the initial studies , numerous discrepancies remain . Together , these results have been taken to suggest that there must be substantial genetic variation acting on posttranscriptional processes . Translation is an important determinant of cellular protein abundance ( [11] , but see [12] ) and the rate of translation was shown to be a better predictor of protein levels than mRNA abundance [13] . Therefore , genetic variants that specifically influence translation are a potential explanation for the reported discrepancies between eQTL and pQTL . Differences in gene expression between individuals can be caused by genetic variants that act in cis or by variants that act in trans [2] . Variants that act in cis influence the expression of alleles to which they are physically linked . In a diploid organism , cis acting variants can be detected as preferential expression of one allele compared to the other ( “allele-specific expression” , ASE ) [14]–[19] . By contrast , trans acting variants influence the expression of both alleles of a gene to a similar extent . Both cis and trans acting variants might have effects on translation . To affect translation in cis , a variant needs to reside within the mRNA of the given gene . By contrast , genetic variation in the various translation factors [20] might influence translation in trans . Further , mutations in ribosomal proteins can lead to highly specific differences in translation of small groups of mRNAs during mouse development [21] , suggesting that genetic differences in genes beyond classic translation factors could affect translation in trans . In this paper , we explored the influence of genetic variation on translation . We measured genome-wide translational activity in two genetically different strains of the yeast S . cerevisiae – the laboratory strain BY and the wine strain RM – as well as their diploid hybrid . Translation was measured by massively parallel sequencing of “ribosome footprints” , i . e . of mRNA fragments that are associated with translating ribosomes [13] , [22] . By comparing the footprint data to measures of mRNA abundance gathered in parallel , we determined translation-specific influences on gene expression . In what follows , we distinguish three quantities . “mRNA abundance” quantifies RNA fragments from the polyadenylated transcriptome , irrespective of whether these molecules are translated . We denote as “footprint abundance” the number of RNA fragments bound by ribosomes , which is a measure of the total protein production for the given gene [13] . Finally , we refer to the ratio of footprint abundance to mRNA abundance as “translational efficiency” ( TE ) [13] ) . TE measures the extent to which the mRNA molecules of a given gene are translated . We found that the differences in footprint abundance between BY and RM were highly correlated with the differences in mRNA abundance , both when comparing the parents and for ASE in the hybrid . Against this largely concordant backdrop , there were a small number of genes with evidence for strong translation-specific genetic effects on their expression , and hundreds of genes with more modest effects .
We used ribosome profiling and mRNA sequencing to compare genome-wide patterns of translation in protein coding regions between the BY and the RM yeast strains . Alignment statistics are presented in Supplementary Tables S1 & S2 and discussed in Supplementary Note 1 . There was excellent agreement between our measures in BY and those obtained by re-aligning the reads from a published yeast ribosome profiling dataset [13] ( Supplementary Figure S1 ) . This agreement is in spite of different growth media , slightly different strain backgrounds , several minor differences between library protocols , and substantially deeper sequence coverage in the current dataset ( Methods ) . mRNA abundance across the 6 , 697 genes annotated as coding or potentially coding in the yeast genome database varied by 3–4 orders of magnitude for the central 95% of genes . Footprint abundance spanned 4–5 orders of magnitude and , as expected , was highly correlated with mRNA abundance ( Figure 1 ) . TE varied by ∼100 fold among the central 95% of genes , in line with previous observations in yeast [13] . Across genes , footprint abundance increased more rapidly than mRNA abundance such that genes with high mRNA abundance tended to have higher TE , while genes with lower mRNA abundance tended to have lower TE ( Figure 1 ) . This pattern is reminiscent of coordinated changes in mRNA levels and translation rates of yeast genes in response to diverse environmental stressors [23]–[26] as well as at steady state [27] . At the extreme end of this distribution were open reading frames ( ORFs ) categorized as “dubious” in the yeast genome database ( Figure 1 ) . Low or absent translation for dubious ORFs is consistent with the definition of ORFs in this category as “unlikely to encode an expressed protein” ( www . yeastgenome . org ) . Together , these observations suggest that our data are of high quality and recapitulate known aspects of gene expression and translation . mRNA and footprint levels in BY were highly correlated with those in RM ( Figure 2A–B ) . Consequently , while 54% and 58% of genes had significant ( binomial test , Bonferroni corrected p<0 . 05 ) mRNA and footprint differences between the strains , more than 90% of these differences had small magnitudes of less than 2-fold ( Table 1 & Figure 2A–B ) . While we chose to employ the stringent Bonferroni multiple testing correction for the analyses presented in the main text , similar patterns were seen when using a more permissive threshold based on the false-discovery rate ( FDR [28] ) ( Supplementary Table S3 ) . That the majority of genes shows significant expression differences between BY and RM is in line with published estimates for mRNA levels – for example , 69% of genes were differentially expressed in a microarray based experiment [29] . To explore differences in translation between strains , we compared the mRNA differences to the footprint differences . The magnitudes of these differences were correlated ( Spearman's rho = 0 . 71 , Figure 3A ) , and this correlation became stronger when restricting the analyses to genes with a significant mRNA or footprint difference ( rho = 0 . 77 , Figure 3B ) . As expected from this correlation , genes with a significant mRNA difference were highly likely to also have a significant footprint difference ( Fisher's exact test ( FET ) : odds ratio = 4 . 6 , p<2 . 2e-16 ) and the direction of differences agreed for 82% ( 3 , 108 out of 3 , 798 ) of genes with a significant mRNA and/or footprint difference . Thus , a gene that differs significantly in mRNA abundance between BY and RM typically also has a significant footprint difference in the same direction . For most genes , translation carries forward mRNA differences to differences in protein synthesis . While a more conservative [30] significance testing method ( the DESeq package , [31] ) yielded substantially fewer differentially expressed genes ( Table 1 ) , the high agreement between mRNA and footprint differences remained intact ( Supplementary Note S2 and Supplementary Figure S2A ) . These observations leave open the possibility that translation may exert subtle , quantitative effects . Indeed , 42% of genes showed significant ( Bonferroni corrected G-test p<0 . 05 ) differences in TE , i . e . , at these genes the ratio of footprint levels between strains differed from the respective mRNA ratio . As expected given the concordance between mRNA and footprint differences noted above , most of the TE differences were of small magnitude ( Figure 2C & Table 1 ) . Thus , while translation typically does not override mRNA differences between BY and RM , for many genes it subtly alters the degree to which mRNA differences are reflected at the level of protein synthesis . To investigate the extent to which translation is influenced by cis-acting variants , we gathered ribosomal footprint and mRNA data from the diploid hybrid between BY and RM and compared the expression of the two alleles . Reproducibility of the allele-specific measurements was high , as judged by comparing two biological replicates processed at the same time ( Supplementary Figure S3 ) . As expected , the number of genes with significant ASE was less than the number of genes with differences between the BY and RM parental strains ( Figure 2 & Table 1 ) : we detected significant ( Bonferroni corrected binomial test p<0 . 05 ) ASE for the mRNA levels of 6% of genes and for the footprint levels of 6% of genes . The mRNA estimate is lower than previous estimates for these two yeast isolates ( ∼14% [17]–∼20% [32] ) because of the stringent Bonferroni correction we applied here; when we use a FDR-based cutoff we obtain higher numbers of genes with significant ASE ( Supplementary Table S3 ) . Variants that act in cis to alter translation should result in footprint ASE that is not a direct reflection of mRNA ASE . By contrast , we found that the magnitudes of mRNA and footprint ASE were correlated ( rho = 0 . 72 ) when considering genes with significant mRNA or footprint ASE ( Figure 3E & Supplementary Figure S2B ) . Consequently , genes with significant mRNA ASE were very likely to also have significant footprint ASE ( FET odds ratio = 25 , p<2 . 2e-16 ) and 83% ( 260/312 ) of genes with significant mRNA or footprint ASE agreed in the direction of ASE . Thus , allele specific footprint levels mostly reflect allele specific mRNA expression , suggesting that cis acting variants with strong effects on translation are rare in BY and RM . To search for genes that may carry cis acting variants that influence translation , we tested for allele-specific TE , i . e . for genes where the ratio of mRNA ASE differs from the ratio of footprint ASE . Significant ( G-test , Bonferroni corrected p<0 . 05 ) allele-specific TE was found for 3% ( n = 106 ) of genes ( Table 1 ) . While most of these effects had small magnitude ( Figure 2F ) , 26 genes had significant allele-specific TE greater than 2-fold ( Table 2 ) . For three of these genes , allele-specific TE is due to premature translation termination in one strain compared to the other ( Table 2; see Supplementary Table S4 and Supplementary Figure S2C for genes with allele-specific TE differences identified using DESeq ) . The remaining 23 of these 26 genes appear to carry cis acting variants that substantially alter the translation rate without disrupting gene structure . The overall contribution of cis vs . trans acting variants to gene expression variation can be determined by comparing differences in expression between two strains to allele-specific expression in their diploid hybrid [14] , [17] . For genes that are entirely regulated in cis , the parental difference should be completely recapitulated by allele-biased expression in the hybrid . For genes that are entirely regulated in trans , expression of the two alleles should be the same in the hybrid , irrespective of the parent difference . Consequently , in the absence of noise , if all genes in the genome were exclusively affected by cis-acting variants , the slope of the relationship between allelic differences and parental differences should equal one , and if all genes were exclusively affected by trans-acting variants , the slope should equal zero . In our data , the slopes of these relationships ( calculated using major axis estimation [33] ) were 0 . 35 for mRNA and 0 . 37 for footprint abundance ( Figure 4A ) . Because the data are noisy , we used bootstrap analysis to show that these estimates are significantly different from zero or one ( Figure 4C ) , implying the presence of both cis- and trans-acting variants . In order to avoid potential biases associated with different quantifications ( all mapped reads in parents vs . only reads overlapping SNPs in the hybrid ) , the parental fold changes used in these analyses were based on allele counts at the same set of SNPs used to analyze the hybrid . Bootstrapped distributions of the mRNA and footprint slope estimates overlapped substantially ( Figure 4C ) . Thus , the relative contributions of cis- and trans-acting variants on mRNA abundance are faithfully represented in footprint abundance . Next , we asked if the effects of previously identified eQTL ( i . e . , individual loci identified through their effects on mRNA levels ) are reflected in our data . We stratified genes according to whether they were influenced by a local ( likely cis-acting ) eQTL , a distant ( likely trans-acting ) eQTL , or by both types of eQTL , using eQTL reported in [29] . Genes with local eQTL had a significantly steeper slope ( the 95% confidence intervals from 1 , 000 bootstraps did not overlap ) than genes with distant eQTL ( Figure 4B & C ) . Thus , the effects of known genetic variants are recapitulated in our mRNA data . The effects of known eQTL were also seen in footprint abundance ( Figure 4B & C ) . Together , these analyses suggest that the relative importance of cis vs . trans acting genetic variation on footprint abundance is largely similar to that on mRNA abundance . There are several possible relationships between mRNA differences and footprint differences ( Figure 5A ) . First and most obviously , mRNA and footprint differences can be statistically indistinguishable; these genes fall near the diagonal in Figure 5A and will not be considered further . When TE differs significantly between BY and RM , this can have the following effects . A significant mRNA difference can correspond to a larger ( reinforced ) or smaller ( buffered ) significant footprint difference in the same direction , no significant footprint difference ( complete buffering ) , or a footprint difference in the opposite direction ( “inverted” genes ) . Finally , a significant footprint difference can appear in the absence of a significant mRNA difference ( “FP only” genes ) . To examine the relative frequencies of these scenarios , we partitioned the genes with significant TE differences between BY and RM ( “TE genes” , 42% of genes using the Bonferroni significance cutoff ) ( Table 3 , Figure 3C & Figure 5B & C ) . For 5% of the TE genes , neither the mRNA nor the footprint difference was significant , providing little further information . Thirty-one percent of the TE genes showed reinforcement and 27% of the TE genes had a footprint difference in the absence of an mRNA difference . Conversely , 29% of the TE genes showed partial or complete buffering . Only 7% of the TE genes fell in the “inverted” category . The results for the hybrid data are given in Table 3 and Figure 5B & C . Results based on a less stringent FDR significance cutoff as well as those based on the more conservative DESeq significance testing framework [31] are shown in Supplementary Table S5 . Recent work comparing the two yeast species S . cerevisiae ( two isolates of which we study here ) and Saccharomyces paradoxus reported an excess of directional effects of translation . An mRNA difference between species was typically accompanied by a difference in translation in the opposite direction , resulting in a smaller or inverted footprint difference [34] , [35] . We sought to test if this excess also exists between BY and RM . Because the results of this inference could depend on the precise choice of how mRNA differences and footprint differences are compared , we systematically tabulated the four possible comparisons of directional effects of translation ( Figure 5 and Tables 3 & 4 ) . These four comparisons arise by including or excluding the “FP only” genes , and by including or excluding the “inverted” genes . The detailed results are presented in Supplementary Note S3 . Taken together , they suggest that variation in translation between BY and RM more often increases than decreases footprint differences compared to mRNA differences , although this conclusion depends on the precise way in which TE genes are compared . To ensure a consistent comparison with these results between BY and RM , we reanalyzed the inter-species results provided in [34] as well as p-values and fold-changes kindly provided by the authors of [35] ( Table 4 , Figure 5B & C , Supplementary Table S6 ) . In line with the published results for the McManus et al . dataset [34] , there was a strong excess of opposing over increasing effects of translation in both the parent and the hybrid data . This effect was robust to the precise choice of tested gene groups ( Table 4 ) . In the hybrid data from Artieri & Fraser [35] , we replicated the reported excess of opposing genes when compared against reinforced genes . This result was less robust to the precise choice of comparison ( Supplementary Note S3 ) . Overall , the signal of translational buffering is strong in one of the two published datasets and more dependent on the precise analyses in the other dataset . Discussion of additional analyses in the interspecies comparisons is provided in Supplementary Note S3 and Supplementary Figure S4 . Next , we asked to what extent the mRNA and footprint differences between BY and RM correspond to differences in protein abundance that are due to individual genetic loci . We recently used a bulk-segregant approach to map trans–acting pQTL in BY and RM [8] . The high statistical power of that approach resulted in the identification of multiple pQTL for many of the analyzed genes . Of the 160 genes in our earlier study , 114 could be analyzed in the present parent and hybrid datasets . For each of these 114 genes , we summed the allele frequency differences at the pQTL , a measure related to the genetic effects of the pQTL ( Methods ) . These summed measures provide a rough expectation of protein level differences between BY and RM that are due to trans-acting variation . As expected given that all the pQTL considered here act in trans , there was no correlation between the predicted protein differences and ASE in the hybrid data ( Figure 6C & D ) . By contrast , we found that the summed pQTL effects correlated significantly with differences in mRNA abundance between BY and RM ( Spearman rank correlation rho = 0 . 36 , p = 0 . 0001; Figure 6A ) . Thus , the aggregate effects of loci that were detected through their effects on protein levels are reflected in mRNA differences between strains . There was a slightly stronger correlation between the summed pQTL effects and the strain differences in footprint abundance ( rho = 0 . 46 , p = 3e-7; Figure 6B ) , but this difference in the strength of correlations appeared to be primarily driven by a few genes ( 112 of 1 , 000 bootstrapped datasets showed a larger mRNA than footprint correlation ) . These observations further support the hypothesis that most pQTL arise from genetic influences on mRNA levels , perhaps augmented by minor additional contributions by genetic effects on translation . The single base pair resolution of ribosome-profiling data permits detailed examination of footprint abundance along the length of a gene , and of how these patterns are affected by genetic variation . In particular , variants that create or disrupt a stop codon are of interest due to the potentially large phenotypic consequences . In addition , insertions and deletions ( indels ) that lead to a shift in the reading frame of a coding gene will usually result in premature translation termination or , more rarely , in translation proceeding beyond the original stop codon . All these types of variants should lead to detectable differences in the pattern of footprint coverage . Among our set of high quality coding SNPs in 3 , 376 genes that had a status of “verified” according to the SGD database , we identified 18 sites where RM relative to BY has gained a premature stop codon and 10 sites where RM has lost the annotated stop codon ( Supplementary Data S1 ) . In addition , we catalogued 32 short indels predicted to lead to a frameshift . We visually examined the footprints patterns in BY and RM at these sites . Data tracks showing these patterns across the genome are available for interactive browsing in the UCSC browser under the URL http://genome . ucsc . edu/cgi-bin/hgTracks ? db=sacCer3&hubUrl=https://labs . genetics . ucla . edu/kruglyak/trackhub/hub . txt . Of the 18 gained stops in RM , three were in genes that were not expressed and could not be analyzed . Of the 15 sites in expressed genes , nine led to clearly visible premature termination ( Figure 7A for an example ) . Four putative premature stop codons were located in a part of the ORF that , while annotated as part of the coding sequence , is in fact upstream of the region we found to be translated in both BY and RM ( Figure 7B ) . These four sites therefore do not affect the protein , but reflect errors in gene annotation . The remaining two sites were situated in the translated part of the coding sequence , but did not lead to a visible reduction in translation . Closer inspection of these two SNPs showed that both of them are part of multi-base substitutions that together lead to an amino acid substitution instead of a nonsense mutation . Thus , only 60% ( 9/15 ) of our list of predicted nonsense mutations in expressed genes had detectable effects on protein sequence . We note that six of these truncating mutations were close to the 3′ end of the coding sequence , where they may be less likely to severely disrupt protein function [36] . Of the ten sites where a BY stop codon was absent in RM , three resided in genes with no or very low expression . Four did lead to visible ribosomal readthrough , and two of these sites are in fact known instances of difference in primary protein structure between BY and RM . For example , the gene NIT1 and the gene annotated to lie immediately downstream ( YIL165C ) form a single ORF in other yeast species [37] and other S . cerevisae strains [38] . The two genes are annotated as two separate ORFs because the yeast genome annotation is primarily based on BY , which carries a premature stop codon inside the ORF ( Figure 7C ) . The remaining three lost stop codons did not visibly result in translational readthrough . Closer examination of the sequence context reveals that these codons are immediately followed by a secondary stop codon that compensates for the lost stop codon [39] . We made similar observations for the 32 putative frame shifting indels: ten were in genes with low expression , five were in untranslated regions erroneously annotated as coding , three were in repetitive regions and may be due to alignment errors , six were close to the end of the ORF , one was downstream of a premature stop and therefore of no consequence itself , and only seven led to visible early termination or extension of the frame-shifted protein .
We used ribosome profiling [13] to explore how genetic differences between the two yeast strains BY and RM influence mRNA abundance and translation . We found that most genes with significant differences in mRNA levels had footprint differences in the same direction . Thus , translation typically carries forward genetic influences on mRNA levels into differences in protein synthesis . While we did detect hundreds of genes that showed evidence for genetic effects on translation , most of these effects subtly modulate rather than override mRNA differences . Genetic variants that induce strong , specific effects on translation appear to be infrequent in BY and RM . We made similar observations in the hybrid between BY and RM . Significant allele-specific mRNA expression was highly correlated with allele-specific footprint abundance . Therefore , with a few exceptions ( e . g . those listed in Table 2 and Supplementary Table S4 ) , most genes do not carry cis-acting variants that have large , specific influences on translation . By comparing the parental differences to ASE in the hybrid [14] , we found that the relative contribution of cis- vs . trans-acting variants on footprint levels was similar to that on mRNA levels . Further , individual local and distant eQTL that had earlier been identified based on their effects on mRNA levels [29] influence the cis vs . trans contribution in both the mRNA and footprint data presented here . These eQTL therefore are carried forward to translation and would be expected to also affect protein levels . Analyses of a mass spectrometry dataset have reported substantial discrepancies between genetic influences on mRNA and protein differences between BY and RM [4] , [5] . Our ribosome profiling data provides little evidence that genetic effects on translation might be responsible for these discrepancies . This observation is in line with recent pQTL studies in yeast that leveraged improvements in protein measurements and experimental design [8] , [9] , [40] and found that eQTL and pQTL are not as discordant as reported previously . To the extent that the remaining discrepancies between eQTL and pQTL are real ( as opposed to , for example , due to experimental variation [41] ) , our results here suggest that they are more likely caused by genetic influences on protein degradation rather than on translation . Two recent papers examined the evolution of mRNA and footprint levels between the yeast species S . cerevisiae and S . paradoxus [34] , [35] . Both studies reported that mRNA differences are more often opposed than reinforced by translation . Motivated by these reports , we conducted similar analyses in our data . We found some evidence that genes with strain differences in TE between BY and RM tend to more often have footprint differences larger than the corresponding mRNA differences , the opposite pattern of what was reported for the species comparisons . A similar pattern was recently observed for allele-specific translation in Candida albicans [42] . However , in BY/RM , this pattern was dependent on the precise fashion in which the analysis is performed ( Table 4 & Supplementary Note S3 ) , in line with the observation that the mRNA and FP differences are similar in magnitude ( Figure 3 ) . Given that this inference in BY and RM was dependent on the exact way in which TE genes are grouped , we performed the same set of comparisons in the two published interspecies data sets . In one of the two studies , the reported excess of opposing effects of translation was robust across comparisons , while in the other study the results were more ambiguous . In our opinion , none of the groupings of the TE genes we used to compare the directional effects of translation is obviously more correct than the others . For example , while it would be our preference to include genes with a footprint but no mRNA difference in the analyses , these genes were excluded in the published inter-species analyses that only analyzed genes with a significant mRNA difference [34] , [35] . It is also unclear whether ( and where ) genes with a footprint difference that is inverted compared to the mRNA difference should be included . In addition , differences in the precise experimental protocols between studies may contribute to the different results . For example , the technical variance in footprints is typically higher than that in mRNA , and also differs among the different datasets ( Supplementary Figure S5 ) . We are thus hesitant to draw strong conclusions about the relative importance of opposing/buffering or reinforcing/increasing effects of translation within and between yeast species , although we cannot rule out genuine evolutionary differences between these intra- and inter-specific comparisons . Which cellular mechanisms might explain the observed cases where translational differences reinforce , buffer or invert an mRNA difference ? The simplest explanation is that these cases involve two or more variants: one altering mRNA levels and another altering translation . Alternatively , more parsimonious explanations might involve a single mutation that affects both mRNA levels and translation rates . There is growing evidence for coordination among the stages of gene expression [43] . For example , positive correlations between mRNA abundance and translation rates have been observed during unperturbed growth [27] , as well as for mRNA abundance changes in response to various stressors [23]–[26] . Recent evidence suggests that in addition to transcription , promoter sequences influence the subcellular localization and translation rates of yeast mRNAs [44] . While the precise mechanisms that mediate these coordinated effects are not fully understood , there is some evidence that the transcription machinery can influence the translational fate of mRNAs through the RNA polymerase II subunits Rpb4 and Rpb7 [45] . Translation can also stabilize mRNA molecules by protecting them from degradation [27] , [46] , so that a higher translation rate per se can result in higher mRNA levels at steady state . A sequence variant that increases TE of a given gene could then not only result in higher footprint levels but also increase mRNA levels , even if the variant has no effect on transcription . Careful study of the dynamics of translation ( e . g . [47] ) will be needed to further address this question . Our analyses of nonsense and frameshift polymorphisms showed that these variants indeed result in detectable differences in translation . However , the results serve as a reminder to exercise caution when interpreting the potential functional impact of variants identified in next generation sequencing datasets , especially for variants with putative large effects [36] . Sequence context ( e . g . secondary stop codons downstream of a lost stop [39] ) and multi-base substitutions can obscure the true consequences of a variant called from a high-throughput pipeline when considered in isolation . Further , even in an extremely well annotated genome such as that of S . cerevisiae , errors in gene annotation can generate the illusion of severe differences in protein sequence between strains when in fact the corresponding variants reside outside of the coding region . Our list of variants between BY and RM with validated effects on translation as well as of problematic gene annotations ( Supplementary Data S1 ) can be useful to assess the consequences of genetic differences between these yeast strains . We have also made available tracks in the UCSC genome browser [48] that allow easy visualization of translation patterns between BY and RM for any gene ( http://genome . ucsc . edu/cgi-bin/hgTracks ? db=sacCer3&hubUrl=https://labs . genetics . ucla . edu/kruglyak/trackhub/hub . txt ) . Molecular phenotypes such as mRNA and protein levels ( as well as others [49] , [50] ) provide crucial intermediates for connecting DNA sequence variation to organismal phenotypes . New measurement technologies will allow an increasingly fine-grained view of the mechanistic connections between the levels of molecular traits and illuminate how genetic variation shapes organisms .
We studied the same strains as in Bloom et al . [51] . The common laboratory BY strain we used had mating type MATa . The RM strain was originally isolated from a vineyard . Our RM strain had genotype MATα hoΔ::hphMX4 flo8Δ::natMX4 AMN1-BY . Both strains were prototrophic , i . e . they did not carry any engineered deletions of metabolic genes . These deletions are commonly used as genetic markers that can have strong effects on gene expression [1] . The haploid parental strains were crossed to generate the diploid hybrid . BY and RM differ in cycloheximide resistance at a dose several orders of magnitude lower than those used in the ribosome profiling protocol [13] , [51] . To confirm that the parents and the hybrid were equally sensitive to the high cycloheximide dose used here to block translation , we attempted to grow them at 30°C in triplicates in liquid yeast nitrogen base ( YNB ) medium with a range of cycloheximide concentrations centered on the dose used in the ribosome profiling protocol . While growth was normal in negative controls without cycloheximide , there was no growth within 24 hours in any of the cycloheximide doses tested . Libraries for RNA-seq and ribosome profiling were prepared as described in [13] , with the following exceptions: ( 1 ) cells were cultured in YNB , ( 2 ) the reverse-transcription step was primed by ligating miRNA Cloning Linker 1 ( IDT ) onto the RNA fragments , and ( 3 ) highly abundant rRNA species were hybridized to biotinylated oligos and subtracted using streptavidin-coated DynaBeads ( Invitrogen ) as in [52] . Deep sequencing was performed on the Illumina HiSeq 2000 platform . Raw reads are available in the NCBI Gene Expression Omnibus under accession GSE55400 . We employed a set of filters to ensure unbiased estimates of ASE . We used the program BWA [53] to align high coverage ( >50× ) 94 bp paired-end whole genome Illumina sequencing data from the BY and the RM strains used in this study [51] to the reference yeast genome version sacCer3 downloaded from the UCSC genome browser ( http://genome . ucsc . edu ) . We used a custom python script kindly provided by Martin Kircher to remove PCR duplicates . Samtools [54] was used to extract a preliminary set of SNPs with variant quality score >30 and with an estimated alternative allele frequency of 1 ( “AF1 = 1” flag in the vcf file ) . Next , we retained only biallelic SNPs where our RM strain carries an alternative allele and our BY strain carries the genome reference allele . There were 43 , 154 SNPs in this initial set . We sought to restrict this set to those SNPs where short sequencing reads ( such as those obtained in ribosome profiling ) can be aligned to unique positions in both the BY and the RM reference genome . For each SNP , we extracted the 30 bp up- and downstream sequence from the BY genome reference ( sacCer3 ) , from both the plus and the minus strand . The SNP allele itself was set to the RM allele . The resulting 61 bp sequences were aligned to the RM reference genome downloaded from the Broad Institute ( http://www . broadinstitute . org/annotation/genome/saccharomyces_cerevisiae . 3/Info . html ) using BWA [53] . We removed SNPs whose flanking sequences mapped to more than one position in the RM genome as well as SNPs where multiple SNPs mapped to the same position in the RM genome . The number of SNPs after these filters was 38 , 706 . Next , we sought to remove SNPs with alignment biases towards one or the other reference genome by examining the alignment behavior of publicly available DNA sequence data obtained from a BY/RM hybrid [55] . Any allelic bias seen in hybrid DNA sequences necessarily is of technical origin and indicates problematic SNPs . We trimmed the hybrid DNA reads to 30 bp single end , aligned them to both the BY and RM reference genomes and counted the number of reads that overlapped the BY or RM reference alleles at each SNP , exactly as described below for our mRNA and footprint reads . To identify SNPs with allelic bias beyond that expected by chance , we simulated an unbiased dataset as follows . For each SNP , we generated allele counts assuming a binomial distribution with p = 0 . 5 , and at a depth of coverage drawn from the observed data . We determined criteria for SNP exclusion based on visual comparison of the observed hybrid DNA dataset to the simulated unbiased data . We removed SNPs with very high ( >100 fold ) and very low ( <30 fold ) coverage , as well as any remaining SNPs where the frequency of the BY ( and , equivalently , the RM ) allele was less than 0 . 3 or more than 0 . 7 . After these filtering steps , 36 , 089 SNPs remained . We noted a population of SNPs with hybrid DNA allelic ratio centered at ∼1/3 , i . e . a 2∶1 bias towards the RM genome . Further inspection revealed that these SNPs all resided in regions where DNA sequencing coverage in our RM parent was twice as high as that in our BY parent . Nearly all of these regions were situated at chromosome ends and likely reflect segmental duplications of these distal regions in the RM strain compared to the BY reference genome . These regions extended for several kb and contained annotated protein coding genes , in line with the recent observation that subtelomeric regions contain large structural variants that segregate among wild yeast [56] . We visually examined the coverage across our BY and RM parent DNA sequences and excluded any regions with evidence for segmental duplications in the RM but not the BY parent . This removed 821 SNPs , for a remaining set of 35 , 268 . Finally , because we are interested in quantifying expression of protein coding genes , we retained only the 23 , 412 SNPs in ORFs annotated in the SGD database ( www . yeastgenome . org , accessed on 06/28/2013 ) . SNPs in ORFs annotated as overlapping on the same strand were removed . However , because the mRNA and footprint data are strand-specific , we were able to retain 395 SNPs that overlap ORFs on different strands for a total of 23 , 807 quantifiable positions in 4 , 462 ORFs ( Supplementary Data S2 ) . Because the reference yeast genome is based on a strain with the BY background , sequence differences between the reference and RM make read alignments from an RM sample more difficult , especially with short reads such as the ∼32 base pair ( bp ) ribosomal footprint fragments . To counter this problem , we implemented a computational pipeline that uses “personalized” genome references for the BY and the RM strain to allow unbiased read mapping . Prior to mapping , we removed sequences corresponding to the Illumina adapter sequence ( CTGTAGGCACCATCAAT ) opposite the sequencing priming site and discarded all reads that did not contain these adapter sequences . We also removed the first base from each read as these often corresponded to adenosines introduced during ligation in the library preparation protocol . The trimmed reads were mapped using BWA [53] as follows ( see Supplementary Table 1 for alignment statistics ) . For the comparisons between the BY and RM strain , reads can be considered irrespective of whether they cover a SNP or not . Reads from the BY strain were mapped to the BY reference genome ( version sacCer3 ) . Reads from the RM strain were mapped to a modified version of the BY reference where the 43 , 154 SNPs between BY and RM as described the section above were set to the RM allele . The rationale for using this strategy was to maximize the number of RM reads that can be mapped to the BY reference without penalizing reads that contain a sequence difference between BY and RM , while still being able to directly use the BY gene annotations . We counted only uniquely mapping reads on the correct strand in genes . For the ASE analyses , we are only interested in reads that span a SNP between the BY and RM strains . We noted that the short reads produced in ribosome profiling are heavily biased against mapping RM reads to the BY reference ( not shown ) . We therefore mapped all reads to both the BY reference and the RM reference available from the Broad Institute . We considered only reads that mapped to one of these two reference sequences uniquely and without mismatch . This strategy guarantees that reads that span a sequence difference between BY and RM can be unambiguously assigned to the parental chromosome they originated from . At each of the 23 , 807 high quality coding SNPs ( s . section above ) , we counted the number of reads that mapped to the correct strand of the BY or the RM genome . When a read overlapped multiple closely linked SNPs , it was randomly counted towards one of them . Because we excluded reads with mismatches , our strategy excludes all reads with sequencing errors . For comparison of ASE in the hybrid to differences between the parent strains , we re-mapped the BY and RM parent reads and quantified allele-specific expression as described in this section . For determining the genomic source of reads in the libraries ( ORFs , UTRs , ncRNAs , etc . ) as well as for the comparison between the parent strains reported in the main text , we used htseq-count ( http://www-huber . embl . de/users/anders/HTSeq/doc/overview . html ) and annotations extracted from SGD ( www . yeastgenome . org ) . For the analyses of allele specific expression , we added the allele counts for all SNPs in a gene . For the hybrid , we summed the counts from the two replicates , with the exception of reproducibility analyses . While all statistical analyses were performed directly on count data ( s . below ) , the figures show gene abundance as log10-transformed fractions of total counts for the given sample . Translation efficiency ( TE ) for a gene was calculated as the difference between the log10-transformed mRNA fraction and the log10-transformed footprint fraction . All quantifications , both for whole ORF and SNPs , are available in Supplementary Data S2 . All statistical analyses were performed in the R programming language ( www . r-project . org ) . Unless stated otherwise , we calculated slopes using major axis estimation [17] , [33] as implemented in [57] . Correlations were calculated as nonparametric Spearman's rank correlations to avoid making assumptions about the distributions of the data . We used two different count-based approaches to gauge statistical significance . In the main text , we report the results from binomial tests while in Supplementary Note S2 we describe results obtained with the DESeq analysis framework [31] . Count-based binomial tests have higher power when the absolute number of counts is high . The number of reads mapped was different between different samples and between different data types . Specifically , the parental footprint libraries had 30%–70% more reads than the parental mRNA libraries ( Supplementary Table S1 ) . We removed these differences in total read counts by downsampling as follows , largely following the procedures described in [58] . For each comparison ( parental analyses based on all mapped reads or allele-specific analyses counting only reads spanning SNPs ) , we identified the sample with the lowest total number of counts . This sample remained as observed , while in the other samples we randomly sampled read counts to match the smallest library size . We call this dataset the “downsampled” dataset and the unadjusted data prior to downsampling the “raw” dataset . To test for differential expression of mRNA or footprints , we generated a second datatset from the downsampled data using hypergeometric resampling [58] . The goal is to avoid differences in power to call differential mRNA or footprint expression when the absolute counts are different for these two data types . For example , if a gene is highly transcribed but has a low rate of translation , the mRNA data might have more counts and therefore higher power . For each gene , we grouped the samples into pairs of samples to be compared . These pairs are the counts from the two alleles in the hybrid , or the counts from the respective parental samples . mRNA and footprint counts formed separate pairs . For each gene , we identified the pair with the smallest sum of counts . This pair remained as observed . For the other pairs , we used hypergeometric sampling to generate data with the same sum of counts as the pair with the lowest sum of counts . We call the resulting dataset the “hypergeometric” dataset . We removed all genes from the analyses where , in the hypergeometric dataset , any pair had a sum of counts <20 ( following [58] ) , and all genes where any individual sample had a count of zero ( these two criteria are not redundant because in a few cases , one member of a pair had >20 counts while the other had zero counts ) . Genes that did not satisfy this filter were also excluded from the downsampled dataset . After filtering , there were 5 , 316 and 3 , 342 genes available for analysis in the parent dataset and the SNP-based hybrid dataset , respectively . No genes were removed from the raw dataset . For the analyses of reproducibility in the hybrid we performed the same down-sampling approach but kept the two hybrid replicates as separate samples . We tested for differential mRNA of footprint expression using binomial tests on the “hypergeometric” dataset . For most analyses the p-values were adjusted for multiple tests using Bonferroni correction . We also estimated false discovery rates by calculating q-values [28] , [59] . For DESeq ( s . below ) , multiple tests were corrected using the Benjamini-Hochberg procedure [60] that is the default in DESeq [31] . To test for differential TE , we used G-tests on the “downsampled” data to test if the ratio of footprint and mRNA counts differed between strains . This test cannot be performed in the “hypergeometric” data because in these , mRNA and footprints have been sampled to the same level , overriding the TE signal . As an independent approach to test for statistical significance , we used the DESeq analysis framework [31] . DESeq models the counts using a negative binomial distribution and asks if , for a given gene , the observed mean difference between strains is more than expected given the variance for a gene of the given abundance . As such , DESeq takes into account the fact that more highly expressed genes have higher counts and therefore higher statistical power than less abundant genes . We used DESeq version 1 . 16 on raw count data , i . e . without any prior normalization , but excluding genes where all samples had a count of zero . In the parent data and hybrid , 6 , 697 and 4 , 462 genes were available for analysis in DESeq . To analyze the BY and RM parent data , we calculated dispersion factors using the method = “blind” option together with the sharingMode = “fit-only” option to approximate experimental ( noise ) variance by treating the BY and RM samples as “replicates” as recommended in the DESeq manual . This procedure is known to be conservative [31] . In the hybrid data , we used the allele counts from the two replicates separately ( without any normalization , and without summing the replicates ) and compared them using the default DESeq settings . Further , in the hybrid data it is possible to use DESeq to test for differential TE by using the nbinomGLMTest ( ) function to compare models specified by the formulae:andWhere count is the count data , molecule is either mRNA or footprint , strain is either BY or RM , and molecule:strain indicates an interaction term . In words , this test asks for each gene whether the ratio of mRNA to footprint counts is different for the two alleles in the hybrid . All data and results , including raw , downsampled and hypergeometric count datasets , p-values and fold changes , are available in Supplementary Data S2 . Recently , we showed that the expression of many proteins is influenced by multiple loci that segregate between the BY and the RM isolates [8] . The Albert et al . and the present datasets overlap for 114 proteins when considering only genes that can be analyzed in the hybrid data ( i . e . , that are expressed and that contain at least one SNP ) . To generate a rough expectation for the aggregate effect that the multiple pQTL have on a given protein , we added their effects . The pQTL in our earlier study were obtained by comparing allele frequencies in populations of cells with high and low protein expression , so that direct estimates of QTL effects ( i . e . the expected magnitude by which protein expression differs between the different alleles ) are not available . Instead , we used the observed difference in allele frequency at the pQTL location as a measure of effect size . Note that the locus effects can cancel each other: two pQTL with the same absolute allele frequency difference , but with opposite sign will result in an expected aggregate effect of zero . The summed pQTL effects were compared to mRNA and footprint differences from the present study using nonparametric Spearman rank correlation . To test whether the footprint differences or the mRNA differences correlated better with the pQTL effects than the mRNA differences , we constructed 1 , 000 bootstrap datasets . In each of these datasets , we randomly sampled from the 114 genes with replacement and calculated both the mRNA and footprint correlations . We calculated the p-value as the fraction of bootstrap datasets where the mRNA difference/pQTL effect correlation exceeded the footprint difference/pQTL effect correlation . For Artieri & Fraser [35] , we used p-values and fold changes kindly provided by the authors . P-values had been calculated as reported in [35] . We excluded all genes where any of the mRNA , footprint or TE p-values could not be calculated due to low coverage or due to the two replicates not agreeing in direction of effect . This resulted in 1 , 861 genes available for analysis in the species comparison and 1 , 451 genes available in the hybrid comparison . We corrected the p-values for multiple testing by calculating q-values [28] and treated q-values of <0 . 05 as statistically significant . For McManus et al . [34] we used the read counts , p-values and fold-changes provided in their Supplementary Table S5 . There were 4 , 863 genes available in both the parent and hybrid analyses of the McManus et al . data . The p-values reported in [34] are already corrected for multiple testing and we deemed p-values<0 . 05 as statistically significant . We restricted these analyses to coding genes with a “verified” status according to the SGD database . For nonsense SNPs , we considered only the set of high-quality SNPs we used in our ASE analyses . For indels , we considered indel calls produced by samtools mpileup [54] . Because our sequencing coverage of the BY and RM strains is very deep and often exceeds the default limit of 250 fold coverage , we used mpileup with parameters –d 1000 and –L 1000 . The resulting indels were further filtered using the bcftools vcfutils . pl varFilter script to only retain indels with coverage of at least 10-fold , variant quality of at least 30 , an estimated allele frequency in the sample of 1 ( as the sequenced strain was haploid ) . Finally , we only considered indels in “verified” genes . To identify nonsense SNPs and frameshift indels , we used the standalone perl version of the Ensembl Variant Effect Predictor tool [61] with Ensembl cache data for the yeast sacCer3 genome build available in Ensembl build 72 . To examine the effects on translation of the resulting nonsense and frameshift variants , we generated custom tracks in the bedGraph format for display in the UCSC genome browser [48] . These tracks show the start coordinate of each read . The tracks are available for interactive browsing in the UCSC browser at the URL http://genome . ucsc . edu/cgi-bin/hgTracks ? db=sacCer3&hubUrl=https://labs . genetics . ucla . edu/kruglyak/trackhub/hub . txt . The track plots in Figure 7 were generated using the Gviz R package [62] .
|
Individuals in a species differ from each other in many ways . For many traits , a fraction of this variation is genetic—it is caused by DNA sequence variants in the genome of each individual . Some of these variants influence traits by altering how much certain genes are expressed , i . e . how many mRNA and protein molecules are made in different individuals . Surprisingly , earlier work has found that the effects of genetic variants on mRNA and protein levels for the same genes appear to be very different . Many variants appeared to influence only mRNA ( but not protein ) levels , and vice versa . In this paper , we studied this question by using a technique called “ribosome profiling” to measure translation ( the cellular process of reading mRNA molecules and synthesizing protein molecules ) in two yeast strains . We found that the genetic differences between these two strains influence translation for hundreds of genes . Because most of these effects were small in magnitude , they explain at most a small fraction of the discrepancies between the effects of genetic variants on mRNA and protein levels .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"functional",
"genomics",
"quantitative",
"trait",
"loci",
"variant",
"genotypes",
"quantitative",
"traits",
"genetic",
"mapping",
"next-generation",
"sequencing",
"genome",
"analysis",
"trait",
"locus",
"analysis",
"gene",
"expression",
"evolutionary",
"genetics",
"genetic",
"loci",
"protein",
"translation",
"heredity",
"transcriptome",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"evolutionary",
"biology",
"genomics",
"statistics",
"computational",
"biology",
"complex",
"traits"
] |
2014
|
Genetic Influences on Translation in Yeast
|
Muller's ratchet is a paradigmatic model for the accumulation of deleterious mutations in a population of finite size . A click of the ratchet occurs when all individuals with the least number of deleterious mutations are lost irreversibly due to a stochastic fluctuation . In spite of the simplicity of the model , a quantitative understanding of the process remains an open challenge . In contrast to previous works , we here study a Moran model of the ratchet with overlapping generations . Employing an approximation which describes the fittest individuals as one class and the rest as a second class , we obtain closed analytical expressions of the ratchet rate in the rare clicking regime . As a click in this regime is caused by a rare , large fluctuation from a metastable state , we do not resort to a diffusion approximation but apply an approximation scheme which is especially well suited to describe extinction events from metastable states . This method also allows for a derivation of expressions for the quasi-stationary distribution of the fittest class . Additionally , we confirm numerically that the formulation with overlapping generations leads to the same results as the diffusion approximation and the corresponding Wright-Fisher model with non-overlapping generations .
In an asexual population of finite size , weakly deleterious mutations can fix by chance . This phenomenon is due to stochastic fluctuations originating from the finiteness of the population , which can lead to a loss of the fittest class of individuals . If one assumes that the mutation rate does not scale with the length of the genome and that the genome is very long , back mutations are unlikely and can be ignored . In this case the fittest class is lost forever and the number of fixed deleterious mutations increases irreversibly . This process has been termed Muller's ratchet [1] , [2] and has been observed experimentally in several studies [3]–[9] . Furthermore , it has been thought to account for the degeneration of non-recombining parts of sexually reproducing organisms such as the Y-chromosome [10] and mitochondrial DNA [6] . Muller's ratchet can also be used to explain the absence of long-lived asexual lineages [11] . Since in the absence of back mutations mutation-free genomes can only be recreated by recombination between mutation-loaded classes , Muller's ratchet provides an appealing explanation for the evolution of sex [12] , . Each time the least-loaded class , i . e . the class with the fewest number of deleterious mutations , is lost , it is said that Muller's ratchet has clicked . Since the rate of the ratchet determines the speed of degeneration of the population , this quantity is of central interest . In its simplest form the rate of Muller's ratchet depends only on the selection coefficient , the mutation rate and the size of the population , where it is assumed that each mutation has the same effect so individuals with mutations have fitness . In this case the fitness space is equivalent to an axis counting the number of deleterious mutations and the population can be organized into discrete classes labeled by the number of mutations they carry . The deleterious mutations have the effect of shifting the population to higher values of . Since the fitness of the respective classes is given by , selection works into the opposite direction . In the limit of an infinitely large population these two opposing forces lead to a steady state distribution whose precise form was found by Haigh [14] . If finite populations are considered , however , the calculation of the rate of Muller's ratchet turns out to be an intricate problem , despite its simple formulation . The difficulty arises due to the complex interaction of the fluctuation of the least-loaded class with the rest of the distribution . A detailed quantitative understanding of the behavior of the occupation of the class with the fewest mutations , however , is necessary to determine the mean time to extinction of this class , i . e . the inverse of the ratchet rate . Despite of considerable efforts and recent advances [15]–[23] , a quantitative understanding of the ratchet rate remains a challenging open problem . In its standard form Muller's ratchet was first quantitatively described by Haigh who analyzed a classical Wright-Fisher model of an asexually reproducing population of fixed size . He pointed out that the most important quantity of the ratchet is the average number of individuals in the least loaded class , , because fluctuations of ultimately lead to a click of the ratchet . Haigh also suggested an expression for the rate of the ratchet by fitting to numerical simulations . Gordo and Charlesworth were subsequently able to derive an expression for the ratchet rate by studying deviations from the deterministic equilibrium distribution to obtain approximate expressions for drift and diffusion coefficients , from which they obtained an expression for the ratchet rate that has to be evaluated numerically [18] . Later , again using a diffusion approximation and non-overlapping generations , it was shown by Jain that the ratchet rate cannot depend only on but rather has to depend on [21] . If is small , the ratchet clicks frequently and the populations behaves like a wave in space propagating towards higher values of . The traveling wave approach to Muller's ratchet was discussed in [24] and provides an appealing quantitative theory for frequently clicking ratchets . While this regime of Muller's ratchet is relatively well understood , a quantitative understanding of the opposite case of a rarely clicking ratchet is still lacking and has recently attracted a lot of attention [21] , [23] . In this regime the rate of the ratchet is exponentially small in [21] and extinction of the fittest class occurs as the result of a rare , large fluctuation . In contrast to the fast clicking regime the distribution of the population equilibrates to a metastable state after each click . A wide-spread approach in this regime is therefore to consider only the fittest class and apply a phenomenological model for all the classes , , with more mutations than the least-loaded class . Such an approach leads to a one-dimensional approximation where just the fittest class is taken into account . Generally the rate of the ratchet can then be calculated by means of a diffusion approximation as the result of a one-dimensional mean-first passage problem . Recently it was shown how this approach can be improved by accounting for the interaction of the fluctuations of the fittest class with the tail of smaller fitness which can lead to a delayed feedback [23] . Up to now a quantitative treatment of Muller's ratchet relied either on Haigh's model or on the corresponding diffusion approximation . To our knowledge a Moran formulation with overlapping generations has not been employed so far . This is not surprising as in the diffusive limit any quantity should become independent of the respective microscopic formulation and a Moran formulation of Muller's ratchet is expected to be computationally disadvantageous . A Moran formulation , however , can also lead to interesting new approaches to tackle the problem of the ratchet rate analytically . In the present work we investigate a Moran formulation of Muller's ratchet and show how this model can be approximated by a one-dimensional Moran-process in the regime where the ratchet clicks infrequently . We show that this model allows for an analytical solution for the ratchet rate that agrees almost perfectly with values obtained by numerical simulations of the full ratchet . Furthermore , by employing a recently developed method to treat rare , large fluctuations in stochastic population dynamics , we find analytical expressions for the ratchet rate and the quasi-stationary distribution of the fittest class in the parameter range , which also agree very well with the corresponding results of the full ratchet . Finally , we confirm numerically that the formulation with overlapping generations leads to the same results as the diffusion approximation and the corresponding Wright-Fisher model with non-overlapping generations .
In the standard formulation of Muller's ratchet , as considered by Haigh [14] , mutations in a population of fixed size occur at rate and individuals are classified into different groups according to the number of deleterious mutations they carry , . Each mutation reduces the fitness of the genotype by an amount such that the growth rate of an individual with mutations is proportional to . The ratio of the mutation rate to the mutation effect , which is denoted by , plays a central role in the analysis of the ratchet . Observe that the Haigh model assumes the simplest case of a multiplicative , permutation invariant fitness landscape . An extension to more complicated fitness landscapes with epistatic interactions was discussed in [21] . The reproduction model usually employed in the analysis of Muller's ratchet is Wright-Fisher sampling . It consists of , at each time step , replacing the whole generation of individuals by a multinomial resampling of the current generation [25] weighted by the fitness of the different classes . Thus , according to Haigh [14] , if is the number of individuals in generation t which carry mutations and , then the distribution of is multinomial with parameters and , where ( 1 ) and the mean fitness is given by . Wright-Fisher sampling has the advantage of being very efficient for numerical simulations . The downside of the model is , however , that it does not easily allow for analytical methods to be used . Therefore , the corresponding diffusion approximation of the microscopic Wright-Fisher formulation is usually used to predict the click rates of the ratchet . The second widely applied reproduction model in population genetics is the Moran process , which in contrast to the Wright-Fisher formulation assumes overlapping generations . The Moran process , which we focus on in this article , is amenable to a wider range of analytical methods ( at the cost of being slower in numerical simulations ) [26] . It is a stochastic process in which at each time step one individual is chosen for reproduction and one for removal from the population . The choice of the individual that reproduces is random , but ( similarly to the Wright-Fisher formulation ) weighted by the fitness of the class the individual is chosen from . The probability of removal ( or death ) of an individual is independent of the fitness . Applied to Muller's ratchet this therefore embodies the following procedure: An individual with mutations is chosen according to the abundance and selection preference of the class with weight . This individual spawns one offspring with mutations that can then mutate to mutations with probability . The probability to mutate is thus , which is the same as in the Wright-Fisher model . Also , one individual with mutations is chosen for removal with probability ( this may be the one that reproduced ) . Since on average every individual is chosen for removal once every time steps , it is natural to define one generation in the Moran model as time steps . In all figures , the ratchet click times are thus expressed in generations . Although different on the microscopic scale , both Wright-Fisher and Moran models usually converge to the same mesoscopic diffusion regime when is large and fitness advantages and mutation rates are of order . In this limit , the equation describing the evolution of the population is given by ( 2 ) where [23] . The uncorrelated Gaussian white noise with models the stochastic fluctuations due to the finiteness of the population ( genetic drift ) . In the infinite population limit , this equation becomes deterministic and has the steady state solution [14] . Also , a time dependent solution of the deterministic model has been obtained [20] . In this paper , we solve Eq . ( 2 ) numerically using stochastic Runge-Kutta methods [27] .
A mathematical analysis of the Moran model for Muller's ratchet is complicated and even the formulation of the corresponding Markov chain [28] is involved and rarely leads to new insights . The important advantage of Moran models , however , is that they can be formulated in terms of a master equation which is a first-order differential equation describing the time-evolution of the probability of a system to occupy each of a number of states [29] . Many methods have been developed to analyze master equations analytically and therefore Moran models are analytically tractable even beyond the diffusion approximation , if only two species are considered . Thus an appealing approach to the analysis of the rate of Muller's ratchet is to approximate the full ratchet by a model consisting only of two species , see Fig . 1 . Since we are interested in the loss of the fittest class with zero mutations a natural choice is to consider individuals with zero mutations as one species , and to combine all others in a class which contains all individuals with one or more mutations which in this approximation all have the same fitness where has to be adjusted to account for the actual fitness distribution of the full ratchet model . We discuss this non-trivial approximation in detail below . The constraint of a fixed population size then leads to a one-dimensional model since it is sufficient to consider only the dynamics of the fittest class . Since in the Moran model the number of individuals can only change by one without further approximations the master equation for the probability to find individuals in the fittest class is ( 3 ) with the transition rates ( 4 ) ( 5 ) where is the mutation rate away from the fittest class and the boundary conditions are imposed [25] . Unless specified otherwise , the initial condition is chosen to be concentrated on the equilibrium value of for large ( see below for details ) . The biological significance of the terms in the equations above are as follows . The probability for one individual of the fittest class to be chosen for birth or death is . For the mutated class , the probability to be chosen for birth is multiplied with the selection disadvantage , and to be chosen to die . The probability for an offspring to mutate is , and not to mutate . The denominators are normalization factors . Note that we apply the convention of the Moran process where the mutation is divorced from the birth/death process [25] . Here and in the following lowercase letters denote the parameter values of the approximate two-class model , while capital letters denote the parameters of the full ratchet ( see also Table 1 for a list of symbols ) . It is important to note that and are effective parameters which need to be related to the biologically relevant parameters and . The idea of representing all classes but the fittest as one class was first introduced in [22] for a Wright-Fisher model of Muller's ratchet . A crucial step in the reduction of the full model of Muller's ratchet to the one-dimensional formulation is the relation of the two mutation rates and fitness disadvantages in the respective models . This mapping is not unique and two reasonable assumptions have to be invoked to relate the two parameters pairs . A plausible approach is to compare the steady state distributions in the infinite population limit of the respective models . For the full ratchet whose dynamics is given by Eq . ( 1 ) the well-known steady state distribution for the probability of an individual to have mutations is . A non-zero steady state of the fittest class in the two-class system can only be obtained in the parameter regime and is given by . To relate the parameters we now demand that the mean fitness of the full population and the mean fitness of all individuals carrying a mutation is equal in both models . The mean fitness of the full population in the steady state of the full ratchet is while the mean fitness of all individuals in the two state model is . Condition ( ) accordingly suggests the relation ( 6 ) The mean fitness of all individuals carrying mutations is in the full ratchet model given by . In the two state model this corresponds to . Employing condition consequently yields the relation ( 7 ) We can also introduce the parameter which is related to according to ( 8 ) Relation ( 8 ) shows that the restriction of the two-class model does not restrict the range of . Before we present the analytical solution for the ratchet rate of this model , let us shortly discuss the validity of the approximation used . To correlate the parameters of the full ratchet and the two state model , we have related properties of the equilibrium solution of an infinite population in both models . This certainly makes sense as long as the typical time that it takes for the population to relax to a metastable state after each click is much smaller than the mean time between two successive clicks . This condition is fulfilled in the case of the slowly clicking ratchet , which is the regime we focus on in this work . If the ratchet clicks rapidly the population does not equilibrate after a click and relating the parameters based on equilibrium distributions is clearly not valid . In the Haigh model mutations are Poisson distributed . It follows that the mutation rate out of the fittest class is . Consequently , from Eq . ( 6 ) , the mutation rates out of the fittest class are equal in both models which certainly is a reasonable assumption . We note that the occupancy of the fittest class and therefore the rate at which the ratchet clicks is the result of a complicated interplay of all fitness classes . Thus , although the mutation rate out of the fittest class is the same in both models , their rates will differ due to the different fitness distributions . Furthermore , our second relation ( 7 ) entails that the number of individuals which are not in the fittest class is the same in the equilibrium states of both models . Consequently the same holds true for the number of individuals in the fittest class , i . e . . Thus , although the parameter mapping is not unique , it is hard to think of any other relation in the slowly clicking regime as this would consequently violate the properties specified above . Furthermore , our relations ( 6 ) and ( 7 ) are the same as the expressions previously obtained by Waxman and Loewe [22] . It is important to keep in mind that the parameter mapping is only valid in the rare clicking regime and that other mappings might be more appropriate in the fast clicking regime [22] . The expression ( 9 ) for the mean time to extinction is exact . It gets , however , unwieldy and impractical when larger population sizes are considered . Furthermore , it does not allow for any analytical statements about the distribution of the frequency of the fittest class . Therefore , we want to gain quantitative insight into the ratchet rate and the distribution of the frequency of the fittest class by an approximate treatment of Eq . ( 3 ) . The most widely applied approach certainly is the diffusion approximation from which by standard methods the mean time to extinction ( MTE ) can be calculated analytically [29] . The resulting expression usually has to be evaluated numerically . While the diffusion approximation provides faithful results in the regime where an extinction event is the outcome of a typical fluctuation of the process , it in general may fail to describe the MTE correctly when extinction occurs as the result of a rare , large fluctuation [30]–[32] . In the rare clicking regime the relaxation time to the metastable state is much shorter than the mean time between the clicks and the population equilibrates after each click . It is important to note that in such a scenario the click of the ratchet is due to a rare , large fluctuation away from the metastable state . An approach to the treatment of master equations which is especially well suited to account for rare event statistics is the WKB- ( Wentzel-Kramers-Brillouin ) theory . This approximation scheme which is sometimes referred to as the eikonal approximation was first developed for a semi-classical treatment of quantum mechanics and has recently attracted a lot of attention in the context of stochastic population dynamics [33]–[36] . Similar to the diffusion approximation , it replaces the master equation of the Moran process by an analytically tractable equation which in addition allows for a mathematically controlled approximation in terms of powers of the inverse population size . Recently the WKB-approximation has also found its way into evolutionary modeling [30] , [37]–[39] . The approach we apply in the following was first considered in [40] and later considerably extended and generalized in [37] . The basic idea relies on the fact that the process can be characterized by a metastable state around which the frequency of the fittest class resides . After a long average time the fittest class is eventually lost and Muller's ratchet clicks . For the approach to work two crucial assumptions have to be made . First , the population size has to be finite and not too small , i . e . . Second the typical relaxation time to the metastable state should be much shorter than the MTE , i . e . . We note that here this condition has to hold anyway in order for the two state approximation to be meaningful . It can be shown that the metastable state , which is sharply peaked around , is encoded in the first excited eigenvector of the master equation ( 3 ) which has not decayed at a time [37] . Thus the shape of the PDF of the metastable state , which is referred to as the quasi-stationary distribution ( QSD ) , is given by . Furthermore , the decay rate of this distribution , i . e . the ratchet rate , is determined by the first non-zero , positive eigenvalue of the master equation ( 3 ) . As was shown in [41] , the decay of the QSD for times can therefore be obtained as ( 10 ) Accordingly , the click probability distribution behaves as ( 11 ) Using Eqs . ( 3 ) , ( 10 ) and ( 11 ) the click rate is given by ( 12 ) which is just the probability flux into the absorbing state . In the remainder of this section we present an approach to calculate the QSD based on a WKB-type approximation . Before employing the WKB ansatz we insert ( 10 ) into ( 3 ) to obtain , after introducing , ( 13 ) where and . Since we consider the rare-clicking regime of the ratchet , the term on the left-hand side is exponentially small in and we can neglect it . The resulting quasi-stationary master equations reads ( 14 ) Now we are ready to employ the WKB approach by expressing the solution of this equation by the ansatz [40] ( 15 ) where both and are assumed to be of order unity and is a normalization constant . Inserting this ansatz into ( 14 ) , expanding around to first order and neglecting terms of order , we obtain in leading order ( 16 ) where . The solution of this equation is given by ( 17 ) After insertion of into the ansatz ( 15 ) the lowest order solution for the QSD is obtained up to the -independent normalization constant . To determine one exploits the fact that the QSD is strongly peaked around and then assumes it to be of Gaussian shape centered at which is normalized to unity . Around the maximum this leads to an approximation of the QSD by whose normalization yields . Hence in leading order we obtain for the QSD ( 18 ) Using this expression of the QSD we can calculate using Eq . ( 12 ) the leading order behavior of the click rate ( 19 ) where we have used that and for large . In leading order we thus obtain the anticipated exponential dependence of the ratchet on in the rare clicking regime . These results are valid as long as because the WKB-ansatz requires the ratchet rate to be exponentially small . Furthermore , the normalization procedure can be expected to fail if the metastable state is close to the boundary because the Gaussian approximation does not hold anymore . So far we have obtained the ratchet rate to exponential accuracy only . The next order -corrections of the WKB-approximation provide the pre-factor of the QSD . They are obtained by expanding to second order and to first order around . The calculation of the sub-leading corrections is more involved and shall not be carried out in detail here . The crucial step in the calculation is to note that the WKB-solution in leading order is not valid close to the absorbing state at . Therefore , the WKB solution has to be matched with an exact recursion solution of the quasi-stationary master equation ( 14 ) . A detailed account of this method is given in [37] . Following the steps in [37] we obtain for the QSD ( 20 ) where the transition rates can be obtained from Eqs . ( 4 ) and ( 5 ) and is given by Eq . ( 17 ) . The WKB solution for the inverse of the ratchet rate is given by ( 21 ) with . Inserting the respective transition rates , we obtain for the mean time to extinction of the fittest class , i . e . the inverse of the ratchet rate ( 22 ) This expression provides an exact result to order . To gain a deeper understanding of the WKB-solution one can simplify the unwieldy expression ( 22 ) for . Keeping and constant and expanding in , we obtain to leading order in the approximation ( 23 ) which is almost indistinguishable from the WKB-solution ( 22 ) for . In Fig . 3 we have compared this result for different parameters to the numerical results of the full ratchet . The WKB approximation of the mean time to extinction in the two state model agrees in the range corresponding to almost perfectly with the numerical results of the full ratchet . While the WKB-prediction is still quite good for it starts to deviate for increasing values of . The parameter range in which the WKB-theory works thus is more restricted than in the two-class model . This can be explained by noting that for the two-class approximation is still valid if the ratchet operates in the rare clicking regime , i . e . if is chosen to be large enough , see Fig . 2 . The WKB-theory on the other hand breaks down if is close to the absorbing state at independent of . A comparison of the exact solution ( 9 ) , the WKB-solution ( 22 ) and the approximation ( 23 ) is provided as supporting information ( Fig . S1 ) . The WKB-theory not only yields results for the ratchet rate but is also capable of describing the frequency distribution of the fittest class in the metastable state , i . e . the QSD , because the parameters of the two-class model were chosen such that the size of the fittest classes match in both models . One can therefore also expect that the QSD of the fittest class is approximately the same in both models . In Fig . 4 we have compared the numerically obtained size of the fittest class in the full ratchet model with the WKB-solution ( 20 ) and observe a striking agreement . As anticipated , the WKB-theory starts to deviate if the deterministic fixed point is close to the absorbing point at and if . Let us now discuss how the presented analysis is related to previous studies on the rate of Muller's ratchet . Preceding works have mostly considered the diffusion approximation in the form of the stochastic differential equation ( 2 ) to approach Muller's ratchet analytically , while numerical simulations have relied on Haigh's model with non-overlapping generations using Wright-Fisher sampling , Eq . ( 1 ) . For this reason it is first of all necessary to check that the Moran model of the full ratchet yields the same rates as the Wright-Fisher model and the diffusion approximation . We note that some care has to be taken to ensure that the diffusive limit of the Wright-Fisher model has the same diffusion constant as the corresponding Moran formulation , since these usually differ by a factor of two [28] . Since fluctuations scale with , one possibility to take this into account is to consider the Wright-Fisher model with individuals , which is what we do in the simulations presented below . We have performed numerical simulations of the Wright-Fisher model and have numerically integrated the stochastic differential equation ( 2 ) using stochastic Runge-Kutta methods . To compare the three different approaches , the click times averaged over 1000 realizations for each model for different values of and similarly to the previous sections are presented in Fig . 5 . We observe excellent agreement of the two macroscopic models and the diffusive description for slow and fast ratchets . After ensuring that the rate of Muller's ratchet is independent of the microscopic reproduction model , let us now explain why a Moran model is nevertheless essential for the presented approach . The Moran model is exclusive because it can be formulated in terms of a master equation for which well-known analytical methods exist that allow alternatives to the diffusion approximation . The WKB-approximation is one example of these methods that is particularly useful to describe rare , large fluctuations . Most classical and recent works , however , have considered a one-dimensional diffusion approximation to analyze the dynamics of the fittest class and calculated the ratchet rate as the mean time to extinction of this process . Given the fact that the rare , large fluctuations are responsible for the ratchet clicks , a Moran model certainly deserves a detailed analysis . In order to compare our solution for the click rate Eq . ( 23 ) with the preceding works , we rewrite this expression as ( 24 ) where and . This expression exhibits the same scaling behavior in the effective parameters , and as the one found by Jain in the rare clicking regime with the parameters , and of the full ratchet , who was the first to show that the ratchet rate cannot depend only on but has to depend on [21] . Thus by using Eqs . ( 6–8 ) and noting that the form of Eq . ( 24 ) agrees with findings of Jain for small and small . Our solution exhibits -dependent functions in the exponent and the pre-factor which is in contrast to the result of Jain where these factors have to be replaced by a constant which is sometime referred to as the Haigh factor . Since and the values of both functions are close to the values between and which were ad hoc chosen for this constant . In a recent work , Neher and Shraiman investigated the propagation of fluctuations in the fitness distribution [23] . In the course of their work they also found the Haigh factor to be -dependent which they could attribute to a time delay between the fluctuations in the fittest class and the fluctuations of the mean fitness , thereby extending the classical work of Haigh [14] , Stephan et al . [15] , Gordo and Charlesworth [18] , and also the more recent work by Jain [21] . While Neher and Shraiman used path integral techniques to obtain an expression similar to Eq . ( 24 ) , and had to calculate the value of the Haigh factor numerically , our approach determines the ratchet rate including the Haigh-factor analytically , at the cost of being restricted to small . Neher and Shraiman plot their result for , for which the WKB is not a good approximation any more . This and the fact that also is no longer similar to may explain that our result for the Haigh factor decays faster than the numerical results of Neher and Shraiman . Muller's ratchet has been proposed as a simple model for the degeneration of asexual populations and non-recombining parts of sexually reproducing populations . The quantitative understanding of the ratchet rate is complicated due to the significant influence of rare , large fluctuations of the number of individuals in the fittest class . This effect is most prominent in the important regime where the ratchet clicks infrequently , which is characterized by a relaxation of the ratchet to a metastable state after each click . The fact that the extinction of the fittest class is due to such a rare , large fluctuation and not the cause of a typical fluctuation prohibits simple diffusive treatments of the ratchet and thus generates difficulties in finding an analytical expression for the ratchet rate . In this article , we have obtained such an analytical expression by considering a simplified Moran model of Muller's ratchet that reduces the calculation of the ratchet rate to the simpler problem of calculating the mean time to fixation of a deleterious allele . We have shown that in the rare clicking regime the rates predicted by this two-class Moran model agree almost perfectly with the rates of the full ratchet obtained numerically . Furthermore , the formulation of the two-class model in terms of a one-dimensional master equation allows for the application of an approximation scheme which specifically accounts for the effects of rare , large fluctuations . This WKB-theory is a controlled approximation in terms of the inverse population size and provides a closed analytical solution without any free parameters . Our method yields the same scaling form of the ratchet rate as previously obtained by Jain [21] . In contrast to Jain , we find a -dependent exponential prefactor . This supports the findings of Neher and Shraiman who also suggested a -dependence of the “Haigh factor” [23] . While in their work the factor had to be estimated numerically , our theory yields an analytical prediction for this quantity . Additionally , we have been able to obtain analytical results for the frequency distribution of the fittest class in the metastable state that are in excellent agreement with numerical simulations . This distribution has been alluded to in several of the previous works on Muller's ratchet , but has remained elusive up to now . Our analytical description of the distribution provides a more complete understanding of the ratchet , particularly because the distribution is formed at a fraction of the ratchet click time . Also , the fact that distinguished non-Gaussian tails can be observed in the frequency distribution again emphasizes the necessity to go beyond simple diffusion approximations to describe the ratchet rate analytically . We have shown that a Moran formulation in conjunction with a reduction to a two-class model and the subsequent application of a WKB-type approximation can provide a viable route for the quantitative prediction of rare , but crucially large fluctuations in simple models of population genetics . We anticipate that models covering additional effects such as epistasis can be included in this framework and , more generally , that the methodology presented here can also be applied in other areas of computational biology where a process is driven by rare stochastic fluctuations .
|
Muller's ratchet is a paradigmatic model in population genetics which describes the fixation of a deleterious mutation in a population of finite size due to an unfortunate stochastic fluctuation . Obtaining quantitative predictions of the ratchet rate , i . e . the frequency with which such a mutation fixes , is believed to be important for understanding a broad range of effects ranging from the degeneration of the Y-chromosome to the evolution of sex as a means of avoiding the fixation of deleterious mutations . To obtain a better understanding of how Muller's ratchet operates , we have considered a model with overlapping generations , which allows for the application of methods specifically tailored for the analysis of rare stochastic fluctuations which drive the ratchet . We obtain concise and accurate results for the rate of Muller's ratchet . Additionally , we are able to predict the full distribution of the frequency of the fittest individuals , a quantity of central interest in understanding the ratchet rate and possibly experimentally much more accessible than the rate , in particular when the ratchet rate is very large .
|
[
"Abstract",
"Introduction",
"Models",
"Results/Discussion"
] |
[] |
2013
|
Distribution of the Fittest Individuals and the Rate of Muller's Ratchet in a Model with Overlapping Generations
|
IL-10 is a critical regulatory cytokine involved in the pathogenesis of visceral leishmaniasis caused by Leishmania donovani and clinical and experimental data indicate that disease progression is associated with expanded numbers of CD4+ IFNγ+ T cells committed to IL-10 production . Here , combining conditional cell-specific depletion with adoptive transfer , we demonstrate that only conventional CD11chi DCs that produce both IL-10 and IL-27 are capable of inducing IL-10-producing Th1 cells in vivo . In contrast , CD11chi as well as CD11cint/lo cells isolated from infected mice were capable of reversing the host protective effect of diphtheria toxin-mediated CD11c+ cell depletion . This was reflected by increased splenomegaly , inhibition of NO production and increased parasite burden . Thus during chronic infection , multiple CD11c+ cell populations can actively suppress host resistance and enhance immunopathology , through mechanisms that do not necessarily involve IL-10-producing Th1 cells .
Dendritic cells ( DCs ) are widely recognized as being the most important myeloid cell involved in antigen presentation and the initiation and regulation of CD4+ T cell-dependent protective immunity against a variety of intracellular parasites ( reviewed in [1] , [2] ) , and show promise for the development of new approaches in vaccination and immunotherapy [3] , [4] . Initially based largely on in vitro studies , the key role of DCs in antigen presentation has been borne out in recent years through the availability of mice in which DCs can be ablated in a conditional manner [5] . Hence , diphtheria toxin ( DTx ) -mediated ablation of DCs results in a significant reduction in T cell priming following various infectious challenges , including with Mycobacterium tuberculosis , Plasmodium yoelli , Listeria monocytogenes , Streptococcus pyogenes and LCMV [6] , [7] , [8] , [9] . In contrast , the role of DCs during later stages of infection and their contribution to the immune imbalance that is often associated with chronic infection are less well understood , in spite of the known ability of DCs to induce tolerogenic or regulatory responses [4] , [10] , [11] , [12] . CD11c+ DCs play multiple roles in the pathogenesis of leishmaniasis , including experimental visceral leishmaniasis ( EVL ) caused by Leishmania donovani ( reviewed in [13] ) . Dermal DC [14] and Langerhans cells [15] have been implicated in the early stages of L . major infection , and as this infection progresses , many parasites are found in the draining LN within CD11c+ cells that resemble TipDCs [16] . Expression of MHCII on DCs is both necessary and sufficient for the induction of effective immunity to L . major , suggesting macrophage antigen presentation may not be required for effector T cell function [17] . In EVL , splenic DCs belonging to the CD8α subset of conventional DCs ( cDCs ) are the first cells to produce IL-12 within the splenic microenvironment [18] , and are activated for heightened expression of a variety of costimulatory molecules through both direct interactions with Leishmania parasites and through inflammatory signals [19] . In chronic EVL , however , cDC cytokine production is modulated in a subset-specific manner [18] and migration through lymphoid tissue is disrupted [20] . In addition , CD11c expression is found on other cells known to contribute to anti-leishmanial resistance , including NK cells [21] , and inflammatory monocytes/TipDCs [16] . However , the relative contribution of these different CD11c+ cell populations to disease progression and the regulation of T cell effector and regulatory function is poorly understood . Visceral leishmaniasis is also noted for the production of the immunoregulatory cytokine IL-10 , and targeting of IL-10 signaling has been identified as a potential therapeutic strategy [22] . Although multiple cellular sources of IL-10 have been identified in VL , the identification of a population of IFNγ-producing CD4+ T cells that also produces IL-10 and its association with progressive disease in both mice [23] , [24] and in humans [25] has drawn particular attention . The co-production of IL-10 by IFNγ-producing CD4+ T cells is not novel for leishmaniasis , however , and is now a recognized feature of Th1 cell differentiation . Considerable attention has been focused , therefore , on dissecting the molecular signals required for expression of this mixed effector/regulatory phenotype . In vitro studies using transgenic CD4+ T cells and repeated exposure to antigen and APCs have suggested that the induction of IL-10 is a consequence of sustained antigen presentation , requiring the presence of high levels of IL-12 [26] . The cytokine IL-27 is also implicated in the generation of IL-10-producing CD4+ T cells in vitro [27] , [28] , [29] , [30] , [31] , via signaling pathways dependent on STAT3 [29] , and optimal generation of CD4+IL-10+ cells by IL-27 requires the co-ordinate initiation of c-Maf , ICOS and IL-21 expression [32] , [33] . In addition , emerging evidence suggests that IL-27 may directly alter methylation patterns around the il10 promoter in CD4+ T cells , thus allowing greater IL-10 expression [34] . IL-27 also favors the production of IL-10 by IFNγ-producing Th1 cells through an alternate signaling pathway that involves STAT1 , STAT4 and Notch [35] , [36] . In spite of these advances , the cellular sources of IL-27 in vivo have been poorly defined . A direct role for DC-derived IL-27 in the generation of IL-10+ T cells has been described in vitro , where production of this cytokine in response to galectin-1 , and during ovalbumin-induced oral tolerance , both favored the differentiation of IL-10-producing T cells with potent regulatory capacity [37] , [38] . Nevertheless , the cellular requirements for generating CD4+IFNγ+IL-10+ T cells in vivo remain obscure and no studies to date have addressed this question in the context of chronic infection . We therefore sought to address two inter-related questions: i ) what is the role of CD11c+ cells during chronic L . donovani infection and ii ) do these cells contribute to the emergence of IL-10-producing Th1 cells . Given recent concerns over off target effects induced by DTx treatment of mice [8] , [39] , we used a functional complementation approach to independently examine the role of CD11chi cDCs and CD11cint/lo cells in determining host resistance and Th1 cytokine production . We show that CD11chi cDCs , as well as a mixed CD11cint/lo cell population , are capable of inhibiting host resistance and promoting disease-associated pathology . Our study also provides the first formal evidence that IL-10+IL-27+ cDCs are able to promote IL-10 production by Th1 cells in vivo . Our data therefore highlight CD11c+ cells as potential targets for immunotherapy and also demonstrate an important discordance between disease progression and the emergence of IL-10-producing Th1 cells .
C57BL/6 mice infected with L . donovani amastigotes developed pronounced splenomegaly from day 21 post infection ( p . i . ) ( Figure 1A ) , associated with an increasing tissue parasite burden ( Figure 1B ) . As assessed by polyclonal activation ex vivo ( Figure 1C and D ) , the frequency of splenic CD3ε+CD4+ IFNγ+ T cells increased from 2 . 4±0 . 4% in naïve mice to 44 . 4±3 . 0% ( p<0 . 001 ) by day 28 of infection which , when taking into account splenomegaly , reflected a >500-fold increase in absolute number of cells committed to IFNγ production in the spleen . Infection was also associated with emergence of a population of splenic CD3ε+CD4+ T cells capable of producing both IFNγ and IL-10 . This population increased in frequency ∼50 fold , comprising 0 . 1±0 . 02% of CD3ε+CD4+ cells in naïve mice and 5 . 0±0 . 9% at day 28 of infection ( p<0 . 001 ) , equating to a ∼170 fold expansion in the total number of splenic T cells capable of simultaneous production of IFNγ and IL-10 ( 7 . 6±1 . 2×103 vs 1 . 3×106±1 . 7×105 in naïve and day 28 infected mice , respectively; p<0 . 001 ) . Whilst the frequency and number of CD3ε+CD4+ T cells producing IL-10 alone also increased , this population of IL-10-producing cells remained modest compared to those that also made IFNγ . Similar results were also obtained following in vitro re-stimulation of splenic CD4+ T cells using Leishmania-antigen pulsed BMDCs ( Figure 1E and F ) . In order to determine the lineage origin of these IFNγ+IL-10+ CD4+ T cells , cytokine producing cells ( Figure 1G ) were examined for intracellular expression of the Th1-associated transcription factor Tbx21 ( T-bet ) , the regulatory T cell-associated transcription factor forkhead box transcription factor 3 ( Foxp3 ) , and for surface expression of the IL-7 receptor alpha chain ( CD127 ) . CD3ε+CD4+ T cells capable of simultaneous production of IFNγ and IL-10 were exclusively T-bet+ , CD127− and Foxp3− ( Figure 1H ) . In contrast to the expansion of IL-10-producing CD4+ T cells , the frequency of splenic Foxp3+ Treg did not increase during infection ( Figure S1 ) , but instead decreased as previously reported [23] . Therefore , chronic infection with L . donovani was associated with the expansion of antigen-specific CD4+ T-bet+CD127− Foxp3− ‘Th1-like’ cells capable of simultaneously producing IFNγ and IL-10 . To address the cellular mechanisms underlying the expansion of CD4+IFNγ+IL-10+ T cells , we focused on alterations within the splenic CD11chiMHCIIhi cDC compartment ( Figure 2A ) . First , we characterized the entire cDC population for the expression of cell surface ligands associated with costimulation . CD80 and CD86 expression was not changed compared to that on cDCs from naïve mice and only a 2–2 . 5 fold induction of CD40 and PD-L1 expression was noted on cDCs at day 28 p . i . , relative to naïve mice ( Figure 2B and Figure S2 ) . The three major lymphoid-resident cDC subsets , as determined by CD4 and CD8 expression ( Figure 2A ) , were all found at the expected frequencies [40] , but individual subsets showed some level of differential regulation of co-stimulatory molecule expression during infection ( Figure 2C–F ) . Of note , CD8α+ cDCs showed the least activation as judged by CD40 , CD80 and CD86 expression , yet conversely had the greatest increase in PD-L1 expression . In light of the critical role for APC-derived cytokines in shaping CD4+ T cell lineage commitment , we next sought to determine how chronic infection impacted upon cDC cytokine production ex vivo . Highly purified CD11chiMHCIIhi cDCs from spleens of naïve and day 28-infected mice ( Figure 3A ) showed differential cytokine profiles at both the whole population level and when separated into distinct subsets . CD11chiMHCIIhi cDCs from day 28 infected mice had significantly reduced levels of spontaneous and LPS-induced IL-12p70 secretion , when compared to cDCs from naïve mice ( p<0 . 01; Figure 3B ) . Similar results were also obtained from isolated cDC subsets by analysis of IL-12p70 [18] and IL-12/23p40 secretion ( Figure 3C and D ) . In contrast , cDC production of IL-27p28 was significantly enhanced during infection , both at the bulk population level ( Figure 3E ) and when assessed for each individual cDC subset ( Figure 3F ) . Similarly , cDCs from infected mice also produced elevated levels of IL-10 when compared to cDCs from naïve mice ( Figure 3G and H ) . Although more pronounced by day 28 p . i . , similarly altered cytokine responses have been observed in the early stages of acute L . donovani infection [18] . Autocrine IL-10 signaling is known to influence DC cytokine production , with splenic cDCs particularly sensitive to this form of regulation [41] , [42] . Therefore , we next sought to determine whether the altered cDC cytokine profile was in part dependent on autocrine IL-10 and/or IL-27 production . cDCs from infected mice cultured in the presence of αIL-10R mAb spontaneously secreted IL-12p70 to a similar extent to those treated with LPS ( Figure 4A ) . However , maximal IL-12p70 production was achieved by simultaneous IL-10R blockade and LPS stimulation . IL-10R blockade also significantly enhanced the accumulation of IL-10 in the culture medium , again most pronounced when combined with LPS stimulation ( Figure 4B ) . These data indicate a potent negative regulatory function for autocrine IL-10 produced by cDCs isolated from infected mice . In contrast , neutralization of IL-27p28 alone had no impact on ex vivo IL-12p70 or IL-10 production by cDCs isolated from infected mice , nor was there any additive effect when combined with IL-10R blockade ( Figure 4A and B ) . Hence , IL-27p28 does not auto-regulate cDC cytokine production under these conditions . Finally , we examined whether the immunomodulators TGFβ or Indoleamine 2 , 3-dioxygense ( IDO ) were substantially regulated in cDC subsets as a result of infection with L . donovani . At the transcriptional level , only CD4+ cDCs from infected mice showed any significant increase in accumulation of Tgfβ mRNA ( Figure S3A ) and in no population of cDC did we observe any accumulation of Ido mRNA as a result of infection ( Figure S3B ) . In summary , therefore , chronic L . donovani infection is associated with muted co-stimulatory molecule expression , increased IL-10 and IL-27p28 production and a dramatic impairment in IL-12p70 production by splenic cDCs , with IL-12p70 secretion regulated in part by autocrine IL-10 signaling . To assess the in vivo impact of cDCs during chronic infection , we generated ( CD11c-cre×Rosa26iDTR ) F1 mice . In these mice , expression of Cre recombinase is driven by CD11c promoter activity , resulting in cleavage at loxP sites flanking a ubiquitously expressed STOP cassette upstream of a simian diphtheria toxin receptor ( DTR ) . Past or current expression of CD11c initiates DTR expression and thus provides inducible sensitivity to diphtheria toxin ( DTx ) . We administered either saline or DTx i . p . to ( CD11c-cre×Rosa26iDTR ) F1 mice at 48 hour intervals for a period of 7 days , beginning on day 21 p . i . ( Figure 5A ) . Unlike in CD11c-DTR mice [5] , we found no evidence of toxicity using this regimen . ∼80–90% of CD11chiMHCIIhi cDCs were ablated after 7 days of treatment ( Figure 5B and C ) . Depletion was almost 100% complete for the CD4+ and CD8α+ subsets , with most residual cDCs belonging to the DN subset ( Figure 5D–F ) . In addition to depletion of cDCs , we also observed depletion of some CD11cint/lo cells ( Figure 5B and C ) . ∼20% of CD11cint/loMHCII− cells and 40% of CD11cint/loMHCIIhi cells were lost , most likely including both CD11cint/lo DCs and NK cells [21] , [43] . Additional off target effects of DTx treatment were also noted . CD169+ marginal zone macrophages were depleted , as determined by immunofluorescence staining of tissue sections ( data not shown ) . Although others have observed loss of marginal zone macrophages in CD11c-DTR mice [39] , these cells are already largely absent in mice infected with L . donovani [44] . Other significant off-target effects of DTx treatment were restricted to a decrease in the frequency and number of NK1 . 1+CD11b+ NK cells and an increase in the frequency of splenic CD11bhiGr-1hi neutrophils ( 1 . 46±0 . 18% vs . 2 . 38±0 . 29% in infected vs . DTx-treated infected mice respectively; p<0 . 05 , Figure S4 ) . Depletion of CD11c+ cells in chronically infected ( CD11c-cre×Rosa26iDTR ) F1 mice had a profound impact on splenic pathology , reflected by a dramatically reduced spleen size ( from 3 . 94±0 . 22% vs . 2 . 14±0 . 14% of total body weight in saline-treated and DTx-treated mice , respectively; p<0 . 0001 ) . In contrast , treatment of naïve ( CD11c-cre×Rosa26iDTR ) F1 mice with DTx had no impact on spleen size ( Figure 6A ) . Ablation of CD11c+ cells also significantly reduced splenic parasite burden ( 101±23 vs . 27±9 LDU in saline vs . DTx treated mice , respectively; p<0 . 05; Figure 6B ) . Conversely , mice treated with DTx had an almost 5-fold increase in nitric oxide production , as measured by spontaneous release from adherent splenocytes ( 23 . 04±2 . 51 vs . 92 . 36±21 . 89 µM control and DTx treated mice; p<0 . 05 ) . NO production was not detectable from adherent splenocytes in naïve animals , irrespective of DTx treatment ( Figure 6C ) . By each of these criteria , therefore , CD11c+ cells appeared to be playing a role in promoting disease progression in vivo . The experimental approach outlined above provided an opportunity to determine whether there was a causal link between the appearance of the phenotypically distinct cDCs described above and the induction of IL-10-producing Th1 cells . We therefore examined CD4+ T cells from these DTx-treated mice for their capacity to produce IFNγ and IL-10 ( Figure 7 ) . Depletion of CD11c+ cells from day 21 to day 28 of infection did not affect the frequency of antigen-specific IFNγ+ T cells ( Figure 7A; 35 . 78±4 . 18% vs . 30 . 33±5 . 32% after saline or DTx , respectively ) , although absolute numbers were decreased by approximately 2-fold in keeping with the reduction of spleen size ( Figure 7D ) . In contrast , the frequency of antigen-specific CD4+ T cells capable of the simultaneous production of IFNγ and IL-10 was significantly reduced following DTx administration ( Figure 7C; 2 . 62±0 . 31% vs . 1 . 28±0 . 11% in saline and DTx treated mice respectively; p<0 . 001 ) and the absolute number per spleen was reduced by four fold ( Figure 7F ) . Antigen-specific CD3ε+CD4+ T cells capable only of IL-10 production also showed a trend towards a reduction in frequency , but this was not significant ( 0 . 57±0 . 23% to 0 . 10±0 . 03%; p = ns; Figure 7B ) . The absolute number of these cells in the spleen was significantly reduced , however ( Figure 7E ) . In combination , these data demonstrate that depletion of CD11c+ cells during chronic infection dramatically reduces splenic pathology , allows NO-dependent parasite clearance and impairs the generation of antigen-specific CD3ε+CD4+ T cells capable of simultaneous IL-10 and IFNγ production . As neutrophils play a role in the control of established L . donovani infection [45] and splenic neutrophil numbers increase after DTx treatment ( this study and [46] , [47] ) , we repeated these experiments using DTx-treated infected mice co-treated with a Ly6G-specific mAB ( 1A8 ) to deplete neutrophils . The frequency of IFNγ+IL-10+ CD4+ T cells was similar in DTx-treated mice irrespective of whether neutrophils were present or absent ( 1 . 69±0 . 37% vs . 1 . 89±0 . 40% in control and 1A8-treated mice respectively ) . Neutrophil depleted DTx-treated mice also showed a similar reduction in splenomegaly and increased NO production as seen in neutrophil replete DTx-treated mice ( Figure 8A and B ) . As previously noted in wild type mice [45] , neutrophil depletion of DTx-treated mice led to increased parasite burden ( Figure 9C ) , illustrating that immunopathology is not strictly associated with parasite load . Nevertheless , changes in neutrophil number cannot account for the changes in Th1 cell differentiation or immunopathology observed in DTx-treated mice . Hence , these data demonstrate that depletion of CD11c+ cells during ongoing infection dramatically reduces splenic pathology , promotes NO-dependent parasite clearance and significantly impairs the generation of antigen-specific CD4+ T cells capable of simultaneous IL-10 and IFNγ production , whilst only slightly reducing the abundance of other Th subsets . To definitively address whether cDCs or other populations of CD11c+ cells were responsible for the induction of IL-10 production in Th1 cells and for changes to host resistance , we employed a functional complementation approach . We adoptively transferred ( in accordance with their population abundance ) wild type DTx-resistant CD11chiMHCIIhi cDCs and CD11cint/lo cells from infected B6 . CD45 . 1 mice into infected DTx-treated ( CD11c-cre×Rosa26iDTR ) F1 mice ( Figure 9A and B ) . cDCs obtained at d21 p . i . had similar patterns of co-stimulatory molecule expression to cDCs obtained at d28 p . i . , particularly with regard to PD-L1 expression and this was also similar to the phenotype of CD11cint cells ( in spite of heterogeneity within this population; Figure S5 ) . We also characterised CD11cint cells and cDCs for IL-27p28 ( Figure S6A ) and IL-10 ( Figure S6B ) mRNA accumulation at d21 p . i . and day 28 p . i . and observed significantly lower accumulation of both cytokines within the CD11cint population , suggesting a sustained difference in their capacity to regulate IL-10 and IL-27 expression during infection . Ablation of endogenous CD11c+ cells was maintained after transfer by repeated DTx treatment . Strikingly , transfer of either CD11cint/lo cells or CD11chi cDCs was sufficient to restore splenomegaly ( Figure 9C ) , parasite burden ( Figure 9D ) and NO production ( Figure 9E ) to levels similar to that observed in non-treated infected mice . The potency of these cells to restore disease progression was all the more remarkable given that at the time of assay ( d7 post transfer ) donor CD45 . 1 CD11c+ cells could not be detected , suggesting long term engraftment had not occurred ( data not shown ) . Adoptive transfer had no impact on the frequency of IFNγ single-producing CD4+ T cells ( in keeping with the limited effect of CD11c depletion on this T cell response ) or on the frequency of IL-10+ single producing T cells , although absolute numbers were increased as a result of the changes in splenomegaly after these interventions ( not shown ) . In contrast , adoptive transfer of cDCs , but not CD11cint/lo cells , restored the frequency ( Figure 9F ) and absolute numbers ( Figure 9G ) of IFNγ+IL-10+ CD4+ T cells to that observed in untreated infected mice . cDCs are therefore required to promote in vivo expansion of IFNγ+IL-10+ CD4+ T cells during L . donovani infection .
This study is the first to demonstrate that CD11c+ cells act to promote disease progression during the chronic phase of infection with L . donovani . Furthermore , by combining conditional DTx-mediated depletion with adoptive transfer during ongoing infection , we could show that whereas both CD11chi and CD11cint/lo cells contribute to disease progression and suppress host protective immunity , only CD11chi cells are capable of promoting the expansion and/or maintenance of Th1 cells that produce IL-10 . In addition to providing the first evidence that cDCs are required to promote expansion of CD4+ T cells with mixed effector/regulatory phenotype in vivo , our data suggests that the emergence of Th1 cells producing IL-10 is not essential for disease progression . Unlike previous studies assessing the role of DCs in acute L . donovani infection [48] or Langerhans cells in acute L . major infection [15] , sustained ablation of CD11c-expressing cells during ongoing infection was required in this study . Under the conditions used , we could deplete 80–90% of splenic CD11chiMHCIIhi cells in mice chronically infected with L . donovani , similar to the efficacy reported for DC depletion during Schistosoma mansoni infection [49] . The impact of CD11c+ cell ablation on disease progression was striking , with reduced splenic pathology , enhanced nitric oxide ( NO ) production and enhanced parasite clearance . Indeed , the impact of CD11c depletion was of a similar magnitude to that observed after a variety of chemotherapeutic and immunotherapeutic interventions [50] . Neutrophil influx has recently been noted following DTx treatment of mice [46] , [47] and was also observed by us here . It is unclear whether CXCL2-mediated egress from the bone marrow underlies the neutrophilia in ( CD11c-cre×Rosa26iDTR ) F1 mice , as suggested for other strains [47] . In preliminary studies , we have also noted a significant increase in the abundance for Il17α mRNA after DTx treatment ( data not shown ) , but further studies are required to determine whether this cytokine also impacts on neutrophil recruitment . In spite of this influx of phagocytes , simultaneous depletion of neutrophils in vivo demonstrated that even in their absence , the ablation of CD11c+ cells resulted in a marked reduction in splenomegaly and increased levels of NO production . Of note , parasite burden was increased in neutrophil depleted DTx treated mice , even though NO levels also increased , suggesting that the leishmanicidal effect of neutrophils may be mediated through NO-independent mechanisms . Nevertheless , to more directly overcome the pitfalls associated with off-target effects of DTx treatment , and to more formally address the role of cDCs in disease progression , we employed a strategy that allowed simultaneous depletion of endogenous CD11c+ cells and reconstitution with subpopulations of wild type CD11c+ cells . Similar to a recent study of allergic inflammation [51] , the transfer of relatively small numbers of highly purified CD11chi cDCs to CD11c-ablated mice resulted in substantial modulation of disease . All parameters of host resistance that we measured were suppressed after CD11c+ cell transfer and splenomegaly returned to the level observed in untreated infected mice . Importantly , we could not distinguish between the effects of transfer of CD11chi cDCs and CD11cint/lo cells based on these criteria . Splenic cDCs capable of promoting disease progression had a number of characteristics that distinguish them from cDCs found in naïve mice , with only minor differences seen between subsets . CD80 and CD86 expression is muted during chronic infection , similar to what has been observed at early times ( ∼5 hr ) post infection [18] , [52] . CD40 expression was somewhat higher at day 28 p . i . than at d21 p . i . , but there are conflicting reports as to whether the CD40-CD40L axis is required [53] , [54] or redundant [55] , [56] with respect to anti-Leishmania responses . It has been suggested that signaling downstream of CD40 may perpetuate IL-10 production and enhance productive infection of macrophages [57] but this has not been evaluated in DCs . cDCs from chronically infected mice showed enhanced expression of Programmed Death Ligand 1 ( PD-L1 ) , a negative costimulatory molecule involved in regulating functional exhaustion of CD8+ T cells during L . donovani infection [58] . IL-12p70 production was severely impaired , whereas high levels of spontaneous IL-10 and IL-27 were observed , with some differences between subsets also reflected in previous data at the mRNA level [18] . IL-10 produced by cDCs in infected mice appears to play an important autocrine regulatory role in limiting IL-12p70 production . Previous work has identified splenic cDCs as being particularly sensitive to autocrine regulation of IL-12 production by IL-10 in vitro [42] and although the molecular mechanisms for such a process are not fully described , Stat3 has been shown to mediate some of the inhibitory effects of IL-10 on cDC activation in vivo [59] . In contrast , we found no evidence supporting a role for IL-27 in the regulation of IL-12p70 or IL-10 production in cDCs , despite evidence that BMDCs generated from IL-27Rα−/− mice show enhanced production of IL-12p40 and p70 in response to TLR ligation [60] and that autocrine IL-27 is required for optimal macrophage IL-10 production [61] . The CD11cint/lo population capable of transferring disease progression includes NK cells and CD11cloCD45RB+ ‘regulatory’ DCs , two populations that we have previously shown to produce IL-10 and contribute to immunopathology using other assay systems [21] , [43] . These data are also consistent with evidence that IL-10 production by innate cells , rather than T cells , is the dominant negative regulator of effector responses after vaccination against L . major [62] , and that therapeutic infusion of LPS-activated BMDCs reduces pathology and parasite load , irrespective of whether the splenic IL-10-producing CD4+ T cell frequency is reduced or maintained [23] . IL-10 is a well known suppressor of IFNγ-induced NO production [63] . Hence , the loss of IL-10 expressing cDCs and CD11cint cells in an IFNγ-replete environment may underlie the increased NO production observed after therapeutic depletion of CD11c+ cells . Unlike previous data showing a requirement for myeloid DCs in the generation of effector responses to acute L . donovani infection [48] , the ablation of CD11c+ cells during chronic infection did not significantly affect IFNγ production by CD4+ T cells , suggesting that neither cDCs nor other CD11c+ cells are essential for the maintenance of effector T cell responses , at least over the 7 day time frame studied here . This result is in keeping with data showing the effects of DC ablation on CD4+ T cell responses to M . tuberculosis , where DCs are critical for initial priming of CD4+ effector T cell responses but dispensable for recall Th1 responses after vaccination [6] . However our data are in contrast to a report showing that CD11c-depletion during established infection with S . mansoni resulted in a significant reduction in IFNγ+ production by CD4+ T cells after ablation of CD11c-expressing cells [49] . Hence , the requirement for CD11c+ cells to maintain T cell IFNγ production would appear to be context and/or infection-specific . Chronic L . donovani infection is associated with the expansion of CD4+ T cells that co-express both IFNγ and IL-10 ( this manuscript and [23] , [24] ) . The further detailed characterization of this population provided here indicates that during L . donovani infection , these IL-10-producing CD4+ IFNγ+ T cells are Foxp3− , T-bet+ and CD127− . Hence , they are likely to be related to the IL-10-producing effector T cells found in experimental Toxoplasma gondii and Listeria monocytogenes infection [35] , [64] and Plasmodium infection [31] , [65] . Although T-bet and CD127 expression were not directly addressed , Foxp3− ‘effector’ CD4+ T cells also appear to be the major CD4+ T cell subset producing IL-10 during infection of mice with L . major [66] , [67] . However the increase in Foxp3+ natural Treg frequency and suppressive activity seen in cutaneous L . major infection [68] appears to be absent during EVL ( this manuscript and [23] ) . IFNγ+IL-10+ cells are also associated with L . donovani and Mycobacterium tuberculosis infection in humans [69] , [70] . Multiple lines of evidence have suggested a link between IL-27 and the production of IL-10 by CD4+ T cells in vitro or during autoimmune processes [27] , [28] , [29] , [30] , [32] , [33] . IL-27 plays a role in the development of IFNγ+IL-10+ CD4+ T cells during experimental L . major [71] and Listeria monocytogenes [35] infection in vivo , as assessed using IL-27Rα−/− mice . Such mice also display enhanced resistance to infection with L . donovani , although T cell IL-10 production was not assessed [72] . Furthermore , systemic IL-27 levels are elevated in humans infected with L . donovani , with splenic myeloid cells providing a major source of IL-27 mRNA that was proposed to enhance IFNγ+IL-10+ T cell responses via the induction of T cell-derived IL-21 [73] . Although there is some in vitro evidence of DC-derived IL-27 inducing T cell IL-10 production [37] , [38] , none of the studies described above provided a formal and causal link between IL-27-producing cDCs and IFNγ+IL-10+ CD4+ cell polarization in vivo , as we have now demonstrated here . The capacity for cDCs to drive polarization of IFNγ+IL-10+ cells may be as a result of their higher expression of IL-27 than CD11cint cells during infection . However , we believe it is likely that a combination of cytokine profile and the capacity for sustained antigen presentation underlies the essential requirement for CD11chi cDCs in the generation of IFNγ and IL-10 co-producing CD4+ T cells [64] , [74] , rather than sole production of IL-27 . Further study and the development of mice with targeted deficiency of IL-27p28 will be required to delineate the relative contribution of these events in vivo . Although T cell-derived IFNγ has long been known as a critical mediator of parasite clearance in EVL [75] , the role of Th1 cells producing mixed effector/regulatory cytokines still remains to be clearly established . During infections where pro-inflammatory responses would otherwise be rampant , this phenotype appears essential to minimize host-mediated pathology [31] , [64] . However , the association of IFNγ+IL-10+ CD4+ T cells with disease progression in leishmaniasis has suggested that through IL-10 production , these cells may contribute to parasite persistence and/or disease pathology . The data provided in this manuscript suggest that at least in EVL this association is not causal , as splenomegaly and loss of host resistance were equally well promoted by CD11cint/lo cells as by CD11chi cells , even though the former failed to promote the expansion/maintenance of IL-10-producing Th1 cells . A similar conclusion was also drawn from reciprocal studies using a model of DC immunotherapy [23] . Our data indirectly suggest , therefore , that IL-10 derived from other cellular sources is sufficient to counterbalance the otherwise potentially fatal consequences of Th1-derived effector cytokines . On a cautionary note , whilst it is tempting to conclude that there is a causal link between the phenotype of cDCs ( and CD11cint cells ) , the regulation of T cell immunity , parasite containment and the development of pathology , this may not be the case , given the complexity of potential interactions in vivo . For example , we have shown that cDCs from infected mice retain the capacity for antigen presentation in vitro ( data not shown ) and altered activation of T cells is clearly a consequence of cDC transfer . Studies involving the transfer of MHC-deficient cDCs ( and CD11cint cells ) would be required , however , to discover whether cDCs and CD11cint cells also have the potential to regulate pathology independently of their capacity to interact in a cognate manner with T cells , e . g . by influencing local stromal or myeloid cell function or directly regulating vascular remodeling [43] , [45] , [50] . Finally , the recognition that cDCs switch from a host protective role in the induction of immunity [48] to one in which they may hinder parasite elimination may provide exciting new opportunities for targeted immunotherapy . Indeed , the impact of CD11c+ cell depletion was of a similar magnitude to that observed after a variety of chemotherapeutic and immunotherapeutic interventions ( reviewed in [76] ) . To our knowledge , DC depletion with the aim of overcoming immunosuppression has not been attempted in the clinic , though removing excessively stimulatory DCs has been suggested as a therapeutic approach to prevent GVHD after allogeneic hematopoietic stem cell transplantation [77] . Whether DCs would serve as potential targets for short-term antibody-based immunotherapy in human VL remains to be determined and will require further concerted efforts to first characterize human DC subsets and their function during this disease .
All animal care and experimental procedures were carried out after review and approval by the University of York Ethical Review Process , and conducted under the authority of United Kingdom Home Office Project Licence PPL 60/3708 ( ‘Immunology and Immunopathology and visceral leishmaniasis’ ) . All experiments were designed and conducted to minimise suffering and to comply with the principles of replacement , refinement and reduction . C57BL/6 and B6J . CD45 . 1 mice were obtained from the Biological Services Facility ( University of York ) or supplied by Charles River Laboratories . C57BL/6J-Tg ( Itgax-cre-EGFP ) 4097Ach/J ( CD11c-cre ) mice and C57BL/6-Gt ( ROSA ) 26Sortm1 ( HBEGF ) Awai/J ( Rosa26iDTR ) mice were obtained from The Jackson Laboratory ( Bar Harbor , Maine , USA ) and Cre and eGFP genotype positive F1 mice were used between 6 and 12 weeks of age . Mice were infected via the lateral tail vein with 3×107 amastigotes of the Ethiopian strain of Leishmania donovani ( LV9 ) . Splenomegaly was calculated relative to body weight and parasite burdens were quantified as Leishman-Donovan Units ( LDU ) [23] . ( CD11c-cre×Rosa26iDTR ) F1 mice were treated with 4 ng/g Diphtheria toxin from Corynebacterium diptheriae ( DTx , Sigma ) i . p . and/or treated with mAb 1A8 or control [45] every other day from day 21 of infection , as required . Administration of 1A8 resulted in depletion of ∼90% of splenic neutrophils , as judged by CD11b , Ly6C and Gr-1 staining ( data not shown ) . Spleen cells were restimulated for either 90–120 min with 10 ng/ml PMA and 1 µg/ml Ionomycin ( Sigma-Aldrich , UK ) or for 3 h with BMDCs pulsed with fixed L . donovani amastigotes , and then further incubated with 1 µg/ml Brefeldin A for 4 h . After restimulation , cells were labeled for 30 minutes on ice with mAbs: CD3ε-PE-Cy7 ( 145-2C11 ) , CD4-FITC ( RM4-5 ) , CD127-PE ( A7R34 ) ( eBioscience ) or CD4-PerCP ( RM4-5; BD Pharmingen ) . Cells were washed and incubated for 30 min on ice in PBS containing Fixable Viability Dye eFluor780 ( eBioscience ) . After washing , cells were fixed ( 15 min on ice ) in 2% paraformaldehyde ( PFA ) . Cells were then permeabilised using 1% Saponin ( Sigma PERM buffer ) . Cells were subsequently labeled ( 45–60 min on ice ) in PERM buffer containing mAbs: IFNγ-PacificBlue ( XMG1 . 2 ) , IFNγ-eFluor450 ( XMG1 . 2 ) , IL-10-APC ( JES5-16E3 ) , IL-10-PE ( JES5-16E3 ) T-bet-AlexaFluor647 ( ebio4BIO ) and Foxp3-FITC ( FJK-16a ) , ( eBioscience ) . All cells were analyzed on a CyAN-ADP flow cytometer using Summit Software ( Beckman Coulter , USA ) . BMDCs were generated from femurs of C57BL/6 mice using standard methods . On day 7 of culture , cells were pulsed for 24 hrs with paraformaldehyde-fixed Leishmania donovani amastigotes at a ratio of 100 amastigotes to 1 BMDC . Antigen-pulsed BMDCs were subsequently used to restimulate T cells for 3 hours , prior to addition of Brefeldin A for 4 hours and subsequent assessment of CD4+ T cell cytokine production by intracellular flow cytometric analysis , as previously described . Spleens were dissociated mechanically and digested in 0 . 2 mg/ml collagenase type IV/DNAse1 mix ( Worthington Biochemical , NJ , USA ) for 30 minutes at room temperature . Staining was performed as above using mAbs: CD11c-PE-Cy7 ( N418 ) , Major Histocompatibility complex class II ( MHCII ) -APC ( M5/114 . 15 . 2 ) , MHCII-eFluor450 ( M5/114 . 15 . 2 ) , CD8α-FITC ( 53-6 . 7 ) , CD4-APC ( RM4-5 ) , CD40-PE ( 1C10 ) , CD80-PE ( 16-10A1 ) , CD86-PE ( GL1 ) and B7-H1-PE ( MIH5 ) . For purification , CD11c+ cells were enriched by magnetic separation [18] and CD11chiMHCIIhi cDCs or individual cDC subsets were sorted to ∼98–99% purity on a BeckmanCoulter MoFlo cell sorter . After sorting , cells were washed , counted and plated in triplicate in complete RPMI at 1×106 cells/ml . Where indicated , LPS ( 1 µg/ml; Sigma ) , anti-mouse IL-27p28 or Goat IgG ( 10 µg/ml; R&D Systems ) or anti-mouse IL-10R ( 1 . 3 µg/ml; gift of Dr . M . Kullberg ) were added . After 24 h , supernatants were harvested and stored at −80°C until assessed by ELISA for levels of IL-12p40 , IL-10 ( Mabtech , Sweden ) , IL-27p28 and IL-12p70 ( R&D Systems ) . Splenocytes from naïve and infected PBS or DTx-treated mice were incubated in RPMI at 5×106 cells/ml for 60 min at 37°C , and non-adherent cells removed by vigorous washing . Adherent cells were cultured for 24 h at 37°C and supernatants were assayed using Greiss Reagent ( Promega , Madison , WI , USA ) . 12 h after the first administration of DTx , infected ( CD11c-cre×Rosa26iDTR ) F1 mice received 1 . 5×105 CD11chiMHCIIhi cells or 6 . 0×105 CD11cintMHCII+ cells isolated from day 21-infected B6J . CD45 . 1 mice . DTx administration was continued at 48 h intervals to maintain depletion of endogenous CD11c+ cells . Statistical analysis was performed using a students t test or one-way ANOVA as appropriate , with p<0 . 05 considered significant . All experiments were conducted independently at least twice .
|
Dendritic cells are well known as myeloid cells that bridge innate and adaptive immunity , and play an important role in the induction of cell-mediated immunity to a variety of pathogens . However , very little is known about the function of dendritic cells after infection has become established . In this study , we have examined the role of dendritic cells during the later phases of experimental visceral leishmaniasis , caused by infection with Leishmania donovani . We show for the first time that dendritic cells are responsible for promoting the development of splenic pathology , and that their removal during established infection leads to improved host resistance . Furthermore , our studies provide the first formal evidence in vivo that dendritic cells making IL-27 can induce production of the regulatory cytokine IL-10 by effector Th1-like CD4+ T cells . Surprisingly , we also found that other populations of CD11c+ cells were able to induce pathology and suppress host resistance , yet did not stimulate IL-10 production in CD4+ T cells , suggesting that the latter T cell population may not play an essential role in disease progression . Our studies provide new insights into dendritic cell function in chronic parasite infection and suggest potential new avenues for immunotherapy against visceral leishmaniasis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"immune",
"cells",
"cytokines",
"antigen-presenting",
"cells",
"immunology",
"microbiology",
"host-pathogen",
"interaction",
"parasitology",
"parastic",
"protozoans",
"neglected",
"tropical",
"diseases",
"immunomodulation",
"infectious",
"diseases",
"t",
"cells",
"microbial",
"pathogens",
"biology",
"pathogenesis",
"immune",
"response",
"immune",
"system",
"immunopathology",
"immunity",
"leishmaniasis",
"protozoology"
] |
2012
|
IL-10-Producing Th1 Cells and Disease Progression Are Regulated by Distinct CD11c+ Cell Populations during Visceral Leishmaniasis
|
To generate a cytopathic effect , the catalytic A1 subunit of cholera toxin ( CT ) must be separated from the rest of the toxin . Protein disulfide isomerase ( PDI ) is thought to mediate CT disassembly by acting as a redox-driven chaperone that actively unfolds the CTA1 subunit . Here , we show that PDI itself unfolds upon contact with CTA1 . The substrate-induced unfolding of PDI provides a novel molecular mechanism for holotoxin disassembly: we postulate the expanded hydrodynamic radius of unfolded PDI acts as a wedge to dislodge reduced CTA1 from its holotoxin . The oxidoreductase activity of PDI was not required for CT disassembly , but CTA1 displacement did not occur when PDI was locked in a folded conformation or when its substrate-induced unfolding was blocked due to the loss of chaperone function . Two other oxidoreductases ( ERp57 and ERp72 ) did not unfold in the presence of CTA1 and did not displace reduced CTA1 from its holotoxin . Our data establish a new functional property of PDI that may be linked to its role as a chaperone that prevents protein aggregation .
Protein disulfide isomerase ( PDI ) is a member of the thioredoxin superfamily with an abb'xa'c structural organization that consists of two catalytic domains ( a & a′ ) separated by two non-catalytic domains ( b & b′ ) and an short x linker , along with an acidic C-terminal c extension [1]–[3] . It is mainly located in the endoplasmic reticulum ( ER ) where it exhibits linked but independent oxidoreductase and chaperone activities . These activities allow it to facilitate the proper folding of nascent secretory proteins as well as the disposal of terminally misfolded proteins through the quality control mechanism of ER-associated degradation ( ERAD ) . The structure and function of PDI is regulated by its redox status: it is a dynamic , flexible molecule which assumes a compact conformation in the reduced state and a more open conformation in the oxidized state [4]–[6] . PDI thus acts as a redox-dependent chaperone in its interactions with certain substrate proteins [7]–[10] . The chaperone function of PDI is defined by its ability to prevent the aggregation of misfolded proteins [11]–[14] . The importance of this activity is highlighted by the link between PDI dysfunction and neurodegeneration: a S-nitrosylated form of PDI that cannot prevent protein aggregation is found in the brains of individuals with Parkinson's or Alzheimer's disease [15] . PDI also prevents the aggregation of α-synuclein which occurs in Parkinson's disease [16] , [17] . PDI can even prevent protein aggregation when added to the substrate 40 minutes after aggregation has begun [18] . This strongly suggests the chaperone function of PDI involves something other than simply binding and masking the solvent-exposed hydrophobic amino acid residues of a misfolded protein . However , the molecular mechanism of PDI chaperone function remains unknown . PDI also plays a key role in cholera intoxication . Cholera toxin ( CT ) is an AB5 toxin that consists of a catalytic A1 subunit , an A2 linker , and a cell-binding B pentamer ( Fig . S1 ) [19] . It moves by vesicle carriers from the cell surface to the ER where the A1 subunit dissociates from the rest of the toxin [20] . The free A1 subunit then shifts to a disordered conformation which allows it to exploit the ERAD system for export to the cytosol [21]–[24] . The translocated pool of CTA1 interacts with host factors in the cytosol to regain an active conformation , and it avoids proteasomal degradation long enough to effectively modify its Gsα target [24]–[27] . CTA1 is anchored to CTA2 by a single disulfide bond and numerous non-covalent interactions . Reduction of the A1/A2 disulfide bond can occur at the resident redox state of the ER [28] , yet reduction alone is not sufficient for holotoxin disassembly [29]: PDI must displace the reduced A1 subunit from the rest of the toxin [7] , [8] . This process is essential for intoxication , as PDI-deficient cells are completely resistant to CT [7] . PDI was originally thought to actively unfold the holotoxin-associated CTA1 subunit and to thereby displace CTA1 from the rest of the toxin [8] . This model was based upon the results of a protease sensitivity assay that only provided an indirect measure of protein structure . An alternative explanation for the “unfoldase” activity of PDI was suggested by our later work which demonstrated the intrinsic instability of CTA1 will allow it to spontaneously unfold upon its separation from CTA2/CTB5 at physiological temperature [24] . Thus , PDI could trigger toxin unfolding simply by removing CTA1 from the CT holotoxin . Our recent biophysical analysis provided experimental support for this alternative model and demonstrated that PDI does not unfold CTA1 [7] . We also found that PDI exhibits conformation-dependent interactions with CTA1: PDI recognizes the folded conformations of CTA1 present at low temperatures and in the CT holotoxin , but it does not bind to the disordered , 37°C conformation of free CTA1 [7] . Consistent with previous reports [8] , we also noted only the reduced form of PDI will interact with CT and CTA1 . This interaction did not appear to involve the oxidoreductase activity of PDI , as PDI did not form mixed disulfides with CTA1 and could bind to cysteine-free CTA1 deletion constructs [7] , [8] . In this work we employed a biophysical and biochemical approach to define the structural basis for PDI-mediated disassembly of the CT holotoxin . Using isotope-edited Fourier transform infrared ( FTIR ) spectroscopy and circular dichroism ( CD ) , we have demonstrated that PDI unfolds upon contact with CTA1 . The substrate-induced unfolding of PDI provides a molecular explanation for holotoxin disassembly: the expanded hydrodynamic radius of unfolded PDI would act as a lever to dislodge reduced CTA1 from its non-covalent association with the rest of the toxin . In support of this model , we found the displacement of reduced CTA1 from CTA2/CTB5 does not occur when PDI is locked in a folded conformation or when PDI chaperone function is disrupted by ribostamycin treatment . Additional drug treatments with bacitracin indicated the oxidoreductase activity of PDI is not required for holotoxin disassembly . Consistent with our model , the substrate-induced unfolding of PDI is blocked by ribostamycin but not bacitracin . Two other ER-localized oxidoreductases ( ERp57 and ERp72 ) did not unfold in the presence of CTA1 and did not displace reduced CTA1 from its holotoxin . Substrate-induced unfolding thus appears to be a unique property of PDI that is linked to its chaperone function and could explain its ability to disrupt protein aggregation .
Far-UV CD spectroscopy was used to assess conformational changes in reduced PDI and CTA1 upon their interaction at 10°C and neutral pH ( Fig . 1 ) . The spectrum of PDI alone was dominated by two components around 208 and 221 nm that can be ascribed to ππ* and nπ* backbone α-helical electronic transitions , respectively . The spectrum of CTA1 displayed a major component around 221 nm and a shoulder between 210 and 216 nm , consistent with its α/β secondary structure identified by X-ray crystallography [30] , [31] . When PDI and CTA1 were combined in an equimolar ratio , the resulting spectrum was different from the sum of the two individual spectra of PDI and CTA1 , suggesting that protein-protein interactions result in conformational changes in either CTA1 , PDI , or both proteins . The spectral difference ( i . e . , [CTA1+PDI] - [CTA1] - [PDI] ) revealed two well defined peaks at 207 and 223 nm , as well as a deep minimum just above 190 nm , indicating a significant loss in the α-helical structure and a gain in the unordered structure . Loss of the PDI “double minima” α-helical signature from the spectrum of PDI+CTA1 combined sample implies that significant conformational changes occur in PDI . However , the spectral overlap of CD signals generated by both proteins prevents unambiguous assignment of the structural changes to one or the other protein . To resolve individual conformational changes in PDI and CTA1 upon their interaction , we used isotope-edited FTIR spectroscopy as described below . Isotope-edited FTIR spectroscopy allows the conformation of a protein to be monitored in the presence of a second , 13C-labeled protein [32]–[34] . 13C labeling does not alter the conformation of a protein . However , the heavier nuclear mass of the stable 13C isotope generates a spectral downshift which allows the FTIR spectrum of a 13C-labeled protein to be resolved from the spectrum of an unlabeled protein . The structures of both unlabeled and labeled proteins can thus be determined with isotope-edited FTIR spectroscopy . This technique allowed us , for the first time , to specifically and directly monitor the conformation of CTA1 in the presence of PDI . Our studies demonstrated that PDI does not unfold CTA1 [7] . Here , we used isotope-edited FTIR spectroscopy to examine the impact of toxin binding on the structure of PDI ( Fig . 2 ) . Our data indicated that PDI unfolds after it contacts CTA1 . In the absence of CTA1 , the 10°C structure of PDI exhibited a folded conformation with 37% α-helical and 42% β-sheet content ( Fig . 2A , Table 1 ) . These percentages were consistent with the secondary structural content predicted from the crystal structure of PDI [35] . In the presence of CTA1 at 10°C , reduced PDI lost substantial α-helical and β-sheet content ( Fig . 2B ) . The percentage of irregular PDI structure rose from 12% in the absence of CTA1 to 42% in the presence of CTA1 ( Table 1 ) . The toxin-induced loss of PDI structure did not occur in the absence of GSH ( Fig . 2C–D , Table 1 ) , which confirmed the specificity of our data: only reduced PDI can bind to CTA1 [7] , [8] . An additional set of FTIR experiments were performed at 37°C to further validate our findings . Reduced PDI does not bind to CTA1 at 37°C in pH 7 . 0 buffer [7] and did not exhibit an increase in irregular structure when incubated with CTA1 under this condition ( Fig . 2E–F , Table 1 ) . However , reduced PDI can bind to CTA1 at 37°C in pH 6 . 5 buffer [7] . Acidified medium stabilizes the CTA1 polypeptide , allowing it to retain a substantial amount of its secondary structure at physiological temperature [21] . Thus , as expected , PDI shifted to a disordered conformation with 40% irregular structure when mixed with CTA1 at 37°C in pH 6 . 5 buffer ( Fig . 2G–H , Table 1 ) . These collective observations demonstrated the unfolding of PDI resulted from its physical interaction with CTA1 . The substrate-induced unfolding of PDI was a reversible event . When CTA1 was added to reduced PDI at 10°C , the folded conformation of PDI shifted to a disordered state ( Fig . 2A–B , Table 1 ) . However , PDI returned to a folded conformation upon warming the PDI/CTA1 complex to 37°C ( Fig . 3 , Table 1 ) . CTA1 unfolds at 37°C , and this unfolding event displaces its PDI binding partner [7] . The displacement of reduced PDI from CTA1 at physiological temperature thus allowed PDI to regain a folded structure . An interaction with the CT holotoxin also resulted in the unfolding and refolding of PDI ( Fig . 4 , Table 1 ) . For this experiment , 13C-labeled PDI was mixed with the CT holotoxin at 37°C and pH 7 . 0 . The holotoxin-associated CTA1 subunit maintains a folded conformation at 37°C and neutral pH [36] , so reduced PDI can bind to holotoxin-associated CTA1 under physiological conditions . However , the spontaneous unfolding of CTA1 which occurs after holotoxin disassembly at 37°C results in the displacement of its PDI binding partner [7] . This process allowed us to monitor both the unfolding of PDI upon its interaction with the CT holotoxin and the refolding of PDI after its release from the dissociated CTA1 subunit . One minute after exposure to the CT holotoxin , reduced PDI exhibited a disordered conformation containing 40% irregular structure ( Fig . 4A , Table 1 ) . This was similar to the percentage of irregular structure in reduced PDI upon its binding to free CTA1 at 10°C and neutral pH or at 37°C and pH 6 . 5 . After holotoxin disassembly , PDI could no longer interact with CTA1 and consequently returned to a folded conformation within 25 minutes of exposure to the CT holotoxin ( Fig . 4B , Table 1 ) . As seen from Table 1 , the conformational changes in PDI between 1 and 25 minutes of its interaction with the CT holotoxin at 37°C involve an increase in the β-sheet fraction by 16% and an increase in the α-helical structure by only 3% , implying that the refolding of PDI begins with a gain of α-helical structure which is followed by a gain of β-sheet structure . The PDI-mediated displacement of reduced CTA1 from the CT holotoxin and subsequent release of PDI from the dissociated , unfolded CTA1 polypeptide are both extremely rapid events ( [7] , see also Fig . 5 ) . Our data suggest these events had already occurred within 1 minute of combining PDI with CT , and the process of PDI refolding was already underway at our first time point . Technical limitations prevented the measurement of PDI structure before 1 minute of incubation with CT . Nonetheless , the data from Figures 2–4 collectively demonstrated that PDI unfolding occurs upon binding to either free CTA1 or holotoxin-associated CTA1 , and the results further indicated that PDI will return to a folded state after the release of its bound substrate . The unfolding of PDI provides a molecular basis for the PDI-mediated disassembly of CT: unfolding will expand the hydrodynamic radius of PDI , thus acting as a wedge to dislodge reduced CTA1 from its non-covalent association with CTA2/CTB5 . To test this model , we locked PDI in a folded conformation by treating it with 400 mM EDC . Previous work has shown this “zero-length” intramolecular cross-linker will activate carboxylic side chains for reaction with nearby primary amines on lysine residues [37] , [38] . A standard cross-linking molecule is absent from this process , so the reactive side chains can only act on intramolecular targets . SDS-PAGE and size exclusion chromatography were used to confirm the absence of PDI dimers or oligomers after EDC treatment ( Fig . S2A–B ) . Isotope-edited FTIR spectroscopy further demonstrated that EDC-treated PDI did not unfold in the presence of CTA1 ( Fig . S2C–D , Table 1 ) : the secondary structure content of EDC-treated PDI in either the absence or presence of CTA1 was similar to the secondary structure content of the isolated , untreated PDI polypeptide . To examine the functional role of PDI unfolding in toxin disassembly , untreated PDI and EDC-treated PDI were perfused over a surface plasmon resonance ( SPR ) sensor coated with the CT holotoxin ( Fig . 5 ) . A functional interaction between PDI and the toxin will result in the displacement of CTA1 from the SPR sensor and a corresponding drop in the refractive index unit ( RIU ) below the baseline value corresponding to the mass of the initial sensor-bound holotoxin [7] . When PDI was perfused over the CT-coated sensor under reducing conditions , we detected a rapid rise in RIU which was indicative of PDI binding to the toxin . This was followed by a drop in the RIU signal to a point below the initial baseline value ( Fig . 5A ) . Identical results were obtained with either 30 mM GSH ( left panel ) or 1 mM GSH ( right panel ) in the perfusion buffer; non-reducing SDS-PAGE with Coomassie staining demonstrated the CTA1/CTA2 disulfide bond is reduced at 30 mM GSH but not 1 mM GSH ( Fig . 5A , right panel inset ) . The loss of signal around 200 seconds occurred even though PDI was still present in the perfusion buffer , suggesting that both PDI and CTA1 had been lost from the sensor . Sequential perfusions of anti-PDI , anti-CTA1 , and anti-CTB antibodies over the PDI-treated slide confirmed this interpretation: only the anti-CTB antibody gave a positive response ( Fig . 5A ) . Perfusion of an anti-KDEL antibody over a PDI-treated sensor also gave a positive response ( not shown ) , which indicated the KDEL-containing CTA2 subunit remained associated with CTB5 after the release of CTA1 . This observation was consistent with previous reports [7] , [39] , and it demonstrated that PDI specifically removes CTA1 from the sensor-bound CTA2/CTB5 complex . EDC-treated PDI bound tightly to the CT holotoxin under reducing conditions but did not displace CTA1 from CTA2/CTB5 ( Fig . 5B ) . Identical results were obtained with either 30 mM GSH ( left panel ) or 1 mM GSH ( right panel ) in the perfusion buffer . Positive signals for the anti-PDI , anti-CTA1 , and anti-CTB antibodies demonstrated that EDC-treated PDI remained associated with the intact CT holotoxin after removal from the perfusion buffer . The locked conformation of PDI thus exhibited a high affinity interaction with CT , but it lacked the mechanism required to separate CTA1 from the rest of the toxin . Our FTIR data strongly suggested this missing mechanism involves the toxin-induced unfolding of PDI . Intramolecular cross-linking could disrupt the enzymatic activity of PDI , although previous work has suggested an oxidoreductase function is not required for PDI to displace CTA1 from CTA2/CTB5 . To confirm this observation , bacitracin-treated PDI was perfused over a CT-coated SPR sensor under reducing conditions . Bacitracin is a peptide antibiotic that inhibits the reductive activity of PDI [40] , but it did not inhibit the toxin-induced unfolding of PDI as assessed by isotope-edited FTIR spectroscopy ( Fig . S3A–B , Table 1 ) . Bacitracin-treated PDI could displace reduced CTA1 from its non-covalent association with CTA2/CTB5 ( Fig . 5C , left panel ) , but it could not separate CTA1 from the rest of the toxin when the CTA1/CTA2 disulfide bond was intact ( Fig . 5C , right panel ) . In contrast , untreated PDI could mediate the disassembly of a CT holotoxin with an intact disulfide bond [7] , [8] ( Fig . 5A , right panel ) . These results indicated that the oxidoreductase activity of PDI could cleave the CTA1/CTA2 disulfide bond and that this activity was inhibited by bacitracin . Thus , bacitracin-treated PDI did not require an enzymatic function to dislodge reduced CTA1 from its non-covalent association with the rest of the toxin . By extension , the inability of EDC-treated PDI to displace reduced CTA1 from its holotoxin ( Fig . 5B , left panel ) could not be attributed to a loss of enzymatic function . PDI can , independently of its oxidoreductase function , act as a chaperone to prevent the aggregation of misfolded proteins [12] , [13] . To determine whether the toxin-induced unfolding of PDI was related to its role as a chaperone , we used S-nitrosylation and ribostamycin to disrupt the chaperone activity of PDI [15] , [41] . Nitrosylated PDI and ribostamycin-treated PDI were then used in our toxin disassembly assay . As shown in Fig . 6A , nitrosylated PDI could not bind to the CT holotoxin . The inability of nitrosylated PDI to prevent protein aggregation [15] , [42] thus appears to result from the loss of substrate binding . Ribostamycin-treated PDI could bind to the CT holotoxin but could not separate reduced CTA1 from CTA2/CTB5 ( Fig . 6B ) . Removal of ribostamycin-treated PDI from the perfusion buffer resulted in a rapid drop in RIU to the initial baseline value corresponding to the mass of the CT holotoxin . Anti-PDI , anti-CTA1 , and anti-CTB antibody controls confirmed that ribostamycin-treated PDI had dissociated from the intact CT holotoxin . Furthermore , isotope-edited FTIR spectroscopy demonstrated that ribostamycin-treated PDI did not unfold in the presence of CTA1 ( Fig . 6C–D , Table 1 ) . The loss of chaperone activity for ribostamycin-treated PDI thus corresponded to an inhibition of both PDI unfolding and holotoxin disassembly . Cells treated with 50 µM ribostamycin were almost completely resistant to CT ( Fig . 7A ) . However , no protective effect was observed in cells transfected with a plasmid encoding a CTA1 construct that is co-translationally targeted to the ER before dislocation back into the cytosol ( Fig . 7B ) . This expression system mimics the translocation events occurring after holotoxin disassembly [43] and was used to ensure ribostamycin treatment did not affect the intoxication process downstream of PDI-mediated toxin disassembly . We also found that the CTA1/CTA2 disulfide bond could be reduced in ribostamycin-treated cells ( Fig . 7C ) . This result indicated that ribostamycin does not inhibit toxin transport to the ER , as reduction of the CTA1/CTA2 disulfide bond takes place in the ER [28] , [44] . In this experiment , brefeldin A ( BfA ) was used a positive control to demonstrate that CTA1/CTA2 reduction does not occur when toxin transport to the ER is blocked [45] . Ribostamycin did not block CT transport to the ER or reduction of the CTA1/CTA2 disulfide bond , yet only a minimal quantity of CTA1 could be detected in the cytosol of ribostamycin-treated cells ( Fig . 7D ) . This was consistent with the toxin-resistant phenotype of ribostamycin-treated cells and indicated ribostamycin prevents the in vivo displacement of reduced CTA1 from CTA2/CTB5 . Collectively , our data demonstrated that ribostamycin does not disrupt ( i ) toxin trafficking to the ER; ( ii ) reduction of the CTA1/CTA2 disulfide bond in the ER; ( iii ) translocation of the free CTA1 subunit from the ER to the cytosol; or ( iv ) CTA1 activity in the cytosol . The block of intoxication in ribostamycin-treated cells was therefore most likely due to the inhibition of PDI unfolding which facilitates holotoxin disassembly ( Fig . 6 ) . This result demonstrated the critical role of PDI unfolding in the CT intoxication process . Reduced CTA1 could not be separated from its holotoxin by ERp57 ( Fig . 8A ) or ERp72 ( Fig . 8B ) , two ER-localized oxidoreductases with an overall domain structure similar to PDI [1] , [2] . ERp57 and ERp72 bound to the CT holotoxin under reducing conditions , but they did not dislodge CTA1 from CTA2/CTB5 . Indeed , as demonstrated with our antibody controls , ERp57 and ERp72 remained stably associated with the intact CT holotoxin . Additional SPR experiments documented direct binding of ERp57 and ERp72 to the CTA1 subunit ( Fig . S4 ) . ERp57 usually binds to glycosylated substrates in a complex with calnexin or calreticulin , but a direct interaction between ERp57 and its substrate has been reported as well [46] . Isotope-edited FTIR spectroscopy further demonstrated that ERp57 and ERp72 do not unfold upon contact with CTA1 ( Fig . 8C–F , Table 1 ) . The toxin-induced unfolding of PDI and the PDI-mediated displacement of CTA1 from CTA2/CTB5 thus appear to be unique , linked properties of PDI that are not shared by other oxidoreductases . This is consistent with the inability of CT to affect PDI-deficient cells [7]: if other resident ER oxidoreductases could perform the same function as PDI , then PDI-deficient cell lines would not be resistant to CT .
CT moves from the cell surface to the ER as an intact AB holotoxin . The CTA1/CTA2 disulfide bond is reduced in the ER , but this is insufficient for holotoxin disassembly: PDI must displace reduced CTA1 from the rest of the toxin . Our biophysical analysis has provided a structural explanation for this event . We have shown by isotope-edited FTIR spectroscopy and CD that PDI unfolds upon contact with CTA1 . A real-time holotoxin disassembly assay demonstrated that the displacement of reduced CTA1 from CTA2/CTB5 does not occur when PDI is locked in a folded conformation or when the substrate-induced unfolding of PDI is blocked due to the loss of its chaperone function . However , the oxidoreductase activity of PDI was not required for this event . The toxin-induced unfolding of PDI provides a molecular basis for holotoxin disassembly: the expanded hydrodynamic radius of unfolded PDI would act as a wedge to physically displace reduced CTA1 from the rest of the toxin . ERp57 and ERp72 did not unfold in the presence of CTA1 and did not displace reduced CTA1 from its holotoxin . Substrate-induced unfolding thus appears to be a unique property of PDI . PDI does not directly interact with CTA2 or the CTB pentamer; it only recognizes the folded conformation of the CTA1 subunit [7] . In order for PDI to dislodge CTA1 from the CT holotoxin , it must bind to a region of CTA1 near the CTA2/CTB5 interface . The expanded hydrodynamic radius of PDI resulting from its toxin-induced unfolding would then push against two components of the holotoxin and thereby dislodge the A1 subunit from its non-covalent association with the rest of the toxin . PDI was originally thought to interact with the C-terminal hydrophobic A13 subdomain of CTA1 [47] , which is distal to the CTA2/CTB5 interface ( Fig . S1 ) . However , binding assays with CTA1 deletion constructs have demonstrated the A13 subdomain is not required for PDI-CTA1 interaction [7] . Binding instead occurred in a region of CTA1 ( residues 1–133 ) that is , in part , proximal to CTA2/CTB5 . The exact location of the PDI binding site on CTA1 remains to be determined , and this information is important for further elucidation of the CT disassembly mechanism . However , the current data are consistent with our model for the physical , PDI-mediated displacement of reduced CTA1 from its holotoxin . PDI binds to the folded conformations of CTA1 that are present at low temperature and in the CT holotoxin [7] . This induces the partial unfolding of PDI ( Figs . 1–2 , Fig . 4 , Table 1 ) , but disordered PDI still remains associated with CTA1 [7] . The modular structure of PDI , which consists of a rigid b′ substrate binding domain flanked by other more flexible domains [2] , [4] , [48] , likely accounts for the ability of partially disordered PDI to remain associated with its folded CTA1 partner . As shown in our recent publication , PDI is only displaced from CTA1 when the toxin unfolds [7] . CT disassembly thus appears to involve the following events: ( i ) the CTA1/CTA2 disulfide bond is reduced at the resident redox state of the ER [28] , but CTA1 remains associated with CTA2/CTB5 through non-covalent interactions [29]; ( ii ) reduced PDI binds to holotoxin-associated CTA1 [7] , [8]; ( iii ) the substrate-induced unfolding of PDI results in the separation of CTA1 from CTA2/CTB5 ( this work ) ; and ( iv ) the dissociated CTA1 subunit spontaneously unfolds at 37°C [24] , which consequently displaces its PDI binding partner [7] . PDI regains its native conformation after release from CTA1 ( Figs . 3–4 ) . PDI and other oxidoreductases may assist reduction of the CT disulfide bond ( Fig . 5A , right panel ) [28] , [44] as observed for other ER-translocating AB toxins [49]–[51] . However , the essential and specific role of PDI in holotoxin disassembly appears to be the physical separation of reduced CTA1 from its holotoxin . The toxin-induced unfolding of PDI suggests a molecular mechanism for its role as a chaperone that prevents protein aggregation: by unfolding in the presence of an aggregation-prone substrate , PDI would act as a lever to dislodge individual proteins from the forming aggregate . With this model , treatments that block the chaperone activity of PDI should prevent the substrate-induced unfolding of PDI . S-nitrosylation and ribostamycin represent two such conditions , as it has already been shown that S-nitrosylated PDI and ribostamycin-treated PDI can no longer prevent protein aggregation [15] , [41] . S-nitrosylation and ribostamycin treatment also blocked the PDI-mediated disassembly of CT . In the case of S-nitrosylation , the inhibition of toxin disassembly resulted from the loss of substrate binding . Ribostamycin-treated PDI could bind to CT , but it did not undergo substrate-induced unfolding . Many chaperones prevent protein aggregation by a simple physical mechanism that involves binding and masking the exposed hydrophobic amino acid residues of a disordered protein . Given that ribostamycin-treated PDI could still bind to CTA1 , it is unlikely that ribostamycin disrupts the chaperone function of PDI through an inhibition of substrate binding . This strongly suggests the chaperone function of PDI involves an activity in addition to substrate binding; we propose this activity is linked to the substrate-induced unfolding of PDI . Consistent with this model , we also demonstrated that EDC-treated PDI binds to CTA1 but does not unfold in the presence of CTA1 and does not displace reduced CTA1 from CTA2/CTB5 . Bacitracin inhibited the enzymatic activity of PDI , but it did not affect the toxin-induced unfolding of PDI and did not prevent the displacement of reduced CTA1 from CTA2/CTB5 . This again suggested that the substrate-induced unfolding of PDI is linked to its chaperone function , as unfolding was blocked by ribostamycin ( an inhibitor of PDI chaperone function ) but not bacitracin ( an inhibitor of PDI oxidoreductase activity ) . Since holotoxin disassembly does not require the oxidoreductase function of PDI , the inhibitory effects of ribostamycin and EDC cannot be attributed to the potential disruption of PDI enzyme activity . Our work has established a new functional property of PDI that is linked to its role as a chaperone . This property of substrate-induced unfolding would not have evolved for the benefit of a bacterial pathogen . There must therefore be a normal , physiological role for PDI unfolding . We propose the action of PDI in CT disassembly is related to its established role as a chaperone that prevents protein aggregation: in both cases , the expanded hydrodynamic radius of unfolded PDI would act as a wedge to disrupt non-covalent macromolecular complexes . This event could also be related to the structural changes that occur when PDI expands from a compact , reduced conformation to its more open , oxidized state . Thus , in addition to elucidating the molecular details of PDI-mediated toxin disassembly , our data provide a possible mechanistic basis for the known but structurally uncharacterized chaperone function of PDI .
Ribostamycin , PDI , BfA , GM1 , GSH , CTA , S-nitrosoglutathione , and anti-CTB antibodies were purchased from Sigma-Aldrich ( St . Louis , MO ) . Bacitracin was purchased from Calbiochem ( La Jolla , CA ) , CT was from List Biological Laboratories ( Campbell , CA ) , and phosphate-buffered saline ( pH 7 . 4 ) with 0 . 05% Tween 20 ( PBST ) was from Medicago ( Uppsala , Sweden ) . ERp57 , the anti-ERp57 antibody , and the anti-ERp72 antibody were from Abcam ( Cambridge , MA ) . ERp72 and the anti-PDI antibody were purchased from Enzo Life Sciences ( Farmingdale , NY ) . The anti-CTA1 monoclonal antibody 35C2 [52] was a generous gift from Dr . Randall K . Holmes ( University of Colorado School of Medicine ) . The pOLR130 plasmid encoding mature human PDI with an N-terminal His6 tag [53] was generously provided by Dr . Lloyd Ruddock ( University of Oulu , Finland ) . Uniformly 13C-labeled 13C6-D-glucose and D2O were purchased from Cambridge Isotope Laboratories ( Andover , MA ) . Uniformly 13C-labeled CTA1-His6 was produced as described in [7] and purified as described in [22] . Escherichia coli strain BL21 pLysS transformed with pOLR130 was inoculated into 5 mL M9 minimal media containing 100 µg/mL of ampicillin and was grown at 37°C with shaking . The culture was then expanded in 200 mL M9 minimal media supplemented with 100 µg/mL of ampicillin and uniformly 13C-labeled 13C6-D-glucose as the sole metabolic carbon source . Incubation at 37°C with shaking continued until the culture reached an O . D . 600 of 0 . 2–0 . 3 . The culture was then induced with 1 mM IPTG for 4 hr at 37°C , followed by centrifugation of the cells at 3 , 500× g and 4°C for 20 min . The supernatant was discarded , and the cells were resuspended in extraction buffer containing 20 mM Tris-HCl ( pH 7 . 0 ) , 300 mM NaCl , 0 . 1% sodium deoxycholate , 100 µg/mL of lysozyme , and 0 . 1 µL/mL of DNAse . Following three freeze/thaw cycles oscillating between −80°C and 37°C , the lysed cells were centrifuged at 13 , 800× g and 4°C for 30 min . The supernatant was collected , syringe filtered to remove any remaining cellular debris , and supplemented with 10 µL/mL of His-PIC ( Sigma-Aldrich ) . For every 5 mL of crude lysate , 1 mL Talon resin beads were prepared as described by the manufacturer ( Clontech , Mountain View , CA ) . After equilibration of the resin with extraction buffer , lysate was added to the resin and agitated at room temperature for 45 min . Unbound proteins in the supernatant were removed after a 5 min room temperature spin at 700× g , and the pelleted resin was resuspended in wash buffer containing 20 mM Tris-HCl ( pH 7 . 0 ) , 600 mM NaCl , and 0 . 1% Triton X-100 . Following 15 min of agitation at room temperature , the resin was centrifuged at 700× g and room temperature for 5 min . The supernatant was removed , and the resin was washed two more times with wash buffer . The washed resin was then resuspended in 20 mM Tris-HCl ( pH 7 . 0 ) with 600 mM NaCl , transferred to a gravity flow column , and allowed to settle . After washing the resin bed with 5 mL 20 mM Tris-HCl ( pH 7 . 0 ) containing 600 mM NaCl , His6-PDI was eluted from the column using 2 mL quantities of increasing imidazole concentrations ( 10 , 15 , 20 , 25 , 35 , 40 , and 50 mM ) . Collected fractions of 0 . 5 mL volume were stored at −20°C until needed . Eluted fractions as well as samples from each step of the purification process were visualized by SDS-PAGE and Coomassie staining to verify the purity of His6-PDI . Samples of the 13C-labeled protein were dialyzed in sodium borate buffer with decreasing concentrations of NaCl , lyophilized , and stored at −80°C before reconstitution in a D20-based sodium borate buffer for use in FTIR spectroscopy . As previously described in [7] , FTIR spectra for PDI , ERp57 , or ERp72 were collected using a Jasco 4200 FTIR spectrometer at 0 . 964 cm−1 spectral resolution and a set resolution of 1 cm−1 . Samples were prepared in a D2O-based 10 mM sodium borate pH* 6 . 6 buffer that corresponds to pH 7 . 0 . Where indicated , a D2O-based 10 mM sodium borate pH* 6 . 1 buffer that corresponds to pH 6 . 5 was used for measurements at acidic pH . The buffers also contained 100 mM NaCl and , unless otherwise noted , 1 mM GSH . The 70 µL samples contained either unlabeled oxidoreductase ( 40 µg ) or a 1∶1 molar ratio of unlabeled oxidoreductase ( 90 µg ) and uniformly 13C-labeled CTA1 . Studies involving PDI and the CT holotoxin used 20 µg of 13C-labeled His6-PDI and 25 µg of CT; readings for this experiment were taken from 1–25 minutes after mixing the two proteins . Absorbance spectra were determined using the matched buffer as a reference and were corrected by subtraction of water vapor contribution , smoothing , and baseline correction in the amide I region . Wavenumbers were assigned to specific protein secondary structures for unlabeled PDI as detailed in [54]: 1655±5 cm−1 , α-helix; 1630±10 cm−1 , β-sheet; 1644±4 cm−1 , irregular . For 13C-labeled PDI , wavenumber assignments were shifted 40–50 cm−1 . Deconvolution of protein secondary structures was performed as previously described [7] . The percentages of oxidoreductase secondary structure calculated by FTIR spectroscopy were consistent with the secondary structure content predicted from crystal structures for PDI ( PDB 2B5E and PDB 4EKZ ) , ERp57 ( PDB 3F8U ) , and ERp72 ( reconstructed from PDBs 2DJ3 , 2DJ2 , 2DJ1 , and 3EC3 ) . Experiments were conducted with 15 µM CTA1 and equimolar PDI in 10 mM borate buffer ( pH 7 . 0 ) containing 100 mM NaCl and 1 mM GSH . Measurements were recorded at 10°C using a 0 . 1 mm optical path-length quartz cuvette and a J-810 spectrofluoropolarimeter equipped with a PFD-425S Peltier temperature controller ( Jasco Corp . , Tokyo , Japan ) . Samples were equilibrated at 10°C for 4 min before measurement , and each spectra was averaged from 5 scans . To monitor disassembly of the CT holotoxin , a gold plated Reichert ( Depew , NY ) SPR sensor slide was coated with ganglioside GM1 and subsequently appended with CT as previously described [22] . PBST was perfused over the sensor for 10 min at 37°C with a 41 µL/min flow rate to generate a baseline RIU signal corresponding to the mass of the bound holotoxin . A PBST solution containing GSH ( 1 or 30 mM ) and PDI , ERp57 , or ERp72 ( all at 100 nM final concentration ) was then perfused over the sensor at 37°C and a flow rate of 41 µL/min . After removal of the oxidoreductase from the perfusion buffer , sequential additions of antibodies were perfused over the sensor at the following dilutions: anti-PDI antibody , 1∶10 , 000; anti-ERp57 antibody , 1∶500; anti-ERp72 antibody , 1∶500; anti-CTA1 monoclonal antibody , 1∶500; anti-CTB antibody , 1∶15 , 000 . To detect the physical association of CTA1 with ERp57 or ERp72 , each oxidoreductase was perfused at 10°C over a SPR sensor appended with CTA1-His6 as previously described for ARF6-CTA1 interactions [24] . All experiments were performed with a Reichert SR7000 SPR refractometer . S-nitrosylated PDI was generated by treatment with S-Nitrosoglutathione for 20 min at room temperature . S-Nitrosoglutathione was prepared by incubating 5 mM reduced GSH with 5 mM sodium nitrate for 1 hr at room temperature . Other in vitro drug treatments involved exposing PDI for 30 min at room temperature to 0 . 2 mM EDC , 50 µM ribostamycin , or 0 . 1 mM bacitracin . A PDI stock concentration of 1 . 6 mg/mL was used for all treatments . CHO cells grown to 80% confluency in a 24-well plate were incubated in serum-free medium with 1 , 10 , or 100 ng/mL of CT for 2 hr at 37°C . Toxin-challenged cells were either left untreated or were co-incubated with 50 µM ribostamycin . The cAMP content of intoxicated and unintoxicated control cells was quantified with a commercial kit ( Perkin-Elmer , Boston , MA ) following the manufacturer's instructions . Values obtained from unintoxicated cells were background subtracted from the results with intoxicated cells , and the experimental data were then expressed as percentages of the maximal cAMP response obtained from cells exposed to 100 ng/mL of CT in the absence of ribostamycin . Triplicate samples were used for each condition . CHO cells grown to 75% confluency in 6-well plates were exposed for 3 hr to a mixture of pcDNA3 . 1/ssCTA1 [55] and Lipofectamine according to manufacturer's instructions ( Invitrogen , Carlsbad , CA ) . At 4 hr post-transfection , cAMP levels in untreated and ribostamycin-treated cells were quantified with a commercial kit ( Perkin-Elmer ) . Resting levels of cAMP from cells mock transfected with the empty pcDNA3 . 1 vector were also recorded . CHO cells grown to 80% confluency in 6-well plates were pulse-labeled for 30 min at 4°C in serum-free medium containing 1 µg/mL of CT . Unbound toxin was removed by washing with PBS , after which the cells were returned to toxin- and serum-free medium for a 2 hr incubation at 37°C . Chase conditions included untreated cells , cells co-incubated with 50 µM ribostamycin , and cells co-incubated with 5 µg/mL of BfA . Membrane fractions from digitonin-permeabilized cells were collected at the end of the pulse and the end of the chase conditions . Samples were resolved by non-reducing SDS-PAGE with 15% polyacrylamide gels and probed by Western blot with an anti-CTA antibody as described in [23] . As described above for the CT transport assay , untreated and ribostamycin-treated CHO cells were pulsed-labeled with CT at 4°C and chased at 37°C for 2 hr . Cytosolic fractions from digitonin-permeabilized cells were then perfused over an SPR sensor coated with the anti-CTA1 35C2 monoclonal antibody [52] . A detailed protocol for this SPR-based translocation assay has been provided in [56] . CT , P01555 and P01556; PDI , P17967and P07237; ERp57 , P30101; ERp72 , P08003 .
|
Protein disulfide isomerase ( PDI ) is a luminal endoplasmic reticulum ( ER ) protein with related but independent oxidoreductase and chaperone activities . The molecular mechanism of PDI chaperone function remains unidentified . Here , we report that PDI unfolds upon contact with the catalytic A1 subunit of cholera toxin ( CT ) . This unfolding event dislodges CTA1 from the rest of the multimeric toxin , which is a prerequisite for the ER-to-cytosol export of CTA1 and toxin activity against the host cell . The substrate-induced unfolding of PDI is linked to its chaperone activity . Our work has established a new property of PDI that is required for CT disassembly and provides a possible structural basis for the broader role of PDI as a chaperone that prevents protein aggregation .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"protein",
"folding",
"host-pathogen",
"interaction",
"biology",
"microbiology",
"pathogenesis",
"biophysics"
] |
2014
|
Substrate-Induced Unfolding of Protein Disulfide Isomerase Displaces the Cholera Toxin A1 Subunit from Its Holotoxin
|
Recurrent and feedback networks are capable of holding dynamic memories . Nonetheless , training a network for that task is challenging . In order to do so , one should face non-linear propagation of errors in the system . Small deviations from the desired dynamics due to error or inherent noise might have a dramatic effect in the future . A method to cope with these difficulties is thus needed . In this work we focus on recurrent networks with linear activation functions and binary output unit . We characterize its ability to reproduce a temporal sequence of actions over its output unit . We suggest casting the temporal learning problem to a perceptron problem . In the discrete case a finite margin appears , providing the network , to some extent , robustness to noise , for which it performs perfectly ( i . e . producing a desired sequence for an arbitrary number of cycles flawlessly ) . In the continuous case the margin approaches zero when the output unit changes its state , hence the network is only able to reproduce the sequence with slight jitters . Numerical simulation suggest that in the discrete time case , the longest sequence that can be learned scales , at best , as square root of the network size . A dramatic effect occurs when learning several short sequences in parallel , that is , their total length substantially exceeds the length of the longest single sequence the network can learn . This model easily generalizes to an arbitrary number of output units , which boost its performance . This effect is demonstrated by considering two practical examples for sequence learning . This work suggests a way to overcome stability problems for training recurrent networks and further quantifies the performance of a network under the specific learning scheme .
There are many human behaviors which unfold over time . Our limb movement , speech and even our internal train of thought appear to involve sequences of events that follow one another in time . We are capable of performing an enormous number of sequences , and we can perform the same action in a variety of different contexts . Hence the concept of generating temporal patterns or sequences by neural networks draw a lot of attention over the years . Early work relied on cyclic inhibition [1–3] which formed the basis of networks that function as ring oscillators [4] . These models could only be applied to small number of neurons and are restricted in the complexity of the output they can generate . The complexity of a sequence is determined by the number of actions that must be remembered in order to know to correct successor . Later work [5 , 6] produced temporal sequences in an arbitrary large network , using associative neural network with Hebb learning rule [7] , encompassing the relation between output pattern and synaptic connections . The main idea in this model was to functionally separate the synaptic connection into two components , slow and fast , such that the slow component encoded transition between patterns and the fast component stabilized the current pattern . This model , in its basic form , only encodes transitions between neighboring states in a sequence . Hence it is also limited in the complexity of outputs it can produce . Specifically , in order to learn two partially overlapping sequences one should introduce another component in the synaptic connection , with time scale proportional to the amount of overlap between the sequences . Jordan first considered a clear distinction between the state of the network and the output [8] . Moreover , applying recurrent links within the network , provides it a dynamic memory by which “time” is implicitly encoded in the state of the network [9] . This kind of network architecture ( i . e . recurrent and feedback connections ) is common in cortical microcircuit [10 , 11] , hence various training schemes for such network architectures arose along the years . A generalization of this approach considered reading out target information from randomly connected network , was first suggested in [12] and later developed to the notion of echo state networks ( ESN ) [13] and liquid state machines ( LSM ) [14] . Typically these networks consists of non-linear activation function for units within the network “reservoir” which linearly combines the output signal . These models do not need an internal pacemaker for producing a temporal sequence , in addition , learning a complex sequence is deduced to effectively learning a simple sequence , as two highly overlapping sequences end up as distinct in the high dimensional phase space of the network . None the less , it has been found as a challenging task to establish a successful learning procedure for these networks , one in which the network is capable of reproducing a desired target sequence for an arbitrary number of cycles , yet exhibiting robustness to errors and noise which are assumed to be common in biological networks . The main difficulties in this context are: In order to achieve a stable solution one should use a long training period involving noise over the output unit . During training the network will sample various fluctuations which improves the final network stability [15] . The second difficulty is assigning credit to output errors , i . e . which neurons and synapses are most responsible for the output error . Previous work settled this issue by restricting modification to synapses which project directly to the output unit [16] . This assumption was supported by [17] , in which they showed that even in the case that all synapses were subject to modification during training , the synapses to the output tended to change the most . In our model we suggest a variation of the ESN , i . e . For a recurrent network with a feedback loop , we consider linear activation function for neurons within the network and a binary output unit . In such case , given a target sequence on the output unit , one may easily solve for the corresponding activity in the network . Following previous work [16] we restrict ourselves on modifying synapses which project directly to the output unit . Even though it causes the solution space to shrink , it makes the learning problem straight forward , as it can be reduced for solving a simple perceptron [18 , 19] problem . This approach settles the problem of feeding back erroneous output to the network . Robustness to errors and noise naturally emerges from the finite margin of the perceptron problem , thus reproduction of a target sequence for an arbitrary number of cycles is possible , even in the presence of noise . In addition , considering a binary output unit helps in better quantifying the network performance , hence providing a different view on the computational power of this class of networks . In our model , quantifying the memory capacity ( MC ) of the network is mathematically equivalent to calculating the capacity of a perceptron with correlated patterns presented to it . Where correlation induced by the network dynamic , as such , this is a challenging task analytically . Similar problems had been tackled in [20 , 21] for the simplified case in which each neuron maintained an activity trace consisting of a decaying sum over all previous inputs presented to it . In [22] they considered correlated input-output associations , where temporal correlations between binary input patterns were modeled as Markov chain . In this case analytic result could only be obtained for the case of no temporal correlation between input patterns . Both models form a simple feed-forward architecture , hence temporal correlation do not depend on the state of other neurons in the network . In our model temporal correlation are of higher complexity due to the recurrent connectivity , hence we use numeric simulations in order to quantify the memory capacity for both the discrete and continuous time cases . Specifically , we solved the soft margin perceptron problem ( Methods ) with matlab standard quadratic programming function . An analytic estimation is given to noise robustness of the system .
In the presence of noise the networks trajectory in phase space will have a probabilistic nature . Each point in phase space , obtained by Eq ( 3 ) will be smeared to a N-1 ball of possible states . Hence noise robustness in the system stems from the finite margin of the perceptron problem . The quantity D ( J ) = 1 | J | min n J · x ( n ) ( 6 ) Defined in [24] quantifies the difficulty level of the classification problem at hand . It is the worst projection from the set { x ( n ) } n = 1 T on the hyperplane perpendicular to J . The best solution , i . e . with largest margin is obtained by maximizing the value of D over all possible weight vectors J: D m a x = max J D ( J ) ( 7 ) This is the margin , κ , obtained by the learning algorithm we used . The robustness to noise will be the order of magnitude of noise ( σ n o i s e 2 ) we can apply on each neuron , such that a learned sequence is still stable i . e . the hyperplane J classifies correctly the the set { x n o i s e ( n ) , z ( n ) } n = 1 T , for many cycles before an error occurs . In order to quantify this we use Eq ( 1 ) to solve recursively for the activation pattern over the generator neurons , this time taking into account the noise term . x n o i s e ( n ) = W n x ( 0 ) + ∑ k = 0 n - 1 W k V z ( n - k ) + ∑ k = 0 n - 1 W k η ( n - k ) = x ( n ) + R ( n ) ( 8 ) where we denoted the noisy activity pattern by xnoise . As expected , Eq ( 8 ) implies that the effect of noise is to drive the original activity pattern by R ( n ) = ∑ k = 0 n - 1 W k η ( n - k ) , which represent the accumulation of noise at the nth time step . Thus in order for the output weights , J , to classify correctly the noisy dynamics we need to find σnoise for which ∀n , ‖x ( n ) − xnoise ( n ) ‖ < κ , i . e . by Eq ( 8 ) we need to satisfy: ∥ R ( n ) ∥ < κ ∀ n ( 9 ) Calculations show ( S1 Text ) that under the annealed approximation , the amount of noise that the network can tolerate is given by: σ n o i s e 2 < ( κ 2 N ) 1 - N σ W 2 1 - ( N σ W 2 ) n ( 10 ) Where σW denotes the variance of an element in the connectivity matrix W , ( W i j ∼ N ( 0 , σ W 2 ) ) . Note that in the large N limit σW is tightly related to |λ|–the maximal eigenvalue of W , through: σ W 2 = λ 2 N ( 11 ) Plugging this relation in Eq ( 10 ) yields: σ n o i s e 2 < κ 2 N ( 1 - λ 2 1 - λ 2 n ) ( 12 ) lim n → ∞ { κ 2 N [ 1 - λ 2 ] λ < 1 0 λ > 1 ( 13 ) This result is confirmed by numerical simulations in which we calculated ‖R ( n ) ‖ for σnoise that saturates the bound predicted by Eq ( 12 ) . The average value of ‖R ( n ) ‖ over the noise is compared to our prediction in Fig 5 . On average results coincide , supporting the prediction for the scale of noise the network can tolerate . From Eq ( 13 ) we note that for λ > 1 , the noise in the system grows exponentially resulting in an unstable system to noise perturbations . It further suggests that the robustness to noise depends both on the normalization of W and on κ . We can explicitly determine λ , but κ is given per certain realization of W , V and a specific target sequence {zt} . On the one hand , if we fix κ we get that increasing λ lowers the robustness to noise . But note that , as expected , simulations shows that κ is monotonically increasing function in λ ( Fig 3D ) . This indicates that there exist an optimal λ , such that the robustness to noise of the system is ideal . This value will represent the counter balance between the ability to forget the errors and memorizing the desired sequence . So far our analysis focused on random connectivity , as we avoided from constraining the connectivity . In this section we will mention other classes of connectivity suggested for short term memory [25] . In this section we consider two tasks that our model , in its generalized form , can easily accomplish , without the need for any parameters fine tuning .
We presented a simple solution to the stability problem in learning temporal sequences by recurrent networks . By considering a linear activation function for a recurrent neural network with feedback loop , we could cast the problem of learning a temporal sequence of actions over the output unit to a simple perceptron problem . Using our method we could get a perfect reproduction of a target sequence for many trials , even in the presence of noise . The robustness to noise was calculated in terms of the perceptron solution margin . Non-linear classification This work only considered linear classification for simplicity . Allowing for non-linear classification , e . g . by the kernel method [29] , one can potentially improve the performance of the network . Indeed , we managed to improve the memory capacity by roughly tenfold , using a radial basis kernel ( Methods ) . In order to best exploit this method one should systematically search for the best kernel and an optimal procedure to determine its parameters values , which was out of scope in this work . Parallel learning A dramatic effect has been observed when facing the task of learning several sequences in parallel . In our model it is possible to learn many sequences , such that the total number of actions the network can learn , substantially overcomes the length of the single sequence maximal length . While the core reason for this property remains mysterious for us , we would like to discuss a mathematical explanation and what it biologically infer . Mathematically the ability to learn sequence or sequences depends on the distribution in phase space of the training set . In our model , strong correlations are induced by the feedback loop , i . e . by the statistics of the target sequence . As mentioned in [20–22] the memory capacity , of the perceptron , monotonically increase with increased correlations in both input and output . Even though they considered different model , we believe this finding stands in one with the effect observed in parallel learning . In our model different sequences are correlated in some manner through the dynamics ( W , V ) and the similar statistics of the target sequence . In the biological aspect , our finding suggest that it is economical to use a neural microcircuit for learning several short sequences rather than a long single sequence , which is a desirable property from a memory circuit . Other weight matrices Our work focused on random connectivity , as we avoided from making assumptions on the internal structure . Nonetheless , major improvement in the network performance has been observed for other types of weight matrices . The special structure of these matrices better exploit the N degrees of freedom available to the network for memory embedding . This property also makes the memory capacity extensive . This finding should motivate future work that might consider learning procedures allowing for internal synapses modifications . Multiple output units In this work we considered two examples of generalizing our model for multiple output units ( two and three ) . Generally , the model will generalize to an arbitrary number of output units . But , since the output states available for the network are exponential in the number of output units , only small number of these are sufficient to produce a fairly rich output sequence . The performance for multiple output units haven’t been studied systematically in this work . Nonetheless , we note that considering multiple output units is beneficial for the network performance . e . g 21 neuron are sufficient to study a 48 step periodic sequence ( melody of the “House of the rising sun” ) , while with a single output unit , a network with 20 neuron could maximally learn a 30 step periodic sequence . Note that in the case of multiple units the network is driven , by its own feedback , in various directions . That compared to the case of a single output unit , which only feeds back on a single vector , V . Thus multiple output units encourage the dynamic of the network to span larger volume in phase space , making the perceptron problem easier . Continuous time case Extension of our model to a Continuous time representation is considered in S2 Text . Nonetheless , such an extension turned out yielding a major drawback . While in the discrete time case our method succeed in providing a robust solution , in the continuous time case it failed , as the margin approaches zero every time there is a jump in the target sequence . As a result , in the continuous case , the network was only able to reproduce the sequence with small jitters . It was numerically evident that the network is vulnerable to noise in the initial condition alone—hence it is vulnerable to noise in general . Changing the learning procedure , i . e . allowing modification in all synapses , internal and feedback connections , might help stabilizing the solution . Note that by taking this route the problem isn’t a simple perceptron problem any more , so a new learning rule should be obtained . One should note that modifying connections within the network , also changes the itinerary of the neural dynamic in phase space . This fact is what turns such an approach to a challenging one . Other works [15] used this approach , but as mentioned before , the network activity will eventually deviate from its target function outside the training window . Timing is fundamental component for many of our day to day tasks . Yet , the neural mechanism underlying temporal processing remain unknown and controversial . It is not clear whether timing is dedicated to certain brain areas , or it is a general property , emerging from the neural activity . In our approach we used the dynamics of a recurrent neural network to implicitly represent time . That is , we encoded the timing of actions in the dynamics of the network . From our results it is hard to be conclusive regarding this question . On the one hand , from the discrete time case it is evident , that indeed it is possible to encode time in a robust manner within the neural activity . On the other hand , in the continuous case we did face stability problem , which might only be a property of our mathematical solution . In our mind , if it possible to robustly encode time in the discrete case , it should also be possible in the continuous case . As a consequence we do believe that this work support the claim in which timing is a general property of the brain , emerging from the neural activity .
In simulations we solved both the primal and dual form of the soft margin perceptron problem , as defined in [29 , 30] . For learning multiple sequences in parallel we used the primal formulation , for the longest single sequence we used the dual formulation . The primal problem takes the following form m i n i m i z e : 1 2 J · J + λ ∑ i = 1 T ξ i s u b j e c t t o : ξ i ≥ 0 ∀ i ∈ 1 , . . , T z t i ( J · x i + b ) ≥ 1 - ξ i Where J is the separating hyperplane and ξi’s are slack variables , yielding ξi = 0 for patterns on the correct side of the margin . 0 < ξi < 1 for patterns in the margin and ξi > 1 for wrongly classified patterns . In this formulation choosing small values of λ will encourage a large margin , with possible not optimal performance on the training data , while large values of λ will encourage a solution that performs well on the training data . The advantage of solving the “soft” problem is that a solution that minimizes the objective function exists . We used λ = 1048 which effectively serves as λ → ∞ , to encourage correct classification over the training data . The dual problem takes the following form m i n i m i z e : 1 2 ∑ i , j = 1 T α i α j z t i z t j ( x i · x j ) - ∑ i = 1 T α i s u b j e c t t o : 0 ≤ α i ≤ λ ∀ i ∑ i α i z t i = 0 from the dual formulation the separating hypeplane is given by the support vectors J = ∑ i = 1 T α ^ i z t i x i ( 17 ) the bias is calculated as a weighted average of the α i ′s , to deal with roundoff errors b = ∑ i = 1 T α ^ i ( z t i - J · x i ) / ∑ i α ^ i ( 18 ) Numerically we used the the matlab function quadprog , to solve both types of optimization problems . We set it with interior-point-convex algorithm and maximum number of iterations of 9000 , to prevent it from terminating prematurely . Non-linear classification Solving the dual problem generalizes easily for solving a non-linear classification problem by choosing an appropriate kernel [29] , i . e substituting xi ⋅ xj with a general kernel K ( xi , xj ) . Individual simulations with radial basis kernel , K ( x i , x j ) = exp [ - | | x i − x j | | 2 2 σ 2 ] , and σ set to typical distance between vectors , could increase the memory capacity by an order of magnitude ( Not shown ) . This observation is based on single trials and not studied systematically . The connectivity matrix , W , was constructed such that its largest eigenvalue is of particular value λ . To do so we first draw a random matrix with elements W ˜ i j ∼ N ( 0 , λ 2 N ) , and applied normalization such that W = λ λ m a x W ˜ . Where λmax is the largest eigenvalue , in absolute value of W ˜ . Every element in the feedback weight vector , Vi , was drawn from a standard normal distribution and normalized such that ‖V‖ = 1 . In each trial of the simulation we where interested in learning a specific binary sequence {zt} of length T , such that , zt ( t ) = ±1 with probability 1 2 . For each setup of random connections W , V we let the network learn various random sequences {zt} of different length T . After learning the output weight vector J , we have simulated the network dynamic with Eqs ( 1 ) and ( 2 ) for 5 cycles ( e . g . for target sequence T , we have simulated the network dynamics for 5T time steps ) . Eventually we compared the simulated output versus the target sequence {zt} counting for erroneous actions . For each sequence length we averaged the error over 300 repetitions of different random sequences . In addition we have done so for a given network setup over different normalization of W , i . e . different values of λ , note that that we have trained each specific random pattern over all different normalization of W . Following this procedure we have constructed the memory curve for a given network of size N , see S1 Fig for example . From such figure we extracted the memory capacity ( MC ) for each normalization of W , we have done so by taking the point on which the derivative of the curve is largest . Doing so for different realization of network setups we have plotted the memory capacity normalized by the network size ( N ) for different normalization of W , as can be seen in Fig 3 . Several sequences in parallel In order to construct Fig 4 we used Simulations of equal length sub-sequences . Given a number ( denote by s ) of sequences we wish to learn in parallel . We look for the maximal length , T s m a x , for which we can learn this set of s sequences . Sequences are again binary with equal probability to be in each state . The memory capacity for a set of s sequences is just s T s m a x . Multiple output units A generalization of the model is to consider an arbitrary number , l , of output units , zi , i = 1 , 2 , … , l , generally satisfying l ≪ N . Each output unit has its own feedback loop Vi , keeping the total feedback O ( 1 ) , requires | | V i | | = ½ 1 l . In the learning phase the perceptron problem is solved for each output unit separately , i . e . finding the best hyperplane , Ji for each unit . Note that the margin in this case is defined by min κ i i , i . e . the minimal margin from all of the l perceptron problem solved .
|
The ability to learn and execute actions in fine temporal resolution is crucial , as many of our day to day actions require such temporal ordering ( e . g . limb movement and speech ) . Indeed , generating stable time-varying outputs , using neural networks has attracted a lot of attention over the last years . One of the core problems , when facing such a task , is the solution stability , hence it was only possible to produce the sequence for a limited number of cycles . Here we propose a robust approach for the task of learning time-varying sequences .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"learning",
"medicine",
"and",
"health",
"sciences",
"statistical",
"noise",
"neural",
"networks",
"nervous",
"system",
"social",
"sciences",
"electrophysiology",
"neuroscience",
"learning",
"and",
"memory",
"simulation",
"and",
"modeling",
"cognitive",
"psychology",
"mathematics",
"statistics",
"(mathematics)",
"cognition",
"algebra",
"memory",
"research",
"and",
"analysis",
"methods",
"gaussian",
"noise",
"computer",
"and",
"information",
"sciences",
"animal",
"cells",
"psychology",
"cellular",
"neuroscience",
"cell",
"biology",
"anatomy",
"synapses",
"linear",
"algebra",
"physiology",
"neurons",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"physical",
"sciences",
"cognitive",
"science",
"eigenvalues",
"neurophysiology"
] |
2017
|
Stabilizing patterns in time: Neural network approach
|
Two-component signal transduction systems ( TCS ) are used by bacteria to sense and respond to their environment . TCS are typically composed of a sensor histidine kinase ( HK ) and a response regulator ( RR ) . The Vibrio cholerae genome encodes 52 RR , but the role of these RRs in V . cholerae pathogenesis is largely unknown . To identify RRs that control V . cholerae colonization , in-frame deletions of each RR were generated and the resulting mutants analyzed using an infant mouse intestine colonization assay . We found that 12 of the 52 RR were involved in intestinal colonization . Mutants lacking one previously uncharacterized RR , VCA0566 ( renamed VxrB ) , displayed a significant colonization defect . Further experiments showed that VxrB phosphorylation state on the predicted conserved aspartate contributes to intestine colonization . The VxrB regulon was determined using whole genome expression analysis . It consists of several genes , including those genes that create the type VI secretion system ( T6SS ) . We determined that VxrB is required for T6SS expression using several in vitro assays and bacterial killing assays , and furthermore that the T6SS is required for intestinal colonization . vxrB is encoded in a four gene operon and the other vxr operon members also modulate intestinal colonization . Lastly , though ΔvxrB exhibited a defect in single-strain intestinal colonization , the ΔvxrB strain did not show any in vitro growth defect . Overall , our work revealed that a small set of RRs is required for intestinal colonization and one of these regulators , VxrB affects colonization at least in part through its regulation of T6SS genes .
Vibrio cholerae causes the diarrheal disease cholera that affects 3 to 5 million people worldwide every year , resulting in 100 , 000–120 , 000 deaths annually [1] . V . cholerae produces a number of virulence factors which facilitate colonization of the intestine and subsequent disease . Major virulence factors are cholera toxin ( CT ) , which is responsible for production of profuse watery diarrhea , and a type IV pilus called the toxin-coregulated pilus ( TCP ) , which is required for intestinal colonization [2] . V . cholerae virulence factors are well known to be under extensive transcriptional control . CT and TCP production are controlled by the transcriptional activator ToxT [3 , 4] . Expression of toxT , in turn , is controlled by a virulence regulatory cascade involving the membrane-bound transcriptional activators ToxRS and TcpPH . These two regulators activate toxT transcription directly [5–7] . TcpPH expression is activated by the transcriptional activators AphA and AphB [8 , 9] . The quorum sensing ( QS ) regulatory system is also linked to the virulence gene regulatory cascade through HapR , the master QS regulator , which represses aphA expression [10] . Recently , the type VI secretion system ( T6SS ) has been identified as a new virulence factor in V . cholerae [11 , 12] . T6SSs deliver effector proteins into both eukaryotic and bacterial cells in a contact-dependent manner [12 , 13] . V . cholerae has one T6SS system with multiple T6SS effectors: VrgG1 and VrgG3 ( valine-glycine repeat protein G ) , which have actin cross-linking activity and peptidoglycan-degrading activity , respectively [14–17]; TseL , which has lipase activity [15 , 18]; and VasX , which perturbs the cytoplasmic membrane of target cells [15 , 19] . Activity of these effectors is antagonized by corresponding immunity proteins: TsiV3 , TsiV1 , and TsiV2 , respectively , to prevent killing by strains bearing these proteins [15 , 16 , 20 , 21] . The T6SS can be divided into functional sections consisting of the core structural components , the T6SS effector and immunity proteins , and transcriptional regulators . The base of the T6SS apparatus spans the cell envelope , and is a tube within a tube . The inner tube is composed of polymers of the hemolysin coregulated protein ( Hcp ) . The outer tube , also called the contractile sheath , is formed by polymers of VipA and VipB [14 , 22] . The Hcp inner tube is capped with a spike complex of trimeric VgrG proteins . The effectors are delivered by contraction of the VipA/VipB sheath , which in turn results in ejection of the inner tube along with VgrG and the effectors towards the target cell [12] . The genes encoding the T6SS components are organized into one large cluster ( VCA0105-VCA0124 ) and two auxiliary clusters ( VCA0017-VCA0022 and VC1415-VC1421 ) [11 , 23] . A key positive transcriptional regulator of the V . cholerae T6SS is VasH ( VCA0117 ) , which is related to enhancer binding proteins that activate transcription in a σ54 ( RpoN ) dependent manner [24 , 25] . VasH acts on the T6SS auxiliary clusters and vgrG3 of the large cluster , but does not affect expression of the structural genes encoded in the large T6SS gene cluster [24 , 26] . Additionally , Hcp production is positively regulated by the master quorum sensing regulator HapR and the global regulator cyclic AMP ( cAMP ) receptor protein CRP , and negatively regulated by QS regulator LuxO and by global regulator TsrA , a protein homologous to heat-stable nucleoid-structuring ( H-NS ) [27 , 28] . These studies have thus shown that numerous global regulators control T6SS expression , as well as one specific regulator ( VasH ) . V . cholerae T6SS studies have mainly focused on the V . cholerae O37 serogroup V52 strain because it assembles a T6SS apparatus constitutively [11] . In this strain , the T6SS is required for cytotoxicity towards Dictyostelium discoideum and J774 macrophages , and induces inflammatory diarrhea in the mouse model [29] . In V . cholerae O1 strain C6706 , the T6SS is not constitutively produced and conditions that promote T6SS production are unknown . However , production of T6SS can be achieved in other O1 strains by inactivating mutations in genes encoding the LuxO and TsrA negative regulators . In O1 strains , the T6SS translocates T6SS effectors into macrophages , and increases fecal diarrhea and intestinal inflammation in infant rabbits [27] . It was also shown that the V . cholerae O1 C6706 strain T6SS mediates antagonistic interbacterial interactions during intestinal colonization . A strain unable to produce the TsiV3 immunity protein , which provides immunity against the effector VgrG3 , exhibited an intestinal colonization defect only when co-infected with strains harboring an intact T6SS locus and VrgG3 [30] . Although T6SS is regulated and expressed differently between V . cholerae strains , production of this system in multiple strains promotes virulence against both eukaryotic and bacterial cells , suggesting the function is largely conserved but the regulation varies . Pathogenic bacteria experience varying conditions during infection of human hosts and often use two-component signal transduction systems ( TCSs ) to monitor their environments [31] . TCSs play important roles in the regulation of virulence factors , metabolic adaptation to host environments , and response to numerous environmental stresses including pH , osmolarity , oxygen availability , bile salts , and antimicrobial peptides [32] . TCS rely on a phosphorelay-based signal transduction system . The prototypical TCS consists of a membrane-bound histidine kinase ( HK ) , which senses environmental signals , and a corresponding response regulator ( RR ) , which mediates a cellular response . Response regulators are typically multi-domain proteins harboring a conserved receiver domain ( REC ) and C-terminal output domain such as DNA-binding , diguanylate cyclase , or methyltransferase [33–35] . Upon environmental stimulation , the HK catalyzes an ATP-dependent autophosphorylation reaction on a conserved histidine residue . The phosphoryl group is transferred from the HK to a conserved aspartate residue on the RR , eliciting a conformation change and subsequent cellular response [32 , 34 , 35] . The V . cholerae genome reference genome of O1 EL Tor N16961 strain is predicted to encode 43 HK and 49 RR ( http://www . ncbi . nlm . nih . gov/Complete_Genomes/RRcensus . html and http://www . p2cs . org ) . We also included 3 additional RRs ( VpsT , VpsR , QstR ) which were not annotated in these databases . Thirteen of these 52 putative RRs have been previously characterized and eight have a role in virulence factor production and host colonization: VarA , LuxO , VieA , PhoB , ArcA , FlrC , CarR , and CheY-3 [36–44] . VarA and LuxO repress production of quorum sensing regulator HapR , which represses expression of aphA and , in turn , TCP and CT production [36 , 37] , VieA regulates ctxAB expression indirectly by affecting production of ToxT through cyclic diguanylate ( c-di-GMP ) signaling [38 , 39] . The RR for phosphate limitation , PhoB , directly controls expression of a key transcriptional regulator , TcpPH , which activates toxT transcription [40] . The RR ArcA controls adaptation to low oxygen environment of the intestine and positively controls the expression of toxT [41] . CarR regulates glycine and diglycine modification of lipid A , confers polymyxin B resistance , and is required for intestinal colonization , although this phenotype is strain dependent [42] . FlrC controls flagellar biosynthesis and CheY-3 is needed for control of chemotactic motility [43 , 44] . Both motility and chemotaxis are known colonization factors for V . cholerae [43] . Together , these results show that RRs shown play a role in intestinal colonization have three basic targets: known virulence regulators and concomitant CT and TCP production; lipid A modification enzymes; or motility and chemotaxis . 39/52 , however , were not yet analyzed at the time of this study . To systematically evaluate the role of V . cholerae TCSs in intestinal colonization , we generated in-frame deletion mutants of each RR gene and analyzed the in vivo colonization phenotypes of the resulting mutants . We found 12 RR were required for wild-type intestinal colonization . One RR in particular had a very strong defect , encoded by genomic locus VCA0566 . We determined that VCA0566 ( now termed Vibrio type six secretion regulator , vxrB ) controls expression of several genes including the T6SS genes . We used multiple methods to substantiate that VxrB is required for expression of the T6SS in vitro and in vivo . Lastly , we report that the T6SS contributes to colonization of the V . cholerae O1 strain used in this study .
We have a limited understanding of the V . cholerae TCSs and their role in colonization and adaptation to host environments . To evaluate the importance of the 52 TCS RRs in colonization , we generated in-frame deletion mutants of the 40 RRs . For this analysis , we excluded 12 RR that were either predicted to be involved in chemotaxis ( 11 CheY , CheV , and CheB proteins ) or that we were unable to mutate ( VC2368 , ArcA ) [43 , 45] . We then analyzed the ability of 40 RR deletion mutants to colonize the small intestine in an in vivo competition assay where in vivo fitness of a mutant strain is compared to that of wild type strain using the infant mouse infection model ( Fig 1A ) [46] . While the vast majority of mutants—28—were not different from wild type , we identified 12 RR mutants that had a statistically significant colonization difference as compared to wild type ( Fig 1A ) . We focused on 8 mutants with a statistically significant colonization difference and exhibited at least 1 . 2-fold difference in CI ( Fig 1B ) . Consistent with previous studies , we identified that ΔVC0719 ( phoB ) , ΔVC1021 ( luxO ) , ΔVC1213 ( varA ) , and ΔVC2135 ( flrC ) were defective in colonization [36 , 37 , 40 , 44] . The competitive indices ( CI ) for ΔphoB , ΔluxO , ΔvarA , and ΔflrC were 0 . 01 , 0 . 02 , 0 . 16 , and 0 . 43 , respectively ( Fig 1B ) . Additionally , we identified a set of genes whose absence slightly but statistically significantly enhanced colonization ( at least 1 . 2 fold higher CI ) , suggesting that inhibition of their expression and activity may be needed for wild-type colonization . These mutants were ΔVC1050 , ΔVC1086 , and ΔVC1087 , which exhibited subtle and enhanced colonization phenotypes with CIs of 1 . 43 , 1 . 24 , and 1 . 48 , respectively ( Fig 1B ) . VC1050 is classified as an Hnr-type RR , [47] but its function is yet to be determined . VC1086 and VC1087 are part of a predicted eight gene operon encompassing VC1080-VC1087 . Both VC1086 and VC1087 have domains that suggest they function in cyclic guanylate ( c-di-GMP ) regulation . Specifically , VC1086 contains an EAL domain with conserved residues required for enzymatic function , while VC1087 harbors an HD-GYP domain , but this domain lacks the conserved residues required for enzymatic activity . We also identified one RR that was defective for colonization that had not been previously characterized . This mutant , ΔVCA0566 , had a colonization defect with a CI of 0 . 14 ( Fig 1A and 1B ) . Because this uncharacterized RR was important for colonization , we focused the rest of our studies on this protein . VCA0566 is the second gene of a predicted five gene operon and had been previously annotated as a RR of the OmpR family . The encoded protein , which we named VxrB for reasons described below , is 245 amino acids in length with an N-terminal REC domain and a C-terminal winged helix-turn-helix DNA-binding domain ( Fig 2B ) . Previously characterized members of the OmpR family in V . cholerae include PhoB , CarR , and ArcA [40–42] . Amino acid sequence alignment of the V . cholerae RRs in the OmpR family and the previously characterized E . coli OmpR [48] was used to identify the aspartate residue that is predicted to be phosphorylated in the REC domain ( Fig 2A ) . Since the phosphorylation state of a RR is likely to determine its activity , we mutated the aspartate residue in the REC domain of VxrB to mimic constitutively active ( D78E ) and inactive ( D78A ) versions , as used in other work [48] , and replaced the wild-type gene in the chromosome with these altered genes . These mutants were competed against wild type in the infant mouse colonization assay to determine if the phosphorylation state of VxrB is important for colonization . In accordance with our initial colonization screen , ΔvxrB displayed a CI of 0 . 15 ( Fig 2B ) . Somewhat surprisingly , the CI for vxrB::D78A ( inactive form ) was 0 . 53 , indicating a modest defect in colonization . This result indicates that the “inactive” form of VxrB does not phenocopy the ΔvxrB mutant , suggesting that VxrB harboring D78A substitution is not fully inactive . The CI for vxrB::D78E ( active form ) is 1 . 07 , suggesting that constitutive activation of VxrB does not significantly impair V . cholerae ( Fig 2B ) . Collectively , these findings suggest that in vivo phosphorylation of VxrB at D78 is likely to be important for its colonization function , but apparently not absolutely required . It is also likely that VxrB may not function by conventional phosphorylation-dependent signal transduction [49] . The first gene of the vxr loci , VCA0565 , is annotated as an HK . The other three genes ( VCA0567-69 ) are predicted to encode proteins of unknown function ( Fig 3A ) . We now termed these genes Vibrio type six secretion regulator ( vxr ) ABCDE and determined that these genes are co-transcribed using RT-PCR and RNAseq analysis ( Fig 3A and S1 Fig ) . Both the genomic context and organization is conserved in the Vibrio species ( S2–S4 Figs ) and vxr gene products do not share significant sequence similarity with previously characterized proteins . To gain a better understanding of the role of the vxr operon in colonization , we investigated whether the cognate HK and other genes in the vxr operon also contributed to mouse colonization . In-frame unmarked deletion mutants of vxrA , vxrB , vxrC , vxrD , and vxrE ( Fig 3B ) were generated and analyzed in an in vivo competition assay . Each mutant was outcompeted , with CIs of 0 . 35 , 0 . 16 , 0 . 44 , 0 . 66 , and 0 . 70 , respectively ( Fig 3B ) . These findings suggest that while vxrA and vxrB genes are critical for colonization in the infant mouse model , contribution of vxrCDE genes appears to be minor . To further confirm the phenotype of ΔvxrB colonization defect , a wild type copy of vxrB whose expression was driven from its native promoter was inserted into the Tn7 site ( located between VC0487 and VC0488 ) on the chromosome of ΔvxrB . In vivo competition assay of ΔvxrB-Tn7vxrB had a CI of 0 . 93 , similar to wild type levels , where ΔvxrB had a CI of 0 . 16 ( Fig 3B ) . Thus , the ΔvxrB colonization defect is restored to wild-type levels by introduction of the wild-type copy of vxrB . To gain a better understanding of the contribution of VxrB to V . cholerae pathogenesis , we performed high throughput transcriptome sequencing ( RNA-seq ) analysis to identify the V . cholerae genes controlled by VxrB . We used cells grown under virulence inducing AKI conditions , to mimic the intestinal conditions encountered when we know VxrB is important . 149 genes showed statistically significant differences in gene expression between the wild type and mutant ( S2 and S3 Tables ) . Of these , 80 genes were expressed to greater levels in the ΔvxrB mutant relative to the wild type ( S2 Table ) , while 69 were expressed to lower levels in the ΔvxrB mutant relative to wild type ( S3 Table ) . Of particular interest was the observation that message abundance of most of the T6SS genes in both the large cluster ( VCA0105-VCA0124 ) and the two auxiliary clusters ( VCA0017-VCA0022 and VC1415-VC1421 ) were less in the VxrB mutant relative to wild type ( Table 1 ) ( S5 Fig ) . This finding suggests that VxrB activates expression of the T6SS genes . To further analyze the role of vxrB in T6SS expression and function , we compared the levels of the major T6SS structural component , Hcp , between wild type and ΔvxrB mutant V . cholerae . Quantitative real-time PCR analysis of hcp revealed that the transcript abundance of hcp was decreased by 3 . 7-fold under AKI conditions and 4 . 1- fold under LB conditions in the ΔvxrB mutant relative to wild type ( Fig 4A ) . This finding supports that VxrB regulates expression of hcp and is consistent with the RNA-seq analysis . Additionally the levels of the Hcp protein in ΔvxrB were lower than wild type , in both whole cell samples and culture supernatants ( Fig 4B ) . We also determined that complementation of the vxrB mutation ( ΔvxrB-Tn7vxrB ) restored Hcp to wild-type levels . Because we found lower amounts of Hcp in the supernatant as well as in whole cells , this finding suggests that VxrB is needed to express and secrete Hcp . As negative controls , we included a Δhcp1Δhcp2 mutant because it is unable to produce the Hcp proteins [11 , 50] . As expected , no Hcp production was observed in this mutant . Furthermore , complementation of hcp1 in the Δhcp1Δhcp2 mutant partially restored Hcp levels ( Fig 4B ) . Overall these findings suggest that Hcp production is decreased in ΔvxrB mutant . Next we analyzed whether VxrB was needed for T6SS function , by examining T6SS-mediated interbacterial killing . Killing assays between the V . cholerae and the target E . coli K-12 strain MC4100 showed that wild-type V . cholerae decreased the numbers of E . coli compared to control experiments . This killing was dependent on the T6SS , as shown by greater numbers of E . coli obtained when incubated with V . cholerae Δhcp1Δhcp2 mutant and ΔvasH mutants , consistent with the findings reported by Ishikawa et al . ( Fig 4C ) [50] . This phenotype was complemented by introduction of either hcp1 or hcp2 into the Tn7 site on the chromosome . Consistent with our transcriptional and protein analysis presented above , we found that ΔvxrB mutants mediated less E . coli killing . These findings suggest that T6SS regulation by VxrB contributes to interbacterial killing . Since VxrB regulates T6SS expression and is required to for intestinal colonization , we next asked whether the T6SS itself is required for intestinal colonization . We performed in vivo competition assays of a T6SS null mutant ( Δhcp1Δhcp2 ) against wild type in the infant mouse model . We found that the in-vivo CI for Δhcp1Δhcp2 was 0 . 17 ( Fig 5A ) . In addition , ΔvgrG3 also had an in-vivo CI of 0 . 26 suggesting that T6SS components are important for intestinal colonization ( Fig 5A ) . This suggests that structural components of the T6SS are needed to colonize the intestine . Furthermore , this finding also suggests that the colonization defect associated with the ΔvxrB mutant could be caused by diminished T6SS production . To evaluate this possibility , we tested the in vivo competition of Δhcp1Δhcp2 against ΔvxrB and found that these strains competed nearly equally with each other ( Fig 5C ) . Furthermore , in-vivo CI of ΔvxrBΔhcp1Δhcp2 triple mutant against ΔvxrB was 0 . 07 and ΔvxrBΔhcp1Δhcp2 against wt was 0 . 10 . This finding suggests that the colonization defect by ΔvxrB was not solely due to altered expression of T6SS genes and other factors regulated by VxrB also contribute to colonization . It is also likely that T6SS expression is not completely abolished by the vxrB mutation . Indeed , western analysis ( Fig 4B ) shows that in vxrB mutant Hcp production is reduced but not completely eliminated . Similarly in vitro killing assay shows that vxrB mutant’s interbacterial killing ability is not identical to that of the strain lacking T6SS . We next asked whether VxrB plays a role in growth in vitro , by performing an in vitro competition assay . ΔvxrB mutants grew equally well as wild type , suggesting that neither had a competitive advantage over the other in vitro ( Fig 5B ) . This outcome suggests that there may be an in vivo signal produced in the infant mouse that triggers T6SS activity and colonization . We also performed single-strain colonization assays in the infant mouse model with ΔvxrB . There was a 12 . 7-fold decrease in colonization for ΔvxrB compared to wild type ( Fig 5D ) . This finding suggests that the colonization defect by ΔvxrB was not solely dependent on wild type , and possibly could be caused by competition with the normal flora or ability of the mutant to adapt to the infection microenvironment .
Systematic mutational phenotypic characterization of TCSs has been performed in only a few bacteria , including Vibrio fischeri , E . coli , Bacillus subtilis , Streptococcus pneumoniae , and Enterococcus faecalis [51–55] . In this study , we systematically analyzed the role of all V . cholerae TCS in colonization of the infant mouse small intestine and identified the RRs that play roles in mouse intestinal colonization . Specifically , ΔVC0719 ( phoB ) , ΔVC1021 ( luxO ) , ΔVC1213 ( varA ) , and ΔVC2135 ( flrC ) , and ΔVCA0566 ( vxrB ) exhibited intestinal colonization defects while ΔVC1050 , ΔVC1086 , and ΔVC1087 showed enhanced colonization . Many of the RRs had either no statistically significant defect or minor defects in the infant mouse colonization assay . It remains possible , however , that these RRs have a role in colonization in other infection models . In vivo transcriptome analysis has been performed on different strains of V . cholerae in the infant mouse and rabbit ileal loop infection models . The analysis of the whole genome expression of V . cholerae O1 El Tor C6706 cells accumulating in the ceca of orally infected infant rabbits and the intestines of orally infected infant mice revealed that expression of the genes encoding RRs is altered during in vivo growth conditions as compared to in vitro growth in nutrient broth and that in vivo expression of TCS also differed between the model systems [56] . In the infant rabbit infection model , expression of seven RR ( VC1081 , VC1082 , VC1155 , vieA , VC2702 ( cbrR ) , VCA0210 , and VCA1105 ) was increased and 1 RR ( carR ) was decreased by more than 2-fold significantly in comparison to V . cholerae cells grown in vitro in nutrient broth . In the infant mouse infection model , expression of 17 RR ( vpsR , VC1050 , VC1081 , VC1082 , VC1086 , VC1087 , VC1155 , VC1522 , flrC , cbrR , ompR , dct-D2 , vxrB , uhpA , vpsT , VCA1086 , and VCA1105 ) and 9 RR ( qstR , phoB , VC1348 , VC1638 , vieB , cpxR , ntrC , VCA0532 , pgtA ) were either decreased and increased significantly by more than 2-fold , respectively , in comparison to V . cholerae cells grown in vitro in nutrient broth [56] . vxrA and vxrB transcript levels were decreased 2 and 3-fold , respectively , in the experiments reported by Mandlik and colleagues , but did not reach statistical significance [56] . This work all used the V . cholerae O1 El Tor strain C6706 , and so it is yet unknown whether vxrB expression is similarly regulated in the O1 El Tor A1552 strain used here . There have been two other studies that analyzed V . cholerae infection phenotypes on a global scale , although they did not specifically target RR . Together , these studies and ours suggests there is a set of genes required for intestinal colonization across multiple models . Fu et al . used random transposon mutants coupled with insertion site sequencing ( Tn-seq ) in a rabbit model [30] . They identified insertions in two genes—VC1021 ( luxO ) and VC1155—that showed 8-15-fold reduction in colonization , while strains harboring insertions into RRs VC1348 , vieA , vieB , arcA , VCA0256 , uhpA , and pgtA had less than a 5-fold reduction in colonization ( p<0 . 001 ) . Another Tn-seq study using the infant rabbit model identified defects associated with luxO and arcA as above , and additionally phoB and varA [57] . Combining the results of these studies with ours identifies luxO , phoB , and varA , as required for in vivo fitness , and others that are variably identified . Because the Tn-seq work used transposon libraries , it is not known whether all RR were eliminated , so it is possible that their studies missed some critical RR . There are hints in their data , however , that the vxr locus is necessary in these other models as well . While Fu et al . did not identify vxrA or vxrB mutants , they did determine that a strain with an insertion into VCA0567 ( vxrC ) exhibited a 9-fold reduction in colonization ( p<0 . 0001 ) [30] . Additionally , Kamp et al . found that a strain with a transposon insertion in VCA0565 ( vxrA ) had a disadvantage in fitness ( mean fitness value of 0 . 6 ) when the bacteria from rabbit cecum fluid was placed into pond water for 48 hours at 30°C [57] . Collectively , these studies suggest that the Vxr genes play important roles in V . cholerae colonization and environmental dissemination . Our study revealed that the RR VxrB plays a significant role in colonization and in vitro inter-bacterial competition through its ability to regulate expression of T6SS genes . Neither vxrB nor any of the vxrABCDE operon members show similarity to previously characterized proteins . The vxr loci are conserved among the Vibrio species Vibrio parahaemolyticus , Vibrio vulnificus , Vibrio harveyi , and V . fischeri . BLAST analysis revealed that VxrA protein exhibits 67–80% , VxrB 79–84% , VxrC 56–68% , VxrD 58–74% , and VxrE 68–81% sequence similarity to the same proteins in other Vibrio species ( S2–S4 Figs ) . We also analyzed the predicted structure and function of the VxrCDE proteins using the protein homology/analogy recognition engine ( Phyre ) [58] . While VxrC and VxrE could not be modeled with high confidence and sufficient coverage , VxrD exhibited structural similarity to outer membrane protein transport proteins ( 100% confidence , 90% coverage ) . These analyses suggest that vxr genomic loci are a part of the ancestral Vibrio genome , and therefore likely have an evolutionarily conserved role in Vibrio biology . Expression and production of T6SS are tightly regulated at the transcriptional and posttranscriptional levels in a variety of bacterial systems [12 , 13 , 59] . Environmental signals such as iron limitation , thermoregulation , salinity , envelope stress , indole , and growth on surfaces regulate T6SS expression [59] . In V . cholerae A1552 , the strain used here , T6SS genes are expressed when cell are grown in high-osmolarity and low temperature conditions [50] . A recent study revealed that the V . cholerae A1552 T6SS genes are part of the competence regulon and their expression is induced when the bacterium grows on chitinous surfaces in a TfoX- , HapR- , and QstR-dependent manner [60] . Our work presented here identified VxrB as a regulator of the T6SS large gene cluster and the two auxiliary clusters . The predicted cognate HK of VxrB , VxrA , does not exhibit similarity to previously characterized sensory domains . The signals that govern expression and activity of VxrAB and how the VxrAB TCS is integrated into the T6SS regulatory network of V . cholerae are yet to be determined . We determined that while the wild-type strain has a competitive advantage in vivo over ΔvxrB , neither strain had a competitive advantage over the other in vitro . Furthermore , single infection studies showed that ΔvxrB had a significant colonization defect compared to wild type , suggesting that VxrB could be involved in competition with normal flora and that ΔvxrB could have a reduced fitness in infection environment . These observations also suggest that there may be an in vivo signal produced in the infant mouse that triggers T6SS activity and colonization . Our studies thus provided significant new insights into the regulation of T6SS in V . cholerae and provided further support that the T6SS is critical for V . cholerae virulence .
All animal procedures used were in strict accordance with the NIH Guide for the Care and Use of Laboratory Animals [61] and were approved by the UC Santa Cruz Institutional Animal Care and Use Committee ( Yildf1206 ) . The bacterial strains and plasmids used in this study are listed in S1 Table . Escherichia coli CC118λpir strains were used for DNA manipulation , and E . coli S17-1λpir strains were used for conjugation with V . cholerae . In-frame deletion mutants of V . cholerae were generated as described earlier [62] . All V . cholerae and E . coli strains were grown aerobically , at 30°C and 37°C , respectively , unless otherwise noted . All cultures were grown in Luria-Bertani ( LB ) broth ( 1% Tryptone , 0 . 5% Yeast Extract , 1% NaCl ) , pH 7 . 5 , unless otherwise stated . LB agar medium contains 1 . 5% ( wt/vol ) granulated agar ( BD Difco , Franklin Lakes , NJ ) . AKI medium contains 0 . 5% NaCl , 0 . 3% NaHCO3 , 0 . 4% Yeast Extract , and 1 . 5% Peptone , as previously described [63] . Antibiotics were used at the following concentrations: ampicillin 100 μg/ml; rifampicin 100 μg/ml; gentamicin 50 μg/ml; streptomycin 50 μg/ml . An overlapping PCR method was used to generate in-frame deletion constructs of each RR genes using previously published methods [62] . Briefly , a 500–600 bp 5’ flanking sequence of the gene , including several nucleotides of the coding region , was PCR amplified using del-A and del-B primers . del-C and del-D primers were used to amplify the 3’ region of the gene including 500–600 bp of the downstream flanking sequence . The two PCR products were joined using the splicing overlap extension technique [64 , 65] and the resulting PCR product , which lacks 80% of amino acids , was digested with two restriction enzymes and ligated to similarly-digested pGP704sacB28 suicide plasmid . Construction of vxrB plasmid harboring point mutations were performed using a similar technique [66] with the following alterations: primers containing the new sequence harboring the point mutations were used in place of the del-B and del-C primers . The deletion constructs were sequenced ( UC Berkeley DNA Sequencing Facility , Berkeley , CA ) and the clones without any undesired mutations were used . The deletion constructs are listed in S1 Table . The deletion plasmids were maintained in E . coli CC118λpir . Biparental matings were carried out with the wild type V . cholerae and an E . coli S17λpir strain harboring the deletion plasmid . Selection of deletion mutants were done as described [64] and were verified by PCR . The Tn7 complementation V . cholerae strains were generated by triparental matings with donor E . coli S17λpir carrying pGP704-Tn7 with gene of interest , helper E . coli S17λpir harboring pUX-BF13 , and V . cholerae strains . Transconjugants were selected on thiosulfate-citrate-bile salts-sucrose ( TCBS ) ( BD Difco , Franklin Lakes , NJ ) agar medium containing gentamicin at 30°C . The Tn7 complementation V . cholerae strains were verified by PCR . An in vivo competition assay for intestinal colonization was performed as described previously [46] . Briefly , each of the V . cholerae mutant strains ( lacZ+ ) and the fully virulent reference strain ( lacZ-otherwise wild-type ) ) were grown to stationary phase at 30°C with aeration in LB broth . Mutant strains and wild-type were mixed at 1:1 ratios in 1x Phosphate Buffered Saline ( PBS ) . The inoculum was plated on LB agar plates containing 5-bromo-4-chloro-3-indoyl-β-D-galactopyranoside ( X-gal ) to differentiate wild-type and mutant colonies and to determine the input ratios . Approximately , 106–107 cfu were intragastrically administered to groups of 5–7 anesthetized 5-day old CD-1 mice ( Charles River Laboratories , Hollister , CA ) . After 20 hours of incubation , the small intestine was removed , weighed , homogenized , and plated on appropriate selective and differential media to enumerate mutant and wild-type cells recovered and to obtain the output ratios . In vivo competitive indices were calculated by dividing the small intestine output ratio by the inoculum input ratio of mutant to wild-type strains . For single strain infections , 107 cfu of each strain , including otherwise wild type ( lacZ- ) strain , were intragastrically administrated to 5-day old CD-1 mice . After 20 hours of incubation , the small intestine was harvested and plated on selective media as previously described above . Statistical analyses for competition infections were performed using Wilcoxon Signed Rank Test . Statistical analyses were performed using Prism 5 software ( GraphPad Software , Inc . , San Diego , CA ) using Wilcoxon Signed Rank Test . P values of <0 . 05 were determined to be statistically significant . RNA was isolated as described below . The reverse transcription reaction to generate cDNA was carried out using the SuperScript III Reverse Transcriptase ( Invitrogen , Carlsbad , CA ) according to the manufacturer’s instructions at 25°C for 5 min , 50°C for 1 h , and 70°C for 15 min using 1 μg of RNA in a 20 μl final volume . The product was used in a PCR using suitable primers , and RNA without RT treatment was used as a negative control . For qRT-PCR expression analysis , RNA was isolated as described below . cDNA was synthesized using iScript cDNG Synthesis Kit ( Bio-Rad , Hercules , CA ) from 1 μg of total RNA . Real-time PCR was performed using a Bio-Rad CFX1000 thermal cycler and Bio-RAD CFX96 real-time imager with specific primer pairs ( designed within the coding region of the target gene ) and SsoAdvanced SYBR green supermix ( Bio-Rad , Hercules , CA ) . Results are from two independent experiments performed in quadruplicate . All samples were normalized to the expression of the housekeeping gene 16S using the Pfaffl method [67] . Relative expression was calculated by normalizing expression at ΔvxrB by that of wt . Statistical analysis was performed using two-tailed student’s t test . V . cholerae cells were grown aerobically overnight in LB at 37°C , then diluted 1:100 in fresh 10 ml AKI media in borosilicate glass test tubes ( diameter , 15mm; height , 150 mm ) and incubated at 37°C without shaking for 4 hours . After 4 hours , 10 ml cultures were transferred to 125 ml flasks ( for maximal aerated growth on an orbital shaker ( 250 rpm ) for 2 hours . Aliquots ( 2 ml ) of the cultures were collected and centrifuged for 2 min at room temperature . The cell pellets were immediately resuspended in 1 ml of TRIzol ( Invitrogen , Carlsbad , CA ) and stored at -80°C . Total RNA was isolated according to the manufacturer’s instructions . To remove contaminating DNA , total RNA was incubated with RNase-free DNase I ( Ambion , Grand Island , NY ) , and an RNeasy mini kit ( Qiagen , Valencia , CA ) was used to clean up RNA after DNase digestion . Five micrograms of total RNA was treated with a MICROBExpress Kit ( Ambion , Grand Island , NY ) to remove ribosomal RNA , and the efficiency was confirmed by Bioanalyzer analysis ( Agilent Technologies , Santa Clara , CA ) . Three biological replicates were generated for each condition . Libraries for RNA-seq were prepared using NEBNext Ultra Directional RNA Library Prep Kit for Illumina ( New England Biolabs , Ipswich , MA ) . Twelve indexed samples were sequenced per single lane using the HiSeq2500 Illumina sequencing platform for 50 bp single reads ( UC Davis Genome Center , UC Davis , CA ) and subsequently analyzed and visualized via the CLC Genomics Workbench version 7 . 5 ( Qiagen , Valencia , CA ) . Samples were mapped to the V . cholerae genome N16961 . Differentially regulated genes were identified as those displaying a fold change with an absolute value of 1 . 5-fold or greater . Statistical significance was determined by Empirical analysis of Digital Gene Expression ( edgeR ) test where p<0 . 05 was deemed significant [68] . V . cholerae strains were grown to an OD600 of 2 . 0 , and the culture ( 25 ml ) was centrifuged at 20 , 000 g for 10 min to obtain whole cell pellets . The culture supernatant containing secreted proteins were filtered through 0 . 22 μ membranes ( Millipore , Billerica , Massachusetts ) and secreted proteins in the culture supernatant were precipitated with 13% trichloroacetic acid ( TCA ) overnight at 4°C , pelleted by centrifugation at 47 , 000 g for 30 min at 4°C , wash with ice cold acetone and resuspended in 1x PBS containing Complete protease inhibitor ( Roche , Basel , Switzerland ) . Bovine serum albumin ( BSA , 100 μg/ml ) was added to the culture supernatant prior to TCA precipitation as a control . Protein pellets from whole cell were suspended in 2% sodium dodecyl sulfate ( SDS ) and protein concentrations were estimated using a Pierce BCA protein assay kit ( Thermo Scientific , Rockford , IL ) . Equal amounts of total protein ( 20 μg ) were loaded onto a SDS 13% polyacrylamide gel electrophoresis ( SDS-PAGE ) . Western blot analyses were performed as described [69] using anti-Hcp polyclonal antiserum provided by the Sun Wai [28] , anti-CRP ( Neoclone Inc . , Madison , WI ) , and anti-BSA ( Santa Cruz Biotech , Santa Cruz , CA ) . OneMinute Western Blot Stripping Buffer ( GM Biosciences , Frederick , MD ) was used to remove the Hcp antibodies and the same blot was used again to probe for CRP or BSA . These experiments were conducted with at least three biological replicates . Killing assays were performed as described previously [20] . Briefly , bacterial strains were grown overnight on LB plates and resuspended in LB broth containing 340 mM NaCl , as V . cholerae strain A1552 displayed enhanced interbacterial virulence towards E . coli under high osmolarity [50] . V . cholerae and E . coli MC4100 were mixed at a 10:1 ratio and 25 μl was spotted onto LB agar plates containing 340 mM NaCl and incubated at 37°C for 4 hours . Spots were harvested , serially diluted , and plated onto LB plates containing 50 μg/ml of streptomycin to enumerate surviving E . coli prey cells . The following assay was performed similarly as the intestinal colonization assay except no animal models were used . The V . cholerae mutant strains with wild-type lacZ allele ( lacZ+ ) and reference strain ( lacZ- ) were grown to stationary phase at 30°C with aeration in LB broth . Mutant strains and wild-type were mixed at 1:1 ratios in 1x PBS . The inoculum was plated on LB agar plates containing X-gal to differentiate colonies formed by the wild-type and mutant strains and to determine the input ratios . The inoculum ( 50 μl ) was spotted on to a LB agar plate and incubated at 37°C . After 20–24 hours of incubation , the 50 μl spots were scraped off the agar plate and resuspended in 1x PBS . The resuspension was serially diluted and plated on appropriate selective and differential media to enumerate mutant and wild type cells recovered and to obtain the output ratios . In vitro competitive indices were calculated by dividing the output ratio by the inoculum input ratio of mutant to wild type strains . Statistical analyses were performed using Wilcoxon Signed Rank Test .
|
Pathogenic bacteria experience varying conditions during infection of human hosts and often use two-component signal transduction systems ( TCSs ) to monitor their environment . TCS consists of a histidine kinase ( HK ) , which senses environmental signals , and a corresponding response regulator ( RR ) , which mediates a cellular response . The genome of the human pathogen V . cholerae contains a multitude of genes encoding HKs and RRs proteins . In the present study , we systematically analyzed the role of each V . cholerae RR for its role in pathogenesis . We identified a previously uncharacterized RR , VxrB , as a new virulence factor . We demonstrated that VxrB controls expression of the type VI secretion system ( T6SS ) , a virulence nanomachine that directly translocates effectors into bacterial or host cells , thereby facilitating colonization by competing with sister cells and intestinal microbiota . This study represents the first systematic analysis of the role of all RRs in V . cholerae pathogenesis and provides a foundation for understanding the signal transduction pathways controlling V . cholerae pathogenesis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[] |
2015
|
Vibrio cholerae Response Regulator VxrB Controls Colonization and Regulates the Type VI Secretion System
|
During herpes simplex virus 1 ( HSV1 ) egress in neurons , viral particles travel from the neuronal cell body along the axon towards the synapse . Whether HSV1 particles are transported as enveloped virions as proposed by the ‘married’ model or as non-enveloped capsids suggested by the ‘separate’ model is controversial . Specific viral proteins may form a recruitment platform for microtubule motors that catalyze such transport . However , their subviral location has remained elusive . Here we established a system to analyze herpesvirus egress by cryo electron tomography . At 16 h post infection , we observed intra-axonal transport of progeny HSV1 viral particles in dissociated hippocampal neurons by live-cell fluorescence microscopy . Cryo electron tomography of frozen-hydrated neurons revealed that most egressing capsids were transported independently of the viral envelope . Unexpectedly , we found not only DNA-containing capsids ( cytosolic C-capsids ) , but also capsids lacking DNA ( cytosolic A-/B-capsids ) in mid-axon regions . Subvolume averaging revealed lower amounts of tegument on cytosolic A-/B-capsids than on C-capsids . Nevertheless , all capsid types underwent active axonal transport . Therefore , even few tegument proteins on the capsid vertices seemed to suffice for transport . Secondary envelopment of capsids was observed at axon terminals . On their luminal face , the enveloping vesicles were studded with typical glycoprotein-like spikes . Furthermore , we noted an accretion of tegument density at the concave cytosolic face of the vesicle membrane in close proximity to the capsids . Three-dimensional analysis revealed that these assembly sites lacked cytoskeletal elements , but that filamentous actin surrounded them and formed an assembly compartment . Our data support the ‘separate model’ for HSV1 egress , i . e . progeny herpes viruses being transported along axons as subassemblies and not as complete virions within transport vesicles .
Herpes simplex virus type 1 ( HSV1 ) is the prototype of the Alphaherpesvirinae , a subfamily of the Herpesviridae . Viruses of this subfamily establish lifelong latent infections in the nervous system of the host organism . About 80 percent of the human population is infected with HSV1 . The infection is typically manifested by cold sores near the oral cavity but can also provoke ocular lesions and in rare cases encephalitis . The pleomorphic HSV1 virion has a complex structure [1] , [2]: the viral capsid encloses the double-stranded DNA genome and is surrounded by an amorphous layer of more than 20 different tegument proteins . A membrane envelope with inserted glycoproteins forms the outer boundary of the viral particle . The capsid is an icosahedrally symmetric protein shell with a diameter of 125 nm and composed of 162 capsomers [3]–[7] . Of these capsomers , 150 are hexons , 11 are pentons and one is the portal [8]–[10] , responsible for DNA packaging into the capsid after procapsid assembly . The pentons and the portal are located at the 12 vertices . The structures connecting adjacent capsomers are termed triplexes . HSV1 capsids assemble in the nucleus with the help of a scaffold protein [11] . The nuclear capsids have been classified into four types: round procapsids and angular A- , B- and C-capsids . Both A- and B-capsids are devoid of DNA , but B-capsids retain the scaffolding protein [12] . C-capsids are mature capsids containing the viral genome [11] . C-capsids have been reported to contain larger amounts of the minor capsid proteins pUL17 and pUL25 than A- and B-capsids [13]–[16] . When compared to virions , nuclear capsids are virtually devoid of tegument [16]–[20] , although it has been suggested that association of the tegument proteins pUL36 and pUL37 might occur already in the nucleus [21] , [22] . Primary envelopment of the newly assembled capsids takes place at the inner nuclear membrane ( reviewed in [19] ) . It is followed by fusion of the primary envelope with the outer nuclear membrane leading to de-envelopment of capsids . Once released into the cytosol , progeny capsids are transported towards the site of secondary envelopment [19] , [23]–[25] . In neurons , egressing HSV1 particles are transported over long distances . For alphaherpesviruses the nature of the particles undergoing axonal transport from the nucleus to the cell periphery ( termed anterograde transport ) is debated , in particular for HSV1 and pseudorabies virus ( PrV ) . Two models of viral assembly have been suggested . According to the ‘separate model’ ( synonym: subassembly model ) [24] , [26] , [27] , cytosolic capsids lacking an envelope are transported along microtubules while the viral tegument proteins and glycoproteins travel separately or in association with transport vesicles . In this model , assembly of mature virions occurs at axonal varicosities or at axon terminals [17] , [26]–[32] . Conversely , according to the ‘married model’ , viral particles are transported from the cell body towards the axon terminal already fully assembled and inside transport vesicles [33]–[37] . In both models , the HSV1 particles are transported in the cytoplasm along microtubules by cellular microtubule motors ( reviewed in [24] , [38]–[40] ) . One particular feature of all cytoskeletal motors is that they move actively only in one direction: either to the plus-ends of microtubules close to the plasma membrane , or towards the minus-ends that are clustered at the microtubule-organizing center situated in close proximity to the cells nucleus . Recent biochemical data have provided evidence that tegumented capsids can recruit several copies of microtubule motors of opposing polarity simultaneously [16] . Furthermore , the intracellular movement of individual particles occurs in both directions with occasional changes in direction; nevertheless , there is an overall preferred transport direction for all viral and host cargos [41] , [42] . The microtubule motor dynein is responsible for transport of capsids from the cell periphery towards the nucleus ( termed retrograde transport ) [43] , [44] . For this function , dynein requires the interaction with its cofactor dynactin [45]–[47] . Anterograde transport , i . e . transport in the opposite direction , is mediated by plus-end-directed microtubule motors , such as kinesin-1 or kinesin-2 [16] , [39] , [43] , [44] . Several tegument proteins are essential for intracellular transport of capsids and may contribute to forming viral motor binding sites . In particular , it has been shown that the tegument proteins pUL36 and pUL37 are essential for capsid transport during entry and egress [48]–[50] . Moreover , HSV1-GFPVP26 capsids lacking most of the outer tegument proteins , but still containing inner tegument proteins such as pUL36 and pUL37 , are transported along microtubules in vitro in the presence of cytosol [18] . Furthermore , Radtke et al . [16] have shown that pUL36 and pUL37 are accessible on the surface of capsids that recruit motors in vitro . It has also been suggested that the capsid protein VP26 is involved in retrograde transport of capsids [51] , although other studies have demonstrated that this protein is not essential for dynein-mediated transport [52]–[54] . Furthermore , VP26 is also not required for recruiting dynein and kinesin onto isolated capsids in vitro [16] , [18] . The tegument protein pUS11 was shown to bind to kinesin-1 [55] , although it does not appear necessary to recruit kinesin-1 to capsids in vitro [16] . Little is known about the location and identity of the tegument proteins bound to capsids in situ during transport . A previous study of HSV1 virions using cryo electron microscopy and single particle icosahedral reconstruction has revealed only a small ordered density of tegument located at the vertices of the capsid [56] . This density has been suggested to be formed by the inner tegument protein pUL36 . Furthermore , earlier conventional electron microscopy studies have shown capsids inside cells with substantial densities bound at the vertices [45] . Together , these results have suggested that the molecular motor complexes might attach to the vertices of the capsid but this binding platform has remained uncharacterized . Here , we applied cryo electron tomography ( cryoET , [57] ) to analyze the three-dimensional structure of HSV1 particles during anterograde axonal transport . By virtue of this technique , the rapidly frozen specimen is kept vitrified in near-native conditions [58] , [59] and does not suffer from structural re-arrangements caused by chemical fixation , dehydration or heavy metal staining . The vast majority of progeny capsids found in axons were non-enveloped; hence , our data support the separate model of anterograde axonal transport . Surprisingly , not only cytosolic capsids containing DNA but also capsids devoid of DNA had been transported along the axons despite significant differences in the amount of tegument associated with them . Further , three-dimensional cryoET snapshots of capsid assembly by secondary envelopment in axon terminals suggest that secondary envelopment might involve vesicle fusion to form a sufficiently large enveloping compartment .
Intracellular transport of alphaherpesviruses is often studied in neurons of dissected nervous ganglia explants , e . g . in sympathetic neurons of rat superior cervical ganglia ( SCG ) . Unfortunately , this system is not accessible to analysis by cryoET for two reasons . First , the size of the explants is typically prohibitive as , upon freezing , it causes an ice thickness exceeding the penetration limit of electrons ( ∼1 µm ) [57] . Placing the EM grids further away from the explant to analyze flat regions of the outgrown neurons is likewise not applicable because the grids then needed to be removed prior to freezing . In this case , the neurons would be damaged thus preventing a native in situ analysis . Therefore , we chose to use primary neurons cultured after dissociation . In early experiments , we analyzed primary neurons from dissociated rat dorsal root ganglia ( DRG ) by cryoET . These peripheral sensory neurons present a near-native model for studying HSV1 transport . Unfortunately , also in this system , the prominent thickness of the cell bodies resulted in a specimen thickness that impeded cryoET . The hippocampus is a brain region that is typically infected during herpes simplex encephalitis in humans [60] . Therefore , we next analyzed hippocampal neurons , which are also a close-to native system for the study of HSV1 infection and , in addition , provided extended areas suitable for cryoET of axons . Primary hippocampal neurons cultured on electron microscopy grids ( Figure 1A ) were infected with HSV1 . Infection was followed by live cell imaging using HSV1 ( KOS ) -GFPVP26 [52] . VP26 is a small capsid protein located on top of the hexons [61] , [62] . At 2 h post infection , the axons contained only occasionally fluorescent HSV1 particles ( not shown ) . Around 16 hours post infection ( p . i . ) , after synthesis of progeny viruses , massive egress of fluorescent viral particles occurred ( Figure 1B ) . Around 60% of the fluorescently labeled particles were motile . The average velocity of the viral particles during anterograde transport was 2 . 4 µm/s ( Figure 2 , Table S1 ) , consistent with earlier reports [28] , [42] , [48] , [49] . Despite frequent changes in direction , capsid transport along axons had a preferred orientation towards the cell periphery and anterograde transport on average was faster than retrograde transport ( Figure 2 , Table S1 ) . Samples vitrified at 16 h p . i . were analyzed by cryo electron microscopy . Often , at least two capsids per field of view ( 1 . 65×1 . 65 µm ) were recognized by cryo electron microscopy ( cryoEM ) 2D projection images ( Figure 1C ) . Using cryoET , we identified the transported particles as non-enveloped cytosolic capsids ( Figure 1D–F ) . In the axon , they were consistently found in close proximity to microtubules ( Figure 1D–F; arrows ) . Besides cytosolic DNA-containing C-capsids , unexpectedly , there were also cytosolic A- and B-capsids present in the axons , although in lower numbers than C-capsids . In 2D projections of thicker cellular specimens , structural features were superimposed upon one another , and thus difficult to interpret . In contrast , in three-dimensional tomographic reconstructions , the capsids were clearly discernible from cytoplasmic vesicles and finer features such as individual capsomers as well as the structural elements in the cytoplasm became recognizable ( Figure 1D-F ) . We could clearly identify cytosolic A-capsids as well as cytosolic B-capsids . While the former were angular and empty , the latter contained densities of the scaffold protein in the capsid lumen . Their morphology was clearly different from that of fully DNA-packaged C-capsids ( Figs . 1D–F , S1 ) . We noted that cytosolic A- and B-capsids were not observed after infection with HSV1 ( KOS ) or the HSV1 ( KOS ) -GFPVP26 variant , whereas they comprise the majority in HSV1 ( F ) ( Table 1 ) . The microtubules had luminal densities consistent with earlier observations [63] ( Figure 1D–F ) . While dense material was associated occasionally with cytosolic capsids ( data not shown ) , we could not assign unequivocally such densities to cellular microtubule motors . Overall , the capsids seem to contain very little tegument , and the capsomers were not obscured by tegument or cellular protein complexes but were clearly recognizable . Earlier studies have shown that some tegument proteins are essential during intracellular capsid transport [48]–[50] , [64] , [65] . To analyze the interactions between capsid and tegument during transport , we averaged the densities of cytosolic capsids that were computationally extracted from tomograms of infected neurons . The 14 tomograms acquired of neurons infected either with wild type strains HSV1 ( F ) or HSV1 ( KOS ) , or with HSV1 ( KOS ) -GFPVP26 contained a total of 67 cytosolic capsids ( Table 1 ) . We calculated separate capsid averages for 41 cytosolic DNA-containing C-capsids ( 24 HSV1 ( KOS ) -GFPVP26 , 4 HSV1 ( KOS ) and 13 HSV1 ( F ) ; Table 1 ) ( Figure 3Ai-iv ) and for 26 cytosolic DNA-lacking A- or B-capsids ( 6 A-capsids and 20 B-capsids , all HSV1 ( F ) wild-type , Table 1 ) ( Figure 3Bi-iv ) . The resolution of the averages was 6 . 9 nm and 9 . 7 nm , respectively . Cytosolic C-capsids were compared to the average of 143 nuclear C-capsids , i . e . C-capsids biochemically purified from the nuclei of infected cells ( Figure 3Ci-iv; Figure 4A , C ) . Cytosolic A-/B-capsids were compared to the average of 158 nuclear A-capsids , likewise biochemically purified from nuclei of infected cells ( Figure 3Di-iv; Figure 4B , D ) . The resolution for both groups of nuclear capsid averages was 5 . 6 nm . Nuclear capsids are known to be virtually devoid of tegument proteins [16]–[19] . Therefore , a comparison of native cytosolic capsids to biochemically purified nuclear capsids could reveal features on cytosolic capsids that correspond to tegument proteins acquired shortly before nuclear egress or in the cytosol . Indeed , this comparison revealed a prominent extra density , located exclusively at the C-capsid vertices ( Figure 4A , C; blue ) . It was present on top of the pentons and connected further to the positions of the two adjacent triplexes and to one side of the neighboring hexons . Thus , these extra densities were only positioned on hexon-penton interfaces but not on hexon-hexon interfaces . The comparison of cytosolic A-/B-capsids to the nuclear A-capsids showed that cytosolic A-/B-capsids comprised only a small amount of extra density ( Figure 4B , D; green ) . Nevertheless , this extra density was also located exclusively at the vertices , in particular towards one side of the peripentonal hexons . On cytosolic A-/B-capsids , no extra density was present on top of the pentons . Occasionally , there were enveloped virions in regions of the axons that were quite some distance away from both , the soma and the axon terminals ( Figure 5 ) , although at much lower frequency than non-enveloped capsids . Of the 73 capsids located in middle regions of axons at 16 h p . i . , less than 10% were enveloped while the others were non-enveloped ( ratio 6∶67 ) . Nevertheless , such enveloped virions were also located in close proximity to microtubules ( data not shown ) . Secondary envelopment sites were characterized by capsids being in close proximity to groups of vesicles ( Figure 6 ) . Three-dimensional analysis revealed that such assembly sites lacked any cytoskeletal elements , but that filamentous actin rather surrounded these assembly sites ( Figure 6C , D ) . In contrast , there were no microtubules in these areas . Notably , the vesicles in assembly sites had different sizes and were characterized by two different morphologies . Some of the vesicles were studded with spike-like densities , protruding from the membrane into the lumen of the vesicle ( Figure 6C , black arrowhead , Figure 6D , yellow densities ) . In contrast , other vesicles showed a smooth luminal surface . Typically , electron-dense material , presumably tegument , was accreted on the cytoplasmic face of vesicles with spike-like densities on their interior side ( Figure 6C , black arrows ) . When those vesicles had a concave cytoplasmic side it was typically facing towards a capsid . By virtue of our three-dimensional reconstructions , we revealed that at least in some cases the volume of individual spike-studded vesicles appeared not to suffice to fully enclose a capsid .
In this study , we used cryoET to visualize HSV1 capsid-tegument interactions in 3D during axonal transport in vitrified hippocampal axons . To this end , we first established a close-to-native experimental cell system enabling us to follow intra-axonal herpesvirus transport by both fluorescence microscopy and cryoET . By culturing dissociated hippocampal neurons directly on electron microscopy grids , we were able to circumvent the practical limitations that dissected nerve ganglia explant systems pose for cryoET . Furthermore , this type of primary neurons provides a relevant model for HSV1 since the hippocampus in the brain is infected during herpes simplex encephalitis in humans [60] . Culturing the hippocampal neurons on electron microscopy grids did not impair the course of HSV1-infection . Live-cell imaging using HSV1 ( KOS ) -GFPVP26 revealed a peak in anterogradely transported virus particles at 16h p . i . . This time point is consistent with earlier reports on egressing HSV1 in infected neurons [29] , [36] , [66] , [67] . The average anterograde speed of 2 . 4 µm/s of the HSV1 particles ( Figure 1C , Table S1 ) agrees well with earlier observations [41] . These transport rates suggest active transport by a kinesin , e . g . kinesin-1 or kinesin-2 which have been shown to bind to isolated , tegumented HSV1 capsids in vitro [16] . The characteristic pattern of net anterograde transport despite intermittent changes in directionality has likewise been reported before for herpesvirus egress [41] , [42] . It most likely reflects the engagement of motor complexes with opposite directionality on the same viral particle [16] , [68] . A direct correlation between the live-cell fluorescence imaging and cryoET at the level of individual viral particles was impossible since the handling steps between fluorescence imaging and the time point of vitrification took about 20 to 30 s . CryoET is not the adequate tool for a systematic , statistical analysis , since it is limited to axon areas thin enough to be penetrated by the electron beam . Nevertheless , on a population level , the number of egressing HSV1 particles was fully sufficient to characterize particles in transit at higher resolution by cryoET . The majority of the viral particles in mid-axon regions were non-enveloped cytosolic capsids . We next focused on the organization of tegument proteins on these cytosolic , axonal capsids to identify the interaction platform for the attachment of microtubule motors mediating intracellular transport . For HSV1 ( F ) , we detected all three intra-axonal capsid types – 52% contained DNA ( cytosolic C-capsids ) , while the remaining ones lacked DNA ( 15% cytosolic A- and 33% B-capsids , respectively ) . We also frequently observed A- and B-type capsids during and after secondary envelopment in axon terminals ( data not shown ) . In contrast to a prevailing hypothesis [13] , [69] , our results indicate that A-/B-capsids leave the nucleus and are actively transported to the cell periphery as it had been suggested previously [3] , [70] . Cytosolic B-capsids have also been reported for another herpesvirus , simian cytomegalovirus [71] . Whether the fact that the majority of intra-axonal capsids were cytosolic C-capsids reflects their higher efficiency in nuclear egress or just the ratio of nuclear C-capsids to A-/B-capsids remains to be determined . Formally , we cannot exclude that the packaged DNA genome had been lost at a later stage in the cytosol resulting in the appearance of cytosolic A-type capsids . However , it seems very unlikely that cytosolic C-capsids would give rise to cytosolic B-capsids . Interestingly , we observed cytosolic A-/B-capsids only when infecting with HSV1 ( F ) , while neurons infected with HSV1 ( KOS ) ( wild-type or -GFPVP26 ) lacked them ( Table 1 ) . Negatsch et al . reported recently for HSV1 ( KOS ) a lack of pUS9 expression [72] . The US9 region of our HSV1 ( KOS ) has the same mutations as reported by Negatsch et al . ( 2011 ) , while our HSV1 ( F ) lacks any mutations in the pUS9 region ( data not shown ) . pUS9 had earlier been reported to play a role in herpesvirus axonal transport [32] , [73] . Our results suggest that also nuclear egress of the HSV1 ( KOS ) capsids may be either highly specific for C-capsids or impaired for A-/B-capsids when compared to HSV1 ( F ) . Whether this difference due to the changes in the US9 gene remains to be established . Thus , in a situation where nuclear capsid egress is highly specific for C-capsids as observed here for HSV1 ( KOS ) , A- and B-capsids might be retarded in the nucleus . The late time point of vitrification at 16 h p . i . suggests that both cytosolic , axonal progeny C-capsids and A-/B-capsids presented an adequate tegument composition to be transported towards the cell periphery for assembly and exit . The abundance of potential parental , incoming capsids derived from newly produced viruses superinfecting these axons is very low , because there were virtually no capsids in cells that had been vitrified 5 to 20 minutes p . i . even when using an MOI of up to 200 ( data not shown ) . The combination of cellular cryoET with subvolume-averaging allowed visualizing the tegument density distribution on intra-axonal capsids in unprecedented detail . Subvolume-averaging for macromolecules inside cells has been barely performed so far [74] . The reasons for this are that it is difficult to identify macromolecular complexes within a cellular context , and that the number of complexes of interest within cells is low compared to in vitro particle preparations . Comprehensive knowledge on the protein composition of the intra-axonal cytosolic capsids is lacking since so far they could not be purified for quantitative mass spectrometry analysis . The existing information on these particles is based on data using fluorescently tagged proteins and immunolabelling experiments and therefore incomplete . Here , we obtained novel information by averaging subviral structures in their native surroundings . This will enable future studies on their interactions with other host factors , and will allow to correlate such data derived from in situ / in vivo experiments with the results of biochemical systems reconstituting key intermediate steps in vitro [75] . In our study , subvolume-averaging in situ including icosahedral symmetrisation showed that tegument proteins associated exclusively with the capsid vertices . In accordance with previous studies from isolated virions [61] , the capsid protein VP26 did not contact the extra density present on such cytosolic capsids . This supports the notion that VP26 , located on top of the capsid hexons [6] , but not the vertex pentons , is dispensable for capsid transport [51]-[54] . Our results furthermore agree with biochemical studies that VP26 is not required for recruiting dynein or kinesin-1 onto capsids [16] , [18] . In turn , the tegument material exclusively located around the capsid vertices supports the notion that the molecular motors mediating transport might bind to the vertex region as it has been suggested previously [45] . The striking differences in the tegument structure between cytosolic C-capsids and cytosolic A-/B-capsids ( Figs . 3 and 4 ) , most notably the presence of extra density on top of the pentons of the cytosolic C-capsids , provided valuable insights into the complex capsid–tegument interaction network . The minor capsid protein pUL25 forming a heterodimer with the protein pUL17 has been attributed to a density termed “elongated C-capsid specific component ( CCSC ) ” in cryoEM reconstructions of nuclear C-capsids [13] , [76] . The CCSC is barely visible on nuclear A- and B-capsids , consistent with a lower abundance of both proteins that has also been confirmed by proteomic data [13] , [77] . Both pUL25 and pUL17 have been localized on nuclear A- and B-capsids by cryoEM and TAP pulldown assays [76] , [78] . In these studies the protein complex has been termed “capsid-vertex specific component ( CVSC ) . Furthermore , pUL25 can interact with pUL36 , the largest herpesvirus tegument protein , in HSV1 [50] , and in PrV [22] that in turn interacts with the tegument proteins pUL37 and VP16 [65] , [80]–[82] . In accordance with these studies , our results show that cytosolic C-capsids were associated with a higher amount of tegument than cytosolic A-/B-capsids . This is consistent with the cytosolic C-capsids comprising higher amounts of pUL17/pUL25 , and therefore binding more tegument than cytosolic A-/B-capsids . Nevertheless , this low tegumentation on cytosolic A-/B-capsids appeared to be sufficient for at least some capsid transport from the neuronal soma into the axons . Thus , full coverage of all capsid vertices by tegument seems not to be required for microtubule transport , and tegument recruitment onto even one vertex might be sufficient albeit barely detectable in the icosahedral average reconstruction of cytosolic A-/B-capsids . Non-enveloped HSV1 cytosolic capsids detected inside hippocampal axons are in agreement with the ‘separate model’ of alphaherpesvirus axonal anterograde transport [17] , [24] , [26]-[31] . Further supporting this model , we identified sites of secondary envelopment at axon terminals ( Figure 6 ) . We also observed enveloped virions in mid-axon regions ( Figure 5 ) , but at a much lower rate than non-enveloped particles ( Table 1 ) . CryoET is not an adequate tool for a statistical analysis , but the ratios between enveloped and non-enveloped capsids nevertheless indicate a trend . Two different scenarios may explain this . First , it is possible that the virions in middle regions of axons underwent secondary envelopment in a varicosity , and that they would eventually exit the cell also from here as reported previously [29] . This would imply that even though enveloped viral particles were sporadically observed , they might not undergo long distance transport . Although axon terminals appeared as the main envelopment and exit sites for HSV1 , some enveloped particles may have been generated in the soma , and only afterwards entered the axons . A recent report comparing PRV and several strains of HSV1 reports that in explanted primary neurons from rat superior cervical ganglia , for HSV , about 75% of the viral particles in the axon and growth cone were enveloped and 25% non-enveloped [67] . Thus , the assembly pathway of HSV1 may be more complex than anticipated by the ‘married’ or ‘separate’ models for HSV1 axonal transport . Further studies comparing a wider range of neurons derived from different structures of the nervous system and other strains of alphaherpesviruses will ultimately reconcile these apparent discrepancies . Furthermore , combinations of different tags on VP26 with additional mutations in US9 or other herpesviral genes may result in complex phenotypes in axonal transport that may not be recognized or remain silent during infection of epithelial cells . CryoET is the method of choice for visualizing filamentous actin . Our native three-dimensional analysis of the axon terminals revealed that the secondary envelopment sites themselves were devoid of filamentous actin while the actin meshwork surrounding them seemed to form the boundary of an assembly compartment . Given that the dimension of these compartments was around ( 1 µm ) 3 , these surrounding actin filament structures remained nevertheless unnoticed by fluorescence microscopy so far . Future dedicated studies of such actin cages using correlative fluorescence and electron microscopy are needed to further characterize this feature of assembly sites . The assembly sites contained numerous vesicles studded on their luminal inside with glycoprotein-like densities , presumably having being transported to these sites independently of cytosolic capsids . Classical electron microscopy techniques cannot visualize these spikes as unequivocally as it has been achieved here . Tegument proteins accumulated on the cytosolic surface of these vesicles and might be the cause of a vesicle indentation to form a concave surface towards the capsids . Further , some of these vesicles did not appear to be large enough to fully envelope one capsid . The close proximity of several of these vesicles suggests that secondary envelopment might involve vesicle fusion to form a sufficiently large enveloping compartment . In summary , we have characterized a new neuronal infection model that enables investigating axonal transport , assembly and egress of HSV1 in 3D in a close-to-native state . CryoET revealed that the axonal viral particles were predominantly non-enveloped cytosolic capsids . We found that in addition to cytosolic C-capsids , unexpectedly cytosolic A-/B-capsids also underwent axonal transport . The prominent differences in tegumentation between these two capsid types suggest that efficient transport of capsids does not require large amounts of tegument , and occurs in the presence of different amounts of tegument . For both capsid types , the capsid-to-inner-tegument interactions were exclusively limited to the capsid vertices . These interactions are likely crucial for transport by forming a binding platform for microtubule motors . The higher abundance of non-enveloped over enveloped capsids in middle regions of axons , and the secondary envelopment sites at axon terminals favor the separate model for HSV egress for this combination of HSV1 strains and hippocampal neurons . The three-dimensional visualization of secondary envelopment sites revealed insights into a level of detail that allowed us to propose novel aspects of this process like formation of an actin bound compartment and a possible role for fusion of smaller vesicles during envelopment .
HSV1 ( F ) , HSV1 ( KOS ) and HSV1 ( KOS ) -GFPVP26 [52] virions were amplified in BHK-21 cells , and the viral titers were determined by plaque titration on Vero cells as described previously [1] , [54] . The virus stocks had a titer of 109 PFU /ml . Hippocampal neurons were isolated from 17 days old rat embryos ( provided by Boyan Garvalov , MPI Neurobiology , Germany ) . IBIDI slides ( µ-slide 8 well , Ibidi GmbH ) were coated with 1 mg/ml poly-L-lysine ( Sigma ) in borate buffer ( 1 . 24 g boric acid + 1 . 9 g borax in 400 ml distilled water , pH 8 . 5 ) overnight . They were then washed with distilled water three times before adding MEM horse serum medium ( Gibco ) , which was replaced next day by neurobasal medium ( Gibco ) supplemented with B27 ( Gibco ) and glutamine . Neurons were then seeded at a density of 4 , 500 cells per 1 cm2 . They were incubated for 7 days at 37°C , 5% CO2 and then infected with HSV1 ( KOS ) -GFPVP26 at an MOI of 50 PFU/cell . Infected neurons were imaged using a 63x oil objective on an Axiovert 200 M light microscope ( Zeiss ) equipped with an AxioCam HRm camera ( Zeiss ) and controlled by the Axiovision 4 . 1 software ( Zeiss ) . For the long time-lapse experiments , viral infection of neurons was monitored by wide field phase contrast and fluorescence imaging every hour over a period of 24 hours . For the short time-lapse experiments , fluorescent pictures were taken every 2 seconds at the same region for 10 min . Fluorescence was detected with a GFP blue band excitation/green band emission filter set ( HQ-EGFP; F41-017; AHF Analysentechnik AG ) . An incubation chamber around the microscope allowed time-lapse observations at 37°C , 5% CO2 and high humidity ( EMBL workshop; No . 530010; Cell Biology Trading ) . In the short time-lapse experiments , the speed and length of several continuous runs were measured in two different observations from two different neurons ( Table SI ) . Some of the particles came in or moved out of the field of view during the observation time . These were also taken in consideration in the measurements . Au grids of 200 mesh with holey carbon support films ( Quantifoil GmbH , Jena , Germany ) were sterilized on a Petri dish under UV light for 15 min and then coated as described above for the IBIDI slides . Dissociated rat hippocampal neurons were prepared as described [83] . Neurons were plated over the grids at a density of 100 , 000 cells in a 60 mm diameter Petri dish and incubated 7 days at 37°C and 5% CO2 to enable the growth of axons and dendrites . Neurons were then infected with HSV1 at a MOI of 50 . At 16 hours post infection ( p . i . ) , neurons were prepared for cryoET as described below . BHK-21 cells were infected with 0 . 01 to 0 . 02 PFU/cell for 2 to 3 days until detached and collected by sedimentation . They were then washed once in MNT buffer ( 30 mM MES , 20 mM Tris , pH 7 . 4 , 100 mM NaCl ) , snap-frozen and stored at -80°C . Nuclear capsids were purified as previously described [16] , [18] , [21] , [84] . Capsids were diluted in three volumes TNE ( 20 mM Tris , pH 7 . 5 , 500 mM NaCl , 1 mM EDTA ) with 10 mM DTT and protease inhibitors , and sedimented by centrifugation ( Beckman TLA100 . 2 rotor , 15 min , 50 krpm , 4°C ) . The pellets were resuspended in BRB80 buffer ( 80 mM PIPES , 1 mM EGTA , 2 mM MgCl2; pH 6 , 8 with KOH ) with 10 mM DTT , 1 mg/ml soybean trypsin inhibitor , protease inhibitors , 100 mg/ml RNase ( Roth , Germany ) and 0 . 1 U/ml DNase I ( M6101 , Promega , USA ) ) . Capsids were then incubated for 30 min at 37°C and overnight at 4°C , repelleted ( TLA100 . 2 rotor , 8 min , 50 krpm , 4°C ) and resuspended in BRB80 buffer by tip sonification on ice ( 3×10 seconds , 40 W ) . Hippocampal neurons growing adherently on the holey carbon support film on Au grids were prepared for cryoET as follows: 2 µl of colloidal gold suspension ( 10 nm diameter in HBSS buffer , coated with BSA ) was added on the EM grid . Excess of liquid was removed by blotting the grid with a filter paper . Specimens were vitrified by plunge-freezing into liquid ethane and transferred into liquid nitrogen for storage . In the case of isolated capsids ( salt-treated capsids and nuclear capsids ) , 5 µl droplets were added onto the holey carbon support film on Cu 200 mesh Quantifoil grids . To avoid formation of aggregates , salt-treated capsids were sonicated before addition on the grid for 3×10 seconds using a Sonopuls HD3200 sonicator with BB6 cup horn ( Bandelin , Berlin ) at 60% max output . Capsids were prepared for cryoET as described for neurons . Data was collected on a Tecnai Polara ( FEI , Eindhoven , The Netherlands ) transmission electron microscope equipped with a GIF 2002 post-column energy filter ( Gatan , Pleasanton , CA ) . Images were collected with a 2K×2K Multiscan CCD camera ( Gatan ) . The microscope was operated at 300 kV and the pixel size was 0 . 81 nm at the specimen level . Tilt series were collected from −60° to 60° , with an angular increment of 2° or 3° . Defocus was measured along the tilt axis after each tilt and automatically maintained at −8 µm for isolated capsids and at −12 µm for neurons to gain phase contrast and to distinguish structures more accurately inside the cell . The total electron dose received at the specimen level was kept between 60 and 90 electrons/A2 . The applied electron dose was kept proportional to 1/cosα of the tilt angle ( α ) . Tilted images were aligned using 10 nm gold beads as fiducial markers . Three-dimensional reconstructions were calculated with the software IMOD [85] . The volume of the reconstructions for visualization was typically 512×512×256 pixels , after the images obtained in the microscope ( 2048×2048 pixels ) were down-sampled by a factor of four ( IMOD ) . Subsequent processing steps were done using Bsoft [86] . Capsids were located in the original unbinned tomograms , and subvolumes with a size of 180×180×180 pixels were extracted . The orientation of all subvolumes was determined using a 22 Å resolution structure of the HSV1 capsid [87] as a template and the oriented subvolumes were averaged . Icosahedral symmetry was applied to the averages . Symmetrized averages were used as templates for the next iteration of orientation refinement . Three iterations were performed . The resolution of the averages was determined by Fourier shell correlation ( FSC ) using the 0 . 5 criterion , after splitting the data in two halves , calculating two separate averages and imposing icosahedral symmetry . To calculate the difference map of the two averages gray values were scaled to the same radial density maximum within the capsid and minimum just outside of the capsid . The difference of densities was then calculated by subtracting the capsids without tegument from the capsids with tegument . All capsid reconstructions were first scaled against the nuclear C-capsid reconstruction . Magnification differences up to 3 . 5% were detected and these were compensated for by creating up-scaled or down-scaled maps . The following HSV-1 capsid maps have been deposited in the Electron Microscopy Data Bank ( EMDB ) at PDBe ( http://www . ebi . ac . uk/pdbe/emdb/ ) : EMD_1956 , cytosolic C-capsids; EMD_1957 , cytosolic A-/B-capsids; EMD_1958 , nuclear A-capsids; EMD_1959 , nuclear C-capsids . The ID numbers for genes mentioned in the text ( source: ncbi . nlm . nih . gov/gene ) are: US9: Gene ID 2703452; UL17: Gene ID 2703388; UL25: Gene ID 2703377; UL35 ( VP26 ) : Gene ID 2703356; UL36: Gene ID 2703357; UL37: Gene ID 2703358 .
|
Herpes simplex virus 1 ( HSV1 ) establishes lifelong latent infections in the peripheral nervous system . After reactivation , progeny viral particles travel within sensory neurons towards sites of initial infection . There are conflicting reports what type of viral structures are transported: some studies observed non-enveloped capsids traveling while others reported transport of fully enveloped viruses within vesicles . Here , we used cryo electron tomography to analyze the three-dimensional architecture of HSV1 in axons of hippocampal neurons . In mid-axonal regions we found predominantly non-enveloped capsids . Interestingly , we observed both genome-containing and empty capsids that differed significantly in the amount of bound proteins . Viral protein recruitment thus varied between the different cytosolic capsid types , but effective transport occurred despite these differences . Furthermore , we observed three-dimensional snapshots of secondary capsid envelopment in axon terminals . Altogether , this study provides valuable structural detail on axonal HSV1 particles supporting the notion that viral subassemblies are conveyed along the axons to be assembled only after axonal transport .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"cell",
"motility",
"cellular",
"structures",
"microtubules",
"macromolecular",
"assemblies",
"microbiology",
"host-pathogen",
"interaction",
"viral",
"structure",
"neuroscience",
"motor",
"systems",
"protein",
"structure",
"cytoskeleton",
"proteins",
"biology",
"biophysics",
"tegument",
"proteins",
"biochemistry",
"cytoskeletal",
"proteins",
"cell",
"biology",
"virology",
"molecular",
"cell",
"biology",
"nucleocapsid"
] |
2011
|
Cryo Electron Tomography of Herpes Simplex Virus during Axonal Transport and Secondary Envelopment in Primary Neurons
|
Estimates of genetic diversity in helminth infections of humans often have to rely on genotyping ( immature ) parasite transmission stages instead of adult worms . Here we analyse the results of one such study investigating a single polymorphic locus ( a change at position 200 of the β-tubulin gene ) in microfilariae of the lymphatic filarial parasite Wuchereria bancrofti . The presence of this genetic change has been implicated in benzimidazole resistance in parasitic nematodes of farmed ruminants . Microfilariae were obtained from patients of three West African villages , two of which were sampled prior to the introduction of mass drug administration . An individual-based stochastic model was developed showing that a wide range of allele frequencies in the adult worm populations could have generated the observed microfilarial genetic diversity . This suggests that appropriate theoretical null models are required in order to interpret studies that genotype transmission stages . Wright's hierarchical F-statistic was used to investigate the population structure in W . bancrofti microfilariae and showed significant deficiency of heterozygotes compared to the Hardy-Weinberg equilibrium; this may be partially caused by a high degree of parasite genetic differentiation between hosts . Studies seeking to quantify accurately the genetic diversity of helminth populations by analysing transmission stages should increase their sample size to account for the variability in allele frequency between different parasite life-stages . Helminth genetic differentiation between hosts and non-random mating will also increase the number of hosts ( and the number of samples per host ) that need to be genotyped , and could enhance the rate of spread of anthelmintic resistance .
In recent years there has been a substantial increase in the use of mass drug administration ( MDA ) to reduce the morbidity associated with helminth infections of humans [1] , increasing the probability that anthelmintic resistance may become a public health concern in the future . One such annual MDA programme is the Global Programme to Eliminate Lymphatic Filariasis ( GPELF ) which , in 2005 , treated over 145 million people with albendazole ( a broad spectrum benzimidazole anthelmintic ) in combination with either ivermectin or diethylcarbamazine [2] . GPELF targets mainly Wuchereria bancrofti , the most widely distributed of the filarial parasites of humans . Sensitive molecular assays are required to detect the presence of anthelmintic resistance before widespread treatment failure is apparent , drug resistance becomes disseminated and disease control is jeopardised [3] . Surveys of helminth parasites of humans are being conducted to establish whether genetic changes at certain polymorphic loci ( associated with resistance to the same or related drugs used against veterinary helminths ) , are present in these populations and subject to detectable selection under chemotherapeutic pressure [4]–[13] . A phenylalanine to tyrosine substitution at position 200 on the β-tubulin isotype 1 molecule has been identified in a number of helminth parasites of farmed ruminants including Haemonchus contortus [14] , [15] , Cooperia oncophora [16] , and Teladorsagia circumcincta [17] and is associated with benzimidazole ( BZ ) resistance in these species . Worryingly , this genetic change has also been identified in W . bancrofti [13] , though the phenotypic studies relating the substitution to a decreased albendazole efficacy have not been undertaken in this species . To aid clarity the two alleles at position 200 on the β-tubulin isotype 1 molecule shall be referred to as allele F ( phenylalanine ) for susceptibility and allele Y ( tyrosine ) for putative resistance . Inbreeding , the mating of related individuals , influences parasite genotype distribution and can affect the selection of adaptive traits . Facets of a species' biology may cause parasite inbreeding , such as population structure or assortative mating ( when mate choice is determined by phenotype ) . Parasite allele frequency can differ between infrapopulations ( the populations of parasites within individual hosts ) due to the ecology of the infection or through the random nature of infection events ( all groups may have an equal probability of having a rare allele , but actual numbers may vary between groups by chance ) . Helminth parasites have a particularly subdivided population structure as adult worms are confined within their definitive host , and only able to mate with other worms that belong to the same infrapopulation . The population genetic structure of most helminth species remains unknown . The few studies that have been undertaken indicate that whilst some species appear to have no apparent genetic structure others exhibit a high degree of parasite genetic differentiation between hosts [18] . The degree of genetic differentiation in the parasite infrapopulation can shed insight into the microepidemiology of parasite transmission [19]–[23] . Infrapopulation genetic differentiation will also influence helminth population genetics as it causes a reduction in the frequency of heterozygote offspring , a phenomenon known as the Wahlund effect [24] . Studies investigating the inheritance of benzimidazole resistance are lacking , though evidence indicates that thiabendazole resistance in H . contortus may be a semi-dominant trait [25] . Other authors have postulated that alleles conferring anthelmintic resistance , including allele Y , are likely to be recessive [17] , [26] , which would make heterozygote worms susceptible to treatment . If an allele conferring drug resistance is recessive , excess parasite homozygosity will increase the probability that a resistance allele will survive treatment . This has been shown using genetic metapopulation models investigating nematodes of grazing animals; these models indicate that the spread of rare recessive genes is promoted by hosts accumulating multiple related infections simultaneously [27] , [28] . The degree of parasite genetic differentiation among hosts can be quantified using FST ( or related analogues; see [18] and references therein ) . The adult stages of the majority of parasitic helminths of humans cannot be obtained routinely for direct investigation , so genetic surveys ( including those investigating drug resistance ) resort to sampling transmission stages , i . e . those ( immature ) life-stages that gain access to the environment to be transmitted to and from hosts or through vectors [13] , [29]–[32] . However , the results of these surveys should be interpreted with caution , as the underlying allele frequency of the adult worm population may differ from the allele frequency of the sampled transmission stages . Variations in transmission stage allele frequency and genotype distribution could be generated randomly or be a product of the parasite's spatial structure and life-history traits . For example , population subdivision will cause random variation in adult worm allele frequencies between hosts at low parasite densities . Filarial parasites have separate sexes and are thought to be polygamous [33] , which may accentuate the variability in microfilarial allele frequency , e . g . a rare allele may be highly over-represented in the subsequent generation if , by chance , a male worm with this allele inhabits a host harbouring females but no other males . In addition , the inherent random sampling of gametes during sexual reproduction [34] , and the overdispersed distribution of parasite numbers among hosts [35] may cause the allele frequency and genotype distribution to vary by chance from generation to generation . This paper analyses population genetic data collected for a study by Schwab et al . [13] who identified the presence of the β-tubulin allele Y in populations of W . bancrofti . Firstly , the extent of parasite inbreeding is estimated from W . bancrofti microfilarial samples taken from patients in Burkina Faso , West Africa . Samples were obtained from different villages , some of which had received a single round of MDA with ivermectin and albendazole , under the auspices of the GPELF . Secondly , an individual-based stochastic model is presented which simulates microfilarial genetic diversity from adult worm allele frequencies . The model generates sample allele and genotype frequencies using the same number of hosts , and the same number of microfilariae per host as in Schwab et al . [13] . This model is then used to assess whether the observed level of parasite inbreeding is the result of a sampling artefact or a true biological phenomenon . Finally , the model is used to assess the likely range of adult worm allele frequencies which could have given rise to the observed microfilarial data , providing some insight into how genetic surveys which sample transmission stages should be interpreted . We discuss the implications of our results in terms of the development and detection of anthelmintic resistance .
Table 1 summarises the data collected for the study by Schwab et al . [13] and indicates the number of microfilariae and hosts sampled . The village of Gora was removed from the F-statistic analysis since only one host was sampled in this village . In some hosts it was possible to genotype only a few microfilariae , increasing the uncertainty associated with estimation of underlying infrapopulation allele frequencies in these hosts . Results are grouped according to parasite treatment history . The average frequencies of allele Y in microfilarial samples from untreated and treated hosts were 0 . 26 and 0 . 60 , respectively [13] . The degree of parasite heterozygosity ( the proportion of microfilariae with the heterozygote genotype ) is estimated for each village . The table also indicates the deviation of each population from the Hardy-Weinberg Equilibrium ( HWE ) , which gives the proportion of heterozygote microfilariae that would be expected in a randomly mating population . This reveals a strong deficit of heterozygotes in all three populations . In this paper , we refer to two different types of allele frequency: ( 1 ) the underlying frequency of the allele putatively associated with BZ resistance , with ql denoting the allele frequency of the entire parasite population of a given locality , and ( 2 ) the parasite allele frequency within the host population that is sampled , denoted by Hql . The superscript l denotes the parasite life-stage under investigation , be it microfilariae ( l = M ) or adult worms ( l = W ) , and H denotes definitive host . The allele frequency estimated from the sample , , may not correspond to the true underlying allele frequency , ql , either because the hosts sampled are not representative of the whole host population , or because the parasites genotyped do not represent adequately the allele frequency within the host . By genotyping transmission stages before they leave the definitive host prior to the introduction of mass chemotherapy , insight can be gained into the different causes of microfilarial excess homozygosity . If it is assumed that the number of microfilariae produced , their survival , and their probability of being sampled are independent of their genotype ( as we do in the null model ) , it can be assumed that deviation from the HWE may be the result of non-random mating . If the locus being investigated is not under selection , the excess microfilarial homozygosity will most likely be the result of either infrapopulation genetic differentiation or non-random parasite mating within hosts . Genotyping transmission stages would allow the relative contributions of each of these two sources of inbreeding to be estimated . The variation in the allele frequency between hosts will account for some of the excess homozygosity whilst deviation from the HWE in the microfilariae within an individual host will indicate possible non-random mating within the infrapopulation . The Wright's hierarchical F-statistic is used to investigate the correlation of parasite genes within and between human hosts [29]–[31] , [36] . It is assumed that the infrapopulation is the first hierarchical group in the parasite population , and FIS is defined as the correlation of genes between microfilariae within the infrapopulation; , as the correlation of microfilarial genes between different hosts living in the same village; , as the correlation of microfilarial genes between different villages within the overall microfilarial population; and FIT , as the correlation of genes between individual microfilariae relative to the overall microfilarial population of the region . The different inbreeding terms introduced are summarized in Table 2 . A value of FIS is significantly greater than zero points towards adult worm non-random mating , indicates variation in worm allele frequency between hosts , and suggests differences in the worm allele frequency between villages . The same statistical frameworks used to estimate Wright's F-statistic were employed here , taking into account variable sample sizes [34] . Estimates of the 95% confidence intervals for FIS , and FIT , were generated by bootstrapping simultaneously worms within each host and bootstrapping over hosts within each village [37] . F-statistics , and their associated uncertainty , were calculated for each village . A dioecious adult worm helminth population with a 1:1 male to female ratio was randomly generated for a given mean number of worms per host and degree of parasite overdispersion ( as determined by the k parameter of the negative binomial distribution , parameterized following [35] ) . Each adult worm infrapopulation was randomly allocated an allele frequency , as analysis of pre-treatment data did not detect any significant relationship between the host's frequency of allele Y and microfilarial burden . The adult worm allele frequency of each host was randomly selected according to the given underlying allele frequency , qW , and the degree of parasite genetic differentiation between hosts , . For a description of a method for generating the distribution of allele frequencies in a subdivided population using the beta distribution [38] , see Porter [39] . It is again assumed that microfilarial production and survival is independent of genotype , allowing a microfilarial population for each host i to be generated according to the size and allele frequency of the adult worm infrapopulation . Worms were assumed to be polygamous; implying that if only one male parasite were present within a host , all fertile females within that infrapopulation would be mated . The number of microfilariae produced by each parasite infrapopulation was assumed to be proportional to the number of fertilised females within that host . It was also assumed that gametes separate independently and re-assort according to the degree of non-random mating ( FIS ) . The probability with which a microfilaria within host i , will be of genotype j is denoted , and given by the equations , ( 1 ) ( 2 ) ( 3 ) where and are , respectively , the frequency of allele Y in the male and female adult worms within host i , and and are the corresponding susceptible allele F frequencies . To allow random stochastic fluctuations in genotype distribution , the actual number of microfilariae in host i with genotype j follows a binomial distribution , with the number of trials being equal to the number of microfilariae produced by host i , with genotype probability equal to . Microfilarial allele frequencies and genotype distributions were generated by sampling a specific number of microfilariae from the generated hypothetical population according to the sampling scheme used in Schwab et al . [13] . The exact number of samples taken from each of the 30 hosts was: 11 , 10 , 15 , 9 , 11 , 9 , 13 , 10 , 10 , 7 , 10 , 10 , 7 , 1 , 11 , 9 , 1 , 7 , 4 , 1 , 10 , 9 , 8 , 6 , 4 , 6 , 9 , 10 , 10 , 8 , for a total of 246 microfilariae . Analysis of pre-treatment data had indicated that the number of samples taken from each host by Schwab et al . [13] was independent of host microfilaraemia and host allele frequency , allowing the number of microfilariae sampled per host to be randomly allocated . The program code for the simulations implemented was written in C++ and run 100 , 000 times , with each run generating a new helminth population and genotype distribution from which 95% confidence limits ( 95% CL ) were calculated . The model was parameterised for the untreated villages of Tangonko and Badongo , Burkina Faso , which had an initial prevalence of microfilaraemia of 25% . The mean adult worm burden was estimated from observed microfilarial counts using the functional relationship given in the deterministic model EPIFIL ( see original formulation and parameter values in Norman et al . [40] ) , giving a mean adult worm burden of 13 . 5 host−1 . The degree of adult worm overdispersion was estimated from the recorded microfilarial prevalence ( taken here as a proxy for the prevalence of adult worms producing microfilariae ) and the mean adult worm burden , using the prevalence vs . intensity relationship that derives from assuming a negative binomial distribution of worms among hosts [35] , yielding a k value of 0 . 07 . The model outlined above will only be valid for comparisons against the pre-treatment data , since chemotherapy is known to impede microfilarial production and / or survival [41] . The null model assumes that mating is random between male and female worms within each infrapopulation and that allele Y is randomly distributed across hosts , i . e . . Results of the inbreeding analysis can be incorporated into the individual-based model described in equations ( 1 ) to ( 3 ) to explore the range of adult worm allele frequencies which can give rise to the observed microfilarial data .
The observed microfilarial genotype distribution was found to deviate from HWE . Villages with no history of mass anthelmintic chemotherapy had an overall inbreeding coefficient of FIT = 0 . 44 ( 95% CL = 0 . 17 , 0 . 68 ) , indicating strong inbreeding . Fifteen percent of the microfilariae were found to be homozygous for allele Y , an estimate 2 . 3 times higher than would be expected in a random mating parasite population . Results indicate the occurrence of a significant degree of genetic differentiation in worm allele frequency among the host population . Infrapopulation allele Y frequency , , varied from 0 to 0 . 77 in the villages with no history of treatment , indicating an increase in microfilarial homozygosity of 60% above HWE . The results also suggest a degree of non-random mating within hosts measured by FIS = 0 . 29 ( −0 . 09 , 0 . 54 ) , which is however is not significantly greater than zero . No difference was observed in the microfilarial allele frequency between the two treatment-naïve villages . The data from the two treatment-naïve villages of Tangonko and Badongo were analysed separately . Both showed a high level of microfilarial homozygosity , with overall inbreeding coefficient of FIT = 0 . 51 ( 0 . 16 , 0 . 76 ) and FIT = 0 . 33 ( −0 . 10 , 0 . 78 ) , respectively ( Figure 1 ) . The degree of parasite genetic differentiation between hosts varied between the two villages , though the difference was not statistically significant ( p = 0 . 38 , calculated from the square of the normalized difference in FST estimates [42] ) . For the purpose of the following analysis the two treatment-naïve villages have been grouped together to increase the study sample size . A similar degree of parasite inbreeding was observed in the village of Perigban which had received one round of MDA . Parasite inbreeding increases the range of underlying adult worm allele Y frequencies , qW , which can give rise to the observed microfilarial allele Y frequency of 0 . 26 ( Figure 2 ) . Results from the null model , where mating was assumed to be random and allele Y is randomly distributed amongst hosts , indicate that qW in the untreated villages of Tangonko and Badongo could range from 0 . 21 to 0 . 32 . If we use the excess inbreeding estimate reported in pre-treatment villages ( FIT = 0 . 44 ) , then model simulations suggest that qW could range from 0 . 18 to 0 . 37 . The microfilarial genotype diversity model indicates that the observed homozygosity is unlikely to be solely a result of genetic sampling , demographic stochasticity , population subdivision , or the sampling scheme employed , suggesting that true biological mechanisms are operating in the parasite population even before the introduction of anthelmintic therapy . Figure 2 indicates the range of likely microfilarial genotype distributions that can be generated from a given qW value using the null ( random ) model . The observed excess homozygosity in the untreated villages was greater than the 95% confidence interval estimates generated by the null model ( Figure 3 ) . It is interesting to note the wide range of microfilarial genotype distributions that can be generated by the null model . Despite the large increase in microfilarial homozygosity attributable to parasite inbreeding , there is only a modest increase in the prevalence of hosts who have microfilariae that are homozygous for allele Y ( and therefore putatively resistant if the allele confers drug resistance were recessive , Figure 4 ) . Parasite overdispersion reduces the number of hosts who are microfilaria-positive and concentrates allele Y into a small proportion of the host population . A high degree of parasite non-random mating and infrapopulation genetic differentiation increases the number of hosts ( and the number of samples per host ) that need to be sampled , in order to detect or quantify reliably parasite genetic diversity ( Figure 4 ) . The model is used to investigate how parasite inbreeding may influence the sampling scheme of genetic surveys seeking to identify the presence of a known marker for drug resistance ( Figure 5 ) . Results indicate that the observed level of parasite inbreeding markedly increases the minimum number of hosts , and the overall number of samples necessary to be 95% confident of detecting a rare allele . The sampling scheme used within Figure 5 assumes that the number of parasites genotyped per host is weighted by the host's microfilarial load . This improves the accuracy of allele frequency estimates by allowing heavily infected hosts to have a greater contribution to the sampled microfilarial population , something which is particularly important in overdispersed parasite populations . To date there is no phenotypic evidence that allele Y causes albendazole resistance in W . bancrofti . However , if an allele conferring drug resistance existed in populations of this parasite then the consequences on the spread of such an allele of parasite non-random mating and genetic differentiation between hosts will depend on the frequency and the relative dominance of the resistance allele . If the resistance allele were recessive , helminth inbreeding would greatly increase the probability that a parasite survives anthelmintic treatment . This is evident from Figure 6 which shows the influence of parasite inbreeding on the relative proportion of resistant genotypes for a given allele frequency . With a recessive resistance allele at a frequency of 0 . 05 , the degree of inbreeding within the W . bancrofti population reported here , would on average increase the number of worms with the homozygote resistance genotype nine-fold . Conversely , if the resistance allele was dominant , inbreeding would reduce the probability that a parasite survives chemotherapy , as fewer worms would have the resistant allele ( the deficiency of heterozygous parasites caused by parasite inbreeding will be greater than the increase in resistant homozygous worms ) .
Our results suggest that adult W . bancrofti worms do not mate randomly within the infrapopulation . This is in agreement with ultrasonography studies that show adult parasites congregating in ‘worm nests’ along lymphatic vessels , which remain stable over time [44] . Spatial heterogeneity within the host may produce multiple reproducing populations within each infrapopulation , which would increase host microfilarial homozygosity . Evidence of an apparent relationship between β-tubulin genotype , the same gene analyzed by Schwab et al . [13] , and female worm fertility in the related filaria O . volvulus has been reported by Bourguinat et al . [10] . If such a relationship exists in W . bancrofti , the excess within-host homozygosity reported above may result from the increased fertility of homozygous adult worms . Anthelmintic treatment , prior to the introduction of MDA for lymphatic filariasis , may also have increased non-random mating depending on the selective advantage that allele Y may confer to the parasite at the time of treatment . The degree of genetic differentiation in the parasite infrapopulation can shed insight into the microepidemiology of parasite transmission [19]–[23] . The metapopulation transmission dynamics of W . bancrofti will depend on the transmission efficiency and biting behaviour of the mosquito vector . Anopheles gambiae sensu stricto and An . funestus are thought to be the main vectors of W . bancrofti in Burkina Faso [45] . Hosts can acquire multiple L3 larvae during the same bite . Although density-dependent processes are known to operate on the uptake and development of W . bancrofti in An . gambiae , infective vectors will regularly transmit multiple related L3 larvae simultaneously [46] . Other mosquito vectors of W . bancrofti have even greater vector competence . For example , up to 32 L3 larvae were recovered from an experimental host after it was bitten by a single Culex quinquefasciatus [47] , a main vector in East Africa . Mark-recapture studies and bloodmeal analysis indicate that various mosquito species appear to have high site fidelity , regularly biting multiple members of the same household [48] , [49] . These aspects of W . bancrofti transmission increase the likelihood that a host will be infected with closely related parasites and will contribute to the observed genetic differentiation . More generally , drug treatment may increase infrapopulation genetic heterogeneity , as those parasites within treated hosts which survive treatment may have a higher resistance allele frequency than those harboured within untreated hosts . In Burkina Faso , lymphatic filariasis is treated with albendazole and ivermectin . Evidence indicates that the albendazole plus ivermectin combination has some macrofilaricidal and reproductive effects ( mainly associated with albendazole [41] ) , as well as the microfilaricidal effect ( mainly associated with ivermectin ) . It is possible that a degree of the genetic differentiation between hosts observed in the untreated villages may have resulted from individual members of the community seeking , for instance , treatment for geohelminth infection prior to the introduction of GPELF . Population subdivision and non-random mating will influence the outcomes of selection under chemotherapeutic pressure in different ways , depending on the initial frequency of the allele under selection and the ecology of the infection . Before the rate of spread of drug resistant parasites can be predicted reliably and accurately , greater knowledge would be required regarding the number , linkage , dominance , and possible negative pleiotropic effects of putative resistance allele ( s ) , as well as regarding the pharmacodynamic properties of the drugs administered singly and in combination . However , useful biological insights can be obtained from mathematical models that make reasonable assumptions concerning the above [50] , [51] . If the resistance allele is recessive and it has a low initial frequency , inbreeding will increase parasite homozygosity and as a result , the spread of drug resistant worms across the parasite population ( see Figure 6 and [50] ) . If drug resistance is a semi-dominant trait then parasite inbreeding will either increase or decrease the spread of drug resistance , depending on the efficacy of the drug against heterozygote parasites . Parasite genetic differentiation between hosts will also increase the spread of resistance even when the resistance allele is initially present at a very low frequency , as it increases the probability that male and female resistant worms will inhabit the same infrapopulation . This work is consistent with mathematical models of veterinary helminths which indicate that spatial heterogeneity and aggregated infections between hosts increase the spread of rare recessive genes [27] , [28] . The operation of a strong degree of parasite genetic differentiation between hosts reduces the prevalence of infection with drug resistant parasites and would therefore increase the number of hosts and parasites that should be sampled to detect and quantify the frequency of resistance-conferring alleles reliably . Even at high resistance allele frequencies , some hosts will have no phenotypic signs of resistance , particularly if the resistance allele is recessive , and therefore hosts respond to treatment . In practice the number of parasites that can be genotyped will be restricted , so surveys should carefully consider the sampling scheme they employ in order to maximise the accuracy of allele frequency estimates . Repeatedly sampling from the same host increases the chance of detecting a resistance mutation if it is present in that infrapopulation . However , sampling transmission stages from as many hosts as possible should be considered the optimum strategy , even in a population with low parasite genetic differentiation between hosts , as it reduces the chance of repeatedly sampling offspring of the same adult worm . Prior to the introduction of chemotherapy , studies investigating the presence and frequency of putative resistance markers through genotyping transmission stages alone should weight the number of samples they take per host by the host's infection intensity . However , after the start of chemotherapy the best sampling scheme will depend on the pharmacodynamics of the drug and the nature of the questions under investigation . For human helminth infections , the importance of parasite genetic differentiation between hosts stretches beyond population genetics and will influence the outcomes of parasite elimination campaigns such as the GPELF . The ability of a parasite species to persist in a host population following prolonged MDA will depend in part on the metapopulation dynamics of helminth transmission , the patterns of host compliance with treatment regimes and the pharmacodynamic properties of the drugs used . The aggregated nature of the passage of transmission stages between hosts will make parasite elimination harder to achieve by lowering the breakpoint density ( the unstable equilibrium below which the parasite population will tend naturally to local extinction [52] ) , as overdispersion of parasites will result in fewer hosts with a single-sexed infection .
|
During the last decade , there has been a substantial increase in the use of mass drug administration to reduce the disease caused by parasitic worms . With so many people regularly receiving treatment , there is a risk that drug resistance may develop . As a result , the number of studies looking for genetic markers of drug resistance has increased noticeably . In this paper we analyse the results of one such study that investigated the presence of genes associated with drug resistance in parasites responsible for elephantiasis . This study , like many other studies of human parasitic infections , relies on analysing parasite immature stages ( such as eggs ) because the adult worms are often inaccessible within the human body . Using computer models we show how the gene frequency in the immature stages may vary from that in the adult worm population . Parasites with these markers for drug resistance might also be unevenly distributed across the host population even prior to treatment . This may increase the spread of drug resistance and make it harder to detect . We suggest that studies conducted only on parasite immature stages should be interpreted with caution and should carefully consider the number of people and the number of parasites they analyse .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"helminths",
"variant",
"genotypes",
"population",
"genetics",
"mathematical",
"models",
"parasitic",
"diseases",
"alleles",
"animals",
"genetic",
"mapping",
"infectious",
"diseases/neglected",
"tropical",
"diseases",
"population",
"biology",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases",
"research",
"and",
"analysis",
"methods",
"wuchereria",
"bancrofti",
"wuchereria",
"mathematical",
"and",
"statistical",
"techniques",
"infectious",
"diseases/helminth",
"infections",
"genetic",
"loci",
"inbreeding",
"infectious",
"diseases/antimicrobials",
"and",
"drug",
"resistance",
"eukaryota",
"heredity",
"genetics",
"nematoda",
"biology",
"and",
"life",
"sciences",
"evolutionary",
"biology",
"organisms"
] |
2008
|
An Analysis of Genetic Diversity and Inbreeding in Wuchereria bancrofti: Implications for the Spread and Detection of Drug Resistance
|
The mucosal immune system identifies and fights invading pathogens , while allowing non-pathogenic organisms to persist . Mechanisms of pathogen/non-pathogen discrimination are poorly understood , as is the contribution of human genetic variation in disease susceptibility . We describe here a new , IRF3-dependent signaling pathway that is critical for distinguishing pathogens from normal flora at the mucosal barrier . Following uropathogenic E . coli infection , Irf3−/− mice showed a pathogen-specific increase in acute mortality , bacterial burden , abscess formation and renal damage compared to wild type mice . TLR4 signaling was initiated after ceramide release from glycosphingolipid receptors , through TRAM , CREB , Fos and Jun phosphorylation and p38 MAPK-dependent mechanisms , resulting in nuclear translocation of IRF3 and activation of IRF3/IFNβ-dependent antibacterial effector mechanisms . This TLR4/IRF3 pathway of pathogen discrimination was activated by ceramide and by P-fimbriated E . coli , which use ceramide-anchored glycosphingolipid receptors . Relevance of this pathway for human disease was supported by polymorphic IRF3 promoter sequences , differing between children with severe , symptomatic kidney infection and children who were asymptomatic bacterial carriers . IRF3 promoter activity was reduced by the disease-associated genotype , consistent with the pathology in Irf3−/− mice . Host susceptibility to common infections like UTI may thus be strongly influenced by single gene modifications affecting the innate immune response .
Despite significant advances in the understanding of genetic variation , common infections are often regarded as too complex for genetic analysis . While single gene defects have a major impact on host susceptibility to classic infections like malaria [1] , the extent to which susceptibility to diarrhea , respiratory tract and urinary tract infection ( UTI ) is genetically controlled remains unclear . Critical to the understanding of host resistance and genetic control is the mucosal route of these infections and the molecular interactions through which mucosal tissues are perturbed . UTI serve as a particularly useful model to identify genetic variants contributing to host susceptibility , as innate immunity controls the antimicrobial defense and molecular mechanisms of host parasite interaction are understood in great detail [2] , [3] . The disease response to uropathogenic Escherichia coli is initiated through fimbriae-mediated adherence , and the expression of P fimbriae distinguishes the pathogenic strains from non-virulent bacteria , which colonize the same mucosal sites . TLRs control the survival of complex organisms by balancing protective against destructive forces of innate immunity . During infection , each TLR recognizes a relatively small number of ligands , including conserved microbial patterns ( PAMPs ) [4] . The horseshoe-shaped , leucine-rich , extracellular TLR domain and its co-receptors are involved in recognition of proteins , as well as lipids , carbohydrates and nucleic acids [5] , [6] , [7] . At mucosal sites , where the bulk of microbial challenge occurs , PAMP recognition is non-functional , however , and does not explain how mucosal TLRs distinguish pathogenic microbes from members of the normal flora [8] . Pathogen-specific TLR responses to mucosal pathogens require receptors that exclusively engage virulence ligands and signaling pathways that activate a pathogen-specific defense [8] . For example , uropathogenic E . coli adhere to mucosa via glycosphingolipid receptors for P fimbriae , thereby activating a TLR4-dependent but LPS/CD14-independent innate immune response in epithelial cells [9] . Signaling through cell surface sphingolipids involves ceramide , the membrane anchor and a ubiquitous component of cell membranes [10] , [11] . The generation of ceramide within rafts alters their biophysical properties and results in the formation of large ceramide-enriched membrane platforms , clustering receptor molecules and facilitating signal transduction following receptor stimulation [12] . Endogenous SMases , activated by many infectious agents , cleave ceramide from the extracellular choline-rich domain of sphingomyelin [13] , [14] , [15] , [16] , [17] and activate the “ceramide-signaling pathway” , which is conserved from yeast to humans [18] . In addition , pathogens that utilize the extracellular domain of glycosphingolipids as receptors may release ceramide after bacterial binding , as first described for P-fimbriated , uropathogenic E . coli [9] , [17] , [19] . Ceramide activates a TLR4-dependent innate immune response [17] , similar to infection-mediated activation , and we have proposed that ceramide acts as a signaling intermediate between the pathogen-specific receptors and TLR4 [9] , [17] , [19] . The molecular mechanisms in this important signal transduction need to be identified , however . UTIs affect >150 million adults each year and about 5% of children <12 years of age . Severe kidney infections like acute pyelonephritis ( APN ) are accompanied by life-threatening urosepsis in about 30% of adults . Children may develop renal scars , which are associated with long-term morbidity including hypertension , complications of pregnancy , and renal failure if scarring is extensive . Despite the urgent need , no tools exist at present to identify children at risk of developing recurrent acute pyelonephritis and renal scarring . Host resistance to UTI is controlled by the innate immune system , through Toll-like receptor ( TLR ) activation [8] , [9] , [20] . Previous studies have shown that TLR4-deficient mice develop asymptomatic carriage rather than severe disease [8] , [20] , [21] , suggesting that disturbances in TLR4 signaling may alter the innate immune dependent host defense [22] . This study examined TLR4 activation by ceramide and P fimbriated E . coli and characterized this signaling pathway . We propose that ceramide interacts directly with TLR4 , activates TRAM phosphorylation followed by nuclear translocation of IRF3 . Furthermore , we show that IRF3 dependent innate immunity is essential for the host defense , as Irf3 knockout mice develop severe kidney infection . Finally , we show that IRF3 promoter polymorphisms are more common in APN prone patients than in those who become asymptomatic bacterial carriers . We propose that this pathway offers a model of how TLR4 may distinguish pathogens from commensals at the mucosal level , through modification of pathogen recognition receptors , adaptors and transcription factors .
In a genetic screen of innate immune effector genes downstream of TLR4 , we identified IRF3 as a major determinant of host susceptibility . Irf3−/− and Irf3+/+ mice were infected via the urinary tract mucosal route with the uropathogenic E . coli strain CFT073 [23] . The Irf3−/− mice developed more severe disease than wild type ( wt ) Irf3+/+ mice . Acute mortality was higher ( 50% after 24 hours ) and bacterial clearance was significantly impaired , with higher bacterial counts in urine , kidneys and bladders ( Figure 1A , p<0 . 001 ) . Abscess formation was also more extensive in Irf3−/− than in wt mice ( day 7 post-infection , Figure 1B–C , p<0 . 001 ) . In Irf3−/− mice , abscesses were diffuse , destroying large tissue areas while in wt mice abscesses were morphologically distinct from surrounding healthy tissue ( Figure 1B ) . Renal abscess formation is caused , in part , by an imbalance between neutrophil recruitment and exit from the tissues [24] , [25] . The kinetics of early neutrophil recruitment did not differ between wild type and Irf3−/− mice ( Figure 1A ) , but later , neutrophil recruitment subsided in wt mice but remained elevated in Irf3−/− mice . In tissue sections from Irf3−/− mice , neutrophils were detected throughout the abscesses and P-fimbriated bacteria were interspersed among the neutrophils , as shown by PapG adhesin-specific antibody detection ( Figure 1D , for negative control , see Figure S1 ) . Wt mice , in contrast , had discrete , neutrophil aggregates with fewer bacteria . The results suggest that Irf3 is essential for a functioning innate immune defense against UTI , to maintain tissue integrity and to clear mucosal E . coli infection . To examine if the IRF3-dependent immune response discriminates uropathogenic E . coli from non-pathogenic bacteria , we inoculated wt and Irf3−/− mice with the prototypical asymptomatic bacteriuria strain E . coli 83972 , which lacks functional UTI-associated virulence factors , including P fimbriae [26] , [27] , [28] , [29] . Both wt and Irf3−/− mice cleared infection rapidly , with no difference in bacterial counts ( Figure 1E ) and no significant neutrophil recruitment ( data not shown ) . As P fimbriae are essential virulence factors , present in up to 100% of E . coli strains causing urosepsis [30] , [31] , we subsequently examined if P fimbriae activate the IRF3 pathway . The asymptomatic carrier strain E . coli 83972 was transformed with a chromosomal copy of the pap gene cluster . We compared disease severity and bacterial counts between wt and Irf3−/− mice infected with E . coli 83972pap . The Irf3−/− mice developed acute , symptomatic disease with sepsis and had dramatically increased bacterial numbers in bladders , kidneys and spleens ( Figure 1E , p<0 . 05 ) , compared to wt mice , which were resistant to infection with E . coli 83972pap . The results show that the IRF3-dependent response distinguishes pathogenic E . coli from non-pathogenic strains and suggest that the expression of a single virulence factor like P fimbriae enables the host to recognize a potential pathogen and to activate this response . P fimbriae bind to glycosphingolipid receptors and trigger ceramide release [9] . To investigate the mechanism of pathogen-specific TLR4/IRF3 signaling activation , we examined if ceramide and TLR4 interact after ceramide release from membrane glycosphingolipids . We treated A498 kidney epithelial cells with sphingomyelinase ( SMase ) for one hour , to release ceramide ( r-ceramide ) from the extracellular phosphocholine domain of sphingomyelin [32] ( Figure 2A–D ) . We labeled TLR4 and native ceramide with specific primary antibodies followed by Alexa fluor-488 ( donor ) and Alexa fluor-568 ( acceptor ) -labeled secondary antibodies , respectively . In unstimulated cells ( no SMase treatment ) , where ceramide remains bound to sphingomyelin , we detected no FRET signal . After SMase treatment , we recorded a significant FRET signal ( Figure 2A–D , 50% FRET-positive cells compared to 8% for unstimulated cells , p<0 . 05 ) , with most of the FRET-positive regions localized in the plasma membrane . LPS and soluble CD14 ( sCD14 ) stimulation , in contrast , did not stimulate a FRET signal above background ( p>0 . 05 , compared to unstimulated cells ) . sCD14 was used , as the uroepithelial cells lack membrane-bound CD14 and respond poorly to LPS [33] . These results suggest that ceramide interacts with TLR4 after release from membrane glycosphingolipids . To examine the ceramide-induced TLR4 signaling pathway , we used RNA interference to suppress specific genes ( Figure 2E , siRNA used for transfection are listed in Table S1 in Supporting Information S1; for knockdown efficiency compared to control cells transfected with irrelevant siRNA , see Figure S2 ) . First , suppression of TLR4 expression abrogated the innate immune response to r-ceramide ( p<0 . 001 ) , confirming that this pathway is TLR4 dependent . Secondly , TRAM siRNA inhibited the responses to r-ceramide ( p<0 . 05 compared to the siRNA control ) . MyD88-specific siRNA did not alter the ceramide response ( p>0 . 05 compared to the siRNA control ) but did reduce the response to LPS+sCD14 ( p<0 . 05 ) , as did TLR4- and TRAM-specific siRNAs . To further investigate ceramide-induced TLR4 signaling , TRAM phosphorylation ( TRAM-P ) was quantified by confocal microscopy , using polyclonal phospho-specific anti-TRAM antibodies ( Figure 2F–G and Figure S3 ) . We detected an increase in TRAM-P staining in cells exposed to r-ceramide or exogenous , water-soluble C6 ceramide; staining had a granular appearance and was most intense in the perinuclear area . By Western blot analysis ( Figure 2H ) , a band corresponding to TRAM-P was increased in cells exposed to C6 and r-ceramide compared to unstimulated cells but total TRAM levels were not altered . LPS+sCD14 triggered weaker TRAM phosphorylation , as shown by confocal microscopy ( p<0 . 001 compared to r-ceramide ) and by Western blot . The results indicate that ceramide triggers TRAM phosphorylation more efficiently than LPS+sCD14 . As TRAM phosphorylation was virtually absent in unstimulated cells , this pathway may need to be activated by exogenous or endogenous stimuli . To define signaling downstream of ceramide/TLR4 and TRAM , we examined kinase phosphorylation , using phosphoarrays specific for 46 protein kinases and substrates ( Figure 3A ) . Ceramide release stimulated the phosphorylation of twelve protein kinases: p27T198 , eNOS , CREB , Fyn ( all 2 . 3-fold ) , Hck , PLCγ1 , Jun ( all 2 . 1-fold ) , Pyk2 ( 2-fold ) , ERK1/2 and Src ( 1 . 9-fold ) , RSK1/2/3 ( 1 . 8-fold ) , p27T157 ( 1 . 7-fold ) , and p53 ( 1 . 6-fold ) . Antibacterial effectors included eNOS , which regulates nitric oxide and related antibacterial effector functions [34] and Hck , a Src-family tyrosine kinase associated with secretory lysosomes in neutrophils and phagosome-lysosome fusion [35] . A number of the significantly phosphorylated proteins activate IRF3- and AP1-dependent transcription . PLCγ1 catalyzes the formation of inositol 1 , 4 , 5-trisphosphate and diacylglycerol from phosphatidylinositol 4 , 5-bisphosphate , leading to PKC activation and CREB ( cAMP response element binding ) phosphorylation [36] , [37] . CREB is then phosphorylated and binds to CBP ( CREB-binding protein ) , which preferentially associates with phosphorylated IRF3 [38] , [39] , leading to IRF3 . Fyn is a Src family tyrosine kinase implicated in the activation of PKA , a protein kinase involved in CREB phosphorylation [40] . Jun in combination with Fos bind to and are a part of the AP-1 transcription factor complex [41] , which induces the transcription of proinflammatory cytokines . Pyk2 activation is highly correlated with the stimulation of c-Jun N-terminal kinase ( JNK ) . Identified phosphorylation targets also included ERKs ( ERK1/2 , extracellular signal-regulated kinases ) which activate downstream protein kinases and transcription factors , including IRF3 and AP-1 [42] . CREB phosphorylation in r-ceramide-activated cells was confirmed by confocal microscopy ( Figure 3D , E , p<0 . 001 compared to control ) , but was not detected in LPS-stimulated cells . We obtained similar results using antibodies specific for phosphorylated Fos ( Figure 3D , E ) . JNK phosphorylation , in contrast , was similar after r-ceramide and LPS+sCD14 stimulation ( Figure 3D , E ) , suggesting that JNK signaling was not ceramide-specific ( p<0 . 001 compared to the control ) . The results suggest that ceramide-induced TLR4 signaling causes rapid phosphorylation/transcription of proteins involved in IRF3 and AP-1 transcription , including CREB , Fyn , PLCγ , MAP kinases , ERK1/2 and Fos/Jun ( Figure S4 ) . LPS+sCD14 , in contrast , caused a weaker phosphorylation response , comprising p27T198 ( 2-fold ) , eNOS ( 1 . 8-fold ) , PLCγ1 ( 1 . 7-fold ) , Pyk2 ( 1 . 7-fold ) , and Jun ( 1 . 6-fold ) , but not the remaining targets that were phosphorylated in response to r-ceramide ( Figure 3A ) . Innate immune activation in response to ceramide was further examined by TLR SuperArrays and compared to LPS+sCD14 ( Figure 3B , Figure S5 ) . After one hour , five genes in A549 cells had responded to r-ceramide: Fos ( 6 . 5-fold ) and Jun ( 2 . 1-fold ) , IL-8 ( 4 . 4-fold ) , IL-6 ( 2 . 9-fold ) and IL-1α ( 2-fold ) . The response showed a similar , restricted repertoire in A498 carcinoma cells after three hours ( Figure 3C ) . r-Ceramide upregulated TRAM , Fos and Jun transcription ( 10 . 2- , 2 . 2- and 2 . 4-fold , respectively ) . Ceramide activated IL-6 transcription ( 6 . 2-fold ) , MAP3K1 and MAP2K3 ( 5 . 9- and 2 . 5-fold ) , as well as IL-8 and CSF2 transcription levels ( about 3-fold ) . In contrast , LPS+sCD14 did not significantly stimulate Fos ( 1 . 5-fold ) , Jun ( 1 . 2-fold ) or IL-1α after one hour . After three hours , only IL-8 transcription was higher in response to LPS+sCD14 than to r-ceramide . The transcriptional profile confirmed the difference between ceramide and LPS+sCD14 activated cells , consistent with a different transcription factor usage . IRF3 is an interferon regulatory transcription factor and following TLR4 activation , phosphorylated IRF3 homodimers translocate from the cytosol to the nucleus [43] , [44] , [45] , [46] . By confocal microscopy ( Figure 4A , C ) we observed that r-ceramide triggered IRF3 translocation to the nucleus ( p<0 . 001 compared to unstimulated cells , 90 min ) . We confirmed the results in a human bladder epithelial cell line ( J82 , Figure S6A ) in which ceramide release caused rapid IRF3 translocation . In cells exposed to LPS+sCD14 , the nuclear IRF3 translocation was weak ( p>0 . 05 compared to control ) and fewer dimers were formed after exposure to LPS+sCD14 . In the bladder epithelial cells , the IRF3 response to LPS+sCD14 or LPS alone was low . LPS+sCD14-induced NF-κB p65 translocation , but the NF-κB response to r-ceramide or C6 ceramide was weak ( Figure 4 B , D ) . For a broader field of view see Figure S6B . Signaling through p38 MAPK has previously been shown to stimulate proinflammatory responses , including IL-8 , IL-6 and TNF [47] , [48] , [49] . The activation of MAP3K1 and MAP2K3 by r-ceramide exposure of A549 cells suggested that this pathway might be involved upstream of IRF3 . Pretreatment of the cells with a p38 inhibitor ( SB202190 ) reduced the IL-8 response to r-ceramide ( Figure 4E ) and prevented nuclear translocation of IRF3 ( Figure 4F , G ) . NF-κB p65 translocation was not affected by the p38 inhibition ( Figure 4F , G ) . The results suggest that ceramide/TLR4 activates IRF3- rather than NF-κB-dependent transcription , and that the IRF3 response involves p38 MAPK-dependent mechanisms . CREB-phosphorylation was also markedly reduced after p38 inhibition , as shown by confocal microscopy ( >99% in A498 cells , >50% in A549 cells , Figure 4H ) and Western blots ( Figure S7 ) . Previous work has suggested that the phosphorylation of TRAM is mediated by PKC-epsilon , which is activated downstream of TLR4 [50] . Given that PKC-epsilon is also essential for IRF3 activation , this pathway was examined , using the pan-PKC inhibitor Bisindolylmaleimide II . The inhibitor reduced the response to PMA , which was used as a PKC dependent , positive control . In contrast , the response to ceramide was not impaired ( Figure S8 ) . To further examine the relationship of the ceramide/TLR4 pathway to the classical IRF3 activation pathway , cells were transfected with TBK1 siRNA and responses were compared to irrelevant siRNA transfected cells . In parallel , the cells were transfected with TLR4 and TRAM siRNAs ( Knock down efficiency for TLR4 and TRAM was >90% and 64% for TBK1 , Figure S9 ) . IRF3-P responses to r-ceramide were reduced by the TLR4 and TRAM siRNAs but were less affected by suppression of TBK1 expression . The response to LPS+sCD14 showed a similar pattern ( Figure S9 ) . These results suggest that the pathway of IRF3 activation identified here has several new features , including p38 dependence and PKC independence . The involvement of TBK1 needs further study . To examine if the IRF3 response is triggered in a pathogen-specific manner involving P fimbriae , we stimulated primary cultures of human CD14+ renal tubular epithelial cells ( HRTEC ) with isogenic P-fimbriated ( E . coli S1918pap ) or type 1-fimbriated ( E . coli S1918fim ) E . coli strains and examined IRF3 by confocal microscopy . Non-fimbriated E . coli S1918 was used as a control . E . coli S1918pap induced higher nuclear IRF3 translocation and IRF3 phosphorylation than E . coli S1918 , consistent with results in ceramide-stimulated cells ( Figure 5A and Figure S10 , p<0 . 01 for a broader view ) . There was less IRF3 translocation in response to E . coli S1918fim or to the non-fimbriated control E . coli S1918 . All three strains stimulated an NF-κB response , but NF-κB translocation was higher in cells infected with P-fimbriated E . coli compared to type 1-fimbriated E . coli ( Figure 5A , p<0 . 05 ) . Uninfected cells showed no evidence of nuclear IRF3- or NF-κB translocation . The same phenomenon was observed in A498 kidney epithelial cells ( Figure S11 ) . In addition , preliminary Western blot analysis of IRF3P in infected cells suggested that S1918pap and S1918fim stimulated a higher response than S1918 ( Figure S11 ) . IRF3 target gene expression was examined by microarray analysis . We infected A498 kidney epithelial cells in vitro with virulent CFT073 or non-pathogenic E . coli ( 4 hours , 108 CFU/ml ) and complementary RNA was hybridized to Illumina whole genome microarrays . There was a dramatic IFNβresponse to infection ( 24-fold above uninfected control cells , ≥log 2 cut off ) ( Figure 5B ) . By Ingenuity Pathway analysis , we detected significant activation of several members of the interferon-signaling pathway such as IFIT1 , STAT1 ISG15 , IP-10 and IFNAR2 . A weaker ISG15 response was observed ( 1 . 8-fold above background ) . By RT-PCR , a strong IFNβ response to r-ceramide was confirmed in human kidney cells ( A498 , Figure 5 B and C ) . To examine if the effects of IRF3 on host susceptibility are IFNβ-dependent , we infected Ifnβ−/− mice with E . coli CFT073 and examined parameters of disease and bacterial persistence ( Figure 5E ) . Bacterial clearance was drastically impaired in Ifnβ−/−mice compared to wt controls ( Figure 5E , p<0 . 001 between Ifnβ−/−and wt mice in urine , kidneys and bladders ) . The Ifnβ−/− mice also developed abscesses . Positive spleen cultures confirmed systemic spread of infection in these mice , which also developed symptomatic disease and were sacrificed on day three ( Figure 5E ) . The results suggest that IFNβ is activated by infection and that IFNβ might be an essential effector molecule in IRF3-dependent bacterial clearance . To examine if UTI susceptibility is associated with differences in IRF3 promoter efficiency , IRF3 promoter sequence variation was studied in two highly UTI-prone patient populations . Sample 1 comprised children in southern Sweden , with a consistent UTI pattern over several years: either severe recurrent kidney infections ( APN; n = 21 ) or asymptomatic carriage of E . coli with no prior symptomatic infection ( primary asymptomatic bacteriuria , ABU , n = 16 ) . These children were identified after prospective , long-term follow-up of a larger patient group . Sample 2 comprised adults in western Sweden , with a history of childhood UTI ( n = 82 ) . They were enrolled in a prospective study of febrile UTI ( APN ) in the 1970s and were recently re-evaluated , after about 30 years , to investigate UTI morbidity and long-term effects on health and kidney function . Both samples included additional patients who developed ABU secondary to an APN episode ( secondary ABU , n = 16 in sample 1 and n = 61 in sample 2 ) . Controls were children without UTI or related morbidity ( n = 27 ) and adult blood donors ( n = 62 ) from the same areas . DNA sequencing of IRF3 promoters from UTI patients revealed variation at the −925 and −776 positions . SNPs −925 and −776 were linked in the study population ( r2 = 1 . 0 ) but the IRF3 genotype varied with UTI severity ( Figure 6A–B ) . Genotype counts for −925 and −776 were in Hardy Weinberg Equilibrium across both case and control samples apart from the APN group ( χ2 = 47 , p<0 . 001 ) , indicating effects of genetic drift in the APN group . We observed significant differences for the two studied markers between cases and controls in allelic or genotypic models . In sample 1 , most of the APN patients were homozygous for the two positions ( A/A–C/C , 79% vs . 25% in primary ABU , Figure 6A p = 0 . 0017 ) . The results in APN patients were confirmed in sample 2 , with 75% homozygous and 13% heterozygous SNPs compared to 53% and 37% in adult controls . The differences were confirmed when the two samples were combined , as shown in Figure 6B . Furthermore , the minor allele frequency was decreased in APN compared to primary ABU ( p = 0 . 0103 ) and controls ( p = 0 . 0239 ) ( Figure 6B ) . The minor allele frequency for paediatric UTI patients , adult UTI patients and the relevant controls are demonstrated in Supplemental Table S4 in Supporting Information S1 . The IRF3 genotype of the secondary ABU patients resembled the APN groups in both samples , consistent with their prior APN episodes . To examine if the IRF3 promoter variation influences transcription efficiency , we cloned promoters from one patient with APN and one with ABU into a luciferase reporter vector . We then changed the APN haplotype at positions −925 , −776 to the predominating ABU haplotype ( A-C to G-T ) , or the ABU haplotype to the APN haplotype ( G-T to A-C ) by site-directed mutagenesis . We then transfected A498 human kidney epithelial cells with the different promoter constructs and determined luciferase activity ( Figure 6C ) . The promoter was functional in these cells , resulting in luciferase activity above the vector control . Transcriptional activity from the APN promoter ( A-C ) was about 50% lower compared to the ABU ( A-T ) promoter ( Figure 6C ) . This difference could be attributed to the polymorphic sites , as the promoter activity increased when the APN A-C haplotype was mutated to G-T and decreased when the ABU G-T haplotype was mutated to A-C ( p<0 . 001 ) . This difference was confirmed by cloning the IRF3 promoters from three additional APN ( A-C ) and three ABU ( G-T ) patients ( p = 0 . 001 ) . The results show that the IRF3 promoter efficiency is reduced by the SNPs occurring in about 80% of APN patients , consistent with the human SNPs reducing IRF3 expression and increasing the risk for APN .
IRF3 was originally described as a transcription factor controlling interferon responses to viral infection [51] . More recently , the involvement of IRFs in antibacterial defense and immunoregulation by TLRs has received more attention , since NF-κB , IRF3 and AP-1 form transcriptional complexes that regulate innate immune responses in monocytes [52] . The relevance of IRF3 to human pathology has not been investigated , however . We show that IRF3 is activated in a pathogen-specific manner by P-fimbriated , uropathogenic E . coli , through a new signaling pathway involving TLR4 , TRAM , CREB and p38 . In the absence of IRF3 , acute morbidity and extensive tissue damage are dramatically augmented , consistent with the need for this pathway to maintain a functional antimicrobial defense . Host susceptibility to common infections like UTI may thus be strongly influenced by single gene modifications affecting the innate immune response . Mucosal pathogens exploit the extracellular domains of sphingolipids as receptors for AB toxins such as Shiga and cholera toxin , as well as attachment ligands for Pseudomonas aeruginosa , HIV gp120 and uropathogenic E . coli [53] , [54] , [55] . This study provides evidence that P-fimbriated E . coli , SMase and exogenous , free ceramide all activate the IRF3-dependent innate immune response . Soluble , exogenous ceramide and SMase were used in parallel in these experiments to ensure that the synthetic , short-chained form of ceramide and the endogenous , long-chained form adequately represented the membrane-anchored species in intact cells . SMase is contaminated by low amounts of LPS , but these trace amounts were insufficient to activate an innate immune response in the CD14-negative mucosal cells used in this study [9] , [14] . FRET analysis showed that ceramide is approximated to TLR4 in the cell membrane , suggesting that a direct interaction with TLR4 and/or the early adaptors trigger this pathway . Such ceramide-induced TLR4/IRF3 signaling might offer a general mechanism for host sensing of pathogens that perturb membrane sphingolipids in mucosal cells . To provide further evidence that ceramide signaling via TLR4 to IRF3 discriminates virulent from commensal bacteria , we infected Irf3−/− mice with a commensal-like E . coli strain from a patient with asymptomatic bacteriuria . This strain did not trigger a significant response and was cleared efficiently , suggesting that the IRF3 pathway was not alerted . In constrast , a P-fimbriated transformant triggered rapid , septic infection in the Irf3−/− but not in wt mice linking this virulence factor that recognizes glycosphingolipid surface receptors , to the IRF3-dependent host response . In contrast , IRF3 was not activated by a type 1 fimbriated isogenic strain , suggesting a preference for glycosphingolipid rather than glycoprotein receptors . This does not negate the previous finding that FimH acts as an immune inducer , protecting against viral infection associated with TLR4 and type 1 interferon signaling , but suggests that the mechanisms differ [56] . The results confirm the pathogen specificity of the IRF3 response and the role of P fimbriae as a virulence ligand triggering this response . As a consequence of this selective IRF3 activation , the uropathogenic or P-fimbriated , commensal E . coli strains influenced epithelial gene transcription in a pathogen-specific manner . IRF3 phosphorylation in response to ceramide was controlled by TLR4 and TRAM , as shown using specific siRNA knock down . Activation was not TBK1 or PKC dependent , however , suggesting alternative activation compared to previously described mechanisms of IRF3 activation [50] . A schematic of the identified kinases and targets is given in supplemental Figure S4 . Although this signaling pathway has not been entirely deciphered , a strong involvement of TRAM and CREB was detected as well as involvement of p38 MAPK-dependent events . In this model , IRF3 activation was not controlled by PKC dependent mechanisms , however . The involvement of TBK1 is not clear , but preliminary experiments did not provide evidence that TBK1 controlled IRF3 phosphorylation in this pathway . In addition , P-fimbriated E . coli strains and ceramide significantly activated NF-κB , thus providing a broad basis for the innate immune response to the intact , complex pathogen . Importantly , the IRF3 response differed after LPS+sCD14 stimulation , further suggesting that pathogen recognition and pattern recognition agonists trigger partially different signaling pathways . The phenotype of Irf3−/− mice predicted that reduced IRF3 expression could also increase human susceptibility to severe kidney infection . In support of this hypothesis , there were marked promoter sequence differences between children with ABU or APN in a long-term prospective study and we confirmed an association of polymorphisms to disease severity in adult patients who were followed for about 30 years after their first febrile UTI episode . In the past , we have shown that genetic variation affecting innate immunity modifies human UTI susceptibility [2] . Chemokine receptor expression and neutrophil function are modified by CXCR1 expression , and promoter variants reducing TLR4 expression are coupled to asymptomatic bacteriuria [22] , [57] , [58] , [59] , [60] . The present study adds IRF3 to this short list of polymorphic innate immune response genes that distinguish asymptomatic carriers from APN-prone patients . The human IRF3 promoter has a number of transcription factor binding sites , including a HOX box , three SP1 sites , NF1 , USF , SRF and IRF1-like site and functional elements are within a 113-nucletide long fragment , containing one Sp-1 site , the IRF1-like site , NF1 and HOX box . SNP −925 is located within this region , indicating a possible role in promoter efficiency . IRF3 promoter SNPs were first described in patients with systemic lupus erythematosus ( SLE ) [61] . It was speculated that the A-C haplotype increased IRF3 transcription and that the G-T haplotype might protect against SLE by reducing type I IFN production . The effect of the Irf3 deletion on disease susceptibility in mice suggested , however , that risk might be associated with reduced , rather than increased , IRF3 function . This idea was supported by luciferase reporter assays designed to test the IRF3 promoter sequences typical of APN- or ABU-prone individuals . UTIs are among the most common bacterial infections in man , and remain a major cause of morbidity and mortality [62] , [63] . A subset of disease-prone individuals is at risk for recurrent severe pyelonephritis and renal dysfunction . Therefore , there is a need to identify and treat these patients , preferably in infancy , when many of them experience their first febrile UTI episode . Although predictive diagnostic tools have been suggested [58] , [59] , [60] , the present study identifies IRF3 for the first time as an innate immune response gene involved in UTI . Thus , IRF3 may be a new molecular target in the diagnosis of UTI susceptibility , potentially creating more precise approaches for detection and prevention of severe , recurrent kidney infection and associated debilitating morbidity .
For research involving humans , informed written consent was obtained from all participants or their parents/guardians . The study was approved by the Ethics Committee of the medical faculty , Lund University , Sweden ( LU106-02 , LU236-99 ) . All the animal experiments were performed with the permission of the Animal Experimental Ethics Committee at the Lund District Court , Sweden ( numbers M166-04 and M87-07 ) . Experimental UTI was performed in a level P2 biohazard laboratory within the MIG animal facility and was governed by the following directive , law , ordinance and provisions: Council Directive EG 86/609/EEC , the Swedish Animal Welfare Act ( Djurskyddslag: 1988:534 ) and the Swedish Animal Welfare Ordinance ( Djurskyddsförordning: 1988 :539 ) . Provisions regarding the use of animals for scientific purposes: DFS 2004:15 , DFS 2005::4 , SJVFS 2001:91 , SJVFS 1991:11 . SMase ( Staphylococcus aureus ) , bovine serum albumin , SDS , LPS ( Salmonella typhimurium ) , C6 ceramide , SB202190 and Bisindolylmaleimide II were from Sigma Aldrich , St Louis , MO , USA . Soluble CD14 ( sCD14 ) was from Biometec , Greifswald and IL-8 was quantified by Immulite 100 , Siemens , Germany . Lipofectamine 2000 transfection reagent was from Invitrogen . siRNA downregulation ( Supplemental Table S1 in Supporting Information S1 ) was validated by qRT-PCR , using primers: TLR4 ( Hs00152939 , Applied Biosystems ) , MyD88 ( QT00203490 , Qiagen ) , TRAM ( QT00033341 , Qiagen ) , TBK1 ( QT00078393 ) . TBK1 siRNA ( sc-39058 ) , a pool of 3 target-specific 19–25 nt siRNAs was from Santa Cruz Biotechnology ( USA ) . Human Phospho-Kinase Array Kit ARY003 was from R&D Systems , Abingdon , Oxford , UK . Transcriptome analysis of r-ceramide or LPS+sCD14 activated cells ( 1 or 3 hours ) was by Superarray ( PAHS018 , SaBioscience ) for 84 TLR signaling pathway genes . IRF3 promoter SNPs were identified by Pyrosequencing using the PSQ 96 SNP Reagent Kit ( Biotage , Uppsala , Sweden , Supplemental Table S1 in Supporting Information S1 ) . Rabbit anti-human TLR4 primary antibodies were from eBioscience , CA , USA , mouse anti-ceramide primary antibodies , clone MID 15B4 from ALEXIS Corporation , Lausen , Switzerland . Rabbit anti-human primary antibodies against CREB-P ( Ser 133 ) , Fos-P ( Thr 232 ) , JNK-P ( Thr 183/Tyr 185 ) and IRF3 and mouse anti-human-NF-κB p65 antibodies from Santa Cruz Biotechnology ( USA ) , rabbit anti-IRF3-P ( Ser 396 ) antibodies were from Cell Signaling Technology . Rabbit anti-human-TRAM-P ( raised against the N-terminal end of the protein , aa 7–21 containing phosphoSer at aa 16 ) and TRAM antibodies were from FabGennix Inc . , Frisco , TX , USA , NIMP-R14 rat anti-mouse neutrophil specific antibodies from Abcam , Cambridge , USA , polyclonal rabbit antiserum raised against a peptide within the PapG adhesin ( CRPSAQSLEIKHGDL ) was used to detect P-fimbriated E . coli . Alexa 488 anti-rat IgG , Alexa 568 anti-rabbit IgG , Alexa 568 anti-mouse IgM and Alexa 488 anti-mouse-IgG secondary antibodies were from Invitrogen , Eugene , Oregon , USA . Swine anti-rabbit immunoglobulins-HRP secondary antibodies were from DAKO A/S , Glostrup , Denmark and Santa Cruz Biotechnology ( USA ) . FRET and fluorescence microscopy was by LSM510 META confocal microscope ( Carl Zeiss , Oberkochen , Germany ) . The human lung adenocarcinoma A549 ( ATCC CCL-185 ) and kidney carcinoma A498 ( ATCC HTB-44 ) epithelial cell lines were grown in RPMI 1640 supplemented with 1 mM sodium pyruvate , 1 mM non-essential amino acids , 50 µM/ml gentamicin , and 5% FBS . Human renal tubular epithelial cells ( HRTEC ) were isolated as described [64] . Cells were maintained at 37°C+5% CO2 in a humidified atmosphere , split weekly and exposed to P-fimbriated , type I fimbriated or non-fimbriated E . coli , 0 . 1–1 U/ml of SMase , freshly prepared C6 ceramide or LPS+sCD14 . IL-8 secretion was quantified by Immulite 100 ( Siemens , Bad Nauheim , Germany ) . LSM 510 Meta confocal laser-scanning microscopy was used for FRET acceptor photobleaching and imaging of epithelial cells . Cell stimulation/infection was in 8-well chamber slides ( LabTek , Nunc , RPMI+5% foetal calf serum ) . The cells were first stimulated with r-ceramide ( 1U of SMase/ml ) or LPS+sCD14 ( 10+1µg/ml ) , fixed with 3 . 7% formaldehyde and stained with a mouse antibody specific for native ceramide ( sphingosine-trans-D-erythro-2-amino-4-octadecene-1 . 3-diol ) and with a rabbit polyclonal antibody , specific for the extracellular domain of TLR4 ( amino acids 6–169 ) . Secondary antibodies to TLR4 and free ceramide were conjugated with Alexa-488 ( donor ) and Alexa-568 ( acceptor ) , respectively . FRET efficiency was estimated in percent of fluorescence increase calculated by: FRET efficiency = ( ( IDA-IDB ) /IDA ) ×100% where IDA is the donor intensity after bleaching and IDB the donor intensity before bleaching . A549 human epithelial cells in 24-well plates ( TPP ) were siRNA transfected , using Lipofectamine 2000 ( Supplemental Table S1 in Supporting Information S1 ) . Knockdown efficiency was validated by qRT-PCR . After a 72 h incubation , transfected cells were stimulated with SMase ( 1 U/well ) or LPS+sCD14 ( 10+1µg/ml ) . Supernatants were collected after 24 h . Total extracted mRNAs were converted to cDNA using RT2 First Strand Kit ( SABioscience Corporation , Fredrick , MA , USA ) . The transcriptomic profile of cells exposed to r-ceramide or LPS+sCD14 was examined using a RT-PCR-based superarray , containing 84 genes involved in TLR signaling ( SABiosciences , PAHS018 ) . Gene expression levels were calculated by the ΔCt method and normalized to five housekeeping genes . RT-PCR was used to determine the efficiency of siRNA knockdown . The TaqMan system was used to quantify TLR4 and GAPDH cDNA and the QuantiTect SYBR Green systems was used to quantify other genes of interest . cDNA was quantified by RT-PCR using a Rotor gene 2000 instrument ( Corbett Life Science , Sydney , Australia ) and normalized against GAPDH . Protein phosphorylation was quantified using the Human Phospho-Kinase Array Kit ( Proteome Prolifer Array , R&D Systems , Abingdon , Oxford , UK ) . Protein extracts were prepared from 100% confluent A549 cells cultured in 6-well plates and treated with 1U/well SMase or LPS+sCD14 ( 10+1µg/ml ) . Untreated cells were used as control . The signals were detected with the ECL Plus Western Blotting Detection System ( GE Healthcare ) . In order to detect phosphorylated TRAM , A549 cells grown in 6-well plates were stimulated with 1U/mL SMase , 0 . 1 µg/mL LPS+sCD14 ( 10+1µg/ml ) , 15 µg/mL C6 ( Sigma ) or RPMI medium alone for 45 and 90 min . Cells were lysed in ice-cold buffer ( 10 mM HEPES-KOH , 5 mM EDTA , 0 . 5% Nonidet P-40 and 10 mM KCl , pH 7 . 9 ) containing a protease inhibitor mix ( Complete , Roche Diagnostics GmbH , Mannheim , Germany ) and 1mM Na3VO4 . After 10 min incubation , lysates were centrifuged for 5 min at 12000 g at 4°C and protein concentrations in the collected supernatants were quantified using the DC protein assay kit ( Bio-Rad , Hercules , USA ) . Proteins in the lysates were separated by SDS-PAGE ( 4–12% NuPAGE Bis-Tris gels , Invitrogen ) on ice with NuPAGE MES SDS running buffer ( Invitrogen ) . Proteins were transferred to a polyvinylidene difluoride ( PVDF ) membrane using NuPAGE transfer buffer ( Invitrogen ) and the membrane was probed with phospospecific primary antibodies followed by HRP-labeled , swine anti-rabbit IgG . Bound antibodies were visualized with the ECL Plus Western Blotting Detection System . IRF3 dimerization was detected by native PAGE and immunoblotting . A549 cells cultured in 6-well plates were stimulated with medium alone , 1 U/mL SMase and 0 . 1 µg/mL or LPS+sCD14 ( 10+1µg/ml ) , for 90 min . Whole cell lysates in a buffer containing 50 mM Tris HCl , pH 7 . 5 , 400 mM NaCl , 1mM EDTA , 1% Nonidet P-40 were separated by electrophoresis on a 7 . 5% native Tris-glycine gel [65] . Membranes were incubated with primary antibodies against human IRF3 ( Fl-425 , Santa-Cruz ) diluted 1∶1000 and anti-rabbit IgG-HRP ( 1∶1000 ) . IRF3 monomers and dimers were detected with the ECL Plus Western Blotting Detection System . In brief , A498 cells ( n = 350000 ) were seeded in 6-well plates and infected with CFT073 ( 108 CFU/ml ) , total RNA was extracted ( Trizol , Invitrogen , USA ) and cleaned by a Qiagen RNeasy . RNA was reverse-transcribed to biotin-labeled cRNA using a TargetAmp Nano-g Biotin-aRNA Labeling kit ( Epicentre Biotechnologies , Madison , USA ) . Labeled cRNAs were hybridized onto an Illumina HumanHT-12 Expression Beadchip for 16 hours at 58°C . The arrays were then washed and stained ( Illumina Wash Protocol ) and scanned using a BeadArray Scanner 500GX . The background-subtracted data were pre-processed to correct negative and non-significant intensities . Pre-processed data was normalized using the cross-correlation [66] and genes with a log fold change of 2 were identified as differentially expressed . Data was preprocessed using RMA implemented in the free software packages R and Bioconductor ( http://www . r-project . org ) . For more details , see Yadav et al . Differentially expressed genes were categorized using the Functional Annotation Clustering Tool in the Database for Annotation , Visualization and Integrated Discovery ( DAVID ) [67] and the EASE score ( a modified enrichment score derived from Fisher exact P-value ) was used to judge the enrichment . To further study signaling pathways altered by CFT073 , the differentially expressed genes were submitted for Ingenuity Pathway Analysis ( Ingenuity Systems , Redwood City , CA ) . Escherichia coli S1918 [68] ( KanR ) lacks genes encoding known adhesins and was used as a recipient strain for recombinant plasmid pPIL 110-75 ( AmpR ) carrying the papAD1100 gene cluster ( pap+ ) [69] or PKL-4 carrying the entire fim gene cluster from E . coli PC31 ( type I+ ) [68] . The papGX deletion mutant of E . coli strain 83972 ( ABU 83972ΔpapGX ) was generated using the lambda red homologous recombination technique [70] . Clones with the reconstituted pap determinant were screened using PCR and verified by DNA sequencing . Primers used for reconstitution of a functional pap gene cluster in E . coli 83972 are shown in Supplemental Table S5 in Supporting Information S1 . The growth rates of the reconstituted mutant strain and the 83972 wild type strain were shown to be identical . In addition , the ability of the reconstituted mutant to agglutinate sheep blood erythrocytes was also confirmed by agglutination assays . Mice were bred at the MIG animal facilities , Lund , Sweden . Female C57BL/6 wild type or Irf3−/− ( from T . Taniguichi ) and Ifnβ−/− ( from F . Ivars , Lund University ) mice were used at 9–15 weeks . After anesthesia ( Isofluorane ) , mice were infected by intravesical inoculation with E . coli CFT073 ( 109 CFU in 0 . 1 mL ) through a soft polyethylene catheter ( outer diameter 0 . 61 mm; Clay Adams , Parsippany , NJ , USA ) . Animals were sacrificed while under anesthesia , kidneys and bladders were removed and prepared for hematoxylin-eosin staining or immunohistochemistry . Viable counts in homogenized tissues ( Stomacher 80 , Seward Medical , UAC House , London , UK ) were determined after overnight growth on tryptic soy agar plates at 37°C . Urine samples collected prior to and daily after infection were cultured and recruited neutrophils were quantified in uncentrifuged urine by use of a hemocytometer . Kidney sections were examined by immunohistochemistry [71] . Tissue sections were dried and permeabilized in 0 . 2% Triton X-100 , 5% goat normal serum ( DAKO ) in PBS , incubated with NIMP-R14 rat anti-mouse neutrophil specific antibodies ( 1∶200 ) and a polyclonal rabbit antiserum to the Pap G adhesin ( 1∶200 ) to detect P-fimbriated E . coli and to Alexa 488 anti-rat IgG and Alexa 568 anti-rabbit IgG secondary antibodies and nuclei were counterstained with DAPI ( 0 . 05 µM ) . After mounting , coverslipped slides were examined by fluorescence microscopy ( AX60 , Olympus Optical , Hamburg , Germany ) at the Department of Pathology , Lund University , Sweden . Confocal fluorescence immunocytochemistry was performed on cells grown to 70–80% confluence on 8-well chamber slides . After stimulation , cells were fixed and permeabilized with 0 . 25% Triton X-100 , 5% FBS in PBS and incubated with primary antibodies diluted 1∶50 in 5% FBS in PBS overnight at 4°C . Alexa fluor-labeled secondary antibodies were applied for 1 hour at RT in the dark . In order to control specific staining of neutrophils and bacteria , slides were stained with only secondary antibodies ( Figure S1 ) . Slides were covered with mounting medium ( M1289 , Sigma ) and cover glasses and the cells were examined with a LSM510 META confocal microscope . The IRF3 promoter from patients with UTI or healthy controls was sequenced ( PSQ96 , Biotage , Uppsala , Sweden ) and examined for −925 and −776 polymorphisms [61] . Genomic DNA was extracted from heparinized peripheral blood using the QIAamp DNA Blood midi kit . More detailed descriptions of inclusion critera and diagnosis are provided in [58] , [59] , [72] . The IRF3 promoter SNPs ( −925; −776 ) were genotyped using Pyrosequencer PSQ 96 after PCR amplification of chromosomal DNA and a second biotinylated PCR for each SNP ( for primers see Supplemental Table S2 in Supporting Information S1 ) . The promoter sequences from extracted chromosomal DNA derived from APN and ABU patients were PCR-amplified using the Infusion primers 5′ IRF3 NheI and 3′IRF3 NcoI ( Supplemental Table S3 in Supporting Information S1 ) and Phusion hot start polymerase according to the manufacturer ( Finnzymes Oy , Finland ) . Amplicons were introduced by recombination , using the Infusion cloning technique ( Clontech ) , into a NheI- and NcoI-cleaved and gel-purified luciferase reporter vector , pGL3 basic ( Promega ) . The recombinant DNA was transformed into E . coli and recombinant clones were screened for the presence of cloned promoter insert . Plasmids of the correct size were further analyzed by DNA sequencing using the Big Dye terminator v3 . 1 cycle sequencing chemistry and ABI capillary sequence . A quick change Multi Site-directed Mutagenesis kit ( Stratagene ) was used according to the manufacturer's instructions to create the various IRF3 promoter constructs ( Supplemental Table S3 in Supporting Information S1 ) . 5498 human kidney epithelial cells were cultured in 24-well plates at a density of 2 . 5×105 cells per well . The cells were transiently transfected with wild type or mutant IRF3 promoter driven firefly luciferase constructs ( pGL3 ) together with a constitutively expressed internal control construct with Renilla luciferase-thymidine kinase promoter ( pRL-TK , Promega ) using Fugene HD ( Roche ) Transfection reagent at 4∶2 ratio . Luciferases were measured using the Dual Luciferase Reporter System Assay ( Promega ) with a Glomax Integrated Luminometer ( Promega ) . Firefly luciferase data were normalized against transfection efficiency of Renilla luciferase and expressed as a ratio . Student's t test or Wilcoxon's rank-sum test were used for paired comparisons , Mann-Whitney test was applied for unpaired comparisons . P values below 0 . 05 were considered to indicate statistical significance . Deviations from Hardy-Weinberg equilibrium ( HWE ) for genotypes at individual loci in patients and controls , as well as differences in genotype and allele distributions between groups , were assessed using the χ2 test . Fisher's exact test was used where appropriate . Gene ID number for human TLR4 is 7099 , human MyD88 is 4615 , human TRIF is 148022 , human TRAM is 353376 , human TBK1 is 29110 , human CREB is 1385 , human IRF3 is 3661 , mouse Irf3 is 54131 , human IFNB is 3456 and mouse Ifnb is 15977 .
|
The host immune system must identify pathogens and defeat them through TLR-dependent signaling pathway activation , while distinguishing them from commensal flora . Contrary to current dogma , the host cannot solely use “pattern recognition” since the microbial molecules involved in such recognition are present on pathogens and commensals alike . We identify here a pathogen-specific mechanism of TLR4 activation and signaling intermediates in this pathway , leading to IRF3-dependent transcription of innate immune response genes . We show in knockout mice that Irf3 deficiency causes severe tissue pathology and that effector functions controlled by IFNβ are involved . Finally , in highly disease-prone pyelonephritis patients we found a high frequency of IRF3 promoter polymorphism compared to asymptomatic bacterial carriers or controls . The polymorphisms influenced promoter activity in reporter assays , suggesting that they are functionally important . Urinary tract infections are among the most common bacterial infections in man , and are a major cause of morbidity and mortality . A subset of disease-prone individuals is at risk for recurrent disease , severe renal dysfunction and end-stage renal disease . At present , there is no method to identify disease-prone infants and to prevent future morbidity and renal damage . The genetic and functional studies described here indicate that genetic variation in IRF3 influences individual susceptibility to kidney infection and might serve as a new tool for future risk assessment in this patient group .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"infectious",
"diseases/urological",
"infections",
"immunology/immune",
"response",
"microbiology/innate",
"immunity"
] |
2010
|
Pathogen Specific, IRF3-Dependent Signaling and Innate Resistance to Human Kidney Infection
|
The origin and evolution of new microRNAs ( miRNAs ) is important because they can impact the transcriptome broadly . As miRNAs can potentially emerge constantly and rapidly , their rates of birth and evolution have been extensively debated . However , most new miRNAs identified appear not to be biologically significant . After an extensive search , we identified 12 new miRNAs that emerged de novo in Drosophila melanogaster in the last 4 million years ( Myrs ) and have been evolving adaptively . Unexpectedly , even though they are adaptively evolving at birth , more than 94% of such new miRNAs disappear over time . They provide selective advantages , but only for a transient evolutionary period . After 30 Myrs , all surviving miRNAs make the transition from the adaptive phase of rapid evolution to the conservative phase of slow evolution , apparently becoming integrated into the transcriptional network . During this transition , the expression shifts from being tissue-specific , predominantly in testes and larval brain/gonads/imaginal discs , to a broader distribution in many other tissues . Interestingly , a measurable fraction ( 20–30% ) of these conservatively evolving miRNAs experience “evolutionary rejuvenation” and begin to evolve rapidly again . These rejuvenated miRNAs then start another cycle of adaptive – conservative evolution . In conclusion , the selective advantages driving evolution of miRNAs are themselves evolving , and sometimes changing direction , which highlights the regulatory roles of miRNAs .
MicroRNAs ( miRNAs ) are a class of small , endogenous RNAs that regulate gene expression post-transcriptionally [1] , [2] . Each miRNA gene is first transcribed as a stem-loop ( hairpin ) RNA structure , 70–90 nt in length in animals , and then processed in several steps into the ∼22-nt mature product , referred to as miR [3] . In animals , miR binds to the 3′ untranslated region ( UTR ) of target mRNAs through perfect base-pairing of the seed region ( position 2–8 of a miR ) , inducing translation repression or mRNA degradation [4] . As the seed is only 7 nt long , each miRNA may potentially regulate hundreds of transcripts while each transcript may in turn be regulated by more than one miRNA [5] . The emergence of new miRNAs is of special interest in evolutionary biology for two reasons . First , they buffer gene expression noises and thus have been hypothesized to be a key player in canalization [6] , [7] . As proposed by C . H . Waddington [8] , [9] , canalization contributes to developmental stability and , in a recent interpretation , it may also contribute to evolvability via hidden genetic variations [10] , [11] . Second , due to their small size , miR-producing hairpins can form readily and de novo emergence of miRNAs from non-miRNA transcripts is a frequent phenomenon [12] , [13] . There are hundreds of thousands of potential miRNA structures in each Drosophila genome [12] and millions in a mammalian genome [14] . Given such a propensity for new miRNAs to emerge , the birth , death and adaptation of new miRNAs are a significant part of understanding the evolution of transcriptional regulation [12] . In contrast , protein-coding genes require long open reading frames to yield functional peptides . Hence , local duplication or retrotransposition [15] , rather than de novo origination , is the common mode for the formation of coding genes . In Drosophila , the birth and death rates of miRNAs have been estimated to be about 12 and 11 . 7 genes per Myr , respectively , with a net gain of about 0 . 3 per Myr [12] . It is generally agreed that the net gain is low , ranging between 0 . 3 and 1 new gene per Myr [16] , [17] . Despite this , the total repertoire of miRNAs should still be increasing dramatically over long periods . While the net gain ( birth – death ) is not in dispute , there is disagreement over the estimated birth and death rates of new miRNAs [12] , [16] , [17] . Because numerous putative miRNAs are found in the transcriptome , these lowly expressed , evolutionarily neutral , and short-lived miRNAs account for the bulk of the estimated births and deaths . The debate is about which ones should be counted as new miRNAs . To resolve the issue , we propose to define new miRNAs in an evolutionary context by a set of stringent criteria , requiring a signature of initial adaptive evolution soon after their birth . Numerous small RNAs that emerge and vanish with the dynamics of neutral sequences are excluded from the evolutionary analysis . Given this definition , only a small fraction of miRNA-like sequences in any species would qualify as new miRNAs . We collected extensive small RNA-seq data available for four Drosophila species ( D . melanogaster , D . simulans , D . pseudoobscura and D . virilis ) [12] , [16] , [18]–[23] and three mosquitoes ( Aedes albopictus , Aedes aegypti and Culex quinquefasiatus ) [24] , [25] . We further generated small RNA-seq data for sex organs and imaginal discs in D . simulans and D . pseudoobscura . The extensive dataset permits systematic identification of new miRNAs and in-depth analyses of their long-term fates . Our first objective is to understand the origin and early evolution of new miRNAs in the species D . melanogaster . The second objective is to track the long-term evolutionary trajectory of new miRNAs , which may be in any of the following four modes after their initial adaptive evolution:
Starting with the youngest genes , we first analyzed the 22 new miRNA genes that emerged in the last 30 Myrs , since D . melanogaster diverged from D . pseudoobscura ( Fig . 1 ) . Among them , 21 originated de novo; only miR-983-2 in D . melanogaster ( dme-miR-983-2 ) was duplicated from another miRNA ( dme-miR-983-1 ) . More than half of the 22 new miRNAs are found in clusters – five in the miR-972 cluster ( abridged as miR-972s ) , two in miR-310s and five in miR-982s . Members in a cluster have significantly higher expression levels than the orphan miRNAs ( Mann-Whitney U test , p<0 . 05 ) . The miR-982 cluster consists only of members emerging in the last 30 Myrs , whereas both miR-310s and miR-972s are mixtures of old and new miRNAs ( Table S3 ) . Thus , the former is most informative about the birth and early evolution of new miRNAs . The miR-982s is X-linked , comprising five distinct miRNA families: miR-982 , -2582 , -303 , -983 and -984 . With the exception of the recently duplicated dme-miR-983-1/-2 , miRNAs in this cluster do not share a seed sequence ( Fig . 2A &B ) . Against the 12 Drosophila species [27] , copies of this cluster can be found in D . simulans , D . sechellia , D . yakuba and D . erecta but are absent in all other more distantly related species . The expression of miR-982s members was confirmed by RT-PCR ( Fig . S1 ) . The evolution of this cluster in the D . melanogaster subgroup is depicted in detail in Fig . 2A . As shown in Fig . 2A , each member of miR-982s appears to emerge in situ from local non-miRNA sequences . Due to their small sizes , unstructured genomic sequences evolving into miRNA-like transcripts have often been suggested [28] but have not been convincingly proven . The cluster of miR-982/2582/303/983/984 appears to be a good example of de novo origin ( see below ) with point mutations improving miRNA processing step by step ( Fig . 2B and Fig . S2 and S3 ) . For example , the secondary structure of miR-982 in D . erecta can only form a poor hairpin ( −18 . 20 kcal/mol ) . Many nucleotide substitutions , accumulated subsequently in the stem and loop regions , have greatly improved the thermodynamic stability of the hairpin in D . melanogaster ( −24 . 00 kcal/mol ) and in the three paralogs of D . simulans ( −21 . 52 to −27 . 50 kcal/mol; Fig . 3A and Fig . S2 and S3 ) . After each miRNA emerges from the unstructured sequence , gene duplication appears common [29] , [30] . miR-2582 and miR-982 were expanded by whole-gene ( Fig . 2A , Duplication 1 , 2 and 3 ) or segment duplication ( Duplication 4 ) in D . melanogaster and D . simulans , followed by gene conversion in D . sechellia ( Fig . 2A ) . Moreover , miR-983 was duplicated in D . melanogaster ( Duplication 5 ) . In this species alone , miR-984 emerged de novo next to miR-983 ( Fig . 2A ) . These duplicates soon accumulated many nucleotide substitutions ( Fig . 2B ) . Meanwhile , seed shifting and arm switching occurred in the miR-982/2582/303/983 families ( Fig . 2B ) . These modifications presumably lead to new targets , resulting in neo-functionalization after gene duplication [28] . After new miRNAs emerged de novo , the question is whether the subsequent evolution is driven by natural selection . A greater level of divergence in miRNA genes than in flanking regions might suggest positive selection ( Ref . [31]; Fig . S4A ) . A proper analysis would require the comparison of between-species divergence ( D ) and within-species polymorphism ( P ) using a modified McDonald-Kreitman ( MK ) test [32] . In this study , we generated DNA sequences from 42 D . melanogaster ( ∼7 . 5 kb from each line ) and 25 D . simulans lines ( ∼8 . 1 kb ) ( Table S4 ) . The D/P ratios for each precursor miRNA from miR-982s , as well as the 1 kb upstream flanking regions , were compared [33] . As shown in Table 2 , all the miRNA genes from the miR-982 , miR-303 and miR-983 families have a significantly higher D/P ratio than the flanking regions in both D . melanogaster and D . simulans ( p<0 . 05 ) , suggesting positive selection . Members of the miR-2582 family show significantly higher D/P ratios in D . melanogaster , but not in D . simulans ( Table 2 , also see next section ) . Because each individual miRNA gene , being small , would yield a significant result in the MK test only when the selection is extremely strong , we also performed the test on new miRNAs collectively , relative to the genome-wide 4-fold degenerate sites ( from Drosophila Population Genomics Project ( DPGP ) ; see Materials and Methods ) . Table 3 shows that the new miRNAs emerging in the last 30 Myrs have a higher D/P ratio than in the genome-wide 4-fold degenerate sites . In fact , more than 79% of the observed divergence in the precursors and more than 89% in the mature regions is estimated to have been fixed adaptively ( see Materials and Methods and Table 3 ) . A higher D/P ratio could also be attributed to an increase in selective constraint , rather than positive selection [34] . However , we excluded such possibility in Text S1 . Due to the large number of adaptive sites , every new miRNA is likely to carry one or more of them . As expected , signatures of positive selection are much weaker for the lowly expressed miRNAs and mirtrons ( Table S5 ) . Other lines of evidence for recent adaptive evolution include the pattern of polymorphism within species and the differentiation between populations . The miR-982 cluster was examined further by the sliding window analysis of Fay and Wu's H ( θH ) , an estimator of nucleotide diversity sensitive to positive selection [35] , [36] . The profile of θH peaks near miR-983/984 and miR-303 in both species , a common footprint of hitchhiking with positive selection [35] . The signature is stronger in D . simulans for miR-982 than in D . melanogaster ( Fig . S4B and S4C ) . In addition , we analyzed the M and Z populations of D . melanogaster [37]–[39] using the Fst statistic [40] . For dme-miR-984 and dme-miR-303 , the precursor sequences are strongly differentiated between M and Z lines ( Fst = 0 . 318 for dme-miR-984 and Fst = 0 . 252 for dme-miR-303 ) compared to all SNPs within the miR-982s region ( Mann-Whitney U test , p = 0 . 057 for dme-miR-984 and p = 0 . 068 for dme-miR-303 , Table S6 ) or the 238 D . melanogaster miRNAs ( Mann-Whitney U test , p = 0 . 046 for dme-miR-984 and p = 0 . 008 for dme-miR-303; data were obtained from DPGP2 [41] , see Materials and Methods ) . The analyses collectively suggest that the rapid evolution of new miRNAs is driven by natural selection . After the initial adaptive evolution , one might reasonably expect these new adaptive miRNAs to be integrated into the transcriptional network and begin evolving at a slower rate . Surprisingly , the most likely fate of these new miRNAs was death , rather than integration . This can be seen in the number of observable new miRNAs from two different time periods – 22 surviving miRNAs from the last 30 Myrs but only 9 from the preceding 30 Myrs ( 30–60 Myrs before present ) . By assuming a constant birth rate , we can estimate the number of newborn miRNAs in each time interval , which can then be compared with the surviving miRNAs from that time period . Using the estimated rate of 3 newborn miRNAs per Myr ( 12 in the last 4 Myrs ) , Figure 4A shows that 87% of new miRNAs disappeared in 4–30 Myrs ( 68 out of 78 ) . The proportion of death in older miRNAs increased only marginally , to 90% , for the period of 30–60 Myrs ( 81 out of 90 ) . Therefore , most miRNAs seem to die quickly at an early stage of evolution , soon after the initial adaptive evolution . Only 6 . 0% of new miRNAs ( 34 out of 570 ) survived after 60 Myrs . It is unexpected that new adaptive miRNAs favored by natural selection should suffer such quick and massive death , albeit at a somewhat lower rate than neutrally evolving new miRNAs [12] . The former has a survival rate of 6 . 0% while the latter has a lower rate , at 2 . 5% [12] . Apparently , the initial adaptation is evolutionarily transient and the continual adaptation toward integration is not a common fate . We should note that alternative explanations have been considered . A most obvious one concerns the possibility of a bust of adaptive new miRNAs in D . melanogaster since its split from D . simulans . These explanations are compared in Discussion . Interestingly , miRNA death may sometimes be an adaptive process . The miR-2582-like gene in D . melanogaster is shown to be evolving adaptively in Table 2 , but its evolution is toward degeneracy . Three lineage-specific mutations that disrupt the duplex structure are shown in Fig . 3B , probably associated with the degeneration of dme-miR-2582 . Presumably , conditions changed causing the adaptive function initially performed by the new miRNA to become deleterious at a later time . Upon survival , new miRNAs eventually became integrated into the transcriptional network and evolved conservatively . There is a transitional phase after the adaptive phase , but before either integration or death . During the transition , these miRNAs often appeared to have a neutral evolutionary rate . Figure 4A shows that all the surviving miRNAs began to evolve either neutrally or conservatively ( three transitional and six conservative miRNAs , respectively ) within 30–60 Myrs ( See Materials and Methods ) . The miR-2582 gene in D . simulans appears to be in such a transition ( Table 2 ) . It is interesting that miR-2582 orthologs in sibling species may be at different stages of evolution . Over long periods of time , new miRNAs will have died or have been integrated into the transcriptional network and are now conservatively evolving . miRNAs born 60–250 Myrs ago have largely vanished ( 94% have disappeared , see Fig . 4A ) . However , some of the cohort of the 34 surviving miRNAs are not behaving as expected . In fact , only 26 of them are evolving conservatively . Nearly a quarter of them ( 8 out of 34 ) are evolving either neutrally or adaptively ( Fig . 4A ) and most of these ( 7 out of 8 ) come from miR-972s or miR-310s ( Table 4 ) . At this rate of evolution , none of them should have been recognizable as homologs between D . melanogaster and D . virilis . We suggest that the 8 unusual miRNAs may have been conservatively evolving for most of their evolutionary history . Four of them have been adaptively evolving once again and the remaining four appear to be in transition , away from the previous selective constraints . If the hypothesis is correct , we expect to see stronger evolutionary conservation in more distant comparisons than in recent ones . We use KmiR/KS , where KmiR denotes the divergence in the precursor region of the miRNA , to measure conservation . Table 4 shows their KmiR/KS values for the last 4 Myrs and for the distant past ( 60 Myrs after the split between D . melanogaster and D . virilis ) . The evolutionary conservation has indeed been relaxed substantially in the last 4 Myrs with the average value increasing from 0 . 337 to 0 . 825 , a 2 . 5-fold difference . Such fold-changes of KmiR/KS were significantly high in the eight miRNAs , compared with the whole repertoire of 238 miRNAs ( Mann-Whitney U test , p = 0 . 00014 ) . The rate increase appears to be true in both D . melanogaster and D . simulans lineages when the homologous sequences from D . yakuba and D . erecta were used as outgroups to calculate the rate in each lineage separately . Among the eight genes , two and six are evolving slightly faster in D . melanogaster and D . simulans , respectively ( see Table S7 ) . It is interesting that some old miRNAs go through the reverse transition ( or rejuvenation ) from conservative to adaptive evolution , the latter being the hallmark of young miRNAs . Rejuvenation can also lead to the death of old miRNAs . The miR-972s may be such an example . Some members of this cluster emerged 60–250 Myrs ago and should have been integrated into the ancestral genome by the time D . pseudoobscura split from D . melanogaster . However , the entire miR-972s region was lost in D . pseudoobscura since the split . Taken together , new miRNAs ( such as miR-310s and miR-972s ) may go through cycles of adaptation , integration ( if escaping death ) and rejuvenation , which would start another cycle of adaptation and integration ( Fig . 4B ) . To study the evolution of new miRNAs sequences , we characterized their expression patterns . We did so by using the global small RNA profiling datasets ( see Table S1 and Materials and Methods ) . Figure 5 shows young miRNAs ( <30 Myrs ) are lowly expressed in specific tissues , generally in the testes and larval brain/gonads/imaginal discs . Middle-aged miRNAs ( 30–60 Myrs ) broadened their expressions to include ovaries and embryos . The older miRNAs ( 60–250 Myrs ) showed moderate and even broader expressions , which then evolved to become highly abundant in all tissues and developmental stages as seen in the oldest miRNAs ( >250 Myrs ) . The simplest explanation is that new miRNAs increase the expression level and expand the breadth as they get older . The change in expression parallels that in sequence evolution ( Fig . 4A and Table S8 ) . There are other explanations that may also account for the different expression patterns between new and old miRNAs ( see Text S2 ) . Detailed descriptions of the evolution in expression patterns are given in Text S3 .
During Metazoan evolution , the miRNA repertoire expanded dramatically from a few genes to several hundreds [28] , [42] . By limiting the analysis to new miRNAs that evolve adaptively soon after their birth , we avoided the large number of lowly expressed miRNA-like sequences . These sequences may or may not be considered miRNAs and are generally thought to be evolutionarily ephemeral and adaptively insignificant [43] , [44] . The inclusion of only new miRNAs that evolve adaptively at emergence reveals an unexpected pattern of an excess of such miRNAs in the last 4 million years of the D . melanogaster lineage . The possible explanations are therefore either a burst of birth since D . melanogaster split from D . simulans , or a decline in the survivorship of adaptive new miRNAs as they age . We consider the latter explanation as more plausible for several reasons . First , the birth rate of miRNA-like sequences indeed appears constant because different Drosophila species have comparable numbers of such new transcripts [16] . Given the ease in forming precursor-like hairpins , the constant rate is hardly surprising . Second , as a result , the birth rate of adaptive new miRNAs may not deviate much from a constant value either . Indeed , the burst of adaptive new miRNAs is observable in D . simulans as well as the common ancestor of D . melanogaster and D . simulans , as is evident in the miR-982 cluster ( Fig . 2A ) . Third , the proportion of adaptive miRNAs born in the period of 4–30 Myrs is also higher than that in the 30–60 Myrs period . Overall , an excess of new adaptive miRNAs appears to be a decreasing function of time , rather than of particular lineages; hence , their death over time is a simpler explanation . Because only a small number of new adaptive miRNAs remain active after cycles of evolution through phases of adaptation and degeneration , the repertoire of miRNAs in the D . melanogaster genome has been nearly static in 40 Myrs of evolution , with only 0 . 18 miRNA integrations per Myrs . We should note that this low rate may still be an over-estimate because not all death has been accounted for . This ( near ) steady state echoes the view of a correlation between morphological complexity and the size of miRNA repertoire [45] , as the Drosophila genus has been relatively invariant in form since its diversification . Despite the low integration rate , many new miRNAs continue to emerge and some briefly evolve adaptively before their demise . This “transient utility” is puzzling as gene functions are lost usually through environmental changes ( such as vision genes in caves [46] ) or redundancies [47] . A possible explanation may be the suggested role of miRNAs in evolutionary canalization [7] . In such a role , the regulators and their targets need not be stringently wired as long as the system remains properly buffered . By this scheme , new miRNAs may emerge to fill in the transiently vacated role created by the shifting interactions between established miRNAs and their targets [7] . They disappear when the role is no longer needed . A small number of new miRNAs that become integrated into the transcriptional network begin this process in the testis , in parallel with new protein coding genes [48]–[54] . Since sexual selection driving male reproduction is a very potent force of evolution , this expression pattern may not be all that surprising [50] , [55]–[58] . In the example of miR-982s , the predicted targets are indeed enriched in genes of male courtship behavior and other male sexual traits ( Table S9 ) . Once a new miRNA is established , its expression is often broadened to other tissues . Testis may be the beachhead that permits the new miRNA to gradually modulate its expression and interactions with potential targets . In addition , new miRNAs with distinct seeds often emerge in clusters , which presumably facilitate their co-expression [29] , [30] , [59] . Unlike protein coding genes , miRNAs can easily emerge de novo , thanks to their small size , but can often be derived from existing genes as well [60] . The simple structure of miRNAs may permit general inferences on features and dynamics of genic evolution . A previous example is the rate of evolution as a correlate of expression level [61] . It would be interesting to see if the inferred cycles of evolution experienced by new miRNAs are a general process .
Total RNA was extracted from D . simulans ( NC48S ) and from D . pseudoobscura using TRIzol ( Ambion ) . Ovaries and testes from 3 to 5-day adults were dissected and collected for both NC48S and D . pseudoobscura . Imaginal discs including central nerve system ( CNS ) were dissected from wandering third-instar larva of D . pseudoobscura . Small RNA libraries were generated from each RNA sample using Illumina Small RNA Sample Preparation kit , and sequenced with the Illumina HiSeq 2000 at the Beijing Genomics Institute ( Shenzhen ) . The data were deposit at Gene Expression Omnibus ( GEO ) database ( http://www . ncbi . nlm . nih . gov/geo/ ) under the accession numbers GSM1165052-GSM1165056 . The publicly available small RNA sequencing reads from four Drosophila species ( D . melanogaster , D . simulans , D . pseudoobscura and D . virilis ) were downloaded from GEO database ( http://www . ncbi . nlm . nih . gov/geo/ , Table S1 ) . The miRNA sequences of three Culicinae species ( Aedes albopictus , A . aegypti and Culex quinquefasiatus ) were adopted from two previous small RNA sequencing studies [24] , [25] . Drosophila genome sequences were retrieved from UCSC ( http://genome . ucsc . edu ) ; the Whole Genome Alignment ( WGA ) and CDS alignment were obtained from 12 Drosophila Assembly/Alignment/Annotation ( http://rana . lbl . gov/drosophila ) . The genome versions used were: D . melanogaster , dm3; D . simulans , droSim1; D . sechellia , droSec1; D . yakuba , droYak2; D . erecta , droEre2; D . ananassae , droAna3; D . pseudoobscura , dp4; D . persimilis , droPer1; D . willistoni , droWil1; D . mojavensis , droMoj3; D . virilis , droVir3; D . grimshawi , droGri2 . The genome coordinates and sequences of miRNA genes were retrieved from miRBase Release 19 ( http://www . mirbase . org ) . The genome coordinates and sequences of intron , rRNA , tRNA , snRNA and transposon elements were obtained from FlyBase ( r5 . 41 , http://flybase . org , ) We defined canonical miRNAs and mirtrons according to Ruby et al . [62] . Mirtrons were defined as pre-miRNAs with both 5′ and 3′ ends matching the splicing sites of host introns . The rest of the miRNAs were then classified as canonical miRNAs . When more than three miRNAs were located within a 20 kb region , these miRNAs were considered as a cluster . Small RNA reads ( 18–30 nt ) were extracted from sequencing data . Firstly , we excluded reads mapped to transposon elements and structural RNAs ( rRNA , tRNA and snRNA ) using bowtie [63] , allowing no mismatch . Next , we annotated novel miRNAs by miRDeep2 [64] with default parameters . Finally , miRNAs with no read matching miR* were removed following previous practice [23] . We combined novel miRNAs sequences and known miRNA sequences for expression analysis . For each species , small RNA reads ( 18–30 nt ) were mapped to miRNA precursor sequences using bowtie [63] , allowing no mismatch . Each read count was divided by the number of matches to miRNA precursors . The miRNA expression was normalized by total miRNA counts and scaled to reads per million ( RPM ) , as previous described [18] . We examined phylogenetic distributions of the D . melanogaster miRNAs in three other Drosophila species ( D . simulans , D . pseudoobscura and D . virilis ) and three Culicinae species ( Aedes albopictus , A . aegypti and Culex quinquefasiatus ) , where small RNAs have been profiled via deep sequencing [12] , [16] , [22] , [24] , [25] . Based on the comprehensive dataset , miRNA homologs were determined by homology search using either the whole genome alignment ( WGA ) within the Drosophila group or BLAST ( threshold E<10−5 ) between Drosophila species and mosquitoes , and cross-checked with small RNA reads in the species in query ( at least one read matching mature and miR* ) . The homologous sequences of the D . melanogaster miRNA precursors in D . simulans ( droSim1 ) , D . pseudoobscura ( dp4 ) and D . virilis ( droVir3 ) were extracted from UCSC pairwise WGAs using LiftOver ( http://hgdownload . cse . ucsc . edu/ , minMatch = 0 . 6 ) . The precursors failing to obtain hits in the genomes were subjected to BLASTN search against NCBI trace archives ( http://www . ncbi . nlm . nih . gov/Traces/home/ ) . Matched sequences with E-values <10−5 were also considered as miRNA homologs and recovered for the analysis below . The WGA output was compared with miRNA annotation by miRDeep2 [64]; miRNA orthologs confirmed by miRDeep2 were retained . The miRNA precursor sequences in Aedes albopictus , Culex quinquefasiatus and A . aegypti were adopted from the studies of Li et al . [24] and Skalsky et al . [25] . These sequences were combined and subjected to BLASTN search against miRNA precursors in D . melanogaster . The best reciprocal hits with E-values <10−5 were retained as the corresponding miRNA homologs in the Culicinae lineage . According to the phylogenetic distribution , maximum parsimony method was used to infer the origination of each miRNA along the main trunk of the phylogenetic tree of D . melanogaster , D . simulans , D . pseudoobscura , D . virilis and Culicinae . An miRNA is assumed to emerge in the most recent common ancestor of all the species bearing the authentic homologs . The branch lengths of the phylogenetic tree ( in Myrs ) were adopted from previous estimations [27] , [65] , [66] . The 238 miRNAs were classified into five age groups , corresponding to the time intervals of 0–4 Myrs , 4–30 Myrs , 30–60 Myrs , 60–250 Myrs and >250 Myrs . The genomic coordinates and precursor sequences of dme-miR-982/303/983-1/983-2/984 and dsi-miR-982c/2582b/982b/2582a/982a/303/983 were retrieved from miRBase ( Release v19 ) . Based on the WGA of 12 Drosophila genomes [27] , genomic sequence of the whole miR-982s cluster ( ∼9 Kb ) in D . melanogaster ( dm3 ) was extracted and used as a query to search against the other 11 Drosophila genomes using BLAT [67] with an E-value threshold of 0 . 001 . We only detected hits in D . simulans , D . sechellia , D . yakuba and D . erecta , indicating that miR-982s is specific to the melanogaster subgroup . Homologous sequences of the miR-982s cluster from the five species were aligned using MUSCLE [68] . Homologs of miR-982s members in each species were identified using BLAST with the query of known precursor sequences ( miRBase Release v19 ) and an E-value threshold of 0 . 001 . The hits were further inspected in the alignment of the whole miR-982s cluster . The phylogenetic tree of each family of miR-982 , miR-2582 , miR-303 , and miR-983 was reconstructed using the maximum likelihood method as implemented in MEGA 5 . 0 [69] . To validate the existence of miR-982s members in D . yakuba and D . erecta , we first predicted the secondary structure and thermo-stability of each miRNA homolog using RNAfold ( http://rna . tbi . univie . ac . at/ ) with the default parameters [70] . A good hairpin with minimum free energy ( MFE ) >15 kcal/mol was considered as a potential miRNA candidate . There were four such candidates: dya-miR-2582-anc , dya-miR-303-anc , der-miR-982-anc , and der-miR-983-anc , where “anc” indicates ancestor . Then , we validated the expression of each candidate by amplifying the potential miRNA precursor from cDNA because the mature miRNA is hard to define . Total RNAs were extracted from testes of D . yakuba and D . erecta using TRIzol ( Ambion ) and treated with TURBO DNase Kit ( Ambion ) . 0 . 5 ug RNA was reverse transcribed ( RT ) in a 20 ul reaction volume using PrimeScript II 1st Strand cDNA Synthesis Kit ( TaKaRa ) . 1 ul RT products were used for PCR with Ex Taq DNA Polymerase ( TaKaRa ) . PCR primers used are listed in Table S10 . A total of 25 D . simulans lines and 42 D . melanogaster lines , including 29 M lines and 13 Z lines , were used for population sequencing of the miR-982s cluster . The fly strains used were listed in Table S4 . The genomic sequences of D . simulans ( droSim1 ) and D . melanogaster ( dm3 ) were used to design primer pairs that amplify a ∼8 Kb region spanning the whole miR-982s cluster and ∼1 . 5 Kb each of the upstream and downstream flanking regions . The PCR product of each primer set was designed to be about 2 Kb in length and overlapped with each other by at least 300 bp . The primers used are listed in Table S10 and their genomic coordinates are displayed in Fig . S5 . PCR was carried out using LA Taq DNA Polymerase ( TaKaRa ) . PCR products were subject to direct sequencing or clone sequencing on an ABI 3730xl DNA Analyzer ( Applied Biosystems ) . DNA sequences were assembled using SeqMan software ( DNASTAR Inc . , USA ) and aligned using MUSCLE [68] with manual inspection . Haplotypes were inferred with the PHASE program when heterozygous sites were present [71] . The sequences obtained in this study have been deposited in GenBank under the accession numbers JX648211-JX648278 . Using the population sequencing data , several methods were used to detect positive selection of miR-982s in D . melanogaster and D . simulans , respectively . First , MK tests were applied on each member of miR-982s based on the divergence between D . melanogaster and D . simulans consensus sequences and polymorphism within either species . Each miRNA precursor was tested against a 1 kb region about 1 . 5 kb upstream of the 5′ end of miR-982s . Second , sliding window analysis of divergence and polymorphism was applied to the whole miR-982s cluster and its flanking region . The divergence was calculated using Kimura's 2-parameter model [72] based on the genomic sequences of D . simulans ( droSim1 ) and D . melanogaster ( dm3 ) . The polymorphism within either species was estimated using the method described previously [35] , [36] , [73] , [74] . D . simulans ( droSim1 ) and D . melanogaster ( dm3 ) were used as the outgroup for each other reciprocally , in order to polarize the derived alleles . The window size is 100 bp and the step width is 25 bp . Finally , based on our miR-982s population data or DPGP2 data ( see below ) [41] , the pattern of population differentiation ( Fst ) between Z and M lines was estimated for each miRNA precursor using Weir's method [40] . We used the McDonald-Kreitman test ( MK test ) [32] framework to detect positive selection in miRNAs from each age group based on the polymorphisms within D . melanogaster and the divergence between D . melanogaster and D . simulans . Precursor or mature sequences of each miRNA group were combined and treated as the functional category , while the 4-fold degenerate sites in the whole genome were used as the neutral control . The divergence is calculated by counting the number of changed nucleotide sites between D . melanogaster ( dm3 ) and D . simulans ( droSim1 ) based on the UCSC whole genome alignment . Polymorphism data was retrieved from Drosophila Population Genomics Project ( DPGP , http://www . dpgp . org/ , release 1 . 0 ) . SNPs that were detected on more than thirty individuals and exhibited a derived allele frequency ( DAF ) >5% were used for the MK test . The proportion of adaptively fixed mutations ( α ) was estimated as previously described [75] . To estimate the evolutionary fate of each miRNA , we first screened for adaptive miRNAs among the 238 candidates by using each miRNA's precursor together with the 50 bp flanking sequences on both sides as the functional sites . The p-values of multiple MK tests were adjusted by the Benjamini-Hochberg method [76] and the adaptive significance of each candidate is re-validated by using the precursor alone in the MK test . We then identified the conservative miRNAs by comparing the number of substitutions in the miRNA precursors ( KmiR ) with the number of substitutions in the synonymous sites ( KS ) between D . melanogaster and D . simulans . miRNAs with KmiR/KS<0 . 5 were considered to be conservatively evolving . Kimura's 2-parameter model [72] and the Nei-Gojobori model [77] were used to calculate KmiR and KS , respectively . Finally , excluding the adaptive and conservative miRNAs , the remaining were considered to be in transition between adaptive to conservative/death . Data processing of small RNA deep sequencing libraries from different development stages and tissues of D . melanogaster [12] , [16] , [18]–[21] , [23] was conducted as described above . The read counts of each miRNAs were normalized to Reads Per Million ( RPM ) , which is the read number of each miRNA per million mapped reads in each library . The normalized counts were log2 transformed and subject to hierarchical clustering using R package heatmap2 . miR-982s targets were predicted by seed match using TargetScan ( v5 . 0 http://www . targetscan . org/fly_12/ ) [5] . Taking all the miRNA members together , 1 , 002 targets were obtained in D . melanogaster and 3 , 563 in D . simulans , of which 454 were shared by both species . We used DAVID to perform a Gene Ontology ( GO ) enrichment test for the predicted targets in the two species ( DAVID v6 . 7 , http://david . abcc . ncifcrf . gov/ ) [78] . Only the GO terms for biological processes were used for the enrichment test .
|
During Metazoan evolution , the architecture of the genome changed dramatically in size , gene number and regulatory elements . Genomic architecture is often assumed to be correlated with morphological complexity . However , it is still not known whether the gene repertoire , both for protein coding and non-coding genes , is continually increasing . In the last decade , a large family of small non-coding RNAs , or microRNAs ( miRNAs ) , has been shown to play an important role in diverse developmental processes . The genes controlled by miRNAs often evolve rapidly , potentially contributing to functional novelty , diversity and speciation . Here we estimated the birth and death rate of new adaptive miRNAs in Drosophila melanogaster . We found most new adaptive miRNAs disappear over long periods of time; hence , the miRNA repertoire stays close to that of a steady state . This steady state is commensurate with the morphological constancy of the genus of Drosophila .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"adaptation",
"sexual",
"selection",
"genome",
"evolution",
"natural",
"selection",
"population",
"genetics",
"comparative",
"genomics",
"biology",
"genomics",
"evolutionary",
"biology",
"genomic",
"evolution",
"evolutionary",
"processes",
"evolutionary",
"genetics"
] |
2014
|
New MicroRNAs in Drosophila—Birth, Death and Cycles of Adaptive Evolution
|
Wolbachia are maternally inherited endosymbiotic bacteria , widespread among arthropods thanks to host reproductive manipulations that increase their prevalence into host populations . The most commonly observed manipulation is cytoplasmic incompatibility ( CI ) . CI leads to embryonic death in crosses between i ) infected males and uninfected females and ii ) individuals infected with incompatible Wolbachia strains . CI can be conceptualized as a toxin-antidote system where a toxin deposited by Wolbachia in the sperm would induce embryonic death unless countered by an antidote produced by Wolbachia present in the eggs . In Drosophila melanogaster , transgenic expression of Wolbachia effector cidB revealed its function of CI-inducing toxin . Moreover in Culex pipiens , the diversity of cidB variants present in wPip strains accounts for the diversity in crossing-types . We conducted cytological analyses to determine the CI mechanisms that lead to embryonic death in C . pipiens , and assess whether diversity in crossing-types could be based on variations in these mechanisms . We revealed that paternal chromatin condensation and segregation defects during the first embryonic division are always responsible for embryonic death . The strongest observed defects lead to an exclusion of the paternal chromatin from the first zygotic division , resulting in haploid embryos unable to hatch . The proportion of unhatched haploid embryos , developing with only maternal chromatin , which reflects the frequency of strong defects can be considered as a proxy of CI intensity at the cellular level . We thus studied the putative effect of variations in crossing types and cidB diversification on CI defects intensity . Incompatible crosses involving distinct wPip strains revealed that CI defects intensity depends on the Wolbachia strains hosted by the males and is linked to the diversity of cidB genes harbored in their genomes . These results support that , additionally to its implication in C . pipiens crossing type variability , cidB diversification also influences the strength of CI embryonic defects .
Wolbachia are maternally-inherited endosymbionts , widespread among arthropods and filarial nematodes [1 , 2] , and the most frequent endocytobiotic bacteria detected in arthropods [3] . This high prevalence is attributed to their ability to manipulate their host reproduction to spread within arthropod populations [1] . The main reproductive manipulation strategy used by Wolbachia is named cytoplasmic incompatibility ( CI ) [4] . CI is a form of conditional sterility resulting in embryonic lethality [5] . In most of the host species , CI occurs when males infected with Wolbachia fertilize uninfected females whereas the reciprocal cross remains compatible . This difference in the production of viable offspring between infected and uninfected female reproduction enhances the spread of Wolbachia in host’s populations [6] . CI can also occur between males and females both infected with different and incompatible Wolbachia strains [7–12] . In such situations , incompatibility can be either unidirectional ( one cross direction is compatible while the reciprocal one is incompatible ) or bidirectional ( both cross directions are incompatible ) [7–9] . The penetrance of CI , i . e . the number of embryos affected by CI in a cross , varies depending on the Wolbachia strain and the host involved in the interaction . Indeed , in the same host Drosophila simulans , wRi induces complete CI ( i . e . crosses in which all the embryos are affected by CI ) , while wNo and wHa strains induce lower levels of CI , i . e . some embryos can develop [13] . Complete CI penetrance was also described in Nasonia spp . depending on the Wolbachia strain involved and in all Culex pipiens incompatible crosses [14 , 15] . The variability of CI penetrance has been correlated to different factors such as the Wolbachia density in the sperm and eggs [16 , 17] , host age [18] and host nuclear genotype [19] . Cellular consequences of Wolbachia-induced CI have been monitored during embryogenesis in D . simulans , D . melanogaster and N . vitripenis [5 , 20–23] . These studies revealed common cellular defects in these three species: a delay in paternal chromatin condensation and segregation defect during the first mitotic division of the embryo [23–25] . In D . melanogaster , a delay in histone H3 . 3 deposition after protamine removal on the paternal chromatin was observed and linked to chromatin remodeling defects [22] . This remodeling defect was associated with the persistence of the DNA replication factor PCNA ( Proliferating Cell Nuclear Antigen ) during mitosis , reflecting incomplete replication of paternal DNA [22] . It has been envisioned that these DNA replication defects might be responsible for the paternal chromatin bridges and segregation failure during the first mitotic division , which result in early embryonic arrest [22 , 24 , 25] . Nevertheless , some embryos reaching late development stages have been reported in CI crosses [21 , 25 , 26] . Late development in CI embryos was interpreted as resulting from a complete paternal chromatin exclusion during the first division , which allows successful maternal chromatin segregation and the formation of two haploid nuclei [25] . These haploid nuclei which further divide , lead to gynogenetic development ( i . e . haploid development with only maternal genetic material ) until late embryonic stages [27] . These haploid embryos are never viable in diploid species such as D . simulans [25] . However , in the haplodiploid parasitoid wasps Leptopilina heterotoma and N . vitripenis , CI-induced paternal chromosome defects can lead either i ) to the death of the embryos or ii ) to the production of healthy males [19 , 21 , 28 , 29] . It has been proposed that these two CI developmental outcomes could result from different degree of paternal chromatin defect ( improper condensation ) before the first division [29–31] . Severe defects would lead to complete elimination of male chromosomes from the first zygotic division resulting in haploidization and male development , whereas less severe defects would lead in partial exclusion of the paternal chromatin resulting in incomplete elimination of male chromosomes and early arrest of the aneuploid development [29–31] . In diploid species such as C . pipiens and D . melanogaster , the proportion of unhatched developed haploid embryos observed in fully incompatible CI crosses would be a proxy of the frequency of total paternal chromatin exclusion during embryogenesis due to strong CI intensity at the cellular level . However , this hypothesis is counter intuitive as one could expect that strong CI defects would prevent any development to occur while soft CI defect would allow development . The molecular mechanism underlying CI can be conceptualized as a toxin-antidote system in which i ) a toxin produced by Wolbachia in the testes , more generally called a “mod factor” , and introduced in the sperm during spermatogenesis would interfere ( “modify” ) with the paternal chromatin and induce embryonic perturbations , and ii ) an antidote released by Wolbachia in the egg , more generally called “resc factor” , would “rescue” these paternal chromatin defects to allow normal embryogenesis to occur [32 , 33] . The recent discoveries of i ) Wolbachia genes cidA and cidB ability to recapitulate the CI phenotypes when expressed in transgenic Drosophila [34 , 35] , and ii ) the link between specific allelic cidAwPip/cidBwPip variations in worldwide natural C . pipiens populations and the capacity of males to sterilize females [36] , open new paths into understanding CI mechanisms . CidA and cidB genes are syntenic genes within the WO phage region ( S6 Table ) [34 , 35 , 37–39] . CidB encodes a deubiquitylating enzyme ( DUB ) and when a cidBwPip construct bearing this catalytically inactivated DUB domain was expressed in D . melanogaster males , CI was no longer observed , showing the implication of the deubiquitylating activity in the mod function [34] . The role of cidA in the CI mechanism is more debated because i ) both cidAwMel and cidBwMel are required to induce CI in transgenic Drosophila [35] and ii ) in natural populations of C . pipiens , specific cidAwPip allelic variations were found to be linked to mod variations [36] . However , the implication of cidA in the resc function is supported by the capacity of cidA to prevent cidB toxicity in yeast [34] and the capability of transgenic uninfected females expressing cidAwMel throughout oogenesis to rescue the effect of cidBwMel [40] . In C . pipiens , all individuals are infected with different Wolbachia strains belonging to the monophyletic wPip group , but divided in five subgroups wPipI to wPipV . MLST ( Multi Locus Sequence Typing ) genes from Baldo et al . ( 2006 ) [41] were not polymorphic between wPip strains , thus a wPip specific MLST with more polymorphic genes MutL , ank2 , pk1 , pk2 , GP12 , GP15 , and RepA was used to resolve wPip phylogeny ( S6 Table ) [12] . Mosquitoes hosting wPip from the same group are likely to be compatible with each other but incompatible with mosquitoes infected with Wolbachia from other wPip groups [42] . This diversity of wPip strains distributed all around the world is responsible for the unique complexity of CI crossing types described in this host species [12 , 43] . Unlike wVitA and wMel , which harbors only one cidA/cidB copy , and wRi , which harbors two identical copies of cidA/cidB , high intra and inter-genomic diversities of cidAwPip/cidBwPip genes were uncovered between and within all wPip strains studied [36] . This diversity certainly explains the unrivaled diversity of crossing types described in C . pipiens [36] . This cidAwPip/cidBwPip genes amplification and diversification within the same Wolbachia genome may also account for the impressive CI penetrance described in C . pipiens . Indeed , expression of multiple cidAwPip and cidBwPip variants in males could i ) be responsible for differences in CI cellular phenotype ( s ) and ii ) influence the penetrance of CI . Here , we investigated the putative impact of crossing type variations and cidAwPip/cidBwPip diversification on CI cellular phenotypes and CI intensity during C . pipiens embryogenesis . To this end , we monitored the development of embryos derived from various incompatible crosses involving males from C . pipiens lines infected with Wolbachia strains from distinct wPip groups and exhibiting different crossing types .
Three different types of crosses were performed using different laboratory mosquito lines: i ) fertile crosses between individuals from the same line , representing our control to monitor normal embryonic early development , ii ) sterile crosses between mosquito lines harboring different wPip strains , and iii ) sterile crosses between infected males and uninfected females ( TC lines ) , to test the effect of Wolbachia absence on embryo development and CI cellular mechanism ( S1 and S2 Tables ) . The cellular phenotype during embryogenesis in fertile intra-line crosses is illustrated in Fig 1 . To differentially visualize the paternal from the maternal chromatin , we used propidium iodide to mark both maternal and paternal chromatin and an anti-acetylated histone H4 labelling that preferentially marks the de novo assembled paternal chromatin after protamine removal [22] . Paternal chromatin appears in green/yellow ( acetylated histone H4 labelling is dominant ) and maternal chromatin appears in red ( propidium iodide labelling is dominant ) . After fertilization , maternal and paternal pronuclei migrated toward each other and apposed ( documented embryos with confocal microscopy images n = 4 , Fig 1A ) . Then , paternal and maternal chromatins condensed and entered into first mitotic division ( n = 3 , Fig 1B ) . During the first division , paternal and maternal chromosomes aligned in separate region at the metaphase plate ( n = 1 , Fig 1C ) . Both sets of chromosomes segregated equally during anaphase ( n = 3 , Fig 1D ) to produce two diploid nuclei ( n = 1 , Fig 1E ) that proliferate mitotically ( n = 16 , Fig 1F ) . After 24 hours of development , organogenesis was ongoing and segmentation was clearly visible ( n = 2 , Fig 1G ) . In sterile crosses between two infected incompatible C . pipiens lines ( Fig 2 ) as well as in crosses between infected males and uninfected females ( Fig 3 ) , paternal and maternal pronuclei migrated and apposed normally ( n = 2 , Figs 2A and 3A ) . However , during the early prophase , paternal chromatin appeared under-condensed compared to maternal chromatin ( n = 2 , Fig 2B and 2C ) . Then the paternal chromatin failed to segregate properly during anaphase ( n = 16 , Figs 2D , 2E , 3B and 3C ) . In telophase paternal chromatin can either i ) form chromatin bridges between the two maternal nuclei ( n = 10 , Figs 2D and 3B ) , certainly causing the early arrest of embryogenesis and production of undeveloped embryos ( Figs 2G1 and 3E ) , or ii ) appear fully excluded ( n = 6 , Figs 2E and 3C ) , allowing maternal chromatin to successfully segregate and eventually formed unhatched haploid developed embryos presenting eyes and segments ( Fig 2G2 ) . Eight distinct CI crosses were monitored between males and females infected with wPip strains belonging to different wPip groups , and two distinct CI crosses were monitored between infected males and uninfected females ( S2 Table ) . However , despite this diversity of CI crosses , condensation and segregation defects of the paternal chromatin were the only observed cellular defects resulting in embryonic death , and were never observed in any embryo resulting from fertile crosses ( seven intra-line fertile crosses , S2 Table ) . 2 hours after oviposition , some embryos resulting from CI crosses between infected lines pursued their embryogenesis ( n = 2 , Fig 2F ) , and after 48 hours these embryos exhibited visible development , as segmentation was clearly observable under optical microscope ( Fig 2G2 ) . However , more than 99 . 9% of these developed embryos did not hatch . In the sterile cross between ♂ Slab x ♀ Ichkeul 13 , unhatched developed embryos only displayed maternal markers ( see Material and methods , "Ploidy determination in CI developed embryos" , S1 Fig ) , showing that they were composed of only haploid maternal DNA , as previously described in Duron and Weill ( 2006 ) [44] . All the seven different crosses performed between males infected with different wPip strains and uninfected females from different TC-treated lines produced 100% of non-developed embryos ( Fig 3E and S3 Table ) . Confocal observations of these embryos showed that only few and abnormal nuclei were observed in the cytoplasm 2 hours post oviposition ( n = 5 , Fig 3D ) , indicating an early arrest of the embryogenesis ( S2 Table ) . It has been previously proposed that the production of haploid or aneuploid embryos in CI crosses represented a proxy of intensity of CI defects that leads to more or less complete paternal chromatin exclusion [29–31 , 44] . Severe defects would lead to the complete exclusion of the paternal chromatin during the first embryonic division ( i . e . strong cellular CI intensity ) , which would allow maternal chromatin successful segregation and the production of a developed haploid embryo . Thus , unhatched developed haploid embryos reflect the occurrence during the first zygotic division of strong CI defects while unhatched non-developed embryos illustrate the occurrence of weak CI defects . We used this link between the degree of paternal chromatin exclusion ( i . e . weak or strong cellular CI ) and the proportion of unhatched developed embryos in eggs-rafts from incompatible crosses to investigate the variability of CI intensity ( i . e . frequency of strong versus weak CI defects ) . Using this proxy , we studied the variation in CI intensity between 20 incompatible crosses between infected lines ( S3 Table ) . These 20 crosses involved i ) males from four different isofemale lines ( Mal lines ) infected with wPip strains from different wPip groups all exhibiting distinct mod profiles , and ii ) females from five isofemale lines ( Fem lines ) all harbouring wPip strains from the wPipIV group and exhibiting the same resc profile [36 , 42] ( S4 Table ) . Significant differences were found regarding the proportion of unhatched developed embryos between these incompatible crosses ( generalized linear model ( GLM ) , χ 2 = 245 . 695 , df = 19 , p< 0 . 001 , Fig 4 , Table 1 and S3 Table ) . While no effect of Fem lines was detected on this proportion ( GLMM , χ2 = 2 . 508 , df = 4 , p = 0 . 643 , Fig 4 ) , the Mal lines involved in the crosses had a significant effect ( GLMM , χ2 = 16 . 211 , df = 3 , p = 0 . 001 , Table 1 and Fig 4 ) . Males from Tunis ( wPipI mod ii ) and Slab ( wPipIII mod iii ) lines induced the highest proportion of developed embryos ( 72% and 73% , respectively ) but were not significantly different from one another ( GLMM , χ2 = 0 . 002 , df = 1 , p = 0 . 968 ) ; males from Utique ( wPipI mod iv ) and Lavar ( wPipII mod vi ) lines induced significantly different and lower proportions of unhatched developed embryos ( respectively 42% and 18% , Table 1 ) . The nuclear genetic background of the males seems not to be involved in the variability of CI defects intensity: males from backcrossed line Sl ( wPipI-Tunis ) and males from the Tunis line , which host the same wPipI strain in different genetic backgrounds , indeed induced similar unhatched developed embryos proportions when crossed with the five Fem lines ( 0 . 71 ± 0 . 22 and 0 . 72 ± 0 . 19 respectively; GLMM , χ2 = 0 . 008 , df = 1 , p = 0 . 927 ) . Consequently variability in CI defects intensity appears to be only dictated by the wPip strain harbored by the different males . The results from the previous section indicate that the proportion of unhatched developed embryos in CI crosses likely depended on variations in the males' mod profiles . To investigate the sources of such variation in CI defects , we tested the putative influence of several variables: i ) the density of Wolbachia in the testes , ii ) the copy numbers of cidA and cidB genes in the different wPip genomes , iii ) the expression levels of cidA and cidB , and iv ) the cidA and cidB variants repertoires in the genomes of the different wPip strains hosted by the males . Correlative analyses were conducted to assess the potential links between variations in CI defects intensity and genetic variations . We found no significant correlations between the proportion of unhatched developed embryos in CI crosses and i ) Wolbachia density in the testes ( Spearman , ρ = 0 . 4 , p = 0 . 750 ) , ii ) cidA copy number ( Spearman , ρ = -0 . 2 , p = 0 . 917 ) , iii ) cidB copy number ( Spearman , ρ = -0 . 2 , p = 0 . 917 ) , iv ) cidA/cidB copy number ratio ( Spearman , ρ = -0 . 4 , p = 0 . 750 ) , v ) cidA expression levels ( Spearman , ρ = -0 . 4 , p = 0 . 750 ) , vi ) cidB expression levels ( Spearman , ρ = 1 , p = 0 . 083 ) , vii ) cidA over cidB expression levels ( Spearman , ρ = -0 . 2 , p = 0 . 917 ) and viii ) the number of different cidA variants in the repertories ( Spearman , ρ = 0 . 8 , p = 0 . 333 ) . However , males infected with wPip strains with 4 cidB variants induced significantly higher proportions of unhatched developed embryos ( wPipI-Tunis and wPipIII-Slab mean: 0 . 72 ± 0 . 17 ) than males infected with wPip strains with only 2 cidB variants ( wPipII-Lavar and wPipI-Utique; mean: 0 . 30 ± 0 . 26 , Wilcoxon , W = 1159 , p<0 . 001 , S4 Fig ) .
To investigate whether the high diversity of cidA/cidB variants within wPip could be responsible for variations in the cellular phenotype of CI , we studied the development of C . pipiens embryos resulting from various incompatible crosses . The early embryogenesis was assessed using fluorescence confocal microscopy in i ) fertile intra-line crosses , ii ) incompatible crosses between infected males and infected females , and iii ) incompatible crosses between infected males and uninfected females . Despite the diversity of performed crosses between males and females infected with wPip strains harboring different cidA/cidB variants repertoires or uninfected female , a unique and recurrent embryonic phenotype was detected , consisting in paternal chromatin condensation and segregation defects during the first embryonic division ( Figs 2B , 2E , 3B and 3C ) . This phenotype was never detected in any embryos derived from intra-line crosses ( Fig 1 ) . Hence the diversity of cidA/cidB variants repertoires describes in C . pipiens does not seem to influence the CI mechanism itself , which is consistent with all CidB variants carrying a conserved DUB domain [36] . Similar defects were already reported in both Drosophila and Nasonia [23 , 25] , suggesting an universality of Cid induced-cellular CI mechanism whenever cid genes are diversified or not in the Wolbachia genome . An unsolved question is the molecular pathway ( s ) targeted by CidA and CidB . Most protein domains within CidA and CidB remain to be characterized and how they interact with each other and host targets to induce CI remains unclear . However , a first tangible element is that the catalytically active DUB domain ( involved in deubiquitination ) in CidB proteins , which is considered as involved in the mod function , is necessary to induce CI in transgenic Drosophila [34] . Ubiquitination pathways have been shown to be crucial for many essential cellular processes , such as the regulation of the chromatin dynamics and the cell cycle progression [45] . Changes in ubiquitination could for instance directly or indirectly affect H3 . 3 histone incorporation after protamine removal and DNA replication as suggested by PCNA persistence on the paternal chromatin [22] , which would result in an asynchronous mitotic entry of paternal and maternal pronuclear chromatin [30] . Interestingly , Cardinium , an endosymbiont phylogenetically distant from Wolbachia , induces CI with quite similar embryogenesis defects in the hymenoptera Encarsia suzanna [46] . Moreover , an ubiquitin specific protease USP classified as a DUB protein has also been detected in Cardinium genome , suggesting a convergent implication of DUB in CI induced by insect endosymbionts [47] . However , some Wolbachia strains able to induce CI do not carry DUB domain ( i . e . no cid ) in their genomes , but display instead a paralog gene with a nuclease domain called cinB [34 , 35 , 38] . DUB ( Cid ) and Nuclease ( Cin ) domains do not have the same predicted functions suggesting that distinct molecular pathways may be responsible for CI [34 , 38] . The CI cellular defects caused by Wolbachia strains harboring only cin genes remain unknown and could differ from the one induced by cid genes . Our study showed that wPip strains , which carry both cid and cin genes in their genomes , induce similar defects during embryogenesis as wMel , which carries only a cid gene . This suggests that the association of cid and cin does not change the cellular phenotype of CI , but the molecular mechanism induces by DUB and Nuclease which must be different due to the biochemical nature of the proteins might converge on a similar cellular defect ( i . e . paternal chromatin condensation defect ) . However , the presence of DUB and Nuclease domains in the same Wolbachia genome could still contribute to CI by modifying its penetrance: wRi ( D . simulans ) and wPip ( C . pipiens ) , which harbor both paralogs , have indeed a strong CI penetrance ( almost no hatched embryos ) , while wNo and wHa ( D . simulans ) , which carry either cin or cid genes , respectively , induce lower CI penetrance [13 , 38] . Our cytological investigation in C . pipiens evidenced a link between the paternal chromatin exclusion degree during the first zygotic division and the existence of two developmental fates following first-division defects . In fact , unhatched embryos can either reach advanced developmental stages , exhibiting segments and visible eyes , or display no visible development ( Figs 2G1 , 2G2 and 6 ) [26 , 44] . We confirmed Duron and Weill ( 2006 ) [44] findings that the unhatched developed embryos resulting from CI were haploid , and carried genetic material from maternal origin only ( Fig 6 and S1 Fig ) . Confocal observations showed that such haploid development likely occurred when paternal chromatin was fully excluded during the first zygotic division , allowing the successful segregation of the isolated maternal chromatin ( Fig 6 ) . In contrast , unhatched non-developed embryos would be due to partial exclusion of the paternal chromatin , which would result in aneuploid nuclei and early arrest of embryogenesis ( Fig 6 ) . It has been previously proposed for other arthropod models that the participation of paternal chromatin to the first division would depend on the intensity of paternal chromatin defects ( i . e . improper condensation ) [29 , 30 , 44 , 48] . Severe defects would lead to complete paternal exclusion ( i . e . strong cellular CI ) and to the production of haploid developed embryos , while less severe defects would lead to a partial paternal chromatin exclusion ( i . e . weak cellular CI ) and to the production of aneuploid non-developed embryos . We used this link between the degree of paternal chromatin exclusion and the ratio of unhatched developed and non-developed embryos in eggs-rafts from incompatible crosses to investigate the variability of cellular CI intensity between different incompatible crosses . We first studied the variability of CI intensity using males and females both infected with incompatible wPip strains . Developed embryos were observed in all these incompatible crosses , with two possible outcomes: i ) less than one per thousand of these embryos were apparently not affected by CI and hatched into diploid larvae [14 , 44] , and ii ) from 11% to 85% of the unhatched embryos , depending on the crosses , reached late embryonic developmental stages showing that they experienced strong CI defects ( Figs 4 and 6 ) . We then studied the influence of the absence of Wolbachia in the oocytes on the cellular CI intensity . As in Duron and Weill ( 2006 ) [44] , i ) we confirmed that not a single larvae was produced in such crosses , and ii ) all the seven CI crosses between infected males and uninfected females ( TC lines ) resulted in 100% of non-developed embryos suggesting that in such crosses , CI phenotype was always weak ( Fig 6 ) . In crosses between infected individuals , it clearly appeared that Mal lines harboring Wolbachia from different wPip groups ( wPip I , II , III ) and displaying distinct mod induced significant variation in CI defects intensity when crossed with females harboring distinct wPipIV strains displaying the same resc ( Figs 4 and 6 ) . Variation in CI defects intensity has already been reported in Nasonia species , where the production of haploid viable males in N . vitripenis was interpreted as resulting from severe paternal chromatin defects , while the production of unviable aneuploid embryos in N . longicornis and N . giraulti was interpreted as resulting from weak paternal chromatin defects . However , variation of CI intensity in these host species was not associated with the different Wolbachia strains , but to variation in host genetic backgrounds [19] . The backcross experiment performed in the present study suggests that CI intensity is not impacted by nuclear genetic variations in C . pipiens . While it was already established that Wolbachia drives alone the observed variation in crossing types in C . pipiens [11 , 49 , 50] , Wolbachia also seems to dictate the intensity of CI defects . Consequently , the variation in CI intensity observed when two infected individuals are crossed seems to be under the major influence of the wPip strain infecting the Mal line via the degree of paternal chromatin exclusion they trigger . In C . pipiens , when females from tetracycline-cleared lines ( TC females ) were crossed with the four Mal lines , 100% of unhatched non-developed embryos only exhibiting few degenerated nuclei were observed , even 2 hours after oviposition ( Figs 3D and 6 ) . Such CI phenotype suggests that the defects caused by the wPip infecting all the Mal lines are always weak ( Fig 6 ) . This result is counter intuitive because one would expect that when Wolbachia is absent from the eggs CI should be always strong and many haploid embryos should be produced . We mentioned above that all the Mal lines can induce strong CI defects in variable proportion of the embryos when crossed with infected females . Consequently , the constant weak CI phenotype observed when females are not infected is linked to the absence of Wolbachia during egg maturation . Our results suggest that in incompatible crosses between infected C . pipiens individuals , the presence of maternal Wolbachia somehow interferes with early embryogenesis allowing haploid development to occur . It seems very unlikely that the presence of incompatible Wolbachia in the egg would enhance the mechanisms leading ultimately to paternal chromosome condensation defects ( i . e . accentuate the mod function ) to result in its total exclusion during the first embryonic division . Instead , the presence of incompatible Wolbachia in the eggs may have an additive effect on the incompatibility between pronuclei , not by directly affecting the paternal chromatin but by influencing the cell cycle timing . For instance , maternal Wolbachia could modulate the maternal kinetics for DNA replication or the mitotic entry during early development , increasing the incompatibility between pronuclei and therefore favoring the haploid development . Thus , while paternal Wolbachia-induced CI defects always occur regardless of the infection status of the eggs , the absence of incompatible maternal Wolbachia would block haploid development resulting in weak CI phenotype . We then investigated the putative genetic determinism of CI intensity variation in embryos derived from infected parents . We assessed whether it could result from difference in Wolbachia density , cidA-cidB gene expression , copy numbers , or variant diversity between the wPip strains . As previously described in Drosophila [35 , 38] , we found in C . pipiens that cidA was always significantly more expressed than cidB , whatever the wPip strain ( Table 1 ) . This is in accordance with the hypothesis that cidA and cidB form a toxin-antidote system where CidA is the antidote of CidB [34 , 36] . Indeed , in such system the antidote was always found more expressed than the toxin to prevent the host from toxicity [51] . No significant difference between Mal lines was found for cidA and cidB expression levels per Wolbachia cell ( Table 1 ) , suggesting that the cidA and cidB expression does not influence CI defects intensity . However , while the cidA and cidB expression levels per Wolbachia cell did not significantly vary between C . pipiens lines , the total amount of CidA and CidB proteins in the host mainly depends on the density of Wolbachia . Since the mod factors are most likely deposited on the sperm in the testes during spermatogenesis [32 , 33] , we measured the density of Wolbachia in the male gonads . We found that Lavar males hosted significantly less Wolbachia in their testes than males from the three other lines ( Fig 5 ) ; Lavar males were also those that generated the lowest proportion of unhatched developed embryos in their offspring , whatever the Fem lines ( Fig 4 ) . Due to lower Wolbachia density in the testes , the global amount of CidB protein could be lower in Lavar line compared to the other lines . This low dosage of CidB would more likely result in weak CI defects leading to only few haploid development . However , this hypothesis relies on a single line and requires more C . pipiens lines with distinct testicular Wolbachia densities to be confirmed . Lavar was also the line with the highest cidA expression relatively to cidB ( Table 1 ) ; as CidA has been proposed as the CidB antidote [34 , 36] , its overexpression could reduce CidB-induced CI defects , and contribute to the low frequency of developed haploid embryos observed in crosses involving males from Lavar line . We previously demonstrated that the amplification followed by the diversification of cidA and cidB variants in wPip certainly constitutes the source for CI diversity profiles in C . pipiens while cinA and cinB did not exhibit any polymorphism [36] . Indeed , specific variations in cidA and cidB repertoires ( number and/or nature of the variants ) clearly seemed to determine the compatibility outcome of crossings between wPipIV-infected males and any infected females , pointing out the putative role of these variations in the prodigious CI complexity recorded in this species [36 , 42] . Here , we tested the putative consequence of cidA and cidB gene amplification ( i . e . number of copies per genome ) on variation of CI defects intensity , and demonstrated no significant correlation between the two parameters . When the quantification of genomic copies obtained by q-PCR are put in relation to the number of different variants in the same isofemale line obtained by cloning-sequencing , some of cidA results appear discordant . This is especially true for the Slab line , which exhibits ten distinct cidA variants for ~5 copies per genomes quantified ( Table 1 and S2 Fig ) . Even taking into account technical limits of q-PCR to quantify high level of gene amplification , this discordance suggests that , at least in the Slab line , some of the Wolbachia cells do not harbor the same cidA variants . We found that the different wPip strains carried by the four Mal lines exhibiting different mod profiles harbored distinct cidB variants . Any variant of this gene could certainly trigger CI alone , as the DUB domain is perfectly conserved between all variants [36] . However , their diversity can modulate CI defects intensity . We thus tested whether cidB repertoire diversity could play a role in CI intensity variability . Supporting this hypothesis , we found that males from the two C . pipiens lines harboring wPip strains with four different cidB variants induced higher proportions of unhatched developed embryos compared to lines harboring wPip with only two different cidB variants ( S4 Fig ) . Each distinct cidB variants could differentially impact the paternal chromatin ( i . e . like different locks ) , putatively leading to an additive mod effect: the more different cidB variants present in a wPip strain , the more likely strong CI defects . However , more wPip strains varying in their diversity of cidB are required to further test this hypothesis . In conclusion , despite the diversity of crossing types observed in C . pipiens , linked to the diversity of cidA/cidB variants repertoires , a single cellular phenotype of CI , was observed in this species . In all crosses ( i . e . uni-bidirectionnal ) , CI results in early developmental defects in the paternal chromatin condensation and segregation during the first zygotic division similar to that observed in other insects . Our study demonstrates that in CI crosses between two infected individuals , the CI intensity ( i . e . frequency of strong and weak CI defects ) is influenced by the male-carried wPip . However , when the female is not infected , and despite the variability of the distinct wPip strains carried by the males , no unhatched developed embryos ( strong cellular CI ) were ever found , suggesting that the weak CI phenotype observed in such crosses is instead due to the absence of Wolbachia in the eggs . Genetic investigation reveals that the variability of CI defects intensity may be linked to cidB variant diversity in wPip strains . While the putative functional role and the singularity of cidB amplification and diversification in wPip remains yet to be fully solved , it clearly appears that it deeply modifies the wPip-induced CI phenotype at different scales , from crossing types [36] to its intensity at the cellular level .
To characterize CI cellular phenotype ( s ) in C . pipiens , several crosses were performed ( S2 Table ) . For every crosses , to avoid confounding age effects , two-day old adults were released in cages . Cages containing 100 females and 50 males were then put into a closet at 25°C where day-night cycle was inverted to allow collection of early developmental stage eggs during the day . After six days in these cages , females were fed with turkey blood in heparin sodium ( bcl Wholly Wild World ) using a Hemotek membrane feeding system ( Discovery Workshops , United Kingdom ) . Five days after blood meal , water-pots were placed into the cages to collect the eggs-rafts . For C . pipiens eggs , at 25°C , the meiosis is approximatively completed 30 minutes after the oviposition and the first mitotic nucleus division 15 minutes after the end of the meiosis , while four hours after oviposition the embryos normally reach the syncytial blastoderm stage [53] . Since , the CI defects described in D . simulans [25] and N . vitripenis [23 , 24] occurred during the first nucleus mitotic division , we mainly collected eggs aged from 30 minutes to 1hour . Older eggs were also harvested to monitor further developmental stages in both fertile and sterile crosses . Eggs-rafts were then placed into commercial bleach ( active ingredient , 9 . 6% of sodium hypochlorite ) to dissociate eggs , and then washed in distilled water . They were then fixed by being shaken for 2 hours in a solution of 3 . 2% para-formaldehyde in PBS 1X with Tween 0 , 02% ( PBS-T ) and washed with PBS 1X . For each fixed egg , the chorion was removed manually with a needle under an optical microscope ( Leica MZ 8 ) . Dechorionated embryos were then collected and treated with RNAse A ( 10 mg/mL , Sigma ) overnight . To differentially visualize the paternal from the maternal chromatin , we used propidium iodide to mark both chromatin and an anti-acetylated histone H4 labelling that preferentially marks the de novo assembled paternal chromatin after protamine removal [22] . Thus maternal and paternal chromatin will be respectively predominantly marked with propidium iodide ( mosty red fluorescence ) and with anti-acetylated histone H4 antibodies ( mostly green fluorescence ) . For immunolabeling , embryos were first incubated overnight at 4°C with primary antibodies ( Polyclonal anti-acetylated histone H4 primary antibody ( 1:1000 , Upstate ) ) , washed during one day with PBS-T 1X , then incubated overnight at 4°C with the secondary antibody ( Alexa Fluor 488 goat anti-rabbit IgG secondary antibodies ( 1:250 , Invitrogen ) ) then washed with PBS-T 1X . Embryos were then incubated in PBS-T 1X for 20 minutes with propidium iodide a DNA intercalating agent ( Molecular Probes , 10μL/1mL ) . Finally , embryos were washed for 5 minutes and mounted between slide and coverslip in Fluoroshield Mounting Medium ( Vector ) . Confocal microscope images were captured on an inverted photoscope ( DMIRB; Leitz ) equipped with a laser confocal imaging system ( TCS SP5; Leica ) using an HCX PL APO 1 . 4 NA 63 oil objective ( Leica ) . Images from fixed , immunostained embryos are merged confocal z-stacks taken sequentially in the green and red channels for the anti-acetylated histone H4 labelling and the propidium iodide signal respectively . Crosses from which confocal microscope images were obtained ( Figs 1–3 ) are listed in S3 Table . To study the proportion of unhatched developed embryo in CI crosses , we performed a total of 32 crosses: 20 crosses involving four lines for the males ( Mal lines ) and five lines for the females ( Fem lines ) , 5 involving Sl ( wPipI-Tunis ) for the males and the five Fem lines , and 7 involving the four Mal lines and females from different TC lines ( S2 Table ) . All these crosses were performed using 50 females and 25 males . After 6 days in the cages , females were blood-fed and after 5 days eggs-rafts were collected in water pot and deposited into 24 wells plates . As hatching normally occurs approximately 48 hours after oviposition , developmental status in non-viable rafts was characterized at least two days after eggs-rafts collection . To attribute a developmental status to each egg , eggs-rafts were mounted between slide and coverslip , observed and documented with an optic microscope ( Axiophot2 equipped with a CCD camera , Zeiss ) . Two developmental statuses were discriminated i ) unhatched embryos harboring no visible development ( Fig 2G1 ) , or ii ) unhatched embryos with visible development ( Fig 2G2 ) . For each cross , we calculated the proportion of embryos showing development for 50 embryos per eggs-raft in 10 eggs-rafts ( total of 500 eggs observed per cross ) . To assess the ploidy status in unhatched developed embryos , we used a PCR/RFLP diagnosis kdr/RsaI that allowed discriminating between C . pipiens and C . quinquefasciatus lines , as previously described in Duron and Weill ( 2006 ) [44] . Slab ( C . quinquefasciatus ) and Ichkeul 13 ( C . pipiens ) were chosen because they exhibit an unidirectional sterile cross: fertile in the direction ( ♂Ichkeul 13 x ♀Slab ) and sterile in the other direction ( ♂Slab x ♀Ichkeul 13 ) . This PCR/RFLP test was performed on DNA extracted as describe above from i ) a pool of larvae from Slab and Ichkeul 13 parental lines and ii ) from eggs-rafts resulting from the two reciprocal crosses between those two lines . To describe the diversity of cidAwPip/cidBwPip repertoires for the two C . pipiens lines Utique and Slab not yet investigated , cloning and Sanger sequencing of the cidA and cidB variants were performed as described in Bonneau et al . ( 2018 ) [36] on DNA from pools of larvae extracted as described above . Variant sequences were aligned , using the Muscle algorithm implemented in Seaview 6 . 4 . 1 software [55] . Variability of unhatched developed embryo proportion in sterile crosses was analyzed using a generalized linear model ( GLM ) : Udep = Cross + ε , with Udep the proportion of unhatched developed embryos for each cross ( Cross , which represent the interaction between the Mal and Fem lines ) and ε the error parameter , following a binomial distribution . To test the specific effect of the four Mal lines and the five Fem lines separately , GLMs with mixed effects ( GLMM ) were used: Udep = Male + Female + 1|Cross + ε with Male and Female respectively the Mal and Fem lines involved in each cross as fixed effects , with Cross as the interaction between Mal and Fem lines as a random effect ( as crosses to produce embryos necessary require an interaction between females and males ) , and ε the error parameter , following a binomial distribution . To test for a specific effect of the host genetic background in crosses involving males from Sl ( wPipI-Tunis ) and Tunis lines which host the same Wolbachia in two different genetic background we used a GLMM: Udep = MalBack + Female + 1|Cross + ε with Udep the unhatched developed embryos proportion for each cross involving males from Sl ( wPipI-Tunis ) and Tunis lines ( MalBack ) and the five Fem lines ( Female ) as fixed effects , with Cross as a random effect , and with ε the error parameter , following a binomial distribution . For several variables ( Wolbachia density in testes , cidA and cidB expressions and copy number ) obtained with q-PCR , variability between the four Mal lines was analyzed using GLMs in the form Var = Male + ε , with Var one of the estimated variable of the Mal line ( Male ) and ε the error parameter , following a Gaussian distribution . Spearman correlation tests [56] were used to test for correlation between these variables ( Wolbachia density in testes , cidA and cidB expressions and copy number ) and the proportion of unhatched developed embryos for each Mal line . We did the same for the relation between the number of different cidA variants and the proportion of unhatched developed embryos for each Mal line . Finally , Wilcoxon test [57] was used to compare mean proportions of unhatched developed embryos between the two Mal lines harboring only two different cidB variants and the two Mal lines harboring four different cidB variants . All computations were performed using the R version 3 . 4 . 4 [58] . Computed models were simplified by testing the significance of the different terms using likelihood ratio tests ( LRT ) and starting from the higher-order terms , as described in Crawley [59] . Factor levels of qualitative variables that were not different in their estimates ( using LRTs ) were grouped as described by Crawley [59] . The normality of the residuals was tested using Shapiro test for models with Gaussian error [60] . For models with Binomial error , overdispersion was calculated using the “dispersion_glmer” function from the package blmeco for GLMM , and by dividing the residual deviance by the residuals degree of freedom of the model for GLM [61]; when detected , overdispersion was taken into account in the LRTs [62 , 63] .
|
In some crosses , mosquito males belonging to the species Culex pipiens prevent their females from having live progenies . This phenomenon called cytoplasmic incompatibility ( CI ) is caused by intracellular bacteria named Wolbachia . CI occurs when males infected with Wolbachia fertilize females infected with genetically distinct incompatible Wolbachia resulting in the death of all the embryos . At the world scale , crossing relationships between C . pipiens are quite puzzling . Despite this complexity in crossing relationships and the diversity of cidB genes involved in CI mechanisms in C . pipiens , we demonstrate a single shared CI cellular phenotype leading to the death of the embryos: the paternal chromatin exclusion from the first embryonic division . If paternal chromatin is fully excluded , embryos developed with haploid set of chromosomes . We show that the frequency of haploid development varies according to the Wolbachia strains hosted by the males which differ in the cidB variants harbored in their genomes . Absence of Wolbachia in the eggs totally block haploid development showing that maternal Wolbachia presence interplays with CI mechanisms in a way that allows haploid development to occur . Understanding CI mechanism in mosquitoes is the corner stone to build new sustainable and adaptable Wolbachia based strategies for vector control .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"reproductive",
"system",
"condensation",
"condensed",
"matter",
"physics",
"dna-binding",
"proteins",
"animals",
"wolbachia",
"animal",
"models",
"developmental",
"biology",
"drosophila",
"melanogaster",
"model",
"organisms",
"experimental",
"organism",
"systems",
"epigenetics",
"embryos",
"bacteria",
"chromatin",
"drosophila",
"research",
"and",
"analysis",
"methods",
"embryology",
"chromosome",
"biology",
"proteins",
"animal",
"studies",
"gene",
"expression",
"histones",
"phase",
"transitions",
"insects",
"arthropoda",
"physics",
"testes",
"biochemistry",
"eukaryota",
"cell",
"biology",
"anatomy",
"embryogenesis",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"organisms",
"genital",
"anatomy"
] |
2018
|
The cellular phenotype of cytoplasmic incompatibility in Culex pipiens in the light of cidB diversity
|
In the recent 2014–2016 Ebola epidemic in West Africa , non-hospitalized cases were an important component of the chain of transmission . However , non-hospitalized cases are at increased risk of going unreported because of barriers to access to healthcare . Furthermore , underreporting rates may fluctuate over space and time , biasing estimates of disease transmission rates , which are important for understanding spread and planning control measures . We performed a retrospective analysis on community deaths during the recent Ebola epidemic in Sierra Leone to estimate the number of unreported non-hospitalized cases , and to quantify how Ebola reporting rates varied across locations and over time . We then tested if variation in reporting rates affected the estimates of disease transmission rates that were used in surveillance and response . We found significant variation in reporting rates among districts , and district-specific rates of increase in reporting over time . Correcting time series of numbers of cases for variable reporting rates led , in some instances , to different estimates of the time-varying reproduction number of the epidemic , particularly outside the capital . Future analyses that compare Ebola transmission rates over time and across locations may be improved by considering the impacts of differential reporting rates .
Despite the unprecedented scale of the 2014–2016 Ebola epidemic in West Africa [1] , significant uncertainty remains about the precise number of cases involved , and about the spatiotemporal dynamics of transmission [2–8] . The numbers of non-hospitalized , community-based cases over time and among locations , are particularly uncertain . Ebola cases who become sick and die in the community are at increased risk of onward transmission , with caregiving and fluid contact as especially important transmission routes [9–11] . Accurate case counts over space and time , which include non-hospitalized cases , are important for estimating disease transmission rates , and identifying response strategies [6 , 12–14] . A central problem is that non-hospitalized cases are less likely to be reported , compared to hospitalized cases: the WHO 2014–2016 Ebola case line-list predominantly includes cases who have been hospitalized [15] . As a result , non-hospitalized cases are underrepresented in Ebola surveillance data , and the observed pattern of cases over time reflects the dynamics of hospital bed capacity and access to formal care as well as the underlying trajectory of the epidemic [16] . A community case may be detected outside of clinical care through interaction with other epidemic response measures , for instance by inclusion in contact tracing , or by laboratory testing of a specimen collected before or after death . However , during the recent epidemic these measures were implemented heterogeneously , due to constraints on public health systems in the affected countries . Estimates of the reporting rate—the proportion of the total number of infected individuals over a specified time period that are reported as cases—range from 0 . 33 [17] to 0 . 83 [18] , which bound the initial estimate by the United States Centers for Disease Control ( CDC ) of 0 . 40 [19] . The differences among these estimates may stem in part from the different analytical methods used , including capture recapture methods [17] , inference from viral sequence data [18] and comparison of hospitalized cases with the projections of a compartmental epidemic model [19] . Reporting rates may vary significantly over time and location . For instance , temporal fluctuations in the number of available hospital beds may cause corresponding fluctuations in case ascertainment rates [11] . Increased case ascertainment through increased access to care may itself lower transmission rates , because hospitalized cases are less likely to transmit [10] . The number of non-hospitalized cases , and the proportion of non-hospitalized cases relative to the total case burden , are also important indicators of the impact of interventions [20] , as fewer community based cases reflects better access to hospital care , or a downturn in transmission rates , or both . Estimates of reporting rates for non-hospitalized cases may not be a priority at the beginning of an outbreak . During an outbreak , such infrastructure is difficult to establish at an appropriate scale , beyond informal reports , until resources are allocated according to case data from patient hospitalizations . As a result , the non-hospitalized cases that occur in the initial weeks of an outbreak , in a new area , are often not extensively captured . Thus , transmission and reporting may be intertwined , making it difficult to tell whether a change in incidence is driven by contagion or surveillance thereof , although the relative contributions of the two processes can have important implications for public health responses . Estimates of disease transmission dynamics during the 2014–2016 Ebola epidemic drove recommendations for a variety of containment strategies—including contact tracing , quarantine , and safe burial—based on estimates of the reproductive number of the disease in different contexts [21] . However , it is unclear how robust these analyses are to temporal and spatial fluctuations in ascertainment bias in case time series , especially in situations where many cases are unreported . Here we disentangle reporting and epidemic dynamics for the recent Ebola epidemic in Sierra Leone using individual-level records of burials performed in the Safe and Dignified Burial ( SDB ) program that was coordinated by the International Federation of Red Cross and Red Crescent Societies ( IFRC ) in Sierra Leone . Safe and dignified burial was required by law for every community death in Sierra Leone , during the timespan of the data analyzed here ( Oct 20 , 2014 –March 30 , 2015 ) . Safe burials were to be conducted in the same way regardless of the suspected cause of death , including collecting a skin swab sample for Ebola testing [22] . However , in practice safe burial was not conducted for every community death , and not all community deaths were tested for Ebola . Non-hospitalized Ebola deaths that did not interface with the SDB program , or where a swab sample was not tested , represent a primary source of underreporting in the epidemic , as well as a significant potential source of onward transmission . Below we describe how the prevalence of Ebola in burial swab samples can be used to estimate the total number of non-hospitalized cases , which can in turn be used to estimate location- and time-specific reporting rates . Using these data , we examine how reporting rates varied over time and across districts , and reconstruct the epidemic curves in each district , accounting for unreported cases . Finally , we address the question of how spatiotemporal dynamics in reporting might affect estimates of disease transmission rates . Our results agree with previous studies [17–19] showing that a substantial fraction of Ebola cases in the 2014–2015 epidemic were unreported . In addition , we find substantial systematic variation across geographic regions and over time in the level of underreporting . In some cases , this variation was sufficient to systematically alter estimates of the reproduction number of the epidemic .
This study relies solely on retrospective analysis of de-identified data , which was collected as part of a humanitarian response and not for research purposes , and is exempt from ethics committee approval .
Ebola prevalence in the SDB data was strongly correlated with reported prevalence in the WHO situation reports , however , the relationship between burial prevalence and reported prevalence varied across districts ( Fig 2 , Table 1 ) . In the capital district of Western Area Urban and the adjacent district of Western Area Rural , burial prevalence accrued more slowly under increasing reported prevalence , while in the districts of Bo and Bombali , burial prevalence rose more steeply under increases in reported prevalence . This is consistent with the hypothesis that reported Ebola prevalence outside of the capital was indicative of a proportionally larger number of unreported , non-hospitalized cases ( Fig 3 , top row ) . Fitting the time-varying hierarchical binomial model to the data quantified variation in reporting rate over time and among districts ( Fig 3 , bottom row ) . Midline estimated reporting rate in Western Area Urban on Oct 20 , 2014 was 0 . 68 ( credible interval: {0 . 45 , 0 . 73} ) . However , estimates for the same date in other districts were significantly lower , at 0 . 55 , 0 . 27 and 0 . 33 , for Western Area Rural , Bombali , and Bo respectively . Reporting rates increased over the course of the epidemic in all districts in the data , so that by March 30 , 2015 , reporting rates in all districts were estimated to be at or above Scarpino et al . ’s estimate of 0 . 83 [18] . However , reporting rates outside the capital district stayed lower for longer before beginning to increase . For instance , while the midline estimated reporting rate in Western Area Urban on the week of January 5 , 2015 had risen to 0 . 96 ( credible interval: {0 . 93 , 0 . 98} ) , estimates for the same date in Bombali and Bo remained close to their initial values ( see Fig 3 , bottom row ) . The reproduction number of Ebola estimated from the WHO data varied over time and among districts , as in previous estimates [4] . The transmission rate of Ebola may be particularly variable in a community context , due in part to variation in the relative risk of different transmission routes that may occur in a community setting , particularly care-giving outside of hospital [29] . To quantify the effect of reporting variation on estimates of Rt in Ebola incidence data , we compared results for time series that were corrected for underreporting ( using our midline estimates of reporting rate in each district over time ) with uncorrected time series ( Fig 4 ) . While in the Western Area districts ( around the capital of Freetown ) correcting for temporally variable underreporting did not significantly change estimates of Rt , in Bo and Bombali the corrected estimates diverged significantly from the uncorrected estimates , particularly earlier in the epidemic . Accounting for variation in reporting can significantly modify our understanding of disease spread and control , but variation in reporting rate over space and time is rarely accounted for during analyses of epidemic dynamics . In particular , most models of Ebola transmission to date have assumed constant reporting rates across space and over time [5 , 6 , 12 , 17–19 , 21 , 30 , 31] . Here we used community-based data on non-hospitalized deaths to infer variation in patterns of reporting , finding significant spatiotemporal variability in case ascertainment . This spatiotemporal variability echoes recently described heterogeneity in transmission patterns in non-hospitalized Ebola data , where the importance of superspreading events was demonstrated [32] . Correcting for these reporting variations improved the accuracy and precision of estimates of transmission patterns . How do fluctuations in reporting rate influence estimates of the reproductive rate of an epidemic ? As the number of cases becomes large , the posterior mean of the estimated reproductive number approaches Rt=∑s=t−τ+1tρsIs∑s=t−τ+1t∑u=1twuρt−uIt−u ( 5 ) where τ is the number of time units over which the reproductive number is assumed constant , ρt is the reporting rate at time t , It is the number infectious at time t , and wt is the generation time distribution of the disease , representing the fraction of secondary cases that originate from the primary case t time units after the primary case becomes infectious [28] . If the reporting rate is constant , ρt = ρ0 , then reporting does not affect estimates of Rt , because the constant ρ0 cancels out in the numerator and denominator of ( 5 ) . However , variable reporting rates are confounded with generation time , and shape estimates of Rt in the same ways that variation in generation time can [33] . For example , a sharp increase in reporting rate will lead to an overestimate of Rt by inflating the numerator relative to the denominator . In the context of field outbreak response , such an increase in reporting might be caused by increased allocation of resources to contact tracing , or an increase in hospital bed capacity . Variation in reporting rates can also affect measures of uncertainty in estimates of Rt . For large numbers of cases , the coefficient of variation in the estimate is given by CV ( Rt ) =1∑s=t−τ+1tρsIs ( 6 ) which is equivalent to the standard deviation of the estimated Rt divided by its mean , and thus can affect the width of the credible intervals for Rt . The denominator of ( 6 ) is proportional to the covariance between reporting and incidence over a time interval of length τ . Thus , if reporting and incidence covary , the credible interval on estimates of the reproductive number will shrink when corrected for underreporting , all else equal . Future epidemic models may be improved by incorporating a process-based representation of reporting dynamics . More specifically , future models could treat reporting rate as a state variable , driven by human behaviors associated with both disease spread and public health response , and including inequities in access to medical care . Improving our quantitative understanding of what determines reporting rates could also allow stronger links between field outbreak response teams and modeling teams , which would improve contextualization and understanding of data limitations , with the potential to improve predictive models of epidemics and enhance the design of control measures .
|
Epidemics are defined by a surge of cases of a disease , yet often a significant number of cases in an epidemic are never reported , for example because not all infected individuals have access to medical care . This underreporting can introduce bias into analyses of disease spread , by distorting patterns in where and when the most cases are observed . Conversely , quantifying underreporting can improve epidemic forecasts and containment strategies . In this study , we analyze data from the recent Ebola epidemic in West Africa , including the time , location and Ebola status of 6491 individual community burials , conducted over 25 weeks in four districts in Sierra Leone . We quantify how reporting rates varied over space and time , and show that estimates of transmission rates that are corrected for dynamic underreporting diverge significantly from uncorrected estimates , particularly earlier in the epidemic and outside the capital .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"and",
"discussion"
] |
[
"death",
"rates",
"medicine",
"and",
"health",
"sciences",
"infectious",
"disease",
"epidemiology",
"geographical",
"locations",
"health",
"care",
"population",
"biology",
"infectious",
"disease",
"control",
"africa",
"public",
"and",
"occupational",
"health",
"infectious",
"diseases",
"geography",
"epidemiology",
"sierra",
"leone",
"hospitals",
"people",
"and",
"places",
"population",
"metrics",
"health",
"care",
"facilities",
"urban",
"areas",
"disease",
"dynamics",
"earth",
"sciences",
"geographic",
"areas",
"biology",
"and",
"life",
"sciences"
] |
2018
|
Unreported cases in the 2014-2016 Ebola epidemic: Spatiotemporal variation, and implications for estimating transmission
|
p53 tumor suppressor has been identified as a protein interacting with the large T antigen produced by simian vacuolating virus 40 ( SV40 ) . Subsequent research on p53 inhibition by SV40 and other tumor viruses has not only helped to gain a better understanding of viral biology , but also shaped our knowledge of human tumorigenesis . Recent studies have found , however , that inhibition of p53 is not strictly in the realm of viruses . Some bacterial pathogens also actively inhibit p53 protein and induce its degradation , resulting in alteration of cellular stress responses . This phenomenon was initially characterized in gastric epithelial cells infected with Helicobacter pylori , a bacterial pathogen that commonly infects the human stomach and is strongly linked to gastric cancer . Besides H . pylori , a number of other bacterial species were recently discovered to inhibit p53 . These findings provide novel insights into host–bacteria interactions and tumorigenesis associated with bacterial infections .
p53 protein has been receiving significant attention for more than 30 years . This interest originates from the protein’s prominent role in tumor suppression that was eloquently paraphrased in the scientific literature as “the guardian of the genome” [1] . p53 is a key component of the cellular mechanisms controlling cellular responses to various cellular stresses , including DNA damage , aberrant oncogene activation , loss of normal cell–cell contacts , nutrient deprivation , and abnormal reactive oxygen species ( ROS ) production . Following cellular stresses , p53 is activated and primarily functions as a transcriptional regulator of expression of multiple effector proteins and miRNAs , which , in turn , regulate key cellular processes such as apoptosis , cellular proliferation , and autophagy . Since regulation of cellular stress responses is tightly intertwined with metabolic regulation , there is an interplay between p53 and multiple pathways involved in regulation of metabolism and cellular homeostasis that is complex and not fully understood . One prominent example is a reciprocal signaling between p53 and mTOR [2] . The latter pathway plays a key role in cell growth and proliferation . p53 is also directly involved in regulation of the cellular energy metabolism and the redox balance regulating glycolysis , oxidative phosphorylation , and the pentose phosphate pathway ( PPP ) . Through multiple mechanisms , p53 can dampen glycolysis and the PPP and promote oxidative phosphorylation . The metabolic functions of p53 are likely to significantly contribute to its tumor suppression activity ( Fig 1 ) . Inactivation of p53 is a hallmark of tumorigenic changes . More than half of all tumors carry p53 mutations , rendering the p53 gene ( tp53 ) the most mutated gene in human tumors . p53 can also be inhibited by mutation-independent mechanisms . Inhibition of wild-type p53 by the SV40 virus was one of the first reported examples . SV40 is a small DNA tumor polyomavirus that induces cellular transformation in cell culture and an array of different tumors in animals . In infected cells , viral protein ( SV40 large T antigen [T-Ag] ) binds p53 and inhibits p53-dependent transcription , resulting in accumulation of inactivated p53 protein [3 , 4] . Inhibition of p53 by large T-Ag is closely linked to the ability of the SV40 virus to induce tumorigenic transformation; SV40 mutants , which are defective in inhibition of p53 , are also defective in cellular immortalization and transformation [5 , 6] . p53 by itself was originally identified as a protein binding to SV40 large T-Ag [7 , 8] . Later studies have shown that SV40 T-Ag is not unique in this sense , and other small tumor DNA viruses ( adenoviruses and papillomaviruses ) also produce similar proteins ( E1B-55K and E6 ) that interact with p53 [9 , 10] . Although adenoviral protein E1B-55K and human papillomavirus ( HPV ) protein E6 are different in their amino acid sequences , they converge at the same function , forming protein complexes with p53 to inhibit its activity . HPV and adenovirus ( Ad ) can also induce ubiquitination and proteasomal degradation of p53 [11] . The ability to degrade p53 varies among viruses . For example , high-risk genital HPV types 16 and 18 , which cause around 70% of cervical cancers , efficiently degrade p53 , while low-risk viruses such as HPV types 6 and 11 are unable to do so [12 , 13] . Similarly , p53 is degraded by human adenovirus serotypes 12 and 5 ( Ad12 , Ad5 ) , while Ad9 and Ad11 do not have this ability [14 , 15] . To degrade p53 , both HPV and Ad use the host protein degradation machinery . HPV E6 protein interacts with the host E3 ubiquitin ligase , E6AP , causing its substrate specificity to be altered so that it ubiquitinates p53 and induces its degradation by the 26S proteasomes [16] . In Ad-infected cells , viral proteins E1B-55K and E4orf6 interact with cellular proteins Cullin5 ( or Cullin2 ) , Rbx1 , and Elongins B and C to form a Cullin-containing E3 ubiquitin ligase that targets p53 for proteasomal degradation [14 , 17 , 18] . A similar degradation strategy is also used by the Epstein–Barr virus ( EBV ) , which forms a complex containing viral protein BZLF1 and cellular Cullin2/5-containing E3 ubiquitin ligase to degrade p53 [19] . Due to a relatively simple organization of the viral genomes , viruses have to rely on host resources for most aspects of their life cycle . In the process of interacting with host cells , they alter the intracellular environment to make it suitable for viral replication . These drastic alterations , however , may cause cellular stress and activate p53 , resulting in cell cycle arrest or apoptosis of host cells; both outcomes are detrimental to viral replication . It is plausible that inhibiting p53 may provide advantages to viruses that have evolved to do so . Recently , this concept was further expanded to include additional microorganisms . These novel data are discussed in this review , focusing on specific mechanisms of bacterial inhibition of p53 .
Recent studies have found that it is not only viruses , but also some pathogenic bacteria , that actively inhibit p53 and induce its degradation . This phenomenon was initially described in gastric cells co-cultured with Helicobacter pylori [20] . H . pylori is a gram-negative , spiral-shaped pathogen that lives in the stomachs of approximately half of the world’s population . The infection is typically acquired during childhood and causes lifelong chronic infection . Because of the association between H . pylori infection and the incidence of gastric cancer , the International Agency for Research on Cancer ( IARC ) has classified this bacterium as a Group 1 carcinogen . H . pylori infection is considered to be the strongest known risk factor for gastric cancer , and epidemiological studies have estimated that , in the absence of H . pylori , 75% of gastric cancers would not occur [21] . Pathogenesis associated with H . pylori infection is determined by interactions between bacterial factors and host cells . The most well characterized bacterial virulence determinants are the vacuolating cytotoxin A ( vacA ) and the cag pathogenicity island ( cag PAI ) . The cag PAI is a 40 kb region of DNA that encodes a type IV secretion system ( T4SS ) that forms a syringe-like pilus structure used for the injection of a bacterial protein CagA ( cytotoxin-associated gene A ) into gastric cells . Following the delivery , intracellular CagA is localized to the plasma membrane and triggers complex alterations of the host signaling pathways [22] , including activation of cellular oncogenes ( Fig 2 ) . CagA itself functions as an oncoprotein . In laboratory tests , CagA promoted anchorage-independent growth and , when transgenically expressed in mice , led to spontaneous development of gastrointestinal and hematopoietic neoplasms [23 , 24] . Oncogenic potential of CagA has also been demonstrated using Drosophila and zebrafish experimental models [25 , 26] . H . pylori infection results in conditions of cellular stress because the bacteria induce DNA damage and disturb normal cellular homeostasis ( including aberrant activation of multiple oncogenic pathways ) , all of which are conditions that typically activate p53 [27 , 28] . However , initial studies of the p53 stress response revealed that H . pylori is able to dampen activity of p53 protein by inducing its rapid degradation [20] . The ability of H . pylori to suppress the p53 response was also demonstrated when DNA damage was experimentally induced by DNA-damaging agents [20 , 29 , 30] . The bacteria specifically target p53 , as p73—another member of the p53 protein family , which has significant functional and structural similarities to p53—is not down-regulated by H . pylori but rather induced [31] . The ability to induce degradation of p53 varies between H . pylori strains , with CagA-positive bacteria being more potent [20 , 29] . Although CagA likely does not directly bind to p53 , it induces its degradation [29] . Notably , ectopic transfection of CagA is sufficient to inhibit p53 activity and induce its degradation [20 , 30] . Recent studies pointed out a complex nature of CagA–p53 interactions . It was shown that levels and natural variability of CagA protein highly affect p53 degradation [32] . Among other bacterial factors , VacA was also reported to regulate p53 [33–35] . Down-regulation of p53 was found to facilitate autophagy in infected cells [35] . The kinetics of p53 in infected cells in vivo appears to be complex . In infected Mongolian gerbils , which are commonly used for studies of H . pylori infection , expression of p53 was changed in a bimodal fashion , with an accumulation after initial infection that was followed by a rapid down-regulation of p53 protein in gastric epithelial cells . A second peak of p53 was observed later , when gastritis ( inflammation of the lining of the stomach ) developed . These findings led to a hypothesis that , at a certain time , levels of p53 reflect a balance between p53 degradation induced by the bacteria and p53 induction caused by cellular stress [20] . A down-regulation of p53 protein , but not p53 mRNA , was observed in H . pylori-infected mice [36] . In contrast to small DNA tumor viruses , H . pylori takes advantage of host mechanisms normally regulating p53 [20 , 35] . The bacteria enhance proteasomal degradation of p53 mediated by E3 ubiquitin ligase HDM2 by increasing its phosphorylation at serine 166 . An increased phosphorylation of HDM2 was found in gastric epithelial cells co-cultured with H . pylori in vitro and H . pylori-infected animals and humans in vivo [20 , 35 , 37] . Inhibition of HDM2 activity with siRNA or chemical inhibitor Nutlin3 suppresses bacterial degradation of p53 [20 , 35 , 38] . A similar effect can be achieved by inhibition of Akt and Erk kinases , showing that these enzymes mediate phosphorylation of HDM2 protein in infected cells [35 , 38] . Expression of HDM2 was found to correlate with phosphorylated Akt ( pAkt ) in patients infected with H . pylori [37] . In addition to HDM2 , recent studies reported that another cellular E3 ubiquitin ligase , Mule/ARF-BP1 , is involved in degradation of p53 in H . pylori-infected cells [32] . It remains unclear how this enzyme is activated by the bacteria . p14ARF tumor suppressor ( termed p19ARF in rodents and p14ARF in humans ) , which functions upstream of p53 , was found to be a critical modulator of p53 protein stability in infected cells [32] , as ARF inhibits activities of both HDM2 and ARF-BP1 proteins [39–41] . It was shown that cells expressing functional ARF are significantly more resistant to degradation of p53 ( Fig 3 ) . However , when ARF protein levels are decreased due to hypermethylation or deletion of the ink4a/ARF locus , H . pylori efficiently degrades p53 [32] . Loss of ARF occurs during gastric tumorigenesis and can be found in gastric precancerous lesions . Methylation of the p14ARF gene is also increased with age [42] . Given these findings , it was hypothesized that older people with gastric precancerous lesions , who are infected with H . pylori , may be particularly vulnerable to degradation of p53 [32] . Among other cellular factors , ASPP2 protein ( apoptosis-stimulating protein of p53 ) , which normally activates p53 , was identified to regulate p53 in H . pylori-infected cells [29] . Buti et al . showed that binding of CagA protein to ASPP2 results in inhibition of transcriptional and proapoptotic activities of p53 and induction of proteasomal degradation of p53 . Recent studies suggest that bacterial degradation of p53 may contribute to gastric tumorigenesis . It was reported that clinical isolates of H . pylori varied greatly in their ability to degrade p53 , but that , generally , isolates associated with a higher gastric cancer risk more strongly affect p53 when compared to low-risk counterparts [32] . H . pylori inhibits p53 through multiple mechanisms , implying that inhibition of p53 activity is an important factor for successful infection . The bacteria not only induce degradation of p53 , but also alter the expression profile of p53 isoforms [43] . Interaction of H . pylori with gastric epithelial cells , mediated via the cag PAI , induces N-terminally truncated Δ133p53 and Δ160p53 isoforms , which inhibit transcriptional and proapoptotic activities of p53 , resulting in activation of NFkB . Induction of proinflammatory cytokine Macrophage Migration Inhibitory Factor ( MIF ) by H . pylori was suggested to inhibit p53 by decreasing its phosphorylation [44] . It was also shown that H . pylori can facilitate mutagenesis of the p53 gene . Infection with H . pylori leads to aberrant induction of activation-induced cytidine deaminase ( AID ) , which deaminates cytosine residues , leading to accumulation of p53 mutations in gastric tissues [45] . Interestingly , AID and other cytidine deaminases are induced by a number of viruses such as HPV , HTLV-1 , HCV , and others [46–48] . SV40 and influenza A viruses have been shown to affect expression of p53 isoforms [49 , 50] . A new and exciting development in this area is that other bacteria induce degradation of p53 using a similar mechanism to that of H . pylori ( Fig 4 ) . Two research groups have recently reported that the intracellular bacterial pathogen Chlamydia trachomatis , and potentially other Chlamydia species , induces degradation of p53 by activating HDM2 protein [51 , 52] . C . trachomatis is a common cause of bacterial sexually transmitted disease ( STD ) and blinding trachoma . Similar to H . pylori , C . trachomatis activates the PI3K/Akt pathway and increases phosphorylation of HDM2 ( Ser166 ) , leading to activation of HDM2 and proteasomal degradation of p53 . Down-regulation of p53 allows Chlamydia to enhance activity of the PPP that provides bacteria with necessary metabolites , such as nucleotides precursors , and protects against oxidative stress by increasing the cellular NADPH pool [52] . Enforced expression of p53 in infected cells results in strong inhibition of chlamydial growth , while overexpression of glucose-6-P-dehydrogenase , a key enzyme in the PPP that is inhibited by p53 , rescues the bacterial growth . The authors reported that degradation of p53 by Chlamydia interferes with the host’s response to genotoxic stress and may contribute to cancerogenesis in the female genital tract [51 , 52] . Inhibition of p53 through the HDM2-dependent mechanism is also employed by enteropathogen Shigella flexneri , which causes bacillary dysentery in humans . Infection with Shigella is accompanied by strong genotoxic stress and cellular damage [53] . To prevent activation of p53 , Shigella causes rapid degradation of p53 using two distinct mechanisms . During the early phase of infection , the bacterial virulence effector IpgD promotes activation of the host PI3K/Akt pathway and phosphorylation of HDM2 at serines 166 and 186 , causing activation of HDM2 and degradation of p53 . The second mechanism for p53 inhibition comes into play during the late phase of infection . p53 is proteolytically cleaved by the calpain protease system , in which activation is facilitated by the Shigella virulence effector VirA . The VirA activates calpain by promoting proteolysis of the calpain inhibitor calpastatin . Bergounioux et al . suggested that Shigella inhibits p53 to prevent apoptotic cells death that saves energy and preserves its own epithelial niche [53] . Interestingly , not all enteric pathogens inhibit p53 . Activation of p53 was reported in the context of Salmonella typhimurium infection [54] . Outside the Enterobacteriaceae family , down-regulation of p53 protein was reported in studies of Neisseria gonorrhoeae , which is responsible for the sexually transmitted gonorrhea that may increase the risk of genital neoplasms [55] . Similar to the aforementioned pathogens , N . gonorrhoeae causes strong genotoxic stress and induces both single and double strand DNA breaks . The mechanism of p53 down-regulation is not fully understood , but Vielfort et al . reported that the bacteria can inhibit transcription of the p53 gene [56] . Inhibition of p53 may provide certain benefits to bacteria . One particular mechanism that may be targeted by bacteria is the p53 DNA damage response . Inhibition of p53 may allow bacteria to subvert the host cell cycle control and apoptosis mechanisms , resulting in inhibition of cell death and survival of host cells damaged by infection . This is in agreement with the findings of antiapoptotic and prosurvival effects produced by bacterial pathogens , which inhibit p53 [20 , 29 , 52 , 53] . In the case of H . pylori , expression of the CagA virulence factor is sufficient to inhibit p53 and extend short and long term survival of gastric epithelial cells that underwent DNA damage [20] . Besides the DNA damage response , bacteria may also target the metabolic control of p53 . Inhibition of the p53 metabolic regulation may be particularly important for obligatory intracellular pathogens such as Chlamydia . As described above , degradation of p53 allows C . trachomatis to release inhibition of the PPP elicited by p53 . When bacterial degradation of p53 was experimentally inhibited , the development and formation of infectious progeny was blocked , suggesting that metabolic control of p53 provides antibacterial protection . It is possible to draw a parallel between Chlamydiae and viruses since both are obligatory intracellular pathogens , which strictly rely on the host resources . Similar to viruses , inhibition of p53 allows Chlamydia to reprogram the host cell signaling to create a metabolic environment necessary for chlamydial survival and growth . To some extent , this may also be applied to obligate parasitic Mycoplasma bacteria , which inhibit activity of p53 [57] . A more complex picture emerges in regards to the role of the p53 signaling in the context of chronic infections with extracellular pathogens such as H . pylori . One proposed possibility is that inhibition of p53 helps H . pylori to compromise the gastric epithelial barrier , allowing the bacteria to acquire nutrients from the host or get access to the lamina propria . This concept is supported by recent findings showing that H . pylori inhibits activation of p53 induced by disruption of the adherens junctions , which stabilize cell–cell adhesion [38] . It was also suggested that suppression of p53 responses may help H . pylori adapt during the early phase of infection and prevent the host immune response [20] . The p53 pathway is known to affect immune response [58] . Among direct transcription targets of p53 are a number of proteins regulating innate immunity and cytokine and chemokine production . p53 is also known to affect NF-κB activity and pro-inflammatory signaling . Although immunomodulatory function may play a role , there is no direct evidence yet that bacterial inhibition of p53 affects the host immune response . Additional studies are needed to further explore these mechanisms .
Interaction of bacterial pathogens with the host cells induces DNA damage , alters intracellular signaling , and profoundly affects normal cellular homeostasis . To prevent the cellular stress response , which may be detrimental to a successful infection , some bacteria have evolved to inhibit p53 , a key component of the stress response machinery . Bacteria inhibit p53 through multiple mechanisms , including protein degradation , transcriptional inhibition , and post-translational modifications . Current research revealed that p53 has a role in controlling the bacterial infections and that inhibition of p53 may confer certain selective advantages to bacteria . Unfortunately , this may have grave consequences for the hosts , increasing the risk of tumor development . It is particularly relevant to prolonged chronic infections . Initial experiments with inhibition of protein degradation of p53 demonstrate that p53 activities can be restored in infected cells using specific chemical inhibitors . These findings may offer new and exciting opportunities for therapeutic targeting of p53 in infected cells . Future studies of the bacterial regulation of p53 hold the promise of a better understanding of pathogenesis and tumorigenesis associated with bacterial infections .
|
This review focuses on a novel aspect of host–bacteria interactions: the direct interplay between bacterial pathogens and tumor suppression mechanisms that protect the host from cancer development . Recent studies revealed that various pathogenic bacteria actively inhibit the major tumor suppression pathway mediated by p53 protein that plays a key role in the regulation of multiple cellular stress responses and prevention of cancerogenesis . Bacterial degradation of p53 was first discovered in the context of Helicobacter pylori infection , which is currently the strongest known risk factor for adenocarcinoma of the stomach . This phenomenon , however , is not limited to H . pylori , and many other bacterial pathogens inhibit p53 using various mechanisms . Inhibition of p53 by bacteria is linked to bacterial modulation of the host cellular responses to DNA damage , metabolic stress , and , potentially , other stressors . This is a dynamic area of research that will continue to evolve and make important contributions to a better understanding of host–microbe interactions and tumorigenesis . These studies may offer new molecular targets and opportunities for drug development .
|
[
"Abstract",
"Historical",
"Perspective",
"of",
"Microbial",
"Inhibition",
"of",
"p53",
"If",
"Viruses",
"Can",
"Do",
"It,",
"Why",
"Can’t",
"Other,",
"More",
"Complex",
"Microorganisms?",
"Summary"
] |
[] |
2015
|
Microbial Regulation of p53 Tumor Suppressor
|
Within the context of a field trial conducted by the Cuban vector control program ( AaCP ) , we assessed acceptability of insecticide-treated curtains ( ITCs ) and residual insecticide treatment ( RIT ) with deltamethrin by the community . We also assessed the potential influence of interviewees’ risk perceptions for getting dengue and disease severity . We embedded a qualitative study using in-depth interviews in a cluster randomized trial ( CRT ) testing the effectiveness of ITCs and RIT in Santiago de Cuba . In-depth interviews ( N = 38 ) were conducted four and twelve months after deployment of the tools with people who accepted the tools , who stopped using them and who did not accept the tools . Data analysis was deductive . Main reasons for accepting ITCs at the start of the trial were perceived efficacy and not being harmful to health . Constraints linked to manufacturer instructions were the main reason for not using ITCs . People stopped using the ITCs due to perceived allergy , toxicity and low efficacy . Few heads of households refused RIT despite the noting reasons for rejection , such as allergy , health hazard and toxicity . Positive opinions of the vector control program influenced acceptability of both tools . However , frequent insecticide fogging as part of routine AaCP vector control actions diminished perceived efficacy of both tools and , therefore , acceptability . Fifty percent of interviewees did feel at risk for getting dengue and considered dengue a severe disease . However , this did not appear to influence acceptability of ITCs or RIT . Acceptability of ITCs and RIT was linked to acceptability of AaCP routine vector control activities . However , uptake and use were not always an indication of acceptability . Factors leading to acceptability may be best identified using qualitative methods , but more research is needed on the concept of acceptability and its measurement .
Acceptability is the perception among individuals , organizations and entities involved in implementation that a given treatment , service , practice , or innovation is agreeable or satisfactory [1] . It is one of eight concepts identified in the literature for labeling and assessing outcomes of implementation processes . Implementation outcomes are preconditions to attain further intervention effectiveness [1 , 2] . Acceptability studies of interventions or of specific tools used within an intervention ( e . g . , drugs , insecticides , vaccines , diagnostic procedures ) are common in the health development field [3 , 4] . Most of these studies tend to take the form of quantitative surveys and often measure acceptability through proxies such as uptake , short-term use , coverage and willingness to pay , among others . Dengue is a vector-borne disease that is mainly transmitted by female Aedes aegypti mosquitos [5] . A mild episode of the disease can evolve to a severe and fatal hemorrhagic illness [6] . The disease is of growing public health importance in tropical and subtropical areas [7] . There is no effective antiviral therapy for the disease and a vaccine is still under research [8–11] . At present , vector control remains the cornerstone of prevention and control efforts [5 , 10 , 12–14] . Ae . aegypti control methods include the application of chemical products , the use of biological agents and environmental management of mosquito breeding sites [15] . However , it has become increasingly difficult to effectively control the vector and to counter expansion of the disease using existing tools [15] . Thus , developing new vector control methods is high on the dengue research agenda [15] . Some new tools ( e . g . , insecticide-treated materials , biological agents and genetic methods ) are already the subject of operational research [16–20] . Some studies have shown promising results of their efficacy on dengue vector densities and potentially on dengue transmission , but are site and uptake dependent [16 , 21 , 22] . In Cuba , a dengue outbreak in Santiago de Cuba in 1997 [23] signaled that the national Aedes aegypti Control Program ( AaCP ) was facing difficulties in sustaining its earlier successes in controlling the vector . Since then , the AaCP has investigated the utilization of new vector control methods to impact dengue transmission and prevent outbreaks [24] . We conducted a qualitative study to assess acceptability of insecticide-treated curtains ( ITCs ) and residual insecticide treatment ( RIT ) by the community for control of Ae . aegypti and dengue . Our study was part of a field test of the tools conducted by the AaCP in Santiago de Cuba . We also explored acceptability of the AaCP and its routine vector control activities , and risk perceptions of getting dengue and severity of the disease to assess their potential influence on interviewee acceptability of ITCs and RIT .
Santiago de Cuba is the second largest city in Cuba and the capital of the province of the same name . It is located in the south-east area of the island and has 513 , 784 inhabitants [25] . Ae . aegypti proliferation is favored by the presence of , on average , four water-storage containers per household [26] , high population density , uncontrolled urbanization , deficient solid and liquid waste management , high temperatures ( 28–34°C ) , and significant rainfall ( 1037 . 9 mm annually ) , among others . Despite ongoing vector control activities , infestation by Ae . aegypti persists , with an average house index for Santiago de Cuba of 2% , which can be substantially higher at the block level , leading to sporadic dengue outbreaks since 1997 [27–29] . A cluster-randomized controlled trial ( CRT ) was conducted in Santiago de Cuba from March 2011 to October 2012 to evaluate the effectiveness of ITCs and RIT to control Ae . aegypti [30] . The tools were distributed or applied by AaCP field workers . During the trial , AaCP routine vector control activities continued in both intervention and control clusters . Routine vector control activities included entomological surveillance , source reduction through periodic inspection of houses , larviciding ( with temephos ) of water-holding containers , selective adulticiding ( fogging with cypermethrine and chlorpyrifos or perifocal residual spraying with deltamethrin ) when Aedes breeding sites or dengue cases were detected , provision of health education information , promotion of community-based environmental management , and enforcement of mosquito control legislation [31] . Fines are applied to individuals who “obstruct enforcement of sanitary measures by the corresponding authority” ( Legislative Decree 272 , enacted February 20 , 2001 ) by not allowing AaCP field workers to enter the premises or do not permit spraying/fogging . Fines range from 100 to 300 Cuban pesos [32]; a significant financial impact on a household as in 2016 , 300 Cuban pesos was nearly half the average monthly income [33] . We conducted a small-scale qualitative study embedded [34] in the CRT conducted in Santiago de Cuba to test ICTs and RIT . ITCs were made from PermaNet polyester netting ( Vestergaard-Frandsen , Switzerland ) treated with a long-lasting formulation of deltamethrin ( 55mg/m2 ) and coated with a protectant ( no details disclosed by the manufacturer ) to prevent degradation of the insecticide when exposed to UV light . The ITCs retain their insecticidal properties and efficacy for about 2 years ( information from manufacturer ) . The curtains were white patterned netting , 1 . 1m wide by 2 . 9m long . Placement of the ITCs took into consideration: Ae . aegypti resting behaviors; perceptions of the heads of households on the areas inside the house with greatest mosquito nuisance; and manufacturer instructions ( i . e . , ITCs should have little to no contact with sunshine; they should not be cut ) . Hanging places were negotiated with the head of household in order to increase acceptability . Heads of households preferred to hang the ITCs by windows or door openings , on closets or the wall ( e . g . , behind the bed ) , or attaching them to existing decorative curtains ( Figs 1 and 2 ) . A maximum of 3 ITCs were distributed per household; which was equal to the number of rooms in a standard house . In public spaces such as schools and offices , more curtains could be distributed . Residual insecticide treatment ( RIT ) was conducted by spraying inside and outside the house every 4 months . K-Othrine 25 WG , supplied by Bayer Environmental Sciences Co . ( 25% deltamethrin formulation ) , was dissolved in water ( 20 grams/8 liters water ) for a concentration of 25mg a . i . /m2 sufficient to treat 200m2 . While duration of the residual activity depends on the type of surface treated , it was expected to last 12 weeks on non-porous surfaces if not manipulated . The insecticide was sprayed on the outside of ground-level water tanks and the walls behind them , and on Ae . aegypti mosquito resting sites ( e . g . , under beds , inside and under closets ) in the intra-domestic area . Before each application round , AaCP workers were trained how to carry out the RIT activities , with support of a short video showing the correct application procedures . Supervisors conducted quality control visits during the application rounds . In-depth interviews were used to explore the following three topics: 1 ) acceptability of the tools , such as appearance ( in the case of the ITCs ) , disturbance ( e . g . , smell ) , perceived efficacy , effects on health , ecological and economic issues , and sources of information about the tools ( e . g . , formal , informal , rumors ) ; 2 ) acceptability of the AaCP , such as perceived efficacy , reputation , use of fines , house inspection ( e . g . , frequency , timing ) , routine control methods and their perceived efficacy; and 3 ) fear of dengue , risk perception of getting sick and of disease severity . Descriptive notes which included observations of interviewees’ facial expressions , tone of voice , and body language during the interview were taken and used for interpretation of the data [35] . The sampling unit was households in the intervention clusters where ITCs had been distributed and RIT rounds had been initiated . As part of the CRT , intervention clusters were randomized to receive only one of the two tools , thus individuals were interviewed about their experience with ITCs or RIT . Households were purposively selected from field trial records , and interviews were conducted four and twelve months after distribution ( ITCs ) or first application ( RIT ) of the tools . Initially the interview categories were households which had: 1 ) refused ITCs or RIT from trial onset , 2 ) accepted ITCs or RIT , 3 ) accepted ITCs with reluctance , or 4 ) stopped use of ITCs or RIT after one year ( Table 1 ) . As no information was available on households which accepted ITCs with reluctance , three user categories were created: 1 ) households which accepted the tools from trial onset , 2 ) households which accepted the tools and stopped using them after a period of time and 3 ) households which did not accept the tools at any time ( YES , YES-NO and NO user categories , respectively ) ( Table 1 ) . While a total of 20 interviews was initially planned , the total number increased during implementation as interviews were conducted until saturation , that is , no new information emerged from the data . Of the 41 households approached for an interview , 38 completed the interview , two refused to participate and one interviewee terminated their interview . The interviews were conducted during two-week periods in September 2011 and April 2012 . Interviews took place in a quiet place within the household , which allowed observing the context where the tools were deployed , verification of actual use of the ITCs and identification of use-related issues such as housing conditions . The interviews were conducted in Spanish by the principal investigator ( DP ) , a sociologist with experience with qualitative inquiry . Interviews were conducted at random times in the morning , afternoon and early evening in order to reach individuals at home during the day as well as men and women who worked outside the home . The person who answered the door and was over 18 years of age was asked to participate in the interview . Those who did not wish to be interviewed were asked to identify another adult member of the household present at the time of the visit to be interviewed . All interviews were audio-taped after informed consent was obtained , and transcribed verbatim by DP and a professional transcriber . Data analysis was deductive , allowing for themes to emerge from the data . Data analysis was conducted with the support of QSR NVivo 9 software ( QSR International Pty LTD , Melbourne , Australia ) . To increase internal validity , a sociologist ( PL ) not involved in data collection , reviewed the data and consistency of the coding system . Additionally , findings from the data analysis were discussed during the study period with the CRT research team , composed of professionals with diverse backgrounds such as epidemiology and vector control . The research protocol was approved by the Research Ethics Committee of the Institute of Tropical Medicine Pedro Kourí , the Infectious Diseases Research Committee of the Cuban Ministry of Health and by the Institutional Review Board of the Institute of Tropical Medicine in Antwerp , Belgium . The Cuban National Vector Control Program ( AaCP ) was provided with verbal and written explanations of the objectives and procedures of the study . Obtaining written informed consent for the interviews was proposed in the initial protocol . However , during the first stage of data collection the written informed consent was replaced by oral informed consent due to fear and resistance of the interviewees to sign an official document . Participants were assured that their data would be confidential and only used for research purposes . Quotations are anonymous to protect participants' identity .
None of the interviewees reported previous experiences with ITCs as the AaCP had not previously used ITCs as a vector control method . Reasons for accepting ITCs were mentioned much less frequently after 12 months of use as compared with after 4 months . Interestingly , interviewees from households classified as NO users also noted reasons for use of the ITCs ( Table 3 ) . Perceived efficacy , not being harmful and attractiveness were given as the main reasons to use ITCs , primarily by interviewees in the YES user category ( n = 6 ) : Both women and men found attractiveness to be an added value for the ITCs , as shown in the following quotations: However , the ITC did not always meet people´s preferences for the appearance and fabric quality of a curtain . A respondent who stopped using the curtains for aesthetic reasons stated: Constraints linked to manufacturer instructions on use of the ITCs ( e . g . , ITCs could not be washed until after six months of use , could not be cut and should have little to no contact with sunshine ) was reported by all NO users as the reason for not accepting ITCs ( Table 3 ) : YES user interviewees also noted reasons for not using ITCs based on their experience , such as allergy , health hazard because of the insecticide , smell , and causing a local skin reaction ( rash ) , among others . The main reasons for removing the ITCs reported by YES-NO users were allergy , toxicity and poor perceived efficacy ( Table 3 ) . Poor perceived efficacy of ITCs was primarily related to two issues: 1 ) lack of information on how the curtains work and 2 ) the combination of different control methods as part of the routine activities of the AaCP . The quantity and quality of information on the ITCs varied by the source of information . As reported by interviewees , information about the ITCs was provided either in community meetings prior to their distribution , by vector control field workers or by other users such as relatives and friends . However , regardless of source , interviewees did not have a clear idea of how the ITCs were supposed to work nor how to ascertain their effectiveness: Some interviewees expected to avoid indoor spraying as an advantage of using the ITCs . That was not , however , the case as illustrated below: There were twelve , seven and one interviewees classified as YES , YES-NO and NO users , respectively . It was difficult to find interviewees who refused RIT at the start of the field trial ( NO user category ) in the trial records . Households classified as YES-NO users were reported in one out of four RIT applications , with the most frequent rejection of RIT being in the third round . Reasons for accepting RIT were mentioned much less frequently after 12 months than after 4 months of application . Perceived efficacy and no disturbances related to the insecticide were most frequently mentioned for accepting RIT , primarily by YES users ( 9 interviewees ) ( Table 4 ) . Perceived efficacy was not only related to mosquitos , but to other insects as well: The only NO user interviewee noted allergy , smell , toxicity , permanent contact with the insecticide , and poor perceived efficacy of fogging as reasons for rejecting RIT . Interviewees in the YES and YES-NO user categories also mentioned previous negative experiences with other residual treatment applications ( Table 4 ) . However , regardless of user category , interviewees revealed a lack of information on the specificities of RIT and confusion with previous experiences with different insecticides applied in a similar manner . In general , interviewees had a positive opinion of the AaCP and accepted the application of all the routine control methods employed by the program . Independent of the specific routine activities considered , most of the interviewees mentioned “usefulness” as the main reason for accepting AaCP interventions , followed by the professionalism of the program staff . Conversely , negative opinions towards AaCP program staff together with household lack of compliance with vector control recommendations were noted as the main reasons for possible rejection of the AaCP , especially by interviewees with RIT application and interviewees in the YES user category ( Table 5 ) . Other reasons why people might reject the AaCP , such as the frequency and hours of field worker visits , lack of privacy and the use of fines , were linked to the modus operandi of the AaCP . Although not numerous , interviewees held strong opinions: Outdoor and indoor fogging with insecticides was strongly identified with the AaCP and the presence of dengue cases or Ae . aegypti breeding sites in the area . The main reason for acceptance or rejection of fogging was perceived efficacy . However , environmental damage resulting from periodic intensified fogging campaigns was highlighted by half of the interviewees: Even those who gave reasons for rejecting indoor fogging accepted it , as it was perceived as mandatory . The following quotation reveals fears of the population to refuse its application: Interviewees who did not mention whether they personally felt at risk for dengue or whether dengue is a serious disease were asked specifically about these two issues . A majority of interviewees stated they perceived a risk for getting dengue and considered dengue to be a severe disease . However , this does not appear to have influenced acceptability of ITC or RIT ( Table 6 ) . In some cases , risk perception was linked to personal and familial experiences with the disease and extra concern for the health of small children: Some YES-NO and NO user category interviewees did not feel at risk of contracting dengue if they refused to use the ITCs or allow RIT , stating that people who follow “hygiene” and “preventive practices” in their households have less chance to get infected . Others felt protected by the application of other vector control methods: It is important to note that there was no awareness on part of the interviewees of the actual risk of dengue transmission as most ( 27 of 38 ) did not know or were not sure about the epidemiological situation in their neighborhood . Interviewees complained about the lack of official information on Ae . aegypti breeding sites identified by AaCP field workers and the presence of dengue cases in their neighborhood or the municipality , as illustrated below:
This study used in-depth qualitative research methods which allowed for the collection of “thick” , detailed data on the factors underlying acceptability of insecticide-treated curtains ( ITCs ) and residual insecticide treatment ( RIT ) . Despite the many reasons provided for actual or potential rejection of ITCs and RIT by the interviewees , it was difficult to identify heads of households who refused to apply the tools or who stopped using them 12 months after distribution ( ITCs ) or application ( RIT ) . In the cluster randomized trial conducted in Santiago de Cuba , RIT coverage was on average 97 . 2% [30] . In a similar trial conducted in Guantánamo Province , coverage reported for ITCs was 98 . 4% [24 , 36] . These high coverage levels could be a contextual issue linked to the fact that both tools were distributed through the Cuban routine vector control program ( AaCP ) . Our results indicate that the AaCP was highly accepted by the community in the intervention clusters in Santiago de Cuba , thus acceptability of the tools cannot be analyzed independently of the acceptability of the program and its routine vector control methods . This is in line with what has been reported in the literature on the importance of intervention distribution channels as a potential confounding factor in the assessment of acceptance or rejection of innovative control tools [37] . The perception of heads of households on the characteristics of the tools played a role in their usage . However , studies have shown that the acceptance of innovative vector control tools is not only determined by their technical attributes [37 , 38] . ITCs were seen as innovative and very different from the control methods that the interviewees were used to , and this favored uptake of ITCs at the start . However , these perceptions affected long-term use as misconceptions on how the ITCs work led to low perceived efficacy as householders expected to observe dead mosquitos surrounding the curtains . A study conducted in Guatemala reported that the greatest perceived benefit of ITCs was that users could clearly see dead mosquitos beneath the curtains [39] . Follow-up surveys at 9 and 12 months post-ITC distribution in Iquitos , Peru reported a high percentage of participants who found ITCs effective in reducing mosquitos [40] . Likewise , a study conducted on the use and acceptance of screens made with netting ( long-lasting insecticidal nets , LLINs ) fitted to doors or windows in Mexico , found that the most notable benefit reported by users was the reduction in the amount of mosquitos at home , linked to a perceived reduction in mosquito biting [41] . Indeed , perceived efficacy has been reported as a strong motivation to use insecticide-treated materials [37 , 40 , 42 , 43] . Most of the referenced studies used curtains or screens made with long-lasting insecticidal netting only on doors or windows [39 , 41 , 44] . In Peru , locations for placing the ITCs were suggested to the research staff by heads of households [45] . Similarly , in our study hanging places were negotiated with each head of household , with walls considered to be the primary place to hang the ITC . This could be a reason why “limited ventilation during the day time” was not mentioned among reasons for ITC rejection , as reported by Rizzo and colleagues [39] . In Cuba , insecticides targeting adult Ae . aegypti mosquitos have been widely and frequently applied by AaCP as a routine control activity for dengue prevention . From 1981 to 1986 , malathion and fenthion were used for fogging and perifocal residual spraying , respectively . From 1986 to date , pyrethroids ( cypermethrin and lambda-cyhalothrin ) have been used when Ae . aegypti positive breeding sites or dengue cases are detected ( e . g . , Santiago de Cuba in 1997 and 2006 , and Havana 2001–2002 ) . Other products such as bendiocarb have also been occasionally used [46–48] . The continuous use of insecticides as a core AaCP control method could be the reason why interviewees in the RIT intervention cluster considered it to be a routine AaCP control program activity , and perceived use of the tool as compulsory . Although in Cuba RIT is indeed compulsory during dengue outbreaks , it is not routinely used during interepidemic periods . Thus , even though all but one of the 20 interviewees accepted RIT , we cannot ignore the fact that interviewees also provided many reasons for rejection of residual insecticide treatment in and around the household . Hence , acceptability of RIT may be influenced by factors such as enforcement of mosquito control legislation ( i . e . , fines ) and social pressure in a setting where avoiding dengue transmission is a high priority for the government and health authorities . Therefore , we cannot determine whether RIT was acceptable or not to the community as a dengue prevention tool . RIT has been strongly recommended and widely implemented for malaria control [49] , and a majority of national malaria control programs in Africa still rely on RIT [50] . It has been considered effective conditional on population willingness to accept its application , with acceptance varying according to location [49] . In some settings , political and social factors also became important drivers for acceptance that overrode factors that could motivate non-acceptance , as seen in a study on indoor residual spraying against malaria conducted in Manhiça district , rural Mozambique [51] . From an epidemiological point of view , continuation of routine vector control program activities , including frequent insecticide fogging in both intervention and control clusters , might not bias the results of the CRT in terms of effectiveness . However , it could have diminished perceived efficacy of both tools and , thus , acceptability for some heads of households . Indeed , intensive fogging could have led to low insect abundance , in spite of the fact that the intervention clusters were areas with the highest Ae . aegypti infestation levels according to baseline entomological surveillance data . This could have implications for integrated vector management strategies advocated in the literature for long-term sustainable vector control [5 , 15] as the population may not be aware of or understand the technical reasons ( e . g . , resource availability , multi-disease control , increased effectiveness ) for which a combination of approaches , methods and tools are being used . Rizzo and colleagues reported that the acceptance of insecticide-treated materials ( i . e . , ITCs and drum screens ) was particularly high in families who had experienced dengue [39] . Unlike other Latin American countries [52] , in Cuba there is greater awareness of the implications and severity of dengue . However , we did not find an association between risk perception of getting sick and perceived severity of disease with acceptability of the vector control tools . This could be due to study limitations discussed below ( e . g . , sample size ) and the presence of confounding factors specific to Cuba . In Cuba , dengue is not considered to be endemic , the outbreaks that occur are rapidly controlled with few deaths [53 , 54] , and information , education and communication ( IEC ) activities on dengue and Aedes aegypti decrease during inter-epidemic periods [54] . As shown in the present study , interviewees’ knowledge of the entomological and epidemiological situation in their neighborhood was very low; and risk perception of getting sick and perceived severity of disease were mainly driven by prior personal experiences with dengue . Over the last three years , Latin America has confronted several outbreaks of viral diseases transmitted by Ae . aegypti; in particular the emergence of chikungunya in 2014 and Zika in 2015 [55] . In the particular case of Zika , even if disease risk is primarily focused on pregnant women and women of reproductive age , the threat of Zika is perceived by the general population as greater than that of dengue [56] . The potential influence of perceived risk of Zika on the acceptability of ITCs and RIT could be an important topic for future research in Cuba . It has been acknowledged in the literature that acceptability of health interventions is a dynamic process , that changes with time and users’ experience [1] . This points to the need , when conducting acceptability studies , to identify the array of factors that could play a role in acceptability and their evolution over time . For this reason , we included a YES-NO category of users in the study and explored acceptability after 4 and 12 months of use or application of the tools . Many acceptability studies are “product biased” , that is , it is often assumed that because the proposed tools are “good” for the health of the population , the beneficiaries will also perceive them as such . Thus , the focus is on how the products can be modified , if necessary , to be more readily accepted . But , as shown in this study , acceptability varies depending upon socio-cultural context . Reasons for rejection of ITCs in Santiago de Cuba included community preferences and expectations of how an “ordinary curtain” should look . This is in line with preferences and cultural connotations of color for insecticide-treated materials reported in the literature [39 , 40 , 57] . Our study findings also highlight the importance of beneficiaries’ values in understanding acceptability , which is especially important when there is rejection of or a strong reluctance towards an intervention . We identified the emergence of ecological values in Cuban society as well as issues of privacy associated with rejection of the AaCP routine vector control actions . Limitations of the study include the purposeful sampling of households and low number of interviewees in some of the user categories; thus , the context of the study should be kept in mind when interpreting the findings . We were not able to compare acceptability of the tools between the two data collection stages or within the three categories of users because of the smaller numbers in some of the categories and stages . This could also have affected the findings in terms of the relationship between acceptability of the tools and risk perception of the disease and disease severity . While purposive sampling does not allow generalization of results in terms of acceptability of ITCs and RIT outside the study area , a strength of the study is that it provided a detailed , in-depth understanding of the array of factors influencing acceptability of ITCs and RIT in this particular context . We found that reasons for acceptance or rejection of the tools were not isolated nor mutually exclusive . Thus , it is important that research themes or categories not be limited to what is already known or assumed by the research team , but include opportunities to identify factors related to the perceptions of the various intervention stakeholders and the socio-cultural context where the interventions or tools are deployed . These factors and the relationships among them could be better identified and explored through the use of qualitative methods in studies conducted preferably before designing quantitative surveys . Finally , the issue of the appropriateness of using proxy indicators for measuring acceptability is one that needs to be better addressed in the literature . Significant efforts are made , especially in cluster randomized trials ( CRTs ) , to reach high coverage and uptake levels in order to demonstrate intervention efficacy . However , the common assumption that there is congruence between reasons for using/not using and acceptance/rejection of interventions is an over simplification . We found RIT YES users who did not seem to really accept the RIT applications as well as ITC NO users who could potentially find ITCs acceptable . In order to avoid mistaken conclusions on what makes a health intervention acceptable or unacceptable for the beneficiaries , more research is needed on the concept of acceptability and its measurement .
|
We aimed to understand what makes insecticide-treated curtains ( ITCs ) and residual insecticide treatment ( RIT ) with deltamethrin acceptable or not to users of these tools . In-depth interviews were conducted as part of a field trial conducted by the Cuban vector control program ( AaCP ) to test the effectiveness of these tools in Santiago de Cuba . Perceived efficacy was the main reason for interviewees who accepted the tools . Constraints linked to manufacturer instructions were the main reason for not using the ITCs when offered at the start of the trial . People stopped using the ITCs due to perceived allergy , toxicity and low efficacy . Few heads of households refused RIT despite identifying various reasons for rejection , such as allergy , health hazard and toxicity . Positive opinions of the Cuban vector control program influenced acceptability of both tools . On the contrary , perceptions of dengue risk did not appear to influence acceptability of ITCs or RIT . Our findings add on the importance of the growing body of qualitative research assessing acceptability of health interventions .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"smell",
"invertebrates",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"immunology",
"geographical",
"locations",
"social",
"sciences",
"neuroscience",
"animals",
"toxicology",
"north",
"america",
"research",
"design",
"toxicity",
"clinical",
"medicine",
"infectious",
"disease",
"control",
"caribbean",
"insect",
"vectors",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"agrochemicals",
"aedes",
"aegypti",
"qualitative",
"studies",
"disease",
"vectors",
"insects",
"agriculture",
"arthropoda",
"insecticides",
"people",
"and",
"places",
"cuba",
"mosquitoes",
"psychology",
"allergies",
"eukaryota",
"clinical",
"immunology",
"biology",
"and",
"life",
"sciences",
"species",
"interactions",
"sensory",
"perception",
"organisms"
] |
2018
|
Insecticide treated curtains and residual insecticide treatment to control Aedes aegypti: An acceptability study in Santiago de Cuba
|
To determine the spatial and temporal dynamics of influenza A virus during a single epidemic , we examined whole-genome sequences of 284 A/H1N1 and 69 A/H3N2 viruses collected across the continental United States during the 2006–2007 influenza season , representing the largest study of its kind undertaken to date . A phylogenetic analysis revealed that multiple clades of both A/H1N1 and A/H3N2 entered and co-circulated in the United States during this season , even in localities that are distant from major metropolitan areas , and with no clear pattern of spatial spread . In addition , co-circulating clades of the same subtype exchanged genome segments through reassortment , producing both a minor clade of A/H3N2 viruses that appears to have re-acquired sensitivity to the adamantane class of antiviral drugs , as well as a likely antigenically distinct A/H1N1 clade that became globally dominant following this season . Overall , the co-circulation of multiple viral clades during the 2006–2007 epidemic season revealed patterns of spatial spread that are far more complex than observed previously , and suggests a major role for both migration and reassortment in shaping the epidemiological dynamics of human influenza A virus .
Intensive study of the molecular evolution of influenza A virus has provided important insights into its seasonal genesis and spread in human populations [1]–[4] . The rapidity with which both epidemics and pandemics of influenza A virus arise and spread globally has also generated great interest in understanding the spatial-temporal dynamics of this important human pathogen [5]–[9] . Phylogenetic trees of the epitope-rich HA1 domain of subtype H3N2 influenza A viruses sampled since its emergence in 1968 exhibit a distinctive ‘cactus-like’ pattern , in which most lineages go extinct within a few years of their genesis , so that usually only a single lineage persists between seasonal epidemics [10] , [11] . This is most likely the result of strong host-mediated selection pressure , resulting in continual evolution at key antigenic sites , a process termed ‘antigenic drift’ [11] , [12] . This antigenic evolution is also episodic , with major changes in antigenicity occurring with a periodicity of approximately 3 years [13] . A variety of epidemiological and evolutionary models have been developed to explain this phylogenetic pattern [14] , [15] , and how the evolution of the HA1 domain relates to that in the rest of the viral genome [16] . Although antigenic drift is clearly a key determinant of influenza A virus evolution , this process has rarely been observed in a single locality over a single epidemic season [17] , [18] . Rather , multiple viral introductions appear to drive evolution at the scale of local epidemics , allowing for the co-circulation of multiple clades of the same subtype [16] , [18] . At a global scale , viral migration from regions characterized by more persistent influenza transmission , notably East and South-East Asia , appears to be important in determining large-scale epidemiological patterns [19] , [20] , [21] . In addition , reassortment events between viruses of the same subtype occur frequently , and are sometimes associated with major antigenic changes in both the A/H3N2 [22] and A/H1N1 subtypes [23] . However , a complete understanding of the evolutionary and epidemiologic dynamics of influenza A virus at all spatial and temporal scales remains an important goal [24] . Every winter , epidemics of human influenza recur in the United States , and are associated with an annual average of 226 , 000 hospitalizations and 36 , 000 deaths , mainly caused by secondary bacterial pneumonia in the elderly and young children [25] , [26] . Epidemiological models have found a strong correlation between the regional spread of influenza virus infection in the United States and the movement of people to and from their workplace [9] . In addition , US influenza epidemics tend to originate in California , which may reflect this region's interconnectivity to Asia and Australia [9] . Although of great importance , most spatial models have utilized mortality cases due to pneumonia and influenza ( P & I ) and hence do not consider the evolutionary history of the viruses involved . Indeed , it is striking that detailed phylogenetic analyses of influenza A viruses from a single season at a national level have not been undertaken , even though the rapid rate of influenza A virus evolution [14] , [27]–[29] means that viral genome sequences may contain important information on country-wide spatial dynamics . Our goal here is to determine the spatial-temporal dynamics of influenza A virus during a single epidemic season ( 2006-2007 ) in the United States through the phylogenetic analysis of whole-genome sequence data . Since the 1968 pandemic , A/H3N2 viruses typically dominate most influenza seasons , including 16 of the past 20 US epidemics ( [30] , for example ) , and are associated with higher levels of morbidity and mortality [31] , higher rates of evolutionary change [14] , and greater synchrony in the timing of local epidemics across the United States than A/H1N1 viruses [9] . However , during the 2006–2007 US influenza epidemic , more viruses reported by the CDC were of the A/H1N1 ( 62 . 3% ) than the A/H3N2 subtype ( 37 . 7% ) [30] . The evolutionary dynamics of this epidemic were particularly complex , including a late-season switch in dominance from the A/H1N1 to the A/H3N2 subtype , the co-circulation of multiple antigenically distinct lineages within both A/H1N1 and A/H3N2 , an A/H3N2 vaccine mismatch , and the co-circulation of adamantane resistant and sensitive viral lineages in both subtypes [30] , [32] . Our analysis of 353 whole-genome influenza A virus sequences of both the A/H1N1 ( n = 284 ) and A/H3N2 ( n = 69 ) subtypes from this 2006–2007 US season represents the first attempt to investigate the spatial-temporal spread of a nationwide influenza virus epidemic within the context of genomic-scale evolutionary dynamics .
Our phylogenetic analysis of 284 whole-genome A/H1N1 influenza viruses sampled between December 2006 and March 2007 in 17 US states revealed substantial genetic diversity for all eight segments of the viral genome . In particular , eight phylogenetically distinct clades ( denoted A–H ) , defined by both high bootstrap values and long branch lengths , are evident on the trees of each genome segment , as exemplified by the HA phylogeny ( Figure 1 ) . The phylogenies of the seven other genome segments contain clades identical to those on the HA phylogeny ( Figures S1 , S2 , S3 , S4 , S5 , and S6 , with the PB1 phylogeny presented in Figure 2 ) . Previous studies [20] , [22]–[23] suggest that each clade is likely to represent a separate introduction of the virus into the United States , although the small sample of sequences available mean that individual clades may sometimes represent multiple introduction events . One clade , herein denoted clade A , was clearly dominant , as it comprised the majority of isolates ( 175/284 isolates , 61 . 6% , Table 1 ) . Minor clades B , C , D , E , F , G , and H contained only 47 , 12 , 35 , 6 , 6 , 1 , and 2 isolates each , respectively ( Table 1 ) . Clade A was the most geographically and temporally pervasive of the eight clades , circulating in 24/30 localities and 14/15 weeks studied , although allowall clades were sampled over wide temporal and geographic scales ( Table 1 , Figures 3 , 4 ) . Notably , there was no association between the phylogenetic positions of isolates and their week of collection ( Figure 3 ) or geographic region ( Figure 4 ) . Rather , clades co-circulated in both time and space , with small clades that are detected in only a single region ( E , G , and H ) to likely be an artifact of limited sampling . The largest clades A and B were highly geographically dispersed , containing isolates collected from both relatively isolated areas and major US cities spanning all six US regions , including 24 and 18 out of 30 localities sampled , respectively ( Table 1 , Figure 4 ) . However , in contrast to a simplified spatial model in which a single lineage spreads in a unidirectional manner , we observed no strong signal for viral migration among the co-circulating clades , even when individual clades were studied in isolation ( Table 2 ) . Indeed , a parsimony-based analysis in which the US state of origin of each isolate is coded as an extra character and mapped onto each ML tree revealed a strong clustering by US state ( p<0 . 001 ) , but only weak evidence for movement among states ( data not shown; available from the authors on request ) . The number of isolates collected from different US localities varied widely ( ranging from 1 isolate from Detroit , Michigan to 42 isolates from Houston , Texas , Table 1 ) , and such geographical biases in our data had a profound effect on spatial patterning . Accordingly , the number of clades identified in a locality was strongly associated with the number of isolates sampled from that locality ( Spearman rho = 0 . 77 , P<0 . 0001 ) , while the population size of each locality was not associated with the number of viruses or clades identified ( P>0 . 69 ) . In addition , the first virus isolated in our A/H1N1 sample was from Cincinnati , Ohio ( Table 2 ) , likely an artifact of the relatively large sample collected from this city ( 30 isolates , Table 1 ) . The peak in A/H1N1 genetic diversity occurred during early February ( corresponding to week 10 , Table 2 ) , with six of the eight clades co-circulating during this week . Geographically uneven sampling also meant that the most clades were detected in the most heavily sampled localities . For example , six clades ( A , B , C , D , E , and F ) were detected in Houston , Texas , the most intensively sampled locality ( Table 1 ) . Extensive genetic diversity was also detected within a single week: as a case in point , at least four clades ( A , B , D , F ) , representing two major antigenically distinct lineages circulating globally ( see below ) , were all present in Houston , Texas during week 10 ( Table 2 ) . Abundant viral diversity was also detected in localities that contributed relatively few ( 6–14 ) isolates , including both urban and remote areas . Three different clades circulated in all of the following localities: Los Angeles , California ( clades A , C , F ) ; Denver , Colorado ( clades A , C , G ) ; New York City , New York ( clades A , C , H ) ; Tullahoma , Tennessee ( clades A , B , D ) , Aberdeen , Mississippi ( clades A , D , F ) ; and Weber City , Virginia ( clades A , B , D ) ( Table 2 ) . In fact , more than one clade was observed in every locality from which >1 viral sample was obtained ( Table 2 ) . To view the phylogenetic relationships among A/H1N1 clades from the 2006–2007 epidemic in a wider geographical context , we included 48 background A/H1N1 influenza viruses sampled from the northern and southern hemispheres between 2001–2006 , years that were dominated by viruses antigenically similar to A/New Caledonia/20/1999 ( ‘New Caledonia-like’ ) [30] . These sequences were available for the HA and NA segments , including three antigenically distinct influenza vaccine reference strains selected for 2006–2007 ( A/New Caledonia/20/1999 ) , 2007–2008 ( A/Solomon Islands/3/2006 ) , and 2008–2009 ( A/Brisbane/59/2007 ) ( Figure 1 ) [30] , [33] , [34] . Of the eight clades that co-circulated during the 2006–2007 season , five ( A , B , C , D , and E ) appear to be descendents of New Caledonia-like viruses from 2002–2005 , ( Figure 1 ) , while three ( F , G , H ) are separated from all other isolates by a very long branch with high ( 100% ) bootstrap support ( Figure 1 ) . Due to their extensive phylogenetic divergence , we define clades F , G , and H as ‘set 2’ clades , in contrast to the ‘set 1’ clades A , B , C , D , and E . We inferred the antigenic characteristics of these eight clades based on their phylogenetic relationships and the number of amino acid differences at antigenic sites in the HA from vaccine reference strains of known antigenicity . Accordingly , set 1 clades A , B , C , D , and E are likely to be New Caledonia-like in antigenicity , given that ( a ) 90% of A/H1N1 viruses from this US epidemic were New Caledonia-like ( as characterized by the CDC surveillance [30] ) and set 1 clades were most prevalent , ( b ) set 1 clades are phylogenetically related to other New Caledonia-like viruses from 2002–2005 ( Figure 1 ) , and ( c ) set 1 clades differ by only 3–5 amino acids from A/New Caledonia/20/1999 , 1–2 of which occurred in antigenic or potential glycosylation sites , versus 10–14 amino acids , 3–5 in antigenic sites for set 2 clades ( Figure 5 , Table 3 ) . It is possible that clades C and D represent additional antigenic variants of New Caledonia-like viruses , given the higher number of amino acid changes in antigenic sites ( 2 ) also observed in these viruses ( Figure 5 ) . However , given the uncertainties involved in inferring antigenic properties from genetic data alone , our antigenic assignments should not be considered definitive . In contrast , set 2 clades F , G , and H appear to be related to two emerging antigenic variants . Clade H may be antigenically similar to the A/Solomon Islands/3/2006 vaccine strain selected for 2007–2008 , based on their close phylogenetic relationship ( Figure 1 ) and the low number of amino acid differences in antigenic sites ( 1 site , Table 3 ) . Clade F is more phylogenetically related to the A/Brisbane/59/2007 2008–2009 vaccine strain , and there are no differences at antigenic sites in these viruses . Clades F and G differ by nine amino acids in the HA , but only one difference occurs at an antigenic site , suggesting that , although phylogenetically distinct , clade G may also be A/Brisbane/59/2007-like in antigenicity . Also of note was the observation that of the 284 A/H1N1 influenza viruses sequenced in this study , only one isolate–A/Colorado/UR06-0053/2007–the sole member of clade G , contained the S31N amino acid replacement in the M2 protein that is associated with resistance to the adamantane class of antivirals ( Table 4 ) [35] . Although clade F was classified as a member of clade set 2 due to the phylogenetic relatedness of its HA gene segment to clades G and H , this clade in fact appears to be set 1-set 2 reassortant . Specifically , on trees inferred for the PB2 , PA , HA , and NA segments , clade F isolates are related to Solomon Islands-like set 2 clades G and H ( as exemplified by the phylogeny of the HA gene segment , Figure 1 ) . However , clade F is more closely related ( with high bootstrap support ) to the New Caledonia-like set 1 clades of A , B , C , and D in segments PB1 , NP , M , and NS ( as exemplified by phylogeny of PB1 gene segment , Figure 2; see Figures S1 , S2 , S3 , S4 , S5 , and S6 for phylogenies of other segments ) . As half of the genome ( PB1 , NP , M , and NS ) of these reassortant viruses was acquired from set 1-like viruses that began circulating in 2005 , this reassortment event most likely occurred between 2005–2006 . Although fewer A/H3N2 influenza viruses ( n = 69 ) were available for study due to the dominance of the A/H1N1 subtype during the 2006–2007 season , abundant genetic diversity is evident on all eight segment phylogenies , as exemplified by the HA tree ( Figure 6; see Figures 7–9 and Figures S7 , S8 , S9 , and S10 for trees of remaining segments ) . To obtain greater resolution , we also estimated phylogenetic trees that included 104 whole genome A/H3N2 influenza viruses sampled globally from 2003–2006 [36] , as well as the HA and NA sequences from the A/H3N2 components of influenza vaccines selected for the 2006–2007/2007–2008 ( A/Wisconsin/67/2005 ) and 2008–2009 ( A/Brisbane/10/2007 ) seasons [30] , [33] , [34] . The majority ( 60/69 , 87 . 0% ) of the A/H3N2 isolates from the 2006–2007 US epidemic were members of a major clade ( denoted clade ‘a’ ) . On both the HA and NA phylogenies , clade a contains the antigenically novel A/Brisbane/10/2007 isolate selected for the 2008–2009 vaccine , whereas only three 2006–2007 singleton isolates ( i . e . isolates that were phylogenetically isolated; described below ) belong to the clade that contains the 2006–2007 influenza vaccine strain A/Wisconsin/67/2005 , confirming prior observations of a vaccine mismatch ( Figure 6 ) [30] . This A/Wisconsin/67/2005-like clade first emerged in 2005 and represented a class of viruses that were adamantane-resistant due to the S31N mutation in M2; it was termed the ‘N-lineage’ in previous work [36] . This N-lineage is closely related to some 2003 isolates ( previously termed ‘clade B’ [36] ) in 4 of the 8 segment phylogenies ( PB1 , PA , NP , and M ) ( Figures 7 , 8 , 9 , Figure S8 ) , confirming that a 4+4 reassortment event was responsible for the genesis of the N-lineage [36] . As with the N-lineage , all isolates in clade a contained the S31N mutation in M2 that confers adamantane resistance ( Figure 6 ) . In addition to the major clade a , a minor clade of five A/H3N2 viruses , denoted clade b , also circulated during the 2005–2006 season ( Figure 6 ) . Although clades a and b both descend from the adamantane resistant N-lineage , every isolate in clade b contains the adamantane-sensitive serine ( S ) at position 31 of the M2 , indicating that a reversion has occurred . In addition , clades a and b may vary antigenically , as they differ in numerous amino acids in HA , five of which occur in antigenic sites A , B , and C ( amino acid sites 50 , 140 , 142 , 157 , 173 ) and one–site 142–in the HA1 domain that was previously identified as undergoing positive selection [37] . Four singleton A/H3N2 viruses ( labeled s1 , s2 , s3 , and s4 ) also circulated during this season ( Figure 6 ) . Isolates s1 , s2 , and s3 are members of the older N-lineage and possess the associated adamantane-resistance S31N mutation . In contrast , isolate s4 is adamantane sensitive and clusters with other adamantane sensitive isolates , including clade b . The HA of isolate s4 differs from that of the major clade a by 12 amino acids , 8 of which occur at antigenic sites and 2 at previously identified positively selected sites ( amino acid sites 193 and 275 ) [37] ( Table 5 ) . Similarly , the HA of s4 differs from clade b by 12 amino acids , 7 of which occur at antigenic sites and 2 at positively selected sites ( 142 and 193 ) . In contrast , the HA of isolates s1 , s2 , and s3 differs from clade a by only 6 , 2 , and 4 amino acids in 3 , 1 , and 2 antigenic sites , respectively . In sum , as many as four antigenic variants of A/H3N2 influenza virus may have co-circulated this season ( although this will be to be confirmed experimentally ) , each of which is likely to represent a separate introduction event: A/Wisconsin/67/2005-like ( isolates s1 , s2 , and s3 ) , A/Brisbane/10/2007-like ( major clade a ) , clade b , and isolate s4 ( Table 4 ) . Major topological differences between the eight phylogenies of the A/H3N2 virus genome strongly suggest that several reassortment events took place involving multiple clades from the 2006–2007 US epidemic . Whereas the adamantane-resistant clade a and the sensitive clade b both appear to derive from the N-lineage on the trees for the PB2 , PA , HA , NA and NS segments , clade b instead derives from the adamantane sensitive clades from 2004–2005 on the M and PB1 trees ( Figures 7 , 8 ) . This major phylogenetic incongruity strongly suggests that clade b viruses re-acquired sensitivity to adamantane by acquiring an older adamantane-sensitive M segment ( with a serine at site 31 of the M2 gene ) through reassortment . The NP segment also has undergone a major reassortment event , as on the NP phylogeny both clades a and b descend from adamantane sensitive clades , rather than from the N-lineage ( Figure 9 ) . The varying phylogenetic positions of the s4 isolate across the genome also suggest that this singleton virus resulted from multi-segment reassortment ( Figures 6–9 , Figures S7 , S8 , S9 , S10 ) . The s4 isolate is closely related to clade b on phylogenies of the PB2 , PB1 , NP , M , and NS segments ( having reassorted along with clade b in the PB1 and M segments; Figures 7–9 , Figures S7 , S8 , S9 , S10 ) . In contrast , on phylogenies of the PA , HA , and NA segments this virus is divergent from all other clades ( Figure 6 , Figures S7 , S8 , S9 , S10 ) . No clear signal of the geographical spread of A/H3N2 influenza viruses could be detected due to our small sample size . All clades were geographically widespread and a secondary parsimony character mapping analysis again revealed strong population subdivision and weak migration ( results not shown; available from authors upon request ) . The major clade a was present in all thirteen localities in which A/H3N2 viruses were collected , and the five isolates contained in minor clade b were geographically dispersed across both urban and remote areas spanning four of five US regions: Los Angeles , California; Chicago , Illinois; Hopkinsville , Kentucky; Madison , Alabama; New York City , New York; and Houston , Texas ( Table 2 ) . New York City exhibited the most A/H3N2 diversity , as major clade a , minor clade b , and singleton viruses s3 and s4 all were detected , which is remarkable given that only six total A/H3N2 isolates were collected from this locality ( Table 2 ) . In some cases , multiple clades of both A/H3N2 and A/H1N1 viruses co-circulated over restricted spatial-temporal scales . As a case in point , at least two A/H1N1 clades and two A/H3N2 clades circulated during week 12 in Chicago , Illinois , week 13 in Houston , Texas , and week 14 in Los Angeles , California ( Table 2 ) . Considering both A/H1N1 and A/H3N2 isolates together , large amounts of genetic diversity circulated in both urban and remote areas of the US: a total of 8 clades of influenza A virus circulated in Houston , Texas during the epidemic , 7 in New York City , New York , 6 in Los Angeles , California , and 5 clades each in Denver , Colorado , Cincinnati , Ohio , and Hopkinsville , Kentucky ( Table 2 ) .
This study utilized whole-genome sequence data from a surveillance initiative of unprecedented scope and scale that sampled both A/H1N1 and A/H3N2 influenza viruses across the US over the course of a single season through the Influenza Genomics Sequencing Project [38] . Rather than a single viral lineage spreading across the US , multiple lineages of both A/H3N2 and A/H1N1 influenza virus were separately introduced and co-circulated , allowing for reassortment within subtypes and greatly complicating patterns of spatial-temporal spread . Given the extent of genetic diversity observed during this season , obtaining a strong signal for the spatial-temporal pattern of spread of multiple different lineages clearly would entail a large increase in sampling . Substantial antigenic diversity was also observed during the 2006–2007 season in the US , as at least five antigenically distinct types of influenza A virus co-circulated: three antigenically distinct variants of A/H1N1 viruses ( A/New Caledonia/20/1999-like , A/Solomon Islands/3/2006-like , and A/Brisbane/59/2007-like ) , and at least two antigenically different types of A/H3N2 virus ( A/Wisconsin/67/2005-like and A/Brisbane/10/2007-like ) , while clade b and isolate s4 also may represent additional antigenic variants of A/H3N2 virus ( Table 4 ) . However , analyses based on hemagglutinin-inhibition ( HI ) tests are required to confirm the antigenic status of these viruses . Although A/Solomon Islands/3/2006-like viruses and A/Brisbane/59/2007-like A/H1N1 viruses were represented only by minor clades during the 2006–2007 season ( H and F , respectively ) , Solomon Islands-like viruses achieved global A/H1N1 dominance by the start of the 2007–2008 season , and the reassortant clade of Brisbane-like viruses rose to dominance later during the 2007–2008 season [39] . Given that the antigenic evolution of A/H1N1 influenza virus is thought to be slower than the A/H3N2 virus , as reflected by eight consecutive years of dominance by A/New Caledonia/20/1999-like viruses , the rapid emergence of two new antigenic variants of A/H1N1 virus in a single year was particularly notable ( [30] , for example ) . The extensive genetic diversity present in both A/H1N1 and A/H3N2 viruses suggests that multiple introductions of virus have taken place during the 2006–2007 season , particularly as our method of collecting viruses clearly underrepresented areas that are major ports of international travel . As a case in point , further sampling in the Los Angeles and New York City regions , where our study still detected significant diversity even at very low sampling levels , would likely augment the total number of viral lineages detected , including those imported from South-East Asia . By sampling in both metropolitan and relatively isolated areas , our study yielded important information on the geographic distribution of viral genetic variation: namely , that extensive viral diversity , including multiple antigenically distinguishable lineages , disseminated widely across the entire United States during the epidemic , even into relatively remote areas , so that it was not confined to the major cities where the virus is thought to enter . As a particular case in point , even relatively low-density areas or those distant from major metropolitan areas , such as Hopkinsville , Kentucky ( population size ∼30 , 000 ) , harbor significant amounts of both genetic and antigenic diversity , suggesting that influenza viruses of multiple antigenic ( and other phenotypic ) types extensively infiltrate the United States over the course of a single season . However , it is important to note that our analyses cannot exclude that a single co-infected individual could have introduced multiple clades of influenza virus into the United States , as the frequency of co-infection among patients in this study is unknown and represents a key area for further research . Importantly , it is also possible that the 2006–2007 US epidemic was particularly difficult to reconstruct due to the unusual complexity of its evolutionary dynamics , which likely relates to the incomplete dominance of either the A/H1N1 or A/H3N2 subtype . The dynamics of influenza virus epidemics vary greatly on an annual basis , and influenza epidemics that are dominated by the A/H3N2 virus have been associated with higher disease transmission and more rapid spread than milder A/H1N1-dominated seasons , as well as stronger synchrony in timing across the United States [9] . Hence , epidemics that are dominated by a single A/H3N2 clade ( such as the 2004–2005 season [18] ) may exhibit stronger signals of spatial spread , and repeating this sampling effort during an A/H3N2-dominated influenza season potentially could yield a stronger spatial pattern . A sampling scheme that minimizes geographical biases and maximizes the number of samples collected early in the epidemic also could increase the likelihood of obtaining a stronger spatial signal . Additional sequencing of influenza viruses in areas outside the United States is also essential to understand the global context of the diversity that enters the US during a given epidemic . From this , and previous studies [18] , [21] , it is clear that influenza A virus is introduced into the United States multiple times during an epidemic . However , the availability of global sequences , particularly at the genomic scale , is currently inadequate to draw any conclusions about the geographic origins of each viral introduction . It has been suggested that US epidemics originate more frequently in California than other states , due to high interconnectivity with Asia and Australia [9] , but further whole-genome sequencing of viruses from Asia is clearly needed to test this hypothesis . Although the tendency of US epidemics to originate in the relatively warm state of California suggests that human movements are more important than climatic factors in the seasonal onset of influenza virus epidemics , further documentation of the complex spatial-temporal dissemination of the virus over an epidemic is required to elucidate the seasonality of influenza . Additionally , the extent of viral and antigenic diversity and the frequent circulation of minor clades that is detected by intensified surveillance efforts , such as the present study , suggest that much more diversity circulates at a global scale than is identified by routine surveillance . In particular , early detection of minor clades , particularly in the source populations of East and South-East Asia [21] , could improve recognition of emerging lineages and prediction of future dominant strains for vaccine design . Indeed , the antigenically variant A/Brisbane/59/2007 ( H1N1 ) -like reassortant clade F detected in this study may not have been picked up by routine global surveillance until later , as no other publicly available global isolates from 2006 were found within this clade . Our findings also suggest that the genetic diversity of the A/H3N2 virus is substantial even when A/H3N2 is not the dominant subtype , as was the case for most of the 2006–2007 epidemic . A major clade ( a ) , a minor clade ( b ) , a reassortant singleton ( s4 ) , and three singletons ( s1 , s2 , s3 ) that appear to be descendents of the N-lineage [36] all co-circulated during this epidemic . All these clades differed in numerous amino acids in the HA , including those in antigenic and positively selected sites [37] . Both clades a and b , as well as the s4 singleton , were involved in at least three separate reassortment events: ( a ) clade b and singleton s4 ( PB1 and M segments ) , ( b ) clades a and b ( NP segment ) , and ( c ) singleton s4 only ( PA , HA , and NA ) . As a caveat , because our study does not involve plaque-purified viruses , it is theoretically possible that the amplification of segments from different viruses co-infecting a single patient could produce a false signal for reassortment , particularly for those putative reassortment events that involve a single virus ( for example , the s4 singleton ) . However , even with this potential source of bias , the frequency of definitive reassortment events among A/H3N2 clades is striking , especially compared to the single reassortment event observed among the A/H1N1 viruses that dominated this season . This most likely reflects the usually lower prevalence of A/H1N1 , which in turn means a reduced likelihood of mixed infection and hence reassortment . In addition , given the importance of other geographical regions , particularly South-East Asia , in the evolution of the influenza A virus [21] , as well as the fact that A/H3N2 was the dominant subtype in Canada and Europe during this season [33] , the A/H3N2 virus likely circulated at higher levels outside the US , providing greater opportunity for reassortment . Of further interest is why inter-subtype reassortment between A/H1N1 and A/H3N2 viruses is not observed more commonly , despite the apparent co-circulation of both subtypes over both time and space ( Table 2 ) . In this case , it is possible that a virus produced by inter-subtype reassortment has a lower fitness , because the greater genetic distance between the A/H1N1 and A/H3N2 subtypes means that reassortment events are more likely to disrupt essential functional interactions among segments . Finally , the existence of the adamantane-sensitive clade b ( A/H3N2 ) during this epidemic was surprising , given that global resistance to adamantanes among influenza A/H3N2 viruses has increased dramatically in recent years , with more than 95% of A/H3N2 influenza viruses classified as resistant in the previous 2005–2006 season in the US [32] . Even more striking was that most of the genome of clade b isolates was more closely related to the adamantane-resistant clade a than to older adamantane-sensitive clades , indicating that this clade did not evolve directly from adamantane-sensitive viruses as may have been presumed . Rather , clade b viruses re-acquired sensitivity to adamantane by acquiring two segments ( PB1 and M ) from older adamantane-sensitive viruses through reassortment . This finding supports prior conclusions that sensitivity and resistance to adamantane can be acquired through genomic reassortment , rather than by direct selection on the M2 gene for drug resistant mutations [36] .
All viruses were collected as part of a larger 2006–2007 US surveillance effort conducted by Surveillance Data Inc . , in which a total of 610 influenza virus specimens of both type A and type B were obtained from nasal and nasopharyngeal swabs from patients seen with influenza-like illness . At the time of this study , 353 type A influenza virus genomes had been sequenced and were available for study . Fifty-six participating physicians , primarily located at family practices , were recruited from 21 states that were geographically distributed across the US ( AL , CA , CO , FL , IL , KS , KY , MI , MS , NJ , NY , NC , OH , OK , OR , PA , TN , TX , VT , VA , WA ) . Doctors swabbed all patients ≥ one year of age who presented with fever and upper respiratory symptoms from December 1 , 2006 to April 1 , 2007 . An in-office immunoassay rapid test ( Quidel QuickVue Influenza A+B Test ) was used to identify positive influenza samples ( A or B ) , and positive swabs were sent to a reference laboratory in Rochester , New York , to be typed as AH1N1 , A/H3N2 , or influenza B , following growth in Primary Rhesus Monkey Kidney Cells ( RhMK ) culture . RNA was extracted from viruses via automated nucleic acid extraction using the Roche MagNA Pure instrument and was shipped to the J . Craig Venter Institute in Rockville , Maryland , for whole virus sequencing ( methods described previously [38] ) . The parsimony-based MacClade program [44] was used to determine those amino acid changes in both the HA and NA gene segments ( Table S9 ) that occurred between each of the eight clades of A/H1N1 virus from the US , as well as global background viruses from 2001–2005 and A/H1N1 vaccine strains . Changes were also identified in potential glycosylation sites , antigenic regions ( Sa , Sb , Cb , Ca1 , Ca2 ) [45] , and the receptor-binding site [46] . The MacClade program also was employed to identify amino acid changes between clades of A/H3N2 and influenza virus vaccine strains , including those in antigenic sites and at eighteen sites previously identified as undergoing positive selection [37] .
|
This study is the first of its kind to reconstruct the spread of an epidemic of influenza A virus across a single country , in this case the United States . In contrast to a single viral lineage spreading across this country , a phylogenetic analysis of the whole-genome sequences of more than 300 influenza A viruses of the A/H1N1 and A/H3N2 subtypes sampled from the 2006–2007 epidemic season reveals that multiple phenotypically and antigenically distinct viral lineages of entered and co-circulated in the US during this time . Furthermore , the widespread co-circulation of multiple lineages , even in geographically remote localities , allowed for frequent reassortment between influenza A viruses of the same subtype . Through reassortment , a minor lineage of A/H3N2 viruses surprisingly re-acquired sensitivity to the adamantane class of antiviral drugs , and a new A/H1N1 antigenic variant emerged that later became globally dominant . In sum , these results highlight the complexity of the spread of influenza A virus in time and space , and highlight the need for intensified global surveillance involving whole-genome sequence data .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"infectious",
"diseases/viral",
"infections",
"genetics",
"and",
"genomics/microbial",
"evolution",
"and",
"genomics",
"infectious",
"diseases/epidemiology",
"and",
"control",
"of",
"infectious",
"diseases",
"genetics",
"and",
"genomics/population",
"genetics"
] |
2008
|
Molecular Epidemiology of A/H3N2 and A/H1N1 Influenza Virus during a Single Epidemic Season in the United States
|
The antiparasitic agent niclosamide has been demonstrated to inhibit the arthropod-borne Zika virus . Here , we investigated the antiviral capacity of niclosamide against dengue virus ( DENV ) serotype 2 infection in vitro and in vivo . Niclosamide effectively retarded DENV-induced infection in vitro in human adenocarcinoma cells ( A549 ) , mouse neuroblastoma cells ( Neuro-2a ) , and baby hamster kidney fibroblasts ( BHK-21 ) . Treatment with niclosamide did not retard the endocytosis of DENV while niclosamide was unable to enhance the antiviral type I interferon response . Furthermore , niclosamide did not cause a direct effect on viral replicon-based expression . Niclosamide has been reported to competitively inhibit the mTOR ( mammalian target of rapamycin ) , STAT3 ( signal transducer and activator of transcription 3 ) , and NF-κB ( nuclear factor kappa-light-chain-enhancer of activated B cells ) signaling pathways; however , selective inhibitors of those pathways did not reduce DENV infection . Similar to the vacuolar-type H+-ATPase inhibitor bafilomycin A1 , both niclosamide and other protonophores , such as CCCP ( carbonyl cyanide m-chlorophenyl hydrazone ) , and FCCP ( carbonyl cyanide-p-trifluoromethoxyphenylhydrazone ) , effectively reduced endosomal acidification and viral dsRNA replication . Co-administration of a single dose of niclosamide partially decreased viral replication , viral encephalitis , and mortality in DENV-infected ICR suckling mice . These results demonstrate that niclosamide diminishes viral infection by hindering endosomal acidification .
Dengue virus ( DENV ) , which is transmitted by the bite of mosquitoes of the Aedes genus , causes approximately 390 million infections annually [1] . DENV belongs to the genus Flavivirus of the family Flaviviridae with a single-stranded , positive sense RNA genome approximately 11 kb in length [2] . The genome of DENV contains a single open reading frame encoding a polyprotein precursor , which is further cleaved into three structural proteins ( capsid ( C ) , premembrane ( prM ) , and envelope ( E ) proteins ) and seven non-structural proteins ( NS1 , NS2A , NS2B , NS3 , NS4A , NS4B , and NS5 ) , which have roles in the pathogen-host interaction and pathogenesis [3] . Patients with DENV infection are usually asymptomatic . However , 3 to 14 days after the infective mosquito bite , some patients exhibit extreme symptoms , including headache , vomiting , fever , rash , myalgia , and retro-orbital pain . Moreover , some patients further progress to life-threatening severe DENV infection , which is characterized by CNS impairment , multiple organ failure , plasma leakage and severe bleeding ( dengue hemorrhagic fever and dengue shock syndrome ) [4] . To date , there is no effective antiviral drug available for blocking DENV infection . In addition to its function as an anthelmintic drug , niclosamide has been widely reported to confer broad antiviral activity [5] . Zika virus , a member of the Flavivirus genus , has been reported to be inhibited by niclosamide treatment through an unknown therapeutic mechanism [6] . Niclosamide likely inhibits Zika and DENV infection through an undefined targeting of flavivirus NS2B-NS3 protease [7] . Repurposing the application of niclosamide for anti-flavivirus therapy is a proposed strategy . However , niclosamide also confers multifaceted blocking effects on different virus infection as well as tumorigenesis . In Epstein-Barr virus infection , niclosamide suppresses viral lytic replication by inhibiting mTOR ( mammalian target of rapamycin ) activation [8] . Additionally , niclosamide exhibits anticancer activity by blocking the mTOR , STAT3 ( transducer and activator of transcription 3 ) , and NF-κB ( nuclear factor kappa-light-chain-enhancer of activated B cells ) signaling pathways [9 , 10] . Although these molecules have been reported to be involved in viral infection [11] , and targeting mTOR may facilitate DENV replication through autophagy induction [12 , 13] , the potent antiviral effects of niclosamide against DENV infection warrant further investigation . The infectious processes of DENV include viral adherence ( receptor-mediated ) , entry ( endocytosis-mediated ) , fusion and uncoating from endosomes following endosomal acidification , RNA release and replication , protein translation , virion assembly , and release [14] . Targeting these processes is a proposed antiviral strategy . Following endocytosis , endosomal acidification leads to the fusion of the viral envelope protein with the host membrane , facilitating the release of the viral genome [15] . Vacuolar-type H+-ATPase ( V-ATPase ) , the proton-pumping enzyme that generates the low intra-vacuolar pH , is required for DENV endocytosis and infection in vitro [16] . Genetically and pharmacologically targeting V-ATPase effectively retard DENV infection in vivo [17 , 18] . Jurgeit et al . demonstrated that niclosamide acts as a proton carrier which blocks endosomal acidification to inhibit human rhinovirus and influenza virus infection [19] . Using our previous in vitro and in vivo models of DENV infection [17 , 20] , we investigated the possible antiviral effects and molecular actions of niclosamide on blocking DENV infection .
Murine Neuro-2a cells ( ATCC , CCL131 ) , human A549 ( ATCC , CCL185 ) , and baby hamster kidney ( BHK ) -21 cells ( ATCC , CCL10 ) were cultured in Dulbecco’s modified Eagle's medium ( DMEM; Invitrogen Life Technologies , Rockville , MD ) . Aedes albopictus C6/36 cells ( ATCC , CRL1660 ) were grown on plastic in RPMI medium 1640 ( RPMI; Invitrogen Life Technologies ) . All culture media were supplemented with 10% heat-inactivated fetal bovine serum ( FBS; Invitrogen Life Technologies ) , 50 U/mL penicillin and 50 μg/mL streptomycin . DENV2 PL046 , a Taiwanese human isolate obtained from the Centers for Disease Control in Taiwan , was propagated in C6/36 cells . Viral titers were quantified by plaque assay using the BHK-21 cells accordingly [17 , 20] . The following reagents and antibodies were used in these studies: niclosamide , the mTOR inhibitor rapamycin , the STAT3 inhibitor Cucurbitacin I , the NF-κB inhibitor BAY 11–7082 , the V-ATPase inhibitor bafilomycin A1 , protonophores carbonyl cyanide m-chlorophenyl hydrazone ( CCCP ) and carbonyl cyanide-p-trifluoromethoxyphenylhydrazone ( FCCP ) , Hoechst 33258 , dimethyl sulfoxide ( DMSO ) , acridine orange , and mouse mAb specific for β-actin ( Sigma-Aldrich , St . Louis , MO ) ; antibodies against Akt Ser473 , Akt , p70S6K Thr389 , p70S6K , ERK1/2 Thr202/Tyr204 , and ERK1/2 ( Cell Signaling Technology , Beverly , MA ) ; antibodies against dsRNA , DENV NS1 , NS3 , capsid , and E ( GeneTex , San Antonio , TX ) ; polyclonal anti-rabbit Atg8 ( LC3 ) I/II ( MBL international , Nagoya , Japan ) ; goat anti-rabbit IgG conjugated with HRP ( Chemicon International , Temecula , CA ) ; rabbit anti-mouse IgG conjugated with HRP ( Abcam , Cambridge , MA ) ; and Alexa Fluor 488-conjugated goat anti-mouse ( Invitrogen , Carlsbad , CA ) . The animal experiments were performed according to the guidelines of the Animal Protection Act of Taiwan . Protocols according to guidelines established by the Ministry of Science and Technology , Taiwan were approved by the Laboratory Animal Care and Use Committee of National Cheng Kung University ( Approval number IACUC #104062 ) . Seven-day-old ICR suckling mice were inoculated intracerebrally with 2 . 5 × 105 plaque-forming units ( pfu ) and intraperitoneally with 7 . 5 × 105 pfu of DENV2 ( PL046 ) , which was combined with or without niclosamide ( 2 or 5 mg/kg ) treatment . On day 9 post-infection , brain tissue was harvested for the protein assay . Body weight and disease scoring were carried out according to our previous studies [17 , 20] . Cells were resuspended at a concentration of 7 × 104 or 1 × 105 cells/mL in the appropriate medium with DENV ( MOI = 1 ) and incubated for 2 h at 37°C . The cells were then washed once with culture medium and incubated for the indicated times . The presence of viral supernatants was evaluated using plaque assays . BHK-21 cells were plated onto 12-well plates ( 7 × 104 cells/well ) . After adsorption with a serially diluted virus solution for 2 h , the solution was replaced with fresh DMEM containing 4% FBS and 0 . 5% methyl cellulose ( Sigma-Aldrich ) . Five days post-infection , the medium was removed , and the cells were fixed and stained with crystal violet solution containing 1% crystal violet , 0 . 64% NaCl , and 2% formalin . Cell cytotoxicity was assessed using Cytotoxicity Detection kit assays ( Roche Diagnostics , Lewes , UK ) according to the manufacturer’s instructions . Total cell lysates were extracted with a buffer containing 1% Triton X-100 , 50 mM Tris ( pH 7 . 5 ) , 10 mM EDTA , 0 . 02% NaN3 , and a protease inhibitor mixture ( Roche Applied Science , Indianapolis , IN ) . Proteins were separated using SDS- polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membrane ( Millipore Corporation , Billerica , MA ) . After blocking , blots were probed with the indicated antibodies and developed using enhanced chemiluminescence ( Pierce , Rockford , IL ) . Following densitometry-based quantification and analysis using ImageJ software ( http://rsbweb . nih . gov/ij/ ) , the relative density of each identified protein was calculated . BHK-21 cells harboring a luciferase-expressing DENV replicon ( BHK-D2-Fluc-SGR-Neo-1 ) were generated and maintained according to previous studies [17] . Fluorescent DENV was prepared by labeling with Alexa Fluor 594 succinimidyl ester ( AF594SE , Molecular Probes , Invitrogen ) as referred to the previous studies [21] . The labeled viruses were purified using Amicon Ultra-15 PLTK Ultracel-PL Membrane ( 30 kDa ) centrifugal filter units ( Millipore ) to remove excess dye . Cells were washed twice after an inoculation ( MOI = 1 ) with cells for 2 h at 37°C . Cells were visualized under a laser-scanning confocal microscope ( Leica TCS SP5 confocal microscope ( Leica Microsystems , Mannheim , Germany ) and were analyzed using FACSCanto II Flow cytometer ( BD Biosciences , Franklin Lakes , NJ ) . The three-dimensional images reconstructed from a series of confocal images , along with the z-axis of the cells and the analysis of z-stacks , were reconstructed using the Leica Confocal Software . Cells were fixed with 4% paraformaldehyde , permeabilized with 0 . 5% Triton X-100 , and washed twice with ice-cold phosphate-buffered saline . Cells were first probed with anti-dsRNA antibodies [22] and then probed with Alexa 488-conjugated goat anti-mouse IgG . 4' , 6-diamidino-2-phenylindole ( DAPI , 5 μg/mL ) was used for nuclear staining . Cells were visualized under a fluorescence microscope ( EVOS FL cell imaging system , Thermo Fisher Scientific , Waltham , MA ) or analyzed using flow cytometry ( Attune Nxt ) . The mean fluorescence intensity ( MFI ) of the dsRNA was analyzed with ImageJ software . The concentration of IFN-β in the cell-conditioned culture medium was determined using ELISA kits ( PBL Assay Science , Piscataway , NJ ) according to the manufacturer’s instructions . Cells were treated with 5 ng/mL acridine orange ( AO; Sigma-Aldrich ) in a serum-free culture medium for 30 min at 37°C . After being washed with Hank's balanced salt solution twice , cells were visualized under a fluorescence microscope ( EVOS ) . Data obtained from three independent experiments are presented as the mean ± standard deviation ( SD ) . Two sets of data were analyzed by an unpaired Student’s t test . Three or more sets of data were analyzed by one-way ANOVA with Tukey’s multiple-comparison test . Statistical significance was set at P < 0 . 05 .
Niclosamide confers potential anti-flavivirus activity against Zika virus infection by targeting unknown factors [6] . It is hypothesized that treatment with niclosamide inhibits not only Zika virus but also the flavivirus DENV . To verify the antiviral effects of niclosamide against DENV infection , an in vitro cell model of DENV infection was examined for viral protein expression and viral release [17] . The release of LDH was measured to monitor cytotoxicity in A549 and BHK-21 cells , and it was found that treatment with niclosamide at all sub-lethal doses caused minor cytotoxic effects on these cells ( Fig 1A ) . Evaluation of 50% cytotoxic concentration ( CC50 ) on niclosamide-treated BHK-21 cells by using LDH assay showed that the value was less than 10 μM ( S1 Fig ) . Further cytotoxic response as assessed by rhodamine 123-based staining for monitoring mitochondrial membrane potential loss was carried out to exclude the lethal dose of niclosamide used in this study ( S2 Fig ) . Niclosamide effectively blocked viral protein expression ( Fig 1B ) and significantly ( P < 0 . 05 ) retarded viral release ( Fig 1C ) . Additionally , niclosamide showed a value of half maximal effective concentration ( EC50 ) near 10 μM ( S3 Fig ) . Under usage with niclosamide at 5 μM , pre- and co-administration but not post-treatment significantly ( P < 0 . 05 ) inhibited DENV replication ( Fig 1D ) . These results confirm the antiviral effect of niclosamide treatment against DENV infection in vitro . Next , to investigate the possible antiviral actions of niclosamide , the cellular responses and infectious processes during DENV infection were explored . We previously demonstrated the infectivity of DENV in Neuro-2a cells [17] . The cytotoxic effects of niclosamide at sub-lethal doses were monitored ( Fig 2A ) . The Western blot results showing the inhibition of DENV NS3 protein expression in the presence of niclosamide treatment confirmed the antiviral effect of niclosamide in DENV-infected Neuro-2a cells ( Fig 2B ) . To demonstrate the infection efficacy in Neuro-2a cells , we performed fluorescent DENV staining followed by confocal microscopic observation ( Fig 2C ) as well as flow cytometric analysis ( Fig 2D ) . The results showed viral endocytosis at 2 h post-inoculation , which was not retarded by niclosamide treatment . Our findings exclude the possibility of niclosamide blocking viral endocytosis at the early stage of DENV infection . To identify the target of niclosamide underlying its antiviral capacity , the type I IFN response was monitored regarding its potent antiviral effect in response to DENV infection . In DENV-infected A549 cells , as quantified by ELISA , IFN-β production was significantly ( P < 0 . 01 ) increased ( Fig 2E ) . We next examined whether niclosamide treatment enhances IFN-β production to reduce DENV infection . However , the results revealed that niclosamide significantly ( P < 0 . 01 ) decreased IFN-β production ( Fig 2E ) , probably following an early blockade on viral infection . These data imply a role of niclosamide independent of facilitating the antiviral type I IFN response . We next examined other steps of the viral cell cycle by assessing firefly luciferase activity in BHK-D2-Fluc-SGR-Neo-1 cells , and we found that treatment with niclosamide caused neither direct inhibitory effects on replicon-based assay of viral genome translation or replication ( Fig 3A ) nor cytotoxicity in cells ( Fig 3B ) . These results indicate that blocking DENV infection with niclosamide had no direct inhibitory effects on FLuc activity in BHK-D2-Fluc-SGR-Neo-1 cells . Niclosamide confers multiple therapeutic effects for the treatment of cancers , infections , and metabolic diseases by interfering activation of the mTOR , STAT3 , and NF-κB signaling pathways [9 , 10] . We next examined the effects of niclosamide on mTOR activation . A time-kinetic assay revealed the decreased phosphorylation of AKT and p70S6K in niclosamide-treated cells ( Fig 4A ) . These data indicate that treatment with niclosamide causes mTOR inhibition . Due to its action of abolishing the association of raptor ( regulatory associated protein of mTOR ) with mTOR , rapamycin is used as a classical mTORC1 inhibitor [23 , 24] . Treatment with rapamycin effectively deactivated ERK , AKT , and p70S6K ( Fig 4B ) . We next evaluated the effects of mTOR inhibition on DENV infection in this study . Consistent with the previous studies in which rapamycin was reported to promote DENV infection through autophagy [25] , rapamycin treatment enhanced the expression of viral proteins NS3 and NS1 , induced autophagy with LC3II conversion ( Fig 4C ) and significantly ( P < 0 . 001 ) facilitated viral release ( Fig 4D ) . Furthermore , treatment with STAT3 and NF-κB inhibitors blocked neither viral protein expression ( Fig 4E ) nor viral release ( Fig 4F ) , suggesting the independent roles of these possible targeting pathways . Taken together , these results indicate that niclosamide confers anti-mTOR activity , but the anti-dengue activity of niclosamide is mediated through a mTOR-independent pathway . A structure-activity assay designated niclosamide as a protonophore which lowers the cytoplasmic pH to cause mTOR inactivation [26] . In endosomes , DENV requires a low-pH-dependent fusion for infectious genome entry into the cytoplasm . A pH-sensitive dye , AO , was used to examine whether selected drugs neutralize the low pH of endosomes during DENV infection . The results revealed that the low pH of endosomes ( red ) in DENV-infected cells was attenuated by treatment with niclosamide , the protonophores CCCP and FCCP , and the V-ATPase inhibitor bafilomycin A1 ( Fig 5A ) . Cells with niclosamide treatment are shown in green , suggesting that the pH was neutralized and endosome acidification was blocked . To assess the viral uncoating process , a time-kinetic expression of DENV E protein revealed that niclosamide interrupted E protein degradation during the early fusion stage of DENV infection ( Fig 5B ) . Furthermore , DENV dsRNA replication , as detected by immunostaining ( Fig 5C ) , and viral release , as determined by plaque assay ( Fig 5D ) , were significantly ( P < 0 . 05 ) decreased by niclosamide , CCCP , and FCCP . These results reveal that niclosamide causes endosomal deacidification to inhibit dsRNA replication and viral release during DENV infection . To further verify the antiviral effects of niclosamide against DENV infection in vivo , the viral replication , viral encephalitis , and mortality in DENV-infected ICR suckling mice were monitored accordingly [17 , 20] . For this animal study , seven-day-old ICR suckling mice were inoculated with DENV2 by concurrent intracranial and intraperitoneal injections with or without niclosamide ( 2 or 5 mg/kg ) co-treatment ( Fig 6A ) . According to the Western blot analysis of the NS3 and NS1 viral proteins ( Fig 6B ) and the plaque assays for detecting the production of infectious particles ( Fig 6C ) , DENV caused significant infection and replication in mouse brains at 9 days post-infection , and niclosamide inhibited viral protein expression and replication . We next monitored time-kinetic changes in clinical scores , which were graded according to the severity of illness as follows: 0 for healthy; 1 for minor illness , including weight loss , reduced mobility , and a hunchback body orientation; 2 for limbic seizures; 3 for moving with difficulty and anterior limb or posterior limb weakness; 4 for paralysis; and 5 for death , as previously described [17 , 20] . First , DENV infection caused a dramatic loss in body weight in a time-dependent manner; however , niclosamide did not reverse these effects ( Fig 6D ) . A significant increase in clinical scores ( Fig 6E ) occurred in DENV-infected mice compared to mock-infected mice by 8 days post-infection . The survival rate of DENV-infected mice decreased by day 9 post-infection , and all of the mice died by day 10 post-infection ( Fig 6F ) . Co-treatment with niclosamide slightly reduced the DENV-induced disease progression and mortality . These data indicate that a single-dose treatment of niclosamide partly abolished encephalitic DENV infection in our model , which leads to neural impairment following viral replication .
According to our findings , treatment with the antiparasitic agent niclosamide confers antiviral activity , including effects on viral genome release , viral protein expression , dsRNA replication , and viral release in vitro in several DENV-infected cell lines . We also demonstrated that a single-dose administration of niclosamide partly reduces DENV replication in vivo as well as DENV-induced acute viral encephalitis-like symptoms , including progressive hunchback posture , limbic seizures , limbic weakness , paralysis , and lethality . These findings , along with the current study showing that niclosamide confers antiviral activity against replication of the flaviviruses Zika and DENV [7] , we and others demonstrated the potential application of niclosamide treatment for inhibiting DENV infection in vitro and in vivo . In addition to its action against DENV , niclosamide has been demonstrated to be an antiviral agent against severe acute respiratory syndrome coronavirus [27] , human rhinoviruses , influenza virus [19] , Chikungunya virus [5] , EBV [8] , and Zika [6 , 7 , 28] . Repurposing niclosamide as an antiviral agent is therefore proposed . In this examination of the antiviral action of niclosamide , inconsistent with the previous study [5] showing that niclosamide inhibits the entry of the Chikungunya virus into cells , our results showed a minor effect on the endocytosis of DENV in niclosamide-treated cells . The target of niclosamide for blocking viral entry/binding was not further addressed [5] . Furthermore , monitoring the antiviral IFN-β response in DENV-infected cells did not reveal immune enhancement by niclosamide stimulation . In contrast , niclosamide reduced IFN-β production , likely by suppressing DENV infection prior to the antiviral immune activation . Generally , niclosamide has been shown to block glucose uptake , oxidative phosphorylation , and anaerobic metabolism to kill tapeworm [9 , 10] . Additionally , niclosamide can inhibit the Wnt/β-catenin , mTORC1 , STAT3 , NF-κB and Notch signaling pathways . For investigating niclosamide-induced antiviral actions , more validation is needed . DENV replicon BHK-21 cells have been generated to assess the replication of the DENV genome [29]; however , niclosamide did not alter viral translation in DENV replicon cells in this study . Although the replicon cells contained the host and viral factors needed for viral genome replication , our findings revealed that the anti-DENV activity of niclosamide is independent of those factors . Currently , through the screening of 2 , 816 approved and investigational drugs , niclosamide has been identified as a potential viral inhibitor targeting the formation of the NS2B-NS3 protease of the flaviviruses Zika and DENV [7] . Considering that the maturation of DENV particles requires NS2B-NS3-mediated cleavage of the viral precursor polyprotein , the direct-acting antiviral agents , such as small-molecules , diaryl ( thio ) ethers , and cyclic peptides targeting the NS2B-NS3 protease are promising antiviral candidates [30–33] . As shown by Li et al . [7] , three potent inhibitors of the NS2B-NS3 protease , including niclosamide , temoporfin , and nitazoxanide , have been confirmed to inhibit the complex formation of DENV NS2B and NS3 . Moreover , these compounds have been shown to reduce the viral replication of DENV in vitro in human A549 cells . Given the niclosamide-based antiviral properties , targeting DENV NS2B-NS3 , at least in part , could be implemented for reducing viral replication . Niclosamide also exhibits multiple-targeted effects on cellular signaling pathways , such as mTOR , STAT3 , and NF-κB [9 , 10] . Inconsistent with niclosamide , inhibitors of STAT3 and NF-κB did not reduce DENV infection , but the mTOR inhibitor enhanced DENV replication . A recent study showed that niclosamide inhibits 12-O-tetradecanoylphorbol-13-acetate- and sodium butyrate-induced mTOR activation during EBV lytic replication [8] . Mechanistic studies indicate that niclosamide possesses protonophoric activity to dissipate protons from endosomes/lysosomes to the cytosol . The resulting increase in protons effectively lowers the cytoplasmic pH , causing mTOR inactivation [26] . However , inhibiting mTOR followed by autophagic induction facilitates DENV replication , likely by modulating lipid metabolism and promoting cell survival [12 , 13] . In this study , in comparison with the direct-acting mTOR inhibitor rapamycin , niclosamide also inhibited the phosphorylation of AKT and p70S6K but induced LC3 conversion for autophagy . In contrast , rapamycin effectively enhanced DENV replication , indicating an enhanced role of autophagy in DENV infection . Although DENV promotes increased autophagy , niclosamide treatment should inhibit the infectious process of DENV prior to the induction of autophagy-facilitated DENV replication . Endosomal acidification followed by viral RNA release is required for DENV replication [14 , 15] . Targeting V-ATPase , a proton-pumping enzyme , inhibits the viral release from endosomes in vitro [16 , 18 , 34] and decreases DENV infection and neurotoxicity in vivo [17] . As reported by Jurgeit et al . [19] , niclosamide acts as a proton carrier which blocks endosomal acidification . Here , we also provide evidence to confirm the inhibition of DENV-induced endosomal acidification , viral E protein degradation , dsRNA replication , and viral release by treatment with protonophores ( niclosamide , CCCP , and FCCP ) . Retarding the DENV viral genome release by interfering with endosomal acidification could be another strategy utilizing the antiviral capability of niclosamide . Although Li et al . [7] showed that niclosamide treatment blocks the NS2B-NS3 complex formation in vitro , it is speculated that niclosamide-induced endosomal deacidification retards the early process of DENV infection rather than blocking NS2B-NS3 . Using our previous in vivo model of DENV infection showing that DENV causes replication in the brain followed by acute viral encephalitis-like symptoms in mice [17 , 20] , we showed that niclosamide not only reduces viral replication but also partly retards lethality in DEN-infected mice . In this study , to verify the antiviral activity of niclosamide , several DENV-infected cell lines in vitro and a mouse model of DENV infection in vivo were used . Further possible effects of niclosamide on the infectious processes were verified using the appropriate testing systems . It was found that niclosamide , similar to protonophores such as CCCP and FCCP , inhibits endosomal acidification to reduce viral genome release independent of inhibiting the mTOR , STAT3 , and NF-κB signaling pathways and without effects on DENV endocytosis , antiviral IFN response , and viral translation . Co-administration of a single dose of niclosamide partly decreases DENV-induced acute viral encephalitis-like symptoms and mortality . A modified treatment with multiple doses is needed to validate its therapeutic efficacy . Although DENV replication may be blocked by niclosamide in vitro and in vivo , however , concurrent blocking pro-inflammatory and/or neurotoxic factors induced by DENV infection for neuroprotection is also needed for therapeutic consideration against dengue encephalitis . Furthermore , the blockade of endosomal acidification by niclosamide should be examined for its hazard effects on the synaptic activity in the neuronal cells although niclosamide causes endosomal deacidification independent of V-ATPase blockade [35] . In conclusion , together with the results of a recent study [7] , our findings further highlight the repurposing application of niclosamide for antiviral drug development against DENV infection .
|
Dengue and severe dengue cause global health concerns annually . Without antiviral drugs , supportive care is the only treatment option for patients with DENV infection . A current vaccine has been approved for protection against DENV infection; however , the potential risks and challenges associated with the immunopathogenesis of DENV remain unresolved . For anti-dengue therapy , the repurposing of drugs with antimicrobial and anticancer properties is a possible pharmacological strategy . In this study , we evaluated the potential antiviral effects of the antiparasitic drug niclosamide , considering its current pharmacological efficacy against arthropod-borne Zika virus infection . Using in vitro and in vivo models of DENV infection , we demonstrated that one of the therapeutic effects of niclosamide is to significantly target endosomal acidification . Following safety screening , repurposing niclosamide treatment may facilitate the development of anti-dengue drugs in the near future .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"dengue",
"virus",
"medicine",
"and",
"health",
"sciences",
"pathology",
"and",
"laboratory",
"medicine",
"viral",
"transmission",
"and",
"infection",
"pathogens",
"cell",
"processes",
"microbiology",
"toxicology",
"viruses",
"protein",
"expression",
"rna",
"viruses",
"viral",
"release",
"rhinovirus",
"infection",
"molecular",
"biology",
"techniques",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"medical",
"microbiology",
"microbial",
"pathogens",
"stat",
"signaling",
"viral",
"replication",
"molecular",
"biology",
"cytotoxicity",
"molecular",
"biology",
"assays",
"and",
"analysis",
"techniques",
"gene",
"expression",
"and",
"vector",
"techniques",
"endocytosis",
"signal",
"transduction",
"cell",
"biology",
"flaviviruses",
"secretory",
"pathway",
"virology",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"viral",
"diseases",
"cell",
"signaling",
"organisms"
] |
2018
|
The antiparasitic drug niclosamide inhibits dengue virus infection by interfering with endosomal acidification independent of mTOR
|
Treatment failure after therapy of pulmonary tuberculosis ( TB ) infections is an important challenge , especially when it coincides with de novo emergence of multi-drug-resistant TB ( MDR-TB ) . We seek to explore possible causes why MDR-TB has been found to occur much more often in patients with a history of previous treatment . We develop a mathematical model of the replication of Mycobacterium tuberculosis within a patient reflecting the compartments of macrophages , granulomas , and open cavities as well as parameterizing the effects of drugs on the pathogen dynamics in these compartments . We use this model to study the influence of patient adherence to therapy and of common retreatment regimens on treatment outcome . As expected , the simulations show that treatment success increases with increasing adherence . However , treatment occasionally fails even under perfect adherence due to interpatient variability in pharmacological parameters . The risk of generating MDR de novo is highest between 40% and 80% adherence . Importantly , our simulations highlight the double-edged effect of retreatment: On the one hand , the recommended retreatment regimen increases the overall success rate compared to re-treating with the initial regimen . On the other hand , it increases the probability to accumulate more resistant genotypes . We conclude that treatment adherence is a key factor for a positive outcome , and that screening for resistant strains is advisable after treatment failure or relapse .
Tuberculosis ( TB ) is a key challenge for global health [1 , 2] . At present about one third of the global population is latently infected [3] and every year about 1 . 7 million people die of tuberculosis . A large number of patients live in resource-limited settings with restricted access to health-care . It is imperative that standard treatment measures are assessed for their efficacy and reliability . Understanding the driving forces behind therapy failures is challenging . This is to a large extent the case because of the complex life cycle and population structure of TB: The typical sequence of events leading to acute pulmonary tuberculosis occurs as follows [1 , 4–7] . Upon inhalation , TB bacilli reach the pulmonary alveoli of the lung . There they are assimilated by phagocytic macrophages . In most cases the bacteria are being killed continuously by phagocytosis while the cell-mediated immunity develops . More rarely , they may persist in an inactive state , which is considered a latent infection . Infected macrophages may aggregate and form granulomas by recruiting more macrophages and other cell types . Inside granulomas , increased necrosis of macrophages can lead to the formation of a caseous core . In latently infected hosts , an equilibrium establishes where the immune system prevents further growth but the bacteria persist in a dormant state [8 , 9] . However , especially in patients with a compromised immune system , the bacteria may continue or resume growth [4 , 6] . In this case , the bacterial population steadily increases until the granuloma bursts into the bronchus forming an open cavity . Mycobacterium tuberculosis is an aerobic organism and depends on the availability of oxygen to promote its growth . Because the oxygen levels inside macrophages and granulomas are low , the growth rate is reduced [6 , 9–13] . In open cavities , oxygen supply is not limiting anymore and the population size increases rapidly . The extracellular bacteria in the cavities may also spread to other locations in the lung where they are again combated by the dendritic cells of the immune system . Some bacteria can be expelled with sputum and be transmitted to other individuals or they may enter a blood vessel and cause lesions in other organs . The standard treatment is a six-month short-course regimen [1 , 14–17] , consisting of two months of combination therapy with isoniazid , rifampicin , pyrazinamide and ethambutol followed by a continuation phase of four months with isoniazid and rifampicin only [18] . According to tuberculosis treatment guidelines all drugs are taken daily during the first two months . During the following four months isoniazid and rifampicin are administered three times a week with a 3-fold increased isoniazid dose [15] . For patients with previous TB treatments the WHO recommends a 8-month retreatment regimen containing additionally streptomycin [17] . In recent years , the problem of drug resistance has increased in severity due to the emergence and spread of multi-drug-resistant tuberculosis ( MDR-TB ) [19–21] , where MDR-TB is defined as infection by M . tuberculosis strains conferring resistance to at least isoniazid and rifampicin . Resistant TB is assumed to emerge at least in part due to inappropriate treatment or suboptimal adherence to the treatment regimen [22] . Poor compliance has been associated with treatment failure and the emergence of resistance in previous studies [23–27] . Multi-drug-resistance usually develops in a step-wise manner . These steps are thought to include functional monotherapy; either due to different drug efficacies among certain bacterial populations or due to different pharmacokinetics [28 , 29] . Prevalence data of MDR-TB in Europe ( see Fig 1 ) show that patients who have previously received treatment are on average six times more likely to suffer from MDR-TB than patients who are newly diagnosed . There are several possible explanations for this observation . Individuals who are infected with MDR-TB are more likely to have a treatment failure or a later relapse [30–33] , especially if they are not properly diagnosed . These patients could then come under more accurate scrutiny and eventually be reported as MDR-TB patients with previous treatment history . Another more direct possibility is that a considerable fraction of patients who have contracted susceptible TB develop de novo MDR-TB during the first therapy [34] . The goal of this study is to assess the factors that determine the de novo acquisition of drug resistance and to get a better insight in the underlying dynamics . Specifically , we want to study the contribution of imperfect compliance and retreatment regimens . In some areas , second-line drugs are not easily accessible . Moreover , drug-susceptibility tests may not be performed due to the lack of required infrastructure or questionable reliability of patient treatment history [37] . Hence , we assess the impact of a retreatment that is identical to the first therapy as well as a retreatment that follows the WHO recommendation [17] . To achieve this goal we develop a computational model of a within-host TB infection and its consecutive treatment with currently recommended first-line regimens . The model framework encompasses the population dynamics of various M . tuberculosis genotypes with different resistance patterns in three pulmonary compartments as well as the pharmacodynamics and the pharmacokinetics of the drugs that are used for treatment . The aim is to provide qualitative insights into the infection dynamics of tuberculosis . The parameterization is based on the most recent concepts and individual experimental results found in the literature . Given the current lack of a good animal or in vitro model for TB , a computational model , may help to bridge the gaps arising from the inaccessibility of TB in experimental model systems and allow the hypothetical assessment of treatment scenarios , which would be otherwise ethically inadmissible in patient trials . In particular , problems resulting from imperfect therapy adherence can be usefully addressed with a computational model .
Our model describes pulmonary tuberculosis and assesses the emergence of resistance during multi-drug therapy . A graphical illustration of the model is provided in Fig 2 . The model reflects the compartmentalization of the bacteria into three distinct subpopulations as described by Grosset [5]: intracellular bacteria within macrophages ( M ) , bacteria within the caseating tissue of granulomas ( G ) and extracellular bacteria which mostly reside in open cavities ( OC ) . The compartments differ in their maximum population sizes as well as the bacterial replication rates that they allow . The base replication rate r is modified by a factor γ , which reflects the compartment specific conditions that influence the replication rate . Bacteria have a natural density-dependent death rate in each compartment . The constant replication rate and the density-dependent death rate constitute a logistic growth model that was assumed to describe the basic population dynamics . Bacteria also migrate unidirectionally at a rate m from one compartment to another . Offspring bacteria have a certain chance to acquire or lose a mutation that confers resistance to one out of up to five drugs that may be administered during treatment . Every resistance mutation confers a fitness cost which affects the reproductive success of its carrier . This means that the bacterial population inside a compartment comprises of up to 32 genotypes , which differ in their drug resistance pattern as well as their relative fitness . To outline the population dynamics within a single compartment we describe them first in the form of a deterministic differential equation . The dynamical equation is given by dNc , gdt=r⋅γc⋅ωg⋅Nc , g−mc⋅NcKc⋅Nc , g+mc′⋅Nc′Kc′⋅Nc′ , g− ( dc+κc , g ) ⋅Nc , g ( 1 ) Here Nc , g is the number of bacteria of a specific genotype g in a specific compartment c . The parameter r is the base replication rate of M . tuberculosis and γc is a factor , which modifies the replication rate according to the different metabolic activities in each compartment . ωg represents the relative fitness of the specific genotype . mc is the rate with which bacteria migrate to the subsequent compartment . The migration rate is multiplied by the ratio between the total population size Nc and the carrying capacity Kc . This reflects the increased migratory activity that takes place during an acute infection . Nc´ , Kc’ and mc´ correspond to the overall bacterial population including all genotypes of the supplying compartment , its carrying capacity and its migration rate , respectively . The last term reflects the density-dependent death rate dc and the drug induced genotype specific killing rate κc , g . The bactericidal effects of the drugs contribute additively to the killing rate κc , g ( see S1 Text for further details ) . The dynamics of the bacterial population in the model are actually simulated as stochastic processes . For this reason we translated the underlying deterministic differential equations into a corresponding stochastic framework by applying Gillespie’s τ-leap method [38] . The parameter estimates used in this model are whenever possible drawn or derived from experimental results in the literature . To account for the diversity of infection and treatment courses in different patients we allow some parameters to vary within a certain range . Parameters are summarized in Table 1 . The basic growth dynamics rest upon the replication rate and the carrying capacity of the compartments . Based on recent studies [39–41] we assume a maximum bacterial load between 105 and 107 bacteria each for the macrophage and the granuloma compartment and 108 to 1010 bacteria for the extracellular compartment . Under optimal conditions M . tuberculosis has a replication time of 20h , hence we set the maximum replication rate in the model to 0 . 8 d-1 [5] . Every new bacteria cell has at birth the chance to acquire or lose one or multiple resistance mutations and therefore get a genotype , which is different from the mother cell . The frequency of specific resistance mutations and therefore the mutation rate for the main first-line drugs have been first estimated by David in 1970 [42] to be around 10−7–10−10 . However , more recent observations suggest considerably higher frequencies in the order of 10−6 to 10−8 [5 , 43] . A possible reason for this discrepancy between these estimates are varying mutation rates in in vitro experiments compared to the conditions encountered in vivo due to stress-induced mutagenesis mechanisms or variations among strains [44–46] . Furthermore , we assume that mutations only occur during proliferation while mutations during the stationary phase could serve as an additional source of resistance mutations [47] . Therefore , we choose to allow for patients with the more recent higher mutation rates because this will yield more conservative estimates ( see Table 2 ) . Our model incorporates backwards mutations from the resistant to the sensitive phenotype , which also restore the reproductive fitness . However , we consider a reversion to be ten times less likely than the original forward mutation because the occurrence of any additional mutation within a gene to be an exact reversion is more infrequent . When assessing the prevalence of certain genotypes , fitness costs that come with resistance mutations have to be considered . The cost of resistance against anti-tuberculosis drugs appears generally to be low [48–51] . Drug-resistant mutants isolated in patients have even been found to be on par with susceptible wild type strains regarding their infectivity and replicative potential . Since cost-free resistance mutations are rather rare , the high fitness of resistant strains that have been found in clinical isolates [48 , 49] is assumed to arise due to the acquisition of secondary site mutations which minimize the fitness costs ( so-called compensatory mutations ) [50] . However , there is evidence that at least initially newly acquired drug resistance confers some physiological cost [52] . Because our model simulates the de novo acquisition of resistance mutations and because the time frame of a single patient treatment is rather short we assign a small fitness cost to every possible mutation and neglect the counterbalance of fitness costs by compensatory mutations . The effect of administered drugs depends on the pharmacokinetics and pharmacodynamics of these drugs ( see Table 1 ) . Both influence the killing rate κc , g at any given time point during treatment . While pharmacokinetic parameters describe the course of the drug concentration in the target tissue , pharmacodynamic parameters characterize the effect the drugs have at a given concentration . The minimal inhibitory concentration ( MIC ) describes the minimal drug concentration at which bacterial growth is reduced by at least 99% . Additionally , the EC50 describes at which drug concentration the half-maximal effect ( commonly , bacterial killing ) is observed , while the Emax indicates the maximal effect of the drug . These pharmacodynamic parameters are obtained by fitting the drug action model to killing curves found in the literature [53 , 54] ( see S1 Text ) . The specific efficacy of most drugs in the different compartments is typically not quantified . There are several studies that tried to assess the bactericidal activity inside macrophages [55–59] . Unfortunately , these estimates are highly variable and sometimes even contradictory [55 , 58] . In addition to these experimental difficulties , it is possible that the pharmacodynamics of anti-tuberculosis drugs are again different in the human body [60–64] . To reflect this uncertainty we assign compartment efficacies from a range of values which corresponds to the most recent estimates [56–59 , 65–70] . To investigate the role of treatment adherence on patient outcome , we followed disease progression starting with the infection of macrophages until all compartments approximately reached their maximum bacterial load . For each parameter set , we simulate the outcome of 10’000 patients who vary both in their pharmacokinetic and–dynamic characteristics as well as compartmental attributes . Parameters are generally picked from a normal distribution . If only a range is known the parameters are chosen from a uniform distribution . To measure the actual treatment efficacy we let every patient develop an acute tuberculosis infection during 360 days . This allows for the emergence of mutants prior to treatment initiation and provides enough time for the establishment of an equilibrium in the bacterial population composition . After this period we start the standard short course therapy regimen with four drugs being taken daily for two months followed by four months in which just isoniazid and rifampicin are taken three times per week . If the infection is not completely sterilized after the first treatment we schedule a retreatment . Since the model does not cover the possibility of dormant bacteria the population recovers rather quickly after an unsuccessful treatment . Hence , we begin the retreatment 30 days after completion of the previous treatment . After such a time span the population reaches a bacterial load where acute symptoms would be again suspected . If not stated otherwise the retreatment corresponds to the WHO recommendation for retreatments [31 , 71] . The WHO recommendations include streptomycin , which is used together with the original four first-line drugs during the first two months . Afterwards the therapy is being continued for another month without streptomycin and during the last five months only isoniazid , rifampicin and ethambutol are administered . All drugs are being taken daily during the whole retreatment . The 95% confidence intervals ( CI ) of patient outcomes in the figures is calculated by picking the value for a two-sided 95% confidence limit with n– 1 degrees of freedom from a t-distribution table where n is the number of patients . This value is then multiplied with the standard deviation σ and divided by the square root of n . The resulting value is then added and subtracted from the mean to get the actual confidence interval .
The impact of treatment on the net growth rate of wild-type or MDR bacteria differs strongly between compartments ( Fig 3 ) : Before treatment starts , the growth rates in macrophages and granulomas are lower than in the open lung cavities due to hypoxia and a generally adverse environment for bacterial growth in these compartments . Since we assume that the drug concentration immediately reaches the maximum the impact of combination therapy on growth rate is immediately apparent after the administration of the first dose of drugs . In all compartments the drugs are able to keep the wild-type populations from regrowth during the following days . Especially in granulomas pyrazinamide is able to diminish the population over a long period due to its relatively long half-life . MDR-TB is substantially less affected by the combination therapy because only pyrazinamide and ethambutol are effective . This means that in macrophages or open lung cavities the multi-drug-resistant population remains constant at best or is even able to slowly grow . Only in the granulomas where mostly pyrazinamide is active ( see Table 1 ) the loss of effectiveness of isoniazid and rifampicin is less prominent . The compliance of a patient with the prescribed drug regimen is a key factor for a successful treatment outcome . For the assessment of treatment success we monitor for every patient three different nested treatment outcomes . Firstly , we define treatment failure as the incomplete sterilization of the lung at the end of the therapy . Secondly , the emergence of MDR-TB is defined in our simulations as 10% or more [72] of the remaining bacterial population after treatment failure being resistant against at least isoniazid and rifampicin . Finally , emergence of full resistance ( FR ) is defined as 10% or more of the population being resistant against all drugs that were used in the treatment regimen ( either 4 drugs for first treatment or up to 5 drugs for retreatment ) . Adherence in our simulations refers to the probability with which the patient takes the prescribed drugs at any given day . We assume that failure to take drugs on a given day always affects all drugs of the prescribed regimen . In our simulations , the level of adherence has a strong but complex impact on treatment success ( Fig 4A ) . Under perfect adherence the model shows a very low failure rate . However , if adherence decreases the probability for treatment failure increases rapidly . Between 40% and 80% adherence there is also a small fraction of patients that fail treatment due to the emergence of MDR-TB . Furthermore , at these adherence levels the model also shows only limited treatment success . Thus , failure decreases monotonically with adherence while MDR is maximized at intermediate levels . Patients who fail on the first treatment and who undergo retreatment ( Fig 4B ) have a failure rate of 20% at 80% adherence . However , the probability for treatment failure increases to about 50% under perfect adherence . Patients who fail the first treatment despite high adherence may often have disadvantageous combinations of PK/PD parameters , which also decrease their success probabilities during the retreatment . In Fig 4B , 4C and 4D the number of patients per adherence level undergoing retreatment decreases strongly as can be seen from the frequency of treatment failure in Fig 4A . When comparing Fig 4A and 4E , which shows the combined outcome probabilities for both treatments , we see that the retreatment reduces the probability of failure over the upper half of the adherence spectrum . The additional treatment success of retreatment regimens depends on adherence and the addition of streptomycin to the regimen ( Fig 4B ) . In our model , even under perfect adherence the chance of treatment failure remains substantial , and in the majority of patients who fail under retreatment MDR-TB emerged de novo . Furthermore , at suboptimal adherence levels a considerable proportion of patients even carry strains that are not susceptible to any of the five administered drugs . The outcome of retreatment depends crucially on whether MDR was acquired during initial treatment: Because the majority of patients who fail the first treatment do not carry MDR-TB their outcome probabilities for the retreatment are almost identical to the overall cohort of failed patients ( Fig 4C ) . Even though the vast majority of patients who failed the first treatment did not develop MDR-TB , a substantial fraction of patients who also failed the second treatment harbor MDR or FR strains . This occurs due to increased subpopulations of monoresistant bacteria that accumulate during the first treatment and that are by itself not sufficient to be diagnosed as MDR-TB . When comparing patients who are diagnosed with MDR-TB after the first treatment ( Fig 4D ) and those who are not ( Fig 4C ) we see that patients who develop MDR-TB are very likely to fail the retreatment as well . At higher adherence levels the majority of those patients develops full resistance against all five drugs ( Fig 4D ) . When considering the outcome for both treatments combined ( Fig 4E ) it becomes more evident that the addition of streptomycin and the more intense retreatment has a beneficial effect on the overall success rate but patients who also fail the retreatment are more likely to carry multidrug-resistant TB strains . When second-line drugs are not available or susceptibility test are not performed , it may occur frequently that a previously treated patient is retreated with the first line treatment . Our results in Fig 5 show that such a retreatment with the first line drugs has almost no additional treatment success beyond the initial treatment . Patients all across the spectrum of adherence experience treatment failure . The identical first-line retreament only increases the chances for the bacteria to accumulate resistance mutations and leads between 50% to 100% adherence to nearly all uncleared patients harboring MDR-TB or worse . This outcome is standing out when comparing the cumulative treatment success in Fig 5D with the results after the first treatment . While the overall success curve did not change the fraction of MDR-TB patients over a large adherence range increased substantially .
The aim of this study is to elucidate the effects of treatment adherence and retreatment on the emergence of resistance in TB . The model explicitly incorporates the pharmacodynamics and pharmacokinetics of all drugs that are used for standard therapy and the WHO retreatment recommendation . Depending on the compartment in the lung in which the bacteria reside ( macrophages , caseous centers of granulomas or open cavities ) , M . tuberculosis has different stages of infection and drug-susceptibilities . Therefore , we explicitly include these different compartments to be able capture the effect of heterogeneous selection pressure . Because not all of the parameters used in our model have been quantified with high accuracy , we do not claim that the model has quantitative predictive power . Rather , it aims to qualitatively demonstrate the underlying dynamics of a tuberculosis infection . Our results suggest that poor adherence is a major cause for treatment failure . When considering the predicted rates of treatment failure one also has to take into account that our definition of treatment failure is probably rather conservative . We do not include the possibility of remaining dormant bacteria , which might increase the likelihood of treatment failure or relapse . On the other hand , we also neglect the possibility of the infection being contained at a later time point by the immune system , thus probably underestimating the chance of success . It is also noteworthy that even at perfect adherence some patients may have a negative treatment outcome . This is most likely due to a random aggregation of very adverse pharmacokinetic parameters and unfavorable infection attributes in some patients . Such outcomes due to pharmacokinetic variability and despite good adherence have been predicted in an in vitro study [73] . Furthermore , our results show that over a certain range of adherence a small fraction of patients develop MDR-TB . At intermediate adherence these patients also have a low likelihood of being treated successfully . Thus , good adherence to therapy is crucial: Not only does it increase treatment success , it also decreases the probability for the emergence of MDR-TB . According to our model , the WHO recommendation for retreatment is somewhat of a double-edged sword . While at high adherence levels the recommended treatment is able to cure the majority of patients who failed the first line therapy , it also increases the fraction of patients harboring drug resistant strains across almost the whole spectrum of adherence . Previous studies already raised concerns about the possible amplification of resistance [71 , 74–77] . In the WHO treatment guidelines it is recommended that drug susceptibility test results should be taken into account when deciding upon the retreatment regimen [17] . However , the vast majority of patients in our model would probably not have been diagnosed with MDR-TB after the first regimen even though they may still harbor increased subpopulations of monoresistant bacteria . Therefore it is conceivable that many would have been treated with the WHO recommended regimen . A large fraction of patients who failed this retreatment eventually developed MDR-TB . Considering the results from our model further clinical studies are needed which analyze the treatment success rates and the accompanying risks of the standard retreatment regimen . Retreating failed patients with an identical short course therapy leads to poor outcome in our simulations . A lower success rate for MDR-TB patients treated with the standard short-course therapy has been confirmed in a large cohort study [37] . In our simulations it is rare that patients who failed the previous treatment are cured after undergoing the same therapy again provided that adherence remains unchanged . Retreatment with the same regimen only generates more opportunities for single resistant mutants that emerged during the first treatment to accumulate further mutations , thus minimizing the number of future treatment options . These findings are in accordance with previous studies which found a positive correlation between previous treatment and the occurrence of resistance [78–81] . This might be an indicator that de novo resistance on an epidemiological scale occurs at a significant frequency and that the main contributor to the frequency of MDR-TB is not necessarily the mere transmission of such strains . In summary our data show that patient adherence is a crucial component of treatment success . The probably cheapest and most effective way to ensure a positive treatment outcome while also minimizing the risk for the emergence of MDR-TB is to maintain proper patient compliance with the treatment . This supports the Directly Observed Treatment , Short-Course ( DOTS ) strategy of the WHO , which includes healthcare workers or community health workers who directly monitor patient medication . If treatment fails , thorough tests of drug susceptibility of the remaining infecting population , would be of considerable value . According to our results a retreatment regimen including streptomycin has the potential to increase the overall cure rate , but also increases the fraction of patients who carry drug-resistant strains . A common principle of physicians is to “never add a single drug to a failing regimen” [82] this principle is often not followed in retreatment . A preceding drug sensitivity test might show existing drug resistances and the retreatment regimen could be adapted accordingly . Nonetheless the standard retreatment regimen is still superior to a retreatment with the identical first-line drugs . Such a retreatment is unlikely to achieve a higher overall cure rate and dramatically increases the probability for the emergence of MDR-TB , which reduces further treatment options . This shows that a dependable patient treatment history that is available to the responsible health professional is also important before initiating a treatment regimen .
|
Our ability to treat and control acute pulmonary tuberculosis ( TB ) is threatened by the increasing occurrence of multi-drug-resistant tuberculosis ( MDR-TB ) in many countries around the globe . It is not clear whether MDR-TB occurs predominantly due to transmission , or whether there is a substantial contribution due to de novo emergence during treatment . Understanding the underlying mechanisms that are involved in the emergence of MDR-TB is important to develop countermeasures . We use a computational model of within-host TB infection and its subsequent treatment to qualitatively assess the risks of treatment failure and resistance emergence under various standard therapy regimes . The results show that especially patients with a history of previous TB treatment are at risk of developing MDR-TB . We conclude that de novo emergence of MDR-TB is a considerable risk during treatment . Based on our findings , we strongly recommend widespread implementation of drug sensitivity tests prior to the initiation of TB treatment .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"blood",
"cells",
"antimicrobials",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"granulomas",
"drugs",
"immunology",
"tropical",
"diseases",
"microbiology",
"bacterial",
"diseases",
"streptomycin",
"pharmaceutics",
"antibiotics",
"drug",
"administration",
"pharmacology",
"infectious",
"diseases",
"white",
"blood",
"cells",
"animal",
"cells",
"tuberculosis",
"isoniazid",
"drug",
"therapy",
"cell",
"biology",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"macrophages",
"multi-drug-resistant",
"tuberculosis"
] |
2016
|
The Role of Adherence and Retreatment in De Novo Emergence of MDR-TB
|
Ribosome biogenesis is essential for cell growth and proliferation and is commonly elevated in cancer . Accordingly , numerous oncogene and tumor suppressor signaling pathways target rRNA synthesis . In breast cancer , non-canonical Wnt signaling by Wnt5a has been reported to antagonize tumor growth . Here , we show that Wnt5a rapidly represses rDNA gene transcription in breast cancer cells and generates a chromatin state with reduced transcription of rDNA by RNA polymerase I ( Pol I ) . These effects were specifically dependent on Dishevelled1 ( DVL1 ) , which accumulates in nucleolar organizer regions ( NORs ) and binds to rDNA regions of the chromosome . Upon DVL1 binding , the Pol I transcription activator and deacetylase Sirtuin 7 ( SIRT7 ) releases from rDNA loci , concomitant with disassembly of Pol I transcription machinery at the rDNA promoter . These findings reveal that Wnt5a signals through DVL1 to suppress rRNA transcription . This provides a novel mechanism for how Wnt5a exerts tumor suppressive effects and why disruption of Wnt5a signaling enhances mammary tumor growth in vivo .
Cellular growth and proliferation depend critically on ribosome biogenesis and protein synthesis . The synthesis of ribosomes is initiated in the nucleolus through a complex , highly coordinated process that engages roughly 60–80% of the cell’s metabolic energy [1–4] . The transcription of rRNA is carried out by RNA Polymerase I ( RNA Pol I ) , which transcribes 28S , 5 . 8S and 18S rDNA as a single , 47S precursor ( pre ) -rRNA . RNA Pol I does so by working in concert with auxiliary proteins to form a transcription-competent complex . Upstream binding factor ( UBF ) serves as a key component of the RNA Pol I machinery that aids its recruitment to the promoter region of the rDNA gene cassette [3–8] . The deacetylase SIRT7 facilitates this recruitment step through the de-acetylation of PAF53 [9–13] . SIRT7’s association with RNA Pol I at the rDNA transcription unit is therefore crucial to the regulation of rRNA synthesis . Correspondingly , SIRT7 regulation serves as an important means for modulating ribosome biogenesis and cellular proliferation during normal cell growth and in disease states , including cancer [14–16] . Wnt proteins are secreted signaling factors that regulate many physiological processes at the cell and tissue level [17 , 18] . Intracellular Wnt signaling is typically mediated through multi-functional Dishevelled proteins ( DVLs ) , which serve as signaling hubs in multiple subcellular locations [18–21] . Misregulation of Wnt signaling , especially the canonical Wnt/β-catenin pathway , is often associated with oncogenesis [17 , 18] . Wnt5a signaling plays a less well-characterized role in tumor development but is understood to act via non-canonical , β-catenin-independent , Wnt signaling pathways [22 , 23] . In breast cancer , Wnt5a signaling has been shown to exert tumor suppressive effects and loss of Wnt5a expression has been associated with accelerated tumor growth [23–25] . However , the physiologically and pathologically relevant targets in breast cancer downstream of non-canonical Wnt signaling have yet to be fully elucidated . It has been demonstrated previously that numerous growth and oncogenic signals , including epidermal growth factor ( EGF ) and RAS , directly increase rDNA transcription as means of promoting cellular proliferation [26] . We therefore hypothesized that counter-balancing cellular signals may decrease rDNA transcription in order to maintain homeostasis , and that Wnt5a may exert suppressive effects on tumor cell growth and proliferation through such means . In order to test this hypothesis , we set out to examine whether exogenous Wnt5a could impact tumor proliferation in breast cancer cells through the regulation of rDNA transcription . Here , we demonstrate that Wnt5a represses rRNA synthesis in breast cancer cells by promoting nucleolar localization of DVL1 and its subsequent association with rDNA chromatin . Localization of DVL1 to rDNA gene cassette regions leads to a displacement of SIRT7 from the Pol I transcription machinery . Hence , Wnt5a signals through DVL1 to negatively regulate rDNA transcription . We substantiated this conclusion by showing that Wnt5a-null mammary tumor cells display enlarged nucleoli , increased proliferation and increased SIRT7 expression in vivo . Given that nucleolar size and elevated levels of rRNA synthesis correlate with poor prognosis in breast cancer , we posit that non-canonical signaling of Wnt5a through DVL1 and the consequent reduction in rRNA synthesis contributes to the role of Wnt5a in tumor suppression [24–26] .
The primary transcript of rRNA genes is a 47S pre-rRNA pre-cursor , which is rapidly processed into smaller , mature rRNA molecules that comprise key structural and functional elements of the large and small subunits of the ribosome [1 , 3] . Given the short half-life of the 47S pre-rRNA , its abundance serves as a marker of active rDNA transcription [27 , 28] . We therefore assayed rDNA transcription after treatment with recombinant Wnt5a protein by measuring the levels of 47S pre-rRNA , using human estrogen receptor positive ( ER+ ) breast cancer cells ( MCF7 ) . Strikingly , Wnt5a treatment reduced 47S pre-rRNA by ≥50% within 15 min , demonstrating that Wnt5a acts to rapidly inhibit rRNA synthesis ( Fig 1a ) . This effect was abrogated by the soluble Wnt antagonist sFRP1 , confirming the specificity of Wnt pathway signaling [29 , 30] ( Fig 1b ) . Consistent with Wnt5a signaling through a β-catenin-independent Wnt pathway , Wnt5a failed to induce accumulation of nuclear β-catenin ( S1a Fig ) but modestly induced DVL2 phosphorylation ( S1b Fig ) [22 , 23 , 31] . Given these observations we hypothesized that Wnt5a may signal through a non-canonical Wnt pathway to inhibit rRNA synthesis . To test this hypothesis , we performed in situ nuclear run-on assays in which synthesis of nascent RNA transcripts was monitored by incorporation of 5-Fluorouridine ( FUrd ) [32] . Using this assay , we observed that a 15 min treatment with Wnt5a produced a >60% drop in the proportion of cells exhibiting nucleolar FUrd ( as defined by co-localization with the nucleolar marker Fibrillarin ) suggesting that Wnt5a represses rDNA transcription ( Fig 1c ) . As the size of nucleolus typically reflects levels of rDNA transcription , we next asked whether Wnt5a treatment lead to changes in the nucleolar area as detected by AgNOR silver staining [26 , 33–35] . As a positive control , we treated cells with Actinomycin D ( ActD ) , a potent inhibitor of transcription [36] . As expected , ActD treatment caused a substantial reduction in the total area of the nucleoli after 4 hours ( Fig 1d ) . Wnt5a induced a significant decrease in the nucleolar area within the same time frame ( Fig 1d ) . The relative reduction in nucleolar area mediated by Wnt5a was even more pronounced in the triple negative human breast cancer cell line ( TNBC ) , BT549 ( S1c Fig ) , indicating that these effects of Wnt5a are found in other breast cancer cell lines . Examination of the proliferation marker Ki-67 also revealed that Wnt5a treatment reduced cellular proliferation ( Fig 1e ) . Moreover , BT549 cells constitutively expressing exogenous Wnt5a displayed smaller nucleoli and slower proliferation than control cells , as measured by MTT assay ( S1d and S1e Fig ) . Reduced nucleolar areas were also observed in MCF7 expressing exogenous Wnt5a ( S1d Fig ) . These data argue that Wnt5a signaling has a repressive effect on rRNA synthesis that restrains proliferation in breast cancer cells . To investigate intracellular signaling effects of Wnt5a signaling , we next examined the subcellular distributions of endogenous DVL1 , 2 and 3 proteins , in the presence and absence of exogenous Wnt5a expression in MCF7 cells . In control cells , DVL1 exhibited distinctive nuclear and sub-nuclear distributions , and co-localized with Fibrillarin ( Fig 2a ) . In contrast , DVL2 and DVL3 were preferentially cytoplasmic and excluded from nucleoli as determined by the absence of co-localization with the nucleolar markers Fibrillarin and UBF ( S2b Fig ) . Surprisingly , ActD treatment led to reduced distribution of DVL1 inside nucleoli in both control cells and cells overexpressing Wnt5a ( Fig 2a ) as has been shown previously for both Fibrillarin and UBF [36] . By contrast , the subcellular localization of DVL2 and DVL3 did not change upon ActD treatment ( S2b Fig ) , suggesting a specific role for DVL1 in regulation of rDNA transcription . The nucleolar localization of DVL1 was further confirmed in three distinct breast cell lines ( Fig 2b ) and by using an alternative DVL1 antibody , which also revealed some cytoplasmic staining ( S3a Fig ) . Evidence that DVL1 can be specifically localized to the nucleolus was further shown by ectopic expression of FLAG-tagged DVL1 in fibroblasts lacking endogenous DVL1 protein ( S3b Fig ) as well as by immuno-electron microscopy of MCF7 cells stably expressing Wnt5a , and of MDA-MB-231 breast cancer cells ( S3c Fig ) . Given that the nucleolar localization of DVL1 was more prominent in cells stably expressing Wnt5a ( Fig 2a , right hand panel ) , we next asked whether acute exposure to exogenous Wnt5a protein affects the cellular distribution of DVL1 . We observed that the treatment of MCF7 cells with recombinant Wnt5a resulted in more prominent nucleolar staining of DVL1 at both 15 and 60 minute time points ( Fig 2c ) . Taken together , these observations suggest that DVL1 is actively recruited into the nucleolus in response to Wnt5a signaling . These observations led us to hypothesize that Wnt5a- mediated accumulation of nucleolar DVL1 might be directly involved in the inhibition of rRNA synthesis . To test this hypothesis , chromatin immuno-precipitation ( ChIP ) assays were performed on control MCF7 cells , and MCF7 cells stably expressing Wnt5a , to assess the occupancy of Pol I , and its specific regulators UBF and the de-acetylase SIRT7 , on rDNA regions ( S3d Fig ) DVL1 occupancy on the same rDNA regions was also assessed . Consistent with the hypothesis , MCF7 cells stably expressing Wnt5a showed an increase in DVL1 and UBF localization to the rDNA promoter region as well as 18S and 28S regions and , to a lesser extent , inter-genic sequences ( IGSs ) ( Fig 3a ) . By contrast , relatively modest changes in Pol I occupancy at rDNA regions were detected , particularly outside of the promoter region ( Fig 3b ) . These observations either suggest that the increase in UBF association influences Pol I dissociation in a manner that masks the expected signal or that Wnt5a’s impact is to reduce the transcriptional competency of Pol I while its occupancy is unaffected . However , in cells expressing Wnt5a , SIRT7 occupancy was reduced at the rDNA repeat ( Fig 3c ) . As observed in cells repressed for Pol I transcription , Wnt5a signaling increased UBF occupancy on all regions of the rDNA repeat that were examined ( Fig 3a and 3b ) [37] . Importantly , these effects also correlated with a reduction of histone 3 lysine 4 tri-methylation ( H3K4me3 ) , an epigenetic mark of transcriptionally active chromatin ( Fig 3c ) [37 , 38] . To rationalize DVL1’s apparent distribution throughout the rDNA repeat , we hypothesized that Wnt5a signaling may induce DVL1 to associate with either UBF or Pol I . Consistent with this model , immunoprecipitation assays performed on total nuclear lysates showed that DVL1 co-precipitated with UBF more efficiently in Wnt5a-expressing cells ( Fig 3d , compare lanes 4 and 8 ) . By contrast , DVL1 did not co-precipitate with Pol I , irrespective of Wnt5a expression ( Fig 3e , compare lanes 4 and 8 ) . Conversely , we observed that SIRT7 co-precipitated with UBF , and that this association was ablated upon Wnt5a expression ( Fig 3d , compare lanes 4 and 8 ) . These data argue that Wnt5a signaling enhances interactions between UBF and DVL1 and that nucleolar localization of DVL1 regulates rRNA synthesis by directly or indirectly inducing a loss of SIRT7 from the Pol I transcriptional machinery . We next asked whether DVL1 is an obligate downstream effector of Wnt5a-mediated suppression of rRNA synthesis . To address this question , we generated MCF7 and BT549 cells in which DVL1 expression was reduced by the stable expression of shRNAs specifically targeting DVL1 ( Fig 4a ) , without altering levels of DVL2 and DVL3 ( S4a Fig ) . In both cell lines , the reduction of DVL1 expression resulted in an increase in the steady-state levels of 47S pre-rRNA and enlargements in nucleolar size ( Fig 4b and 4c ) , as well as an increased proliferation rate ( Fig 4d ) . Similar results were obtained when siRNA oligonucleotides were used to reduce DVL1 expression in MCF7 cells ( S4b Fig ) . Importantly , while Wnt5a treatment of control shRNA cells inhibited 47S pre-rRNA synthesis ( as in wild-type cells ) , Wnt5a treatment failed to significantly reduce 47S pre-rRNA levels in DVL1 shRNA cells ( Fig 4e ) . These results demonstrate a requirement for DVL1 in the Wnt5a-induced suppression of rDNA transcription and suggest that nucleolar DVL1 acts as an inhibitor of RNA Pol I-mediated transcription . The ability of Wnt5a to inhibit rRNA synthesis and reduce nucleolar size suggests a possible mechanism by which it could constrain tumor growth in vivo . To address this possibility , we employed a mouse model of breast cancer with tumors driven by the MMTV-PyMT oncogene in wild-type and Wnt5a-null background [39] . As previously reported , Wnt5a-null mammary tumors exhibited increased Ki-67 expression levels compared to those derived from MMTV-PyMT/Wnt5a+/+ mice ( Fig 5b ) [39] . However , we also observed an increase in AgNOR staining of nucleolar areas in the Wnt5a-/- tumors ( Fig 5a ) . These effects suggest that the higher growth rate of Wnt5a-/- tumors depends , at least in part , on elevated levels of rRNA synthesis . MMTV-PyMT/Wnt5a-/- tumors also exhibited increased SIRT7 expression distributed throughout the tumor compared to those derived from MMTV-PyMT/Wnt5a+/+ mice ( Fig 5b ) . Taken together , these data argue that anti-proliferative effects of Wnt5a are mediated through repression of rDNA transcription in vivo . In order to test the relevance of these findings to human disease , we analyzed publicly available expression data sets from cancer patients . We hypothesized that lower Wnt5a and DVL1 expression would correlate with a poorer prognosis and reduced patient survival . Survival curves generated using data from the Cancer Genome Atlas ( TCGA ) indeed corroborated this view ( Fig 5c ) . Interestingly , higher levels of Wnt5a and DVL1 were correlated with higher survival rates irrespective of SIRT7 expression levels .
Ribosome availability is of central importance to the translational capacity of cells and their ability to synthesize biomass [1–3] . In both normal and cancerous cells , signals that attenuate the rate of rRNA synthesis have previously been shown to have a constraining effect on cell growth and proliferation [26] . Our observations in Wnt5a-deficient mammary tumors support this notion . While secreted growth factors are known to have positive effects on rRNA synthesis , and can contribute to oncogenesis , to our knowledge Wnt5a is the first example of a secreted factor that inhibits rRNA gene transcription [26] . In the context of human breast cancer , both epidemiological and functional data have implicated Wnt5a as a tumor suppressor and our data indicating that Wnt5a signals through DVL1 to inhibit rRNA synthesis provide a novel mechanism underlying that role [23–26] . Together with tumor suppressor proteins such as RB , p53 , and other factors known to regulate nucleolar function [23] , we suggest that loss of Wnt5a from breast tumor cells , or from their microenvironment , is one contributor to the enlarged nucleolar and elevated rRNA synthesis levels that are characteristic hallmarks of highly proliferative tumor cells . The molecular details of this Wnt5a-DVL1-SIRT7-rDNA signaling axis may reveal novel targets for drug development in the treatment of cancers whose rapid growth depends on deregulated ribosomal synthesis . The roles of DVL proteins in Wnt signaling have mostly been associated with receptor-proximal aspects , reversible assembly of multi-molecular complexes , and/or interactions with components of the cytoskeleton [20 , 21 , 40 , 41] . However , previous studies have described DVL proteins within the nucleus , where they were implicated in activation of the TCF-responsive target genes of canonical Wnt/β-catenin signaling [19 , 42] . A large number of potential signaling consequences have been invoked for non-canonical signaling by Wnt5a in different contexts , many of which require DVL proteins as signaling intermediates [20 , 21 , 43] . Here we report that Wnt5a signaling leads to increased subcellular localization of DVL1 in the nucleolus , where it negatively regulates rRNA synthesis through loss of the de-acetylase SIRT7 from chromatin regions containing rDNA . Our working model , which rationalizes the available data and explains the present observations , is that DVL1 and SIRT7 may bind to UBF in a mutually exclusive manner at active sites of rRNA synthesis ( Fig 6 ) . The DVL1-induced loss of SIRT7 , which may be a direct or indirect effect , likely prevents the maintenance of the Pol I subunit PAF53 in a hypo-acetylated state , which is required for assembly of the transcription-competent Pol I machinery [13] ( Fig 6 ) . Considering that elevated levels of UBF are generally observed throughout the rDNA region upon transcription inhibition , we speculate that DVL1 may contribute to a mechanism that controls UBF deposition on rDNA chromatin for transcription repression , possibly modulating the number of active rDNA genes [7 , 37] . While further experiments will be needed to examine this mechanism in molecular detail , the present data reveal that extracellular signals can induce a novel pathway for the down-regulation of rDNA transcription . These observations provide further insight into the diverse cellular mechanisms and biological responses that can be regulated by Wnt signaling during development and disease .
MCF7 , MDA-MB-231 , and Rat2 cells were maintained in Dulbecco Modified Eagle’s Medium ( DMEM ) and BT549 in RPMI 1640 medium . All media contained 10% fetal bovine serum ( FBS , Invitrogen ) and 1% penicillin-streptomycin . Recombinant Wnt5a protein ( R&D , #645WN ) and Wnt3a ( Peprotech , #315–20 ) were used at 200 ng/mL . Actinomycin D ( Sigma-Aldrich A-9415 ) was added at 40 ng/mL or 1000 ng/mL MCF7 and BT549 cells expressing exogenous Wnt5a were generated by infection with the retrovirus vector LNCX containing Wnt5a cDNA , followed by selection of pooled colonies in G418 ( Geneticin ) . Control cells were infected in parallel with the ‘empty’ vector . Rat2 fibroblasts transduced with FLAG-tagged DVL1 in an LNCX vector were constructed by Dr . José González-Sancho . Lentivirus particles expressing DVL1 shRNA ( sc-35228-V ) , or non-silencing control shRNA were purchased from Santa Cruz Biotechnology Inc . and used to infect MCF7 and BT549 cells with selection in 5 μg/mL puromycin . ON-TARGETplus Human DVL1 ( 1855 ) siRNA—SMARTpool ( L-004068-00-0005 ) and ON-TARGETplus Non-Targeting Pool ( D-001810-10 from Dharmacon was used for nucleofection ( Lonza Group Ltd , Switzerland ) of MCF7 cells . Cells were plated on glass cover slips at 50% confluency , one day before treatment with Wnt5a , vehicle , or Actinomycin D . Experiments were repeated at least three times . After treatment , cells were fixed in 4% formaldehyde for 15 min , quenched with 10 mM glycine , permeabilized with 0 . 1% or 0 . 3% Triton-X100 for 15 min , and blocked for 1 hour with 5% normal goat serum in PBS . In between these steps , cells were washed three times with PBS . Primary antibodies and their dilutions were as follows: Ki-67 ( 1:1000 , RM-9106S1 , Thermo Scientific ) ; DVL1 , ( 1:100 , monoclonal 3F-12 , Santa Cruz Biotech , SC-8025 ) and ( 1:100 , polyclonal Biomol , DA4170 ) ; DVL2 ( 1:1000 , Cell Signaling 3216 ) ; DVL3 ( 1:200 , 4D3 , Santa Cruz Biotech , SC- 8025 ) ; Fibrillarin , ( 1:1000 , Abcam ab5821 ) ; UBF , ( 1:100 , F-9 , SC-13125 , Santa Cruz Biotech . ) ; β-catenin , ( 1:200 , BD Transduction Labs , 610154 ) , BrdU ( 1:200 , Clone BU33 , B2531 , Sigma-Aldrich ) , FLAG ( 1:250 , F4049 , Sigma-Aldrich ) . Cells were stained for 60 min with primary antibodies diluted in PBS containing 0 . 1% BSA ( PBSB ) . After two washes in PBSB , cells were incubated for 60 min with secondary antibodies diluted 1:1000 in PBSB . Secondary antibodies were: Alexa Fluor 488 goat anti-mouse , Alexa Fluor 546 goat anti-rabbit , Alexa Fluor 647 goat anti-rabbit ( A11001 , A11010 and A21244 Invitrogen Inc . ) . After antibody removal , cells were washed three times in PBSB , incubated with 1 mM TO-PRO ( Invitrogen ) in PBS or with DAPI to stain DNA , and washed twice more in PBS . Cover slips were mounted in SlowFade ( Invitrogen Inc . ) on microscope slides and visualized using either a Zeiss LSM510 or Zeiss LSM 710 confocal microscope . Cells were imaged using a 63X oil immersion lens , there were on average 20 cells per field . Each experiment was done in biological triplicate ( n = 3 ) with each condition done in technical replicate ( n = 2 ) . Cells were seeded in 96-well plates at a density of 104 cells per well in 100 μl of medium and their numbers evaluated over 4 successive days . For each 24 hour time point , 10 μl of 3- ( 4 , 5-dimethylthiazol-2-yl ) -2 , 5-diphenyltetrazolium bromide ( MTT ) solution ( Invitrogen; 5mg/ml in PBS ) was added and the cells incubated for 4 hours at 37°C . After removing the medium , 75 μl dimethyl sulfoxide was added to each well to solubilize the precipitate , and absorbance at 540 nm was measured in a microplate reader ( EL321e; Bio-Tek Instruments , Winooski , VT ) . Where indicated , the equivalent experiments were carried out in MCF7 cells subjected to shRNA-mediated silencing of DVL1 . Silver staining of nucleoli was based on previously described AgNOR procedures [34 , 35] . Briefly , after fixation and rehydration , cells were stained with a freshly prepared AgNOR staining solution for 30 min . After staining , cells were rinsed twice in distilled water , treated with 5% sodium thiosulfate for 2–5 min , rinsed again , and mounted for bright field microscopy and image capture . ImageJ software was used to determine the total area of AgNOR staining in fields of 100 cells each [35] . Values were obtained from at least three experiments prior to analysis by Student’s t-test . As previously described , mammary anlagen from mice carrying the MMTV-PyMT transgene in a Wnt5a-/- or Wnt5a+/+ background were rescued from e18 . 5 day embryos by transplantation into cleared fat pads of 3-week old ICR/SCID hosts [39] . After expansion of the epithelium for one month , the tissue was transplanted again into ICR/SCID recipients and invasive adenocarcinomas of the two genotypes were recovered at 16 weeks . Paraffin-embedded fixed tissue from 3 tumors of each genotype was sectioned and processed for AgNOR staining as described and nucleolar area measured in fields of 100 nuclei as above . IHC staining of SIRT7 , ( 1:150 , 62748 , Abcam , ) and Ki-67 ( 1:150 , RM-9106S1 , Thermo Scientific ) was done according to standard protocols . MCF7 and BT549 cells were grown in 12-well dishes and , where indicated , treated with Wnt5a ( 200 ng/ml ) , sFRP1 ( R&D , 1384-SF , 400 ng/ml ) or a combination of both . In the case of double treatment , the cells were pre-treated with sFRP1 for one hour before addition of Wnt5a for 1 hour . RNA was isolated using an RNeasy Mini kit ( Qiagen ) according to the manufacturer’s protocol and treated with DNase1 . cDNA was then synthesized with an iScript cDNA synthesis kit ( Bio-Rad ) . The concentration of cDNA was determined by spectrophotometry , following which it was diluted in RNase-free water to a concentration of 10 ng/uL . Semi-quantitative RT-PCR was performed using 20 ng of cDNA template , a SYBR Green Fast-Mix PCR kit ( Quanta Biosciences ) , and an MJ Research Opticon2 real-time PCR machine . The annealing temperature for 47S primers was 52 . 5°C and for all other primers was 55°C . All samples were run in triplicate . Primer sequences for 47S pre-rRNA were: forward , 5´-TGTCAGGCGTTCTCGTCTC-3´; reverse , 5´-GAGAGCACGACGTCACCAC-3´ ( both from IDT , Inc . ) [27] . Primers for DVL1 ( QT01672944 ) , β-actin ( QT01680476 ) and Wnt5a ( QT00025109 ) were from Qiagen ( Quantitect ) . Fold changes in RNA levels were calculated using the delta-delta Ct method and analyzed by Student’s t-test [44] . ChIP assays were performed as previously described [33–34] . Formaldehyde cross-linked chromatin obtained from MCF7/Ctrl or MCF7/Wnt5a cells was subjected to immuno-precipitations with the autoimmune serum S57299 against Pol I [45] or with antibodies to DVL1 , UBF , SIRT7 , Histone H3K4me3 ( HistoneH3K4me3 , 39159 , Active Motif ) , and non-specific rabbit IgGs ( Abcam ) as control . DNA-protein complexes were analyzed by qPCR with primers specific for the rDNA promoter ( forward primer 5’- CCCGGGGGAGGTATATCTTT; reverse primer 5’- CCAACCTCTCCGACGACA ) , the 18S ( Forward: 5’- CGGCTACCACATCCAAGGAA; reverse 5’- GCTGGAATTACCGCGGCT ) , the 28S rDNA repeat ( Forward: 5’- CGACGACCCATTCGAACGTCT; reverse 5’- CTCTCCGGAATCGAACCCTGA ) and the IGS ( Forward: 5’- ATCTTGTTGTGCGGGAGTTC; reverse 5’- TTGTTCTGTCACTCGGTTGC ) . The qPCR analysis was performed as previously described and the results displayed as bars diagrams [45] . The values are presented as percentages of the input signal for each primer pair . Immunoprecipitation assays were performed as previously described [45] . Briefly , nuclear extracts of growing MCF7 cells or MCF7 cells expressing Wnt5a were incubated with the autoimmune serum S57299 against Pol I [45] or with antibodies against UBF , DVL1 , or control non-specific rabbit immunoglobulins ( IgGs ) . The antibodies were subsequently precipitated with Protein G Sepharose ( Invitrogen ) . The beads were washed with 1X PBS supplemented with 1 mM PMSF , 0 . 2% NP-40 and then re-suspended in Laemmli buffer and heat denatured . Bound proteins were resolved by SDS-PAGE and analyzed on immunoblots for UBF , DVL1 , SIRT7 and the largest Pol I subunit RPA194 . Active foci of Pol I-mediated transcription in MCF7 cells were revealed by FUrd incorporation as previously described [32] . Sub-confluent cells were pelleted and fixed with 4% paraformaldehyde ( Merck ) , dehydrated and embedded in Agar 100 resin ( Agar Scientific Ltd ) . Thin sections were stained with 2% uranyl acetate and examined in a transmission electron microscope Tecnai G2 Spirit BioTwin ( FEI Company ) at 80 kV . Cells were lysed in RIPA buffer ( 50 mM Tris-HCl pH 7 . 5 , 150 mM NaCl , 1 mM EDTA , 1% NP-40 , 0 . 5% sodium deoxycholate , 0 . 1% SDS ) supplemented with protease inhibitors ( cOmplete cocktail , Roche ) . Protein extracts were boiled in gel sample buffer ( Invitrogen ) , separated by SDS–PAGE under reducing conditions and transferred to nitrocellulose filters ( Hybond , Amersham ) by electro-blotting . Primary antibodies and dilutions used were DVL1 ( 3F-12 , 1:200; SC-8025 , Santa Cruz Biotech ) ; DVL2 ( 1:1000 , 3216 , Cell Signaling ) ; DVL3 ( 1:200 , 4D3 , Santa Cruz Biotech , SC-8025 ) ; Wnt5a ( 1:300; AF645 , R&D ) ; Pol I RPA194 ( C1 , sc-48385 , Santa Cruz Biotech ) ; UBF ( 1:500 , F-9 , Santa Cruz Biotech , SC-13125 ) ; SIRT7 ( 1:500; Abcam #62748 ) ; GAPDH ( 1:1000 , 6C5 , Abcam ab8245 ) ; β-actin ( 1:500 , Abcam ab1801 ) ; Tubulin ( 1:500 , Abcam , ab4074 ) . Secondary antibodies include HRP conjugated , Mouse ( GE , NA931V ) , Rabbit ( NA931V , GE ) or Rat ( NA9340V ) , used at 1:10000 . Blots were exposed using ECL using SuperSignal West Femto Maximum Sensitivity Substrate ( Thermo Scientific , 34095 ) and visualized via BioRad ChemiDoc XRS imaging system . Survival analysis was performed using the PROGgene V2 Prognostic Database ( http://watson . compbio . iupui . edu/chirayu/proggene/database/ ? url=proggene ) [46] . Each analysis used “breast cancer” as cancer type , “death” as survival measure , and bifurcated the gene expression at the median . The gene expression was taken from the TCGA database . The data were not adjusted for clinical status . The survival status was analyzed for expression levels of Wnt5a , DVL1 and SIRT7 .
|
Synthesis of the translation machinery , including the mega-Dalton , RNA-protein ribosome complex , serves as a key driver of cellular growth and proliferation . It is therefore unsurprising that ribosomal biogenesis is under intricate regulation . The process through which ribosomes are made entails the coordination of components from diverse signaling pathways in both normal and diseased cells . Both oncogenes and tumor suppressors can influence this orchestration by impinging upon the rate-determining steps of RNA Polymerase I-mediated transcription of ribosomal RNA ( rRNA ) and the coupled process of ribosome assembly . In this study we investigated whether the secreted protein Wnt5a , an antagonist of mammary tumor growth , regulates rRNA synthesis in breast cancer cells . We find that the induction of Wnt5a signaling disturbs assembly of the RNA polymerase I machinery , leading to a repressive rDNA chromatin state that is not amenable to active rRNA gene transcription . Wnt5a signaling represses rRNA synthesis by stimulating nucleolar accumulation of Dishevelled1 ( DVL1 ) , a downstream effector of Wnt5a signaling , while having no such effect on DVL2 or DVL3 . Wnt5a-induced accumulation of DVL1 in the nucleolus directly interferes with the synthesis of rRNA , suggesting that a tumor suppressive effect of Wnt5a in breast cancer cells is mediated by DVL1-dependent repression of rRNA synthesis .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"medicine",
"and",
"health",
"sciences",
"breast",
"tumors",
"cancers",
"and",
"neoplasms",
"dna",
"transcription",
"oncology",
"epigenetics",
"immunologic",
"techniques",
"cellular",
"structures",
"and",
"organelles",
"chromatin",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"chromosome",
"biology",
"oncogenic",
"signaling",
"gene",
"expression",
"immunoassays",
"breast",
"cancer",
"immunofluorescence",
"ribosomes",
"biochemistry",
"rna",
"signal",
"transduction",
"cell",
"staining",
"ribosomal",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"genetics",
"biology",
"and",
"life",
"sciences",
"wnt",
"signaling",
"cascade",
"non-coding",
"rna",
"cell",
"signaling",
"signaling",
"cascades"
] |
2016
|
Wnt5a Signals through DVL1 to Repress Ribosomal DNA Transcription by RNA Polymerase I
|
Trypanosoma cruzi ( Tc ) infection causes chagasic cardiomyopathy; however , why 30–40% of the patients develop clinical disease is not known . To discover the pathomechanisms in disease progression , we obtained the proteome signature of peripheral blood mononuclear cells ( PBMCs ) of normal healthy controls ( N/H , n = 30 ) and subjects that were seropositive for Tc-specific antibodies , but were clinically asymptomatic ( C/A , n = 25 ) or clinically symptomatic ( C/S , n = 28 ) with cardiac involvement and left ventricular dysfunction . Protein samples were labeled with BODIPY FL-maleimide ( dynamic range: > 4 orders of magnitude , detection limit: 5 f-mol ) and resolved by two-dimensional gel electrophoresis ( 2D-GE ) . After normalizing the gel images , protein spots that exhibited differential abundance in any of the two groups were analyzed by mass spectrometry , and searched against UniProt human database for protein identification . We found 213 and 199 protein spots ( fold change: |≥ 1 . 5| , p< 0 . 05 ) were differentially abundant in C/A and C/S individuals , respectively , with respect to N/H controls . Ingenuity Pathway Analysis ( IPA ) of PBMCs proteome dataset identified an increase in disorganization of cytoskeletal assembly and recruitment/activation and migration of immune cells in all chagasic subjects , though the invasion capacity of cells was decreased in C/S individuals . IPA predicted with high probability a decline in cell survival and free radical scavenging capacity in C/S ( but not C/A ) subjects . The MYC/SP1 transcription factors that regulate hypoxia and oxidative/inflammatory stress were predicted to be key targets in the context of control of Chagas disease severity . Further , MARS-modeling identified a panel of proteins that had >93% prediction success in classifying infected individuals with no disease and those with cardiac involvement and LV dysfunction . In conclusion , we have identified molecular pathways and a panel of proteins that could aid in detecting seropositive individuals at risk of developing cardiomyopathy .
Chagasic cardiomyopathy is caused by Trypanosoma cruzi . According to the World Health Organization report released in 2010 , ~16 million individuals are infected with T . cruzi , and >25 million people are at risk of infection in Latin America and Mexico [1] . New challenges of increased transmission are faced due to lack of sustainability of the vector control programs [2 , 3] , migration of infected individuals to non-endemic areas ( e . g . US , Canada , Europe ) [4 , 5] , and transfer of infection through blood or organ donation [6 , 7] . The Centers for Disease Control reports that >300 , 000 individuals infected with T . cruzi are currently living in the United States [8] . Several years after the initial exposure to the parasite , ~30–40% of the infected individuals develop cardiomyopathy and may progress to heart failure ( reviewed in [9] ) . No vaccine is available for the prevention of infection [10] and the available drugs , benznidazole and nifurtimox , have exhibited no significant effects in arresting the progression of chronic cardiomyopathy [11] . Importantly , tools to assess the effectiveness of new drugs against T . cruzi infection and Chagas disease are currently not available . We have found that T . cruzi elicits oxidative stress of inflammatory and mitochondrial origin in immune and non-immune cells; and sustained oxidative stress plays a crucial role in eliciting left ventricular dysfunction during progressive Chagas disease [9 , 12 , 13] . Our studies showed that myocardial changes in oxidant/antioxidant balance and oxidative adducts were detectable in the peripheral blood of infected mice [14] and chagasic patients [15–17] . The level of oxidative stress markers ( i . e . lipid hydroperoxides ) and inflammation ( i . e . myeloperoxidase ) increased and the level of antioxidants ( e . g . manganese superoxide dismutase ) decreased in both heart and peripheral blood of infected rodents with progressive disease [14] . These studies , thus , support the notion that peripheral blood cells provide a suitable tissue for delineating the pathways that are deregulated during the chronic development of chagasic cardiomyopathy . In this study , we have employed a quantitative saturation fluorescence labeling approach for the detection of the differential protein signature of peripheral blood mononuclear cells ( PBMCs ) in T . cruzi-infected subjects . All enrolled subjects were assessed by electrocardiography and transthoracic echocardiography and characterized for the severity of cardiac disturbances . We employed a thiol-labeling maleimide dye under saturating conditions that exhibits stable , specific , quantitative labeling of cysteine residues in conjunction with two-dimension electrophoresis and mass spectrometry for developing the PBMCs’ proteome of chagasic patients . Up to 92% of the human proteins contain at least one cysteine residue [18] , and thus can be detected using the thiol-labeling maleimide dye . Our findings provide clues to the molecular pathways that may be disturbed with development of chronic Chagas disease . We discuss a panel of proteins that could potentially be useful in classifying the disease state and identifying asymptomatic individuals at risk of developing clinical disease .
All chemicals and reagents were of molecular grade ( >99 . 5% purity ) . BD Vacutainer CPT Cell Preparation Tubes ( heparinized ) containing 8 ml whole blood samples were centrifuged following manufacturer’s instruction . The FICOLL Hypaque™ density gradient was employed to enrich the PBMC fraction , and the latter was pelleted by centrifugation at room temperature at 400 x g for 10 min . The PBMC pellets were suspended in 1 ml of hypotonic buffer to lyse contaminating red blood cells , and 9 ml of complete RPMI-1640 medium / 10% fetal bovine serum ( Invitrogen ) added . After centrifugation as above , final cell pellets consisting of 8-10-million PBMCs were stored at -80°C . PBMC pellets from individual study subjects were lysed in 7 M urea , 2 M thiourea , 2% CHAPS , and 50 mM Tris ( pH 7 . 5 ) , containing benzonase nuclease ( 300-units/ml ) , as described previously [20 , 21] . Protein concentrations were determined by using a Pierce Modified Lowry Protein Assay Kit , and cysteine ( cysteic acid ) levels in all samples were determined by using an Amino Acid Analyzer ( Model L8800 , Hitachi High Technologies America , Pleasanton , CA ) [20] . Samples were incubated for 1 h with 6 mM ascorbate ( Asc ) to ensure all cysteine residues were reduced and available for dye-binding , dialyzed against urea buffer to remove excess ascorbate , and then labeled with BODIPY FL N- ( 2-aminoethyl ) maleimide ( BD from Life Technologies , Grand Island , NY ) at 60-fold excess to cysteine [21] . The mixtures were incubated for 2 h; the reactions were stopped with a 10-fold molar excess of 2-mercaptoethanol over dye . All incubations were carried out at room temperature in the dark in 200 μl reaction volume [20 , 21] . BD-labeled PBMC lysates ( 100 μg protein ) were separated by 2-dimension electrophoresis ( 2DE ) , employing an IPGphor multiple sample isoelectric focusing ( IEF ) device ( GE Healthcare ) in the first dimension , and the Criterion Dodeca cell ( Bio-Rad ) in the second dimension , as we have described previously [22 , 23] . Briefly , samples were loaded on to 11 cm dehydrated precast immobilized pH gradient ( IPG ) strips ( GE Healthcare ) , and strips were rehydrated overnight . IEF was performed at 20°C with the following parameters: 50 V , 11 h; 250 V , 1 h; 500 V , 1 h; 1 , 000 V , 1 h; 8 , 000 V , 2 h; 8 , 000 V , 48 , 000 V/h . The IPG strips were then incubated in 10 ml of equilibration buffer ( 6 M urea , 2% sodium dodecyl sulfate ( SDS ) , 50 mM Tris-HCl , pH 8 . 8 , 20% glycerol ) for 30 min at 22°C , and electrophoresis was performed at 150 V for 2 . 25 h , 4°C using precast 8–16% polyacrylamide gels in Tris-glycine-SDS buffer ( 25 mM Tris-HCl , 192 mM glycine , 0 . 1% SDS , pH 8 . 3 ) [22 , 23] . Gels were fixed in 20% methanol / 7% acetic acid / 10% acetonitrile for 1 h and washed with 20% ethanol / 10% acetonitrile to reduce background . Gel images were acquired at 100 μm resolution using the Typhoon Trio Variable Mode Imager ( GE Healthcare ) to quantify BD-labeled proteins ( Ex488 nm / Em520 nm ) . Up to 92% of the human proteins contain at least one cysteine residue [18] . The Totallab SameSpots software ( formerly Nonlinear Dynamics Ltd . Newcastle , UK ) selects one reference gel according to several criteria , including quality and number of spots with the intent on selecting the gel that best represents all the gels . The reference gel containing the most common features was selected from the pool of gels of the N/H samples , and all data were then derived by comparison to the N/H reference gel . To ensure that the maximum numbers of proteins were detected , the reference gel was also stained with SyproRuby ( Life Technologies Grand Island , NY ) that binds all proteins irrespective of presence or absence of cysteine amino acid , and gel image was acquired at Ex488nm/Em560nm . The exposure time for both dyes ( BD and SyproRuby ) was adjusted to achieve a value of ~55 , 000–63 , 000 pixel intensity ( 16-bit saturation ) from the most intense protein spots on the gel [22 , 23] . In total , 83 BD-stained 2D gels representing 30 , 25 , and 28 samples from N/H , C/A , and C/S subjects , respectively , were scanned and analyzed with the Totallab SameSpots software . After manual and automated pixel-to-pixel alignment , the program performed automatic spot detection on all images . The SyproRuby stained reference gel was used to define spot boundaries; however , the gel images taken under the BD-specific filters were used to obtain the quantitative spot data . This strategy ensures that spot numbers and outlines were identical across all gels in the experiment , eliminating problems with unmatched spots as well as ensuring that the greatest number of protein spots and their spot volumes were accurately detected and quantified [23] . Protein spot abundance ratios were calculated from normalized spot volumes from affected samples versus the matched normal spot volumes ( Δ protein abundance = Asc+chagasic/Asc+ N/H controls ) . Spot volumes were normalized for each sample using a software-calculated bias value assuming that the great majority of spot volumes did not change in abundance ( log ( abundance ratio ) = 0 ) . The scatter of the log ( abundance ratios ) for each spot in a gel ( sample ) is distributed around some mean value that represents the systematic factors that govern the experimental variation . Thus , a gain factor is calculated to adjust the mean spot ratios of a given gel to 0 ( log ( abundance ratio ) = 0 ) and applied to each spot volume [23] . For the purpose of selecting differentially abundant protein spots for mass spectrometry , normalized spot volumes were subjected to statistical analysis using in-built tools in Totallab SameSpots software . Spot volumes were log2 transformed and spot-wise standard deviation , arithmetic mean , and coefficient of variation ( CoV ) values of the standard abundance values were calculated for each spot [24] . Student’s t-tests with Welch’s correction for unequal variances were used to test for differential protein expression between N/H controls and either C/A or C/S chagasic subjects . Benjamini-Hochberg multiple hypothesis testing correction was applied to account for the false discovery rate and significance was accepted at p<0 . 05 . The protein spots identified to be differentially abundant ( p< 0 . 05 ) in at least one of the groups were submitted for mass spectrometry identification . Selected spots on the 2D gels that exhibited significant differential prevalence ( p≤0 . 05 ) in at least one of the group were picked robotically ( ProPick II , Digilab , Ann Arbor , MI ) , and trypsin digested as described by us [19 , 25] . In brief , gel spots were incubated at 37°C for 30 min in 50 mM NH4HCO3 , dehydrated twice for 5 min each in 100-μl acetonitrile , dried , and proteins were digested in-gel at 37°C overnight with 10 μl of trypsin solution ( 1% trypsin in 25 mM ammonium bicarbonate ) . Peptide mixtures ( 1-μl ) were directly spotted onto a MALDI-TOF MS/MS target plate with 1 μl of alpha-cyano-4-hydroxycinnamic acid matrix solution ( 5 mg/ml in 50% acetonitrile ) , and analyzed using a MALDI-TOF/TOF AB Sciex TOF/TOF 5800 Proteomics Analyzer ( Framingham , MA ) . The Applied Biosystems software package included the 4000 Series Explorer ( v . 3 . 6 RC1 ) with Oracle Database Schema ( v . 3 . 19 . 0 ) and Data Version ( 3 . 80 . 0 ) to acquire and analyze MS and MS/MS spectral data . The instrument was operated in a positive ion reflectron mode with the focus mass set at 1700 Da ( mass range: 850–3000 Da ) . For MS data , 1000–2000 laser shots were acquired and averaged from each protein spot . Automatic external calibration was performed by using a peptide mixture with the reference masses 904 . 468 , 1296 . 685 , 1570 . 677 , and 2465 . 199 . Following MALDI MS analysis , MALDI MS/MS was performed on several ( 5–10 ) abundant ions from each protein spot . A 1-kV positive ion MS/MS method was used to acquire data under post-source decay ( PSD ) conditions . The instrument precursor selection window was +/- 3 Da . Automatic external calibration was performed by using reference fragment masses 175 . 120 , 480 . 257 , 684 . 347 , 1056 . 475 , and 1441 . 635 ( from precursor mass 1570 . 700 ) [19 , 25] . For protein identification , the MS and MS/MS spectral data were searched against the UniProt human protein database ( last accessed: March 25 , 2013; 87 , 656 sequences; 35 , 208 , 664 residues ) by using a AB Sciex GPS Explorer ( v . 3 . 6 ) software in conjunction with MASCOT ( v . 2 . 2 . 07 ) as described previously [19] . The protein match probabilities were determined by using expectation values and/or MASCOT protein scores . The MS peak filtering included the following parameters: a mass range of 800 Da to 3000 Da , minimum S/N filter = 10 , mass exclusion list tolerance = 0 . 5 Da , and mass exclusion list for some trypsin and keratin-containing compounds included masses ( Da ) 842 . 51 , 870 . 45 , 1045 . 56 , 1179 . 60 , 1277 . 71 , 1475 . 79 , and 2211 . 1 . The MS/MS peak filtering included the following parameters: minimum S/N filter = 10 , maximum missed cleavages = 1 , fixed modification of carbamidomethyl ( C ) , variable modifications due to oxidation ( M ) , precursor tolerance = 0 . 2 Da , MS/MS fragment tolerance = 0 . 3 Da , mass = monoisotopic , and peptide charges = +1 . The significance of a protein match , based on the peptide mass fingerprint ( PMF ) in the MS and the MS/MS data from several precursor ions , is presented as expectation values ( p<0 . 05 ) . To confirm the identified proteins were of human and not of parasite origin , we also performed a similar search against NCBI non-redundant protein database consisting of T . cruzi sequences . In cases where abundance was ≥ |2| but protein IDs were ambiguous ( protein scores <62 ) , the digested proteins were submitted for analysis by LTQ OrbiTrap Velos ( ThermoFisher , Waltham , MA ) . We used the Ingenuity Pathways Analysis ( IPA ) web-based application ( Ingenuity Systems , Redwood city , CA ) to assess the biological meaning in the proteome datasets . IPA retrieves biological information from the literature—such as gene name , sub-cellular location , tissue specificity , function , and association with disease—and then integrates the identified proteins into networks and signaling pathways with biological meaning and significance [26] . An “e-value” was calculated by estimating the probability of a random set of proteins having a frequency of annotation for that term greater than the frequency obtained in the real set , and a significance threshold of 10−3 was used to identify significant molecular functions and biological processes [19] . With these parameters , we were able to highlight the most informative and significantly over-represented gene ontology terms in the dataset [19 , 27] . For MARS modeling , normalized spot volumes for all spots from 83 gels were exported from SameSpots in to Excel , and analyzed by using R and SPSS ver . 20 software . For modeling the disease state specific response , a stringent cut-off was applied; differentially abundant protein spots were first screened by t test/Welch’s correction and then Benjamini-Hochberg test was employed at p<0 . 001 ( ≥І1 . 5І fold change ) . MARS was employed to model changes in multiple variables for distinguishing between infection and disease status [24] . We used 10-fold cross-validation and 80% ( training ) /20% ( testing ) approaches to predict the protein spots that can distinguish N/H from C/A and C/S subjects . The sensitivity and specificity of the identified models were validated by receiver operator characteristics ( ROC ) curves .
All protein extracts were analyzed for cysteine content by amino acid analysis and labeled with uncharged BODIPY FL-maleimide ( BD , dye-to-protein thiol ratio > 60:1 ) . The saturation fluorescence labeling with BD provided no non-specific labeling , had no effect on the isoelectric point and mobilities of the proteins , and provided a linear dynamic range of over four orders of magnitude in identifying the protein spots ( detection limit: 5 f mol protein in a gel spot at a signal-to-noise ratio of 2:1 ) , as we have also noted in a previous study [23] . PBMC lysates of the normal healthy ( N/H ) controls ( n=30 ) , and of seropositive , clinically asymptomatic ( C/A , n=25 ) and seropositive , clinically symptomatic ( C/S , n=28 ) individuals were resolved by 2D-GE . The representative 2D gel images for these groups are shown in Fig 1A–1C . All protein spots were within the relative molecular sizes 10 to 250 kDa . All of the 2D gel images were assessed for quality control by SameSpots software , and then aligned both manually and automatically against the reference gel ( Fig 2 ) , chosen from the entire set of gel images by the software . The fluorescence intensity of the protein spots was normalized using a bias factor calculated assuming most spots did not change across the experiment . The log2 transformed abundance values for each protein spot on 2D gels were utilized to calculate the mean coefficient of variation ( CoV ) values ( Fig 3 ) for the biological replicates . These data showed the mean CoV values were 49 ± 21 . 7% , 67 ± 26 . 4% , and 77 ± 41 . 3% , for N/H , C/A and C/S groups , respectively ( Fig 3A–3C ) . Up to 75% of the spots in all groups did not exceed the CoV value of 80% indicating that most of the protein abundances are quite stable in the different groups . Protein spots exceeding a CoV of 100% were largely noted in chagasic subjects , indicating a changing and variable protein expression pattern with disease progression . For the purpose of selecting protein spots for identification by mass spectrometry , the protein spot datasets were analyzed in pair-wise manner by t test with Welch's correction that accounts for unequal variances . This analysis yielded 315 ( 162 up-regulated , 153 down-regulated , p<0 . 05 ) and 348 ( 180 up-regulated , 168 down-regulated , p<0 . 05 ) differentially abundant protein spots in seropositive subjects with no disease and those with LV dysfunction , respectively . These datasets were then submitted to Benjamini-Hochberg multiple hypothesis testing correction to adjust the false discovery rate , and the differentially abundant protein spots ( fold change: |≥1 . 5| , p<0 . 05 with B-H correction ) were submitted for MALDI-TOF/TOF analysis . Homology searches were conducted against the UniProt’s human proteome database for protein identification [19] . A total of 213 protein spots ( 102 up-regulated , 111 down-regulated , fold change: |≥1 . 5| ) in seropositive/clinically-asymptomatic subjects; and 199 protein spots ( 97 up-regulated , 102 down-regulated , fold change: |≥1 . 5| ) in seropositive subjects with LV dysfunction were found to be differentially expressed with respect to normal controls , and identified by mass spectrometry ( Table 2 ) . These proteins were predicted to be localized in cytoplasm ( 67% ) , extracellular space ( 14% ) , nucleus ( 8% ) , or plasma membrane ( 9% ) ( Fig 4A ) . The changes in abundance frequency of the identified proteins ranged from > -3-fold to >9-fold in chagasic subjects ( Fig 4B ) . A majority of the identified protein spots were differentially abundant in all chagasic subjects though the extent of change in expression was more pronounced in seropositive subjects with LV dysfunction . When we compared the differential abundance of proteins in seropositive C/A versus C/S subjects , we noted 20 and 10 protein spots that were uniquely changed in abundance in clinically-asymptomatic ( Fig 4C ) and clinically-symptomatic subjects ( Fig 4D ) , respectively , and were relevant to disease state . We performed IPA analysis to predict the molecular and biological relationship of the differential proteome datasets ( Table 2 ) . IPA recognizes all isoforms ( e . g . gel-detected pI and size variants of actin , fibrinogen ) as the same protein and collapsed the dataset to 82 and 78 differentially abundant proteins in seropositive subjects with no heart disease and those with LV dysfunction , respectively . IPA analysis of the differential proteome datasets predicted an increase in cytoskeletal disassembly and disorganization ( z-score: -1 . 091 to -0 . 248 , S1 Fig ) , immune cell aggregation ( ALB↓ , FGA↑ , GSN↓ , MPO↓ , THBS1↑ , z-score: 1 . 521 , p value 1 . 48E-03 ) and recruitment/activation and migration of immune cells in chagasic ( vs . normal ) subjects ( z-score: 0 . 501–1 . 698 , p value: 1 . 94–5 . 29E-04 , S2 Fig ) , though invasion capacity of cells was decreased in C/S subjects ( S2 Fig panel B ) . Molecular and cellular function annotation of the proteome datasets by IPA predicted a balanced cell proliferation/cell death response in C/A subjects ( S3 Fig panel A ) while cell death along with inhibition of cell survival was dominantly predicted in PBMCs of C/S subjects ( S3 Fig panel B , z-score: 0 . 858–2 . 406 ) . IPA also implied a pronounced increase in production of free radicals associated with a decline in scavenging capacity with progressive disease in chagasic subjects ( z-score: 1 . 019 to -1 . 455 , S4 Fig ) . The top up-stream molecules predicted to be deregulated and contributing to the differential proteome with disease progression in chagasic subjects included MYC , SP1 , MYCN , and growth factor ANGPT2 ( z-score -2 . 266 to -2 . 190 ) proteins ( S5 Fig ) . We performed MARS analysis to develop a classification model for predicting risk of disease development . MARS is a nonparametric regression procedure that creates models based on piecewise linear regressions . It searches through all predictors to find those most useful for predicting outcomes , and then creates optimal model by a series of regression splines called basis functions [28 , 29] . For this , MARS uses a two-stage process; first half of the process involves creating an overly large model by adding basis functions that represent either single variable transformations or multivariate interaction terms . In the second stage , MARS deletes basis functions in order of least contribution to the model until the optimum one is reached . End result is a classification model based on single variables and interaction terms which will optimally determine class identity [28 , 29] . Inputs to the model were log2 transformed values for protein spots that were differentially abundant in seropositive/no disease ( 84 spots , n = 25 ) and clinically-symptomatic ( 87 spots , n = 28 ) groups with respect to normal controls ( n = 30 ) at p<0 . 001 with B-H correction . We assessed the model accuracy by looking at the prediction success rate and the ROC curves . To address the possible issue of over-fitting the data , we employed two approaches: 1 ) 10-fold cross validation ( CV ) allowing same number of maximum basis functions as were the differentially abundant protein spots at p<0 . 001 ( with 1 max interaction term ) , and 2 ) testing/training approach in which 80% of the data was utilized for creating the model and the 20% of the remaining data was used to assess the fit of the model for testing dataset . The CV and 80/20 approaches identified 11 and 6 protein spots , respectively , with high importance ( score >20 , Fig 5A & 5B ) for creating the MARS model , detecting differences between the controls and seropositive/no disease subjects . The prediction success showed the CV and 80/20 models fitted perfectly on the training dataset ( AUC/ROC: 1 . 00 ) and by >93% on the testing dataset ( AUC/ROC: 0 . 96 for CV and 0 . 933 for 80/20 ) ( Fig 5C & 5D ) . Likewise , the CV and 80/20 approaches identified 11 and 8 protein spots , respectively , with high importance ( score >20 , Fig 6A & 6B ) for creating the MARS model distinguishing controls from clinically-symptomatic chagasic patients . The prediction success of the CV and 80/20 models were 100% for the training data ( AUC/ROC: 1 . 00 ) . When fitted on testing data , the CV model exhibited very high prediction success ( AUC/ROC: 0 . 926 , Fig 6 ) while the 80/20 model fitted perfectly on the training data ( AUC/ROC: 1 . 00 , Fig 6D ) . These analyses suggested that PBMC changes in the selected protein spots will have high specificity and sensitivity in predicting the disease state in chagasic subjects in comparison to normal/healthy controls .
This study was aimed at assessing the proteomic changes in PBMCs of chagasic subjects grouped as clinically asymptomatic ( C/A , n = 25 ) and clinically symptomatic with heart involvement ( C/S , n = 28 ) in comparison with healthy subjects ( n = 30 ) . 2DE/ MALDI-TOF MS analysis identified 213 and 199 protein spots that were differentially abundant in C/A and C/S subjects in comparison to normal/healthy controls ( Table 2 ) . The major cell populations in PBMCs are lymphocytes ( B , T and NK cells , ≥70% ) and monocytes/macrophages ( 10–30% ) . Very few studies have , however , characterized the role of peripheral immune cells in parasite control vs . cardiac pathology in Chagas disease . For example , a recent study noted detection of no NK cells in early infection [30] . In late acute stage of infection , a selective increase in a distinct lineage of NK cells ( CD16+CD56– ) , as well as a persistent expansion of B cells , possibly indicative of a relationship between B cell activation and a subset of NK cells was noted in humans [30 , 31] . Others have demonstrated a robust expansion of T cell response in patients with progressive chronic disease though their role in parasite control vs . pathology remains controversial [32–35] . A high frequency of T cells is found in peripheral blood of indeterminate ( i . e . C/A ) and cardiac ( i . e . C/S ) patients [35 , 36] , and CD8+ granzyme+ T cells were the main cell type found in infiltrating infiltrate in the myocardium [37] . However , recent studies have suggested that CD8+T cells found in C/A subjects were parasite-antigen specific and functional , while CD8+T cells undergoing immunological exhaustion were noted in C/S patients and their lack of activity contributed to the establishment of pathology [38] . A correlation between the production of inflammatory cytokines ( IFNγ > IL-10 ) by CD4+ T cells and monocytes of C/S patients , and the production of Th2 cytokine profile ( IL-10 and IL-4 ) by the same cells of C/A patients is also shown [39 , 40] . These studies tend to conclude that functional capacity of T cells along with anti-inflammatory activation of monocytes determines the control of parasite and clinically asymptomatic state in chagasic individuals while functionally incapable T cells and consistent proinflammatory activation of monocytes contributes to chronic , clinically symptomatic disease . IPA analysis of the proteome datasets in this study suggested that differential migration and/or invasion capacity of immune cells may also contribute to host’s ability to control T . cruzi and enter C/A vs C/S stage . An increase in cellular disassembly and disorganization associated with disruption of filaments that is central to remodeling of the cytoskeleton and modulation of cell shape for migration was observed in PBMCs of all chagasic patients ( S1 Fig ) . Specifically , the expression profile of Ca2+-dependent phospholipid-binding members of the annexin family that possess phospholipase A2 inhibitory activity [41] , vimentin and actin isoforms ( ACTB , ACTG ) that are the cytoskeletal component responsible for maintaining cell integrity and are mediators of internal cell motility [42] and filamin A ( FLNA ) that interacts with several molecules ( e . g . integrins ) to regulate the actin cytoskeleton organization [43] were all altered in PBMCs of chagasic subjects . However , the expression levels of small G proteins ( Rab14 , RAP1B ) that regulate membrane trafficking across golgi and endosomal compartments [44 , 45] and of Rab13 that controls junctional development by directly binding to F actin and modifying actin cytoskeletal reorganization [46] and cell spreading via filamins [47] were increased and decreased in C/A and C/S subjects , respectively , and might have played an important role in determining the extent of immune cell migration in C/A versus C/S chagasic subjects . Consistent with this , all seropositive chagasic subjects exhibited an expression profile indicative of increase in migration of phagocytes and leukocytes ( S3 Fig ) , though a small subset of molecules identified to be linked to invasion process ( 11 molecules , z score: -2 . 032 , p value: 1 . 43E-03; ANXA1↓ , ANXA2↓ , FLNA↓ , GSN↓ , LTF↑ , PKM↓ , S100A6↑ , SOD2↓ , THBS1↑ , VIM↓ , YY1↑ , S3 Fig panel B ) were decreased in C/S subjects , thus suggesting that functional lymphocytes may be mobilized in periphery but not able to access and kill tissue parasites . What might be the source of low-grade antigenic stimulus that results in persistence of immune cells and whether these surviving immune cells are functional in the context of parasite control is not entirely clear . Some investigators have argued that it is the long-term persistence of parasitic antigens that result in exhaustion of the functional T cell compartment [48 , 49] . The authors noted the frequency of parasite-specific functional CD4+ and CD8+ T cells decreased with more severe stages of clinical disease in human patients , and the T cells that persisted in chronically infected individuals were not metabolically or functionally active and exhibited the phenotypic characteristics of senescence [48 , 49] . Our data showed an increase in free radical synthesis and a decline in free radical catabolism and scavenging capacity in infected individuals that exhibited more pronounced disease state ( S4 Fig , panel B ) . We and others have shown that oxidative stress is persistent in chronically-infected chagasic animals and patients [14 , 17 , 50 , 51] , and oxidized cardiac proteins serve as neo-antigens and recognized by antibody response in chagasic mice and patients [25] . Thus , it is also possible that self-proteins that are oxidized due to persistence of oxidative stress serve as the source of antigenic stimulus for a low-grade but persistent activation of immune cells in chagasic host . The two hypotheses , i . e . , parasite or self-antigens contributing to persistence of non-functional , senescent immune cells are not mutually exclusive and together explain why the persistent chronic inflammation is of pathological importance in Chagas disease . The gene expression studies using global and custom arrays have shown the mitochondrial function-related gene expression is decreased in experimental models of T . cruzi infection and in the cardiac biopsies of chagasic patients [52–55] . A loss in the activity of mitochondrial respiratory complexes ( I and III ) was also noted in cardiac biopsies of chagasic rodents [14 , 56] and peripheral blood of human patients [17] that correlated with decreased coupled respiration and ATP generation [50 , 57] . In this study , PBMCs of chagasic patients showed protein expression pattern indicative of inhibition of glycolysis/gluconeogenesis ( ↓PKM , ↓GAPDH , ↓ENO1 , ↓ADLOA , and ↓PGK1 ) . The abundance of ATP5A1 that contributes to oxidative phosphorylation and ATP synthesis was counter-effected by abundance of MTCH1 that is localized to the mitochondrion inner membrane and induces Bax- and Bak- independent apoptosis [58 , 59] in chagasic PBMCs . Further , all isoforms of TUFM that participate in protein translation in mitochondria were decreased in chagasic PBMCs . Mutations in TUFM are shown to contribute to oxidative phosphorylation inefficiency and lactic acidosis in infantile encephalopathy [60] . These data provide a novel clue , and suggest that decreased translation and/or transport of mitochondria-targeted proteins affecting the functional assembly of electron transport chain complexes might play a major role in mitochondrial energy deficiency during progressive Chagas disease . The top upstream regulators , MYC/MYCN and SP1 were predicted to be inhibited ( z-score: < 2 , p<0 . 001 , all ) , and identified as common link contributing to expression profile of protein datasets related to metabolism , cell death/cell proliferation , ROS scavenging and cytoskeletal remodeling in chagasic subjects . MYC and MYCN are very strong proto-oncogenes that play a role in cell cycle , apoptosis and cellular transformation through diverse mechanisms . Recently , MYC has been reported to induce accumulation of DNA oxidative adducts and impair cell cycle regulatory capacity which potentially can increase the genomic instability and provide an environment conducive to growth of the cancer cells [61] . Others have shown MYC-dependent-ROS increase induced cell death [62] . Whether MYC-induced ROS contribute to tumorigenesis in human cells is not clearly demonstrated; however , in the context of chagasic subjects , our study suggests that the inhibition of MYC was likely an adaptive response to control pathological outcomes related to uncontrolled ROS production and immune cell proliferation . Indeed as early as 1992 , a selective reduction of c-myc and c-fos mRNAs in association with the severe suppression of the IL-2 gene in lymphoid of mice infected by T . cruzi was noted [63] . Like MYC , SP1 transcription factor also modulates the expression of genes involved in cell division , apoptosis , and immune responses . Post-translational modifications of SP1 are suggested to alter its DNA binding and transactivation activity and thereby affect the transcriptional activity [64] . Up regulation of SP1 is shown to be tumorigenic and its reduction was found to be neuroprotective in in vitro and in vivo models of Huntington’s disease [65] . PARP-1 , a member of the poly ( ADP-ribose ) polymerase family , produces poly ( ADP-ribose ) units ( PAR ) [66] and PAR modifications of SP1 suppressed its DNA-binding properties [67] . We have shown hyperactivation of PARP-1 stimulated by oxidative DNA damage in cardiomyocytes infected by T . cruzi [68] . How cross-talk of PARP-1 and SP1 determines the expression and transcriptional function of SP1 in the context of chronic chagasic cardiomyopathy remains to be elucidated in forthcoming studies . In summary , this study demonstrates that unbiased proteomic analysis of PBMCs in a discovery mode is useful in enhancing our knowledge of the pathomechanisms that determine predisposition to and progression of clinically symptomatic Chagas disease . By employing a 2DE and MALDI-TOF/MS approach for developing the PBMC proteome signature of chagasic subjects , we have identified the possible pathologic mechanisms in disease progression would involve host’s inability to recruit immune cells , scavenge free radicals , and prevent cell death . MYC/SP1 transcription factors that regulate hypoxia and inflammatory stress were predicted to be key targets for controlling chagasic pathology . MARS-modeling identified a panel of protein spots that if monitored in infected individuals , will have >93% success in predicting risk of clinical disease development . Our results provide an impetus for further studies in a second independent cohort of patients for confirming the diagnostic potential of suggested panel of proteins .
|
Chagasic cardiomyopathy is elicited by Trypanosoma cruzi infection . T . cruzi transmission is prevalent in Latin American countries , and its transmission is also noted in Mexico and Southern parts of the United States . In this manuscript , we have utilized blood samples from human subjects that were normal healthy or were infected with T . cruzi and exhibited variable symptoms of heart disease . We have employed a highly sensitive approach of protein labeling , developed a detailed proteomic map from all samples , performed comparative analysis of gel images , and identified a panel of proteins that were changed in abundance in clinically asymptomatic ( C/A ) and clinically symptomatic ( C/S ) chagasic individuals with respect to healthy controls . Functional annotation of these proteins suggested that pathologic mechanisms in disease progression would involve host’s inability to recruit immune cells , scavenge free radicals , and prevent cell death . We also describe a panel of proteins that can differentiate C/A from C/S subjects and will potentially be useful in identifying infected individuals at risk of developing clinical disease .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"immunology",
"tropical",
"diseases",
"dna-binding",
"proteins",
"parasitic",
"diseases",
"parasitic",
"protozoans",
"protozoans",
"neglected",
"tropical",
"diseases",
"infectious",
"disease",
"control",
"contractile",
"proteins",
"infectious",
"diseases",
"actins",
"animal",
"cells",
"proteins",
"protozoan",
"infections",
"biochemistry",
"trypanosoma",
"cruzi",
"cytoskeletal",
"proteins",
"trypanosoma",
"chagas",
"disease",
"cell",
"biology",
"proteomes",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] |
2016
|
Changes in Proteome Profile of Peripheral Blood Mononuclear Cells in Chronic Chagas Disease
|
N-Acetyl-L-Glutamate Kinase ( NAGK ) is the structural paradigm for examining the catalytic mechanisms and dynamics of amino acid kinase family members . Given that the slow conformational dynamics of the NAGK ( at the microseconds time scale or slower ) may be rate-limiting , it is of importance to assess the mechanisms of the most cooperative modes of motion intrinsically accessible to this enzyme . Here , we present the results from normal mode analysis using an elastic network model representation , which shows that the conformational mechanisms for substrate binding by NAGK strongly correlate with the intrinsic dynamics of the enzyme in the unbound form . We further analyzed the potential mechanisms of allosteric signalling within NAGK using a Markov model for network communication . Comparative analysis of the dynamics of family members strongly suggests that the low-frequency modes of motion and the associated intramolecular couplings that establish signal transduction are highly conserved among family members , in support of the paradigm sequence→structure→dynamics→function .
Many recent studies , both experimental and computational , point to the inherent ability of proteins to undergo , under native state conditions , large-amplitude conformational changes that are usually linked to their biological function . Proteins have access , via such equilibrium fluctuations , to an ensemble of conformers encoded by their 3-dimensional ( 3D ) structure; and ligand binding essentially shifts the population of these pre-existing conformers in favour of the ligand-bound form [1]–[4] . With the accessibility of multiple structures resolved for a given protein in different forms , it is now possible to identify the principal changes in structure assumed by a given protein upon binding different ligands , which are observed to conform to those intrinsically accessible to the protein prior to ligand binding [5]–[7] . The observations suggest the dominance of proteins' intrinsic dynamics in defining the modes of interactions with the ligands . This is in contrast to the induced-fit model [8] where the ligand ‘induces’ the change in conformation . Instead , the Monod-Wyman-Changeux ( MWC ) [9] model of allostery where a selection from amongst those conformers already accessible is triggered upon ligand binding . Yet , the choice between intrinsic vs induced dynamics , and the correlations between dynamics and function , are still to be established , and presumably depend on the particular systems of study [10] . NMR relaxation experiments provide evidence , for example , for the existence of correlations between the time scales of large-amplitude conformational motions and catalytic turnover [11] , [12]; and collective motions in the low frequency regime appear to be potentially limiting reaction rates . On the other hand , other studies point to the different time scales and events that control catalysis and binding events [13] , [14] . Furthermore , while the intrinsic dynamics in the unbound form is observed to be the dominant mechanism that facilitates protein-protein or protein-ligand complexation , the ligand may also promote structural rearrangements on a local scale at the binding site [2] , [15] , [16] . Given that proteins' collective dynamics , and thereby potential functional motions , are encoded by the structure , proteins grouped in families on the basis of their fold similarities would be expected to share relevant dynamical features [17]–[21] . It is of paramount importance , in this respect , to have a clear understanding of collective motions and their relationship to binding or catalytic activities , if any , toward gaining deeper insights into functional mechanisms shared by members of protein families . Protein dynamics can be explored by means of all-atom force fields and simulations , or by coarse-grained ( CG ) models and methods . All-atom simulations such as Molecular Dynamics ( MD ) describe the conformational fluctuations of the system over a broad range of timescales . Except for small proteins , the main limitation of MD is that the timescales computationally attainable ( below hundreds of nanoseconds ) do not allow for accurate sampling of slow and large-amplitude motions ( low-frequency modes ) that are usually of biological interest . CG approaches , on the other hand , lack atomic details but provide insights into global movements . Among them , Elastic Network Models ( ENMs ) have found wide use in conjunction with normal mode analyses ( NMAs ) in the last decade [22] . ENMs describe the protein as a network , the nodes of which are usually identified by the spatial positions of Cα-atoms . Elastic springs of uniform force constant connect the nodes in the simplest ( most broadly used ) ENM , referred to as the anisotropic network model ( ANM ) [23]–[25] . Despite the oversimplified description of the protein conveyed by the ENMs , a surge of studies have shown that the predicted low-frequency modes describe well experimentally observed conformational changes and provide insights into potential mechanisms of function and allostery [5]–[7] , [24]–[27] , in accord with NMAs performed [28] , [29] with more detailed models and force fields . Additionally , recent studies by Orozco and co-workers [30] , and Liu et al [31] point to the similarities of the conformational space described by the low-frequency modes obtained from MD and that from CG NMA , provided that MD runs are long enough to accurately sample the collective motions . The present study focuses on the amino acid kinase ( AAK ) family . This family comprises the following enzymes on the basis of sequence identity and structural similarities: N-acetyl-L-glutamate ( NAG ) kinase ( NAGK ) , carbamate kinase ( CK ) , glutamate-5-kinase ( G5K ) , UMP kinase ( UMPK ) , aspartokinase ( AK ) and the fosfomycin resistance kinase ( FomA ) . Rubio and co-workers [32] have exhaustively studied this family and proposed that the shared fold among the members is likely to give rise to a similar mechanism of substrate binding and catalysis . NAGK is the most widely studied member of this family taking into account the large amount of structural information gathered [32] , [33] . This enzyme indeed serves as a structural paradigm for the AAK family , such that studying its structure-encoded dynamics can shed light on the mechanisms shared by family members to perform their function [32] . NAGK catalyzes the phosphorylation of NAG , which is the controlling step in arginine biosynthesis . The hallmark of this biosynthetic route in bacteria is that it proceeds through N-acetylated intermediates , as opposed to mammals that produce non-acetylated intermediates . Consequently , NAGK activity may be selectively inhibited and , taking into account that it is the controlling enzyme of arginine biosynthesis , it is a potential target for antibacterial drugs . In many organisms , NAGK phosphorylation is the controlling step in arginine biosynthesis . In these cases , NAGK is feedback inhibited by the end product arginine , and recent studies shed light on this mechanism of inhibition [34] , [35] . NAGK from Escherichia Coli ( EcNAGK ) , on the other hand , is arginine-insensitive . Its mechanism of phosphoryl transfer has been the most thoroughly characterized among the enzymes that catalyze the synthesis of acylphosphates ( EC group 2 . 7 . 2 ) . In particular , crystallographic studies by Rubio and coworkers [32] , [33] have provided insights into its mechanisms of binding and catalysis . EcNAGK is a homodimer of 258 residues , each monomer being folded into an αβα sandwich ( Figure 1 ) . The N-domain of each subunit/monomer makes intersubunit contacts and hosts the NAG binding site ( NAG lid ) , whereas the C-domain binds the ATP . The phosphoryl transfer reaction takes place at the interface between the two domains within each subunit . Kinetic studies show no evidence of cooperativity between subunits [36] , suggesting that the dimeric structure provides thermodynamic stability , only , to the monomeric fold that has been evolutionary selected to perform the catalytic function . The diverse crystallographic structures solved for the bound state of this enzyme indicate two types of functional motions [33]: ( 1 ) X-ray structures of EcNAGK complexed with either ADP or with the inert ATP analogue AMPPNP ( PDB codes 1GS5 , 1OH9 , 1OHA and 1OHB ) have a too narrow active site to let the substrates bind directly; whereas the unbound structure ( PDB code 2WXB; kindly provided by the authors prior to release ) has a more open active site . This suggests that the enzyme undergoes a conformational closure that is likely to be triggered upon nucleotide binding , since all these complexes display a closed structure whether NAG is bound or not . ( 2 ) The ternary complex with ADP and NAG displays the ability to exchange NAG with a sulphate ion in solution without opening the active site . The NAG lid therefore must be able to open and close independently of other structural elements . The aim of the present study is two-fold . Firstly , given the interest in acquiring deeper knowledge on the enzymatic mechanism of EcNAGK and the potential role of slow dynamics in the pre-disposition of the enzymatic function , we analyze here the low-frequency modes of motion of EcNAGK . Secondly , using EcNAGK as the paradigm of AAK family , we assess to what extent the slow modes of motion are shared by other members of the AAK family .
The results are organized as follows . First , results from GNM analysis are presented , which give insights into the functional significance of residue fluctuations and underlying sizes of motions in the most readily accessible ( i . e . softest ) collective modes . Second , ANM modes are described to analyze the directionality/mechanism of these modes . Note that GNM does not provide information on 3N-dimensional structural changes , but on N-dimensional properties such as the mean-square fluctuations ( MSFs ) of residues , their cross-correlations , or movements along normal mode axes , hence the use of the ANM for exploring and visualizing the 3D motions ( see Methods ) . Third , communication properties of the EcNAGK enzyme are assessed based on graph theoretical examination of shortest paths between network nodes representative of the enzyme . Finally , a comparative analysis of the ANM dynamics of different members of the AAK family is made . GNM and ANM modes of EcNAGK are computed for the open form in general , except for the analysis of the intrinsic dynamics of the closed form; and ligands are not included in the calculations . The predicted motions therefore reflect the intrinsic dynamics of the enzyme in the absence of bound ligands . Figure 2 displays the results from the GNM analysis of the equilibrium dynamics of EcNAGK . Panel ( B ) compares the MSFs of residues , < ( ΔRi ) 2> , predicted by the GNM with those indicated by X-ray crystallographic B-factors Bi = 8π2/3 < ( ΔRi ) 2> . For clarity , the different structural elements are numbered and color-coded along the upper abscissa bar in accord with the colors in panel ( A ) . Results for chains A and B are identical as a result of the dyadic axis of symmetry at the intersubunit surface . Calculations and experimental data refer to the open form of EcNAGK . The high correlation coefficient ( r = 0 . 75 ) between the experimental and theoretical curves in Figure 2 is remarkable in view of the simplicity of the GNM , but it is also worth noting that in the case of the closed conformation the correlation coefficient drops to r = 0 . 61 . Indeed , ENMs tend to provide a better description of the dynamics of open forms [24] . The mobility profile in Figure 2B permits us to identify the most mobile and rigid regions of the protein from the maxima and minima , respectively . Mainly two dynamical features are distinguished . First , the β3–β4 hairpin , which corresponds to the NAG-binding site lid , is the most mobile part of the N-domain ( region 3 ) . Second , the β12–β13 hairpin and αF and αG helices ( regions 9 and 10 ) emerge as the most mobile parts of the C-domain; notably , these structural elements are involved in ATP binding . It is remarkable that the topology of the structure provides flexibility near the two binding sites , which may be a functional requirement to accommodate ligand binding . Rigid/constrained elements , on the other hand , include the αC helices making intersubunit contacts , along with the strands β8 and β10 in the N-domain core . Moreover , the N-termini of αB and αE helices ( regions 2 and 8 ) , which point toward the γ-phosphate in the closed form , also show reduced mobility . The lack of mobility in these NAGK sequence motifs [32] is presumably a dynamic requirement to optimally perform their functional role in orienting their dipoles to withdraw negative charge from the transferring phosphate group . These results are consistent with those inferred by Rubio and co-workers from their crystallographic studies [32] , [33] , The decomposition of the global dynamics into a set of GNM modes permits us to identify the different kinds of motions allowed by the structure as well as the couplings between different parts of the protein . Moreover , minima in the low-frequency mode-profiles reveal mechanically important residues . When residues surrounding a given site move in opposite directions , the latter site serves as a hinge . Hinge sites at low-frequency modes , also called soft modes , usually serve as key mechanical site at the interface between domains subject to concerted movements [37] . Figure 3 shows the mobility of different parts of the protein in the first three softest modes . The diagrams are color coded from red ( most rigid ) to blue ( most mobile ) , in accord with the size of motions undergone by the residues along these examined modes' axes ( shown on the right panels ) . All three modes appear to induce motions symmetrically distributed about the inter-subunit interface . In the 1st mode , the mobility increases with distance away from the dyadic axis , such that the C-domain , and in particular the β12–β13 hairpin and αF helix , undergo the largest movements . The 2nd slowest mode involves movements of the C- and N-domains with respect to each other within each monomer . A hinge site at residue A174 is observed , where previous crystallographic studies had exactly set the boundary between the C- and N-domains [32] . This hinge presumably enables the opening/closure of the active site in each subunit . A mutant on residue D162 [36] , which is close to this hinge site , disrupted function and thus confirms this hinge as a key element in the functionality of the enzyme . On the other hand , the 3rd mode involves mainly the β3–β4 hairpin , i . e . , the NAG lid , and suggests an intrinsic ability at this region to move independently with respect to the ATP site ( note that all modes are orthonormal and independent ) . Such local flexibility is consistent with an ability of the NAG lid to open and close the NAG binding site , in support of the hypothesis inferred from crystallographic studies [33] . The anticorrelated motion of the C-domain , due to a hinge at residue E181 , is minimal but , as in mode 2 and together with the movement of NAG lid , might lead to the opening/closure of the active site . Interfacial residues making intersubunit contacts exhibit low mobilities , suggesting that the tightly packed hydrophobic contacts are essential to the thermodynamic stability of the dimeric protein . With the aim of gaining insights into the directionality of these modes , one can map GNM modes into ANM modes by comparing the mean-square fluctuations . A one-to-one correspondence would not be expected , due to differences in the number of modes as well as underlying potentials of the two ENMs . We found that the first GNM mode correlates with the 1st and 3rd ANM modes; the 2nd with the 1st , 2nd and 4th ANM modes; and the 3rd is the counterpart of the 5th ANM mode . The directionality provided by the ANM approach helps us ascertain how well the slowest ANM modes describe the conformational difference observed between the open and closed structures resolved by X-ray crystallography . To this aim , the ANM modes are projected into the deformation vector Δr obtained from the open and closed conformations . Figure 4 displays the cumulative overlap ( see equation ( 7 ) ) between Δr and the ANM modes for both passages ( from closed to open , and vice versa ) . It is worth emphasizing that the first 10 ANM modes of the open and closed forms are able to describe 84% and 76% of the observed conformational change , respectively . On the other hand , the open form requires a smaller set of modes to describe the deformation vector to a given extent . This is in agreement with the fact that the dynamics of open conformations are usually better described with ENM as also noted above . Modes 1 , 3 and 5 are the main contributors to the cumulative overlap and thus the most relevant modes to describe the global dynamics of NAGK ( see Figure S1 ) . As mentioned above three modes play a dominant role in enabling the functional changes in NAGK . Modes 1 and 3 drive a symmetrical opening and closing of the active site . In both modes , the most mobile region is the C-domain , which binds ATP , while the N-domain is practically rigid . Mode 5 ensures the opening/closing of the NAG-binding site by the β3–β4 hairpin that serves as a lid . It is of the utmost importance to examine how active site residues move in these modes . Do they possess an intrinsic ability to adopt the conformation of the bound state , or does ligand binding trigger the conformational change ? To explore this issue , we generated a series of conformations driven by these modes . Figure 5 illustrates the results for mode 1 . This mode entails an anticorrelated movement of the two subunits as shown in panel ( A ) . The following features are distinguished by a closer examination of functional sites . The catalytic residues K8 , K217 and D162 , and those in the vicinity , such as N158 , exhibit minimal changes in their coordinates as seen in panel ( B ) . On the other hand , a number of hydrophobic residues near the ATP binding site move concertedly in a direction required for coordinating ATP , as the subunit reconfigures from the open to the closed form ( see panel ( C ) ) . This mode accessible in the absence of ATP binding thus facilitates the suitable re-positioning of these residues upon ATP binding . NAG-binding residues undergo minimal change during this global motion ( panel ( D ) ) . The movement of residues in mode 3 complement those in the 1st mode to reach the bound conformation from the open form ( Figure S2 ) . K8 and D162 are key catalytic residues on the basis of structural [32] , [33] and mutational [36] studies . D162 is inferred from these studies to play a critical role in properly positioning two lysines ( K8 , K217 ) that stabilize the negative charge of ATP . The minimal displacements of D162 and K8 in these global modes , and the intrinsic tendency of K217 to move toward D162 , are presumably dynamic requirements to optimally perform their catalytic roles ( note that D162 is located close to the hinge site of GNM mode 2 , as pointed out above , and thus its mobility is rather constrained ) . This rigidity is confirmed by the striking similarity in the orientation of these residues in different bound states of the enzyme [33] that characterize the entire catalytic process . The rearrangements of catalytic residues may be necessary to optimally orient , or pre-organize , the ligands to catalyze the chemical reaction [13] , [14] , [38] . In this case , some additional changes appear to occur in the bound form , such as the change in the side chain conformation of K217 , which exchanges a salt bridge between residues E181 and D162 . These rearrangements would presumably take place upon ligand binding , since E181 interacts with ATP via hydrogen bonds . In relation to the NAG binding process , mutants on the NAG binding site revealed that N158 and R66 are key residues that underlie the affinity of EcNAGK for NAG [36] . By examining the NAG binding mode ( 5th ANM mode , see Figure S3 ) , R66 was found to be far more flexible than N158 . This suggests that R66 may play a role in the recognition of the ligand , whereas the less exposed residue N158 might subsequently aid to fix the position of NAG at the active site . Furthermore , upon NAG binding , the size of the hydrophobic pocket ( L65 , R66 , V122 and N160 ) that hosts the methyl group of NAG is reduced upon correlated movements between R66 and L65 toward the closed form . Binding of R66 and N158 to NAG , thus , apparently guides L65 toward the methyl group of the substrate . The correlated movements of L65 together with the rigidity of V122 and N160 fix the size of the hydrophobic pocket , which has been observed to be unable to bind glutamate derivatives with larger N-acyl groups [32] , [39] ( Figure S3 ) . Using the Markov model described in the Methods , we computed the hitting times Hji . Hji provides a measure of the average path length over all possible combinations of edges , required to send information to a given node j , or ‘hit’ residue j , starting from node i . The hitting times for all pairs of residues were evaluated for three different cases: open form ( NAGK ( O ) ) , closed from without ligands ( NAGK ( C ) ) and closed form with ligands ( NAGK ( C ) +ligands ) . Toward gaining an understanding of the communication propensity of individual residues , results have been consolidated , by calculating the mean hitting time , <Hi> = ( 1/N ) Σj Hji , for each residue i . Figure 6A displays the mean hitting times of all residues in the three cases . The main contribution to Hji arises from the MSF of the target residue itself ( via the term [Γ−1]jj in equation ( 8 ) , which in turn is proportional to < ( ΔRj ) 2> - see equation ( 3 ) ) . As a result , the average hitting time profile shares some characteristics with the MSF profile shown in Figure 2B . The minima correspond to the most efficient receivers; these exhibit minimal fluctuations in their positions . It is worth noting that catalytic residues are among those with the lowest hitting times . These results suggests that the structural position and contact topology of the active site have been evolutionary designed to effectively receive signals from the binding sites and other parts of the protein so as to optimize the catalytic activity of the enzyme . On the other hand , the ligand-binding residues exhibit a broader variety of hitting times . A closer comparison of the results obtained for the three structures revealed an interesting feature upon examination of the average hitting times between different substructures . The results in Figure 6C display the average path lengths evaluated for the communication between such particular domains in each structure: the average path lengths over all residue pairs belonging to the respective C-terminal and N-terminal domains ( blue curves ) , those over all the N-terminal domain and catalytic site residues ( red ) , and those over the C-terminal domain and catalytic site residues ( green ) . These results clearly demonstrate that the closure of the structure enhances the communication of residues ( decreases the average hitting times or path lengths ) and upon ligand binding the communication shows a further improvement . Panel ( D ) demonstrates that not only the average path lengths , but the variance in the path lengths decrease upon domain closure , and ligand binding . In all cases , the N- and C-domains exhibit average path lengths longer than those connecting either domain to the catalytic site . This is a natural consequence of the location of catalytic residues - at the inter-domain region , where the phosphoryl transfer takes place . Figure 7 panels ( A ) and ( B ) illustrates the three types of communication pathways in the open and closed states . Three residues have been selected as endpoints representative of the N-terminal domain NAG-binding site ( R66 ) , C-terminal domain nucleotide-binding site ( L209 ) and the catalytic site at the inter-domain interface ( D162 ) , and the residues along the shortest paths evaluated using the Dijkstra's algorithm ( see Methods ) are shown by different dots in each case ( see caption ) . We note in panel ( C ) that the ligand in the closed+liganded structure effectively spans the optimal communication pathway . The enhancement of communication observed in panels ( C ) and ( D ) of Figure 6 is a consequence of the rigidity imparted by the closure of the structure and by ligand binding . The structure obviously becomes more cohesive in the closed conformation and consequently the couplings between residue fluctuations are increased , or the fluctuations in inter-residue distances are reduced . As summarized in the methods and derived in detail in our previous work [40] , the commute times τij between residue pairs directly scale with the fluctuations in the corresponding inter-residue distances ( see equation ( 9 ) ) . Restrictions in inter-residue distance fluctuations acquired upon closure of the structure thus necessarily induce an enhancement in communication . From a catalytic point of view , the closure of the structure upon substrate binding is presumably an efficient way to optimize signal transduction and facilitate the catalytic process . Panel ( B ) in Figure 6 displays the changes in the contribution to hitting times from cross-correlations ( ) evaluated using equation ( 8 ) ( see Methods ) twice , for the open and closed+liganded states , and taking their difference . It is observed that upon domain closure and ligand binding the contribution from cross-correlations within the C-domain decrease ( blue points in the panel ) , whereas those between the N- and C-domains within each subunit increase ( red points in the panel ) . The resultant shorter and more homogeneous communication pathways suggest that ligands tend to centralize the communication between the C- and N- domains . Therefore , the transmission of conformational signals between the flexible domains and the more rigid catalytic residues takes place across the substrates . This might indicate a way to cooperatively optimize substrate binding ( or product release ) or even couple the intrinsic enzyme dynamics to the catalysis of the chemical reaction . It is of interest to determine if the dynamical features observed for EcNAGK are shared by other members of the AAK family . Various approaches can be adopted to this aim [19] , [21] , [41] . Here , we focus on global dynamics and compare the ANM modes predicted for EcNAGK with those predicted for each AAK member . First , each pair of enzymes is structurally aligned using the DALI server [42] . Second , we define our ‘subsystem’ as the aligned portions of the families members and constructed the Hessian submatrices for these portions , and the remaining chain segments are considered as environment , similar to the approach adopted by Zheng & Brooks [43] , described in Methods ) . Third , NMA is performed using equation ( 11 ) for the Hessian of the examined system . The correlation cosines between the collective modes evaluated for each subsystem and those calculated for the EcNAGK dimer are presented in Figure 8 . Results are shown for the top-ranking 10 modes , calculated for the corresponding dimers of three different members of the AAK family ( structures shown in Figure 8 ) : Carbamate kinase from Pyrococcus furiosus ( PfCK ) , NAGK from Thermotoga Maritima ( TmNAGK ) and UMPK from Pyrococcus furiosus ( PfUMPK ) . Further information on the structural and dynamical pairwise comparisons is provided in Table 1 . The present study focused on the EcNAGK dimer in order to provide new insights into the competition between intrinsic vs induced dynamics in controlling enzymatic activity , assessing which residues play a key role in mediating the collective motions , or which conformational mechanisms are shared among members of the AAK family . The most probable modes of motion encoded by the structure have been determined using the available structural data for EcNAGK dimer , which is used as a prototype , as well as other members of the family . The present study illustrates that this family , not only has important sequence and structure similarities , but also shares relevant dynamical features ( Figure 8 ) . The results in Figure 4 demonstrate that the conformational change observed between the open and closed forms of EcNAGK are essentially accomplished by movements along a small subset of modes ( among the complete set of 1542 modes accessible to the enzyme ) ; these are the modes predicted by the GNM to be the softest , i . e . , they incur the least ascent in energy for a given size of motion . This shows that the ‘easiest’ movements intrinsically favoured by the enzyme structure are actually those underlying its ligand binding mechanism , suggesting an evolutionary optimization of structure to favour functional dynamics . Closer examination of the changes on a local scale , on the other hand , shows that the changes induced at residue level , may have the right directions to facilitate the interactions with the substrate , but are not sufficiently large and specific enough to explain the re-positioning of amino acids near active sites . The ATP-bound conformation appears , for example , to result from the intrinsic dynamics of the protein combined with local rearrangements induced by the ligand to optimize its interactions in the complex ( Figure 5 ) . The low frequency modes shared among the examined family members are shown here to enable access to the active site , by opening/closing to the environment the cleft where catalytic residues reside . Notably , these movements have minimal effects on the organization of catalytic residues , which are located near the hinge center that allows for the relative movements of the N- and C-domains in each subunit . The restricted mobility of catalytic residues is consistent with previous observations where catalytic sites have been pointed out to be highly constrained in the global modes and occupy key positions ( near or coinciding with global hinges ) in the structure . The collective modes do not therefore induce distinctive rearrangements at the catalytic residues . However , they appear to modulate their communication to the environment , i . e . they provide exposure to solvent , and/or a loosening or tightening of the packing density in the neighbourhood which apparently plays a role in controlling the propagation of structural or energetic perturbations to/from the active site . Perhaps the most striking results concern the communication properties and the role of domain movements and ligand binding in enhancing allosteric effects . The decrease in the mean values and dispersion of the hitting times and the communication path lengths upon ligand-binding ( Figure 6 ) suggests that the ligands optimize the coupling of domain movements that are already characteristic of the intrinsic protein dynamics . Indeed , the structure may have been evolutionary selected to bind the substrates in an optimal position to maximize the allosteric couplings .
The GNM potential depends on the vectorial distance between each pair of nodes as ( 1 ) where N is the total number of residues , γ the uniform force constant for all springs in the network , Γij is the ijth element of the N×N Kirchhoff matrix Γ that defines the connectivity of the network , equal to −1 if residues i and j are within a cutoff distance Rc , zero otherwise , Rij and Rij0 are the instantaneous and equilibrium distance vectors between residues i and j , residue positions being identified by those of their α-carbons in the PDB files . The GNM approach allows for decomposing the dynamics of the protein into a set of normal modes of motion upon eigenvalue decomposition of Γ . The contribution of the kth mode to the MSF of residue i is expressed as ( 2 ) where λk and uk are the kth eigenvalue and eigenvector of Γ , respectively; ( uk ) i designates the mobility of residue i along the kth mode . The low-frequency modes usually have the highest degree of collectivity , and they make the largest contribution to the observed MSFs ( 3 ) where the summation is performed over all non-zero modes and [Γ−1]ii designates the ith diagonal element of the inverse of Γ . Therefore , the lowest frequency modes usually provide insights into the cooperative motions involved in biological function [22] . GNM provides information on the relative sizes of residue motions in different modes ( equation ( 2 ) ) , the MSFs of individual residues ( equation ( 3 ) ) , or their cross-correlations ( 4 ) ( obtained by rewriting equation ( 3 ) for pairs of residues i and j ) , as but not on their directionality; the fluctuations are implicitly assumed to be isotropic . The 3D characterization of the normal modes is provided by the ANM . The ANM potential is a function of the scalar distance between the interacting pair of nodes and is given by [23] ( 5 ) Both GNM and ANM penalize inter-residue distance changes , but GNM also takes into account the orientational change of the inter-residue vector , which leads to better agreement with experimental B-factors [54] . NMA is carried out using the 3N×3N Hessian matrix H derived from the ANM potential . The diagonalization of H yields 3N-6 non-zero modes ( as opposed to N-1 in the GNM ) . A given ANM mode ( eigenvector ) thus contains information on the x- , y- and z-components of the motion undergone by each residue , thus describing the spatial directionalities of the collective motions . ANM modes can be used to generate alternative conformations sampled along most easily accessible ( lowest frequency ) mode directions . Due to the harmonic character of the potential , two sets of conformers are obtained for a given mode k: ( 6 ) where λk and are the eigenvalue and eigenvector for mode k respectively , R0 is the 3N-dimensional vector representing the initial coordinates and sk is a parameter that scales the amplitude of the deformation induced by mode k . No sidechain atomic coordinates are included in the ANM calculations . An all-atom model for the deformed structure is generated by displacing the backbone and side chain atoms of each residue along the mode component of the corresponding Cα-atom and subsequent energy minimization . Such energy minimization performed with Gromacs [55] was verified to involve negligible conformational change in the backbone . The degree of similarity between a conformational change Δr observed by crystallography and the theoretically predicted direction of the kth mode can be quantified with the correlation cosine , cos ( Δr· ) . Here Δr refers to the 3N-dimensional difference vector between the α-carbon coordinates of the open form and closed form of NAGK , for example , after optimal superimposition of the two structures to eliminate the external degrees of freedom . The cumulative overlap between the experimentally observed deformation Δr and that accounted for by a subset of m modes ( m < 3N-6 ) is given by a summation of squared correlation cosines as ( 7 ) The summation of the squared cosines over all 3N-6 nonzero modes is identically equal to unity as the eigenvectors form a complete orthonormal basis set in the 3N-6 dimensional space of internal conformational changes . In the absence of correlation between Δr and , the average correlation cosine squared contributed by mode k will thus be 1/ ( 3N-6 ) . Note the strong departure from this random behaviour in Figure 4 . Inter-residue communication has been suggested as an essential mechanism in the allosteric regulation of protein function and enzymatic catalysis [56] , and explored in diverse computational studies [57]–[59] . A network-based Markov model has been recently developed [40] , [60] , which reconciles the NMA-based predictions on allosteric changes in conformations ( global modes ) with the shortest path ( s ) analyses based on graph theoretical methods [28] . We use this method to identify communication paths . The interactions between residue pairs are defined therein by the affinity matrix A . The elements of this matrix are defined as [60] aij = Nij/ ( Ni Nj ) ½ where Nij is the number of atom-atom contacts between residues i and j , based on a threshold distance of 4 Å , and Ni , Nj are the number of heavy atoms of both residues . A is related to the Kirchhoff matrix ( same as GNM , except for the adoption of affinities , instead of γ , for the weights of the edges ) as = D−A , where D is the diagonal matrix of elements . In the simplified case where aij = 1 for all Rij<Rc , reduces to the GNM . The network communication is controlled by the Markov transition matrix M = {mij} , where mij = aij/dj represents the conditional probability of transmitting a signal from residue j to residue i in one time step [60] . We define −log ( mij ) as the ‘distance’ between two residues , in terms of communication , and the maximum-likelihood paths associated with each residue pair are evaluated using the Dijkstra's algorithm [60] . This permits us to evaluate a basic communication property: hitting time Hji as the average path length for the passage of signals from node i to node j [40] . Hji can be expressed in terms of the elements of ( or ) as [40] ( 8 ) Given that the elements of Γ−1 scale with the MSFs of residues ( diagonal elements ) or the cross-correlations between residue fluctuations ( off-diagonal elements ) , the above equation establishes the link between the signal transduction properties of the protein and its collective dynamics [40] . Note that the commute time τij = Hij + Hji assumes an even simpler form , using equation ( 8 ) twice , i . e . , ( 9 ) This equation simply states that the communication between two residues takes longer if their inter-residue distances have higher fluctuations [40] . The inter-residue distances , in turn , are readily evaluated from the difference < ( ΔRij ) 2> = < ( ΔRi ) 2> + < ( ΔRj ) 2> −2 < ( ΔRi • ΔRj> , where the respective terms are evaluated using the equations ( 3 ) and ( 4 ) . If the dynamics of a part of the protein ( subsystem , S ) in the presence of an environment ( E ) is of interest , a useful approach is to partition the Hessian into four submatrices [43]: ( 10 ) where HSS is the matrix referring to the subsystem , HEE to that associated with the interactions within the environment and HSE ( and HES ) to those coupling the subsystem to the environment . An effective Hessian for the subsystem can be constructed from these elements as ( 11 ) The NMA of effectively describes the collective dynamics of the subsystem in the presence of coupling to the environment . This approach proved useful in determining the allosteric potential of residues [61] or the location of transition states of chemical reactions by defining a reduced potential energy surface [62] . The above method has been used for evaluating the overlap between the collective modes of different family members in different environments . The cumulative overlap ( expressed in terms of percentage in Table 1 ) between subsets of modes is evaluated using equation ( 7 ) where a double summation over the particular subsets of modes of interest , e . g . top ranking 10 modes of the two systems . All figures depicting molecular structures have been generated with the VMD visualization software [63] .
|
During the last 20 years both the experimental and computational communities have provided strong evidence that proteins cannot be regarded as static entities , but as intrinsically flexible molecules that exploit their fluctuation dynamics for catalytic and ligand-binding events , as well as for allosteric regulation . This intrinsic dynamics is encoded in the protein structure and , therefore , those proteins with similar folding should share dynamic features essential to their biological function . In this work , we have applied an Elastic Network Model to predict the large-amplitude dynamics of different enzymes belonging to the same protein family ( Amino Acid Kinase family ) . Subsequent comparison of the dynamics of these proteins reveals that this protein family follows the same dynamic pattern . The present results are strongly supported by experimental data and provide new insights into the performance of biological function by these enzymes . The investigation presented here provides us with a useful framework to identify dynamic fingerprints among proteins with structural similarities .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] |
[
"computer",
"science/numerical",
"analysis",
"and",
"theoretical",
"computing",
"biophysics",
"biochemistry",
"computational",
"biology"
] |
2010
|
On the Conservation of the Slow Conformational Dynamics within the Amino Acid Kinase Family: NAGK the Paradigm
|
Typhoid fever , caused by Salmonella enterica serovar Typhi ( S . Typhi ) , remains a serious global health concern . Since their emergence in the mid-1970s multi-drug resistant ( MDR ) S . Typhi now dominate drug sensitive equivalents in many regions . MDR in S . Typhi is almost exclusively conferred by self-transmissible IncHI1 plasmids carrying a suite of antimicrobial resistance genes . We identified over 300 single nucleotide polymorphisms ( SNPs ) within conserved regions of the IncHI1 plasmid , and genotyped both plasmid and chromosomal SNPs in over 450 S . Typhi dating back to 1958 . Prior to 1995 , a variety of IncHI1 plasmid types were detected in distinct S . Typhi haplotypes . Highly similar plasmids were detected in co-circulating S . Typhi haplotypes , indicative of plasmid transfer . In contrast , from 1995 onwards , 98% of MDR S . Typhi were plasmid sequence type 6 ( PST6 ) and S . Typhi haplotype H58 , indicating recent global spread of a dominant MDR clone . To investigate whether PST6 conferred a selective advantage compared to other IncHI1 plasmids , we used a phenotyping array to compare the impact of IncHI1 PST6 and PST1 plasmids in a common S . Typhi host . The PST6 plasmid conferred the ability to grow in high salt medium ( 4 . 7% NaCl ) , which we demonstrate is due to the presence in PST6 of the Tn6062 transposon encoding BetU .
Typhoid fever remains a serious public health problem in many developing countries , with highest incidence in parts of Asia ( 274 per 100 , 000 person-years ) and Africa ( 50 per 100 , 000 person-years ) [1] , [2] . The causative agent is the bacterium Salmonella enterica serovar Typhi ( S . Typhi ) . While vaccines against S . Typhi exist , it is mainly restricted groups such as travellers [3] , [4] and individuals enrolled in large vaccine trials [5] who are immunized , and antimicrobial treatment remains central to the control of typhoid fever [3] . However antimicrobial resistant typhoid has been observed for over half a century and is now common in many areas . Chloramphenicol resistant S . Typhi was first reported in 1950 , shortly after the drug was introduced for treatment of typhoid [6] . By the early 1970s , S . Typhi resistant to both chloramphenicol and ampicillin had been observed [7] and multidrug resistant ( MDR ) S . Typhi ( defined here as resistance to chloramphenicol , ampicillin and trimethoprim-sulfamethoxazole ) emerged soon after [8] . The rate of MDR among S . Typhi can fluctuate over time and geographical space , as can the precise combination of drug resistance genes and phenotypes [9] , [10] . However in many typhoid endemic areas , an increasing prevalence of MDR S . Typhi was observed in the late 1990s [11] , [12] , [13] , and MDR typhoid now predominates in many areas [9] , [14] including parts of Asia [15] , [16] , Africa [17] and the Middle East [18] , [19] , [20] , [21] . MDR S . Typhi with reduced susceptibility to fluoroquinolones are increasingly common [9] , [15] , [16] , [22] , leaving macrolides or third generation cephalosporins as the only options for therapy [23] , [24] . In S . Typhi the MDR phenotype is almost exclusively conferred by self-transmissible plasmids of the HI1 incompatibility type ( IncHI1 ) [8] , [11] , [25] , [26] , [27] , [28] , [29] , [30] , although other plasmids are occasionally reported [31] . In the laboratory , IncHI1 plasmids can transfer between Enterobacteriaceae and other Gram-negative bacteria [32] and in nature , IncHI1 plasmids have been detected in pathogenic isolates of Salmonella enterica and Escherichia coli [33] , [34] , [35] , [36] . However it remains unclear whether the increase in MDR typhoid is due to the exchange of resistance genes among co-circulating S . Typhi or to the expansion of MDR S . Typhi clones . Efforts have been made to investigate variability within IncHI1 plasmids [29] , [33] , [37] or their S . Typhi hosts [22] , [38] , [39] , [40] , [41] but little progress has been made in linking the two together to answer fundamental questions of how MDR typhoid spreads . We recently developed a plasmid multi-locus sequence typing ( PMLST ) scheme for IncHI1 plasmids , which identified eight distinct IncHI1 plasmid sequence types ( PSTs ) among S . Typhi and S . Paratyphi A isolates , including five PSTs found in S . Typhi [37] . This pattern was not consistent with a single acquisition of an IncHI1 plasmid in S . Typhi followed by divergence into multiple plasmid lineages , rather it indicated that divergent IncHI1 plasmids have entered the S . Typhi population on multiple occasions [37] . However the phylogenetic relatedness of the S . Typhi hosts was not determined , thus we were unable to estimate how many times plasmids may have been independently acquired . In this study , we aimed to investigate the relative contribution of plasmid transfer , as opposed to the expansion of plasmid-bearing S . Typhi clones , to the emergence of MDR typhoid . We found evidence for plasmid transfer in older S . Typhi . However the vast majority of recent MDR typhoid was attributable to a single host-plasmid combination ( S . Typhi H58-IncHI1 plasmid ST6 ) . We performed further experiments to investigate possible mechanisms for the success of this host-plasmid combination , and identified a transposon in PST6 that confers tolerance to high osmolarity .
The bacterial isolates analyzed by SNP assay are summarized in Table 1 and listed in full in Table S1 . DNA was extracted using Wizard Genomic DNA purification kits ( Promega ) according to manufacturer's instructions . Details of the isolates used for competition experiments are also listed in Table S1 . BRD948 is an attenuated Ty2-derived strain ( also known as CVD908-htrA ) , which has deletion mutations in aroC ( t0480 ) , aroD ( t1231 ) , and htrA ( t0210 ) [42] . The growth of BRD948 on LB agar or in LB broth was enabled by supplementation with aromatic amino acid mix ( aro mix ) to achieve the final concentration of 50 µM L-phenylalanine , 50 µM L-tryptophan , 1 µM para-aminobenzoic acid and 1 µM 2 , 3-dihydroxybenzoic acid . Plasmid sequences were downloaded from the European Nucleotide Archive ( plasmid details and accessions in Table 2 ) . SNPs between finished plasmid sequences were identified using the nucmer and show-snps algorithms within the MUMmer 3 . 1 package [43] , via pairwise comparisons with pAKU_1 . To identify SNPs in S . Typhi PST6 IncHI1 plasmids , 36 bp single-ended Illumina/Solexa sequencing reads from S . Typhi isolates E03-9804 , ISP-03-07467 and ISP-04-06979 were aligned to the pAKU_1 sequence using Maq [44] and quality filters as described previously [45] . SNPs called in repetitive regions or inserted sequences were excluded from phylogenetic analysis , so that phylogenetic trees were based only on the conserved IncHI1 core regions . This resulted in a total of 347 SNPs , which were analyzed using BEAST [46] to simultaneously infer a phylogenetic tree and divergence dates ( using the year of isolation of each plasmid as listed in Table 1 , resulting tree in Figure 1 ) . Parameters used were as follows: generalised time reversible model with a Gamma model of site heterogeneity ( 4 gamma categories ) ; a relaxed molecular clock with uncorrelated exponential rates [46] , a coalescent tree prior estimated using a Bayesian skyline model with 10 groups [47] , default priors and 20 million iterations . The chromosomal haplotype of S . Typhi isolates was determined based on the SNPs present at 1 , 485 chromosomal loci identified previously from genome-wide surveys [41] , [45] and listed in [22] , [39] . IncHI1 plasmid haplotypes were determined using 231 SNPs located in the conserved IncHI1 backbone sequence , listed in Table S2 ( note these do not include SNPs specific to pMAK1 or pO111_1 which were not available at the time of assay design , nor any SNPs falling within 10 bp of each other as these cannot be accurately targeted via GoldenGate assay; however additional SNPs identified via plasmid MLST [37] were included , see Table S2 ) . Resistance gene sequences were interrogated using additional oligonucleotide probes , listed in [16] . All loci were interrogated using a GoldenGate ( Illumina ) custom assay according to the manufacturer's standard protocols , as described previously [16] , [22] , [39] . SNP calls were generated from raw fluorescence signal data by clustering with a modified version of Illuminus [48] as described previously [22] . The percentage of IncHI1 SNP loci yielding positive signals in the GoldenGate assay clearly divided isolates into two groups , indicating presence of an IncHI plasmid ( signals for >90% of IncHI1 loci ) or absence of such a plasmid ( signals for <10% of IncHI1 loci ) , see Figure 2 . SNP alleles were concatenated to generate two multiple alignments , one for chromosomal SNPs and one for IncHI1 plasmid SNPs . Maximum likelihood phylogenetic trees ( Figure 3 ) were fit to each alignment using RAxML [49] with a GTR+Γ model and 1 , 000 bootstraps . PCR primers were designed using Primer3 [50] according to the following criteria: melting temperature 56°C , no hairpins or dimers affecting 3′ ends , no cross-dimers between forward and reverse primers . Primer sequences are given in Table 3 . PCRs were performed on a TETRA DNA Engine Peltier Thermal Cycler ( MJ Research ) with a reaction consisting of 1 . 2 µl of 10X Mango PCR buffer , 1 . 5 mM MgCl2 , 25 µM of each dNTP , 1 . 25 U Mango Taq ( Bioline ) , 0 . 3 µM of each primer , 1 . 0 µl DNA template ( approx . 100 ng ) and nuclease free water in a total reaction volume of 12 µl . Cycling conditions were as follows: 5 min at 94°C , 30 cycles of 15 s at 94°C , 15 s at 58°C , and 60 s at 72°C; final extension of 5 min at 72°C . The transfer of pHCM1 and pSTY7 from respective E . coli transconjugants to the attenuated S . Typhi BRD948 was performed by cross-streaking onto LB agar supplemented with aro mix and incubating at 37°C overnight . The growth was harvested , resuspended in 2 ml of dH20 , plated on MacConkey agar containing streptomycin ( 1 µg/ml or 5 µg/ml ) and chloramphenicol ( 5 µg/ml or 20 µg/ml ) and incubated overnight at 37°C . BRD948 transconjugants were confirmed by antimicrobial susceptibility patterns ( disk diffusion ) and colony PCR specific for BRD948 background ( primers 5939-5′-CGTTCACCTGGCTGGAGTTTG-3′ and5940-5′-CATGCCAGCAGCGCAATCGCG-3′ ) and pHCM1 or pSTY7 plasmids ( Insert1056L- 5′-TAGGGTTTGTGCGGCTTC-3′ and Insert1056R-5′-CCTTCTTGTCGCCTTTGC-3′ ) . The competition between BRD948 ( pHCM1 ) and BRD948 ( pSTY7 ) was started with equal inoculums of roughly 5×103 cfu each in 10 mL of LB broth supplemented with aro mix and chloramphenicol ( 5 µg/mL ) . The culture was incubated for 16 hours at 37°C with shaking . Approximately 104 cfu of this culture were then used to inoculate the next passage . The cultures were passaged for a total of 4 days . Samples were collected at time point 0 ( at the time of initial inoculation ) and after 1 , 2 , 3 and 4 days of passage , diluted and spread on LB agar supplemented with aro mix . Sixty-four colonies from each sample were randomly picked and tested by PCR to identify their plasmid type ( see below ) . The entire competition assay was performed in triplicate , i . e . beginning with three initial cultures of equal inoculums of the two isolates . The colony PCR was perform using standard condition ( see PCR section above ) with three primers ( DF 5′-CGATTTGTGAAGTTGGGTCA-3′ , DR2 5′- CAACCTGGGCAGGTGTAAGT-3′ and DR3 5′- TTCGTTACGTGTTCATTCCA-3′ ) . Expected sizes of PCR products were 511 bp for BRD948 ( pHCM1 ) and 285 bp for BRD948 ( pSTY7 ) . Four individual competitive growth assays were performed using wildtype host-plasmid combinations genotyped using the GoldenGate assay ( isolates listed in Table S1 ) ; H58-C vs . H1 , H58-E1 vs . H1 , H58-C-ST6 vs . H1-ST1 and H58-E1-ST6 vs . H1-ST1 . Bacterial isolates were recovered from frozen stocks onto Luria-Bertani ( LB ) media , supplemented with 20 mg/ml of chloramphenicol for isolates with MDR plasmids . Individual colonies were picked and used to inoculate 10 ml of LB broth , which were incubated overnight at 37°C with agitation . Bacterial cells were enumerated the following day by serial dilution and plating . Equivalent quantities of the two competing S . Typhi isolates were inoculated into 10 ml of LB broth and were incubated as before ( Day 0 ) . The competition assays were conducted by growing the mixed bacteria to stationary phase and then passaging them into 10 ml of LB broth in a 1∶1000 dilution in triplicate over four days . One ml of media containing bacteria from each of the triplicates was stored at −80°C at each time point . DNA was extracted from the frozen samples by boiling for 10 minutes , samples were pelleted , the supernatant was removed and used as template in all of the subsequent competitive real-time PCR reactions ( below ) , which were performed on each template in duplicate . We performed two individual competitive real-time PCRs ( Taqman system ) with LNA probes to calculate the proportions of S . Typhi H1 vs . S . Typhi H58 and S . Typhi H58-C vs . S . Typhi H58-E1 in aliquots of DNA extracted from broth following competitive growth . These assays were performed to accurately calculate the relative proportion of the isolates in all competitive assays , including those that could not be calculated by plating alone . The haplotype specific primers and probes were designed using Primer Express Software ( Applied Biosystems ) and manufactured by Sigma-Proligo ( Singapore ) . Primer and probe sequences were as follows ( capital letters indicate the position of LNA and the letters in square brackets indicate the SNP position ) ; H58 vs H1 ( 99 bp amplicon ) : F ( 71–83 ) -CCGAACGCGACGG , R ( 169-157 ) -TGCGGCACACGGC and probe 5′-FAM-ccggtAat[G]gtAatGaagc-BHQ1 ( S . Typhi H1 ) and 5′-Hex-ccggtAat[A]gtAatGaagc ( S . Typhi H58 ) ; H58-C vs H58-E1 ( 89 bp amplicon ) : F ( 60–75 ) -ACCCTGCACCGTGACC , R- ( 148–135 ) -GCATGATGCCGCCC and probe 5′-FAM-ttcCag[G]ccAtgAcgc –BHQ1 ( S . Typhi H58-C ) and 5′-HEX-ttcCag[A]ccAtgAcgc-BHQ1 ( S . Typhi H58-E1 ) . PCR amplification were performed using a light cycler ( Roche , USA ) , with hot start Taq polymerase ( Qiagen , USA ) under the following conditions , 95°C for 15 minutes and 45 cycles of 95°C for 30 seconds , 60°C for 30 seconds and 72°C for 30 seconds . As the primer locations were identical for the internal competitive PCR assay , the efficiency of the PCR was also considered to be identical . Therefore , proportions of isolates at the various time points throughout the assay were calculated by taking the mean of six Cp values ( each competition assay was performed in triplicate and the PCR was performed in duplicate ) . The Mean Cp values for each competitive assay was converted into a proportion ( isolate A ) using the following calculation: Proportion isolate A = 1/ ( 2−ΔCp +1 ) , where ΔCp = Cp ( isolate B ) – Cp ( isolate A ) . Phenotype microarrays of osmotic/ionic response ( PM 9 ) , pH response ( PM 10 ) and bacterial chemical sensitivity ( PM 11 to 20 ) were performed as described previously by Biolog Inc . ( Hayward , California USA ) [51] . BRD948 was used as a reference for comparison with BRD948 ( pHCM1 ) or BRD948 ( pSTY7 ) test isolates to identify the phenotypes affected by the presence of IncHI1 plasmid pHCM1 ( PST1 ) or pSTY7 ( PST6 ) . The three isolates were pre-grown on LB ( Luria-Bertani ) agar plates supplemented with 1X of an aromatic amino acid mix ( a 50X aromatic amino acid mix consisted of 50 µM L-phenylalanine , 50 µM L-tryptophan , 1 µM para-aminobenzoic acid and 1 µM 2 , 3-dihydroxybenzoic acid ) . Sterile cotton swabs were used to pick colonies and suspend them in 10 ml inoculating media IF-0a ( Biolog ) , the optical density of which was then adjusted to 0 . 035 absorbance units at 610 nm . A total of 750 µl of this cell suspension was diluted 200 fold into 150 ml inoculating media IF-10 ( Biolog ) , containing 1X aromatic acid mix ( 1 . 2X Biolog media , 22 ml of sterile water and 3 ml of 50X aromatic amino acid mix ) . PM microtitre plates 9–20 were inoculated with 100 µl of the inoculating media cell suspension per well . Microtitre plates were then incubated at 37°C for 48 h in the Omnilog ( Biolog Inc ) and each well was monitored for colour change ( kinetic respiration ) . Tests were performed in duplicate and the kinetic data was analyzed using the OmniLog PM software set ( Biolog Inc ) . A lower threshold of 80 omnilog units ( measured as area under the kinetic response curve ) was set , and the phenotypes of each of the three isolates were compared . The fragment of two CDSs within Tn6062 of pSTY7 ( 3405 bp ) was amplified using two primers IS1056-03 ( 5′-CAGGCACCGTTTTCTTATTAGAATCTTCGCCACT-3′ ) and IS1056-04 ( 5′-TCATTGAACTTTGCTACCCTGA-3′ ) . The pACYC184 fragment ( 2033 bp ) containing its p15A ori and chloramphenicol resistant gene ( cmR ) was amplified using pACYC184-01 ( 5′-AAAATTACGCCCCGCCCTGC-3′ ) and pACYC184-03 ( 5′-TAATAAGAAAACGGTGCCTGACTGCGTTAGCA-3′ ) . The two fragments were then fused together by overlapping primer extension PCR ( pACYC184-03 and IS1056-03 were two overlapping primers ) using pACYC-01 and IS1056-04 primers . All three PCRs above were performed using PfuUltra II Fusion HS DNA Polymerase ( Agilent , former Stratagene , UK ) to achieved highly accurate amplification . The PCRs were set up following the manufacturer's manual with the specific annealing temperature of 58°C and extension time of 45 s for Tn6062 and pACYC184 fragments or 1 . 5 min for the fusion fragment . The fused PCR product was re-circularised by T4 ligase ( New England BioLabs , UK ) to form pACYC184Δtet::Tn6062 and electroporated into BRD948 . The pACYC184 fragment was also re-circularised to form the empty vector pACYC184Δtet and electroporated into BRD948 . Overnight bacterial cultures of BRD948 ( pHCM1 ) , BRD948 ( pSTY7 ) , BRD948 ( pACYCΔtet ) and BRD948 ( pACYCΔtet::Tn6062 ) were diluted by distilled water to the cell suspension of 0 . 1 OD600 before 1 µl of the cell suspension was inoculated into 200 µl of 0 . 8 M NaCl LB broth ( supplemented with aro mix ) in a well of a 96-well plate . Each isolate was inoculated into six wells ( i . e . six biological replicates ) . The bacteria were grown at 37°C with shaking at 300 rpm and OD600 was measured automatically every 15 minutes for 24 hours in the Optima plate reader ( BMG Labtech , Germany ) . Absorbance data were collected and saved in Excel format for graphing .
We compared the DNA sequences of eight ∼200 kbp IncHI1 plasmids isolated from enteric pathogens ( Table 2 ) and identified a conserved IncHI1 core region ( >99% identical at the nucleotide level ) that included the tra1 and tra2 regions encoding conjugal transfer [29] , [33] , [37] , [52] . Subsequently , we identified 347 single nucleotide polymorphisms ( SNPs ) within these conserved regions , which were used to construct a phylogenetic tree of IncHI1 plasmids and to estimate the divergence dates of internal nodes of this tree based on the known isolation dates for each plasmid [53] ( Figure 1 ) . The tree topology is in general agreement with that inferred previously using a plasmid MLST approach [37] . The sequences of the three most recent S . Typhi plasmids ( isolated 2003–2004 ) were very closely related and correspond to a previously defined plasmid sequence type ( PST ) known as PST6 [37] ( Figure 1 , red ) . According to our divergence date estimates , the most recent common ancestor ( mrca ) shared by these three plasmids existed circa 1999 ( Figure 1 ) . The PST6 plasmids were also closely related to the PST7 plasmid pAKU_1 from S . Paratyphi A ( Figure 1 , orange ) , with mrca circa 1992 . The plasmids pHCM1 , pO111_1 and pMAK1 formed a distinct group corresponding to PST1 , with mrca circa 1989 ( Figure 1 , green ) . The eighth reference plasmid R27 ( PST5 ) was quite distinct from the others , with an estimated divergence date of 1952 ( Figure 1 , black ) . In addition to the conserved IncHI1 core regions , the plasmids each harbour insertions of drug resistance elements . These include transposons Tn10 ( encoding tetracycline resistance ) , Tn9 ( encoding chloramphenicol resistance via the cat gene ( SPAP0067 ) ) , strAB ( SPAP0152-SPAP0153 , SPAP0230-SPAP0231; encoding streptomycin resistance ) , sul1 and sul2 ( SPAP0132 , SPAP0151; encoding sulfonamide resistance ) , dfrA7 ( SPAP0133; encoding trimethoprim resistance ) and blaTEM-1 ( SPAP0143; encoding ampicillin resistance ) [29] , [33] , [54] . The insertion sites of these elements , confirmed using PCR ( Tables 3 & 4 ) , differed between lineages of the IncHI1 phylogenetic tree ( Figure 1 , grey ) . All plasmid sequences included Tn10 , however three different insertion sites were evident ( Table 4 ) , suggesting the transposon was acquired by IncHI1 plasmids on at least three separate occasions ( Figure 1 , grey ) . Tn9 was present in all plasmids other than R27 , however the insertion site in PST6 and PST7 plasmids differed from that in PST1 , suggesting at least two independent acquisitions . It was previously noted that pHCM1 ( PST1 ) and pAKU_1 ( PST7 ) share identical insertions into Tn9 of a sequence incorporating Tn21 ( including sul1 , dfrA7 ) , blaTEM-1 , sul2 , and strAB [33]; here we found this insertion into Tn9 was conserved in all PST1 and PST6 plasmid sequences . Together , this composite set of drug resistance elements encodes MDR ( resistance to chloramphenicol , ampicillin and trimethoprim-sulfamethoxazole ) . In order to investigate the contribution of distinct IncHI1 plasmid types over time to the emergence of MDR S . Typhi , we performed high resolution SNP typing of S . Typhi chromosomal and IncHI1 plasmid loci in a global collection of 454 S . Typhi , isolated between 1958–2007 ( Table 1 , Table S1 ) . These isolates include 19 S . Typhi isolates sequenced previously [45] and 22 S . Typhi isolated from Kenya in 2004–2007 [22] . We also typed eight IncHI1 S . Typhi plasmids harboured in E . coli transconjugants [29] , [37] . SNP typing was performed using the GoldenGate ( Illumina ) platform to simultaneously assay chromosomal and plasmid SNP loci . We targeted 231 SNPs from the conserved region of the IncHI1 plasmid ( Table S2 , [37]; note 116 of the 347 identified SNPs were not able to be included in the GoldenGate assay , see Methods ) and 119 from resistance genes and associated transposons ( see [16] ) . Of the 454 S . Typhi that we typed , 193 ( 43% ) harboured IncHI1 plasmids , which clustered into nine distinct haplotypes ( Figure 3B ) . As expected , the majority of IncHI1 plasmids harboured multiple resistance genes or elements including Tn10 , Tn9 , strAB , sul1 , sul2 , dfrA7 and blaTEM-1 . Transposon insertion sites were confirmed for representative plasmids using PCR ( Table 4 ) and agree with the patterns of insertion sites determined by sequencing ( Figure 1 & 3B ) . Thirteen IncHI1 plasmids were identified among S . Typhi isolated prior to 1994 ( Table 5 ) , including seven of the total nine distinct IncHI1 plasmid haplotypes ( Figure 3B ) . A total of 26 distinct S . Typhi haplotypes were identified by typing of chromosomal SNPs; their phylogenetic relationships are shown in Figure 3A . The PST2 plasmid was detected in three S . Typhi haplotypes isolated in Asia between 1972 and 1977 ( Table 5 ) , consistent with repeated introduction of closely related IncHI1 plasmids into distinct S . Typhi hosts . Similarly , PST8 was present in two S . Typhi haplotypes from Peru in 1981 ( Table 5 ) [55] , consistent with transfer of the PST8 plasmid among multiple S . Typhi haplotypes co-circulating in Peru at this time . Significantly , from 1995 onwards , nearly all IncHI1 plasmids were type PST6 ( 180/184 plasmids , 98% ) . Remarkably , there was an exclusive relationship between PST6 plasmids and S . Typhi haplogroup H58 , with all PST6 plasmids found in S . Typhi H58 hosts , and no S . Typhi H58 harbouring non-PST6 plasmids ( although 35% of S . Typhi H58 were non-MDR and plasmid-free ) . This strongly suggests that the apparent rise in MDR typhoid since the mid-1990s [11] , [12] , [13] is due to the clonal expansion of H58 S . Typhi carrying the MDR PST6 plasmid . This is in contrast to the longer-term situation described above , which showed that in the years following the first emergence of MDR typhoid ( 1970s–1980s ) , MDR IncHI1 plasmids had transferred repeatedly into distinct co-circulating S . Typhi haplotypes . The clonal expansion of H58 S . Typhi has been documented previously [22] , [41] , however the role of the PST6 plasmid has not been investigated . Among our collection , the oldest S . Typhi H58 isolate dates back to 1995 and carries the PST6 plasmid . To ascertain whether the common ancestor of S . Typhi H58 might have carried the PST6 plasmid , the phylogenetic structure among our 293 S . Typhi H58 isolates was resolved using 45 of the assayed SNP loci that differentiate within the H58 haplogroup ( Figure 4 ) . These SNPs divided the isolates into 24 distinct H58 haplotypes , with the majority ( N = 270 ) in 13 haplotypes ( Figure 4 ) . Most of the H58 haplotypes ( N = 14 ) , including the ancestral haplotype A , included isolates harbouring the PST6 plasmid ( Figure 4 ) . We have previously sequenced the genomes of 19 S . Typhi , including seven isolates from the H58 haplogroup [45] , and observed the insertion of an IS1 transposase between protein coding sequences STY3618 and STY3619 within all sequenced H58 S . Typhi genomes . This transposase was identical at the nucleotide level to the IS1 sequences within Tn9 in IncHI1 plasmids pHCM1 and pAKU_1 , and shared a common insertion site in all seven S . Typhi H58 chromosomes sequenced [45] . In the present study , our SNP assays included a probe targeting sequences within the IS1 gene ( SPAP0007 ) . Nearly all of the S . Typhi H58 isolates gave positive signals for this IS1 target ( Figure 4; coloured or white ) , with the sole exception of six isolates belonging to the H58 ancestral haplotype A ( Figure 4 , grey ) , which also included three isolates that carried the PST6 plasmid and tested positive for IS1 ( Figure 4 , purple ) . This suggests that the PST6 plasmid was likely acquired by the most recent common ancestor of S . Typhi H58 ( Figure 4 , haplotype A ) , followed by transposition of IS1 into the S . Typhi chromosome prior to divergence into subtypes of H58 . Thus the dominance of PST6 over other MDR IncHI1 plasmids ( noted here and previously [37] ) and the dominance of H58 over other S . Typhi haplotypes ( noted here and previously [22] , [41] ) appears to be the result of a trans-continental clonal expansion of MDR S . Typhi H58 carrying the PST6 plasmid . These results indicate that the recent global spread of MDR typhoid is attributable to the emergence of a single plasmid-host combination ( H58-PST6 ) . We were able to transfer the PST6 plasmid pSTY7 from S . Typhi to E . coli [29] and back to S . Typhi ( data not shown ) , confirming that the PST6 plasmid retains the ability to transfer between bacteria via conjugation , yet we found no evidence of PST6 transfer in natural S . Typhi populations ( above ) . This raises the question of why this particular plasmid-host association has been so successful and exclusive . To investigate whether PST6 could confer any selective advantage over other IncHI1 plasmids harbouring similar antimicrobial resistance genes , representative PST6 ( pSTY7 ) and PST1 ( pHCM1 ) IncHI1 plasmids from Vietnamese S . Typhi were introduced into a common S . Typhi BRD948 host , derived from S . Typhi Ty2 ( haplotype H10 ) . The PST1 plasmid pHCM1 was chosen for comparison since its complete sequence is available [54] and it was previously observed to be common in MDR S . Typhi in Vietnam in the early 1990s , just prior to the emergence of PST6 in S . Typhi in Vietnam and elsewhere [29] . BRD948 ( pHCM1 ) grew to three times the number of cfu compared to BRD948 ( pSTY7 ) after 4 days of mixed growth in LB broth ( Figure 5 , black ) . We therefore hypothesized that the advantage conferred by PST6 plasmids , if any , might be related to specific environmental conditions or to plasmid-host compatibility . To test the latter , we compared the growth of wildtype PST1-bearing S . Typhi H1 and PST6-bearing S . Typhi H58 isolated from typhoid patients in Vietnam and Pakistan and genotyped using the GoldenGate assay ( listed in Table S1 ) . The two PST6-bearing S . Typhi H58 isolates tested were both able to out compete the PST1-bearing H1 isolate , so that S . Typhi H1 was barely detectable after four days of competitive growth ( Figure 5 , red ) . However plasmid-free S . Typhi H58 isolates were also able to outcompete a plasmid-free S . Typhi H1 isolate ( Figure 5 , blue ) , thus we cannot confirm the plasmid plays a role in the competitive advantage of H58-PST6 S . Typhi over and above that of the H58 chromosomal haplotype . To screen for conditions under which PST6 plasmids confer an advantage compared to PST1 plasmids , we used Biolog phenotyping arrays to compare the growth of plasmid-free S . Typhi BRD948 to BRD948 ( pHCM1 ) and BRD948 ( pSTY7 ) under a wide variety of conditions including various pH levels and osmotic/ionic strengths , and a wide variety of antibiotics and chemicals [51] . As expected , both IncHI1 plasmids conferred enhanced growth in the presence of a wide range of antibiotics including amoxicillin , azlocillin , oxacillin , penicillin G , phenethicillin , chloramphenicol , streptomycin , gentamicin , tetracyclines and trimethoprim ( Table S3 ) . BRD948 ( pHCM1 ) displayed some minor growth advantages in the presence of additional antimicrobials , however none of these reached clinically relevant levels ( Table S3 ) . The only conditions under which BRD948 ( pSTY7 ) grew better than BRD948 and BRD948 ( pHCM1 ) was under high osmotic stress ( 3-5% NaCl or 6% KCl ) ( Table S3 ) . We confirmed this phenotype by inoculating each isolate into high salt concentration media ( 0 . 8 M NaCl LB broth , approx . 4 . 7% NaCl ) ; only the PST6-bearing isolate BRD948 ( pSTY7 ) was able to grow under these conditions ( Figure 6 , red and grey ) . We hypothesised that the osmotolerant properties of PST6 plasmids may be explained by the presence of two putative transporters encoded within a composite transposon , Tn6062 ( SPAP0100 , SPAP0105 , SPAP0106 , SPAP0110; this transposon was referred to as Ins1056 in [37] ) . Tn6062 was present in all PST6 plasmids , the novel subtype of PST1 ( 57Laos ) and two of the three PST8 plasmids , but absent from all other isolates ( detected via two Tn6062-specific probes included in our SNP typing assay ) . To determine if Tn6062 was responsible for the osmotolerant phenotype of BRD948 ( pSTY7 ) , the two putative transporter genes from Tn6062 ( SPAP0105 and SPAP0106 ) were inserted into the plasmid vector pAYCY184 and we assessed their effect on S . Typhi BRD948 in high salt concentration medium ( 0 . 8 M NaCl LB broth , approx . 4 . 7% NaCl ) . BRD948 ( pAYCY184-Tn6062 ) was able to grow at a slightly lower rate than BRD948 ( pSTY7 ) ( Figure 6 , blue ) , while BRD948 carrying the empty pAYCY184 vector was unable to grow ( Figure 6 , black ) . Therefore the transposon Tn6062 carried by the PST6 IncHI1 plasmids confers an osmotolerant phenotype on its S . Typhi host .
Our analysis of IncHI1 plasmid sequences indicates that plasmids responsible for the MDR phenotype in S . Typhi are closely related to those associated with MDR in other enteric pathogens including S . Paratyphi A , S . Choleraesuis and enterohaemorrhagic E . coli O111:H- ( Figure 1 , Table 2 ) . These plasmids share a recent common ancestor approximately six decades old and have evolved into several distinct lineages via accumulation of point mutations , followed by acquisition of resistance elements and further point mutation ( Figure 1 ) . Simultaneous SNP typing of plasmid and host enabled us to differentiate between the clonal expansion of MDR S . Typhi , and independent acquisitions of related MDR plasmids by distinct S . Typhi hosts . Evidence for the latter includes the detection of PST2 and PST8 plasmids in co-circulating S . Typhi isolates of distinct haplotypes in the 1970s and 1980s ( Table 5 ) . This indicates that the emergence of MDR typhoid during this period was in part due to transfer of IncHI1 plasmids within local S . Typhi populations . One of the PST2-S . Typhi combinations ( chromosomal haplotype H42 ) was later detected among two isolates from Africa in 2003–2004 , suggesting that an individual IncHI1 plasmid may be able to persist in a single host haplotype for decades ( Table 5 ) . In stark contrast , all 193 PST6 plasmids were observed in S . Typhi of the H58 haplotype and virtually all MDR S . Typhi observed after 1995 belonged to the same PST6-H58 clone , indicative of global spread of MDR typhoid via clonal expansion . Since humans are the only known reservoir for S . Typhi [56] , it is likely that trans-continental spread of this clone depends on international travel or migration . If this is the case it will be particularly difficult to control since S . Typhi can be transmitted by asymptomatic carriers [57] , [58] , who are usually unaware of their status and are difficult to detect [59] , [60] . Our data suggest that the PST6 plasmid was acquired by the most recent common ancestor of S . Typhi H58 ( Figure 4 ) , implying that the expansion of S . Typhi H58 did not begin until after acquisition of the plasmid . To our knowledge , the oldest confirmed S . Typhi H58 isolate is 9105928K [41] , which was isolated in India in 1991 and is MDR ( Mia Torpdhal , personal communication ) . This suggests that the initial expansion of S . Typhi H58 may have been associated with the acquisition of the PST6 plasmid , implying a selective advantage over and above that of MDR , which was also conferred by other IncHI1 plasmid types circulating in S . Typhi in the 1990s . The only growth advantage we detected for PST6 plasmids via our phenotype screen was that of osmotolerance , which we showed to be conferred by the Tn6062 transposon carried by PST6 plasmids . The transposon Tn6062 includes betU ( SPAP0106 ) , which encodes a betaine uptake system capable of transporting glycine betaine and proline betaine [61] . It was first described in E . coli isolates causing pyelonephritis ( ascending urinary tract infection ) and is believed to be an osmoregulator , allowing E . coli to survive the high osmolarity and urea content in urine [61] . However the gene is distributed among E . coli with a range of pathogenic phenotypes , so its osmoprotectant properties may be useful in other environmental contexts [62] . It is possible that enhanced osmotolerance may enhance survival of S . Typhi in specific niches within the human body , including the gut , gall bladder , urinary tract or liver . It is also possible that the ability to grow in the presence of high salt concentrations might enable S . Typhi to continue replicating in certain environments outside the host , which may lower the infectious dose or enhance the possibility of transmission by increasing the level of S . Typhi contamination in certain environments . This may have contributed to the selection of PST6 over other IncHI1 plasmids previously circulating among S . Typhi and the initial clonal expansion of H58 S . Typhi , however questions remain as to why the PST6 plasmid has not been detected among non-H58 S . Typhi . The PST6 plasmid appears to have been lost from H58 S . Typhi in some areas where the recommended treatment of typhoid was switched to fluoroquinolones , including Nepal and Vietnam [39] , [63] , [64] , while it has been maintained in areas such as Kenya where chloramphenicol is still commonly used to treat typhoid [17] , [22] . This confirms that antimicrobial use exerts a strong selective pressure for maintenance of the IncHI1 plasmid among S . Typhi and indicates that in the absence of such pressure any additional advantages conferred , including the increased osmotolerance described above , is not enough to maintain the PST6 plasmid indefinitely .
|
Typhoid fever is caused by the bacterium Salmonella enterica serovar Typhi ( S . Typhi ) . Treatment relies on antimicrobial drugs , however many S . Typhi are multi-drug resistant ( MDR ) , severely compromising treatment options . MDR typhoid is associated with multiple drug resistance genes , which can be transferred between S . Typhi and other bacteria via self-transmissible plasmids . We used sequence analysis to identify single nucleotide polymorphisms ( SNPs ) within these plasmids , and used high-resolution SNP typing to trace the subtypes ( termed haplotypes ) of both the S . Typhi bacteria and their MDR plasmids isolated from more than 450 typhoid patients since 1958 . Among isolates collected before 1995 , a variety of plasmid haplotypes and S . Typhi haplotypes were detected , indicating that MDR typhoid was caused by a diverse range of S . Typhi and MDR plasmids . In contrast , 98% of MDR S . Typhi samples isolated from 1995 were of the same S . Typhi haplotype and plasmid haplotype , indicating that the recent increase in rates of MDR typhoid is due to the global spread of a dominant S . Typhi-plasmid combination . We demonstrate this particular plasmid type contains a transposon encoding two transporter genes , enabling its S . Typhi host to grow in the presence of high salt concentrations .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"medicine",
"bacterial",
"diseases",
"infectious",
"diseases",
"bacterial",
"and",
"foodborne",
"illness",
"gastroenterology",
"and",
"hepatology",
"population",
"genetics",
"biology",
"evolutionary",
"biology"
] |
2011
|
Emergence of a Globally Dominant IncHI1 Plasmid Type Associated with Multiple Drug Resistant Typhoid
|
Enteropathogenic Escherichia coli ( EPEC ) represents a major causative agent of infant diarrhea associated with significant morbidity and mortality in developing countries . Although studied extensively in vitro , the investigation of the host-pathogen interaction in vivo has been hampered by the lack of a suitable small animal model . Using RT-PCR and global transcriptome analysis , high throughput 16S rDNA sequencing as well as immunofluorescence and electron microscopy , we characterize the EPEC-host interaction following oral challenge of newborn mice . Spontaneous colonization of the small intestine and colon of neonate mice that lasted until weaning was observed . Intimate attachment to the epithelial plasma membrane and microcolony formation were visualized only in the presence of a functional bundle forming pili ( BFP ) and type III secretion system ( T3SS ) . Similarly , a T3SS-dependent EPEC-induced innate immune response , mediated via MyD88 , TLR5 and TLR9 led to the induction of a distinct set of genes in infected intestinal epithelial cells . Infection-induced alterations of the microbiota composition remained restricted to the postnatal period . Although EPEC colonized the adult intestine in the absence of a competing microbiota , no microcolonies were observed at the small intestinal epithelium . Here , we introduce the first suitable mouse infection model and describe an age-dependent , virulence factor-dependent attachment of EPEC to enterocytes in vivo .
Gastrointestinal infections remain a major cause of morbidity and mortality in the pediatric population worldwide . Among them , infections with enteropathogenic Escherichia coli ( EPEC ) have been recognized to exhibit a great pathogen-attributable risk of death in infants aged 0–11 months [1] . Insight into the interaction between EPEC and the host has mostly been derived from in vitro studies using immortalized cell lines . These studies demonstrated that type IV bundle forming pili ( BFP ) mediate the initial contact between EPEC and the host epithelial cell and are responsible for the typical localized adherence pattern observed at the epithelial surface [2–4] . The bacterium-cell interaction is further strengthened by the translocation of the translocated intimin receptor ( Tir ) via the type III secretion system ( T3SS ) , resulting in the formation of typical attaching and effacing ( A/E ) lesions [5] . Additional effector molecules translocated by the T3SS were shown to induce massive cytoskeletal reorganization , manipulate host cell signaling and induce epithelial apoptosis [6–9] . In the past , the lack of a suitable small animal model has prevented a detailed analysis of the host-microbial interactions during infection in vivo [10] . EPEC infections have already been examined in larger animals such as rabbits , pigs or calves [9 , 11] . These models , however , are not amenable to genetic modifications and germ-free animals are not widely available . In addition , Citrobacter rodentium , a natural mouse enteropathogen , is used as a model organism for EPEC infection . However , although C . rodentium shares many features with EPEC , their tissue tropism , histopathology and clinical symptoms after infection differ . Therefore in vivo epithelial host responses to EPEC , protective antimicrobial host factors as well as the influence of the enteric microbiota and the consequences of EPEC infection on host-microbial homeostasis have all remained ill-defined . Here we present the establishment of a new oral model of EPEC infection using neonate mice . Oral administration induced effective intestinal colonization . Bacterial attachment to the epithelial apical surface in vivo was associated with the generation of A/E lesion-like focal microcolonies dependent on the presence of functional BFP and T3SS . Transcriptome and RT-PCR analysis of wildtype and gene-deficient animals illustrated the epithelial response to EPEC infection and identified the innate immune receptors involved . High throughput 16S rDNA sequencing revealed infection-induced alterations of the developing microbiota . Finally , microcolony formation was shown to be restricted to the neonatal period despite efficient colonization of adult animals in the absence of a competitive enteric microbiota . Thus , we present a new oral EPEC infection model and demonstrate the age-restricted development of typical features associated with EPEC infection .
Initially , 0 . 5 to 1x105 CFU EPEC ( strain E2348/69 ) were orally administered to mice at different ages and bacterial colonization was monitored at 4 days post infection ( p . i . ) . Animals infected during their first week of life exhibited efficient intestinal colonization with high bacterial numbers recovered from small intestinal and colon tissue . Significantly lower numbers of colonizing bacteria were noted in the small and large intestine of animals infected after the age of 10 and 13 days , respectively ( Fig 1A and 1B ) . Oral infection of adult animals even with high bacterial number ( 0 . 5 to 1x108 CFU ) did not lead to detectable colonization ( Fig 1A and 1B ) . Subsequently , we analyzed the duration of the colonization following oral administration to 1-day-old neonates . Efficient colonization was noted during the first 12 days following bacterial challenge ( Fig 1C and 1D ) . Colonization along the length of the small intestine took place primarily in the distal part ( S1A Fig ) . A significant reduction in bacterial colonization was first observed at day 16 p . i . and day 20 p . i . in the small intestine and colon , respectively . Despite the high intestinal burden , no bacterial spread to systemic organs occurred ( S1B and S1C Fig ) . Also , no measurable increase in intestinal permeability , clinical signs of disease , such as diarrhea , weight loss or mortality were observed in wild type mice ( S1D , S1E and S1F Fig ) . These results demonstrate that EPEC efficiently colonizes the intestinal tract of mice during the first two weeks of life but fails to persist in the intestine of adult animals . Next , we visualized the interaction of EPEC with the small intestinal epithelium . Immunostaining and scanning electron microscopy ( SEM ) revealed the formation of typical bacterial microcolonies composed of densely packed bacteria attached to the epithelial surface ( Fig 2Ai–2Aiii ) . Transmission electron microscopy ( TEM ) illustrated the tight contact between the bacterial surface and the apical plasma membrane of the epithelium ( Fig 2Aiv and 2Av ) . The tight attachment of EPEC to the epithelial plasma membrane was further illustrated by the crescent-shaped staining pattern observed on attached bacteria by fluorescence immunostaining for bacterial lipopolysaccharide ( Fig 2Avii ) . Of note , microcolonies were preferentially localized at uncharacterized epithelial asperities or folds of the epithelial surface ( Fig 2Avi and 2Avii ) . This might explain the difficulties to visualize microcolonies by electron microscopy . Microcolonies were observed as early as 4 days p . i . , and their number peaked at 8 days p . i . Fewer microcolonies were observed at later time points and were absent from tissue sections obtained at 20 days p . i . ( Fig 2B ) . No EPEC microcolonies could be observed in colonic tissue ( S2A Fig ) . To investigate the involvement of the two major EPEC pathogenicity factors BFP and T3SS during intestinal colonization and microcolony formation , mutants deficient in the biogenesis of BFP ( ΔbfpA ) , T3SS ( ΔescV ) or both ( Δbfp , A ΔescV ) and two commensal E . coli strains were analyzed ( S2B Fig ) . EPEC bfpA and escV single mutants as well as the bfpA/escV double mutant and the two commensal E . coli strains readily colonized the neonate small intestine and colon ( Fig 2C and S2C Fig ) . Some variation in the colonization efficacy was noted . In particular , the EPEC escV/bfpA double mutant and both commensal E . coli strains exhibited a somewhat reduced colonization efficacy . In contrast , EPEC bfpA and escV single mutants displayed a colonization efficacy similar to WT EPEC . In a direct comparative analysis , however , WT EPEC bacteria outcompeted both , bfpA or escV , mutants when administered simultaneously at a 1:1 ratio to neonates ( Fig 2D and S2D Fig ) . Importantly , bfpA and escV EPEC mutants completely failed to generate microcolonies at the small intestinal epithelial surface at 8 days p . i . ( Fig 2E and 2F ) . Thus , BFP and the T3SS are critical determinants of microcolony formation but only marginally contribute to colonization of the neonate intestine . Subsequently , we analyzed the epithelial transcriptional response to EPEC exposure . Global transcriptome analysis using total mRNA prepared from isolated primary intestinal epithelial cells ( IEC ) revealed a defined set of genes that were upregulated after infection ( Fig 3A ) . Remarkably , transcriptional stimulation of these genes was completely absent following infection with the T3SS-deficient ( ΔescV ) mutant indicating the requirement of an intact T3SS for epithelial stimulation . Genes induced after EPEC infection were mostly involved in metabolic , cellular and regulatory processes as well as responses to exogenous stimulation ( Fig 3B ) . Quantitative RT-PCR for the acute phase reactant serum amyloid A3 ( saa3 ) and the carboxypeptidase N ( cpn2 ) confirmed the EPEC-induced epithelial cell response and additionally revealed the strong influence of BFP expression for epithelial stimulation ( Fig 3C and 3D ) . These results suggest a functional link between microcolony formation and the induction of epithelial gene expression . Next , we examined the expression of the antibacterial c-type lectin RegIIIγ , one of the most highly ( 65-fold ) upregulated genes 8 days p . i . , in more detail . Whereas EPEC-induced epithelial expression of RegIIIγ was completely abolished in the absence of a functional T3SS , a significant , albeit reduced , epithelial innate immune stimulation was observed following infection with BFP-deficient ( ΔbfpA ) EPEC ( Fig 4A ) . This finding was corroborated by immunostaining . RegIIIγ positive goblet cells were only observed in the distal part of the small intestine of animals infected with WT EPEC and EPEC bfpA mutants ( Fig 4B ) . Of note , basal epithelial RegIIIγ expression increases with age after birth [12] . Direct comparison with age-matched control animals , however , demonstrated a significantly enhanced epithelial RegIIIγ expression following challenge ( Fig 4C ) . To characterize the upstream signaling events of the EPEC-mediated innate immune stimulation , WT mice were compared with animals deficient in innate immune receptors known to mediate epithelial innate immune recognition . A significant reduction in epithelial RegIIIγ expression was noted in the absence of MyD88 , Tlr5 or Tlr9 ( Fig 4D ) . In addition , impaired innate host responses enhanced the susceptibility to EPEC infection , demonstrating the critical role played by these molecules in the mucosal response to EPEC infection ( S1F Fig ) . In addition , we carefully analyzed the composition of the enteric microbiota in the small intestine and colon of infected and non-infected animals 8 and 20 days p . i . by 16S rDNA sequencing . At 8 days p . i . , at the height of the infection , EPEC ( OTU 4425571 ) accounted for 2 . 6% or 7 . 1% of the total bacteria present in the small intestine or colon , respectively . At 20 days p . i . , when the infection was being resolved , EPEC could not be detected anymore in either the small intestine or the colon ( Figs 1C , 1D and 5A ) . PCA plot analysis revealed a significant influence of EPEC on the intestinal bacterial community at 8 days p . i . In fact , the microbiota in the small intestine and colon differed between infected ( red squares ) and non-infected control ( blue squares ) animals at day 8 p . i . ( p-value = 0 . 0181 for small intestine and 0 . 0147 for colon [PC2] ) but was superimposable in both organs after recovery at day 20 p . i . ( green triangles versus orange triangles ) ( p-value = 0 . 2006 for small intestine and 0 . 3341 for colon [PC2] ) ( Fig 5B ) . A highly significant age-dependent difference in the microbiota composition was noted , with a decrease in Proteobacteria and a concomitant rise of bacteria of the obligate anaerobic phylum Bacteroidetes between 9- and 21-day-old animals . This is consistent with major changes in nutrients ( breast milk vs solid food ) , the local luminal milieu and the establishment of an increasingly diverse enteric microbiota . The overall phylum composition between the infected samples and their uninfected age-matched controls was , however , remarkably similar for both the small intestine and the colon ( Fig 5C ) . Significant differences in the bacterial composition between infected and control animals were detected at 8 days p . i . and these differences were also observed after removal of the EPEC-related sequences from the analysis . Also , bacterial groups expanded concomitantly with the decrease in EPEC colonization but were absent in uninfected animals ( S1 Table ) . The critical role of the emerging enteric microbiota in restricting EPEC colonization was demonstrated using germ-free ( GF ) mice . Conventional and GF mice were infected at day 1 after birth . Whereas the colonization in conventionally housed mice dropped significantly 20 days p . i . , EPEC colonization in germ-free mice remained high with numbers similar to younger animals ( Fig 6A ) . Additionally , we investigated the role of the enteric microbiota during infection of adult animals . As previously shown , conventionally raised adult mice were resistant to EPEC colonization/infection ( Figs 1A , 1B and 6B ) . However , EPEC was recovered from the feces of infected gnotobiotic mice or of conventionally raised mice treated with antibiotics ( Fig 6B ) . EPEC colonization required continuous antibiotic administration since a sharp decrease in EPEC colonization was observed upon termination of antibiotic treatment ( Fig 6B ) . Importantly , despite shedding high numbers of EPEC at 8 days p . i . , no microcolony formation was observed in the gut of adult GF and streptomycin-treated mice ( Fig 6C and 6D ) . In contrast , neonate GF mice similar to conventionally raised newborn animals , displayed EPEC microcolonies 8 days p . i . ( S3 Fig ) . Also , epithelial RegIIIγ expression remained low upon infection of GF and streptomycin-treated adult animals ( Fig 6E ) . Together , these results illustrate the role of the enteric microbiota in colonization resistance . The generation of microcolonies , hallmark of the EPEC-host cell interaction , was however age-dependent but microbiota-independent .
Here we present the first small animal model amenable to genetic modifications to investigate EPEC infection in vivo . EPEC was originally described to cause outbreaks of diarrhea in pediatric wards in industrialized countries but nowadays remains a major health concern in developing countries , particularly for small children and HIV-infected infants [1 , 13–15] . Although general principles of EPEC pathogenesis have been unraveled using in vitro and large animal models , the emerging picture is still incomplete . Oral infection of neonate mice led to the histological hallmark of EPEC infection in humans: an A/E lesion-like localized adherence pattern on the surface of the small intestinal epithelium . Two previously defined important virulence factors , BFP and T3SS , were critical for microcolony formation as well as mucosal innate immune stimulation . Infection of neonate mice , however , was not accompanied by clinical symptoms such as watery diarrhea observed in humans . The reason is currently not clear but might result from differences in the physiology and ability to develop diarrhea between mice and men or from the degree of the infection and the infection-induced inflammatory tissue response . Nevertheless , we believe that the T3SS and BFP-dependent generation of A/E lesion-like microcolonies and the epithelial stimulation as characteristic features of EPEC infection make this model a useful tool to investigate the underlying mechanisms of EPEC infection and for testing novel therapeutic strategies in the future . Intestinal colonization also occurred in the absence of BFP or of a functional T3SS . Similarly , orally administered commensal E . coli strains were able to colonize the neonate gastrointestinal tract . Thus , the ability to colonize the neonatal intestine did not represent a virulence trait of EPEC but rather reflected the low colonization resistance in the murine neonate intestine also observed in other species [16 , 17] . Most likely , epithelium-attached bacteria accounted for only a minor fraction of the total number of intestinal EPEC bacteria . The presence of a subpopulation of virulent , disease-promoting bacteria has been confirmed during experimental Citrobacter rodentium and Salmonella infection [18 , 19] . Also in our model , competition experiments identified the enhanced colonization capacity of fully virulent T3SS and BFP-positive EPEC bacteria . The ability to firmly adhere to the intestinal epithelial surface might help to avoid shedding by continuous mucus secretion and intestinal peristalsis and allow enhanced proliferation [20] . Future studies need to further address bacterial virulence gene expression in respect to microcolony formation and enhanced colonization . Similar to the situation in the neonate mouse intestine , EPEC was able to colonize germ-free adult mice or adult animals pretreated with antibiotics [21] . Colonization resistance thus protects adult animals from EPEC colonization . Differences in the enteric microbiota might thus explain reports describing EPEC colonization of the adult intestine [22] . Importantly , consistent with previous reports , infection of germ-free or pretreated adult mice failed to induce epithelial stimulation , intimate epithelial attachment and the generation of A/E lesion-like microcolonies [21 , 23 , 24] . Thus , the ability to colonize the intestine represents a prerequisite for infection but is not sufficient to generate the characteristic structural and functional features , microcolony formation and innate immune stimulation . These features seem to be restricted to the neonatal small intestine . Many aspects of the innate and adaptive immune system such as the antimicrobial peptide repertoire , mucus secretion , epithelial-turn over , immune cell maturation as well as metabolic and anatomical features differ between the neonate and adult intestine and may account for the differential susceptibility to infection . Age-dependent susceptibility was also observed for other age-related enteric pathogens such as E . coli , Salmonella or rotavirus but the underlying mechanisms for this observation have not been fully elucidated [25–28] . Currently , the factor responsible for the age-dependent ability of EPEC to form microcolonies and stimulate the epithelium is unknown . A better understanding might , however , improve the clinical management of children suffering from infection with enteropathogenic microorganisms . EPEC infection resulted in a significant upregulation of a distinct set of genes in the intestinal epithelium . Antimicrobial effectors , apolipoproteins and metabolic molecules accounted for the majority of these genes , thereby illustrating the antimicrobial and adaptive changes of the epithelium to the infectious challenge . The induced antimicrobial host response might contribute to the somewhat altered morphology of adhering EPEC bacteria illustrated by TEM . Upregulation of genes occurred in a T3SS-dependent manner , since the response was abolished in the absence of the essential inner membrane protein of the T3SS apparatus EscV . BFP-deficient EPEC mutants exhibited a strongly reduced stimulatory potential but still allowed for RegIIIγ upregulation . BFP-negative bacteria have a functional T3SS and are able to intimately attach to the intestinal epithelium but exhibit a lower binding efficacy and fail to form microcolonies [29 , 30] . The reduced host response induced by BFP-negative EPEC may explain the frequent isolation of so-called atypical EPEC strains that lack expression of BFP and constitute an emerging form of EPEC in humans [31–33] . Notably , our results confirm goblet cells as a major source of RegIIIγ in neonate mice , whereas enterocytes and Paneth cells were shown to be the main producers of RegIIIγ in adult animals [12 , 34] . Interestingly , effector gene expression was found distant to the site of microcolony formation , suggesting that horizontal epithelial communication via direct epithelial-epithelial cell interaction or via the secretion of soluble mediators occurred [35] . Consistent with previous reports , enhanced RegIIIγ mRNA expression required the TLR signaling adapter molecule MyD88 [36] . RegIIIγ expression was induced during the postnatal period , in parallel to the establishment of the enteric microbiota [12] . Our data additionally identify the upstream TLRs involved in EPEC-induced RegIIIγ expression in vivo , namely TLR5 and TLR9 . However , since the comparative analyses were not performed using littermates , an influence of differences in the microbiota composition between the mouse strains used cannot be excluded . Intestinal epithelial cells express TLR5 and TLR9 and recognition of the flagellated EPEC by Tlr5 in vitro has already been described in the literature [37 , 38] . Also TLR9 was shown to protect from mucosal damage but its role during intestinal bacterial challenge has not been investigated [39] . Interestingly , both TLR5 and TLR9 have been proposed to exhibit a polarized expression and EPEC was shown to manipulate the localization of TLR5 in epithelial cells suggesting that it might manipulate the pro-inflammatory epithelial response [37–39] . Further in vivo analyses will be required to identify host and microbial factors contributing to EPEC pathogenesis . Our results identified major changes in the microbiota composition of uninfected mice between 9 and 21 days after birth that could be responsible for the rise in colonization resistance observed during this period . For example , the small intestinal pathobionts , segmented filamentous bacteria [SFB ( Candidatus arthromitus ) ] were only observed in samples from older mice , in accordance with previous reports [40] . Consistent with the ability of EPEC to colonize streptomycin-pretreated mice , SFB have been shown to be highly susceptible to streptomycin treatment [41] . Moreover SFB were less abundant in infected mice and colonization with SFB was shown to inhibit intestinal colonization by rabbit-specific EPEC in the rabbit model [42] . Other significant alterations of the microbiota composition were observed after EPEC infection . Infections associated with tissue destruction , immune stimulation or metabolic changes significantly influence nutritional and antimicrobial aspects of the enteric milieu and microbiota alterations might contribute to the pathogenesis of diarrheal disease [43 , 44] . Finally , infection-induced microbiota alterations might ultimately restrict pathogen colonization [45] . Together our findings highlight the potential influence of the enteric microbiota in EPEC colonization resistance , disease progression and pathogen elimination . Further investigations are , however , required to demonstrate causal relationships and elucidate the potential therapeutic value . In conclusion , we present a new small animal model amenable to genetic modifications to investigate EPEC infection . We confirm the critical role of bacterial virulence factors , characterize the antimicrobial host response and identify the innate immune receptors stimulated by EPEC in vivo . Finally , we demonstrate the infection-induced transient alteration of the neonate enteric microbiota and reveal a critical role for age-dependent , but not microbiota-dependent factors during EPEC infection . Therefore , this model might help to unravel the mechanisms involved in the EPEC-host cell interaction and facilitate a much-needed improvement in the clinical management of infected children worldwide [46] .
Mice were bred locally and held under specific pathogen-free or germ-free conditions at the Hannover Medical School animal facility . C57Bl/6N mice were purchased from Charles River laboratories , TLR4-/- ( B6 . B10ScN-Tlr4lps-del/JthJ ) , TLR5-/- ( B6 . 129S1-Tlr5tm1Flv/J ) , TRIF-/- ( C57BL/6J-Ticam1Lps2/J ) and MyD88-/- ( B6 . 129P2 ( SJL ) -Myd88tm1 . 1Defr/J ) mice were obtained from Jackson laboratories . TLR9-/- ( B6 . 129P2-Tlr9tm1Aki ) mice were kindly provided by M . Brinkmann ( Helmholz Center for Infection Biology , Braunschweig , Germany ) . All animal experiments were performed in compliance with the German animal protection law ( TierSchG ) and were approved by the local animal welfare committee ( approval 13/1256 of the Niedersachsische Landesamt für Verbraucherschutz und Lebensmittelsicherheit Oldenburg , Germany ) . WT EPEC E2348/69 ( StreptomycinR ) , ΔescV EPEC ( escV::miniTn10kan; StreptomycinR , KanamycinR ) , ΔbfpA EPEC ( bfpA::TnphoA; KanamycinR ) , ΔescV/bfpA EPEC ( KanamycinR , ChloramphenicolR ) , E . coli Nissle 1917 and a murine small intestinal E . coli isolate were grown in Luria Broth supplemented with the appropriate antibiotics [47–49] . Both commensal E . coli strains were transformed with a GFP expression plasmid ( pGFP , AmpicillinR ) to confer ampicillin resistance [27] . 1-day-old mice were defined as animals born maximum 24 hours before the infection and presenting a milk spot indicating previous breast-feeding by the dam . 1- , 4- , 7- , 10- , 13- or 17-day-old mice were orally infected with approximately 0 . 5–1×105 CFU bacteria in PBS ( 1μl ) . 5- to 8-week-old adult mice were gavaged with approximately 0 . 5–1×108 bacteria in PBS ( 100μl ) . Streptomycin-treated adult mice were given orally 20μg streptomycin ( SIGMA ) in PBS ( 50μl ) 24 hours before receiving EPEC and had access ad libitum to drinking water containing streptomycin ( 5g/L ) for the first 8 days of the experiment . For competition experiments , bacteria were mixed in a 1:1 ratio and 1-day-old mice were co-infected with 0 . 5–1×105 bacteria of each strain in PBS ( 2μl ) . 4 , 8 , 12 , 16 or 20 days p . i . , the small intestine , colon , spleen , liver and/or mesenteric lymph nodes were collected in PBS , homogenized , diluted and plated on LB agar plates containing the appropriate antibiotic . Small intestines of WT EPEC infected neonates were collected 8 days p . i . , longitudinally opened , flushed and fixed for 1 hour at room temperature in 200mM HEPES , pH 7 . 35 containing 4% formaldehyde and 0 . 1% glutaraldehyde . Samples were then dehydrated using increasing acetone series . Critical point drying was performed using a CPD030 critical point dryer ( Balzers , Lichtenstein ) following manufacturer instructions . SEM was performed using a QuantaSEM ( FEI ) in high vacuum mode . For the ultrastructural analysis , the distal end of small intestines of WT EPEC infected neonates were dissected 8 days p . i . and fixed with 1% glutaraldehyde in 0 . 2M HEPES buffer , pH7 . 4 overnight . For embedding , the intestine was cut into 3mm long segments and these were cut open longitudinally . Samples were post-fixed with 1% osmium tetroxide and contrasted with 2% uranyl acetate , both for 1 hour . Dehydration was performed with a graded ethanol series ( 70-80-90-95-100% ) , followed by progressive infiltration with epoxy resin and polymerization overnight at 60°C . 70nm thin transverse sections were prepared using an ultramicrotome ( Ultracut EM UCT-Leica Microsystems ) and a diamond knife ( Diatome ) , and contrasted with 0 . 2% lead citrate for 15s . Samples were analyzed with JEM1400 transmission electron microscope ( JEOL ) . Images were recorded with TemCam-F216 ( Tvips ) . Paraffin-embedded small intestinal and colonic tissue sections were stained using a rabbit anti-OK127 anti-serum ( Statens Serum Institute ) to visualize EPEC or a rabbit anti-RegIIIγ anti-serum ( gift from Lora Hooper , Southwestern Medical Center , Texas , USA ) and a mouse anti-E cadherin ( BD Biosciences ) antibody and fluorescein-labelled wheat germ agglutinin ( Vector labs ) . Secondary antibodies used in this study were purchased from Dianova . Slides were then mounted in DAPI-mounting medium ( Vectashield ) and pictures were taken with an ApoTome microscope connected to a digital camera ( Zeiss ) . Microcolony quantification was done by assessing the number of “clusters of at least 5 EPEC bacteria attached to the small intestinal epithelium” ( microcolonies ) per tissue section of the full organ . 3 non-consecutive sections from 3 animals were analyzed by experimental condition . IEC were isolated from the small intestine as previously described [28] . Total RNA was isolated using TRIZOL ( Invitrogen ) and cDNA synthesized using RevertAid ( Fermentas ) . Quantitative real-time PCR were performed using Taqman gene expression assays ( Life Technologies ) : hprt ( Mm00446968_m1 ) , regIIIγ ( Mm01181783_g1 ) , saa3 ( Mm00441203_m1 ) and cpn2 ( Mm01169716_m1 ) . Microarray analysis was performed using Whole Mouse Genome Oligo Microarray v2 ( 4x44k ) ( Agilent Technologies ) following the SC_AgilentV5 . 7 protocol provided by the manufacturer . The heat map was generated using Qlucore Omics explorer ( p-value = 0 . 02; q-value = 0 . 21 ) . Cluster of orthologous groups’ analysis was performed using PANTHER ( http://www . pantherdb . org/ ) . Expression array data are available through GEO Series accession number GSE71685 . Litters of 1-day-old mice were orally infected with WT EPEC or left untreated . 8 or 20 days p . i . , full small intestine and full colon were collected , pooled ( 6 animals per sample for 9-day-old; 3 animals for 21-day-old ) and snap frozen in liquid nitrogen . Bacterial DNA was extracted and analyzed by 16S rDNA sequencing . Mice left untreated or orally infected at birth with EPEC were orally fed 1μg of FITC dextran ( MW 3000–5000—SIGMA ) 8 days p . i . 4 hours later , their blood was withdrawn and the fluorescence intensity in their serum measured using a fluorometer ( Victor ) . Small intestinal epithelial m-ICcl2 cells were cultured as previously described [50] . WT EPEC and ΔescV mutant carrying a green fluorescent protein ( GFP ) expression plasmid were used for in vitro infection experiments . Cells were grown on 8 chamber slides ( Lab-Tek ) for 6 days and were then left untreated or infected with WT EPEC or escV mutant at a MOI of 1 for 3 hours . Slides were then fixed with 4% PFA and stained using the CytoPainter F-actin staining kit ( Abcam ) . Slides were then mounted in DAPI-mounting medium ( Vectashield ) and pictures were taken with an ApoTome microscope connected to a digital camera ( Zeiss ) . Escherichia coli 0127:H6 E2348/69 complete genome , Accession: FM180568 . 1 Escherichia coli O127a:H6 bfpA gene for bundlin-2a , complete cds , strain: E2348/69 , Accession: AB247923 . 1 Mus musculus toll-like receptor 4 ( Tlr4 ) , mRNA , Accession: BC029856 . 1 Mus musculus toll-like receptor 5 ( Tlr5 ) , mRNA , Accession: NM_016928 . 3 Mus musculus toll-like receptor 9 ( Tlr9 ) , mRNA , Accession: NM_031178 . 2 Mus musculus IL-1 receptor related protein MyD88 mRNA , Accession: U84409 . 1 Mus musculus mRNA for Trif , mRNA , Accession: AB025010 . 1 Mus musculus regenerating islet-derived 3 gamma ( Reg3g ) , mRNA , Accession: NM_011260 . 1
|
Enteropathogenic Escherichia coli ( EPEC ) is an important causative agent of infant diarrhea associated with significant morbidity and mortality particularly in the developing world . Current knowledge on EPEC pathogenesis has mainly emanated from in vitro studies as research is limited by the absence of a suitable small animal infection model . Here , we use neonate mice and present a new infection model that mimics the hallmark of the EPEC–host cell interaction in humans . We observe microcolonies of EPEC closely attached to the epithelial surface in the infected small intestine dependent on the presence of two well-established bacterial virulence factors , namely the type III secretion system and bundle forming pili . Studying the mucosal host response , we demonstrate enhanced epithelial expression of a distinct set of genes as well as an alteration of the intestinal microbiota composition . In contrast , EPEC fails to induce similar changes in adult animals illustrating the age-dependent susceptibility to EPEC infection . In the future , the new model could help to better understand the underlying mechanisms of EPEC infection and lead to the development of new therapeutic strategies to improve the outcome of infection in children .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"antimicrobials",
"small",
"intestine",
"medicine",
"and",
"health",
"sciences",
"microbiome",
"drugs",
"microbiology",
"animal",
"models",
"bacterial",
"diseases",
"model",
"organisms",
"antibiotics",
"gastroenterology",
"and",
"hepatology",
"pharmacology",
"microbial",
"genomics",
"digestive",
"system",
"research",
"and",
"analysis",
"methods",
"infectious",
"diseases",
"escherichia",
"coli",
"infections",
"medical",
"microbiology",
"mouse",
"models",
"gastrointestinal",
"infections",
"gastrointestinal",
"tract",
"anatomy",
"genetics",
"microbial",
"control",
"biology",
"and",
"life",
"sciences",
"genomics",
"colon"
] |
2016
|
Age-Dependent Susceptibility to Enteropathogenic Escherichia coli (EPEC) Infection in Mice
|
How cells control the overall size and growth of membrane-bound organelles is an important unanswered question of cell biology . Fission yeast cells maintain a nuclear size proportional to cellular size , resulting in a constant ratio between nuclear and cellular volumes ( N/C ratio ) . We have conducted a genome-wide visual screen of a fission yeast gene deletion collection for viable mutants altered in their N/C ratio , and have found that defects in both nucleocytoplasmic mRNA transport and lipid synthesis alter the N/C ratio . Perturbing nuclear mRNA export results in accumulation of both mRNA and protein within the nucleus , and leads to an increase in the N/C ratio which is dependent on new membrane synthesis . Disruption of lipid synthesis dysregulates nuclear membrane growth and results in an enlarged N/C ratio . We propose that both properly regulated nucleocytoplasmic transport and nuclear membrane growth are central to the control of nuclear growth and size .
Much is known about the molecular mechanisms that underpin membrane trafficking and local membrane growth in eukaryotic cells [1] , but how membrane-bound organelles determine their overall growth rate and maintain an appropriate size is not well understood . The simple shape of the nucleus , and the fact that it is generally present in single copy within a cell , makes it a useful model to study overall membrane-bounded organelle growth and organelle size homeostasis . Work in algae and sea urchin embryos led Hertwig in 1903 to propose that there is a constant karyoplasmic ratio characteristic of cells [2]; since then nuclear size has been reported to correlate with cell size across a range of cell types and species [2 , 3] . Budding and fission yeasts exhibit a nuclear size proportional to cell size , resulting in a constant ratio of nuclear and cellular volumes ( N/C ratio ) [4 , 5] . In fission yeast the N/C ratio remains constant throughout the cell cycle , and no increase in the ratio is observed during or after S phase; even a 16-fold increase in nuclear DNA content does not affect N/C ratio [5] . These results indicate that , contrary to the generally accepted view , nuclear size is not directly determined by nuclear DNA content . Increases in ploidy do result in enlarged nuclei but this occurs indirectly , via an increase in cell volume which results in an increase in nuclear size [5] . Study of multi-nucleated cells with nuclei that are unevenly distributed throughout the cell revealed that the volume of each nucleus is proportional to that of its surrounding cytoplasm [5] . Results of an in vitro study of Xenopus egg extracts demonstrated that the available space surrounding a nucleus determines nuclear expansion rate [6] , consistent with the fission yeast results . Cytoplasmic effects on nuclear size were also observed when erythrocyte nuclei injected into the cytoplasm of larger HeLa cells were found to grow in size [7] . Similarly , HeLa nuclei increased in volume when injected into the cytoplasm of X . laevis oocytes [8] . These experiments indicate that nuclear size is determined by the overall size of the cell , and that the cytoplasmic content immediately surrounding a particular nucleus is important for determining its size . However , these studies have given no insight into the molecular mechanisms that control nuclear growth and nuclear size homeostasis . An important contribution to molecular mechanism was provided by Levy and Heald [9] . These authors studied nuclear assembly around exogenous DNA added to egg extracts from two species of Xenopus: X . laevis , with large nuclei , and X . tropicalis , with small nuclei . A GFP-NLS ( nuclear localisation signal ) fusion protein was found to be accumulated at a faster rate into nuclei assembled in extracts from X . laevis . The authors concluded that nuclear transport was key to establishing the differing nuclear sizes assembled in vitro in the egg extracts . The transport factor Impα2 ( an importin ) was found to be at a higher level in X . laevis extracts than in X . tropicalis extracts whereas the transport factor Ntf2 was found to have an inverse relationship . Increasing the level of Impα2 increased the size of the in vitro assembled nuclei and overexpression of Impα2 increased nuclear size in embryonic cells . Addition of Lamin B3 , a cargo of Impα2 , to extracts also increased the size of the nuclei assembled in vitro . This study led the authors to propose that Lamin B3 , transported by Impα2 into the nucleus , plays a key role in determining nuclear size . A further study reported that total lamin concentration , rather than the level of a specific lamin , affects the size of nuclei assembled in X . laevis egg extracts as well as in Xenopus embryos and mammalian tissue culture cells , though different lamin concentrations have different effects in different cell types and developmental stages , sometimes increasing the lamin level increased nuclear size and sometimes decreased it [10] . Fission yeast cells lack lamins yet display nuclear size control [5] , suggesting that there are other key players in nuclear size control that have not , as yet , been identified . To identify , more systematically , other proteins involved in controlling nuclear growth and nuclear size homeostasis , we have carried out a genetic screen to identify genes that when deleted alter the N/C ratio of growing fission yeast cells , and so are candidates for playing roles in nuclear size control . Our screen has identified new factors as having roles in this control , and our studies have led us to conclude that nucleocytoplasmic transport of both RNA and protein and nuclear membrane growth contribute to the overall control of nuclear growth and nuclear size homeostasis in growing cells .
To identify genes involved in nuclear size control , we carried out a screen for fission yeast mutants exhibiting an abnormal N/C ratio . We screened a gene deletion collection consisting of 2 , 969 viable deletion mutants representing approximately 80% of S . pombe viable haploid gene deletion strains [11] . The screen was carried out in three stages ( Fig 1A ) . A visual screen was carried out on solid agar , examining growing cells on the edges of colonies , using the lipophilic fluorescent dye DiOC6 [12] to visualise nuclear and cellular membranes; 366 strains were identified as potentially having N/C ratios greater or smaller than wild type . In a second stage , these potential N/C ratio mutants were visually screened with DiOC6 during steady state growth in liquid media; 97 potential N/C ratio mutants were retained . In a third stage a nuclear membrane marker Cut11-GFP was introduced into cells , and imaging was carried out during steady state growth in liquid media . Cut11 is a transmembrane nuclear pore complex protein orthologous to H . sapiens and S . cerevisiae NDC1 . N/C ratios were determined as described previously [5] . We identified 14 gene deletion strains that displayed a N/C ratio at least 15% higher than that of wild type cells ( Fig 1B , S1 Table ) ; four of these , mlo3 , caf1 , dss1 and trm112 ( Fig 1C ) , exhibited a N/C ratio greater than 0 . 100 , which was more than 25% higher than the wild type value of 0 . 081 . No strains with a N/C ratio significantly smaller than wild type were identified . Mutants could generate aberrant N/C ratios as a consequence of asymmetric nuclear division instead of an interphase nuclear size control defect . If this were the case the N/C ratio would be expected to correct if cells grew for an extended time in interphase . We used the cell cycle mutant cdc25-22 to arrest cells in interphase , and then measured the N/C ratio ( Fig 1D , S1 Table ) . Eight mutants still exhibited significantly aberrant N/C ratios during interphase arrest; these eight mutants carried deletions of the caf1 , crf1 , cut8 , dss1 , nem1 , mlo3 , spo7 and trm112 genes , and included the four gene deletions noted above with the strongest phenotypes in normal exponentially growing cells ( Fig 1B ) . In addition to an increased N/C ratio , two of the candidate mutants , nem1Δ and spo7Δ , exhibited nuclear shape defects ( Fig 1C ) . Strikingly , this unbiased screen independently identified two components of two complexes , Dss1-Mlo3 and Nem1-Spo7 , which are involved in mRNA export from the nucleus and membrane synthesis respectively [13 , 14] . The mRNA export factor Dss1 and the RNA binding protein Mlo3 ( orthologous to S . cerevisiae YRA1 and H . sapiens ALYREF ) are components of a complex implicated in nuclear mRNA export [13] . To investigate this further we looked at other components of the complex . A third component of the complex , Rae1 , targets the Dss1-Mlo3 mRNP to the nuclear pore [13] . The rae1 gene is essential so was not included in our screen of viable strains . To assess the N/C ratio phenotype of cells lacking Rae1 function we used the temperature-sensitive mutant rae1-167 . Rae1-167 cells have been reported to accumulate poly ( A ) +RNA in the nucleus when shifted to the restrictive temperature [15] , and we confirmed that poly ( A ) +RNA accumulates in the nucleus after shift from 25°C to 36°C; this accumulation occurs rapidly , beginning within 15 minutes of temperature shift ( S1A Fig ) . At 25°C , the N/C ratio of this strain was similar to that of wild type ( 0 . 080 ) , but after incubation at 36°C for 4 hours the N/C ratio increased by more than 50% to 0 . 125–0 . 135 , values greater than those of any of the viable gene deletion strains identified in our screen ( Fig 2A and 2B , S2 Table ) . Given this more extreme effect we focused our studies on rae1-167 . A N/C ratio increase was detectable within 1 hour of the temperature shift , and became maximal an hour later ( Fig 2C ) . This N/C ratio increase was caused by an increased nuclear growth rate , which was approximately twice that required to maintain a constant N/C ratio ( Fig 2D ) . To determine whether this increase in the N/C ratio caused by inhibiting mRNA export required continued RNA synthesis , we treated cells with 15 μg/ml thiolutin which completely blocks RNA synthesis in wild type cells [16 , 17] . Although thiolutin treatment did not affect the N/C ratio of wild type cells , the treatment suppressed the increase of N/C ratio observed in rae1-167 cells following temperature shift ( Fig 2A and 2B , S2 Table ) . These results indicate that defects in mRNA export can change the N/C ratio , and that this change requires continued RNA synthesis . In addition to nuclear poly ( A ) +RNA accumulation , at 36°C rae1-167 cells exhibit nuclear accumulation of poly ( A ) -binding protein ( Pabp ) which shuttles between the nucleus and the cytoplasm in wild type cells ( S1B Fig ) [18] . Pabp-GFP accumulated in the nucleus within 1 hour of temperature shift in 100% of rae1-167 cells ( S1C and S1D Fig ) . However , deletion of the pabp gene did not significantly suppress the high N/C ratio of rae1-167 cells , so accumulation of Pabp is not sufficient to cause the N/C ratio increase ( Fig 2A , S2 Table ) . To assess whether proteins more generally accumulated in the nuclei of rae1-167 cells , we used the fluorescent protein-staining dye fluorescein isothiocyanate ( FITC ) . DAPI staining was used to identify nuclei , and nuclei of rae1-167 cells appear to show reduced compaction of chromatin following shift to the restrictive temperature ( Fig 3A ) . Proteins were uniformly distributed throughout nuclei and cytoplasms of both wild type and rae1-167 cells at 25°C . Within 30 minutes of shift to 36°C , rae1-167 cells exhibited significant nuclear protein accumulation , in contrast to temperature shifted wild type cells which still had a uniform protein distribution between the nucleus and the cytoplasm ( Fig 3A and 3B ) . This result suggested protein accumulation following mRNA export inhibition also contributes to the N/C ratio increase observed , consistent with previous observations in which prolonged inhibition of nuclear export of proteins by Leptomycin B ( LMB ) , an inhibitor of exportin Crm1 [19] , increased the N/C ratio [5] . To investigate this possibility further we tested whether the N/C ratio increase following mRNA export inhibition requires continued protein synthesis by treating cells with 100 μg/ml cycloheximide ( CYH ) which inhibits protein synthesis [20] . Although CYH treatment did not affect the N/C ratio of wild type cells , the treatment largely suppressed the high N/C ratio observed in rae1-167 cells following shift to 36°C ( Fig 2A and 2B , S2 Table ) . This result indicates that continued protein synthesis is required for the N/C ratio increase in these cells . We next tested whether it was general bulk protein and mRNA accumulation or accumulation of a smaller number of specific proteins and mRNAs which was taking place during nuclear enlargement . We characterised the protein and mRNA content of rae1-167 nuclei at the level of individual proteins and mRNAs using SILAC mass spectrometry and microarray analysis respectively . To identify proteins accumulated in rae1-167 nuclei with an enlarged N/C ratio , we compared the protein content of wild type and rae1-167 nuclei . We enriched for nuclei to allow detection of low abundance nuclear proteins that may not be detected in whole cell samples . We confirmed that nuclear enriched samples were enriched for nuclear-localised and depleted for cytosol-localised proteins ( P<0 . 02 ) ( S3 Table ) . When we compared the protein content of nuclear enriched samples , there was an increase in a subset of proteins in rae1-167 relative to wild type at the restrictive temperature , that was not observed at the permissive temperature . 22 . 4% of proteins detected at 36°C ( 509/2270 proteins detected ) , in contrast to only 1 . 3% at 25°C ( 30/2270 proteins detected ) , were present in the rae1-167 nucleus enriched sample at a level at least 25% higher than in wild type ( Fig 3C ) . Proteins found at this level at 36°C and not at 25°C are listed in S4 Table . To examine what classes of proteins were enriched in rae1-167 nuclei we carried out 2D enrichment analysis [21] . The proteins enriched in rae1-167 nuclei relative to wild type nuclei at 36°C and not at 25°C showed significant enrichment of proteins reported to be localised to the nucleus and subnuclear structures in S . pombe [19] ( Fig 3D ) . Enrichment of specific gene ontology ( GO ) categories is shown in S2 Fig; nucleic acid binding proteins were enriched in proteins increased in rae1-167 nuclei relative to wild type nuclei at 36°C and not at 25°C . Therefore , our analysis suggests that bulk accumulation of many different proteins localised to the nucleus is taking place in rae1-167 nuclei when the N/C ratio is enlarged , rather than accumulation of a few specific proteins . We next investigated the steady-state levels of individual mRNA transcripts by microarray analysis to assess whether the accumulation observed is of a few specific mRNAs or bulk accumulation of many mRNAs . Levels of 888 mRNAs were increased at least 2-fold in rae1-167 cells at 36°C relative to 25°C ( S5 Table ) . These included mRNA transcripts of 106 common environmental stress response ( CESR ) genes [22] and 64 meiotic genes [23] . To examine whether increased expression of either of these specific groups of genes causes the N/C ratio enlargement observed in rae1-167 cells , we used the transcription factor mutants , atf1Δ and mei4Δ , which are respectively defective in induction of most CESR genes and of the middle meiotic genes [22 , 24] . Deletion of neither atf1 nor mei4 suppressed the high N/C ratio of rae1-167 cells ( Fig 2A , S2 Table ) , indicating that specific transcription-driven increases of CESR gene or middle meiotic gene mRNAs are not the cause of the increased N/C ratio of rae1-167 cells . It is possible that retaining mRNAs in the nucleus could affect their stability; in this situation increased levels of these specific groups of mRNAs could be causative of the nuclear size increase because atf1Δ and mei4Δ mutants would be unlikely to affect their steady state level . Taken together , our mass spectrometry and microarray analyses indicate that general bulk accumulation of a large number of proteins and mRNAs occurs in rae1-167 cells when an N/C ratio increase is observed , and that many of these proteins are normally localised in the nucleus . Two more enlarged N/C ratio candidates identified in our screen were nem1 and spo7 . These encode the catalytic and regulatory subunits of the Nem1-Spo7 phosphatase complex , responsible for dephosphorylation and activation of Ned1 , a lipin family phosphatidic acid phosphatase [14 , 25] . In addition to the nuclear size phenotype in these mutants , nuclear shape deformation suggestive of nuclear envelope overproliferation was also observed [26] . These observations suggest that correct regulation of lipid metabolism leading to changes in membrane synthesis plays a role in nuclear size control , and we hypothesised that the increased N/C ratio of nem1 and spo7 deletion mutant cells may be caused by inappropriate nuclear envelope expansion . If this is the case , then inhibition of fatty acid synthesis , which is required for membrane growth [25 , 26 , 27] , should suppress the increased N/C ratio observed in these mutant cells . To test this , we used a temperature-sensitive mutant cut6-621 , which is defective in acetyl-CoA carboxylase and impaired in fatty acid metabolism [28] . The cut6-621 mutant blocked the increase in nuclear size of nem1Δ cells and suppressed the nuclear shape change ( Fig 4A and 4B ) . There was no effect of cut6-621 on the N/C ratio of nem1+ cells . These results indicate that nem1 deletion leads to overproduction of phospholipid by inactivation of Ned1 , causing inappropriate expansion of the nuclear envelope resulting in a N/C ratio increase . We next investigated whether new membrane synthesis is required for nuclear enlargement in the rae1-167 mutant , by combining the rae1-167 and cut6-621 mutations and measuring the N/C ratio . The N/C ratio of the double mutant is significantly lower than that of the rae1-167 mutant following shift to 36°C ( Fig 4C and 4D , S2 Table ) , indicating that new membrane synthesis is required for the nuclear size increase observed in rae1-167 cells . Combining the rae1-167 mRNA export mutant with the nem1Δ membrane synthesis regulation mutant led to a further N/C ratio increase , greater than that observed in either single mutant ( Fig 4C and 4D ) . This suggests that two distinct processes are implicated in nuclear size control , membrane synthesis and nucleocytoplasmic transport . The shape phenotype of the nem1Δ single mutant was also suppressed by the rae1-167 mutation ( Fig 4D ) .
Our screen of viable fission yeast gene deletion strains identified 8 genes that when deleted lead to an increased N/C ratio in both exponentially growing and interphase arrested cells . Significantly , among these 8 were 4 genes encoding two components of a complex involved in nuclear mRNA export ( dss1 and mlo3 ) and two components of a complex involved in lipid metabolism ( nem1 and spo7 ) . We did not find any gene deletions with a N/C ratio smaller than wild type , suggesting that reduced nuclear size might be deleterious , resulting in lethality . A screen of the diploid heterozygous gene deletion collection [11] could be expected to have less severe effects and may reveal genes that have decreased N/C ratios . It is possible that the rapid growth of unicellular fission yeast cells may require rapid ribosome biogenesis , which results in a limited range of N/C ratio perturbations being viable . A screen of essential genes may identify more severe phenotypes . The four genes , mlo3 , caf1 , dss1 and trm112 , with the strongest deletion phenotypes increased the N/C ratio by 25% , an increase that was also observed in enlarged mutant cells blocked in interphase . All four genes are thought to be involved in RNA metabolism , and two of them , dss1 and mlo3 , encode protein components of a complex required for nuclear mRNA export . We examined a third component of this complex , Rae1 , which associates with the nuclear pore . A temperature-sensitive mutant of this essential gene , rae1-167 , showed a 50% N/C ratio increase at the restrictive temperature , an increase greater than that seen in dss1Δ or mlo3Δ cells . The 50% N/C ratio increase was the result of a doubling of nuclear growth rate , and required continued RNA and protein synthesis . Both the mRNA and protein content of the enlarged nucleus increased . Mass spectrometry and microarray analyses showed that bulk nuclear accumulation of a large number of different proteins and mRNAs , rather than accumulation of a few specific proteins and mRNAs was taking place . This suggests that general accumulation of nuclear content contributes to the N/C ratio increase observed , though it is also possible that one , or a subset , of the many accumulated proteins and mRNAs may more specifically effect the N/C ratio increase . Membrane growth is required for the nuclear size increase of rae1-167 cells . The cut6-621 mutant impairs nuclear membrane growth and also prevents the rae1-167 nuclear size increase . The importance of membrane growth for nuclear size was also shown by our identification of two lipid metabolic genes , nem1 and spo7 , encoding proteins that form a phosphatase complex responsible for activation of Ned1 , a lipin family phosphatidic acid phosphatase [14] . Inactivation of Ned1 by either nem1Δ or spo7Δ induced nuclear envelope expansion , resulting in aberrant nuclear shapes and an increased N/C ratio . These changes were observed in cells blocked in interphase and so are not due to aberrant mitosis . These phenotypes , like the N/C ratio increase of rae1-167 cells , were suppressed by inhibition of nuclear membrane growth by the cut6-621 mutation . These observations are consistent with observation that a phosphomimetic mutant of Ned1 exhibits a 34% increase in nuclear surface area in interphase [14] . The rae1-167nem1Δ double mutant exhibited a N/C ratio increase greater than that of either single mutant , suggesting that two distinct processes are important for nuclear size control , both bulk nucleocytoplasmic transport and nuclear membrane growth . In addition to the components of the Dss1-Mlo3 and Nem1-Spo7 complexes , our systematic screen identified four further gene deletion strains exhibiting enlarged N/C ratios . These carried deletions in caf1 , cut8 , crf1 and trm112 . Caf1 encodes a deadenylase of the CCR4-NOT complex . The multifunctional CCR4-NOT complex has been implicated in many different areas of gene expression , both nuclear and cytosolic . These include regulation of histone modification , regulation of transcription initiation and elongation , nuclear poly ( A ) -RNA degradation , mRNA export , cytosolic poly ( A ) -RNA decay and protein turnover [29 , 30] . It is possible that disruption of one or a few of these roles of the CCR4-NOT complex by caf1 deletion could lead to nuclear poly ( A ) -RNA accumulation and therefore effect nuclear size increase by the same mechanisms as those in play in rae1-167 cells . However , due to the diverse roles this complex plays , a separate mechanism is also possible . It is possible that cut8 deletion may cause nuclear size increase by similar mechanisms to those in rae1-167 cells as it encodes a nuclear proteasome tethering factor [31] so could affect nuclear protein levels . Cut8 mutant cells have also been reported to accumulate poly ( A ) -RNA in the nucleus at the restrictive temperature suggesting a possible role in mRNA export [32] . Although we have focused on the mRNA export role of Dss1 in this study because another component of the Dss1-Mlo3 complex was also identified by our screen , it is of note that Dss1 , like Cut8 , has been implicated in proteasome function [33] . Crf1 is predicted to encode a TOR-responsive transcriptional corepressor , and deletion of its S . cerevisiae orthologue perturbs repression of ribosomal protein gene transcription [34] . The S . cerevisiae orthologue of trm112 regulates methylation of tRNAs , rRNAs , and translation factors , and is required for synthesis of both 40S and 60S ribosomal subunits [35] . As ribosome biogenesis is dominant in rapidly growing yeast cells and involves both nuclear import of ribosomal proteins and export of ribosomal subunits , it is possible that perturbing the biogenesis process , for example by crf1 or trm112 deletion , could lead to the accumulation of dysfunctional ribosomes or their constituents in the nucleus thus influencing nuclear size . Therefore , the remaining candidates from our screen all encode proteins that might have roles in regulating overall RNA and protein levels , and so impact the RNA and protein content of the nucleus . Nucleocytoplasmic transport has also been implicated in nuclear size control in metazoa . The study of nuclear assembly in Xenopus egg extracts discussed in the Introduction , and subsequent studies in Xenopus egg extracts and mammalian cells have implicated the transport factors Impα2 and Ntf2 and the import of lamins in determination of nuclear size [9 , 10 , 36] . Our data indicates that there are roles for other components involved in nucleocytoplasmic transport and also for nuclear envelope growth in nuclear size homeostasis . Our in vivo study of fission yeast cells in steady state growth has revealed a role for nucleocytoplasmic transport of mRNAs and proteins in interphase nuclear size control , which is dependent on continued RNA and protein synthesis . Our work has shown that both the accumulation of nuclear content and membrane synthesis , as well as the linkage between these two processes , must be considered when proposing potential models of nuclear size control . Outlined below are two examples of the types of mechanisms for maintenance of the N/C ratio that take account of these considerations . In one , overall cytoplasmic content determines how much protein and RNA is imported into the nucleus , and as a cell grows the resulting increase in nuclear content stimulates new membrane growth , enlarging the nucleus in balance with the cytoplasm . Nuclear content could promote nuclear envelope expansion in one of two ways: increased bulk RNA and protein import could put pressure on the nuclear membrane altering its tension and inducing its expansion [37] , or the increased bulk RNA and protein import could lead to increased import of one or a group of specific RNAs and proteins that bring about nuclear envelope expansion . In a second model we suggest that overall cytoplasmic content determines the growth rate of the nuclear membrane , perhaps operating through global cellular membrane growth being controlled by cell size , with a certain proportion of total cellular membrane being delivered to the nucleus . The increased nuclear surface area would result in the incorporation of more nuclear pore complexes , resulting in increased import of protein and RNA , increasing nuclear size in balance with the cytoplasm . Both mechanisms implicate nucleocytoplasmic transport and nuclear membrane growth in interphase nuclear size control , and so are compatible with the data presented here and by others elsewhere . However , in the first model it is the accumulation of nuclear content that is the initial driver of nuclear growth , whilst in the second it is growth of the nuclear membrane . Obviously these are only examples of possible mechanisms . We conclude that appropriately regulated nucleocytoplasmic transport and nuclear membrane growth are central to nuclear size control . This may be relevant to disease states given that abnormal nuclear size and shape phenotypes are observed in many diseases [38 , 39] but the role of nuclear size in the pathology of these diseases remains unclear . The genetic identification of processes involved in nuclear size control in fission yeast provides a tractable system in which to investigate this control , and contributes to our understanding of how membrane-bound organelles regulate their overall growth and size .
Strains used are listed in S6 Table . Gene tagging was performed by PCR and homologous recombination [40] . S . pombe media and methods as described previously [41] . For mass spectrometry , cells were grown in SILAC media; heavy labelled samples were grown for >8 generations in media supplemented with heavy arginine ( L-Arginine:HCL ( U13C6 , 99% ) ) and heavy lysine ( L-Lysine:2HCl ( U13C6 , 99% ) ) ( Cambridge Isotope Laboratories Inc . ) . All other strains were grown in YE4S . SILAC media used was EMM ( 6 mM ammonium chloride ) supplemented with 0 . 25 mg/ml leucine , 0 . 15 mg/ml uridine , 0 . 04 mg/ml arginine and 0 . 03 mg/ml lysine [42] . YE4S used was Yeast extract ( Difco ) supplemented with adenine , leucine , uracil and histidine at 225 mg/l . The 3-stage screen was carried out as described above . Firstly , deletion mutants were incubated at 25°C for 12 to 20h in 300 μl of YE4S in 96-well plates then inoculated onto YE4S agar plates containing DiOC6 at 10 μg/ml ( Life technologies ) using a pin tool ( V & P Scientific , Inc ) and incubated at 25°C for 12 to 20h . The N/C ratio of each mutant strain was estimated by comparison with the wild type strain growing in the same plate using a Zeiss Axioskop 40 microscope . 102 of the 2 , 969 deletion mutants failed to grow on plates so were excluded from the screen . 366 mutant strains were selected for a secondary visual screen in liquid medium . Cells were collected from individual exponentially growing cultures of the 366 candidate mutants , stained with DiOC6 and visually screened to estimate the N/C ratio . The 97 strains selected for the tertiary screen were tagged with the nuclear envelope marker protein Cut11-GFP and the nuclear and cellular volumes measured to assess the N/C ratio [5] . Images were analysed using ImageJ ( NIH ) as previously described [5] . Nuclear volume of shape mutants was calculated using ImageJ . Unpaired Student’s t tests were used in Figs 1B and 1D , 2A and 2C , 4A and 4C to test statistical significance in pairwise comparisons . For visual screening , mutant cells were observed using a Zeiss Axioskop 40 microscope equipped with a 63x/1 . 4 NA objective and an AxioCam MRm camera . For protein and poly ( A ) +RNA localisation and N/C ratio measurement , cells were imaged using a DeltaVision Elite microscope ( Applied Precision ) comprised of an Olympus IX71 wide-field inverted fluorescence microscope , an Olympus Plan APO 60x oil , 1 . 42 NA objective , and a Photometrics CoolSNAP HQ2 camera ( Roper Scientific ) . Images were captured in 0 . 3 or 0 . 4 μm z-sections over 5 μm and deconvolved using SoftWorx ( Applied Precision ) . Projections of Cut11-GFP images were combined with bright field images . FITC intensity was measured using ImageJ ( NIH ) . 5 ml of exponentially growing cells were fixed with 70% ethanol at 4°C for at least 30 mins , washed with phosphate buffered saline ( PBS ) then resuspended in PBS containing 1 μg/ml fluorescein isothiocyanate ( FITC ) ( invitrogen ) . 4 , 6-diamidino-2phenylindole ( DAPI ) was used to stain DNA . The in situ hybridisation method used was described previously [43] . Oligo ( dT ) 50 3’-end labeled with cy3 was used as the hybridisation probe . DAPI was used to stain DNA . RNA was isolated by acid-phenol extraction and purified using RNeasy ( Qiagen ) . Biotinylated cRNA was hybridised onto GeneChip Yeast Genome 2 . 0 arrays ( Affymetrix ) which were scanned with a GeneChip Scanner 3000 and analysed with GCOS v1 . 4 ( Affymetrix ) using default analysis settings and global scaling normalisation [44] . NCBI GEO accession number: GSE81666 . Exponentially growing cells were incubated at 36°C or 25°C ( as indicated ) for 1 hour . Wild type ( heavy labeled ( H ) ) and rae1-167 ( light labeled ( L ) ) were mixed 1:1 by optical density ( inverse labels also mixed ) . Whole cell and nuclear enriched protein samples were extracted . For in-gel digestion each SILAC sample was loaded onto a NuPAGE Bis-Tris Protein Gel , 1 . 0 mm , 10-well ( Thermo Fisher ) , and allowed to migrate through the gel before being stained with coomassie blue . Polyacrylamide gel slices were prepared for mass spectrometric analysis using the Janus liquid handling system ( Perkin-Elmer ) . Each lane was excised into eight equally sized protein gel pieces , destained with 50% acetonitrile + 50 mM ammonium bicarbonate , reduced with 10 mM DTT , and alkylated with 55 mM iodoacetamide . After alkylation , the proteins were digested with 6 ng/μl trypsin overnight at 37°C . The resulting peptides were extracted in 2% formic acid/1% acetonitrile . Samples were analysed by LC-MS/MS . An LTQ-Orbitrap Velos coupled to an UltiMate 3000 HPLC system for on-line liquid chromatographic separation was used for data acquisition . The extracted peptides were separated over a 70 min gradient elution ( 75 μm × 50 cm C18 column ) with collision-induced dissociation ( CID ) selected as the activation method . MaxQuant 1 . 3 . 0 . 5 was used for data processing and quantification . Default MaxQuant parameters were used with the following adjustments: Lys6 and Arg6 were the heavy labels , ‘Filter labelled amino acids’ was deselected , re-quantify was selected with the instruction to keep low-scoring versions of identified peptides within parameter groups and match between runs was selected . Data was searched against a UniProt extracted S . pombe FASTA file amended to include common contaminants . Normalised H/L ratios were used for analysis in Perseus 1 . 4 . 0 . 2 . Average log2 rae1-167/WT ratios were calculated , data was annotated with ORFeome localisation data [19] and Gene Ontology terms ( default Perseus GO annotation lists ) , and 2D enrichment analysis was carried out [21] ( Benjamini-Hochberg FDR truncation threshold: 0 . 02 ) . The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [45] partner repository with the dataset identifier PXD004530 . Whole cell samples were produced by quenching cells by adding ice cold 100% ( w/v ) trichloroacetic acid to a final concentration of 10% . Cells were incubated on ice for at least 20 minutes , washed in ice-cold acetone and stored at -80°C . Pellets were washed and resuspended in lysis buffer ( 8 M urea , 50 mM ammonium bicarbonate , 5 mM EDTA and cOmplete Mini EDTA-free protease inhibitor cocktail ( Roche ) ) . 0 . 4 mm diameter acid washed glass beads ( Sigma ) were added and samples beaten to break cells ( FastPrep120 ) . Cell debris was pelleted and supernatant protein extracts stored at -80°C . Nuclear enriched samples were produced using a protocol based on [46] and Experiment 18 [47] . Cells were harvested from 1 L cultures by centrifugation , washed in S buffer ( 1 . 4 M sorbitol , 40 mM HEPES and 0 . 5 mM MgCl2 at pH 6 . 5 ) , resuspended in S buffer + 10 mM β-mercaptoethanol + 1 mM phenylmethanesulfonyl fluoride ( PMSF ) and incubated at 32°C for 10 minutes . Cells were harvested by centrifugation and the pellet was resuspended in S buffer + 1 mM PMSF containing 20 mg/gram cell pellet Zymolyase 100T ( Amsbio ) and incubated at 32°C until cell wall digestion was complete ( confirmed by SDS lysis ) . Remaining steps were performed on ice . Cells were pelleted and washed four times in S buffer then resuspended in 20 ml F buffer ( 18% Ficoll 400 ( w/v ) , 20 mM PIPES and 0 . 5 mM MgCl2 ) + 1 mM PMSF and lysed using a dounce homogeniser . The lysate was layered on 20 ml GF buffer ( 7% Ficoll ( w/v ) , 20% glycerol , 20 mM PIPES and 0 . 5 mM MgCl2 ) and centrifuged at 20 , 000 g for 30 minutes . The pellet was resuspended in 20 ml F buffer and centrifuged at 3 , 000 g for 15 minutes . The supernatant was centrifuged at 20 , 000 g for 25 minutes . Nuclear enriched pellets were resuspended in 250 μl 2X Laemmli buffer ( 100 mM TRIS ( pH 6 . 8 ) , 4% SDS , 20% glycerol and 0 . 2 M dithiothreitol ) , heated to 99°C for 10 minutes and centrifuged . Supernatants were harvested and stored at -80°C . Nuclear enrichment by this protocol was confirmed by preparation of whole cell and nuclear enriched samples of both wild type and rae1-167 cells grown at 36°C ( heavy and light labelled ) . Heavy labelled nuclear enriched sample was mixed 1:1 by protein concentration ( DC Protein Assay ( Bio-Rad ) ) with light labelled whole cell extract for each strain ( inverse label mixes also produced and analysed ) . Samples were analysed by SILAC mass spectrometry as described . Average log2 nuclear enriched/whole cell ratios were calculated , data was annotated with ORFeome localisation data [19] and 2D enrichment analysis was carried out [21] . Benjamini-Hochberg FDR was used for truncation , threshold value 0 . 02 .
|
Membrane-bound organelles are maintained at a size proportional to cell size during cell growth and division . How this is achieved is a little-understood area of cell biology . The nucleus is generally present in single copy within a cell and provides a useful model to study overall membrane-bound organelle growth and organelle size homeostasis . Previous mechanistic studies of nuclear size control have been limited to cell-free nuclear assembly systems . Here , we screened a near genome-wide fission yeast gene deletion collection for mutants exhibiting aberrant nuclear size , to identify , more systematically , components involved in nuclear size control . Roles for protein complexes previously implicated in nuclear mRNA export and membrane synthesis were identified . Molecular and genetic analysis of mRNA nuclear export gene mutant cells with enlarged nuclear size revealed that general accumulation of nuclear content , including bulk mRNA and proteins , accompanies the nuclear size increase which is dependent on new membrane synthesis . We propose that properly regulated nucleocytoplasmic transport and nuclear envelope expansion are critical for appropriate nuclear size control in growing cells .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"deletion",
"mutation",
"messenger",
"rna",
"vertebrates",
"animals",
"xenopus",
"animal",
"models",
"mutation",
"fungi",
"model",
"organisms",
"amphibians",
"experimental",
"organism",
"systems",
"cell",
"nucleus",
"cellular",
"structures",
"and",
"organelles",
"schizosaccharomyces",
"research",
"and",
"analysis",
"methods",
"physical",
"chemistry",
"surface",
"chemistry",
"nuclear",
"membrane",
"schizosaccharomyces",
"pombe",
"chemistry",
"yeast",
"biochemistry",
"rna",
"cell",
"biology",
"nucleic",
"acids",
"genetic",
"screens",
"artificial",
"membranes",
"gene",
"identification",
"and",
"analysis",
"genetics",
"biology",
"and",
"life",
"sciences",
"yeast",
"and",
"fungal",
"models",
"physical",
"sciences",
"frogs",
"organisms"
] |
2017
|
A systematic genomic screen implicates nucleocytoplasmic transport and membrane growth in nuclear size control
|
Sex hormone-binding globulin ( SHBG ) is a glycoprotein responsible for the transport and biologic availability of sex steroid hormones , primarily testosterone and estradiol . SHBG has been associated with chronic diseases including type 2 diabetes ( T2D ) and with hormone-sensitive cancers such as breast and prostate cancer . We performed a genome-wide association study ( GWAS ) meta-analysis of 21 , 791 individuals from 10 epidemiologic studies and validated these findings in 7 , 046 individuals in an additional six studies . We identified twelve genomic regions ( SNPs ) associated with circulating SHBG concentrations . Loci near the identified SNPs included SHBG ( rs12150660 , 17p13 . 1 , p = 1 . 8×10−106 ) , PRMT6 ( rs17496332 , 1p13 . 3 , p = 1 . 4×10−11 ) , GCKR ( rs780093 , 2p23 . 3 , p = 2 . 2×10−16 ) , ZBTB10 ( rs440837 , 8q21 . 13 , p = 3 . 4×10−09 ) , JMJD1C ( rs7910927 , 10q21 . 3 , p = 6 . 1×10−35 ) , SLCO1B1 ( rs4149056 , 12p12 . 1 , p = 1 . 9×10−08 ) , NR2F2 ( rs8023580 , 15q26 . 2 , p = 8 . 3×10−12 ) , ZNF652 ( rs2411984 , 17q21 . 32 , p = 3 . 5×10−14 ) , TDGF3 ( rs1573036 , Xq22 . 3 , p = 4 . 1×10−14 ) , LHCGR ( rs10454142 , 2p16 . 3 , p = 1 . 3×10−07 ) , BAIAP2L1 ( rs3779195 , 7q21 . 3 , p = 2 . 7×10−08 ) , and UGT2B15 ( rs293428 , 4q13 . 2 , p = 5 . 5×10−06 ) . These genes encompass multiple biologic pathways , including hepatic function , lipid metabolism , carbohydrate metabolism and T2D , androgen and estrogen receptor function , epigenetic effects , and the biology of sex steroid hormone-responsive cancers including breast and prostate cancer . We found evidence of sex-differentiated genetic influences on SHBG . In a sex-specific GWAS , the loci 4q13 . 2-UGT2B15 was significant in men only ( men p = 2 . 5×10−08 , women p = 0 . 66 , heterogeneity p = 0 . 003 ) . Additionally , three loci showed strong sex-differentiated effects: 17p13 . 1-SHBG and Xq22 . 3-TDGF3 were stronger in men , whereas 8q21 . 12-ZBTB10 was stronger in women . Conditional analyses identified additional signals at the SHBG gene that together almost double the proportion of variance explained at the locus . Using an independent study of 1 , 129 individuals , all SNPs identified in the overall or sex-differentiated or conditional analyses explained ∼15 . 6% and ∼8 . 4% of the genetic variation of SHBG concentrations in men and women , respectively . The evidence for sex-differentiated effects and allelic heterogeneity highlight the importance of considering these features when estimating complex trait variance .
Sex hormone-binding globulin ( SHBG ) is a protein secreted mainly by the liver that binds to the sex steroids , testosterone , dihydrotestosterone , and estradiol , transports them in the circulation , and influences their action in target tissues by regulating their bioavailability . SHBG thereby influences the expression of sex hormone sensitive phenotypes including sexual characteristics and reproductive function in men and women [1] . In addition to regulating sex steroid hormone effects , SHBG may exert independent effects through its own receptor [2] . Variation in SHBG concentration has also been associated with various chronic diseases including cancers [3] , polycystic ovary syndrome ( PCOS ) [4] , [5] and type 2 diabetes ( T2D ) [6] , [7] . Although SHBG is estimated to have a heritable component ( ∼50% ) [8] , little is known about the genetic regulation of SHBG . Polymorphisms at the SHBG gene locus have been associated with SHBG concentrations [9] , [10] , but much remains unknown about specific genetic variants that may determine circulating SHBG concentrations . Identifying genetic factors that influence SHBG may provide insights into the biology of sex steroid hormone regulation , metabolism and tissue effects that underlie their relationship with chronic diseases such as T2D as well as hormone-sensitive cancers such as breast and prostate cancer .
We identified nine loci associated with SHBG concentrations at the genome-wide significance threshold of p = 5×10−8 ( Table 1 and Figure 1 ) in a genome-wide association study ( GWAS ) meta-analysis of circulating SHBG concentrations in 21 , 791 men and women from 10 studies ( Table S1 ) . All nine lead SNPs at these loci had effects in the same direction ( seven with p<0 . 05 ) in the validation dataset of 7 , 046 men and women from six additional studies ( Table S2 ) . The strongest association was within the SHBG locus ( rs12150660 , p = 2×10−106 ) . Together , these nine lead SNPs explained 7 . 2% of the genetic variance ( assuming 50% heritability ) in SHBG concentrations . We next performed a series of additional analyses to explain more of the phenotypic variance ( Figure 2 ) . First , we hypothesized that genetic effects may be different in men and women , as SHBG concentrations are >50% higher in females than males , and may be differentially regulated between sexes . In a sex stratified analysis , three of the nine loci showed evidence of sex-differentiated effects at p<0 . 02 when we would not expect any signals to have reached this level of significance by chance . The associations at the 17p13 . 1-SHBG and Xq22 . 3 loci were stronger in males whereas the association at the 8q21 . 13 locus was stronger in females . To investigate the apparent differential sex effect for the X chromosome further we ran a recessive regression model for the X chromosome SNP rs1573036 in women in the Framingham Heart Study and found no association with SHBG suggesting the sex-differentiated effect is not the result of a recessive inheritance pattern . Sex stratified GWAS identified one novel signal in men , which showed no association in women ( 4q13 . 2: men p = 2 . 5×10−8 , women p = 0 . 66 , heterogeneity p = 0 . 003 ) . A series of conditional analyses were performed to identify statistically independent signals . At the SHBG locus , three apparently independent additional signals separate from the main index SNP were observed , based on low ( r2<0 . 05 ) pairwise correlations in HapMap ( rs6258 p = 2 . 7×10−46 , rs1625895 p = 1 . 2×10−14 and rs3853894 p = 2 . 5×10−11 ) . A series of iterative conditional analyses ( Table 2 ) involving SNPs at the SHBG locus generated a final regression model including six statistically independent SHBG SNPs . Four of these SNPs ( #1–4 Table 2 ) retained GWS when conditioned against the other five , and two were nominally associated ( SNP#5 p = 0 . 0001 , SNP#6 p = 0 . 01 ) . Re-running the GWAS meta-analysis adjusting for these six SNPs revealed evidence for three additional statistically independent ( pairwise HapMap r2<0 . 01 ) signals at the SHBG locus ( SNP#7 p = 1 . 5×10−7 , SNP#8 p = 4 . 6×10−5 , SNP#9 p = 9 . 9×10−6 ) ( Figure 3 ) . There were also two additional trans signals located at 2p16 . 3 and 7q21 . 3 ( Table 1 ) . Although the 2p16 . 3 signal dropped below GWS when combined with follow-up samples ( p = 1×10−7 ) , the index SNP at 2p16 . 3 is ∼300 kb away from a strong candidate gene , the luteinizing hormone receptor gene ( LHCGR ) . The majority of pair-wise correlations for the nine SHBG locus SNPs highlighted by our conditional analyses showed very low HapMap r2 values . However , the pairwise D′ values are often high ( Table S3 ) indicating that no or few recombination events have occurred between some SNPs , and that combinations of SNPs may be tagging un-typed variants on a common haplotype . To investigate this possibility , we performed more extensive analyses in a single study ( NFBC1966 , n = 4467 ) . We used a denser set of SNPs imputed from the June 2011 version of the 1000 Genomes data and performed model selection analyses . Model selection identifies a set of SNPs that best explain phenotypic variation , while simultaneously penalizing each SNP included in this set , and therefore correlated SNPs tend to be excluded from the final model . These analyses consistently included at least seven SNPs in the model , although it is hard to estimate the false-negative rate of using the reduced sample size . While we are underpowered to accurately pinpoint the exact number of independent signals , these analyses support the results of the conditional analysis and suggest that multiple variants at the SHBG locus are independently associated with SHBG concentrations . Data from an independent study , the InCHIANTI study , was used to calculate the proportion of genetic variance in SHBG concentrations explained when accounting for sex specific effects , the multiple signals of association at the SHBG locus , and the additional trans signals identified post conditional analysis . In men and women we explained ∼15 . 6% and ∼8 . 4% of the heritable component respectively . The SHBG locus accounted for ∼10% and ∼6 . 6% of the genetic variation in men and women respectively with the lead SNP in isolation accounting for ∼7 . 8% and ∼3 . 3% of the variation in men and women , respectively . We identified genes near the associated SNPs and explored their biologic relevance to SHBG . The genes associated with identified SNPs included the SHBG locus ( rs12150660 , 17p13 . 1 , p = 1 . 8×10−106 ) , PRMT6 ( rs17496332 , 1p13 . 3 , p = 1 . 4×10−11 ) , GCKR ( rs780093 , 2p23 . 3 , p = 2 . 2×10−16 ) , ZBTB10 ( rs440837 , 8q21 . 13 , p = 3 . 4×10−09 ) , JMJD1C ( rs7910927 , 10q21 . 3 , p = 6 . 1×10−35 ) , SLCO1B1 ( rs4149056 , 12p12 . 1 , p = 1 . 9×10−08 ) , NR2F2 ( rs8023580 , 15q26 . 2 , p = 8 . 3×10−12 ) , ZNF652 ( rs2411984 , 17q21 . 32 , p = 3 . 5×10−14 ) , TDGF3 ( rs1573036 , Xq22 . 3 , p = 4 . 1×10−14 ) , LHCGR ( rs10454142 , 2p16 . 3 , p = 1 . 3×10−07 ) , BAIAP2L1 ( rs3779195 , 7q21 . 3 , p = 2 . 7×10−08 ) , and UGT2B15 ( rs293428 , 4q13 . 2 , p = 5 . 5×10−06 ) ( Figure 1 ) . We used the online tool STRING ( www . string-db . org ) to perform pathway analyses to explore possible interactions between the SHBG gene and the proteins encoded by the 11 most plausible genes nearest the 11 SNPs listed above . There was an interaction noted between GCKR and JMJD1C which were associated with the lipoprotein fractions VLDL and HDL , respectively [11] . In an expanded analysis , we assessed protein interactions among SHBG and 67 genes within 500 kb of our 11 identified SNPs and uncovered additional protein interaction pathways . An interaction between two proteins encoded by GTF2A1L and STON1 was found; these proteins are co-expressed in testicular germ cells in the mouse [12] . An interaction between LHCGR and BRI3 encoded proteins that are associated with the G-protein coupled receptor complex in the human luteinizing hormone receptor was also identified [13] . Finally , an interaction between LHCGR and IAPP ( amylin ) proteins which are components of a ligand/G-protein receptor/G-protein alpha subunit complex was found ( database: www . reactome . com ) . Targeted analysis of two strong candidate genes , hepatocyte nuclear factor-4α ( HNF4α ) and peroxisome-proliferating receptor γ ( PPARγ ) did not identify any SNPs at HNF4α but did identify one SNP , rs2920502 , at PPARγ that reached statistical significance ( p = 9 . 9×10−5 ) and a second SNP at PPARγ , rs13081389 , that reached nominal significance ( p = 0 . 01 ) .
Several genes near the identified SNPs regulate sex steroid production and function . The NR2F2 locus ( 15q26 . 2 ) encodes a nuclear receptor important in testicular Leydig cell function , the primary source of gonadal testosterone production [15] , and has been linked to male infertility [16] . NR2F2 has also been associated with estrogen receptor alpha ( ERα ) signaling and may influence hormone responsivity in breast cancer [17] . PRMT6 ( 1p13 . 3 ) also encodes a nuclear receptor regulatory protein that mediates estrogen signaling as a co-activator of the estrogen receptor [18] . LHCGR ( 2p16 . 3 ) encodes the luteinizing hormone receptor which was associated with polycystic ovary syndrome ( PCOS ) in a recent GWAS [19] , [20] . PCOS is both a reproductive and metabolic disorder characterized by higher testosterone serum concentrations as well as an increased prevalence of obesity , insulin resistance , and T2D in women . Inappropriate secretion of luteinizing hormone leads to increased ovarian production of testosterone . Coincident lower SHBG concentrations contribute to increased bioavailable testosterone concentrations and the expression of both reproductive and metabolic phenotypes in PCOS [21] , [22] , [23] . The SLCO1B1 locus encodes a liver-specific transporter of thyroid hormone as well as estrogens which impact liver production of SHBG [24] . JMJD1C ( 10q21 . 3 ) , also known as TRIP 8 ( thyroid hormone receptor interactor protein 8 [25] ) , may impact SHBG concentrations via thyroid hormone effects on liver protein production . Thyroid hormone may alter SHBG production through effects on HNF4α which is known to regulate SHBG transcription [26] , [27] . Many of the genes identified are involved in carbohydrate and lipid metabolism and liver function . The GCKR locus ( 2p23 . 3 ) encodes a protein that regulates glucokinase activity and has been associated with T2D in several ethnic populations [28] , [29] , [30] , [31] . GCKR has been associated with metabolic and inflammatory traits including triglyceride concentrations and other lipid fractions [30] , [32] , fasting plasma glucose [33] , [34] , insulin concentrations , uric acid , c-reactive protein ( CRP ) , and non-alcoholic fatty liver disease which are all characteristic of the metabolic syndrome and T2D [28] , [35] , [36] , [37] , [38] , [39] , [40] , [41] , [42] . The SLCO1B1 locus ( 12p12 . 1 ) codes for a protein , hepatocyte protein anion-transporting polypeptide 1B1 , involved in liver metabolism of both endogenous and exogenous compounds [43] . Consistent with SLCO1B1's role in liver metabolism , the same SNP ( rs4149056 ) has been associated with circulating bilirubin concentrations in previous GWAS [44] . BAIAP2L1 ( 7q21 . 3 ) encodes a protein important in cytoskeleton organization [45] that has been associated with the inflammatory marker CRP in patients with arthritis [46] . BAIAP2L1 is also known as IRTKS ( insulin receptor tyrosine kinase substrate ) which is involved in insulin receptor signaling [47] and may relate to insulin resistant states including obesity and T2D [48] , [49] , [50] , [51] , [52] , [53] , [54] . We conducted a targeted analysis of PPARγ , a gene that influences SHBG gene expression in the liver [1] , [55] and is associated with T2D [56] , [57] . Our analysis identified one significant SNP ( rs2920502 , p = 9 . 9×10−5 ) and a second nominally significant SNP ( rs13081389 , p = 0 . 01 ) at PPARγ . Some of the identified genes involved in hepatic metabolism of lipids and carbohydrates may be affect SHBG concentrations indirectly through effects on the SHBG transcription regulator HNF4α although HNF4α itself was not identified in this meta-analyses [27] , [58] , [59] , [60] . The UGT2B15 locus ( 4q13 . 2 ) was significantly associated with SHBG concentrations in men but not women in this meta-analysis . UGT2B15 belongs to a family of genes ( the UGT2B gene family ) that code for enzymes involved in the metabolism of sex hormones through glucuronidation which allows for excretion of sex steroids through the kidney and the gut via bile excretion [61] , [62] , primary clearance mechanisms for sex steroids [63] . UGT2B15 is involved in the conjugation and inactivation of testosterone [64] . An association between rs293428 in the UGT2B15 locus and circulating SHBG concentrations in men is supported by a previous study demonstrating that a non-synonymous SNP in UGT2B15 ( rs1902023; D85Y ) is associated with serum SHBG concentrations in younger adult men [65] . UGT2B15 is thought to play a significant role in local tissue inactivation of androgens in androgen dependent prostate cancer [66] , [67] . The mechanism behind the influence of genetic variants in UGT2B15 on SHBG concentrations is unknown , but one may speculate that UGT2B15 affects the local androgenic environment in selected tissues , which in turn results in regulation of SHBG concentrations . In addition to UGT2B15 , three other genes near the identified SNPs are associated with carcinogenesis , particularly in the prostate and breast . ZBTB10 ( 8q21 . 13 ) , has been linked to breast cancer [68] . In breast cancer cell lines ZBTB10 is suppressed by ROS-microRNA27a thereby enhancing ERα alpha expression and mediating estrogen effects [17] . The ZNF652 ( 17q21 . 32 ) locus codes for a DNA binding protein thought to act as a tumor suppressor gene in breast cancer [69] , [70] , [71] that is also co-expressed with the androgen receptor in prostate cancer [72] . TDGF3 , teratocarcinoma derived growth factor 3 , is the only significant region identified on the X chromosome ( ( Xq22 . 3 ) . TDGF3 is a pseudogene of TDGF1 located on chromosome 3p23-p21 that has been associated with testicular germ cell tumors [73] . This GWAS meta-analysis incorporated data from approximately 22 , 000 men and women from 16 epidemiologic cohorts . The overall size of the study yields power but the meta-analysis of data from different epidemiologic studies requires the inclusion of different laboratory methods . The different studies used a variety of assay methodologies to measure serum SHBG concentrations although the vast majority were immunoassays ( Tables S1 and S2 , Text S1 ) with similar methodologies . Variation introduced by the use of different SHBG assays would result in loss of statistical power and likely bias toward the null . Additionally , the majority of women were post-menopausal as ascertained by self-report in all studies ( Table S1 ) . SHBG concentrations , like testosterone , decline only slightly across the menopause [74] so adjustment for menopause status is not necessary . SHBG may also increase with ovulation and be slightly higher in the luteal versus the follicular phase of the menstrual cycle in premenopausal women , but most studies did not collect data on menstrual phase at the time of SHBG measurement so adjustment for menstrual phase was not possible [75] . Finally , individuals were not excluded based on health status , therefore some individuals with chronic conditions that may affect hepatic production of or clearance of proteins including SHBG such as liver disease , renal disease , or severe malnutrition , may have been included in this analysis . SHBG synthesis in the liver is known to be affected directly or indirectly by estrogens , androgens and thyroid hormones and has been observed to be inversely associated with the higher insulin concentrations characteristic of insulin resistant states such as T2D [1] , [6] . In summary , the results of this GWAS reflect these influences . Three regions map to proteins related to hepatic function ( 12p12 . 1-SLCO1B1 [76] , 2p23 . 3-GCKR [77] and 10q21 . 3-JMJD1C [77] ) . In addition , 2p23 . 3-GCKR and 7q21 . 3-BAIAP2L1 [alias insulin receptor tyrosine kinase substrate ( IRTKS ) ] are involved in susceptibility to T2D [48] and insulin signaling [47] , respectively . Two signals also mapped to loci involved in thyroid hormone regulation ( 10q21 . 3-JMJD1C and 12p12 . 1-SLCO1B1 ) . One signal mapped to the receptor for luteinizing hormone 2p16 . 3-LHCGR [20] , the hormone that stimulates testosterone production . Five regions mapped to genes previously implicated in androgen and estrogen signaling ( 1p13 . 3-PRMT6 [18] , 8q21 . 13-ZBTB10 [17] , 12p12 . 1-SLCO1B1 [76] , 15q26 . 2-NR2F2 [78] , 4q13 . 2-UGT2B15 [63] ) . We have combined a conventional GWAS approach with detailed additional analyses , including sex stratification , conditional analysis and imputation from 1000 Genomes . Our results demonstrate that these approaches can lead to an appreciable gain in heritable variance explained . It does however highlight the complexity of elucidating individual variant causality through statistical approaches . In addition to the extensive allelic heterogeneity at the SHBG locus , our data identify loci with a role in sex steroid hormone metabolism , which may help elucidate the role of sex steroid hormones in disease , particularly T2D and hormone-sensitive cancers .
We performed a sensitivity analysis using samples from the 1966 Northern Finland Birth Cohort ( NFBC1966 ) study to further investigate allelic heterogeneity at the SHBG locus ( Text S1 ) . The conditional meta-analysis showed evidence for up to nine signals at the SHBG locus , but it is possible that these signals could be explaining a much smaller number of causal variants in the region . Since 1000 Genomes imputation allows us to assess the genetic variation associated with a phenotype across a much denser set of markers , it increases our power to detect allelic heterogeneity within a region . Therefore , 1000 Genomes imputation was carried out on all the samples in the NFBC1966 study and forward selection was used to identify the set of SNPs that best explain the variation in the SHBG phenotype . 1000 Genomes imputation was carried out using IMPUTE2 . The mean genotype probabilities for each SNP were calculated and used in the model selection step . Only SNPs 250 kb upstream and 250 kb downstream from the SHBG locus ( 7283453–7786700 bp ) were used in the analysis . All SNPs with MAF <0 . 1% or an imputation quality score less than 0 . 4 were excluded from the analysis . In total , 1978 SHBG region SNPs measured or imputed in 4467 samples from the NFBC1966 study were used in the sensitivity analysis . Forward selection was implemented in R ( version 2 . 13 . 0 ) using the stepAIC package to estimate the Akaikie Information Criterion ( AIC ) , an inclusion parameter . Given the high degree of correlation between the SNPs in this region , we increased the penalty ( k ) on the number of terms included in the model to 12 ( where it is usually two ) , to minimize possible over fitting . The final model included seven SNPs , adjusted for sex and BMI . We examined potential interactions among the proteins encoded by the SHBG locus and the proteins encoded by the 11 genes ( ZBT10 , TDGF1 , ZNF652 , PRMT6 , JMJD1C , GCKR , BAIAP2L1 , LHCGR , SLCO1B1 , UGT2B15 , NR2F2 ) closest to the 11 identified SNPs using pathway analysis with Search Tool for the Retrieval of Interacting Genes/Proteins ( STRING ) Pathways Analysis ( www . string-db . org ) . The interactions explored by STRING include direct ( physical ) and indirect ( functional ) associations . We then expanded the analysis to examine protein interactions among the SHBG gene and the proteins encoded by 67 genes within 500 kb of the 11 identified SNPs . We conducted targeted analysis of two strong candidate genes , hepatocyte nuclear factor-4α ( HNF4α ) and peroxisome-proliferating receptor γ ( PPARγ ) . Statistical significance thresholds were set correcting for the number of SNPs tested in each gene region ( ±100 kb ) .
|
Sex hormone-binding globulin ( SHBG ) is the key protein responsible for binding and transporting the sex steroid hormones , testosterone and estradiol , in the circulatory system . SHBG regulates their bioavailability and therefore their effects in the body . SHBG has been linked to chronic diseases including type 2 diabetes and to hormone-sensitive cancers such as breast and prostate cancer . SHBG concentrations are approximately 50% heritable in family studies , suggesting SHBG concentrations are under significant genetic control; yet , little is known about the specific genes that influence SHBG . We conducted a large study of the association of SHBG concentrations with markers in the human genome in ∼22 , 000 white men and women to determine which loci influence SHBG concentrations . Genes near the identified genomic markers in addition to the SHBG protein coding gene included PRMT6 , GCKR , ZBTB10 , JMJD1C , SLCO1B1 , NR2F2 , ZNF652 , TDGF3 , LHCGR , BAIAP2L1 , and UGT2B15 . These genes represent a wide range of biologic pathways that may relate to SHBG function and sex steroid hormone biology , including liver function , lipid metabolism , carbohydrate metabolism and type 2 diabetes , and the development and progression of sex steroid hormone-responsive cancers .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"medicine",
"public",
"health",
"and",
"epidemiology",
"epidemiology",
"endocrinology",
"reproductive",
"endocrinology",
"diabetes",
"and",
"endocrinology",
"genetic",
"epidemiology",
"endocrine",
"physiology"
] |
2012
|
A Genome-Wide Association Meta-Analysis of Circulating Sex Hormone–Binding Globulin Reveals Multiple Loci Implicated in Sex Steroid Hormone Regulation
|
In a 1997 seminal paper , W . Maddison proposed minimizing deep coalescences , or MDC , as an optimization criterion for inferring the species tree from a set of incongruent gene trees , assuming the incongruence is exclusively due to lineage sorting . In a subsequent paper , Maddison and Knowles provided and implemented a search heuristic for optimizing the MDC criterion , given a set of gene trees . However , the heuristic is not guaranteed to compute optimal solutions , and its hill-climbing search makes it slow in practice . In this paper , we provide two exact solutions to the problem of inferring the species tree from a set of gene trees under the MDC criterion . In other words , our solutions are guaranteed to find the tree that minimizes the total number of deep coalescences from a set of gene trees . One solution is based on a novel integer linear programming ( ILP ) formulation , and another is based on a simple dynamic programming ( DP ) approach . Powerful ILP solvers , such as CPLEX , make the first solution appealing , particularly for very large-scale instances of the problem , whereas the DP-based solution eliminates dependence on proprietary tools , and its simplicity makes it easy to integrate with other genomic events that may cause gene tree incongruence . Using the exact solutions , we analyze a data set of 106 loci from eight yeast species , a data set of 268 loci from eight Apicomplexan species , and several simulated data sets . We show that the MDC criterion provides very accurate estimates of the species tree topologies , and that our solutions are very fast , thus allowing for the accurate analysis of genome-scale data sets . Further , the efficiency of the solutions allow for quick exploration of sub-optimal solutions , which is important for a parsimony-based criterion such as MDC , as we show . We show that searching for the species tree in the compatibility graph of the clusters induced by the gene trees may be sufficient in practice , a finding that helps ameliorate the computational requirements of optimization solutions . Further , we study the statistical consistency and convergence rate of the MDC criterion , as well as its optimality in inferring the species tree . Finally , we show how our solutions can be used to identify potential horizontal gene transfer events that may have caused some of the incongruence in the data , thus augmenting Maddison's original framework . We have implemented our solutions in the PhyloNet software package , which is freely available at: http://bioinfo . cs . rice . edu/phylonet .
Accurate species trees , which model the evolutionary histories of sets of species , play a central role in comparative genomics , conservation studies , and analyses of population divergence , among many other applications . Traditionally , a species tree is inferred by sequencing a single locus ( gene ) in a group of species , its tree , known as the gene tree , is reconstructed using a method such as maximum likelihood , and this tree is declared to be the species tree . The underlying assumption is , obviously , that the gene tree and the species tree are identical , and hence reconstructing the former amounts to learning the latter . However , biologists have long recognized that this assumption is not necessarily always valid . Nevertheless , due to limitations of sequencing technologies , this approach remained the standard method until very recently . With the advent of whole-genome sequencing , complete genomes of various organisms are becoming increasingly available , and particularly important , data from multiple loci in organisms are becoming available . The availability of such data has allowed for analyzing multiple loci in various groups of species . These analyses have in many cases uncovered widespread incongruence among the gene trees of the same set of organisms . Therefore , while reconstructing a gene tree requires considering the process of nucleotide substitution , reconstructing a species tree requires , in addition , considering the process that resulted in the incongruities among the gene trees , so that the species phylogeny is inferred by reconciling these incongruities . In this paper , we address the problem of efficient inference of accurate species trees from multiple loci , when the gene trees are assumed to be correct , and their incongruence is assumed to be exclusively due to ( incomplete ) lineage sorting . We also address the integration of horizontal gene transfer , as a potential cause of gene tree incongruence , into the framework . Let us illustrate the process of lineage sorting and the way it causes gene tree incongruence . From an evolutionary perspective , and barring any recombination , the evolutionary history of a set of genomes would be depicted by a tree that is the same tree that models the evolution of each gene in these genomes . However , events such as recombination break “linkage” among the different parts of the genome , and those unlinked parts may take different paths through the phylogeny , which results in gene trees that differ from the species tree as well as from each other , due to lineage sorting . Widespread gene tree incongruence due to lineage sorting has been shown recently in several groups of closely related organisms , including yeast [1] , Drosophila [2] , Staphylococcus aureus [3] , and Apicomplexan [4] . In this case , gene trees need be reconciled within the branches of the species tree , as shown in Figure 1 . A few methods have been introduced recently for analyzing gene trees , reconciling their incongruities , and inferring species trees despite these incongruities . Generally speaking , each of these methods follows one of two approaches: the combined analysis approach or the separate analysis approach; see Figure 2 . In the combined analysis aproach , the sequences from multiple loci are concatenated , and the resulting “supergene” data set is analyzed using traditional phylogenetic methods , such as maximum parsimony or maximum likelihood; e . g . , [1] . In the separate analysis approach , the sequence data from each locus is first analyzed individually , and a reconciliation of the gene trees is then sought . One way to reconcile the gene trees is by taking their majority consensus; e . g . , [4] . Another is the “democratic vote” method , which entails taking the tree topology occurring with the highest frequency among all gene trees as the species tree . Shortcomings of these methods based on the two approaches have been analyzed by various researchers [5] , [6] . Recently , Bayesian methods following the separate analysis approach have been developed [7] , [8] . While these methods have a firm statistical basis , they are very time consuming , taking hours and days even on moderate-size data sets , which limits their scalability ( for example , the BEST tool of [7] took 800 hours on the yeast data set of [1] ) . In [9] , Maddison proposed a parsimony-based approach for inferring species trees from gene trees by minimizing the number of extra lineages , or minimizing deep coalesces ( MDC ) . A heuristic for this approach was later described in [10] . In [3] , Than et al . provided a two-stage heuristic for inferring the species tree under the MDC criterion . However , no exact solutions for computing the MDC criterion exist . In this paper , we provide a formal definition of the notion of extra lineages , first described in [9] . We then present exact solutions—an integer linear programming ( ILP ) algorithm and a dynamic programming ( DP ) algorithm—for finding the optimal species tree topology from a set of gene tree topologies , under the MDC criterion ( see Methods ) . Our solutions are based on two central observations: ( 1 ) the species tree is a maximal clique in the compatibility graph of the set of species clusters , and ( 2 ) quantifying the amount of incongruence between a set of gene trees and a species tree can be obtained by a simple counting of lineages within the branches of the species tree . The accuracy and computational efficiency of these solutions , as we demonstrate , allow for analysis of genome-scale data sets and analysis of large numbers of data sets , such as those involved in simulation studies . Given that MDC is a parsimonious explanation of the incongruence in the data , it is imperative that sub-optimal solutions are considered . The computational efficiency of our solutions allow for a rapid exploration of sub-optimal solutions . Last but not least , these exact solutions allow us to empirically study properties of MDC as an optimality criterion for inferring the species tree . We have implemented both exact solutions in the PhyloNet software package [11] . We reanalyze the Apicomplexan data set of [4] ( 268 loci from eight species ) , the yeast data set of [1] ( 106 loci from 8 yeast species ) , and a large number of synthetic data sets of species/gene trees ( up to 2000 loci from 8 species ) that we simulated using the Mesquite tool of [12] . For each data set , our method computed the species tree in at most a few seconds ( in some cases , it took 0 . 01 seconds ) , and produced very accurate species trees , as we show . In the case of the Apicomplexan data set , we provide a tree that is slightly different from the one proposed by the authors in [4] , and discuss this tree . For the yeast data set , we obtain a tree that is identical to the one proposed by the authors in [1] , as well as other studies , such as [7] . In addition to the quality of the species trees and efficiency with which our method inferred them , one advantage of our method is that it can be used in an exploratory fashion , to screen multiple species tree candidates , and study the reconciliation scenarios within the branches of each of them . We illustrate the utility of this capability on the yeast and Apicomplexan data sets . Further , for the Apicomplexan data set , we illustrate how to screen for possible horizontal gene transfer events using the reconciliation scenarios computed by other methods . Using the synthetic data sets , we study the statistical consistency , as well as convergence rate , of the MDC criterion . We also show that it may be sufficient to consider only the set of clusters induced by the gene trees , which , in practice , may be much smaller than the set of all clusters of species , thus achieving further reduction in computation time . Nonetheless , we present an example to illustrate that , in certain cases , focusing only on the gene tree clusters may result in a sub-optimal species tree under MDC . The computational efficiency of our methods , coupled with the promising properties of the MDC criterion , makes our methods particularly applicable to large , genome-scale data sets .
In this paper , we reanalyze two biological data sets under the MDC criterion: the Apicomplexan data set of [4] and the yeast data set of [1] . The Apicomplexan data set contains eight species: Babesia bovis ( Bb ) , Cryptospordium pavum ( Cp ) , Eimeria tenella ( Et ) , Plasmodium falciparum ( Pf ) , Plasmodium vivax ( Pv ) , Theileria annulata ( Ta ) , Toxoplasma gondii ( Tg ) , and Tetrahymena thermophila ( Tt ) . The authors in [4] identified 268 single-copy genes suitable for phylogenetic inference . For each gene , they reconstructed its tree using three methods ( maximum parsimony , maximum likelihood , and neighbor joining ) . Among the 268 gene trees , there were 48 different gene-tree topologies , the most frequent of which appears with about 18% frequency . [4] inferred the species tree using two different methods: the concatenation method and the majority consensus method , both of which produced the same tree , shown in Figure 3 , which the author presented as their hypothesis for the species tree of these eight Apicomplexan species . The yeast data set contains seven Saccharomyces species S . cerevisiae ( Scer ) , S . paradoxus ( Spar ) , S . mikatae ( Smik ) , S . kudriavzevii ( Skud ) , S . bayanus ( Sbay ) , S . castellii ( Scas ) , S . kluyveri ( Sklu ) , and the outgroup fungus Candida albicans ( Calb ) . [1] identified 106 genes , which are distributed throughout the S . cerevisiae genome on all 16 chromosomes and comprise about 2% of predicted genes . For each gene , they reconstructed its tree using the maximum likelihood and maximum parsimony methods . Among the 106 trees , more than 20 different gene-tree topologies were observed . The authors in [1] inferred the species tree using the concatenation method on the the sequences of the 106 genes . The resulting tree had 100% bootstrap support for each of its branches , and the tree topology is shown in Figure 4 . Further , to study various properties of the MDC criterion and our exact solutions , we ran the methods on synthetic data . To generate those data , we used the Mesquite [12] “Uniform Speciation” ( Yule ) module to generate 30 species trees , each with 8 taxa and total depth of 1 , 000 , 000 generations . Next , within the branches of each of these 30 species trees , 2000 gene trees were simulated under the colaescent process by using the Mesquite module “Coalescence Contained within Current Tree” . The effective population size Ne is 100 , 000 . These were the parameters also recommended by [10] . Finally , to enable studying the statistical consistency of methods , we simulated sampling of gene trees as follows: for each set of 2000 genes trees simulated within the branches of a species tree , we created random samples of 5 , 10 , 25 , 50 , 100 , 250 , 500 , 1000 , 1500 and 2000 trees; to obtain statistically significant results , we created 30 data sets for each sample size and averaged the results . It is worth mentioning that the parameters we used here , following [10] , produced a considerable amount of gene tree incongruence that was similar to the patterns we observed in the two biological data sets . We have implemented our methods in the PhyloNet software package [11] and analyzed the biological and synthetic data described above by inferring the species tree from the gene trees . In the case of the biological data , and since the “true” species tree is unknown , we compared the species tree inferred by our method to that hypothesized by the authors . We compared the species tree inferred by our method to the one reported in [4] and shown in Figure 3 in the case of the Apicomplexan data set , and to the one reported in [1] and shown in Figure 4 for the yeast data set . It is worth mentioning that the species tree inferred by Rokas et al . for the yeast data set was also inferred by the BEST Bayesian method [8] and reported in [7] . Since the species tree is known for the synthetic data , we studied the performance of methods by comparing the species tree they inferred against the true species tree . For this comparison , we used the normalized Robinson-Foulds ( RF ) measure [13] , which quantifies the average proportion of branches present in one , but not both , of the trees . A value of 0 for the RF distance indicates the two trees are identical , and a value of 1 indicates the two trees and completely different ( they disagree on every branch ) . Applying our method to the Apicomplexan data set , by using the 268 gene trees reported by [4] , there was a single optimal tree , which is shown in Figure 5A . The inferred tree requires in total 440 extra lineages to reconcile all 268 gene trees . This tree differs from the tree reported in [4] , and shown in Figure 3 , with respect to only the single clade ( Cp , ( Et , Tg ) ) . As Figure 3 shows , the tree reported by Kuo et al . places Cp as a sibling of the clade ( ( Et , Tg ) , ( ( Pf , Pv ) , ( Bb , Ta ) ) ) . However , it is important to note that as the authors reported , this placement of Cp has very low bootstrap support values of 38 , 42 , and 40 based on maximum likelihood , maximum parsimony and neighbor joining methods , respectively . Therefore , this grouping is not well-supported , even though both the concatenation and majority consensus methods compute it . Our method differed by placing Cp as a sibling of the clade ( Et , Tg ) . In fact , this grouping was advocated by [14] . To investigate this data set further , and particularly the placement of Cp , we employed our method in an exploratory mode: the method identified all maximal cliques in the compatibility graph of this data set , and for each maximal clique it computed the optimal fitting of all gene trees by minimizing the deep coalescences . The compatibility graph has 37 vertices ( which means there are 37 different clusters induced by all gene trees ) and 297 edges . In this graph , there are 247 maximal cliques , all of which have 6 vertices . This allows us to construct 247 fully binary species tree candidates . Figure 6 plots the number of extra lineages for all 247 species tree candidates , sorted from the lowest ( which is the optimal one with 440 extra lineages ) to the least optimal , which is a maximal clique requiring about 2200 extra lineages to reconcile all gene trees . We observed that next to the optimal maximal clique with 440 extra lineages , the next two sub-optimal maximal cliques within 100 lineage counts from the optimal one had 469 and 542 extra lineages , respectively . In other words , in addition to the optimal maximal clique , whose corresponding species tree is shown in Figure 5A , there were two additional trees very close in terms of the optimality of MDC . These two trees are shown in Figure 5B and 5C . It is worth noting that the tree in Figure 5B is exactly the tree reported in [4] , and that the tree in Figure 5C is the third way to group Cp , ( Et , Tb ) and ( ( Bb , Ta ) , ( Pf , Pv ) ) . In other words , while our method identified a single optimal tree , this tree along with the two close sub-optimal trees differ from each other by the placement of Cp . This fact is already reflected in the community by having two different hypotheses about this placement reported by Levine [14] and Kuo et al . [4] . The MDC criterion , however , supports Levine's hypothesis of the species tree , which has 29 fewer deep coalescence events than that proposed by Kuo et al . Beside the biological significance of the results , this analysis highlights several strengths of our method . It is very fast , and it can be run in an exploratory way , which allows the biologist to investigate multiple hypotheses that , while not all optimal , are very close in terms of the optimality criterion . Our method took a few seconds to compute all the values reported in Figure 6 . In other words , the method took a few seconds for 247 inferences of species tree candidates , each inference entailing the analysis of 268 gene trees . Second , while the majority consensus method reports a single tree , our method , when run in an exploratory manner , allows for exploring the “landscape” of the different tree topologies that could be species tree candidates . Third , the computation carried out very efficiently using our formulation allows us to explore the number of extra lineages on each of the branches of the inferred species tree , or any candidate tree that the biologist may wish to explore . For example , these numbers for the top three trees are shown on the branches of the trees in Figure 5 . Notice that across all three trees , only the number on one branch changes , and that is affected by the placement of Cp . These numbers may be useful in a further analysis aimed at estimating divergence times or population sizes , since these two parameters affect the number of extra lineages . The yeast data set contains 106 genes from eight species , with massive discordance among the gene trees , as reported in [1] . The authors concatenated all gene sequences and used maximum likelihood and maximum parsimony methods to reconstruct the species tree , and produced a species tree all of whose branches had 100% bootstrap support; this tree is shown in Figure 4 . For our analysis , we reconstructed the gene trees using a maximum parsimony heuristic , and used our method to infer the species tree . There was a single optimal tree found by our method , which is shown in Figure 9A . Clearly , the tree is identical to the one reported by [1] . This tree requires 127 extra lineages to reconcile all 106 gene trees . Edwards et . al . [7] also reported the same species tree using their Bayesian tool , BEST [8] . However , while our method took a fraction of a second to infer this species tree , the BEST tool took 800 hours . As we did with the Apicomplexan data set , we also generated all species tree candidates from the compatibility graph built from gene trees ( see Methods ) . The compatibility graph for this yeast data has 17 vertices and 94 edges . We then built 48 binary trees from the 48 maximal cliques in the compatibility graph , and scored the minimum number of deep coalescences required to reconcile all gene trees with each of the trees; these values are shown in Figure 9B . The majority of those species tree candidates require more than 200 extra lineages . The first seven best trees have 127 , 134 , 163 , 170 , 186 , 191 and 193 , respectively . The best tree ( the one with 127 extra lineages ) is the one shown in Figure 9A , while the other six are shown in Figure 10 . A very important point to make here is that these seven trees , while produced by our non-parametric method , include all six maximum posterior probability trees found by BEST in [7] . As before , this exploratory nature of our method allows us to investigate all seven of these trees , not only in terms of their topological differences , but also the implications that selecting one of them has on the reconciliation scenarios of the gene trees in the data set . The simulated data allowed us to investigate other aspects of the performance of our method , since the true species tree is known and we could compare the inferences made by our method against the true trees . One of the questions we sought to investigate is whether we need to use the compatibility graph of all species clusters or whether it is sufficient to focus on the compatibility graph of the gene trees . For n taxa , there are 2n−1 clusters ( including clusters that have a single taxon and the cluster that contains all taxa , but excluding the “empty cluster” ) ; hence , the compatibility graph of all clusters will have 2n−1 clusters , which becomes prohibitive for large values of n . The number of clusters exhibited by the gene trees , on the other hand , may be much smaller than 2n−1 in practice . Indeed , this is what we observed in the case of the Apicomplexian and yeast data sets . For both data sets , we have n = 8 ( the number of species ) , which means the number of all species clusters is 28−1 = 255 . However , the numbers of clusters exhibited by the gene trees were 37 and 17 , for the Apicomplexan and yeast data sets , respectively . This led to drastic reductions in actual running times . Further , this reduction was achieved without compromising the accuracy , as the optimal trees for both data sets were found in the compatibility graphs of the gene trees . To investigate this question further , we analyzed the synthetic data sets with respect to varying the sizes of gene tree samples ( see section Data above ) . For each sample of gene trees , we built the compatibility graph and tested whether the species tree is one of the maximal cliques in the graph . However , rather than the binary question of existence/non-existence , we quantified the proportion of branches in the true species tree that are missing from the closest maximal clique in the graph . If this proportion is zero , then the species tree is one of the maximal cliques . Figure 11A shows the results for this analysis . The results show that when only 25 gene trees are sampled , the compatibility graph provides good “coverage” that the true species tree is already one of the maximal cliques . Even for sample sizes 5 and 10 , the proportion of true species tree branches missing from the best maximal clique are 0 . 02 and 0 . 004 , respectively . These are negligible error rates . Two important observations are in order . First , these results are well-supported under the experimental conditions we used , which are the parameters used by [10] . Investigations of this question under a richer set of parameters is currently under way . In fact , it is safe to state that there will be points in the parameter space under which the species tree may not be a maximal clique , or even a subgraph , of the compatibility graph . Such a scenario occurs , for example , in the “anomaly zone” [5] , which is a setting of specific branch lengths under which the most likely gene tree may be different from the species tree . Second , even though under these parameters the true species tree is one of the maximal cliques in the compatibility graph , this does not imply that optimizing the MDC criterion will correctly identify the species tree . To investigate this issue , we ran our method on the data and compared the optimal tree under MDC with the true species tree . The results are shown in Figure 11B . In phylogenetics , two of the desirable properties of a phylogenetic method are statistical consistency and fast convergence . A method is statistically consistent if the error rate in its inference converges to 0 as the amount of data increases . In our case , a method is statistically consistent if the RF distance between the tree it infers and the true tree converges to 0 as more genes are sampled . Fast convergence means that not only does the method converge , but it does so “fast” , where “fast” here means “from small sample size of the data . ” The results in Figure 11 show that the MDC criterion has very low error rate . It is important to note that while the average RF distance for our method does not go down to zero , even when 2000 gene trees are used , the RF distance is negligible ( about 0 . 04 , which virtually amounts to zero wrong branches in the inferred tree ) . Yet , the interesting observation here is that combining the results of Figures 11A and 11B , we drew the conclusion that the species tree is one of the maximal cliques in the compatibility graph ( particularly for samples of size at least 25 ) , yet it is not the one with the minimum number of extra lineages . Figure 11C shows the difference between the number of extra lineages in the true species tree and that number in the tree inferred by our method . Since our method is guaranteed to find the optimal tree in terms of the number of deep coalescens , this difference ( when subtracting the latter number from the former ) is non-negative . The results in the figure confirmed our hypothesis: in a few cases , the tree that minimizes the number of deep coalescences is not necessarily the true species tree . Instead , in this case , the species tree is sometimes a sub-optimal one . We observed this same issue even with the two biological data sets , particularly the Apicomplexan one . We then investigated how sub-optimal the species tree may be . In all cases when the species tree was not the optimal tree , it was either the first or second sub-optimal one . Once again , this matches results of the analysis of the two biological data sets . It is important to note that in practice only gene sequences are given and gene trees need to be inferred . Error in the inferred gene trees may affect the performance of the method negatively . Under the MDC criterion , error in the inferred gene trees may masquerade as deep coalescence events , but also may “cancel out” some of the incongruence truly caused by deep coalescence . Therefore , extending the simulation study to include inference of the gene trees , rather than assume they are given , is a task we identify for immediate future research . Nonetheless , the analysis of the two biological data sets above includes the inference of the gene trees themselves , and in these two cases , the MDC criterion provides accurate results . We finish by showing an example of three gene trees for which the tree that minimizes their deep coalescences is not one of the maximal cliques in the compatibility graph of these three gene trees . Consider the three trees in Figure 12 . The compatibility graph that is built from their induced clusters is shown in Figure 13 . A minimum vertex-weighted clique of the graph is highlighted with thick lines . Its weight is 1+2+4 = 7 , and it corresponds to the leftmost tree in Figure 12 . This means that this tree requires seven extra lineages to reconcile the three trees in Figure 12 . However , the tree in Figure 14 requires only six extra lineages to reconcile all those three trees . We note that it induces cluster {a , b , c , e} that does not appear in any of the three gene trees . This illustrates that in theory the optimal tree under the MDC criterion may not be found in the compatibility graph of the clusters induced by the gene trees .
Let X be a set of taxa . A phylogenetic tree T = ( V , E ) , where V and E are its nodes and edges , is a tree with a bijection from X to its leaf set . Tree T is said to be rooted if the edges in E are directed and there is a single internal node r with in-degree 0 . Except when explicitly stated , in this paper trees are assumed to be rooted and binary . For a node v∈V , we denote by T ( v ) the clade , or subtree of T , rooted at v . The set of leaf labels of T ( v ) is called a cluster , denoted by CT ( v ) . Cluster X and single-element clusters are called trivial . For a cluster A , we denote by MRCAT ( A ) the most recent common ancestor ( also known as the least common ancestor ) of taxa in A in tree T . For two clusters A , B , we say that they are compatible if either A⊆B , B⊆A or A∩B = Ø . Informally , it means that there exists a rooted tree that induces , or contains , both A and B . A collection of pairwise compatible clusters uniquely defines a rooted tree [18] . In [9] , Maddison introduced the concept of extra lineages and a parsimony approach , which we call the “minimize deep coalescences” approach , for species tree inference based on minimizing the number of extra lineages . We first define a mapping between a species tree and gene tree which allows for a precise definition of the number of extra lineages . We then prove that this number can be computed independently for each cluster in the species tree . Suppose we are given a gene tree T and a species tree T′ on the same taxon set X . We fit tree T into T′ by mapping each node v of T according to three rules below: Figure 15 shows an example of such a mapping . In this figure , we can see that for branch ( u′ , v′ ) there are two lineages , one being the lineage of the common ancestor of species A , B , C , and one being lineage D . In the case where T and T′ are identical topologically , then we can easily see that there is only one lineage in ( u′ , v′ ) , that is one lineage for the common ancestor of A , B , C and D . Therefore , for the branch ( u′ , v′ ) in Figure 15 , the number of extra lineages is 2−1 = 1 . Formally , we define the number of extra lineages in a branch of T′ as the number of lineages exiting it minus 1 , and the number of extra lineages for T′ as the sum of those numbers in all of its branches . Each pv in T′ that is the image of the mapping of an internal node v in T is a coalescence event . In Figure 15 , there are two coalescence events in branch ( v′ , w′ ) , but there are no coalescent events in branch ( u′ , v′ ) . We can establish a relationship between the number of extra lineages and the number of coalescence events as follows . Consider a branch ( u′ , v′ ) of T′ . There are exactly |CT′ ( v′ ) | species in the subtree T′ ( v′ ) . If there were no coalescence among those species , then there would be |CT′ ( v′ ) | lineages exiting ( u′ , v′ ) . However , each coalescence event merges two lineages into one , and we note that under the mapping's conditions whenever there is a coalescence among lineages from species in CT′ ( v′ ) , it must occur either in a branch of T′ ( v′ ) or in ( u′ , v′ ) . Therefore , the actual number of lineages exiting ( u′ , v′ ) is equal to |CT′ ( v′ ) | minus the total number of coalescence events among species in T′ ( v′ ) . We have the following lemma . Lemma 1 . Let n ( v′ ) be the number of coalescence events occurring among species in CT′ ( v′ ) . Then , the number of extra lineages in branch ( u′ , v′ ) is ( 1 ) We note that this lemma may not be true without the conditions of the mapping defined above . If we do not have Conditions 2 and 3 , then lineages A , B , and C in Figure 15 , for example , need not coalesce in branch ( v′ , w′ ) . They can coalesce at a branch above u′ , and in this case there are four lineages ( and therefore , three extra ones instead of one ) in ( u′ , v′ ) . As we have seen , the number of extra lineages reflects the amount of incongruence between two trees . It is small if two trees are quite similar , and in fact zero if they are identical topologically . Given a set of gene trees , one approach to inferring the species tree is to minimize the number of extra lineages: Problem 1 ( Species Tree Inference Using the MDC Criterion ) . Input: A set of gene trees . Output: A tree T such that the total number of extra lineages required to reconcile all gene trees of within T is minimized . Let T be a gene tree and T′ be a species tree . It seems that the number of extra lineages in a branch ( u′ , v′ ) of T′ depends on both T and T′ . The following theorem shows it depends only on the gene tree T and on the cluster CT′ ( v′ ) . Because of its independence on T′ , we can denote it by α ( CT′ ( v′ ) , T ) . Theorem 2 . Let T be a gene tree and T′ be a species tree . Let ( u′ , v′ ) be a branch of the species tree T′ . Denote by t1 , … , tk all the maximal clades of T such that for 1≤i≤k . Then , the number of extra lineages in ( u′ , v′ ) is ( 2 ) Proof . Consider a clade ti , 1≤i≤k . First of all , because ti is clade of T all species in ti must coalesce into a single lineage ( and they must coalesce either in a branch of T′ ( v′ ) or ( u′ , v′ ) under the mapping's conditions defined above ) . Second , because ti is a maximal clade such that , that lineage will not coalesce with any other lineages in T′ ( v′ ) or in branch ( u′ , v′ ) ( for otherwise , we will obtain a bigger clade in T whose leaf set is still a subset of CT′ ( v′ ) , a contradiction ) . By Lemma 1 , the number of coalescence events occurring among species of ti is . We also note that . So , by applying this lemma again , we obtainAs an example , consider trees T and T′ in Figure 15 . From the figure , we see that there are no extra lineages in branch ( v′ , w′ ) . The cluster under w′ is {A , B , C} . The clade ( A , ( B , C ) ) is a maximal clade of T with only species from {A , B , C} . Therefore , the number of extra lineages is 1−1 = 0 . On the other hand , consider branch ( u′ , v′ ) . There are two maximal clades in T with species from {A , B , C , D}: ( A , ( B , C ) ) and D . So , the number of extra lineages in ( u′ , v′ ) is 2−1 = 1 . Linear programming ( LP ) is an algorithmic technique for optimizing a linear objective function , cx , where c is a vector of coefficients and x is a vector of variables , subject to a set of linear constraints Ax≤b , where A is a matrix of coefficients and b is a vector of coefficients . This is usually written in the formWhen the variables x are required to be integers , the problem becomes an integer linear programming ( ILP ) . Solving an ILP problem is NP-hard in general . Nonetheless , software tools for efficiently solving large and hard instances in practice have been developed . One such ( commercial ) tool is CPLEX , which was developed by the company ILOG ( http://www . ilog . com/ ) . In this section , we show how to use ILP to optimize the MDC criterion . Using Theorem 2 , it is possible to compute the number of extra lineages contributed by each individual cluster without the need of prior knowledge of the species tree . We can therefore solve Problem 1 exactly by seeking a maximal set of compatible clusters whose total number of extra lineages is minimum . Based on our empirical observation , we find that the species tree is almost always an agglomeration of compatible clusters , each of which appears in at least one of the input gene trees ( see Results and Discussion ) . Based on these two observations , we propose the following method to approximately solve Problem 1: We now give the details of the method , using the illustration in Figure 16 as the running example . We can find the optimal species tree without the need to find a maximum vertex-weighted clique in the compatibility graph G by employing dynamic programming . Dynamic programming ( DP ) is a divide-and-conquer algorithmic technique that breaks a problem into sub-problems , solves the sub-problems , and then uses those solutions in an efficient way to form the solution to the main problem . For a problem to be amenable to a DP solution , it has to exhibit certain properties . For more details , the reader is referred to any standard textbook on algorithms; e . g . , [20] . We now describe how to solve the MDC optimization problem using a DP algorithm . Let t′ be a rooted binary phylogenetic tree on a fixed taxon subset of X . Given a collection of gene trees , let us denote the sum of for all clusters B in t′ , including A . Further , let be the minimum value of over all possible binary trees t′ on A . If and are the two subtrees whose roots are the children of t′ , then clearly we haveThe quantity is fixed for each A , and therefore , if t′ is an optimal tree on A such that is minimum , then and must also be minimum . This allows us to compute recursively as follows . Although the algorithm described above only returns the number of extra lineages , we can easily modify it so that we can actually reconstruct the optimal species tree . For each i , 3≤i≤|X| , in Step 3 , we also record two pointers to optimal subclusters A1 and A2 . By backtracking those pointers starting with cluster X , we can obtain the optimal set of compatible clusters . Any tree induces exactly |X|−2 nontrivial clusters . Therefore , . For every A⊆X , there are at most subsets of A to look at , and hence Step 3 is executed at most times . The running time of the algorithm is then . The collection described in the algorithm only contains clusters induced by gene trees in . However , we can replace it by the collection of all nonempty subsets of X ( there are 2|X|−1 such subsets ) . In this case , the running time of the algorithm is bounded by . Although it is exponential , it is significantly better than a brute-force approach that examines all ( 2|X|−3 ) ! ! binary rooted phylogenetic trees on X .
|
Inferring the evolutionary history of a set of species , known as the species tree , is a task of utmost significance in biology and beyond . The traditional approach to accomplishing this task from molecular sequences entails sequencing a gene in the set of species under consideration , reconstructing the gene's evolutionary history , and declaring it to be the species tree . However , recent analyses of multiple gene data sets , made available thanks to advances in sequencing technologies , have indicated that gene trees in the same group of species may disagree with each other , as well as with the species tree . Therefore , the development of methods for inferring the species tree despite such disagreements is imperative . In this paper , we propose such a method , which seeks the tree that minimizes the amount of disagreement between the input set of gene trees and the inferred one . We have implemented our method and studied its performance , in terms of accuracy and computational efficiency , on two biological data sets and a large number of simulated data sets . Our analyses , of both the biological and synthetic data sets , indicate high accuracy of the method , as well as computationally efficient solutions in practice . Hence , our method makes a good candidate for inferring accurate species trees , despite gene tree disagreements , at a genomic scale .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] |
[
"computational",
"biology/population",
"genetics",
"computer",
"science/applications",
"genetics",
"and",
"genomics/comparative",
"genomics",
"evolutionary",
"biology/evolutionary",
"and",
"comparative",
"genetics",
"computational",
"biology/comparative",
"sequence",
"analysis",
"evolutionary",
"biology/genomics",
"computational",
"biology/evolutionary",
"modeling",
"computational",
"biology/genomics",
"evolutionary",
"biology/bioinformatics",
"genetics",
"and",
"genomics/bioinformatics",
"genetics",
"and",
"genomics/population",
"genetics"
] |
2009
|
Species Tree Inference by Minimizing Deep Coalescences
|
Leishmania braziliensis is the main causative agent of cutaneous leishmaniasis in Brazil . Protection against infection is related to development of Th1 responses , but the mechanisms that mediate susceptibility are still poorly understood . Murine models have been the most important tools in understanding the immunopathogenesis of L . major infection and have shown that Th2 responses favor parasite survival . In contrast , L . braziliensis–infected mice develop strong Th1 responses and easily resolve the infection , thus making the study of factors affecting susceptibility to this parasite difficult . Here , we describe an experimental model for the evaluation of the mechanisms mediating susceptibility to L . braziliensis infection . BALB/c mice were inoculated with stationary phase promastigotes of L . braziliensis , isolates LTCP393 ( R ) and LTCP15171 ( S ) , which are resistant and susceptible to antimony and nitric oxide ( NO ) , respectively . Mice inoculated with LTCP393 ( R ) presented larger lesions that healed more slowly and contained higher parasite loads than lesions caused by LTCP15171 ( S ) . Inflammatory infiltrates in the lesions and production of IFN-γ , TNF-α , IL-10 and TGF-β were similar in mice inoculated with either isolate , indicating that these factors did not contribute to the different disease manifestations observed . In contrast , IL-4 production was strongly increased in LTCP393 ( R ) -inoculated animals and also arginase I ( Arg I ) expression . Moreover , anti-IL-4 monoclonal antibody ( mAb ) treatment resulted in decreased lesion thickness and parasite burden in animals inoculated with LTCP393 ( R ) , but not in those inoculated with LTCP15171 ( S ) . We conclude that the ability of L . braziliensis isolates to induce Th2 responses affects the susceptibility to infection with these isolates and contributes to the increased virulence and severity of disease associated with them . Since these data reflect what happens in human infection , this model could be useful to study the pathogenesis of the L . braziliensis infection , as well as to design new strategies of therapeutic intervention .
Leishmaniasis comprises several diseases caused by protozoans of the genus Leishmania . The most common disease in Brazil is American tegumentary leishmaniasis ( ATL ) , caused by L . braziliensis . In the Americas , the main invertebrate vectors for this parasite , are sand flies of the genus Lutzomyia [1] . The most studied parasite of the genus Leishmania is L . major . Human infection with this pathogen is generally benign and eventually resolves spontaneously , resulting in lifelong immunity . On the other hand , infection with L . braziliensis is chronic and causes latency , which may lead to parasite dissemination to the nasal and oral mucosa years after resolution . Even chemotherapeutic treatment does not exclude the possibility of developing mucocutaneous leishmaniasis [2]–[4] . Protection against , and susceptibility to L . major , another cutaneous leishmaniasis-causing parasite , have been clearly established in mouse models of infection . In mice , Th1 responses [5] involving nitric oxide ( NO ) , and the cytokines IL-12 , IFN-γ and TNF-α [6] , [7] , result in parasite killing [8] , [9] . In contrast , Th2 responses , which are characterized by the production of IL-4 , IL-13 and IL-10 , result in susceptibility to infection [10]-[13] . CD4+CD25+ regulatory T cells are also important sources of IL-10 and contribute significantly to susceptibility [14] , [15] . Similarly , Th1 responses have been shown to be necessary for parasite killing in mouse models of L . braziliensis infection [16]; however , it has been difficult to develop experimental models for studying susceptibility factors because most mouse strain develop strong Th1 responses that easily control L . braziliensis infection [17] . In humans , cutaneous ( CL ) and mucocutaneous leishmaniasis ( ML ) caused by L . braziliensis infection are also associated with a strong production of Th1 cytokines and marked migration of inflammatory mononuclear cells to lesion sites [18]–[22]; although spontaneous resolution is observed in only 30% of patients [23] . Also a large range of clinical manifestations is observed in patients with L . braziliensis cutaneous leishmaniasis , showing a great difference between the diseases caused by L . major and L . braziliensis . A complex interplay between host traits and intrinsic properties of the parasite L . braziliensis that contribute to the variety of clinical presentations is not clearly understood . Experimental infection of mice with two L . braziliensis isolates from patients with either mild or severe lesions resulted in distinct clinical features and different patterns of chemokine production; however , no differences were observed in parasite replication [24] , [25] . Intrinsic characteristics of the parasites that result in increased capacity to survive inside human macrophages , such as resistance to nitric oxide ( NO ) in some L . braziliensis and L . amazonensis isolates , have been associated with more severe forms of the disease [26] . Despite these findings , the true immunological mechanisms that mediate susceptibility to L . braziliensis remain poorly understood . To address this issue , we developed an experimental model in which BALB/c mice were inoculated with stationary phase promastigotes from L . braziliensis isolates obtained from CL patients that were refractory or responsive to antimony treatment , and that presented different severities of disease manifestations . We characterized the experimental infection with these two isolates in mice and found that similar to the difference in disease severity between the human hosts from which these strains were isolated , the resistant isolate caused a more severe disease in mice than the susceptible isolate . The increased lesion development caused by increased parasitic replication was associated with the production of IL-4 in response to the resistant isolate . This interesting model of infection can be useful to further studies to understand the variability of clinical manifestations of the disease and to design immunological targets to be used to control the infections .
Female 6–8 week old BALB/c mice were used in all experiments . The animals were maintained at the animal holding facility of the department of Biochemistry and Immunology of the Ribeirão Preto Medical School – University of São Paulo , and all procedures were approved by the local ethics committee for animal care and research “Ethics Committee in Animal Research of the FMRP-USP” . Leishmania braziliensis isolates LTCP393 ( R ) and LTCP15171 ( S ) were obtained from cutaneous ulcers of patients from the endemic area of Corte de Pedra , BA , Brazil , specifically for this study , and all human subjects were briefed on procedures and signed informed consent documentation . All work with human subjects was carried out under “Federal University of Bahia Ethical Committee” approval number 5/2006 . L . braziliensis LTCP15171 ( S ) was isolated from a 38-year-old male patient who had a single skin ulcer with an area of 120 mm2 that was cured after one course of antimony therapy . L . braziliensis LTCP393 ( R ) was isolated from a 26-year-old male patient that had several skin lesions ranging in size from 100 to 1575 mm2 and mucosal disease . This patient was treated with several courses of antimony , pentoxifylin , and amphotericin B over a period of 20 years before being cured . These parasites were isolated in N . N . N . media and frozen in liquid nitrogen immediately after isolation . Promastigotes were grown in Schneider's insect medium ( Sigma , Saint Louis , USA ) supplemented with 20% heat-inactivated fetal calf serum ( Cultilab , Campinas , SP , Brazil ) , 4 mM NaHCO3 , 100 U/ml penicillin , 100 µg/ml streptomycin ( all from Gibco , Grand Island , NY , USA ) , and 2% v/v male human urine at 25°C . At 5th day of culture , both isolates reached stationary phase , after inoculation of 1×106 promastigotes in 6 ml of medium . The amount of metacyclic parasites at the chosen day was analyzed using Ficoll 400 ( Pharmacia-GE Healthcare , Uppsala , Sweden ) gradient as described [27] , [28] , and was similar for LTCP393 ( R ) and LTCP15171 ( S ) isolates ( 0 . 8050%±0 . 01528 and 1 . 132%±0 . 2122 , respectively ) . NO resistance assays were performed as described [26] , [29] . LTCP393 ( R ) was resistant to 16 mM NaNO2 ( NO donor ) , and LTCP15171 ( S ) was susceptible to 4 mM NaNO2 . Antimony resistance was tested by culturing BALB/c mice peritoneal macrophages isolated and cultured as described [30] , in the presence or absence of 10 , 30 and 90 µg/ml Glucantime ( Sanofi Aventis Farmacêutica , Suzano , SP , Brazil ) , for 24 , 48 and 72 h . Compared to non-treated infected cells , only in LTCP15171 ( S ) -infected macrophages there was a substantial reduction in intracellular parasite numbers . We therefore will refer to these parasites as resistant to antimony and NO ( LTCP393 ( R ) ) , and susceptible to antimony and NO ( LTCP15171 ( S ) ) . To determine the infectivity of the isolates , macrophages were incubated with 5 and 10 parasites per cell and the percentage of infected cell and number of intracellular parasites determined . Before the beginning of the experiments , the isolates were submitted to passages into hamsters and BALB/c mice , for recovering of virulence and selection of parasites . During the experiments , parasite virulence was assured by constant maintenance and re-isolation from infected BALB/c mice . We used a well established protocol of inoculation [31]–[35] , in which the right ear dermis of BALB/c mice was inoculated with stationary phase promastigotes ( 106 parasites in 10 µl of sterile saline ) using a 27 . 5-gauge needle . Lesion size , which was defined as the difference in thickness between the infected ear and the non-infected contralateral ear , was monitored weekly using a digital caliper ( Mitutoyo , Suzano , SP , Brazil ) . The parasite load was determined using a quantitative limiting dilution assay as previously described [36] . The rat anti-IL-4 mAb was purified from ascites of mice injected with the hybridoma 11B11 . IL-4 was neutralized by intraperitoneal ( i . p . ) injection of 2 mg purified mAb one day before infection . Additional i . p . injections of 1 mg purified mAb were performed twice a week for seven weeks . Controls received 1 mg of normal rat IgG diluted in PBS . Ears from infected mice were collected and incubated at 37°C for one hour in RPMI-1640 medium containing 2 mM L-glutamine , 100 U/ml penicillin , 100 µg/ml streptomycin ( all from Gibco ) and 500 µg/ml Liberase CI ( Roche , Basel , Switzerland ) . The tissues were processed inside Medcons using a Medimachine ( both from Becton & Dickinson Biosciences , San Diego , CA , USA ) . After processing , the cells were filtered through a 50 µm filter , viability was assessed by trypan blue exclusion , and the cell concentration was adjusted . Immunostaining was performed with anti-CD3 , CD19 , CD4 , CD8 , CD25 , CD11b , CD11c and Gr1 antibodies conjugated to FITC , PE or PerCP fluorochromes . For regulatory T cell phenotyping , CD4+CD25+ cells were stained with anti-FoxP3 , CD103 , CTLA-4 and GITR antibodies conjugated to PE . For intracellular staining , the cells were permeabilized with a Cytofix/Cytoperm kit ( BD Biosciences ) according to the manufacturer's guide . For all analyses , the results were compared to those obtained with cells stained with isotype control antibodies ( all antibodies were from BD Biosciences and eBiosciences , San Diego , CA , USA ) . Cell acquisition was performed using a FACSort flow cytometer and CellQuest software ( BD Biosciences ) . Data were plotted and analyzed using Cell Quest ( BD Biosciences ) and FlowJo software ( Tree Star , Ashland , OR ) . Single cell suspensions of draining retromaxilar lymph nodes were prepared aseptically , diluted to a concentration of 2×106 cells/ml and dispensed into 48-well plates in a total volume of 500 µl of complete RPMI-1640 medium ( Gibco ) ( 1×106 cells/well ) with or without live stationary phase L . braziliensis promastigotes at a ratio of five parasites to one cell ( 5×106 parasites/well ) . Cell culture supernatants were harvested after 72 h of culture at 37°C in 5% CO2 , and the levels of IFN-γ , IL-4 , IL-10 and TGF-β were determined by ELISA using commercial kits ( BD Biosciences and R&D Systems , Minneapolis , MN , USA ) . Total RNA was extracted from whole fragments of infected ears using the Promega RNA extraction kit ( Promega , Madison , WI , USA ) . RNA was quantified using a spectrophotometer ( BioMate 3; Thermo Spectronic , Rochester , NY , USA ) and cDNA was synthesized using 1 µg RNA through an RT reaction using ImProm-II reagents ( Promega ) according to the manufacturer's instructions . Real-time PCR quantitative mRNA analyses were performed using the Platinum SYBR Green qPCR SuperMix-UDG with ROX reagents ( Invitrogen , Carlsbad , CA , USA ) and the ABI Prism 7000 sequence detection system ( Applied Biosystems , Warrington , UK ) . The sequences of murine primers were designed using PrimerExpress software ( Applied Biosystems ) and nucleotide sequences present in the GenBank database . Relative gene expression levels were calculated according to the instructions from the Applied Biosystems user's bulletin ( P/N 4303859 ) . Briefly , gene expression levels were normalized to the level of β-actin in each sample using the cycle threshold ( Ct ) method . Negative controls without RNA and without RT were also performed . The threshold for positivity of real-time PCR was determined based on the expression of the specific mRNAs in the ears of uninfected mice using the equation 2−ΔΔCt . Murine primer sequences utilized: β-actin sense: 5′- AGC TGC GTT TTA CAC CCT TT - 3′ , β actin anti-sense: 5′ - AAG CCA TGC CAA TGT TGT CT – 3′; IFN-γ sense: 5′- GCA TCT TGG CTT TGC AGC T – 3′ , IFN-γ anti-sense: 5′ - CCT TTT TCG CCT TGC TGT TG – 3′; IL-4 sense: 5′ - GAA TGT ACC AGG AGC CAT ATC – 3′ , IL-4 anti-sense: 5′ - CTC AGT ACT ACG AGT AAT CCA – 3′; IL-10 sense: 5′ - TGG ACA ACA TAC TGC TAA CCG – 3′ , IL-10 anti-sense: 5′ - GGA TCA TTT CCG ATA AGG CT – 3′; TGF-β sense: 5′ - GCT GAA CCA AGG AGA CGG AAT – 3′ , TGF-β anti-sense: 5′ - GCT GAT CCC GTT GAT TTC CA – 3′; TNF-α sense: 5′ - TGT GCT CAG AGC TTT CAA CAA - 3′ , TNF-α anti-sense: 5′ - CTT GAT GGT GGT GCA TGA GA – 3′; inducible nitric oxide synthase ( iNOS ) sense: 5′ - CGA AAC GCT YCA CTT CCA A – 3′ , iNOS anti-sense: 5′ - TGA GCC TAT ATT GCT GTG GCT – 3′; arginase I ( Arg I ) sense: 5′ - GTT CCC AGA TGT ACC AGG ATT C – 3′ , Arg I anti-sense: 5′ - CGA TGT CTT TGG CAG ATA TGC – 3′ . Murine Arg I primer was tested for cross-hybridization with the parasite ortholog , and no cross-hybridization or amplification occurred in the parasite material . Data were expressed as means ± standard errors of the means ( SEM ) . Student's t test and ANOVA followed by Tukey's multiple comparison test were used to analyze the statistical significance of the observed differences between the mice infected with LTCP393 ( R ) or LTCP15171 ( S ) when samples had a normal distribution , and Mann-Whitney tests were used when the distribution was not normal . Differences were considered significant when P<0 . 05 . All analyses were performed using Prism software version 5 . 0 ( GraphPad San Diego , CA , USA ) .
To examine the kinetics of lesion development , 5th day of culture stationary phase parasites , with similar amount of metacyclics , were inoculated in the right ear dermis of BALB/c mice , and lesions were measured weekly using a caliper for 15 weeks following infection . Mice controlled the lesions caused by both isolates , but kinetics of lesion development was different between the groups and the severity of lesions was higher in LTCP393 ( R ) -infected mice . Fifteen weeks after infection , the lesions in mice inoculated with LTCP15171 ( S ) were completely healed , while small lesions still remained in mice inoculated with LTCP393 ( R ) ( Fig . 1A ) . On the third week after infection , the lesions in animals inoculated with LTCP15171 ( S ) were approximately twice the size of those in mice inoculated with LTCP393 ( R ) ( Fig . 1A and 2B ) . Five weeks post-infection , the lesions in the animals inoculated with LTCP15171 ( S ) reached their peak size and began to decrease in thickness , while the lesions in mice infected with LTCP393 ( R ) continued to develop and reached their peak size seven weeks post-infection . At this time point , the lesions in mice inoculated with LTCP393 ( R ) were approximately three times thicker than those in mice inoculated with LTCP15171 ( S ) and two times thicker than the peak size of the lesions in mice inoculated with LTCP15171 ( S ) , which was reached five weeks post-infection ( Fig . 1A ) . Also in this time point , the lesions were ulcerated in the mice inoculated with LTCP393 ( R ) ( Fig . 1B , arrow ) . Eight weeks post-infection , the lesions in mice inoculated with the LTCP393 ( R ) isolate began to regress , but these lesions remained significantly larger than those in mice inoculated with LTCP15171 ( S ) , and small lesions were present in mice inoculated with LTCP393 ( R ) after 12 weeks ( Fig . 1B , arrowhead ) . We next determined the infective capacity of the stationary phase promastigotes of both isolates , by quantifying the amount of amastigotes inside mouse peritoneal macrophages , after 6 h incubation . LTCP393 ( R ) and LTCP15171 ( S ) parasites infected 57 . 14%±9 . 787 and 58 . 38%±5 . 558 the cells , respectively . Moreover , the amount of intracellular parasites was similar ( 347 . 7±47 . 75 versus 384 . 7±20 . 30 amastigotes of the isolates LTCP393 ( R ) and LTCP15171 ( S ) , respectively , per 100 macrophages , using 5 parasites/cell to the infection ) . Similar data was found using different ratio of parasites/cell . We concluded that both isolates infect macrophages with the same efficacy . The number of parasites in the infected ears and in the draining lymph nodes at specific time points post-infection was determined using a previously described limiting dilution assay [36] . The parasite loads in the earlier phases of the experimental infection were analyzed ( 1 , 3 , 7 and 14 days post-infection ) in ears and draining lymph nodes . Therefore , no difference was found between the groups , showing that in vivo , as observed previously in vitro , both parasite isolates presented the same infectivity rate with the chosen inocula . At time points three and five weeks post-infection , also no differences in parasite load were observed between the groups . Seven weeks post-infection , a significantly higher number of parasites was present in the ears of mice inoculated with LTCP393 ( R ) than those of mice inoculated with LTCP15171 ( S ) ( 2 . 01×108±1 . 63×108 and 1 . 73×104±1 . 1×104 , respectively; p = 0 . 0079 ) . Twelve weeks after infection , no parasites were detected in the ears of animals inoculated with the susceptible isolate , but the ears of mice inoculated with the resistant isolate contained an average of 1 . 79×107±1 . 08×107 parasites ( Fig . 2A ) . Fewer parasites were detected in the draining lymph nodes than in the ears of infected mice . Three weeks post-infection , the number of parasites detected in the lymph nodes of mice inoculated with LTCP15171 ( S ) was significantly higher than that in lymph nodes of mice inoculated with LTCP393 ( R ) ( 103 . 7±45 . 94 and 0 , respectively ) . Five weeks after infection , no significant difference was observed between the number of parasites in lymph nodes from mice inoculated with LTCP393 ( R ) ( 70 . 53±43 . 47 ) and that in lymph nodes from mice inoculated with LTCP15171 ( S ) ( 61 . 32±27 . 55 ) . Despite the regression of lesions in the ears at subsequent time points post-infection , the parasite loads in the lymph nodes increased significantly ( LTCP393 ( R ) , p = 0 . 0112 and p = 0 . 0398 at seven and 12 weeks post-infection , respectively , compared to five weeks post-infection; LTCP15171 ( S ) , p = 0 . 0119 and p = 0 . 0080 at seven and 12 weeks post-infection , respectively , compared to five weeks post-infection ) ; however , no significant differences in parasite loads were detected between mice inoculated with the resistant or susceptible isolates ( Fig . 2B ) . We next characterized the inflammatory infiltrates present in the lesions of mice experimentally infected with each isolate using flow cytometry . There were no differences in the cell populations studied three weeks after infection ( Fig . 3A ) . Five weeks after infection , the numbers of most cell types , and especially of the CD4+ T lymphocytes and neutrophils ( GR-1+ ) , increased significantly in infiltrates in mice infected with either isolate; however , CD8+ T lymphocytes , dendritic cells ( CD11b+CD11c+ ) and macrophages ( CD11b+CD11c− ) were not significantly increased in infiltrates in LTCP15171 ( S ) -inoculated mice . No differences in the cell populations were detected between mice infected with resistant or susceptible isolates at this time point ( Fig . 3B ) . Seven weeks after infection , more B lymphocytes were present in lesions of animals inoculated with LTCP393 ( R ) than in those of animals inoculated with LTCP15171 ( S ) ( 102 , 300±27 , 840 and 18 , 440±3 , 189 , respectively; p = 0 . 0201 ) . Similarly , the lesions of mice experimentally infected with LTCP393 ( R ) contained more CD4+ T lymphocytes than those of mice experimentally infected with LTCP15171 ( S ) ( 1 , 761 , 000±257 , 100 vs . 990 , 900±124 , 200; p = 0 . 0279 ) ( Fig . 3C ) . No differences were observed in the number of CD8+ T lymphocytes or CD4+CD25+ T cells in the lesions of mice inoculated with LTCP393 ( R ) or LTCP15171 ( S ) , and CD4+CD25+ T cells expressed similar levels of FoxP3 , CD103 , CTLA-4 and GITR in both groups ( Fig . 3E ) . The numbers of neutrophils , dendritic cells ( CD11b+CD11c+ ) and macrophages were higher in animals inoculated with LTCP393 ( R ) than in animals inoculated with LTCP15171 ( S ) ( p = 0 . 0086 , 0 . 0339 and 0 . 0006 , respectively ) ; between weeks five and seven , these cell types decreased in number more intensely in lesions from mice inoculated with LTCP15171 ( S ) than in lesions from mice inoculated with LTCP393 ( R ) ( Fig . 3C ) . Twelve weeks post-infection , the numbers of all cell populations in the animals inoculated with LTCP393 ( R ) remained similar to those observed seven weeks post-infection while the numbers of most cell populations decreased in the animals inoculated with LTCP15171 ( S ) . The numbers of B lymphocytes , CD4+ and CD8+ T lymphocytes , CD4+CD25+ T cells , and neutrophils were thus significantly higher in animals inoculated with LTCP393 ( R ) than in LTCP15171 ( S ) -inoculated mice . No statistical differences in the numbers of dendritic cells and macrophages were observed between mice inoculated with resistant or susceptible isolates ( Fig . 3D ) . Furthermore , no significant difference was observed in the frequency ( % ) of FoxP3 , CD103 , GITR and CTLA-4 positive cells among CD4+CD25+ cells between mice inoculated with the resistant or susceptible isolates ( Fig . 3F ) . We have previously observed that the differences between the disease manifestations in mice inoculated with LTCP393 ( R ) or LTCP15171 ( S ) became evident at later time points of the experimental infection . Therefore , immune responses induced by parasites were examined by measuring the cytokines produced by lymph node cells isolated 3–12 weeks post-infection and stimulated in vitro with live stationary phase promastigotes . The average production of IFN-γ was higher in lymph node cells harvested from animals inoculated with LTCP15171 ( S ) three weeks post-infection , than in those from animals inoculated with LTCP393 ( R ) ; however , this difference was not statistically significant . Lymph node cells harvested from animals inoculated with either isolate 5 , 7 or 12 weeks post-infection produced more than 5 ng/ml IFN-γ when stimulated with live parasites . No statistical significant differences were observed between lymph node cells from mice inoculated with resistant or susceptible isolates at these time points ( Fig . 4A ) . Low levels of IL-4 were produced by lymph node cells isolated from animals inoculated with LTCP15171 ( S ) three weeks post-infection and stimulated with live parasites , but no IL-4 was detected in the lymph node cells from animals inoculated with LTCP393 ( R ) at this time point . In contrast , five weeks post-infection , lymph node cells isolated from animals inoculated with LTCP393 ( R ) and challenged with live parasites produced significantly higher levels of IL-4 ( 3 . 987±0 . 1914 ng/ml ) than those isolated from animals inoculated with LTCP15171 ( S ) ( 0 . 2122±0 . 0924 ng/ml; p<0 . 0001 ) . The production of IL-4 by lymph node cells from mice inoculated with the resistant isolate seven and 12 weeks post-infection was lower than that observed at five weeks post-infection ( 1 . 406±0 . 2616 and 0 . 6536±0 . 08217 ng/ml , respectively ) , but still significantly higher than that observed in animals inoculated with the susceptible parasites ( 0 . 135±0 . 04826 and 0 . 1071±0 . 04127 ng/ml; p = 0 . 0088 and 0 . 0040 , respectively ) ( Fig . 4B ) . Upon stimulation with parasites , IL-10 was detected at low levels in lymph node cells isolated from mice inoculated with either isolate three and five weeks post-infection . Seven weeks after infection , however , IL-10 production increased in cells isolated from mice inoculated with LTCP393 ( R ) , and these cells produced higher levels of IL-10 than those isolated from mice inoculated with the susceptible parasite ( 0 . 4911±0 . 07603 and 0 . 1925±0 . 02022 ng/ml , respectively; p = 0 . 028 ) . Twelve weeks post-infection , IL-10 production increased in lymph node cells from mice inoculated with either isolate , but cells from mice inoculated with LTCP393 ( R ) continued to produce higher levels of this cytokine than those from mice inoculated with LTCP15171 ( S ) ( 2 . 190±0 . 1237 ng/ml and 1 . 488±0 . 1190 ng/ml , respectively; p = 0 . 0064 ) ( Fig . 4C ) . TGF-β production did not significantly differ between stimulated and unstimulated cultures or between cells isolated from mice inoculated with LTCP393 ( R ) and those from mice inoculated with LTCP15171 ( S ) at the time points analyzed . The production of this cytokine throughout the experimental infection remained similar to non-infected mice in all time points analyzed in both groups ( Fig . 4D ) . The expression patterns of cytokines and the enzymes iNOS and Arg I at the sites of experimental infection , were examined by processing the ears of the experimentally infected mice for mRNA extraction and quantifying mRNA by real time PCR at the same time points at which cytokine production was measured in lymph node cells . Three weeks post-infection , there was no significant difference in the expression of IFN-γ , IL-4 , IL-10 , TGF-β , TNF-α , iNOS and Arg I mRNA at experimental infection sites between the mice inoculated with the resistant or susceptible parasites ( Fig . 5A–G ) . Similar to our observations in lymph node cultures , five weeks post-infection , the ears of LTCP393 ( R ) -inoculated mice expressed higher levels of IL-4 mRNA than those of LTCP15171 ( S ) -inoculated mice ( 372 . 6±168 . 1 and 43 . 51±20 . 81; p = 0 . 0398 ) ( Fig . 5B ) , and no significant differences were found in the expression of IFN-γ mRNA ( Fig . 5A ) . No differences were found in IL-10 and TGF-β expression and also in the enzymes iNOS and Arg I expression , at five weeks post infection ( Fig . 5C–G ) . At this same time point , TNF-α was expressed at higher levels in the ears of animals inoculated with LTCP393 ( R ) than those of animals inoculated with LTCP15171 ( S ) ( Fig . 5E ) . Seven weeks post-infection , we observed approximately 20-fold increase in IL-4 mRNA expression in mice inoculated with the resistant isolate compared to that in mice inoculated with the susceptible isolate ( 419 . 9±76 . 33 and 16 . 23±5 . 164 , respectively; p = 0 . 0019 ) ( Fig . 5B ) . Also , Arg I expression was significantly higher in ears of mice experimentally infected with LTCP393 ( R ) compared to mice infected with LTCP15171 ( S ) ( 21 . 41±5 . 615 and 3 . 282±1 . 338 , respectively; p = 0 . 02 ) ( Fig . 5G ) . No statistical difference was detected in IFN-γ and iNOS expression between these groups , although a clear tendency of higher expression of the enzyme was observed in ears of mice infected with the resistant isolate ( Fig . 5A and F ) . Albeit IL-10 mRNA expression increased in the ears of animals infected with the LTCP393 ( R ) isolate at seven weeks post infection , it was not significantly different from that detected in the ears of LTCP15171 ( S ) -inoculated mice . ( Fig . 5D ) . No differences were detected in the expression of TNF-α and TGF-β seven weeks after infection ( Fig . 5B and E ) . Twelve weeks post-infection , the expression levels of IFN-γ , IL-4 , IL-10 , TNF-α and Arg I were similar in both groups . The only differences observed were in TGF-β and iNOS expression; the cytokine was expressed at higher levels in ears of mice inoculated with LTCP15171 ( S ) while the enzyme was expressed at higher levels in ears of mice inoculated with LTCP393 ( R ) ( Fig . 5E and F ) . Because the most striking difference between the two groups was the level of IL-4 production , it was reasonable to associate IL-4 production with the increased susceptibility to experimental infection with the resistant isolate . To test this hypothesis , BALB/c mice were treated with anti-IL-4 mAb before the inoculation with 1×106 stationary phase promastigotes of L . braziliensis LTCP393 ( R ) or LTCP15171 ( S ) isolates . The mice were then treated with anti-IL-4 mAb twice a week for seven weeks after infection . During the first six weeks after infection , no differences were detected between mice inoculated with LTCP393 ( R ) and treated with anti-IL-4 mAb and those inoculated with LTCP393 ( R ) and treated with control IgG . Seven weeks after experimental infection , however , lesions in animals treated only with normal IgG increased in thickness while those in animals treated with anti-IL-4 mAb began to decrease significantly in thickness ( 0 . 8233±0 . 151 and 0 . 35±0 . 02 , respectively; p = 0 . 036 ) ( Fig . 6A ) . In mice inoculated with LTCP15171 ( S ) , no difference was observed between the control animals and those treated with the anti-IL-4 mAb at any time point following inoculation of parasites ( Fig . 6B ) . Moreover , the depletion of IL-4 in animals inoculated with LTCP393 ( R ) resulted in lesion development similar to that seen in mice inoculated with LTCP15171 ( S ) , in which little production of IL-4 was detected . Quantification of the parasite load showed that anti-IL-4 mAb treatment resulted in not only decreased lesion thickness , but also a significant decrease in the parasite loads ( ∼100× ) in the ears of LTCP393 ( R ) -inoculated animals . As observed in the lesion measurements , parasite load in anti-IL-4 treated LTCP393 ( R ) -inoculated mice ears was similar to that observed in LTCP15171 ( S ) -inoculated mice ( Fig . 7A ) . Although the average of parasite loads in lymph nodes of anti-IL-4 mAb treated LTCP393 ( R ) -inoculated mice were lower compared to rat IgG treated counterparts , no significant difference was observed ( Fig . 7B ) . In mice inoculated with LTCP15171 ( S ) , no difference in lymph node parasite loads were observed between anti-IL-4 mAb and rat IgG treated animals ( Fig . 7B ) .
In this work , we describe an experimental model for the evaluation of the immunomodulatory effects of different isolates of L . braziliensis that result in distinct disease manifestations . Using this model , we show that immunomodulation by IL-4 contributes to the different disease outcomes observed upon infection with different L . braziliensis isolates . Upon inoculation of BALB/c mice with L . braziliensis isolates LTCP393 ( R ) and LTCP15171 ( S ) we observed clearly two distinct patterns of disease development in the mice . The increased severity of the disease caused in mice by the resistant parasites inoculation was due to both , higher inflammation and increased parasite burdens that could be observed only at later times post-infection . In models of L . major infection , it is clearly established that the type of inflammatory infiltrate may interfere with the infection outcome . CD4+ and CD8+ T cells are important in inducing protection via Th1 responses [5]–[9] , [37] , while neutrophils may act as a vehicle for silent entry of L . major into macrophages , favoring parasite survival [38] , [39] . However , due to few studies concerning about this issue , the importance of each cellular population in L . braziliensis infection remains unknown . Therefore , we analyzed the possible roles of cell populations in inflammatory infiltrate for L . braziliensis disease outcomes , via inoculation of LTCP393 ( R ) or LTCP15171 ( S ) L . braziliensis isolates . Even presenting different disease manifestations , in both groups , the inflammatory infiltrates were composed mainly by CD4+ T lymphocytes and neutrophils . We found differences in the number of cells in some infection time points , however , the frequency of each cell population thus varied following the same pattern in both groups . We observed that all the cell populations analyzed increased or decreased following the same pattern in both groups , according to parasite load . Additionally , no specific cell population was altered differentially between the groups throughout infection , showing no contribution of any specific cell population in the disease outcomes observed in our model . However , our results provide the first evidence that CD4+CD25+FoxP3+ regulatory T cells ( Treg ) , which have the capacity to suppress Th1 effector responses and favor Leishmania survival [40]–[45] , are recruited to the infection sites in L . braziliensis-infected mice . The differences in the disease outcomes in our experimental model could be due to differences in the induction of specific T cell responses to the parasites of the distinct isolates . For this purpose , we analyzed the production of Th1 cytokines , related to protection to cutaneous leishmaniasis [7] , [9] , [16] , [17] , [46]–[49] , and Th2 and Treg cytokines , related to susceptibility in murine models of cutaneous leishmaniasis [50] , [11] , [13] , the two last of which have their role poorly characterized in L . braziliensis infection . In our model , the difference in disease outcomes was not due to differential induction of Th1 responses by the isolates , since IFN-γ production in both groups was similar throughout the infection . Also , IL-10 and TGF-β were expressed similarly in the lesions throughout the critical points of infection , thus discarding the possibility of any strong interference of differential Treg activity in the lesion site could be determinant for the disease outcomes . Therefore , during the later phases of infection , IL-10 production increased in lymph node cultures from mice infected with both isolates , mainly in LTCP393 ( R ) -inoculated animals . Since this increase occurred only when the lesions were healing or healed , we believe that it may reflect the homeostatic control of the pro-inflammatory responses , especially in LTCP393 ( R ) -inoculated mice , that presented a more robust inflammatory response at this time point . On the other hand , increased IL-10 production may also be responsible for the increase in the number of parasites in the lymph nodes during the later phases of infection . Moura et al . [32] also demonstrated an increase in the number of parasites in lymph nodes of L . braziliensis-infected BALB/c mice in which lesions were cured . It has been shown that IL-10 suppresses effector responses , preventing sterile cure which confers immunity to re-infection with L . major [12] . The ability of IL-10 to control parasite numbers may also be important in maintaining immunity to L . braziliensis; however , it could also represent an evolutionary success of L . braziliensis parasites in maintaining a transmissible reservoir in the host without causing disease . TNF-α expression was also characterized , but this is a cytokine that suffers posttranscriptional regulation , through binding of proteins to a region called adenine urenine rich element ( ARE ) , present in TNF-α mRNA transcripts [51] , and a better characterization is needed to establish its role . However , in our case , although TNF-α is related to protection against infection [9] , [48] , [16] , our finding of increased TNF-α mRNA expression in lesions of LTCP393 ( R ) -inoculated mice five weeks post-infection , may be associated to Th2 responses . In the skin , keratinocytes may produce TSLP , that up-regulates OX40L expression in dendritic cells , which in turn drives differentiation of Th2 cells that produce IL-4 , IL-5 , IL-13 and TNF-α [52]–[54] . The role of cytokines produced by keratinocytes in L . braziliensis infection is an important issue that must be investigated in future works . In fact , the most striking difference we found between mice infected with the resistant or susceptible isolates was in Th2 responses . IL-4 , lesion size and parasite load dramatically increased concomitantly in LTCP393 ( R ) -inoculated mice compared to LTCP15171 ( S ) -inoculated animals . Also , IL-4 neutralization dropped lesion size and parasite loads in lesions of LTCP393 ( R ) -inoculated mice to the same levels found in LTCP15171 ( S ) -inoculated animals , while had no effect in the last ones . This showed that induction of IL-4 production by LTCP393 ( R ) was responsible for the differences in lesion development , susceptibility to parasite replication and disease persistence between mice infected with the different isolates . It would be expected that IL-4 neutralization led to higher IFN-γ production , increasing parasite killing . Therefore , we did not observe any increase in IFN-γ production upon IL-4 neutralization ( data not shown ) . In L . major infection , different grades of resistance to infection could be seen in mice that produce similar levels of IFN-γ but varying levels of IL-4 , suggesting that the magnitude of the IL-4 response determines the severity of disease more than IFN-γ [55] . Also in patients infected with L . braziliensis , the concomitant production of IL-4 together with IFN-γ is observed , and it has been proposed that it favors parasite survival and impairs spontaneous lesion healing . [56] . Likewise , even though both isolates induce similar levels of IFN-γ production , experimental infection with the LTCP393 ( R ) L . braziliensis isolate induces production of higher levels of IL-4 , which favors the survival of this parasite and increases disease severity . IL-4 is a potent activator of the enzyme arginase I ( Arg I ) , which competes with iNOS for L-arginine that is used for the synthesis of NO , in macrophages . By consuming L-arginine , Arg I decreases the availability of this substrate for iNOS , resulting in diminished production of NO , and inhibition of inflammation [57] . Therefore , there is not a reciprocal inhibition between the enzymes , and iNOS and Arg I can be concomitantly expressed in macrophages at the mRNA and protein levels upon stimulation [58] , [59] . Especially in a microenvironment where Th1 and Th2 responses are taking place at the same time , that is the case in LTCP393 ( R ) -inoculated mice , this feature occurs , as we showed . Despite expressing the same levels of iNOS than LTCP15171 ( S ) -inoculated mice , the increased Arg I expression in LTCP393 ( R ) -inoculated mice suggests that increased IL-4 production in these animals leads to increases in Arg I in the macrophages , resulting in decreased production of NO , favoring parasite survival , which results in a more severe disease . In lymph nodes , however , IL-4 seems not to interfere in the parasite elimination as in the lesions , since high levels of IL-4 in LTCP393 ( R ) -infected mice did not result in increased parasite load ( Fig . 2 ) . Similarly , IL-4 depletion did not result in significant parasite reduction in the organ . These data show that the control of parasite growth and/or parasite elimination in the periphery is distinct than that of lymphoid organs . We do not know the mechanisms by which this feature occurs , although it is possible that IL-10 could be involved . Also , APCs as dendritic cells and macrophages , are able to kill the parasites in the dependence of the microenvironment [60] , [61] . In peripheral sites , these cells act mainly as “killer cells” , while , upon migration to lymphoid organs , they change their profile to “antigen presenting cells” [62]–[64] . In conclusion , our model allowed us to suggest that in addition to intrinsic resistance to drugs and NO , the immunomodulation towards Th2 response induced by certain parasite strains may account for more severe lesions , increased parasite burdens , and increased length of disease . We may ultimately propose that even in the presence of IFN-γ , IL-4 production increases arginase I expression and impairs the parasite killing in lesion site . This characteristic of some parasite isolates can be considered a susceptibility factor for L . braziliensis infection . The models of infection described here will be useful in the identification of immunological targets to control L . braziliensis infection .
|
Leishmaniasis is a neglected disease that affects more than 12 million people worldwide . In Brazil , the cutaneous disease is more prevalent with about 28 , 000 new cases reported each year , and L . braziliensis is the main causative agent . The interesting data about the infection with this parasite is the wide variety of clinical manifestations that ranges from single ulcerated lesions to mucocutaneous and disseminated disease . However , experimental models to study the infection with this parasite are difficult to develop due to high resistance of most mouse strains to the infection , and the mechanisms underlying the distinct manifestations remain poorly understood . Here , the authors use a mouse experimental model of infection with different L . braziliensis isolates , known to induce diseases with distinct severity in the human hosts , to elucidate immune mechanisms that may be involved in the different manifestations . They showed that distinct parasite isolates may modulate host response , and increased IL-4 production and Arg I expression was related to more severe disease , resulting in longer length of disease with larger lesions and reduced parasite clearance . These findings may be useful in the identification of immunological targets to control L . braziliensis infection and potential clinical markers of disease progression .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"infectious",
"diseases",
"immunology",
"infectious",
"diseases/protozoal",
"infections",
"immunology/immunity",
"to",
"infections"
] |
2011
|
BALB/c Mice Infected with Antimony Treatment Refractory Isolate of Leishmania braziliensis Present Severe Lesions due to IL-4 Production
|
Several studies highlighted the relevance of extrinsic noise in shaping cell decision making and differentiation in molecular networks . Bimodal distributions of gene expression levels provide experimental evidence of phenotypic differentiation , where the modes of the distribution often correspond to different physiological states of the system . We theoretically address the presence of bimodal phenotypes in the context of microRNA ( miRNA ) -mediated regulation . MiRNAs are small noncoding RNA molecules that downregulate the expression of their target mRNAs . The nature of this interaction is titrative and induces a threshold effect: below a given target transcription rate almost no mRNAs are free and available for translation . We investigate the effect of extrinsic noise on the system by introducing a fluctuating miRNA-transcription rate . We find that the presence of extrinsic noise favours the presence of bimodal target distributions which can be observed for a wider range of parameters compared to the case with intrinsic noise only and for lower miRNA-target interaction strength . Our results suggest that combining threshold-inducing interactions with extrinsic noise provides a simple and robust mechanism for obtaining bimodal populations without requiring fine tuning . Furthermore , we characterise the protein distribution’s dependence on protein half-life .
Gene-expression data displays bimodal distributions in several systems ranging from cancer to immune cells [1 , 2] . The two peaks of the distribution are usually associated with different physiological states of the system , be that different stem-cell fates or different disease states or cancer subtypes [3–6] . From a theoretical point of view , a common belief is that bimodality is directly related to bistability , i . e . , to multiple steady states appearing in the absence of noise . However , as shown in [7] and reviewed in [8] , bimodality in some biological systems is solely due to stochastic effects . Environmental fluctuations , usually referred to as extrinsic noise [9 , 10] , can be a source of noise in molecular networks . Together with intrinsic fluctuations due to the probabilistic nature of chemical reactions , extrinsic noise shapes gene expression and may in principle drive cell differentiation . In the past , both theoretical [11 , 12] and in vitro [13] studies , have highlighted the possibility that microRNAs ( miRNAs ) , in particular stoichiometric conditions , may induce bimodality in the expression of their targets simply because of stochastic effects related to their specific titrative interactions and not because of a bistable system . This is similar to the action of miRNAs’ bacterial counterpart [14] . MiRNAs are small molecules of non-coding RNA , found in eukaryotes to act as post-transcriptional regulators . Although they were found in several different eukaryotic kingdoms , their role is known to be vital in multicellular organisms . They perform this function by recognising mRNA targets through Watson-Crick base pairing . Once bound to the target , they prevent its translation and can enhance its instability by degrading it . Interestingly , different levels of miRNA-target interactions can be achieved by different numbers of miRNAs [15–17] . Theoretical predictions [12] together with in vitro single-cell experiments [13] suggested that bimodality in the expression levels of miRNA targets can be achieved with a high miRNA-target interaction strength . In terms of genetic sequences , this would imply a high specificity between target and miRNA , and therefore a high number of complementary binding sites ( bs ) per target . As long as a miRNA molecule is bound to the target , it cannot be translated . It is then possible to define a threshold for the mRNA transcription rate such that below the threshold many of the mRNA target molecules are bound to miRNAs and above the threshold there are molecules of mRNA free for translation [12 , 18 , 19] . The titrative interaction between miRNA and mRNA thus induces a specific dependence between the amounts of the two species in the system . Such a threshold mechanism involves two regimes that can be explored by varying , e . g . , the mRNA transcription rate . The two regimes are characterized by a low and a high expression of mRNA . In more detail , the threshold behaviour of mRNA expression crucially depends on three main parameters: the transcription rate of miRNA , the transcription rate of mRNA , and the interaction strength between miRNA and mRNA , which is related to the affinity between them and to the number of miRNA binding sites on the target . If the amount of miRNA is high compared to that of mRNA , mRNA will be sequestered and repressed . Conversely , if the quantity of miRNA is low compared to the mRNA , mRNA will be expressed and free for translation . The two regimes can be explored by varying the parameters . For instance , starting with very few free molecules of mRNA and gradually increasing the mRNA transcription rate moves the system from the repressed to the expressed state as shown in Fig 1E . The transition from repressed to expressed upon increasing the mRNA transcription rate occurs in a non-linear fashion when the amount of mRNA molecules roughly matches that of miRNAs . The steepness of the threshold at this point depends on the interaction strength between miRNA and mRNA . The presence of the threshold was first detected in [18] where the mechanism described above was shown and explained in terms of a titrative interaction . The dependence of the threshold location on miRNA expression was investigated in two different papers , namely in [13] for the highly expressed miRNA regime and in [20] for the weakly expressed miRNA regime . Close to the threshold the number of both free miRNAs and targets is small . Their fluctuations are highly coupled by the non-linear interaction between the two and a small fluctuation in their amounts may lead the system from the “bound” to the “unbound” state [12 , 13 , 19] . As anticipated , if the interaction strength between miRNA and target is high , then the transition from the bound to the unbound state is sharp . Close to the threshold , a subset of the targets will be bound to the miRNA and a subset will be unbound . This is because of the intrinsic fluctuations in the amount of both miRNA and target . Picturing this in terms of the target distribution would lead to a bimodal distribution whose two modes are associated with the bound and unbound state . It is worth underlining that this kind of bimodality is due to the presence of noise and not to peculiar molecular mechanisms introducing multiple deterministic stable states in the system . MiRNAs are predicted to regulate more than 60% of our genome through a combinatorial action: every single miRNA can regulate several targets and one target can be regulated by different miRNAs [17 , 21] . The variety of targets they regulate is so wide and important for different signalling pathways or developmental stages [22 , 23] that the alteration of their expression levels is thought to contribute to tumour development and metastatisation [24–27] . As discussed in [28] , a role for miRNA in generating expression variability can be remarkable . If miRNA activity increases the cell-to-cell variability of pivotal pluripotency factors , ( consistent with observations in [29] ) , and in turn of pluripotency networks , then miRNA expression variability can provide an efficient mechanism for generating transitions between cell states . Although extrinsic noise may influence gene expression and regulation at different levels , large variability across a cell population seems to be dominated by the population dynamics [30] . Even a monoclonal population has cells in different phases of their cell cycle because of growth and divisions . It is nowadays well established that multiple cell-cycle regulators are controlled by miRNAs , whose regulation could be in turn cell-cycle dependent [31–34] . The expression level of miRNAs may thus change with cell-cycle progression , and there are indeed miRNAs differentially expressed according to the particular phase of the cell cycle [35] . As a consequence , in a population of cells heterogeneous with respect to the cell cycle , such as non-quiescent cancer cells , the amount of miRNAs can strongly fluctuate from cell to cell . This introduces an extra source of noise in the system besides the intrinsic stochasticity of chemical reactions involving gene transcription , translation and regulation . The aim of this work is to understand , with the aid of analytics and numerics , how extrinsic noise on miRNA expression can induce bimodality on the miRNA targets . Of course , such bimodality is not omnipresent but restricted to particular stoichiometric conditions and levels of noise . We show how a distribution of miRNA transcription rates reshapes the threshold between miRNA and target and defines a wider region of bimodality compared to case without extrinsic noise . Such a bimodal distribution can be seen at a “population level” , since the amount of miRNA is heterogenous throughout the different cells . This outcome is significantly different from previous results where differential phenotypic expression is induced by the strong coupling between miRNA and its target at the “single-cell level” . We also show that , if the miRNA target is protein coding , the protein half-life can alter the protein distribution . With respect to the shape of the mRNA distribution , an increased protein half-life leads to a narrowing of the protein distribution around its mean . This may promote or suppress bimodality , suggesting that bimodal distributions at the level of mRNA may still correspond to a specific single phenotype at the protein level . Conversely , repressed heavy tailed mRNA distributions may give rise to bimodal protein distributions . Finally , given the existence of multiple targets competing for one type of miRNA , we ask whether these properties can be maintained in a more complex circuit made of two competing endogenous RNAs ( ceRNAs ) and one miRNA [36] . The different target genes indeed act as sponges for the miRNA molecules and may sequester them from the environment . As a result , the overexpression or underexpression of one of the targets can lead respectively to an increase or decrease in the level of expression of the other competitors . The intensity of such cross regulation depends on the distance from the threshold of quasi equimolarity between miRNAs and targets [12 , 19] . This suggests that , if one target has a bimodal distribution , such bimodality may be influenced by the expression levels of the other miRNA competitors .
Models of microRNA-mediated circuits have been the subject of several recent studies [18 , 37–41] . Here we will describe one of the simplest ways of accounting for microRNA-driven inhibition , depicted in Fig 1A . The molecular species involved in this circuit are miRNAs ( S ) , target mRNAs ( R ) and proteins ( P ) , resulting from the translation of the target mRNA . In the following , we shall assume miRNAs and mRNAs are transcribed from independent genes . For simplicity we neglect all the intermediate reactions leading to the synthesis of mRNAs , therefore assuming they are produced at constant rate kR . For the miRNA , we consider it to be transcribed with a constant rate kS which we let fluctuate between different cells to probe the effects of extrinsic noise on the system . This approach is equivalent to having kS slowly fluctuating in time with respect to the systems’ reactions while different cells are stochastically unsynchronised . We also remark that , in the opposite limit of very rapid extrinsic fluctuations , these variations average out and the system effectively behaves as if subject to intrinsic noise only . A more detailed analysis of the interplay between the time scales of extrinsic fluctuations and those of the system is reported in a devoted section of the Results . MiRNAs and mRNAs can also be degraded by the action of specialised enzymes . Here we assume these reactions are governed by mass-action laws with rates gS and gR . The associated molecular reactions read: ∅⇌gRkRR , ∅⇌gSkSS . ( 1 ) MiRNAs act as post-transcriptional regulatory elements , by binding the target mRNAs in a complex that can be subsequently degraded . Such interactions between miRNAs and mRNAs are quantified by the effective parameter g , which takes into account the strength of the miRNA-target coupling: from a biochemical point of view , it depends on the affinity between the two molecular species and on the number of miRNA binding sites dedicated to a specific target [18] . The formation of the miRNA-mRNA complex reads: R + S ⟶ g R S . ( 2 ) While the mRNAs are always degraded due to the titrative interaction , the miRNAs can be recycled with probability 1 − α in the following way: R S ⟶ 1 - α S . ( 3 ) Whenever the mRNAs are not bound to miRNAs , they can be translated into proteins with translation rate kP and , as assumed for the other molecular species , proteins can also be degraded with rate gP , i . e . : R ⟶ k P R + P , P ⟶ g P ∅ . ( 4 ) From now on , we define as “intrinsic noise” the fluctuations due to the stochasticity of the chemical reactions with constant rates ( Fig 1A ) and as “extrinsic noise” those due to the fluctuating miRNA transcription rate ( see Fig 1C ) . The system can be described by the probability distribution P ( n S , n R , n P , t | K ) of observing nS molecules of miRNA , nR molecules of mRNA and nP proteins at time t given a set of parameters K = { k R , k S , k P , g R , g S , g P , g , α } . This probability distribution follows the same master equation presented in [12] that can be either solved numerically or at the steady-state with some approximations . If the parameters fluctuate , this must be taken into account in order to obtain the full distribution at steady state Pss ( nS , nR , nP ) . This can be achieved by using the law of total probability [42] , which states that P ( n S , n R , n P ) = ∫ P ( K ) P ( n S , n R , n P | K ) d K . As our aim is to test the effects of a fluctuating miRNA transcription rate , we shall assume this to be the only parameter drawn from a probability distribution , specifically a Gaussian centred around 〈kS〉 with variance σ k S 2 and defined only for positive values of kS . As previously mentioned , this is equivalent to assuming the extrinsic noise fluctuates slowly compared to the typical time scale of any other reaction in the system . To obtain the steady-state distribution P ( nS , nR , nP|kS ) conditional on a specific miRNA transcription rate we could choose different approximation methods . Pivotal examples are the Van Kampen [43] and the Gaussian approximations [12] . In the following we focused on the former one , leaving to the Supporting Information a comparison between the two methods ( see S1 Fig ) . We therefore performed a system-size expansion , thus assuming the system distribution at fixed parameters to be Gaussian . The marginal distribution P ( nS , nR , nP ) was then found by using the law of total probability , i . e . , by performing a weighted average over all possible values of kS . Finally , we applied the same approach when considering two targets interacting with the same miRNA ( Fig 5A ) . In this case the conditional distribution is P ( nS , nR1 , nR2 , nP1 , nP2|kS ) from which one can obtain the full distribution by integrating over the values of the miRNA transcription rate .
The understanding of bimodal distributions is usually related to cell-fate determination and differentiation . These mechanisms are at the basis of organism development and mis-development . It is therefore important to address the question of what might be the underlying molecular mechanisms allowing cell diversity and variability . Given the strong involvement of miRNAs in developmental decisions , we focus here on the miRNA network represented in Fig 1A , at the single-cell level . Previous works [18 , 36] showed that the binding and unbinding reactions between miRNA and target give rise to non-trivial threshold effects in quasi equimolar conditions between miRNA and target ( see Fig 1E ) where the threshold is defined , in terms of miRNA and mRNA transcription rates , as k S * = α k R * [18] . If the miRNA is transcribed at a rate above the threshold value , k S > k S * , the system is enriched in microRNA , which tends to bind most of the present mRNA and prevents its translation . In this regard , the system can be seen as below the threshold with respect to the target and we shall refer to it as in the “repressed state” . Above the threshold , the mean amount of free target increases linearly with its transcription rate . The scenario with free mRNA molecules will be denoted as the “unrepressed state” of the system . This threshold effect displaying a transition between the repressed and the unrepressed state gets more marked as the interaction strength between miRNA and targets increases . Close to the threshold value of the target transcription rate , due to the probabilistic nature of chemical reactions , the system will stochastically switch between the repressed and unrepressed state . Such stochastic switching is enough to give rise to bimodal target distributions which appear for a narrow range of the target transcription rate kR [12] ( Fig 1B ) . This bimodality characterises the single cell where the miRNA network is defined: every single cell can jump from the repressed to the expressed target state if the coupling constant with the miRNA is high enough . On the contrary , in presence of extrinsic noise , the miRNA transcription rate is not the same for every cell ( Fig 1C and 1D ) . Hereby , we model the extrinsic noise through a Gaussian-distributed miRNA transcription rate . To understand intuitively the consequences of this kind of extrinsic noise , let us consider the case of a miRNA transcription-rate distribution with fixed average 〈kS〉 . When the mRNA transcription rate ( kR ) is very low and the average miRNA transcription rate is much larger than the threshold value ( 〈kS〉 ≫ αkR ) , most of the drawn transcription rates kS will be larger than the threshold value . This would place the network in the parameter range where the targets are almost all bound to the miRNAs ( see Figs 2 and 3 ) . For larger kR , approaching the threshold , values of kS extracted from the left-tail will correspond to the case with some unbound targets . Below the threshold , as 〈kS〉 < αkR , the majority of the drawn kS will belong to the unrepressed state with the right tail of the distribution sampling from the all-bound region ( Figs 2A–2C and 3A and 3B ) . However , this scheme is the same as the previously mentioned population level scenario . The presence of rates above and below the threshold across the population can give rise to a bimodal distribution in the number of free targets ( Figs 2C and 2D and 3B and 3C ) . In particular , the higher the extrinsic noise , the larger the range of target transcription rates for which bimodality is present and the greater the separation between the two phenotypes ( bound and unbound targets ) , as depicted in Fig 4A . This implies that , in contrast to the case without extrinsic noise , it is no longer necessary to fine tune the transcription rates to obtain a bimodal distribution . Even for high values of kR , the fraction of randomly picked kS that results in the bound state is not negligible and forms a visible peak in the distribution ( Figs 2B–2D and 3A–3C ) . The bimodal distribution is in this case given by the superposition of unimodal distributions obtained for different kS and weighted by the probability P ( kS ) ( see S2 Fig in SI ) . Focusing on one particular value of the variance σ k S 2 and varying the target transcription rate , kR , we monitored the appearance of bimodal distributions . We ran Gillespie’s simulations from which we sampled the number of targets for the histograms shown in Figs 2 and 3 . By using system-size expansion and the law of total probability , we analytically obtained the target number distributions , shown in Figs 2 and 3 . The analytic approximation captures the behaviour of the system for the mean , the Coefficient of Variation and the probability distribution , as testified by the agreement with the simulations ( see Figs 2B–2D , 3A–3D and S3 in SI ) . In Fig 3 we showed that the results are maintained for a set of endogenously meaningful parameters , with low mean amount of free mRNAs . In this case , right because of the small amount of molecules involved , our approximation method is quantitatively less precise , though still keeping the qualitative shape of the distributions ( see SI for details ) . To dissect the properties of bimodal distributions , we first ran Gillespie’s algorithm simulating the network in Fig 1C for different target transcription rates , kR , and different variances of the Gaussian noise on the miRNA transcription rates σ k S 2 . Monitoring the appearance of bimodality , one can build up a phase diagram like the one shown in Fig 4A . In the absence of extrinsic noise but in the presence of intrinsic noise , bimodal distributions appear only for high coupling between the miRNA and the target and this region gets wider upon increasing the coupling constant g . Therefore the interaction strength between the miRNA and the target , g , affects the range of values of kR in which bimodality is present . Since in this case bimodality is a single-cell effect , only those cells having the target interacting strongly with the miRNA have a chance to experience the repressed and unrepressed state when kR is close to its threshold value . Adding some extrinsic noise relaxes the constraint on the interaction strength . Bimodality becomes a population effect , with some cells being locked in the repressed state ( by having large miRNA transcription rates kS ) and some others ( with smaller kS ) displaying free targets . Fig 4B shows how it is possible to have similar bimodal distributions either increasing the miRNA-target interaction strength ( blue histogram ) or increasing the extrinsic noise ( orange histogram ) with respect to a reference case with pure intrinsic noise and low miRNA-target interaction ( black line ) . The extrinsic noise and the miRNA-target interaction strength act at a similar level with respect to bimodality , where a higher extrinsic noise can compensate for a low interaction strength ( small number of miRNA binding sites on the target ) in order to obtain two differentially expressed phenotypes . The study so far led us to consider the possible importance of extrinsic noise in cell phenotypic variability . Given the existence of multiple miRNA-target networks , we now investigate how the results for the one-target case extend to the multiple-target one . Let us consider a minimal model with two targets , R1 and R2 , competing for the same miRNA , S ( Fig 5A ) . We start by investigating the effect of an increase in the expression of target R2 on target R1 with and without extrinsic noise . Upon increasing the transcription rate k R 2 of target R2 , the threshold of the target R1 shifts towards lower expression levels: the miRNAs are indeed sequestered by R2 and a lower amount of R1 is needed to overcome the threshold . If R1 has a high interaction strength g1 with the miRNA , then the range of bimodality shifts towards lower expression levels as well . The width of the range of R1 bimodality is determined by the interaction strength g2 of the target R2 with the miRNA . If g2 ≫ g1 , then the miRNAs are sequestered by the second target with such a high frequency that the net effect is a reduction in the amount of miRNAs available to target R1 . This entails a shift not only of the k R 1 threshold value but also of the range of bimodality . If g2 < g1 , the second target R2 interacts with low frequency with the miRNA , R1 is slightly derepressed and the net effect on its bimodal distribution is a reduction of the range of transcription rates for which it is present . The emerging picture is that , for a given transcription rate k R 1 , it is possible to tune the distribution of target R1 from monomodal to bimodal and from unrepressed to repressed and vice versa via the expression of target R2 . The presence of extrinsic noise also makes such cross regulation possible for cases with lower miRNA-target interaction on both targets . In Fig 5B we show the bimodality phase diagram for R1 at a fixed interaction strength g1 between miRNA and target R1 , and for a fixed level of extrinsic noise ( see S4 Fig in SI for a different noise level ) . The interaction strength is such that in the case of pure intrinsic noise R1 does not show a bimodal distribution . As an explanatory example , Fig 5C shows that the peaks of R1 distribution can be tuned towards the repressed or the unrepressed case by decreasing or increasing the expression of a second target R2 . Here , the two targets R1 and R2 are both coupled through the noisy miRNA with small interaction strengths . Also in this case , the same analytic approach as before gives good agreement between theory and simulations . These observations suggest that even if the miRNA repression is low and diluted over a network of multiple targets , the noisy environment allows cross-regulation between ceRNAs at the population level ( see Fig 5D , with the intersection between the bimodality phase diagrams for both targets for a fixed level of noise and miRNA interaction strengths ) . Given the relevance of the final product of gene expression , it is important to consider what is the effect of extrinsic noise on proteins’ distributions . From the deterministic system , one can see that the mean amount of target protein is proportional to the amount of its unrepressed mRNA . That is , those molecules not bound to miRNAs and free for translation . In the presence of extrinsic noise the target mRNA can show bimodality even without having a clear double steady state in the deterministic system . Here we investigate if this is the case for the protein distribution . A key factor to keep account of is the time scale of protein synthesis and degradation . In general , if the protein dynamics are fast , the protein distribution follows closely that of the mRNA ( see Fig 6A1 , 6B1 and 6C1 ) . Conversely , slower protein dynamics tend to filter out the intrinsic fluctuations of the mRNA and lead to narrower distributions ( see the SI for a detailed discussion of the case without extrinsic noise ) . This is a single cell effect . That is , for a given rate of miRNA transcription ks , the corresponding protein distribution gets more peaked as the protein dynamics get slower . The protein distribution subject to extrinsic noise also tends to concentrate around its mean . This feature has different consequences for the protein distribution shape according to the specific structure of the mRNA distribution . If the mRNA distribution is bimodal , slower proteins will have a distribution condensing around their mean , which is located close to the unrepressed peak . They will therefore preferentially display the unrepressed phenotype and may completely lose their bimodal structure ( see Fig 6A2 ) . Bimodality can persist for strongly bimodal mRNA distributions because the noise reduction mechanism is acting at the single cell level . Hence it cannot overcome the effects of the extrinsic noise ( see Fig 6B1 and 6B2 ) . For a repressed ( unimodal ) mRNA distribution the mode is far from the mean , so the narrowing around the mean implies the rise of a second ( unrepressed ) peak . For moderately slower dynamics ( see Fig 6C2 ) the protein distribution may be bimodal , and for even slower ones it will be unimodal close to its mean ( see Fig 6C3 ) . Altogether these results suggest that slow proteins promote the expression closer to the mean of the corresponding mRNA distribution . This may or may not be sufficient to remove the bimodal feature of the protein distribution depending on the interplay between the amplitude of the extrinsic noise , the coupling between target and miRNA , and the transcription rates . In the previous sections we considered cases in which the average number of molecules at play is not too low so that the analytic approximation is expected to perform well . Here we investigate how likely the titrative mechanism we analysed is to produce bimodal distributions in regimes of fold repression and mean amount of molecules closer to the endogenous case . We define the fold repression as the ratio between the constitutive expression of the target ( i . e . the value the target would have in absence of miRNA regulation , when g → 0 ) and the target itself . As discussed in detail in SI , the fold repression decreases greatly when adding even a little offset in the amount of mRNA molecules ( see S6 Fig ) . Since our model has no offset , the fold repression we measured should be taken as an upper bound to the ones obtainable in experiments . Fig 7 shows the phase diagram of bimodality for two sets of mean miRNA transcription rate kS and miRNA interaction strength g plotted against the mean amount of free mRNAs and proteins ( Fig 7A and 7C ) and against the fold repression ( Fig 7B and 7D ) . Red lines show the bimodality region . As the figure shows , the bimodality region shrinks and shifts upon decreasing the miRNA transcription rate . The mean values of free mRNAs are of order of hundreds in Fig 7A and of order of tens in Fig 7C ( as measured in [58 , 59] ) and the fold repression ranges between 2 and 10 . The amount of free miRNAs in this regime is of the order of tens , while its total amount ( measured as the ratio between its transcription and degradation rates ) is within 250 molecules per cell . This suggests titration interactions and extrinsic noise may give rise to bimodal distributions also in endogenous regimes . For the sake of simplicity , analytical and numerical tractability , the analysis presented so far was performed by modelling the extrinsic noise as a gaussian random draw of the miRNA transcription rate . The transcription rate was therefore “fixed” at the beginning of each simulation and not assumed to vary in time . Such approximation is expected to describe adequately the scenario where any variation of the miRNA expression rate takes place on longer time scales than the typical ones in the system . However , this might not be the case for some systems: variation of gene expression may happen on time scales of minutes or hour depending on cells’ , and more widely organisms’ , needs . One expects that , in the limit of very fast extrinsic fluctuations , their effect will average out and the scenario will reduce to the case of intrinsic noise only where the bimodality region is narrow , located in proximity to the threshold and found only for strong interaction between microRNA and target mRNA , as discussed in [12] . To confirm these expectations and to investigate the case in which extrinsic fluctuations take place on time scales comparable to the ones of the other reactions , we performed Gillespie simulations allowing the transcription rate of the microRNA to fluctuate in time . In more detail , we first set all parameters in the system as in Fig 6 , where bimodality was observed for static extrinsic noise . We then realised a dynamically fluctuating microRNA transcription rate via an auxiliary birth and death process with finite pool N = 100 ( see SI ) . The steady-state distribution of kS is a Binomial that closely approximates a Gaussian distribution with mean k ¯ S = 1 . 2 × 10 - 3 nM min - 1 and standard deviation σkS=2 . 4×10−4nMmin−1 , i . e . the same distribution from which we drew the rates in the static case discussed in Fig 6 . The time scale τ on which the microRNA transcription rate fluctuates can be explored by changing the magnitude of the birth and death rates , keeping their ratio fixed . To probe regimes in which these fluctuations are faster , comparable and slower than the typical time scales of the reactions we let τ take the values of 8 . 3 × 10−2 min , 0 . 83 min , 21 min , 83 min and 830 min ( see Fig 8 ) . These should be compared with the typical time scales of the various reactions at play in the system . For the target mRNA and protein , the time scale on which significative changes of concentration may occur ranges between about 10 min and 45 min ( see the SI for details on how the time scales were estimated ) . The numerical results confirm that if the miRNA transcription rate varies very rapidly , the effect of the extrinsic noise is averaged out . Indeed , the first panel on the left in Fig 8 shows how the distributions with fast fluctuating miRNA and without extrinsic noise are the same . Considering progressively slower transcription rate variations , the bimodality gradually reappears , and eventually recovers the static case in the slow-variation regime . For τ = 830 min the resulting distributions are practically indistinguishable from the static extrinsic noise case treated in the previous sections . This time scale is comparable to ( shorter than ) 24 hours ( 1440 min ) . Then , we expect the results obtained for static noise to be relevant for settings in which the extrinsic noise is caused by variations along the cell cycle , in cells dividing every 24 hours .
Previous studies pointed out the relevance of extrinsic noise in molecular networks in shaping cell decision making and differentiation [9 , 10] . Although extrinsic noise influences gene expression and regulation at different levels , the dominating variability across a cell population seems due to population dynamics [30]: even a monoclonal population has cells in different phases of their cell cycle because of growth and divisions . Such intra-population variability may manifest into heterogeneous expression patterns , which eventually develop a bimodal distribution [62 , 63] . When bimodal distributions are observed in gene expression levels , these modes often correspond to different physiological states of the system [1–6] . In this work we addressed the question of the role of extrinsic noise in shaping bimodal gene distributions in the context of miRNA-mediated regulation , both with stochastic modelling and simulations . As observed in vitro [13] , in particular stoichiometric conditions miRNAs may induce bimodality in the expression of their target genes simply due to peculiar titrative interactions . In a theoretical system with pure intrinsic noise , such bimodal distributions can be observed in conditions of high miRNA-target interaction strength and for a small range of target transcription rates [12] . The binding and unbinding reactions between miRNA and target in conditions of quasi equimolarity let the target “jump” from the bound to the unbound state , giving rise to bimodal distributions . This bimodality is observed at a single-cell level: every single cell can indeed switch from one state to the other so that , at a given time , part of the population will be “bound” and part “unbound” . We showed that introducing some extrinsic noise to the miRNA transcription widens the range of target transcription rates for which one observes target bimodality . In this case the bimodal distributions arise at the population level , made of several cells that are heterogeneous with respect to miRNA expression and therefore amount . Hence , the bimodality arises from the superposition of those unimodal distributions describing the single cells , i . e . , each of them obtained for a different value of miRNA transcription rate . Interestingly , in this framework , a high miRNA-target interaction strength is not necessary to obtain a population-induced bimodal distribution . We showed that extrinsic noise and miRNA-target interaction strength act at similar levels with respect to the bimodality . The interaction strength between miRNA and target in our model takes into account the possibility of different numbers of miRNA binding sites on the mRNA target sequence . Since the miRNA repression on a given target is usually small and possibly diluted over multiple targets , our results suggest that some extrinsic noise can compensate for a low interaction strength in order to obtain differentially expressed phenotypes . Since every single miRNA may have many different targets that in turn compete for the shared pool of miRNAs , a change in the expression level of one of them may alter the expression of the other ones depending on the prioritisation of their interaction strengths [12 , 64] . Once sorted , the interaction strengths would indeed provide the specificity of the interaction among a group of miRNAs and a group of targets , as shown in [12] . While from a purely theoretical point of view , there is in principle no limit on the number of genes that can be indirectly regulated by another one ( it is just a matter of parameters to tune ) , of course in physiological conditions the number of genes involved in the crosstalk may be limited and case specific . If one of these targets is bimodally distributed , then the bimodality may be influenced by the expression level of the other miRNA competitors according to their prioritised interaction strengths , and in turn may induce the expression of other targets to become bimodal . We modelled the simplest version of this scenario considering two targets in competition for the same miRNA and showed that cross regulation is possible even in the case of small miRNA-target interaction strengths if some extrinsic noise is present . In particular , different targets may cross-regulate each other’s bimodal distributions and their interplay is pivotal in stabilising the presence of single phenotypes . This suggests that even if the miRNA repression is low and diluted over a network of multiple targets , the noisy environment makes cross-regulation among them possible . One may then wonder whether the appearance of bimodality is due to unphysiological amount of molecules or it may appear in endogenous situations , with a mean amount of mRNAs in the order of 10 − 1000 [58 , 59] and small fold repressions [13 , 18] . As shown in Fig 7 , the bimodality region may span values of mean mRNA amount in the order of tens and fold repression values ranging between 1 and 10 . The mean amount of free miRNAs corresponding to these values is in the order of tens , while its total amount is below 250 molecules per cell . These values suggest that bimodal distributions of the target might appear in endogenous situations . The importance of miRNAs in increasing protein noise in highly expressed genes was recently suggested in [29] . This result should be read together with the finding that signalling factors and developmental regulators in embryonic stem cells show bimodal expression patterns only in presence of mature miRNAs [62 , 63] . As pointed out in [28] , as a whole these studies suggest a role of miRNAs in generating gene-expression variability . Consistently with the cell-to-cell miRNA variability observed in [29] , they also suggest that variation in miRNA expression may influence the co-variation of factors that are more likely to fluctuate together to trigger transitions between cell states . The outcome of regulatory systems is the control of protein expression . Concerning the effect of extrinsic noise on their distribution , we showed that depending on the time scales of protein synthesis and degradation , the protein distribution may suppress or amplify the bimodality inherited from the mRNA . Altogether our results suggest that the coupling between extrinsic noise and threshold behaviour represents a possible mechanism to obtain bimodal phenotypes without the need for fine tuning the rates of reactions which was required for the case of intrinsic noise only . On one hand , we observed that the system is , for a broad range of parameters , able to buffer the extrinsic fluctuations and channel only one final phenotype ( unimodality region in Figs 4 and 7 ) . On the other hand , we have highlighted how extrinsic noise widens the bimodality region compared to the intrinsic-noise-only case . This suggests how bimodality can arise as a built-in effect of the coupling between the miRNA-based titration and the presence of noise . This may contribute to the understanding of the bimodal distributions observed in [13] . Such bimodality may play a role in view of the advantages related to having a high variability among different cells . Given an estimate of the miRNA-target interaction strengths , the model allows the prediction of the amount of extrinsic noise necessary to induce a bimodal phenotype . A feasible strategy to test the experimental validity of these theoretical results involves building ad-hoc synthetic circuits made of miRNA targets tagged with fluorescent labels , as previously done in [13 , 65] , and performing transfection experiments . Even though the reporters are expressed at artificial levels and with artificial dynamics , with the appropriate reporter-controls the experimental set-up is controlled enough to check the validity of the model . By analysing the fluorescence patterns of the targets interacting with the miRNA throughout the entire population of cells , the shape of the target distributions can then be extracted and compared to the theoretical predictions . MiRNAs are differentially expressed in different tissues and we also expect the amplitudes of their extrinsic fluctuations to vary consequently . Hence , repeating the experiment in cells derived from different tissues would allow the study of cells exposed to different levels of extrinsic noise . This approach would enable the construction of a phase diagram for the bimodality linking extrinsic noise and model parameters , as done in [13] . Nowadays , relevant experiments should use CRISPR-tagged endogenous proteins and inducible systems for miRNA expression to move a physiological system into the desired region of the phase diagram . This would provide a valuable and precise tool to define and control the key variables for the appearance of bimodal target distributions , particularly in disease-related contexts .
|
Phenotypic differentiation often relies on bimodal distributions of gene expression levels , which can normally be achieved by different molecular mechanisms . During the past decade microRNAs , small noncoding RNA molecules , were found to downregulate the expression of preferred mRNA targets by sequestering and successively degrading them , thus influencing the level of gene expression . We theoretically address the question on how microRNA-mediated regulation can induce the appearance of bimodal phenotypes . Our findings show that the presence of extrinsic noise favours bimodal distributions . This suggests a simple mechanism for obtaining bimodal populations where the presence of extrinsic noise relaxes the requirements on parameters fine tuning .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"gene",
"regulation",
"cell",
"cycle",
"and",
"cell",
"division",
"messenger",
"rna",
"cell",
"processes",
"micrornas",
"probability",
"distribution",
"mathematics",
"phase",
"diagrams",
"computer",
"and",
"information",
"sciences",
"gene",
"expression",
"probability",
"theory",
"biochemistry",
"rna",
"biochemical",
"simulations",
"data",
"visualization",
"nucleic",
"acids",
"cell",
"biology",
"protein",
"translation",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"computational",
"biology",
"non-coding",
"rna"
] |
2018
|
On the role of extrinsic noise in microRNA-mediated bimodal gene expression
|
The delta-retrovirus Human T-cell leukemia virus type 1 ( HTLV-1 ) preferentially infects CD4+ T-cells via cell-to-cell transmission . Viruses are transmitted by polarized budding and by transfer of viral biofilms at the virological synapse ( VS ) . Formation of the VS requires the viral Tax protein and polarization of the host cytoskeleton , however , molecular mechanisms of HTLV-1 cell-to-cell transmission remain incompletely understood . Recently , we could show Tax-dependent upregulation of the actin-bundling protein Fascin ( FSCN-1 ) in HTLV-1-infected T-cells . Here , we report that Fascin contributes to HTLV-1 transmission . Using single-cycle replication-dependent HTLV-1 reporter vectors , we found that repression of endogenous Fascin by short hairpin RNAs and by Fascin-specific nanobodies impaired gag p19 release and cell-to-cell transmission in 293T cells . In Jurkat T-cells , Tax-induced Fascin expression enhanced virus release and Fascin-dependently augmented cell-to-cell transmission to Raji/CD4+ B-cells . Repression of Fascin in HTLV-1-infected T-cells diminished virus release and gag p19 transfer to co-cultured T-cells . Spotting the mechanism , flow cytometry and automatic image analysis showed that Tax-induced T-cell conjugate formation occurred Fascin-independently . However , adhesion of HTLV-1-infected MT-2 cells in co-culture with Jurkat T-cells was reduced upon knockdown of Fascin , suggesting that Fascin contributes to dissemination of infected T-cells . Imaging of chronically infected MS-9 T-cells in co-culture with Jurkat T-cells revealed that Fascin’s localization at tight cell-cell contacts is accompanied by gag polarization suggesting that Fascin directly affects the distribution of gag to budding sites , and therefore , indirectly viral transmission . In detail , we found gag clusters that are interspersed with Fascin clusters , suggesting that Fascin makes room for gag in viral biofilms . Moreover , we observed short , Fascin-containing membrane extensions surrounding gag clusters and clutching uninfected T-cells . Finally , we detected Fascin and gag in long-distance cellular protrusions . Taken together , we show for the first time that HTLV-1 usurps the host cell factor Fascin to foster virus release and cell-to-cell transmission .
Human T-cell leukemia virus type 1 ( HTLV-1 ) , which infects approximately 5–10 million people worldwide [1] , is the only human retrovirus causing cancer: adult T-cell leukemia/lymphoma ( ATL ) , a fatal neoplasia of CD4+ T-cells [2–4] . Further , HTLV-1 is the causative agent of a neurodegenerative , inflammatory disease , HTLV-1-associated myelopathy/tropical spastic paraparesis ( HAM/TSP ) [5 , 6] . Both diseases can develop as a consequence of prolonged viral persistence in T-cells after a clinical latency of decades in 1–5% ( ATL ) or 3–5% ( HAM/TSP ) of infected individuals [7 , 8] . Activated CD4+ T-cells are the main and preferential target for HTLV-1 infection , but the virus is also present in very low amounts in other cell types including CD8+ T-cells , monocytes , and dendritic cells ( DC ) [9] . After binding to its receptor , which is composed of the glucose transporter GLUT-1 , neuropilin-1 ( NRP-1 ) and heparan sulfate proteoglycans ( HSPGs ) [10–12] , HTLV-1 integrates into the host cell genome . The virus is mainly maintained in its provirus form ( 9 . 1 kb ) , which is flanked by long terminal repeats ( LTR ) in both the 5’ and 3’ region . In addition to structural proteins and enzymes common for retroviruses , HTLV-1 encodes regulatory ( Tax , Rex ) and accessory ( p12/p8 , p13 , p30 , HBZ ) proteins [13] . HTLV-1 replicates either by infecting new cells or by mitotic division and clonal expansion of infected T-cells [14–16] . Efficient infection of CD4+ T-cells requires cell-cell contacts and coordinated steps of the virus infectious cycle with events in the cell-cell adhesion process . Thus , transmission of HTLV-1 occurs via breast feeding , sexual intercourse , and cell-containing blood products [9 , 17] . Unlike human immunodeficiency virus ( HIV ) or murine leukemia virus ( MLV ) , cell-free transmission of HTLV-1 to T-cells is inefficient . Therefore , only a limited amount of poorly infectious viral particles is produced from infected lymphocytes and free virions can hardly be detected in infected individuals [18–20] . Thus far , two types of cell-cell contacts have been described to be critical for HTLV-1 transmission , tight cell-cell contacts and cellular conduits [21 , 22] . For transmission at tight cell-cell contacts , two non-exclusive mechanisms of virus transmission at the virological synapse ( VS ) , a virus-induced specialized cell-cell contact [23] , have been proposed [17]: ( 1 ) polarized budding of HTLV-1 into synaptic clefts [21] , and ( 2 ) cell surface transfer of viral biofilms [24] . The latter consist of extracellular , concentrated viral assemblies that are surrounded by components of the extracellular matrix and cellular lectins [24] . Beyond , transmission via biofilms seems to be a major route of transmission since removal of biofilms by heparin treatment impairs cell-to-cell transmission by 80% in vitro [24] . Independent of the route of HTLV-1 transmission , viral particles are thought to be transmitted in confined areas protected from the immune response of the host in vivo . Moreover , cytoskeletal remodeling and cell-cell contacts are a prerequisite for all routes of virus transmission [21 , 25] . Although it is known that the viral protein Tax and polarization of the host cell cytoskeleton are crucial for formation of the VS and for HTLV-1 transmission ( for details see: [17 , 23] ) , only little is known about the link between Tax and remodeling of the cytoskeleton to foster viral spread . The regulatory protein Tax is essential for viral replication due to strong enhancement of viral mRNA synthesis by transactivating the HTLV-1 LTR ( U3R ) promoter . Beyond , Tax is a potent transactivator of cellular transcription and important for initiating oncogenic transformation [13] . Tax is also critical for HTLV-1 transmission since Tax cooperates with intercellular adhesion molecule 1 ( ICAM-1 ) , thereby inducing polarization of the microtubule organizing center ( MTOC ) at the VS [26] and thus , enhancing HTLV-1 cell-to-cell transfer . Furthermore , Tax enhances both actin- and tubulin-dependent transmission of virus-like particles ( VLPs; [25] ) . However , only few host cell factors with a role in Tax-induced virus transmission have been characterized . Among those is ICAM-1 , which is induced by Tax and cooperates with Tax in VS formation [26 , 27] . The Tax-induced small GTP-binding protein GEM enhances cellular migration , conjugate formation , and thus , is required for viral transmission [28] . In our search of novel target genes of Tax with a putative role in virus transmission , we have previously identified the evolutionary conserved actin-bundling protein and tumor marker Fascin as a new host cell factor strongly induced by Tax [29] . Fascin cross-links filamentous actin and stabilizes cellular protrusions , filopodia , and invadopodia [30] . Recent work shows that Fascin also interacts with microtubules to regulate adhesion dynamics and cell migration [31] . Fascin has evolved as a therapeutic target in several types of cancer since Fascin expression is associated with metastasis in malignant tumors and it correlates with clinical aggressiveness of some tumors [30] . In hematopoietic cells , Fascin is expressed in mature DC where it is important for stability of dendrites and for formation of the immunological synapse [32] , while no expression of Fascin can be detected in unstimulated human T-cells [33] . We found that expression of Fascin is a common feature of chronically HTLV-1-infected T-cell lines . Fascin colocalizes with actin in the cytoplasm and at the membrane of HTLV-1-infected cells . Furthermore , knockdown of Fascin reduces the invasive capacity of HTLV-1-infected ATL-derived T-cells into extracellular matrix [29] . Since expression of Tax is sufficient to induce expression of Fascin [29 , 34] and Tax enhances actin-dependent virus transmission [25] , we now asked whether Fascin affects HTLV-1 cell-to-cell transfer . Here , we report that Fascin is crucial for release and cell-to-cell transmission of HTLV-1 in different cell model systems . While T-cell conjugate formation is Fascin-independent , cell adhesion of infected cells in co-culture with uninfected cells is impaired upon repression of Fascin . Imaging of Fascin and the viral gag protein at cell-cell contacts suggests a role of Fascin in transmission potentially by redirecting viral proteins to budding sites . Thus , Fascin as a major contributor to HTLV-1 transmission provides a link between Tax’s activity and virus transmission .
293T cells ( kindly provided by Ralph Grassmann ( deceased ) , FAU , Erlangen , Germany ) were cultured in DMEM containing 10% fetal calf serum ( FCS ) , L-glutamine ( 0 . 35g/l ) and penicillin/streptomycin ( Pen/Strep; 0 . 12g/l each ) . For selection of stable 293T cells carrying shRNAs , 4μg/ml puromycin was added to the media . The CD4+ T-cell line Jurkat ( ATCC , LGC Standards GmbH , Wesel , Germany ) from acute lymphoblastic leukemia was cultured in RPMI 1640M , Panserin , 10% FCS , L-glutamine and Pen/Strep [35] . The human Epstein-Barr virus ( EBV ) -positive B-cell line Raji derived from Burkitt’s lymphoma containing the surface receptor CD4 ( Raji/CD4+ ) was a kind gift from Vineet N . Kewal Ramani ( NIH , Frederick , Maryland , USA ) and was cultured in RPMI 1640M , Panserin , 10% FCS , L-glutamine and Pen/Strep containing 500μg/ml geneticin to ensure retainment of the CD4 receptor [36] . The HTLV-1 in vitro transformed CD4+ T-cell line MT-2 [3] and the ATL-derived CD4+ T-cell line HuT-102 [2 , 37] were kindly provided by Ralph Grassmann ( deceased , FAU , Erlangen , Germany ) and were cultured in RPMI 1640M , 10% FCS and Pen/Strep . The HTLV-1 in vitro transformed T-cell line MS-9 ( containing a single , full-length provirus ) [38] was a kind gift from Charles Bangham ( Imperial College , London , UK ) and was cultured in RPMI 1640M , Panserin , 20% FCS , Pen/Strep and 100U/ml interleukin 2 ( IL-2 ) . All cell lines were checked for integrity by DNA profiling of eight different and highly polymorphic short tandem repeat loci ( DSMZ , Braunschweig , Germany ) . In general , 107 Jurkat T-cells were transiently transfected by electroporation using the Gene Pulser X Electroporation System ( BioRad , Munich , Germany ) at 290V and 1500μF . Cells were transfected using a total of 50 or 100μg of DNA . 5x105 293T cells or stable 293T cell lines that carry shNonsense , shFascin5 or shFascin4 were seeded in 6-well plates 24h prior to transfection . Cells were transfected with GeneJuice reagent ( Merck Millipore , Darmstadt , Germany ) according to the manufacturer’s protocol using a total amount of 2μg DNA . HuT-102 cells stably transduced with shRNAs targeting Fascin ( shFascin5 ) or a control ( shNonsense ) were co-cultured with Jurkat T-cells that had been transfected 24h earlier with the luciferase reporter plasmid pGL3-U3R-Luc carrying the luc gene under control of the HTLV-1 core promoter U3R [44] , or with the control plasmid pGL3-Basic ( Promega , Mannheim , Germany ) . After 48h of co-culture at 37°C , luciferase reporter gene assays were performed . Relative light units ( RLU ) were normalized on protein content and on background activity of controls ( pGL3-Basic ) . Values obtained in control cells were set as 100% and at least three independent experiments each performed in triplicate were executed . Cells were washed once with PBS ( without Ca2+ and Mg2+ ) and then lysed in 100μl lysis buffer ( 100mM Tris/HCl ( pH 7 . 8 ) , 1M dithiotreitol ( DTT ) , 0 . 18mM DCTA , 0 . 2% Triton X-100 , 20% glycerol ) . After shaking for 30min at 30°C , samples were centrifuged ( 14 . 000rpm , 15min , 4°C ) and supernatants were kept . Luciferase activities were measured according to the manufacturer’s instructions ( Orion luminometer ) using assay buffer ( 100mM KPO4 , 15mM MgSO4 , 4mM ATP ) and D-luciferin ( 0 . 26 mg/ml; Roche Diagnostics , Indianapolis , IN , USA ) dissolved in assay buffer . 5x105 of the respective cells were seeded and incubated for 48h at 37°C . Cells were centrifuged ( 1200rpm , 5min , 25°C ) , pellets were used for western blot analysis , and supernatants of MT-2 cells or of co-cultures from experiments using the single-cycle replication-dependent reporter vectors ( see Infection assays ) were sterile filtrated , and virus release was measured using gag p19 ELISA according to the manufacturer’s protocol ( ZeptoMetrix Corporation , Buffalo , NY , USA ) . MT-2 cells were either left untreated or treated with DMSO ( solvent control ) , cytochalasin D or nocodazole ( 5μM each ) 48h prior to harvest . Additionally , MT-2 cells that stably carry shRNAs ( shNonsense or shFascin5 ) were analyzed . Data were obtained using Softmax Pro Version 5 . 3 software ( MDS Analytical Technologies , Sunnyvale , California , USA ) . At least , four independent experiments were performed . Cells were washed once with PBS and protein lysates were obtained by lysis of cells in 100μl lysis buffer ( 150mM NaCl , 10mM Tris/HCl ( pH 7 . 0 ) , 10mM EDTA , 1% Triton X-100 , 2mM DTT and protease inhibitors leupeptin , aprotinin ( 20μg/ml each ) and 1mM phenylmethylsulfonyl fluoride ( PMSF; 1mM ) ) . After repeated freeze-and-thaw cycles , lysates were centrifuged ( 14 . 000rpm , 15min , 4°C ) . For detection of Tax , samples were sonicated three times for 20sec before centrifugation . Equal amounts of protein ( 50μg ) were denatured for 5min at 95°C in sodium dodecyl sulfate ( SDS ) loading dye ( 10mM Tris/HCl ( pH 6 . 8 ) , 10% glycerin , 2% SDS , 0 . 1% bromophenol blue , 5% β-mercaptoethanol ) . After SDS-PAGE and immunoblotting on nitrocellulose transfer membranes ( Whatmann , Protran , Whatmann GmbH , Dassel , Germany ) , proteins were detected using the following antibodies: rabbit polyclonal antibodies anti-V5 ( Sigma ) , mouse monoclonal antibodies anti-Fascin ( 55K-2; Dako Deutschland GmbH , Hamburg , Germany ) , anti-β-actin ( ACTB; Sigma ) , anti-Hsp90 α/β ( F-8; Santa Cruz Biotechnology , Heidelberg , Germany ) , anti-HTLV-1 gag p19 ( ZeptoMetrix Corporation ) , and anti-GFP ( Sigma ) , and mouse antibodies to Tax , which were derived from the hybridoma cell line 168B17-46-34 ( provided by B . Langton through the AIDS Research and Reference Reagent Program , Division of AIDS , NIAID , NIH; [45] ) . Secondary antibodies conjugated with horseradish peroxidase ( HRP; GE Healthcare , Little Chalfont , UK ) were used . Peroxidase activity was detected by enhanced chemiluminescence ( ECL; 98 . 9% ECL A , 1% ECL B , 0 . 031% H2O2 ) using a CCD camera ( Kodak Image Station 4000MM Pro camera , Kodak or Fujifilm LAS-1000 Intelligent Dark Box; Fujifilm ) . One of at least three independent western blots per experiment is shown . Intensities of specific bands were quantitated using Advanced Image Data Analyser ( AIDA Version 4 . 22 . 034 , Raytest Isotopenmessgeräte GmbH , Straubenhardt , Germany ) , and values were normalized on those of the housekeeping gene Hsp90 α/β . 107 Jurkat T-cells were transfected with 50μg pEFTax or pEF . 5x105 293T cells were transfected with 2μg of pEFTax or pEF ( see Transfections ) . 48h later , total cellular RNA was isolated from transfected Jurkat or 293T cells ( RNA isolation Kit II , Macherey-Nagel , Düren , Germany ) and reversely transcribed to cDNA using SuperScript II and random hexamer primers ( both Life Technologies GmbH ) . 200ng of cDNA and SensiMix II Probe Kit ( BioLine GmbH , Luckenwalde , Germany ) were used according to the manufacturer’s instructions for quantitative real-time RT-PCR ( qPCR ) in an ABI Prism 7500 Sequence Analyzer ( Applied Biosystems , Foster City , CA , USA ) . Primers and FAM ( 6-carboxyfluorescein ) / TAMRA ( tetramethylrhodamine ) -labeled probes for detection of β-actin ( ACTB ) and Tax transcripts have been described before [46] . A TaqMan Gene Expression Assay ( Hs00979631_g1; Applied Biosystems ) was used for quantitation of Fascin transcripts . Expression levels were computed by interpolation from standard curves generated from plasmids carrying the respective target sequences and calculation of the mean of triplicated samples . Relative copy numbers ( rcn ) were determined by normalizing copy numbers on those of ß-actin ( ACTB ) . At least , three independent experiments were performed . Microsoft Office Excel software was used for statistical analysis using the t-test ( unpaired ) . P<0 . 05 was considered to be significant .
To assess the role of Fascin on HTLV-1 transmission , we made use of a single-cycle replication-dependent reporter system that is transfected into donor cells and allows monitoring of reporter gene activity in newly infected target cells only [25] . Briefly , a virus packaging plasmid encoding all HTLV-1 genes ( wildtype; wt ) and a replication-dependent HTLV-1 reporter vector containing a CMV-promoter driven luciferase ( luc ) gene were co-transfected into 293T cells . Alternatively , an HTLV-1 packaging plasmid carrying a mutation in the envelope ( env ) gene , and a VSV-G-expression plasmid were co-transfected instead of wt . The luc gene is oriented in antisense and is interrupted by an intron oriented in sense , therefore translation of the reporter mRNA in transfected cells is precluded . The vector mRNA is spliced and packaged into VLPs . After infection and replication , a provirus that lacks the intron is generated , and reporter gene expression ( luc activity ) can be measured in the target cell [25] . We previously used this system to assess the role of cellular restriction factors on HTLV-1 [49] . To analyze whether Fascin is important for transmission of these HTLV-1 reporter vectors , stable 293T cells with a knockdown of Fascin were generated . For this purpose , cells were transfected with two different shRNA constructs carrying an IRES-EGFP expression cassette and shRNAs targeting Fascin ( shFascin5 , shFascin4; [29 , 42] ) or a control ( shNonsense ) , and cells were selected with puromycin . Flow cytometry monitoring GFP-expression revealed that approximately 90% of cells carried the shRNA-constructs ( S2A Fig ) . Beyond , vitality of stable cell lines was unaffected by the presence of shRNAs as detected by live/dead staining ( S2B Fig ) . Cell lines were transfected with single-cycle replication-dependent HTLV-1 reporter vectors ( inluc ) , a viral packaging plasmid ( Δenv or wt ) , and as indicated with VSV-G for pseudotyping ( Fig 1A ) . sh293T cells ( shNonsense ) transfected with inluc and Δenv served as negative control ( control , Fig 1B ) for both wt env-carrying and VSV-G-pseudotyped viral particles . After 24h , media were changed , and after another 24h , cells were harvested to measure cell-to-cell transmission in luciferase assays ( Fig 1B ) , virus release by gag p19 ELISA ( Fig 1C ) and protein expression by western blot analysis ( Fig 1D ) . Making use of single-cycle replication-dependent HTLV-1 reporter vectors revealed that stable repression of endogenous Fascin by shRNAs leads to a significant reduction of reporter gene activity ( Fig 1B ) . While shFascin5 resulted in a strong reduction of reporter gene activity by more than 70% ( Fig 1B ) and , in parallel , of Fascin protein ( Fig 1D ) , the influence of shFascin4 on reporter gene activity ( Fig 1B; by 30% ) and on Fascin protein expression ( Fig 1D ) was less pronounced . Overexpression of Tax ( black bars ) did not enhance transmission of VSV-G-pseudotyped HTLV-1 ( Fig 1B , left part of left panel ) , confirming earlier observations [25] . However , overexpressed Tax enhanced cell-to-cell-transmission of HTLV-1 reporters packaged with wt env ( Fig 1B , right part of left panel and enlargement in right panel ) contrary to previous observations [25] . The latter suggests that Tax and wt env cooperate in cell-to-cell transmission . The relative infectivity of VSV-G pseudotyped reporter vectors was about 7-fold higher than that of wt env-pseudotyped reporter vectors ( Fig 1B ) and undetectable , if no envelope was added ( control ) . However , independent of the envelope type used , repression of Fascin significantly reduced the relative infectivity of both wt env-carrying and VSV-G-pseudotyped reporter vectors . Moreover , independent of the experimental condition , Tax could not further induce expression of Fascin protein in 293T cells ( Fig 1D ) . While Tax led to a robust induction of Fascin mRNA ( S4A Fig ) and protein ( S4B Fig ) in Jurkat T-cells confirming our previous work [29] , Tax did not further modulate Fascin expression in 293T cells , which already exhibit high amounts of endogenous Fascin ( S4A and S4B Fig ) . Yet , reporter gene activity as a measure of HTLV-1 cell-to-cell transmission was Fascin-dependent in presence of overexpressed Tax , too ( Fig 1B ) , suggesting that Fascin is also important for HTLV-1 cell-to-cell transmission in cells which express high amounts of endogenous Fascin . To analyze whether Fascin also impairs virus release , the viral gag p19 protein was measured by ELISA in cells that had been transfected with HTLV-1 reporters packaged with wt env . Overexpression of Tax ( Fig 1C , black bars ) did not further enhance gag p19 levels in the supernatants of 293T cells and repression of Fascin led to approximately 40% reduction of virus release only when Tax was supplemented suggesting that Tax and Fascin cooperate in processes that are important for virus release . Since repression of Fascin led to a severe defect of cell-to-cell transmission also in absence of supplemented Tax ( Fig 1B , grey bars ) , but not to a decrease in virus release ( Fig 1C , grey bars ) , this suggests that Fascin’s role in cell-to-cell transmission dominates over its role on virus release in this experimental setup . However , results obtained by the reporter system only provide a signal upon productive infection of a target cell , while the gag p19 ELISA also quantifies non-infectious VLPs . To exclude that virus production in the cell is already impaired by repression of Fascin , western blots detecting gag were performed ( Fig 1D ) . Overall levels of cell-associated gag p55 were comparable between different experimental conditions . Beyond , Fascin was strongly repressed in presence of shFascin5 and only moderately repressed in presence of shFascin4 . We could also detect Tax expressed from the packaging plasmids as tiny band , and an increased expression of Tax upon supplementing a Tax expression plasmid . Taken together , our data indicate that Fascin is important for transmission of HTLV-1 reporter vectors independent of the envelope type in 293T cells . To strengthen our results , we made use of Fascin-specific nanobodies that had been developed and characterized previously [43] . Briefly , nanobodies are antigen-binding domains of camelid heavy-chain antibodies . The employed Fascin-specific nanobodies contain a mitochondrial outer membrane ( MOM ) signal that leads to targeted subcellular delocalization of Fascin to the MOM [43] . Use of these nanobodies allowed us to trigger Fascin protein loss of function without changing its expression . Upon transfection of HTLV-1 reporter vectors and expression plasmids encoding Fascin-specific nanobodies into 293T cells , luciferase assays ( Fig 2A ) , gag p19 ELISA ( Fig 2B ) , western blot analysis ( Fig 2C ) and immunofluorescence stains followed by confocal laser scanning microscopy ( Fig 2D–2F ) were performed . We found that Fascin nanobody 5 ( FASNb5 ) significantly reduced reporter gene activity in a dose-dependent manner compared to a control nanobody ( GFPNb ) and to FASNb2 ( Fig 2A ) . In gag p19 ELISA ( Fig 2B ) , we measured a dose-dependent decrease of released gag p19 into the supernatants in presence of FASNb5 , suggesting that this nanobody also impairs release of HTLV-1 . Expression of the nanobodies and the unaltered expression of Fascin were confirmed by western blot analysis ( Fig 2C ) . Additionally , immunofluorescence was performed confirming earlier studies [43] showing co-localizations of V5-tagged and MOM-expressing nanobodies ( GFPNb , FASNb2 , FASNb5 ) with mitochondria ( Fig 2D ) . Next , we checked whether Fascin-specific nanobodies lead to efficient delocalization of Fascin by staining V5-tagged nanobodies and Fascin . Immunofluorescence analysis revealed that FASNb5 lead to a more efficient delocalization of Fascin ( 90 . 2% of Fascin delocalized ) compared to FASNb2 ( 71 . 1%; Fig 2E ) , which mirrors the different impact of FASNb5 and FASNb2 on cell-to-cell transmission ( Fig 2B ) and virus release ( Fig 2C ) . However , since FASNb5 also impairs Fascin-mediated actin-bundling compared to FASNb2 [43] , these data also suggest that Fascin’s actin-bundling activity could be required for transmission of HTLV-1 . Contrary to delocalizing Fascin , FASNb5 did not delocalize gag to the mitochondria ( Fig 2F ) , suggesting that Fascin and gag do not directly interact , or , if they interact , the interaction is not sustained during delocalization . Moreover , the impact of FASNb5 on virus release as measured by gag p19 ELISA may be indirect ( Fig 2B ) , e . g . by impairing the transport of gag to budding sites or by impairing budding itself . Summed up , not only repression , but also delocalization of Fascin in the cell interferes with HTLV-1 cell-to-cell transmission . Our results obtained thus far do not exclude that repression of Fascin impairs viral entry since we used a one-step transfection/infection co-culture system , where transfected cells produce VLPs that infect neighboring cells [25] , which also harbor a knockdown of Fascin , or which could be impaired by Fascin-specific nanobodies . Further , since HTLV-1 predominantly infects CD4+ T-cells in vivo , we switched to a more physiological system and analyzed the role of Tax and Fascin on HTLV-1 transmission in CD4+ Jurkat T-cells in co-culture with Raji/CD4+ B-cells , a co-culture system that had been described earlier to allow monitoring of HTLV-1 transmission with single-cycle replication-dependent reporter vectors [25] . Upon co-transfection of Jurkat T-cells with HTLV-1 reporters ( inluc ) , packaging plasmids ( wt ) , Tax expression plasmids and shRNAs targeting Fascin ( Fig 3A ) , media were changed at 24h , and Jurkat T-cells were co-cultured for another 48h with Raji/CD4+ B-cells . Measurement of luciferase reporter gene activity reflecting cell-to-cell transmission revealed that repression of Fascin did not affect basal HTLV-1 cell-to-cell transmission ( Fig 3B , grey bars ) . However , upon overexpression of Tax ( black bars ) , reporter gene activity significantly increased confirming earlier observations in this cell type [25] . Interestingly , repression of Fascin led to a reduction of Tax-induced reporter gene activity suggesting that Fascin is a major contributor of Tax-induced cell-to-cell transmission . To exclude and to confirm that the measured reporter gene activity is not due to cell-free virus transmission , we incubated Raji/CD4+ B-cells with supernatants of Jurkat T-cells that had been transfected with the reporter system , which did not result in detectable luciferase signals ( S3 Fig; [25] ) . Measuring of gag p19 release by ELISA mirrored the results obtained by luciferase assays and showed that Tax-enhanced virus release in Jurkat T-cells occurs Fascin-dependently ( Fig 3C ) . Knockdown of Fascin in presence of overexpressed Tax led to a reduction of gag p19 release nearly reaching those levels measured without supplemented Tax . Western blot analysis showed that , contrary to 293T cells ( Fig 1D , S4 Fig ) , Tax is a potent inducer of Fascin transcript and protein expression in Jurkat T-cells ( Fig 3D , S4 Fig ) confirming our previous results [29 , 34] . Thus , too low levels of endogenous Fascin in Jurkat T-cells without overexpressed Tax ( S4 Fig ) could be a potential explanation for the Tax-dependency of the Fascin-effect in this cell type . Concomitant with our findings obtained in 293T cells ( Fig 1D ) , western blot analysis revealed that the levels of cell-associated gag p55 were comparable between different experimental conditions also in Jurkat T-cells ( Fig 3D ) . Contrary to 293T cells , Fascin was induced by Tax in Jurkat T-cells . Further , Fascin was strongly repressed in presence of shFascin5 and moderately repressed in presence of shFascin4 . Tax expressed from the packaging plasmids could be detected as a tiny band , and an increased expression of Tax was detectable upon supplementing a Tax expression plasmid . Thus , Tax enhances virus release , and augments cell-to-cell transmission from Jurkat T-cells to Raji/CD4+ B-cells dependent on Fascin . To substantiate these findings , we tested Fascin-specific nanobodies instead of shRNAs in the Jurkat-Raji/CD4+ co-culture model . Compared to the control nanobody GFPNb and to FASNb2 , FASNb5 led to a significant reduction ( by 61% ) of Tax-induced HTLV-1 reporter gene activity ( Fig 3E ) . This suggests that delocalization of Fascin without changing its expression ( Fig 3F ) and inhibition of Fascin’s actin-bundling activity by FASNb5 [43] impair HTLV-1 cell-to-cell transmission in different cell types ( Figs 2 , 3E and 3F ) . Taken together , use of single-cycle replication-dependent HTLV-1 reporter vectors revealed that stable repression of endogenous Fascin ( 293T cells ) , or of Tax-induced Fascin ( Jurkat T-cells ) by shRNAs and inhibition of Fascin using specific nanobodies impair both gag p19 release and HTLV-1 cell-to-cell transmission . Next , we asked whether Fascin contributes to HTLV-1 cell-to-cell transmission also in chronically HTLV-1-infected T-cells , which express high amounts of Fascin protein . For this purpose , ATL-derived HTLV-1-infected HuT-102 cells were stably transduced with lentiviral vectors expressing either a shRNA targeting Fascin ( shFascin5 ) or a nonsense shRNA ( shNonsense ) [29] . According to a published protocol [50] , HuT-102 cells were co-cultured with Jurkat T-cells that had been transfected with an HTLV-1-LTR ( U3R ) -dependent luc gene reporter system ( pGL3-U3R; Fig 4A ) . Upon infection of Jurkat T-cells , the viral Tax protein should activate expression of the HTLV-1 U3R resulting in enhanced luciferase activity . After 24h of co-culture , luciferase activity was measured and normalized on protein content and on transactivation of a mock luciferase construct ( Fig 4B ) . Transactivation of the reporter in Jurkat T-cells was diminished by more than 50% when Fascin was knocked down in the co-cultured HTLV-1-infected HuT-102 cells hinting at a role of Fascin for HTLV-1-mediated cell-to-cell transfer . In parallel , knockdown of Fascin in HuT-102 was verified in immunoblots and shown to be Fascin-specific since Tax protein and the housekeeping gene ß-actin ( ACTB ) were not affected by shFascin5 ( Fig 4C ) . Further , we excluded detrimental effects of the shRNA on cell vitality by measuring apoptosis and cell death in stable cell lines compared to cells treated with 15μM etoposide , which is known to induce cell death ( S5 Fig ) . Since results from co-culture assays may also reflect cell fusion events or the transfer of Tax-containing exosomes [51] , we decided to measure HTLV-1 infection also directly by a flow cytometry based assay that allows monitoring of newly infected cells . For this purpose , we first stably transduced the chronically HTLV-1-infected T-cell line MT-2 cells with lentiviral vectors expressing either shFascin5 or shNonsense . According to an established protocol [28] , MT-2 cells were then co-cultured with uninfected Jurkat T-cells for 1h and the number of newly infected Jurkat T-cells was detected by flow cytometry by measuring the amount of the viral matrix protein gag p19 in these cells ( Fig 4D and 4E ) . For this purpose , co-cultures were permeabilized and stained with antibodies targeting HTLV-1 gag p19 and the IL-2 receptor alpha chain CD25 , which is present on MT-2 cells , but not on Jurkat T-cells [52] . Flow cytometry revealed that repression of Fascin led to a significant reduction of newly infected , gag p19-positive Jurkat T-cells to 68% compared to the control ( Fig 4E ) . Beyond , western blot analysis confirmed a robust reduction of Fascin protein in MT-2 cells carrying shFascin5 ( Fig 4F ) , while Tax and ACTB were unaffected . Further , cell vitality was also unaffected by repression of Fascin as indicated by live/dead stainings ( S5 Fig ) . We also measured release of gag p19 into culture supernatants and found that knockdown of Fascin not only reduced infection of co-cultured Jurkat T-cells , but also diminished the release of gag p19 ( Fig 4G ) . Similar results were obtained with MT-2 cells treated with cytochalasin D or nocodazole ( 5μM each ) , which interfere with actin or tubulin polymerization , respectively ( Fig 4H ) . Overall , our observations are in line with the data we obtained with the HTLV-1 reporter vectors ( Figs 1–3 ) suggesting that , independent of the cell and test system used , repression of Fascin impairs release and cell-to-cell transmission of HTLV-1 . Immunoblot analysis revealed that cell-associated gag and processing of the gag p55 precursor into gag p19 and gag p27 was unaffected by knockdown of Fascin ( Fig 4I ) . However , treatment of MT-2 cells with the compounds cytochalasin D and nocodazole interfered with processing of gag p55 suggesting that chemical interference with the cytoskeleton acts differently on virus production than Fascin repression . Taken together , we found that Fascin is critical for release and cell-to-cell transmission of HTLV-1 reporter vectors , and for transactivation and infection of co-cultured T-cells indicating an important role of Fascin in HTLV-1 cell-to-cell transmission . To shed light on the mechanism of Fascin’s role during HTLV-1 transmission , we asked whether Fascin enhances conjugate formation between infected and uninfected T-cells similar to the small GTP-binding protein GEM [28] . For this purpose , we performed a flow cytometry-based conjugate formation assay between Jurkat T-cells that had been transfected with Tax and Fascin-specific shRNAs as donor cells , and Raji/CD4+ B-cells as acceptor cells according to a previously described protocol [25] . Briefly , Jurkat T-cells were co-transfected with one of two different Tax-expression constructs ( GFP-Tax; pEFTax ) and one of two different shRNAs encoding IRES-EGFP and targeting Fascin ( shFascin5 , shFascin4 ) or a control ( shNonsense ) . After 24h , cells were co-cultured with Raji/CD4+ B-cells for 1h and the percentage of conjugate formation between the two cell types was quantitated by flow cytometry ( S6 Fig ) . Detection of GFP encoded by GFP-Tax and/or the shRNA constructs was used to differ between transfected ( GFP+ ) and untransfected ( GFP- ) Jurkat T-cells . Conjugates of Jurkat T-cells ( CD3+ ) with Raji/CD4+ B-cells ( HLA-DR+ ) were identified in GFP+ ( Tax-positive ) and GFP- ( Tax-negative ) gates as double-positive signals ( HLA-DR+CD3+ ) and normalized on the total number of Jurkat T-cells . While Tax enhanced conjugate formation between the two different cell types confirming earlier observations ( Fig 5A; [25] ) , we found that Tax-induced cell aggregation was independent of Fascin ( Fig 5B ) . Western Blot analysis confirmed the expression of GFP-Tax , Tax , and the functionality of the Fascin-specific shRNAs ( Fig 5C ) . To validate these findings also in chronically infected T-cells , we asked whether Fascin is important for conjugate formation between HTLV-1-infected MT-2 cells and Jurkat T-cells , or whether Fascin contributes to adhesion of HTLV-1-infected cells on different attachment factors . For this purpose , HTLV-1-infected MT-2 cells with repressed Fascin ( shFascin5 ) and the respective controls ( shNonsense , untreated ) were co-cultured with Jurkat T-cells that had been pre-stained with the life cell dye Calcein-AM ( green ) . After 1h of co-culture at 37°C , cells were spotted on glass slides either coated with poly-L-lysine or fibronectin ( Fig 6A ) . Co-cultures were stained with antibodies targeting the viral matrix protein gag p19 ( blue ) to label HTLV-1-infected MT-2 cells and with antibodies targeting Fascin ( red; Fig 6B ) . Immunofluorescence revealed that gag was detectable in all experimental conditions , while expression of Fascin was repressed in MT-2 shFascin5 cells . Overlay of the respective channels and transmitted light showed that protrusive structures between chronically infected MT-2 cells and uninfected Jurkat T-cells could be detected ( Fig 6Bd and 6Be ) . The length of the protrusions was approximately 8 . 35μm+/-2 . 05μm ( on fibronectin ) or 6 . 16μm+/-2 . 24μm ( on poly-L-lysine ) . In most cells , Fascin and gag localized in close proximity ( Fig 6Bi ) , and occasionally , both proteins co-localized ( Fig 6Bd ) . This was further confirmed by determining fluorescence intensities of both Fascin and gag along arbitrary drawn ROIs ( regions of interest; S7a–S7g Fig ) : ( 1 ) Fascin and gag localize in close proximity , but do not co-localize ( ROI 1 ) . ( 2 ) The parallel shape of Fascin and gag fluorescence intensities at ROI 2 suggests that both proteins co-localize . Since co-localization events were only found in < 5% of all MT-2 cells , our findings suggest rather a transient or indirect than a tight and direct interaction between Fascin and gag during the dynamic process of virus budding . Next , automatic image analysis was performed to count the number of cell-cell contacts between infected MT-2 cells and uninfected Jurkat T-cells and to handle the large numbers of images and cells . An image processing algorithm was developed and applied that allowed for automatic quantitation of the respective cell types , and for counting of cell-cell contacts between infected and uninfected cells [47] . Automatic image analysis discriminated between Jurkat T-cells and MT-2 cells based on a cell segmentation approach using an active contour algorithm incorporating a priori shape information . The accuracy of cell segmentation determined on a single cell basis added up to aJ = 82 . 5% for Jurkat T-cells and aM = 77 . 8% for MT-2 cells . Correctly identified Jurkat cells were segmented with a hit quality hJ = 96 . 4% and correctly identified MT-2 cells with hM = 83 . 2% . In parallel , the micrograph images were checked manually to include cells into the evaluation that had been missed by the algorithm . Applying this algorithm , the following findings were obtained: ( 1 ) The number of cell-cell aggregates between HTLV-1-infected and uninfected T-cells is independent of the attachment factor and of Fascin ( Fig 6C ) , confirming our results from the flow cytometry-based assay ( Fig 5 ) . ( 2 ) Adhesion of HTLV-1-infected MT-2 cells is significantly impaired upon knockdown of Fascin ( Fig 6D ) , while cell vitality is unaffected ( S5 Fig ) . Thus , Fascin seems to be important for proper attachment of MT-2 cells on fibronectin- and poly-L-lysine-coated matrices and could favor dissemination of infected cells in vivo . Taken together , although Fascin seems to be required for proper attachment , Fascin does not affect the quantity of cell-cell contacts to uninfected Jurkat cells . Having found that knockdown of Fascin impairs release and cell-to-cell transmission of HTLV-1 , we performed imaging analysis to shed light on the localization of Fascin and of the viral gag protein ( Fig 7A ) . Most chronically infected T-cell lines harbor more than one copy of HTLV-1 provirus and produce large amounts of gag , which is unfavorable for imaging analysis . Therefore , we decided to analyze the chronically infected T-cell line MS-9 , which harbors only one integrated provirus and thus , reasonable amounts of gag to perform imaging analysis [26] ( Fig 7B ) . MS-9 cells ( Fascin-positive ) were co-cultured with Jurkat T-cells that had been pre-stained with the live cell dye Calcein-AM ( green ) and express only low amounts of endogenous Fascin . Cells were spotted on poly-L-lysine- and fibronectin-coated coverslips and incubated for 0 , 30 , or 60min at 37°C before fixation . Afterwards , cells were stained with antibodies targeting Fascin ( red ) and gag p19 ( blue ) , and confocal laser scanning microscopy was performed ( Fig 7A ) . Imaging revealed different patterns of Fascin localization at cell-cell contacts . First , we examined cells where the viral gag protein polarizes towards the uninfected target cell , suggesting the presence of the virological synapse ( VS ) [21] ( Fig 7Aa–7Aj ) . At cell-cell contacts we found not only clusters of gag ( Fig 7Ab; thin white arrows ) , which are reminiscent of viral biofilms [24 , 53] , but also clusters of Fascin protein ( Fig 7Ac ) . Concomitant with our previous findings in MT-2 cells ( Fig 6B; S7 Fig ) , we could rarely detect co-localizations between Fascin and gag . However , polarized gag clusters were interspersed with Fascin clusters ( Fig 7Ad ) at cell-cell contacts , which was confirmed by analysis of fluorescence intensities across the cell-cell-contact region ( S8 Fig ) . This suggests that Fascin makes room for gag clusters at the VS . Since large aggregates of HTLV-1 virions in the viral biofilm on the surface of infected cells are important for efficient infection of target cells [53] , Fascin could be important for the transport of viral proteins to the budding site , and thus , foster HTLV-1 transmission . This idea is supported by the quantitative evaluation of T-cell conjugates between MS-9 and Jurkat T-cells: If Fascin is localized at the cell-cell contact region , the frequency of polarized gag ( suggesting formation of the VS ) is 79 . 5% or 43 . 8% on poly-L-lysine or fibronectin , respectively . In contrast , if Fascin is dispersed and not accumulated at the cell-cell contact , the frequency of gag polarization ( at the VS ) is much lower ( 1 . 1% on poly-L-lysine , 1 . 8% on fibronectin ) . Thus , this suggests a direct role of Fascin in the local distribution of gag to budding sites and an indirect effect on cell-to-cell transmission . Beyond , we observed short , Fascin-containing membrane extensions that clutched uninfected T-cells ( Fig 7Ai; white-framed arrows ) and made room for these gag clusters ( Fig 7Ai; thin white arrows ) . Finally , we found long-distance connections ( approximately 15 . 21±5 . 48 μm in length on poly-L-lysine and 24 . 14±1 . 29μm on fibronectin ) between infected MS-9 cells and uninfected Jurkat T-cells ( Fig 7Ak–7Av; thick white arrows ) . The frequency of protrusions emanating from infected MS-9 cells was low ( approximately 3 . 28% of all MS-9 cells ) , however , protrusions were found independent of the time point of analysis ( Fig 7Ao and 7Au ) . Interestingly , we found Fascin and gag p19 protein expression in 65 . 3% or 79 . 5% of all protrusions ( on poly-L-lysine or on fibronectin , respectively ) . Among the Fascin-positive protrusions all except one were also stained positive for gag p19 . As depicted in Fig 7A , Fascin and gag partially co-localize in these cellular protrusions ( Fig 7At , 7Av and 7Aw ) , or the proteins are located in clusters in close proximity ( Fig 7An and 7Ap ) between infected MS-9 cells and newly infected Jurkat T-cells . Thus , formation of Fascin-containing protrusions could potentially account for the transfer of gag to target cells . To summarize our data , Fig 8 gives an overview of our current findings and provides a model of Fascin’s role in HTLV-1 transmission . HTLV-1-infected T-cells express the transactivator Tax that upregulates Fascin expression via the NF-κB signaling pathway . Not only Tax-induced Fascin , but also endogenous Fascin seems to be required for virus release and cell-to-cell transmission . Beyond , adhesion of infected cells in co-culture with uninfected cells occurs Fascin-dependently , which may favor dissemination of infected cells in vivo . Functionally , Fascin clusters intersperse with gag clusters suggesting that Fascin makes room for gag clusters reminiscent of viral biofilms at the VS . Furthermore , short-distance Fascin-containing membrane extensions clutch uninfected T-cells , which could favor the transfer of viral material to target cells via budding of enveloped virions at tight cell-cell contacts at the VS . Additionally , Fascin localizes with gag in long-distance connections between chronically infected and newly infected T-cells . It is conceivable that Fascin is required for the proper organization of protrusive structures , which may account for budding of HTLV-1 at the tip of the protrusion towards the target cell via a putative “mini VS” , a structure that had been proposed earlier [22 , 54] . Overall , our data suggest that Fascin could be important for the transport of viral proteins to foster polarized budding , virus release and cell-to-cell transmission of HTLV-1 . Thus , Fascin is an interesting novel target to inhibit HTLV-1 cell-to-cell transmission .
In this work we found that the actin-bundling protein Fascin is critical for HTLV-1 transmission . Fascin is known as a tumor marker , which is highly upregulated in many types of cancer and crucial for invasion and metastasis , potentially by stabilizing invasive structures [30] . We previously showed that Fascin is also important for invasive migration of virus-transformed lymphocytes [29 , 42] . Using different cell culture systems and infection models , we now found that repression of Fascin by shRNA or by Fascin-specific nanobodies severely impairs release and cell-to-cell transmission of the retrovirus HTLV-1 shedding new light on the function of Fascin . To address the role of Fascin in HTLV-1 cell-to-cell transmission , we made use of a single-cycle replication dependent reporter system , which allows automatic quantitation of productive infection in newly infected target cells only [25] . Using this system , our results indicate that Fascin is a major contributor of HTLV-1 cell-to-cell transmission independent of the cell type and the envelope type tested . Despite recent work showing that the reporter system we used even underestimates cell-to-cell transmission events [55] , we still see a significant reduction of reporter gene activity reflecting cell-to-cell transmission upon repression of endogenous or Tax-induced Fascin expression . However , new reporter vectors with improved splicing and packaging of the spliced reporter RNA might allow for better quantitating the role of Fascin on cell-to-cell transmission in lymphocytes and in primary cells in future work [55] . Our data suggest that both virus release and cell-to-cell transmission appear Fascin-dependent , while the amount of cell-associated virus ( as reflected by western blots of gag ) is not affected by repression of Fascin . However , our data also suggest that Fascin’s role in cell-to-cell transmission dominates over its role on virus release . Despite a significant impact on cell-to-cell transmission , virus release was not affected by Fascin-specific shRNAs in every experimental condition ( Fig 1C ) . This may be due to the fact that ELISAs measuring gag p19 also quantify non-infectious VLPs , while reporter assays ( Figs 1–3 ) and flow cytometry measuring gag p19 transfer ( Fig 4 ) quantitate newly-infected cells only . Although the release of virions is impaired upon knockdown of Fascin , cell-free virions carrying the wildtype env of HTLV-1 are severely impaired in infecting target cells ( S3 Fig; [25] ) . Thus , it is very likely that reduced infectivity of co-cultured target cells upon knockdown of Fascin results from cell-to-cell transmission , and not from infection with poorly infectious free viral particles . Thus , the impact of Fascin on direct cell-to-cell transmission could be underestimated . To confirm the relevance of Fascin for HTLV-1 transmission , we also investigated chronically infected T-cells . Both fusion-based assays and measuring of gag transfer to target cells confirmed the relevance of Fascin for HTLV-1 release and cell-to-cell transmission . Use of Fascin-specific nanobodies that target Fascin to the mitochondrial outer membrane ( MOM ) confirmed a role of Fascin in gag p19 release and HTLV-1 cell-to-cell transmission . Nanobodies are new promising stable recombinant antigen-binding domains of camelid heavy-chain antibodies that had already been successfully used to prevent Fascin-dependent invasion and migration of cancer cells [43 , 56] . Thus , nanobodies trigger Fascin protein loss of function without changing its expression [43] . Since the potent nanobody FASNb5 not only delocalizes Fascin to the MOM more efficiently , but also impairs Fascin-mediated organization of actin-bundles [43 , 56] , Fascin’s actin-bundling activity might be required for transmission of HTLV-1 . For HTLV-1 , the role of the actin cytoskeleton on virus transmission has not been analyzed in detail , however , it is known that polarization of the MTOC and transfer of HTLV-1 reporter vectors to target cells is impaired in presence of compounds interfering with actin polymerization [21 , 25 , 57] . We found that not only repression of the actin-bundling protein Fascin , but also interference with actin and tubulin polymerization led to reduced gag p19 levels in the supernatants of HTLV-1-infected T-cells . These observations are in contrast to HIV , where the budding process does not strictly rely on cytoskeleton remodeling . Although filamentous actin co-localizes with budding structures , inhibition of actin does not change localization of budding sites and packaging of actin and actin-binding proteins into virions seems to be a secondary consequence of the high abundance of these molecules at budding sites [58] . Yet , it is unknown , whether Fascin also contributes to release and cell-to-cell transmission of other viruses than HTLV-1 . We found that not only Tax [29] , but also latent membrane protein 1 ( LMP1 ) of Epstein-Barr virus ( EBV ) is a potent inducer of Fascin [42] . Interestingly , LMP1-deleted EBV is severely impaired in virus release into culture supernatants , potentially due to a defect in particle transport [59] . Thus , LMP1-mediated induction of Fascin and its continuous expression suggest a role of Fascin in virus release also of EBV . This is further corroborated by the finding that cell-to-cell transmission of EBV to epithelial cells also depends on canonical NF-κB signaling [60] , which is a prerequisite for efficient Fascin induction by LMP1 [42] . Although it is known that Tax is required for formation of the VS and efficient virus transmission [21] , only little is known about host factors that are regulated by Tax to modulate virus transmission [17] . With regard to pathways important for viral transmission , Tax transcriptionally alters the expression of cell adhesion and surface molecules , leads to cytoskeletal remodeling and complexes with proteins involved in cytoskeleton structure and dynamics . These Tax-interacting proteins include α-internexin , cytokeratin , actin , gelsolin , annexin , γ-tubulin and small GTPases of the Rho family [61] Two of these Rho-GTPases , Rac-1 and Cdc42 , complex with Tax and seem to be important for Tax-induced MTOC-polarization [57 , 62] . Thus , it is conceivable that Tax might connect Rho GTPases to their targets and affect cytoskeleton organization which could favor HTLV-1 transmission . Interestingly , Chevalier et al . found that GEM , which is an upstream negative regulator of ROCK-I Rho kinase , is induced by Tax [28] . GEM is a small GTP-binding protein and enhances cellular migration and conjugate formation between infected and uninfected T-cells . Knockdown of GEM in chronically infected T-cells reduces gag transfer to target cells showing that GEM is required for viral transmission [28] . It had been suggested earlier that not only GEM , but also Fascin and collapsin response mediator protein 2 ( CRMP2 ) , which is induced by Tax and important for migration [63] , might contribute to HTLV-1 transmission [17 , 28] . We now show that this is true for Fascin , however , the mechanism differs from the one described for GEM . Both GEM and Fascin are important for HTLV-1 cell-to-cell transmission , whereas only GEM is required for T-cell conjugate formation between infected and uninfected T-cells [28] . Contrary , Fascin also impairs virus release , which seems to be unaffected by GEM . Our findings that the adhesion of infected cells to different matrices is modulated by Fascin , in co-cultures with uninfected cells , could explain our previous observations where we found Fascin to be important for the invasion of ATL-derived cells through ECM and for the invasive migration of EBV-transformed and LMP-1-expressing lymphocytes [29 , 42] . These results are in line with recent data showing that Fascin forms a complex with focal adhesion kinase ( FAK ) and Src to control adhesion stability [31] Contrary to the test systems used in our manuscript , free viral particles of HTLV-1 are hardly detectable in vivo [20] . Viruses are transmitted at tight cell-cell contacts or via cellular protrusions protected from the host’s immune response [21 , 22] . It is estimated that HTLV-1 buds into a synaptic cleft and is transferred to target cells [21] . Moreover , viruses are tethered to and embedded in extracellular assemblies , viral biofilms , and transmitted at virological synapses to target cells [24] . It is likely that immune pressure and specific signals from uninfected target cells play a role in preventing release of HTLV-1 in vivo . Thus , it remains to be determined how Fascin affects HTLV-1 transmission in natural infection . Imaging revealed that Fascin clusters localize in close proximity to gag clusters at cell-cell-contacts , which are reminiscent of viral biofilms . Viral biofilms are carbohydrate-rich surface assemblies of viral particles which are composed of various components of the ECM and they account for the majority of HTLV-1 cell-to-cell transmission in vitro [24] . The localization of Fascin in close proximity to gag suggests that Fascin makes room for gag clusters at viral biofilms . Beyond , it is also conceivable that Fascin is required for formation , maintenance or tethering of viral biofilms , e . g . by redirecting the transport of viral and cellular proteins to budding sites via reorganization of the actin- or microtubuli-cytoskeleton [30 , 31] . Since Fascin is concentrated at cell-cell contacts , and localizes in close proximity to gag clusters , it is possible that Fascin may be packaged into HTLV-1 particles . We also observed short , Fascin-containing short membrane extensions clutching uninfected T-cells . These potentially support the transfer of virions to target cells , but presumably not due to enhanced conjugate formation , which remains unaffected by Fascin . Surprisingly , we observed Fascin and gag localization in long-distance protrusions between chronically infected and newly-infected T-cells . Long distance connections for the transfer of retroviruses or viral proteins have previously also been found in cells infected with MLV [64] or HIV [65] . For HTLV-1 , the viral p8 protein was identified as inducer of cellular protrusions [22] . Therefore , it remains to be determined , whether p8-induced protrusions are Fascin-dependent , and whether viruses bud from these protrusions at a “mini VS” to target cells . Taken together , our data suggest that Fascin could be important for the transport of viral proteins to budding sites , and thus , foster HTLV-1 transmission . However , the detailed mechanism of Fascin-dependent HTLV-1 transmission remains to be determined . Since repression of Fascin also reduces release of gag p19 into culture supernatants , it is conceivable that either the transport of viral proteins to the budding sites is impaired , or that viral particles are retained inside the infected cell , or at the viral biofilms . Since co-localization events between Fascin and gag were rare , our findings suggest a transient or indirect Fascin:gag interaction during the dynamic process of virus budding . Despite playing a crucial role in cell-to-cell transmission of HTLV-1 , it is not settled yet whether Fascin is also essential for formation of the VS . Localization of Fascin at cell-cell contacts and its association with a high frequency of polarized gag suggests that Fascin is involved in recruiting gag to the VS , and , thus , indirectly affects cell-to-cell transmission . However , it is unclear whether gag protein could localize at the VS in the absence of Fascin . These experiments are not accomplishable with chronically infected MT-2 cells , which can be manipulated by knockdown strategies , since these cells carry several proviral copies- some of them defective [66]- and excessive amounts of cell-associated gag protein . Fascin may also represent an interesting regulator of HTLV-1 cell-to-cell transfer in other cell types than infected T-cells . It is estimated that dendritic cells ( DC ) are the primary cells to be infected in vivo and that they play a pivotal role in transmitting the virus to CD4+ T-cells depending on cell-cell-contacts . Beyond , infection of DC may also be required for the establishment and maintenance of HTLV-1 infection in primate species [67] . Contrary to CD4+ T-cells , DCs are efficiently infected cell-free by highly concentrated viruses or by separated viral biofilms in vitro [68 , 69] . Since Fascin expression is selectively induced in mature DC [70] , important for the stability of dendrites and for formation of the immunological synapse [32] , future work should also investigate whether Fascin plays a role in dissemination of HTLV-1 from DC to T-cells . For a long time , Fascin has been known as an actin-bundling protein only . However , Fascin exerts other functions independent of its role in actin-binding and -bundling . Recent findings have supported this notion showing that Fascin also interacts with microtubules [31] . In light of HTLV-1 transmission , which depends on polarization of the MTOC and on proper actin and tubulin function [21 , 25] , our work identifying Fascin as a critical host factor in HTLV-1 transmission may provide a link between the activity of Tax and regulation of both the actin and microtubule cytoskeleton . Thus Fascin is a promising candidate to counteract HTLV-1 transmission .
|
Human T-cell leukemia virus type 1 ( HTLV-1 ) is the only human retrovirus causing cancer and is transmitted via breast feeding , sexual intercourse , and cell-containing blood products . Efficient infection of CD4+ T-cells occurs via polarized budding of virions or via cell surface transfer of viral biofilms at a tight , specialized cell-cell contact , the virological synapse ( VS ) . The viral protein Tax and polarization of the host cell cytoskeleton are crucial for formation of the VS , however , only little is known about the link between Tax and remodeling of the cytoskeleton to foster viral spread . The actin-bundling protein Fascin has evolved as a therapeutic target in several types of cancer . Here , we show that Fascin is also crucial for release and transmission of the tumorvirus HTLV-1 . Since Fascin is a transcriptional target gene of Tax in T-cells , our work provides a link between Tax’s activity and virus transmission . Visualization of cell-cell contacts between infected and uninfected T-cells suggests a role of Fascin in viral transmission potentially by facilitating the transport of viral proteins to budding sites . Thus , Fascin is not only crucial for metastasis of tumors , but also for transmission of HTLV-1 and is a new cellular target to counteract HTLV-1 .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"blood",
"cells",
"flow",
"cytometry",
"medicine",
"and",
"health",
"sciences",
"immune",
"cells",
"pathology",
"and",
"laboratory",
"medicine",
"293t",
"cells",
"pathogens",
"immunology",
"biological",
"cultures",
"microbiology",
"plasmid",
"construction",
"retroviruses",
"viruses",
"rna",
"viruses",
"dna",
"construction",
"molecular",
"biology",
"techniques",
"infectious",
"disease",
"control",
"research",
"and",
"analysis",
"methods",
"specimen",
"preparation",
"and",
"treatment",
"staining",
"infectious",
"diseases",
"white",
"blood",
"cells",
"spectrum",
"analysis",
"techniques",
"animal",
"cells",
"medical",
"microbiology",
"htlv-1",
"t",
"cells",
"microbial",
"pathogens",
"cell",
"lines",
"molecular",
"biology",
"spectrophotometry",
"antibody-producing",
"cells",
"cytophotometry",
"cell",
"staining",
"cell",
"biology",
"b",
"cells",
"viral",
"pathogens",
"biology",
"and",
"life",
"sciences",
"cellular",
"types",
"organisms"
] |
2016
|
The Tax-Inducible Actin-Bundling Protein Fascin Is Crucial for Release and Cell-to-Cell Transmission of Human T-Cell Leukemia Virus Type 1 (HTLV-1)
|
Transcription factor nuclear factor kappa B ( NF-κB ) regulates cellular responses to environmental cues . Many stimuli induce NF-κB transiently , making time-dependent transcriptional outputs a fundamental feature of NF-κB activation . Here we show that NF-κB target genes have distinct kinetic patterns in activated B lymphoma cells . By combining RELA binding , RNA polymerase II ( Pol II ) recruitment , and perturbation of NF-κB activation , we demonstrate that kinetic differences amongst early- and late-activated RELA target genes can be understood based on chromatin configuration prior to cell activation and RELA-dependent priming , respectively . We also identified genes that were repressed by RELA activation and others that responded to RELA-activated transcription factors . Cumulatively , our studies define an NF-κB-responsive inducible gene cascade in activated B cells .
The family of nuclear factor kappa B ( NF-κB ) transcription factors regulates inducible gene transcription in response to diverse stimuli . Signals from receptors ultimately activate inhibitor of NF-κB kinases ( IKKs ) that phosphorylate a variety NFKB inhibitors ( IκBs ) , targeting them for degradation and leading to accumulation of NF-κB family members in the nucleus . The “classical” pathway , via IKK2 activation , results in nuclear translocation of RelA- or Rel-containing NF-κB proteins , whereas the nonclassical pathway , via IKK1 activation , results in nuclear accumulation of RelB-containing NF-κB proteins [1 , 2] . Some signals only activate IKK1 ( such as B cell activating factor [BAFF]/BAFF receptor [BAFF-R] ) , others only activate IKK2 ( such as tumor necrosis factor alpha [TNFα] and interleukin 1 beta [IL-1β] ) , and yet others activate both IKK1 and IKK2 ( such as CD40/CD40L interaction ) . The cellular response to NF-κB activation therefore depends on the nature of the stimulus and the associated pattern of NF-κB proteins that are driven to the nucleus . Despite identification of a handful of well-accepted NF-κB target genes ( such as NFKBIA , TNFAIP3 , and MYC ) , genome-wide transcriptional responses mediated by NF-κB remain poorly defined . There are several reasons for this . First , the time course of NF-κB induction varies greatly depending on stimulus . For example , classical NF-κB activation by TNFα or IL-1β is rapid and transient , whereas activation via Toll-like receptor 4 ( TLR4 ) is slower and more sustained [1 , 3] . Second , NF-κB proteins consist of several family members . Nuclear factor kappa B subunit 1 ( NFKB1 ) , RELA , and REL proteins respond primarily by the classical pathway , whereas nuclear factor kappa B subunit 2 ( NFKB2 ) and RELB respond to nonclassical activation . Thus , a comprehensive analysis must include gene targets of each family member . This inherent complexity is compounded by observations that some genes that encode Rel family members , such as NFKB1 , NFKB2 , and RELB , are themselves targets of NF-κB . Third , NF-κB responses vary depending on the cell type as well as the initiating stimulus . Cell type specificity of NF-κB targets in monocytes/macrophages has been proposed to be conferred by regulated access of induced NF-κB to a subset of genomic sites by tissue-specific transcription factors [4–6] . Stimulus specificity has been explored largely in the context of TLR signaling and attributed to differences in dynamic patterns of NF-κB induction [7–9] . Yet , connections between NF-κB dynamics and transcriptional output are not well understood . For genes to be classified as NF-κB targets , they must change in expression as a consequence of NF-κB binding in cells that have received 1 or more NF-κB-inducing stimuli . This requires integrating at least 3 variables: transcriptional outcomes , NF-κB binding , and the consequences of abolishing NF-κB binding . The first 2 criteria have been explored by microarray or RNA sequencing ( RNA-Seq ) and by chromatin immunoprecipitation and sequencing ( ChIP-Seq ) to assay transcription factor binding genome-wide . For NF-κB , the majority of ChIP-Seq studies used TNFα as the NF-κB-inducing stimulus in HeLa cells or endothelial cells , with stimulus times ranging from 1 to 6 h . In response to TNFα , RELA bound to approximately 1 , 200–12 , 500 sites genome-wide in different studies , with the majority of binding occurring at sites other than gene promoters [10–19] . One of the earliest RELA ChIP-Seq studies also noted that the factor bound at many genes whose expression was unaffected by TNFα treatment [20] . Sites of RELA binding were enriched not only for the κB motif ( GGGRNYYYCC ) but also for recognition sites of other transcription factors , especially activator protein 1 ( AP1 ) . The latter observation corroborated previously reported interactions between NF-κB and AP1 [21 , 22] . Additionally , E2F and forkhead box M1 ( FoxM1 ) transcription factor motifs have been identified within RELA binding regions [20 , 23] . The RelA response has also been characterized in murine macrophages treated with lipopolysaccharide ( LPS ) for 1–3 h [4 , 5 , 24 , 25] . A substantial proportion of sites to which RelA was recruited in these cells were found to have prebound PU . 1 , a macrophage-enriched transcription factor . In addition , recognition motifs of interferon regulatory factors ( IRFs ) and AP1 were enriched at RelA-bound sites in macrophages . Such differences have been proposed to underlie selectivity of NF-κB responses . The third criterion , that of establishing that transcriptional outcome is a consequence of NF-κB binding , remains largely underexplored . The most precise way to causally connect binding events to gene expression requires mutating these sites in genomic DNA followed by transcriptional analyses . This is virtually impossible to do on a genome-wide scale . A more feasible , yet meaningful , alternative is to monitor transcriptional consequences of depleting the transcription factor , such as the use of RelA-deficient macrophages to validate the role of RelA in the LPS response [24] . Additionally , most studies do not account for the impact of dynamic patterns of RELA induction on inducible gene transcription . For this , both NF-κB binding and RNA levels must be interrogated at multiple time points in response to a stimulus . This is especially true for stimuli that activate NF-κB transiently . Here we probed transcriptional responses to a transient NF-κB-inducing stimulus by combining data from kinetic analyses of RELA binding , RNA polymerase recruitment , transcriptional output , and perturbation of classical NF-κB activation . Using BJAB B lymphoma cells , we demonstrate that NF-κB/RELA is transiently recruited to nearly 3 , 000 sites genome-wide in response to pharmacological mimics of B cell antigen receptor activation . From these sites , we identified several hundred genes that were direct transcriptional targets of NF-κB . Most of these genes were not found in databases of putative NF-κB target genes . Though the majority of functional NF-κB target genes were up-regulated by RELA , we also identified genes whose expression was suppressed by RELA binding . RELA target genes displayed different transcriptional kinetics , and most recruited RNA polymerase II ( Pol II ) in response to cell activation . In querying the basis for kinetic differences , we found that late-activated NF-κB target genes required extracellular signal–regulated kinase ( ERK ) activity , whereas rapidly induced NF-κB target genes were marked by Pol II–containing loops in unactivated cells . Despite relatively short activation times used in our experiments , we also identified many “indirect” targets of NF-κB . These genes appeared to be regulated by NF-κB-induced transcription factors and thereby represented downstream effects of NF-κB activation . Consequently , transcriptional responses of such genes were delayed compared to direct NF-κB target genes . Taken together , our studies define the first steps of an NF-κB-responsive inducible gene cascade in activated B cells and highlight mechanisms by which kinetic patterns of NF-κB-dependent gene induction are established .
NF-κB inducibility in BJAB human B lymphoma cells closely paralleled that seen in primary B cells activated via the B cell receptor ( BCR ) ( S1A Fig ) [26] . Hallmarks of this response were rapid nuclear translocation of RELA followed by exit from the nucleus within 4–6 h after stimulation . Over a 4 h time course of phorbol 12-myristate 13-acetate ( PMA ) and ionomycin ( P+I ) treatment , approximately 1 , 000 genes were up-regulated and 1 , 000 genes were down-regulated more than 2-fold in BJAB cells ( Fig 1A ) . We used k-means clustering based on correlation as the distance metric to partition up- and down-regulated genes into 6 distinguishable patterns ( Fig 1B , S1B Fig ) . Hierarchical clustering of RNA expression profiles also revealed categories similar to those determined by k-means analysis ( S1C and S1D Fig ) . Additionally , Gene Ontology ( GO ) analyses revealed distinct biological functions of genes in each category of up-regulated genes ( S1E Fig ) . Expression of the largest subset of genes increased over the 4 h time course ( Fig 1B , patterns 1A , 3A , and 4A ) . Smaller subsets of genes were rapidly up-regulated at 1 h and then either leveled off at 4 h or were subsequently down-regulated at 4 h ( Fig 1B , patterns 2A , 5A , and 6A ) . Analogously , most genes that were down-regulated by P+I decreased continuously from 0 to 4 h . Examples from these categories are shown in Fig 1C . Amongst these diverse patterns of altered expression , our goal was to identify genes that were responding directly to NF-κB activation . To identify genes that bound inducible NF-κB , we carried out ChIP-Seq using anti-RelA antibodies with unactivated cells or cells activated with P+I for 1 or 4 h . We focused only on those ChIP-Seq peaks that exceeded a threshold peak score of 100 ( after peak annotation in HOMER ) and were replicated in 2 independent ChIP-Seq experiments ( S1F Fig ) . We reasoned that these stringent criteria would increase focus on robust RELA interactions genome-wide , despite decreasing the total numbers of peaks being studied . We identified 345 RELA binding sites prior to cell activation . The number of RELA-bound sites increased to nearly 3 , 000 after 1 h of activation and thereafter fell back to approximately 600 sites at 4 h ( Fig 1D ) , demonstrating that genome-bound RELA closely paralleled total nuclear RELA levels . Sequences related to the κB motif ( recognition site of NF-κB ) were enriched at sites of RELA binding in all conditions ( S1G Fig ) . Most inducible RELA binding at 1 h occurred in parts of the genome annotated as introns and intergenic regions ( Fig 1D ) , a tendency that was noted in previous studies . We identified approximately 500 gene promoters that were newly targeted by RELA in activated cells . As a first step towards identifying functional targets of RELA , we used HOMER to associate RELA peaks with genes . Peaks that were located outside annotated gene promoters and introns usually fell within 50 kb of the transcriptional start sites ( TSSs ) of assigned genes [20] . Applying these criteria to the approximately 1 , 000 genes whose RNA levels increased ≥2-fold with activation , we found inducible RELA bound to 354 genes; conversely , RELA was associated with 201 ( out of 900 ) genes whose expression decreased ≥2-fold upon activation ( Fig 1E ) . The majority of RELA-binding up-regulated genes increased in expression between 1 and 4 h of activation ( Fig 1B , indicated in red ) . Conversely , expression of most RELA-binding down-regulated genes also decreased between 1 and 4 h , the period during which nuclear RELA levels were falling . Within the group of RELA-binding up-regulated genes were recognizable NF-κB target genes ( such as NFKBIA , TNFAIP3 , and RELB ) , as well as others that had not been previously associated with NF-κB ( such as RILPL2 ) ( Fig 1F ) . The 354 up-regulated and 201 down-regulated genes ( such as CYTH4 , Fig 1F ) with inducible RELA binding constituted a working list of putative NF-κB target genes in activated B cells . Only 106 out of 354 up-regulated genes and 36 of 201 down-regulated RELA-binding genes identified in our analysis were present in a list of 1 , 992 putative NF-κB-responsive genes compiled from the “NF-κB Target Genes” list maintained by Thomas Gilmore’s lab ( http://www . bu . edu/nf-kb/gene-resources/target-genes/ ) , NF-κB target gene sets ( https://www . yumpu . com/en/document/view/8327926/the-nfkb-target-gene-sets-are-listed-below-broad-institute ) , and other recent publications [24 , 27] . The newly identified 248 up-regulated and 165 down-regulated putative NF-κB targets are shown in S1 Table . To directly identify functional targets of inducible RELA ( that is , genes whose transcriptional changes depended on RELA binding ) , we attenuated classical NF-κB activation by expressing a degradation-resistant , mutated dominant negative IκBα ( dnIκBα ) [28 , 29] . This form of IκBα is expected to quench the release of NF-κB proteins from all cytosolic IκBs via the posttranslational pathway . For this , we generated 2 clones of BJAB cells in which dnIκBα could be induced by tetracycline ( Tet ) treatment ( Fig 2A ) . In these clones , nuclear RELA induction in response to P+I was similar to that of control BJAB cells in the absence of Tet but was abolished in cells that had been pretreated with Tet for 24 h ( Fig 2B ) . To determine the effects of dnIκBα expression on inducible gene expression , each clone was either pretreated with Tet for 24 h ( to induce dnIκBα ) or not , followed by activation with P+I for 0 , 1 , or 4 h . Replicate experiments were carried out with each clone , and total RNA was prepared for RNA-Seq . Basal gene expression was not affected substantially in the presence or absence of Tet ( S2A and S2B Fig ) . We compared inducible gene expression in each clone in the presence or absence of Tet and focused only on those inducible genes whose response to dnIκBα was replicated in both clones . We identified 806 genes whose inducible expression was significantly reduced ( false discovery rate [FDR] ≤ 0 . 05 ) by dnIκBα at either 1 or 4 h ( Fig 2C , middle ) in both Tet-inducible clones . Of these , 304 genes inducibly bound RELA in our ChIP-Seq experiments and were therefore considered to be direct transcriptional targets of RELA . These genes varied in their kinetic responses to cell activation ( Fig 2C , right and S2C Fig ) and were enriched for NF-κB binding motifs at sites of RELA binding as well as in their promoters ( S2D Fig ) . We found that NF-κB target genes with different kinetic expression patterns enriched for different biological functions ( S2F Fig ) . GO analyses showed that transiently activated genes ( pattern 3Ad ) were associated with processes such as “negative regulation of cellular processes , ” “leukocyte activation , ” and “inflammatory response . ” By contrast , processes that scored high among genes whose expression continued to increase between 1 and 4 h activation ( patterns 2Ad and 4Ad ) included “ribosome biogenesis , ” “immune response , ” regulation of “Type 1 interferon production , ” and “cellular response to cytokines . ” These observations indicate that kinetic patterns were associated with distinct functional categories of NF-κB target genes . Additionally , de novo motif analysis in HOMER revealed distinct transcription factor motifs associated with RELA peaks in different kinetic patterns . RELA peaks in patterns 2Ad and 4Ad genes were enriched for the κB motif as well as the motif for transcription factor AP1 , whereas the latter motif was not evident in RELA peaks of pattern 3Ad genes ( S2E Fig ) . This list of 304 NF-κB target genes included many that had not been previously identified as being NF-κB responsive ( S2 Table ) . To probe the NF-κB response , we focused on 130 of the 304 direct target genes that were induced more than 2-fold in the absence of Tet ( S2 Table ) . Out of these 130 genes , 74 have not been previously categorized as NF-κB targets . We found that virtually all transiently induced genes were amongst these 130 most robustly induced , RELA-binding , and dnIκBα-sensitive genes ( Fig 2C , right , red numbers ) . This category included genes such as TNFAIP3 and HERPUD1 ( Fig 2D , left ) . The prevalence of genes with this expression profile was consistent with reports showing that many NF-κB target genes have short-lived mRNAs [30] . For the majority of these genes , RELA binding occurred close to TSSs ( S4D Fig ) . Second , most ( 95 out of 130 ) of these genes were contained in the set of 354 putative target genes identified in Fig 1 . The remaining 259 ( out of 354 ) genes that were not substantially affected by dnIκBα , despite robust inducible RELA binding and altered gene transcription , reaffirmed the idea that inducible RELA binding and inducible transcription are insufficient criteria to identify functional targets of NF-κB . Third , mRNA levels of many genes continued to increase within the time frame of our studies ( Fig 2C , right , patterns 2Ad and 4Ad ) . These included genes such as STAT5A and NR1D1 ( Fig 2D , right ) . We found that RELA was recruited to these genes at 1 h , but most of it was lost by 4 h . These observations indicated that continued increase in mRNA was RELA-independent , suggesting that RELA was required to initiate but not maintain transcription of these genes . To further explore the basis for continued transcription of NF-κB target genes after chromatin-bound RELA was depleted , we drew upon the observation that the AP1 motif was enriched in RELA peaks associated with genes whose expression levels continued to increase between 1 and 4 h ( patterns 2Ad and 4Ad , S2E Fig ) . We tested the possible involvement of this transcription factor family by pharmacologically inhibiting ERK , a mitogen-activated protein kinase ( MAPK ) required for activation of AP1-like factors . RNA isolated from BJAB cells treated with P+I in the presence or absence of the ERK inhibitor PD0325901 was assayed by quantitative real-time PCR ( qRT-PCR ) for expression of genes from patterns 2Ad and 4Ad . We found that inducible expression of 3 genes from these categories was suppressed by ERK inhibition ( Fig 3A , top line ) . To rule out that PD0325901 affected NF-κB activation by some unanticipated pathway , we also assayed genes whose expression kinetics coincided with RELA induction by P+I ( pattern 3Ad in Fig 2C ) . These genes were not substantially affected by PD0325901 ( Fig 3A , lower line ) . We conclude that ERK-dependent transcription factors confer continued transcriptional activity to a subset of NF-κB target genes after induced nuclear RELA levels dissipate . The kinetically delayed response of these genes is consistent with a “priming” role for RELA , followed by an activator role for AP1-like factors . Such priming may involve recruitment or stabilization of additional transcription factors or coactivators by transiently bound RELA . RELA did not bind the remaining 502 genes whose expression was reduced by dnIκBα in both clones ( Fig 2C , middle , gray ) . Transcriptional responses of these genes in the presence or absence of dnIκBα also clustered into patterns similar to those seen for direct target genes , including 2 in which gene expression was inducibly down-regulated upon cell activation ( Fig 2C left , S3A Fig ) . A pattern that was prominently missing in this gene set compared to direct RELA targets was one in which RNA levels increased transiently in response to activation . To understand the basis for these genes being affected by dnIκBα in the absence of RELA binding , we looked for shared transcription regulatory features in this set . HOMER analysis of promoter regions of these genes revealed an enrichment for the binding motif of transcription factor MYC ( S3B Fig ) . The MYC motif was found in the promoters of 247 of these 502 genes , including BRIXI , DDX18 , and AKAP1 , and the majority of these genes ( 226 out of 247 ) were previously shown to bind MYC in ChIP-Seq assays [31] . Because MYC is a known target of NF-κB ( S3C Fig ) , we hypothesized that this set of genes was induced by NF-κB-activated transcription factors . Among 304 direct RELA target genes , we found 37 that encoded DNA-binding transcriptional regulators ( S3D Fig , left ) . These included genes for KLF10 and IRF1 that were previously linked to NF-κB . We also found many other genes encoding factors such as HES1 and ZNF267 that had not been previously associated with NF-κB ( Fig 3B ) . We confirmed dnIκBα-sensitive RELA recruitment to promoters of these genes by chromatin immunoprecipitation ( ChIP ) followed by quantitative PCR ( qPCR ) ( S3E Fig ) . Thus , many transcription factor genes were induced in activated cells via NF-κB . To further substantiate the hypothesis that NF-κB-induced transcription factors contributed to dnIκBα-sensitive gene expression , we evaluated whether promoters of indirect NF-κB target genes contained recognition motifs for NF-κB-regulated transcription factors . For the promoter analysis , we focused on 78 ( out of 502 ) genes ( S3 Table ) that were dnIκBα-sensitive , did not bind RELA , and were changed more than 2-fold by P+I treatment in the absence of Tet . We searched for transcription factor motifs that were present in more than 20% of these promoters ( S3F Fig ) . This analysis revealed recognition sites for kruppel-like factor ( KLF ) , zinc finger protein ( ZNF ) , and ETS-domain transcription factors . The correspondence between transcription factor genes induced by NF-κB and motifs enriched in promoters of indirect NF-κB target genes supports the notion that the set of dnIκBα-sensitive genes that did not bind RELA were targets of transcription factors activated by NF-κB . However , in the present study , we did not directly evaluate binding of such NF-κB-induced transcription factors genome-wide . We will refer to such genes as indirect targets of RELA . In keeping with their proposed dependence on NF-κB-induced transcription factors , RNA levels of such indirect targets increased at later times compared to direct targets ( Fig 3C ) . GO analyses of the dominant patterns ( 2Ai and 4Ai ) of indirect NF-κB target genes showed that they were enriched for genes involved in RNA processing , ribosome biogenesis , and RNA-associated metabolic processes ( S3G Fig ) . These processes were largely distinct from those associated with direct NF-κB target genes , thereby identifying a hierarchy of biological consequences associated with NF-κB activation . Because MYC has been implicated in activating ribosomal genes [32] , we surmise that many of the identified processes are the consequence of NF-κB-directed MYC expression in activated BJAB cells . Taken together , our kinetic analyses identified gene targets at which inducible RELA binding activated transcription ( direct targets ) and , additionally , revealed secondary transcriptional consequences of NF-κB activation in B lymphoblastoid cells via NF-κB-induced transcription factors ( indirect targets ) . We also uncovered a mechanism by which RELA induced persistent transcriptional activity despite its transient nuclear induction . We identified 263 gene transcripts that were up-regulated by dnIκBα expression in both Tet-inducible BJAB clones . Eighty-five of these genes bound RELA in activated cells ( Fig 4A , green circle ) ; our interpretation is that RELA binding reduced expression of this subset of genes . Such RELA-repressed genes had diverse expression profiles , including genes that were up- or down-regulated in response to activation ( Fig 4A right , patterns 6Rd and 1Rd , respectively , S4A Fig left ) . Some examples are shown in Fig 4B ( see also S4B Fig for complete RNA time courses ) . Of the 53 ( out of 85 ) RELA-repressed genes whose expression changed more than 2-fold in response to activation in the absence of Tet , 36 were not found in NF-κB-related databases and thus represent novel targets of NF-κB activity ( S4 Table ) . In contrast to RELA-activated genes that contained canonical κB motifs within RELA peaks , sequence motifs underlying RELA peaks of RELA-repressed genes were enriched for AP1 binding sites ( S4C Fig ) . Additionally , RELA binding was scattered throughout these genes rather than being enriched in promoter regions ( S4D Fig ) . Our interpretation is that RELA was recruited to these regions primarily by association with DNA-bound AP1 factors rather than direct DNA binding by RELA itself . Such interactions may attenuate transcriptional activation by AP1 , thereby resulting in gene repression . In total , 178 genes up-regulated by dnIκBα expression did not bind RELA ( Fig 4A , left; Fig 4C ) . We surmised that up-regulation of these genes by dnIκBα was also a secondary consequence of NF-κB activation . That is , such genes were either activated by transcription factors that were up-regulated by dnIκBα expression or attenuated by factors that were direct or indirect RELA targets . We found examples of each category in our RNA-Seq database . Among the 85 RELA-repressed genes , we found 23 that encoded transcriptional regulators ( S3D Fig ) whose increased expression in the presence of dnIκBa could be responsible for activating a subset of up-regulated genes that did not bind RELA . To further identify factors that indirectly activated gene transcription by dnIκBa , we screened promoter motifs present in 57 ( out of 178 ) genes whose expression changed more than 2-fold in the absence of Tet . Most of these overlapped with motifs present in genes that were indirectly activated by NF-κB ( S5 Table , S4E and S4F Fig ) . However , a few motifs were selectively associated with dnIκBα-activated genes , such as those for signal transducer and activator of transcription ( STAT ) and SRY-box ( SOX ) factors , and those for nuclear hormone receptors ( S4F Fig ) . As shown above , STAT5 is a direct target of NF-κB in these cells and may negatively regulate a subset of these indirect RELA-repressed genes . We also found SOX8 and PPARG mRNAs to be up-regulated by P+I in dnIκBα-expressing cells ( S4G Fig ) ; however , the associated mechanism ( s ) have not been further addressed in this study . We conclude that NF-κB activation also initiates a cascade of transcriptional down-regulation , both by directly interacting with a subset of genes and by modulating expression of other transcription factors . GO analysis of RELA-repressed genes revealed some interesting features . As noted for RELA-activated genes , there was relatively little overlap between the 6 patterns for the top 10 biological processes ( S4H Fig ) . Among genes that were directly repressed by RELA , we found 1 pattern ( 6Rd ) to enrich for genes involved in transcription termination by Pol II . One interpretation is that NF-κB proteins elevate gene expression by both activating transcription initiation and inhibiting transcription termination . The latter mechanism may apply to genes for which NF-κB has been proposed to push prebound RNA polymerase from abortive initiation state to productive elongation mode . Other pathways that featured in this gene set included modulation of biosynthetic and metabolic processes and negative regulation of cellular processes . Prominent among genes that were indirectly repressed by RELA were those involved in autophagosome organization and assembly and posttranslational protein modifications ( S4I Fig ) . By up-regulating essential autophagy genes such as ATG5 and 7 [33] and suppressing others involved in autophagosome assembly , NF-κB may fine-tune the autophagic response . The emerging patterns reveal synergistic use of RELA-dependent activation and suppression of gene expression to optimize cellular responses . The preceding analysis started by identifying genes whose inducible expression changed significantly in the presence of dnIκBα . While focusing on functional targets of RELA , this approach did not fully utilize our time-dependent RNA analyses . In particular , we missed out on the broader genomic landscape of NF-κB recruitment in response to B cell activation , especially where RELA binding did not affect mRNA levels after dnIκBα induction . Such sites were of potential interest because they vastly outnumbered those where RELA binding had functional consequences and may therefore contribute to B cell biology in unanticipated ways . We did this in 2 steps . First , we identified genes whose expression did not change in both dnIκBα-inducible clones across all time points of activation with or without Tet . Approximately 600 out of 8 , 000 such genes inducibly bound RELA ( S5A Fig ) . These binding sites were associated with canonical κB and AP1 motifs ( S5B and S5C Fig ) . Second , we used k-means clustering with correlation parameter to visualize gene expression patterns of the remaining genes in each clone in the absence or presence of dnIκBα ( S5D Fig ) . Patterns with similar expression characteristics in both clones were combined into 4 patterns of inducible gene expression ( S5E Fig ) . Patterns I and II corresponded to genes whose expression decreased or increased , respectively , in response to dnIκBα expression . Analysis of these gene sets largely recapitulated the conclusions from Figs 2–4 . Patterns III and IV provided insights into patterns of inducible expression that were not identified in the preceding analysis . These groups contained genes that were either up- ( Pattern III ) or down-regulated ( Pattern IV ) with activation but whose expression was not affected by dnIκBα ( S5E and S5F Fig ) . Numerous genes in each set bound RELA ( green circles ) , and NF-κB and AP1 binding sites were again enriched in sequences underlying RELA peaks ( middle column ) . These observations reinforced the idea from Figs 1 and 2 that inducible binding coupled with transcriptional changes was an insufficient criterion to identify functional targets of transcription factors . Differential regulation of these gene sets was also evident from their promoter architecture . Binding motifs for ETS-domain proteins and the transcription factor Yin Yang 1 ( YY1 ) were enriched in gene promoters that were up-regulated with activation ( S5E Fig , right column ) , whereas the motif for IRFs was enriched weakly amongst genes that were down-regulated with activation . Rapid gene induction in response to cell stimulation is programmed in different ways . In the classic example of c-Fos induction , phosphorylation of a promoter-bound transcription factor in unactivated cells triggers RNA synthesis after cell activation [34 , 35] . In other instances , RNA polymerases bound at promoters in unactivated cells can be pushed into elongation mode by phosphorylation of their C-terminal domain in response to stimuli [36 , 37] . This mode of activation has been implicated at some NF-κB-dependent target genes [38–40] . To gain more insight into inducible gene expression by RELA , we carried out ChIP-Seq with antibodies directed against Pol II . We used a threshold peak score of ≥100 in HOMER and reproducibility in replicate experiments to assign Pol II occupancy with confidence ( S6A Fig ) . Using these criteria , we found that 50 out of 130 direct NF-κB target genes contained prebound Pol II at their promoters prior to activation ( S6B Fig , S6A Table ) . However , these genes recruited additional Pol II after cell activation , which was evident in the average profile across all direct target genes ( S6C and S6D Fig ) . Proportionally fewer indirect target genes ( 13 out of 78 genes ) had prebound Pol II prior to P+I treatment , and inducible Pol II recruitment was clearly evident at these promoters in response to activation ( S6E Fig ) . We conclude that Pol II recruitment is a major mechanism of inducible gene transcription by NF-κB . Presence or absence of Pol II at the basal state did not readily explain kinetic differences in patterns of NF-κB target gene induction . To further probe for a possible mechanism , we performed chromatin interaction analysis by paired-end tag sequencing ( ChIA-PET ) in cells prior to stimulation . This assay scores for interaction of Pol II–bound sequences with other parts of the genome [41 , 42] . Biological replicates were processed using ChIA-PET tool software [43] , and the data were divided into 4 groups ( Fig 5A ) . Over the entire dataset , genes that lacked Pol II had the lowest RNA levels at baseline , genes with Pol II–bound promoters with no loops had intermediate RNA levels , and genes whose Pol II–bound promoters engaged in looping interactions had the highest levels of RNA ( S6F Fig ) . Similar trends were observed in previous ChIA-PET studies [42] . Additionally , we confirmed several looping interactions that had been identified in earlier studies , indicating that our assay scored for functionally relevant Pol II–associated interactions ( S6G Fig ) . We then examined chromatin interactions in the context of NF-κB target genes . Approximately half of the 130 robustly induced ( ≥2-fold ) direct RELA target genes had Pol II loops in unactivated cells ( Fig 5B , left; S6B Table ) . On average , RELA target genes with preformed loops reached close to maximum levels of inducible expression rapidly compared to those without loops ( Fig 5B , middle ) , and most of the transiently induced target genes identified in Fig 2 ( 22 out of 30 ) fell in this category ( Fig 5B , right , pattern 3Ad ) . Conversely , RELA target genes that did not have preformed loops were enriched for genes that were induced more slowly and whose expression increased continuously over the time course of activation ( Fig 5B , right , 34 out of 54 genes in pattern 2Ad and 23 out of 43 genes in pattern 4Ad ) . One example of a prelooped RELA target gene is shown in Fig 5D ( left ) . For indirect target genes , the trends were reversed ( Fig 5C ) . Approximately 30% of genes in this category ( 24 out of 78 robustly induced genes ) had preformed Pol II loops ( S6B Table ) . Looped genes in this set had higher basal RNA levels but did not show kinetic differences in RNA induction compared to nonlooped genes ( Fig 5C , middle ) . Instead , indirect target genes that had preformed loops in unactivated cells achieved higher levels of induced RNA compared to genes with no loops . Many indirect target genes with preexisting loops tended to be induced early in the presence of dnIκBα but crashed thereafter ( 18 prelooped genes in patterns 3Ai , 4Ai in Fig 5C , right ) . One example of a prelooped indirect RELA target gene is shown in Fig 5D ( right ) . We propose that preformed loops regulate kinetic patterns of direct RELA target genes , whereas they determine the maximal RNA output for indirect target genes . While RELA target genes were typically involved in single promoter interactions ( Fig 5A , pattern III ) , we also found approximately 1 , 000 genes that were involved in multiple promoter interactions ( ChIA-PET category 4 ) . These genes yielded GO terms such as “purine triphosphate metabolic process , ” “pyrimidine nucleotide biosynthetic process , ” and other comparable metabolic pathways ( S6H Fig ) . Amongst these , we found interactions involving PPP4C , ALDOA , and histone genes ( S6I Fig ) . It is likely that linking gene promoters via Pol II interactions provides a mechanism to coregulate genes that are involved in a common biological pathway [42] . In contrast , the need for greater flexibility in output of RELA-responsive genes depending on the stimulus and cell type may preclude their connection in a preformed network .
By combining time-dependent transcriptional responses with genome-wide recruitment of RELA and Pol II and perturbation of classical NF-κB activation , we sought to identify mechanisms by which kinetic patterns of NF-κB-dependent gene expression are established . In these studies , a pharmacologic equivalent of antigen receptor signaling was used to activate human B lymphoblastoid BJAB cells over a time course during which nuclear NF-κB was transiently induced . Three interesting features emerged from a consideration of RELA target genes . First , close to 60% of the 130 most robustly induced target genes identified here had not been previously connected with NF-κB responses . Our list also contained well-established NF-κB targets such as MYC , TNFAIP3 , and NFKBIA , attesting to the validity of our analyses . The incompleteness of current lists of NF-κB target genes was further accentuated when we included the additional 174 targets identified here that were induced less robustly . Of these 174 genes , 80% were not present in NF-κB-related databases , while the remaining 20% included genes such as TP53 , TNIP1 , and TAP1 that were previously linked to NF-κB . We surmise that NF-κB target genes identified here that are also present in earlier lists may represent more “universal” targets that respond regardless of cell type or stimulus . In contrast , genes uniquely identified in our study may represent cell type–or stimulus-specific responses . While use of a lymphoma cell line for these studies makes it difficult to draw direct connections to transcriptional responses of primary human B cells , the hundreds of new functionally curated NF-κB target genes identified here constitute a unique pool of possible mediators of NF-κB activity in B lymphoid cells . We also identified many genes that we refer to as indirect NF-κB targets . Such genes were sensitive to dnIκBa expression but did not bind RELA . We hypothesize that such genes were activated ( or repressed ) by NF-κB-induced transcriptional regulators . Even within the relatively short time course of our kinetic studies , we identified 37 genes among the 304 direct targets that encoded transcriptional regulators . In addition to previously identified targets such as MYC and IRF1 , this list included many new NF-κB-regulated factors such as hes family bHLH transcription factor 1 ( HES1 ) and ZNF267 . Additional studies are needed to directly evaluate the contribution of such factors in regulating indirect target gene transcription . It is possible that a subset of genes that we classified as indirect RELA targets are controlled by RELA bound to sites that did not score in the program used to connect binding sites to genes . From the studies presented here , we cannot specify the subunit composition of RELA-containing homo- or heterodimers that activate transcription of the identified NF-κB target genes . Because most of the RELA genome binding occurred at 1 h after cell activation , our working hypothesis is that functional NF-κB measured in these assays was generated from cytosolic pools by the classical posttranslational pathway . By electrophoretic mobility shift assays , this form consists largely of p50/RELA heterodimers; however , sequential ChIP is required to verify this model . Additionally , the contribution of REL to dnIκBa-sensitive gene transcription was not experimentally evaluated by depleting REL in BJAB cells . Studies to address this question are in progress using Rel-deficient murine B cells . Second , 2 dominant kinetic patterns of inducible expression emerged for the 130 genes that were most strongly induced . A small number ( 30 out of 130 ) were transiently induced and included genes such as TNFAIP3 and NFKBIA . The expression pattern of such genes can be easily explained by transcriptional activation when bulk RELA is nuclear ( at 1 h ) , followed by transcriptional inactivation when RELA moves back into the cytoplasm ( at 4 h ) , together with rapid degradation of the encoded mRNAs . The short half-life of many of these transcripts has been previously highlighted [30] . More surprising was the observation that inducible expression of the majority of these genes ( 97 out of 130 ) continued to increase between 1 and 4 h of activation . This time period coincided with down-regulation of RELA from the nucleus , and indeed , we found that RELA was lost from gene promoters over this period . For a subset of genes that we tested , ERK activity was required for sustained RNA synthesis after loss of nuclear RELA , pointing to involvement of the AP1 family of transcription factors . Our working hypothesis is that RELA binding “primes” the promoter for subsequent binding and transcriptional activation by ERK-dependent transcription factors . However , continued RELA binding is not required for transcriptional activity , thus distinguishing this mode of gene regulation from synergistic promoter activity by cobound factors . Our observations also provide a novel perspective on the phenomenon of assisted loading , a term used to describe cooperative recruitment of transcription factors that co-occupy gene regulatory sequences [44] . Regarding NF-κB/RelA , it has previously been shown that binding of IRF5 or STAT3 to a subset of genomic sites in activated hepatocytes or macrophages , respectively , requires simultaneous RELA activation [5 , 45] . In the examples presented here , we show instead that RELA does a hit-and-run on gene promoters that have a characteristic kinetic transcription profile . Although RELA is lost from these promoters , our experiments do not distinguish whether the κB site remains empty or is occupied by other factors . For example , p50 homodimers associated with IκBξ , which have been previously proposed to confer transcriptional activity [46] , may substitute RELA-containing complexes at such sites . Alternatively , other proteins that recognize κB motifs may provide transcriptional activity in the absence of RELA . Nuclear factor of activated T cell ( NFAT ) proteins are a distinct possibility because they are induced in P+I-activated B cells [47] , have been shown to bind to κB elements [48] , and function in collaboration with AP1 factors . Earlier studies have analyzed the connection between AP1/ERK and kinetics of gene induction by NF-κB . Natoli and colleagues showed that a subset of inflammatory gene promoters recruited RELA in response to LPS only after being marked by serine phosphorylated histone H3 ( H3S10P ) via p38 MAPK activity [49] . The 2-step process delayed RELA recruitment and transcriptional induction of these genes compared to other inflammatory genes to which RELA bound without requiring H3S10P . By contrast , RELA recruitment was not delayed even at late-induced genes in the studies described here , thereby invoking a novel mode of intersection between NF-κB and MAPK pathways . We note several differences that may underlie mechanistic variations observed for NF-κB-inducible transcription in the 2 studies . These include the different cell types used ( dendritic cells versus B cells ) that could differentially mark RELA recruitment sites , different initiating stimuli ( LPS versus P+I ) that induce distinct cytosolic signaling milieus , and the nature of NF-κB activation ( sustained versus transient ) that could influence gene expression outcomes . Further studies are needed to understand rules by which NF-κB tunes cellular responses to diverse stimuli . In a more recent study , Brasier and colleagues identified AP1 and SP1 motifs in regions surrounding RELA peaks in TNFα-induced A549 ( human pulmonary epithelial ) cells [14] . Genes with SP1 motifs reached maximal inducible expression 30 min after activation , whereas those with AP1 motifs reached maximal levels 60 min after activation . Two interesting features emerged from a comparison of our data with those of Yang and colleagues [14] . First , de novo motif analysis did not reveal SP1 sites near RELA peaks of our most rapidly induced genes ( Fig 2 , pattern 3Ad ) . Thus , rapid gene induction by NF-κB comes in different flavors . One possibility is that SP1 and NF-κB cooperate at promoters where inducibility via NF-κB is coupled with relatively high basal-level expression via SP1 . Second , in TNFα-treated A549 cells , RELA levels at a prototypical NF-κB/AP1 promoter continued to rise even when RNA levels were falling . By contrast , we found that at ERK-sensitive genes in activated BJAB cells , RELA levels fell , while RNA levels continued to rise . These distinctions yet again emphasize the variety of ways in which kinetic patterns of NF-κB-dependent transcription are achieved . Third , half of the strongly induced direct target genes had preformed Pol II–containing loops in unactivated cells . Genes that contained such loops were induced more rapidly on average than genes without loops and reached close to maximal expression levels at 1 h post activation . In contrast , RNA levels of unlooped genes continued to increase in the interval between 1 and 4 h . We propose that kinetic patterns of NF-κB-dependent transcription are determined in part by a poised state reflected in such preformed loops . Interestingly , most of the transiently induced targets ( 22 out of 30 ) fell in the looped category , likely reflecting the need for these genes to reach maximum expression as soon as possible while RELA is still in the nucleus . These genes were also more evolutionarily conserved than prelooped genes that were induced more slowly ( S6K Fig ) [50] . However , several transiently induced genes ( such as TNFAIP3 and NFKBIA ) did not have looped configurations in unactivated cells . One possibility is that these genes have “simple” NF-κB-dependent promoters that do not require interactions with distal regulatory sequences to modulate expression levels . Hao and Baltimore recently demonstrated that genes that are rapidly transcriptionally induced undergo rapid splicing to produce cytoplasmic mRNA [30] . Such mRNAs are also relatively unstable , resulting in transient gene induction . Five out of 7 genes that were shared between our dataset of transiently induced NF-κB target genes and that of Hao and Baltimore were found to have looped configurations in unactivated BJAB cells . Thus , a prelooped configuration may also assist in increasing splicing efficiency of rapidly induced genes . We note the caveat that the transformed state of BJAB cells may affect the distribution of genes with preformed loops . We also identified many genes whose inducible expression increased in the presence of dnIκBα . A subset of these genes bound RELA in our ChIP-Seq analyses , suggesting that RELA binding attenuated transcription of these genes . Identification of AP1 motif as the prominent sequence at sites of RELA binding leads us to hypothesize that RELA is recruited by protein–protein interactions with transcription factors bound at these sites rather than by DNA binding . In doing so , RELA may reduce transcription activation function of the DNA-bound factor . Interestingly , genes encoding several AP1 motif-binding factors , such as FOS , FOSB , MEF2B , and MEF2C , were in this set of RELA-repressed genes , possibly indicating some form of regulatory feedback . Dual-specificity phosphatase 1 ( DUSP1 ) , a regulator of ERK activity , was also in this list , again suggesting cross talk between NF-κB and AP1 signaling cascades . At other genes , DNA-bound RELA might interfere with the progression of RNA polymerases , thereby reducing transcriptional output . Though the mechanism and functional importance of RELA-dependent down-regulation of gene expression remain largely speculative at this time , our studies highlight a mode of gene regulation by this transcription factor that has been largely overlooked .
BJAB cells were cultured in RPMI 1640 medium supplemented with 10% FBS ( HyClone ) , Penicillin-Streptomycin-Glutamine ( Invitrogen ) , and 2-mercaptoethanol . For activation , cells were exposed to 50 ng/ml PMA and 2 μM ionomycin ( Sigma-Aldrich ) for 1 and 4 h . To inhibit ERK signaling , cells were pretreated with 0 . 33 nM PD0325901 ( Selleckchem ) for 1 h before P+I stimulation . To generate dnIκBα-inducible clones , BJAB cells were transfected with pcDNA6/TR ( Invitrogen ) to express Tet repressor ( TetR ) , and stable clones were selected in 15 μg/ml blasticidin ( Invitrogen ) for 6 d . Stable single clones with the highest levels of TetR expression were subsequently transfected with full-length dnIκBα ( S32A and S36A ) cloned into pcDNA4/TO . Stable clones were selected in the presence of both blasticidin and 600 μg/ml zeocin ( Invitrogen ) for 6 d . dnIκBα expression was induced with 1 μg/ml Tet ( Invitrogen ) for 24 h before P+I treatment . All cell lines were maintained at 37°C with 5% CO2 . The following antibodies were used for ChIP , ChIP-Seq , or western blot: RelA ( sc-372 , Santa Cruz ) ; Pol II ( Covance Cat # MMS-126R ) ; hnRNP ( sc-32301 , Santa Cruz ) ; IκBα ( sc-371 , Santa Cruz ) ; β-actin ( sc-47778 , Santa Cruz ) . Horseradish peroxidase–coupled goat anti-mouse IgG and goat anti-rabbit IgG ( Santa Cruz ) were used for immunoblotting . Proteins were separated by electrophoresis through 10% SDS-PAGE and electrophoretically transferred to nitrocellulose membrane ( Millipore ) . After blocking with 5% dried milk in Tris-HCl-buffered saline/0 . 05% Tween ( TBST ) for 1 h , membranes were incubated with primary antibodies overnight , washed in TBST , and incubated for 1 h with horseradish peroxidase–coupled secondary antibodies ( Santa Cruz ) . Proteins were detected by using the enhanced chemiluminescence ( ECL ) systems ( Pierce ) and Syngene Imaging System . ChIP experiments were performed as described by [51] . Briefly , cells were washed twice with PBS and then were fixed at room temperature with either 1% formaldehyde in PBS for 10 min ( for Pol II ChIP ) or 1 . 5 mM EGS ( Pierce Cat # 21565 ) for 30 min followed by 1% formaldehyde ( Sigma-Aldrich ) at room temperature for 15 min in PBS ( RELA ChIP ) . Reactions were quenched by adding glycine to a final concentration of 0 . 125 M , and cells were washed twice with cold PBS . Nuclei were isolated and lysed in buffer containing 50 mM Hepes-KOH , pH 7 . 5; 150 mM NaCl; 1 mM EDTA; 1% Triton X-100; 0 . 1% sodium deoxycholate; 0 . 1% SDS; and protease inhibitors . The crosslinked chromatin was subjected to fragmentation by sonication ( Branson Sonicator ) . ChIP was performed with 2 . 5 μg anti-RelA antibody ( Santa Cruz ) and 2 μg anti-Pol II antibody ( Covance ) prebound to 50 μl Protein A or G Dynabeads ( Invitrogen ) . Sonicated chromatin was added to antibody-bound beads and incubated at 4°C overnight . Beads were collected by centrifugation , washed , and incubated at 65°C for 4 h in elution buffer ( 50 mM Tris-HCl , pH 7 . 5; 10 mM EDTA; 1% SDS ) to reverse cross-linking . ChIP DNA was purified by phenol-chloroform extraction followed by ethanol precipitation . For qPCR quantitation of ChIP , the signal from gene-specific amplicon was compared to an amplicon from the H19 locus according to the equation RE = 2- ( CT ( target gene ) -CT ( H19 ) . Primer sequences are listed in Table 1 . For sequencing , adapters were ligated to the precipitated DNA fragments or the input DNA to construct a sequencing library according to the manufacturer’s protocol ( Illumina , San Diego , CA , United States ) . Adapters with a T overhang were ligated to the DNA fragments and size selected ( approximately 200–350 bases ) on a 4 . 5% agarose gel . Eighteen cycles of PCR amplification were performed to enrich for fragments with an adaptor on both ends . These samples were bound to an Illumina single-read Flowcell , followed by cluster generation on the Illumina Cluster Station and sequencing with Illumina Genome Analyzer ( GA-II ) . Two biological replicate ChIP-Seq experiments were carried out with each antibody . ChIP-Seq data are available on the GEO website ( http://www . ncbi . nlm . nih . gov/geo/ ) ( Accession number GSE117259 ) . Bowtie2 software [52] was used to map quality-filtered reads from demultiplexed FASTQ files to human genome assembly GRCh37/hg19 with the default options . RELA ChIP-Seq peaks were called using standard parameters in MACS 2 . 1 . 0 [53] with input as the control and activated samples as the treatment . Peaks were called at an FDR ≤ 0 . 05 . Peak annotation and motif finding were carried out with HOMER ( http://homer . ucsd . edu/homer/ ) . The HOMER program annotatePeaks . pl was used to annotate peaks with default parameters ( promoter regions were defined from −1 kb to +100 bp ) . In our analysis , most intergenic peaks were located within 50 kb of TSSs . Two biological replicate experiments were carried out , and peaks with peak score ≥ 100 that were common to both replicates were used for all further analysis . All samples were normalized to 10 million reads for visualization . The programs findMotifsGenome . pl and findMotifs . pl were used to identify transcription factor binding motifs within peaks or promoter regions ( −400 to +100 bp from TSSs ) . The program findGO . pl was used to assess the enrichment of various categories of gene function ( GO ) . Total RNA was extracted ( 2 × 106 cells ) using the Qiagen RNeasy Mini Kit ( Qiagen ) . cDNA was synthesized with the SuperScript First Strand Synthesis System ( Invitrogen Life Technologies ) . RT-PCR was performed in duplicates using the ABI PRISM 7000 ( Applied Biosystems , Carlsbad , CA , US ) . Expression of NFKB1 , IL4I1 , MAPK6 , RND1 , TNFAIP3 , and NFKBID was normalized to GAPDH mRNA on the same PCR plate . Relative expression ( RE ) of individual genes was calculated by the equation RE = 2- ( CT ( target gene ) -CT ( GAPDH ) ) . Primer sequences are listed in Table 1 . Total RNA purified from BJAB and transfected derivatives was used for bar-coded library preparation and sequencing at the Johns Hopkins Deep Sequencing & Microarray Core . Two independent dnIκBα-inducible single-cell clones were treated with Tet for 24 h or not , prior to activation with P+I for different times . This experiment was performed twice for each clone , and RNA was prepared for sequencing . RSEM [54] was used to align RNA-Seq reads to the human genome and to quantify transcript abundance . EBSeq [55] was used to compare the aligned reads from multiple conditions to find differentially expressed genes using a cutoff of FDR ≤ 0 . 05 . k-means analysis of RNA expression data was carried out in MATLAB using normalized read counts , with correlation as the distance metric , the number of times to repeat clustering set to 5 , and other parameters set to default . All samples were normalized to 1 million aligned reads for visualization . Integrative analysis of gene expression in relation to ChIP-Seq data was done by ngs . plot ( https://github . com/shenlab-sinai/ngsplot ) . RNA-Seq data are available on the GEO website ( http://www . ncbi . nlm . nih . gov/geo/ ) ( Accession number GSE117259 ) . Heatmaps of gene expression in S1C Fig ( left ) , S1D Fig ( left ) , S2C Fig ( left ) , S3A Fig ( left ) , and S4A Fig were generated using the package “gplots” in R program ( https://CRAN . R-project . org/package=gplots ) by log2-transformed normalized read counts after adding a pseudocount of 1 . Colors represent standardized gene expression for which each gene is standardized across samples to have zero mean and unit standard deviation . The row color bar marks the cluster membership of each gene from the previous k-means clustering results . In addition , hierarchical clustering was applied based on the standardized gene expression . The results are shown in S1C ( right ) and S1D ( right ) Fig , S2C Fig ( right ) , and S3A Fig ( right ) as heatmaps in which the rows are reordered by the hierarchical clustering results , and the row color bar represents the k-means clustering results . The silhouette for each gene in which the expression is up-regulated ≥2 fold ( 1 , 021 genes in Fig 1B ) in BJAB cells is based on correlation as the distance metric [56] . The silhouette value ranges from −1 to 1 , where a larger value means that the gene is better matched to its own cluster than the neighboring clusters . As a baseline , we permuted cluster memberships and calculated the silhouette , which is shown in S1B Fig as the random group . RNA Pol II ChIA-PET was performed as previously described [41 , 42] . A total of 109 BJAB cells were treated with 1 . 5 mM EGS ( Pierce Cat # 21565 ) for 30 min , followed by 1% formaldehyde at room temperature for 15 min and then neutralized using 0 . 2 M glycine . Chromatin was sheared by sonication , and anti-Pol II monoclonal antibody 8WG16 ( Covance , MMS-126R ) was used to enrich Pol II–bound fragments . A portion of ChIP DNA was eluted from antibody-coated beads for quantitation using Picogreen fluorimetry and for enrichment analysis using qPCR . Two biological replicate ChIP samples were used for ChIA-PET library construction [42] . ChIA-PET data analysis was carried out as described by [43] ( https://github . com/GuoliangLi-HZAU/ChIA-PET_Tool ) . Final peak calling was done at FDR ≤ 0 . 05 . After filtering for paired-end tag ( PET ) clusters ≥ 2 counts and FDR ≤ 0 . 05 , approximately 6 , 000 loops per replicate were obtained . The overlap between two replicates was 80%; data analysis was carried out from 1 replicate . ChIA-PET data are available on the GEO website ( http://www . ncbi . nlm . nih . gov/geo/ ) ( Accession number GSE117259 ) . All RNA-Seq , ChIP-Seq , and ChIA-PET data were visualized by preparing custom tracks for the WashU EpiGenome Browser [57 , 58] . All analyses were performed on Biowulf or Helix computer clusters at NIH . Evolutionary conservation of NF-κB binding sites was calculated using a method similar to Iwanaszko and colleagues [50] . Homologous genes were obtained from NCBI HomoloGene ( https://www . ncbi . nlm . nih . gov/homologene ) for human , chimpanzee , rhesus , cattle , dog , mouse , and rat , followed by identification of promoter sequence ( upstream 1 , 000 bp to TSS ) of each gene for each species using R package biomaRt [59] based on the Ensembl database [60] . Promoter sequences of the other species ( i . e . , chimpanzee , rhesus , cattle , dog , mouse , and rat ) were compared to the promoter sequence of human using R package msa [61] . We used R package TFBSTools [62] to identify conserved NF-κB binding sites between human and the other species . Four NF-κB family motifs ( i . e . , NFKB1: MA0105 . 2 , NFKB2: MA0778 . 1 , REL: MA0101 . 1 , and RELA: MA0107 . 1 ) obtained from JASPAR [63] were used for the analysis . In accordance with Iwanaszko and colleagues [50] , the threshold of the motif mapping score was set to 80% , and the conservation cutoff was set to 40% , based on a window size of 51 bp . To obtain the percentage of conserved NF-κB binding sites for each gene , the total number of NF-κB motif sites in the human genome was obtained based on the 4 NF-κB family motifs . The number of conserved NF-κB motif sites was calculated by comparing human and every other species , respectively . The percentage of conserved NF-κB binding sites is calculated as the ratio between the conserved NF-κB motif sites and the total number of NF-κB motif sites for each gene . In Fig 5B and 5C , data are represented as the mean with the SEM . Comparisons between two groups at each time point were assessed by a 1-way ANOVA Kruskal-Wallis test . A p-value of ≤0 . 05 was considered statistically significant .
|
The nuclear factor kappa B ( NF-κB ) family of transcription factors regulates cellular responses to a wide variety of environmental cues . These could be extracellular stimuli that activate cell surface receptors , such as pathogens , or intracellular stress signals such as DNA damage or oxidative stress . In response to these triggers , NF-κB proteins accumulate in the cell nucleus , bind to specific DNA sequences in the genome , and thereby modulate gene transcription . Because of the diversity of signals that activate NF-κB and the ubiquity of this pathway in most cell types , cellular outcomes via NF-κB activation must be finely tuned to respond to the initiating stimulus . One mechanism by which NF-κB-dependent gene expression is regulated is by varying the duration of nuclear NF-κB; some signals lead to persistent nuclear NF-κB , while others lead to transient nuclear NF-κB . Consequently , time dependency of transcriptional responses is a unique signature of the initiating stimulus . Here we probed mechanisms that generate kinetic patterns of NF-κB-dependent gene expression in B lymphoma cells responding to a transient NF-κB-activating stimulus . By genetically manipulating NF-κB induction , we identified direct targets of RELA , a member of the NF-κB family , and provide evidence that kinetic patterns are established by a combination of factors that include the chromatin state of genes prior to cell activation and cofactors that work with RELA .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Methods"
] |
[
"cell",
"binding",
"sequencing",
"techniques",
"cell",
"physiology",
"gene",
"regulation",
"regulatory",
"proteins",
"dna-binding",
"proteins",
"cloning",
"dna",
"transcription",
"transcription",
"factors",
"sequence",
"motif",
"analysis",
"molecular",
"biology",
"techniques",
"rna",
"sequencing",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"bioinformatics",
"proteins",
"gene",
"expression",
"molecular",
"biology",
"biochemistry",
"cell",
"biology",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences"
] |
2018
|
Transcriptional outcomes and kinetic patterning of gene expression in response to NF-κB activation
|
A fundamental challenge in human health is the identification of disease-causing genes . Recently , several studies have tackled this challenge via a network-based approach , motivated by the observation that genes causing the same or similar diseases tend to lie close to one another in a network of protein-protein or functional interactions . However , most of these approaches use only local network information in the inference process and are restricted to inferring single gene associations . Here , we provide a global , network-based method for prioritizing disease genes and inferring protein complex associations , which we call PRINCE . The method is based on formulating constraints on the prioritization function that relate to its smoothness over the network and usage of prior information . We exploit this function to predict not only genes but also protein complex associations with a disease of interest . We test our method on gene-disease association data , evaluating both the prioritization achieved and the protein complexes inferred . We show that our method outperforms extant approaches in both tasks . Using data on 1 , 369 diseases from the OMIM knowledgebase , our method is able ( in a cross validation setting ) to rank the true causal gene first for 34% of the diseases , and infer 139 disease-related complexes that are highly coherent in terms of the function , expression and conservation of their member proteins . Importantly , we apply our method to study three multi-factorial diseases for which some causal genes have been found already: prostate cancer , alzheimer and type 2 diabetes mellitus . PRINCE's predictions for these diseases highly match the known literature , suggesting several novel causal genes and protein complexes for further investigation .
Associating genes with diseases is a fundamental challenge in human health with applications to understanding disease mechanisms , diagnosis and therapy . Linkage studies are often used to infer genomic intervals that are associated with a disease of interest . Prioritizing genes within these intervals is a formidable challenge and computational approaches are becoming the method of choice for such problems . When one or more genes were already implicated in a given disease , the prioritization task is often handled by computing the functional similarity between a given gene and the known disease genes . Such a similarity can be based on sequence [1] , functional annotation [2] , protein-protein interactions [3] , [4] and more ( see [5] for a comprehensive review of these methods ) . When no causal genes are known , the prioritization is done by exploiting the modular view described above , comparing a candidate gene to other genes that were implicated in similar diseases . Approaches in the latter category are often based on a measure of phenotypic similarity ( see , e . g . , [6] , [7] ) between the disease of interest and other diseases for which causal genes are known . This is motivated by the observation that genes causing the same or similar diseases often lie close to one another in a protein-protein interaction network [3] , [5] . Lage et al . [7] score a candidate protein with respect to a disease of interest based on the involvement of its direct network neighbors in a similar disease . The protein and its high-confidence interactors are also suggested to form a putative protein complex that is related to the disease . Kohler et al . [8] group diseases into families . For a given disease , they employ a random walk from known genes in its family to prioritize candidate genes . Finally , Wu et al . [9] score a candidate gene for a certain disease based on the correlation between the vector of similarities of to diseases with known causal genes , and the vector of closeness in a protein interaction network of and those known disease genes . A recent follow-up work by Wu et al . introduces AlignPI , a method that exploits known gene-disease associations to align the phenotypic similarity network with the human PPI network [10] . The alignment is used to identify local dense regions of the PPI network and their associated disease clusters . The authors show the utility of their framework in causal gene prediction . Most of these methods focus on prioritizing independent genes; however , in many cases , mutations at different loci could lead to the same disease . This genetic heterogeneity may reflect an underlying molecular mechanism in which the disease-causing genes form some kind of a functional module ( e . g . , a multi-protein complex or a signaling pathway ) [7] , [11] . For example , Fanconi anemia is a heterogeneous syndrome for which seven of its causing genes are known to form a protein complex which functions in DNA repair [12] . Thus , good prioritizations could potentially lead to the inference of larger disease-related machineries , revealing important mechanistic insights on the disease of interest . While the above methods that integrate protein-protein interaction ( PPI ) information with a phenotypic similarity measure have been shown to outperform previous prioritization approaches , they are limited in their application . Specifically , both AlignPI and the method of Lage et al . consider only small localized regions of the PPI network and do not capture global network signals . The methods of Kohler et al . and Wu et al . tackle the prioritization task only , and do not suggest ways to reveal the protein modules that are affected in a given disease . In this work we tackle both challenges . We present a novel network-based approach for predicting causal genes and protein complexes that are involved in a disease of interest . The method , which is called PRINCE ( PRIoritizatioN and Complex Elucidation ) , generalizes the network-based approaches above by both considering the network signal in a global manner and going beyond single genes to the modules that are affected in a given disease . It receives as input a disease-disease similarity measure and a network of protein-protein interactions . It uses a propagation-based algorithm , a preliminary version of which appeared in [13] , to infer a strength-of-association scoring function that is smooth over the network ( i . e . , adjacent nodes are assigned similar values ) and exploits the prior information on causal genes for the same disease or similar ones . This process is illustrated in Figure 1 . This scoring is then used in combination with a PPI network to infer protein complexes that are involved in the given disease . We apply our method to analyze disease-gene association data from the Online Mendelian Inheritance in Man ( OMIM ) [14] knowledgebase . We test , in a cross-validation setting , the utility of our approach in prioritizing genes for all diseases with at least one known gene . We compare the performance of our method to two state-of-the-art , recently published methods [8] , [9] . In all of our tests PRINCE outperforms the other methods by a significant margin . We then use our method to associate protein complexes with disease . The complexes that we recover are shown to be highly coherent in terms of the function , expression and conservation of their member proteins . According to these measures the collection of protein complexes we infer significantly outperforms a previous collection suggested by Lage et al . [7] , in which each complex was limited to a protein and its direct interactors . Our complete set of predictions of gene- and protein-complex associations is available in the Supplementary Material ( Suppl . Datasets S1 , S2 , S3 ) . We demonstrate the power of PRINCE by studying in detail three multi-factorial diseases for which some causal genes have been mapped already: Prostate Cancer , Alzheimer Disease and Non-insulin-dependant Diabetes Mellitus ( Type 2 ) . For each disease we investigate PRINCE's top-10 predictions when considering the entire network , and when limiting the search to genomic intervals that have been associated with the disease . 69% of these predictions are validated in the literature ( using independent data ) , leaving 18 suggestions for novel causal genes .
In order to perform a comprehensive comparison of our approach to extant ones on the same input data , we reimplemented two state-of-the-art global approaches for gene prioritization introduced earlier: the random-walk based method of [8] and the Cipher algorithm [9] . We could not reimplement the method of Lage et al . [7] , as it has many parameters that had to be returned to fit our data , and a code for this method was not readily available . To evaluate the performance of the different methods , we used a leave-one-out cross validation procedure . In each cross-validation trial , we removed a single disease-protein association from the data , and each algorithm was evaluated by its success in reconstructing the hidden association , i . e . , by the rank it assigned to a protein when querying the disease it is associated with ( for further details on the cross-validation process see Methods ) . To simulate the case of prioritizing proteins encoded by genes inside a linkage interval , we followed [8] and artificially constructed for each protein associated with a disease an interval of size 100 around it . We evaluated the performance of an algorithm in terms of overall precision versus recall when varying the rank threshold . Precision is the fraction of true gene-disease associations that ranked within the top in the corresponding trial of the cross validation procedure . Recall is the fraction of trials in which the hidden association was recovered as one of the top scoring ones . We tested our method on all 1 , 369 diseases with a known causal gene in the OMIM database . The results of applying the different methods are depicted in Figure 2 . Our algorithm achieved the best performance , ranking the correct gene as the top-scoring one in 34% of the cases . Random-walk and Cipher methods achieved inferior results with 28 . 8% and 24 . 7% success rates , respectively . This trend was maintained when performing 2-fold , 5-fold and 10-fold cross validation ( Suppl . Figure S1 ) . Interestingly , even though our score does not directly indicate the probability of a successful prediction , we noticed a significant difference in the score distribution of top-1 correct predictions and top-1 incorrect prediction in the cross validation setting ( see Suppl . Figure S2 ) . Namely , about of our correct top-1 predictions received a score higher than , whereas about of our incorrect top-1 predictions received a score lower than that value . In the top-1 case , if all of the predictions with a score lower than are discarded , PRINCE's precision is boosted to , whereas the recall decreases to . To further validate the predicted associations , we collected recently published gene-disease associations that were not part of our original data set . We obtained 51 new associations for 47 diseases with previously known causal genes , and 10 new associations for diseases where the causal gene was unknown at the time of the original data collection . On the first association set , PRINCE ranked one of the newly associated genes as the top scoring one in 20 of the 47 diseases ( ) . On this set , PRINCE significantly outperforms CIPHER and compares favorably to Random Walk ( Suppl . Figure S3 ) , providing higher precision and recall for . On the second association set , PRINCE ranked the newly associated gene as the top scoring one in two of the ten diseases , and ranked the true causal gene higher than or equal to the other methods in four additional cases , thus providing the best average ranking ( Suppl . Table S1 ) . Having validated our method , we proceeded to execute our algorithm on specific multifactorial diseases that are linked to multiple genomic regions . We selected Prostate Cancer ( MIM: 176807 ) , Alzheimer's disease ( MIM: 104300 ) and Diabetes Mellitus , type 2 ( MIM: 125853 ) as our three case studies . We ranked candidate genes both over the entire PPI network , and over genomic intervals to which the phenotype has been mapped but no causal gene was identified , and analyzed our top-10 predictions in each case ( Suppl . Table S2 ) . We checked whether our predicted genes were already found to be involved with their query disease by searching online databases [14]–[16] and scientific publications . In all of the three test cases , the vast majority of top candidate genes over the entire network were already known to be involved with the disease . These often included the ‘usual suspects’ for the relevant disease family . For example , the top predictions for Prostate Cancer included BRCA1 , TP53 and NBN , which are tumor suppressors involved in several types of cancer . Over half of the top candidates from the associated intervals were already implicated in the corresponding diseases . Our ranking provides further support for their involvement in the investigated diseases . In addition , PRINCE yields several top scoring candidates that were not previously associated with these diseases , providing viable candidates for further research . Going beyond the above three test cases , we applied our algorithm to all 916 disorders in OMIM with an associated interval and for which no causal gene is known . The complete set of results is provided in the Supplementary Material ( Suppl . Datasets S1 ) . Often , as alluded to above , mutations in multiple proteins that form a protein complex or a pathway may lead to the same disease . Thus , we sought to exploit the prioritization function we have developed for the complex inference task . To this end , starting with the set of proteins whose prioritization score is above a threshold , we look for densely connected subsets that may form a protein complex . The search is aided by a likelihood-based scoring of protein complexes that takes into account the reliability of the PPI interactions and the degrees of the network proteins ( Methods ) . As we show in Suppl . Figure S4 and S5 , this score correlates well with the coherency of the identified complexes ( see below ) . Applying this scheme to the OMIM diseases we predicted 566 complexes for diseases in which a causal gene is known and 137 complexes for diseases for which only an associated genomic interval is known . To test the biological plausibility of the identified complexes we evaluated their coherency with respect to several attributes of their member proteins ( Methods ) . These measures quantify the extent to which proteins in a complex share the same functional annotation , have similar expression patterns under multiple conditions , and have similar phylogenetic profiles , respectively . As a baseline , we compared these measures with those computed for: ( i ) a set of manually annotated protein complexes obtained from the Gene Ontology ( GO ) annotation [17]; ( ii ) a set of protein clusters that are not necessarily disease-related , obtained by applying the MCL algorithm [18] to the PPI network; and ( iii ) a set of predicted disease-related complexes ( Lage et al . [7] ) ( Methods ) . To allow a fair comparison between our results and those of Lage et al . , we focused on a subset of the identified complexes of the same size as that provided by Lage et al . ( 80 for the case of a known causal gene , and 59 for the case of a known locus; Methods ) . The subset was constructed by computing the likelihood score of each complex and choosing the highest ranking complexes . We found that the complexes predicted using our propagation approach exhibited higher coherencies than the collection of Lage et al . with respect to most measures ( with the exception of conservation coherency in the known-locus case ) . Notably , both our collection and that of Lage et al . outperformed the PPI-based collection produced by MCL , demonstrating the importance of the disease association data in the protein complex inference process . Moreover , our results were comparable to , and in some cases better than , the manually curated collection , again testifying to its high quality . These results are summarized in Table 1 . As a further validation of the complexes inferred by PRINCE , we searched OMIM for evidence for the possible involvement of the proteins of a complex in the diseases associated with it . Specifically , for each complex , we scanned the OMIM entries of the diseases associated with at least one complex member . For each such disease , we checked whether any complex member that is not known to be associated with that disease , is mentioned in its entry . We found such support for 61% of the predicted complexes , with an average of 3 . 6 genes per complex whose involvement was corroborated in this manner . For comparison purpose , we permuted the gene names and repeated the analysis on the resulting random complexes . Only 7% of these random sets were supported by OMIM , with an average of 1 . 6 evidences per set . Three example putative protein complexes and their associated diseases are shown in Figure 3 . The first putative protein complex ( Figure 3 ( a ) ) was generated for the query disease Ataxia-Telangiectasia ( MIM:208900 ) , which is associated with the gene ATM . The putative complex contains 11 proteins which are all known to be involved in response to DNA damage stimulus . Except for CHEK2 , all of them are directly involved in DNA repair . All 7 diseases associated to those genes ( among them are Breast Cancer , Li-Fraumeni syndrome and Fanconi Anemia ) are known to be tightly coupled with mutations in DNA-repair related genes . In this specific case it may be that these proteins do not form a single complex in-vivo , but rather span a dense region of the PPI network due to their central role as master regulators ( especially ATM and TP53 ) of reactions to DNA damage . The second complex ( Figure 3 ( b ) ) was generated for the query disease Hereditary Prostate Cancer type 8 ( HPC8 , MIM:602759 ) , for which the causal gene is presently unknown . The complex's proteins are associated with several Colorectal Cancer variants and Endometrial cancer . The genes associated with the Colorectal and Endometrial cancers are from the MLH ( MutL analog ) and PMS families which are involved in DNA mismatch repair . MLH1 and PMS2 form a Heterodimer , which interacts via MLH1 with EXO1 ( Exonuclease1 ) , which also participates in DNA mismatch repair . The gene coding for EXO1 is located at genetic locus 1q43 , which lies within the region associated with HPC8 ( 1q42 . 2–q43 ) . Moreover , EXO1 was ranked first by PRINCE in this interval . In this case , the inferred protein complex provides support also to the prediction that EXO1 is a causal gene for prostate cancer ( MIM: 176807 ) discussed in the previous subsection . The last complex ( Figure 3 ( c ) ) was generated for the query disease Microcephalic Osteodysplatic Primordial Dwarfism ( MOPD-I , MIM:210710 ) , which has no known causal genes . Two of the predicted complex's genes are associated with two hereditary diseases characterized by developmental delay and physical deformations: ERCC5 with Cockayne Syndrome type A , and ERCC2 with Cerebrooculofacioskeletal Syndrome 1 . The genes in the complex are all involved in DNA damage repair: ERCC2 , ERCC3 , GTF2H1 and GTF2H2 are subunits of the core-TFIIH basal Transcription Factor , and ERCC5 forms a stable complex with TFIIH enabling recruitment of the Transcription Factor for repairing UV damage [19] . ERCC3 , one of the predicted complex's members , lies within the genetic locus associated with MOPD-I , and is ranked as the best causal gene candidate for MOPD-I among the genes at that locus by PRINCE . PRINCE is a powerful method for prioritizing genes and protein complexes for a disease of interest . We have demonstrated its power both in a cross validation setting and by closely examining its predictions over complex , polygenic hereditary diseases . Key to its successful application is its global network approach , combined with a novel normalization of protein-protein interaction weights and disease-disease similarities . While the results of PRINCE are promising , several of its limitations should be acknowledged . First , PRINCE relies on prior phenotypic information , which limits its application to diseases that are phenotypically similar to diseases with known causal genes . Second , PRINCE uses known gene-disease associations in its computation , but other relevant data , such as genes that are differentially expressed in the disease state , are not taken into account . Combining such data into the prioritization process , e . g . , using the method of [20] , could increase the prediction power . Last , PRINCE depends on accurate and comprehensive protein-protein interaction data . As such data accumulate , the applicability and accuracy of PRINCE are expected to grow .
The input to a gene prioritization problem consists of a set of gene-disease associations; a query disease ; and a protein-protein interaction network , where is the set of proteins , is the set of interactions and is a weight function denoting the reliability of each interaction . The goal is to prioritize all the proteins in ( that are not known to be associated with ) with respect to . For a node , denote its direct neighborhood in by . Let represent a prioritization function , i . e . , reflects the relevance of to . Let represent a prior knowledge function , which assigns positive values to proteins that are known to be related to , and zero otherwise . Intuitively , we wish to compute a function that is both smooth over the network , i . e . , adjacent nodes are assigned with similar values , and also respects the prior knowledge , i . e . , nodes for which prior information exists should have similar values of and . Formally , we express the requirements on as a combination of these two conditions: ( 1 ) where is a normalized form of ( described below ) . The parameter weighs the relative importance of these constraints with respect to one another . The requirements on can be expressed in linear form as follows: ( 2 ) where is a matrix whose values are given by , and and are viewed here as vectors of size . We require the eigenvalues of to be in . Since , the eigenvalues of are positive and , hence , exists . While the above linear system can be solved exactly , for large networks an iterative propagation-based algorithm works faster and is guaranteed to converge to the system's solution . Specifically , we use the algorithm of Zhou et al . [21] which at iteration computeswhere . This iterative algorithm can be best understood as simulating a process where nodes for which prior information exists pump information to their neighbors . In addition , every node propagates the information received in the previous iteration to its neighbors . We chose to normalize the weight of an edge by the degrees of its end-points , since the latter relate to the probability of observing an edge between the same end-points in a random network with the same node degrees . Formally , define a diagonal matrix such that is the sum of row of . We set which yields a symmetric matrix where . Note that is similar to the stochastic matrix . Since similar matrices have the same eigenvalues , and since a stochastic matrix's eigenvalues are in ( according to the Perron-Frobenius theorem ) , the eigenvalues of are indeed in . To determine the prior information vector , we used the similarity metric computed by van Driel et al . [6] . This metric spans diseases in the OMIM [14] knowledgebase and is based on their medical subject headings description . van Driel et al . tested the predictive power of different ranges of similarity values by calculating the correlation between the similarity of two diseases and the functional relatedness of their causal genes . According to their analysis , similarity values in the range are not informative , while for similarities in the range the associated genes show significant functional similarity . These empirical findings motivated us to represent our confidence that two diseases are related using a logistic function , such that for , , and for , . This implies that needs to be close to . We set , which determines as , and tuned the parameter using cross validation ( see Parameter Tuning Section below ) . We used to compute the prior knowledge in the following way: for a query disease and a protein associated with a disease , we set , where is the similarity between and . If is associated with several diseases , we choose the disease which is the most similar to . We extracted known disease-protein associations from GeneCards [15] spanning diseases and proteins . We considered only disease-protein relations that included proteins from the network and such that the relations are known to be causative to avoid associations made by circumstantial evidence . We constructed a human PPI network with proteins and interactions that were assembled from three large scale experiments [22]–[24] and the Human Protein Reference Database ( HPRD ) [25] . The interactions were assigned confidence scores based on the experimental evidence available for each interaction using a logistic regression model adapted from [26] . We used the obtained scores to construct the adjacency matrix . To simulate the case of prioritizing proteins encoded by genes inside a linkage interval , we followed [8] and artificially constructed for each protein associated with a disease an interval of size 100 around it . We used the protein scores obtained from the output of the algorithm to prioritize proteins residing in that interval . To evaluate the performance of the different methods in predicting gene-disease association , we used a leave-one-out cross validation procedure . In each cross-validation trial , we removed a single disease-protein association from the data , and in addition all other disease-protein associations involving protein . An algorithm was evaluated by its success in reconstructing the hidden association , i . e . by the rank it assigned to protein when querying disease . The reason we hid all associations of was to avoid “easy” cases in which is also associated with other diseases that are very similar to . Our algorithm has three parameters that should be tuned: ( i ) – the parameter controlling the logistic regression transformation; ( ii ) – controlling the relative importance of prior information in the association assignment; and ( iii ) the number of propagation iterations employed . We tested the effect of these parameters on the performance of the algorithm in a cross validation setting . The precision-recall plots for the general disease case are depicted in Suppl . Figure S6 . By this figure , the optimal regression coefficient is , implying that similarity values below 0 . 3 are assigned with very low probability ( ) , in accordance with the analysis of [6] . The algorithm is not sensitive to the actual choice of as long as it is above 0 . 5 ( panel b ) . Finally , the algorithm shows fast convergence , achieving optimal results after ten iterations only ( panel c ) . The random-walk based approach requires disease grouping information . To allow it to run on the more comprehensive disease similarity data we had , we generalized the approach to use these similarities ( transformed by the logistic function ) as initial probabilities for the random walk . The parameter of the method , which controls the probability for a restart , as well as our transformation parameter , were optimized using cross-validation ( as in the Parameter Tuning Section above ) . Note that Kohler et al . suggested a second , diffusion-kernel based approach , which was shown to be inferior to the random walk one , hence we did not include it in our comparison . Also note that our propagation-based method reduces to a random walk under appropriate transformations of the edge weights and prior information . The Cipher method [9] is based on computing protein closeness in a PPI network . Two variants of the algorithm were suggested: Cipher-DN , which considers only direct neighbors in the closeness computation , and Cipher-SP , which is based on a shortest path computation . The former was shown to outperform the latter , and hence we implemented this variant ( Cipher-DN ) only . Given a disease and a prioritization score for all the network proteins , we aim at inferring densely connected protein complexes that contain high scoring proteins . To this end , we start with the top 100 scoring proteins within the entire network as complex seeds ( The method is not sensitive to the number of initial top scoring proteins , and produces similar results for numbers in the range 50–150; data not shown ) . We filter all seeds whose score is below a prespecified threshold or that were already associated with the disease in a previously detected complex . To each seed we iteratively add a neighboring protein with the highest score , as long as this score is greater than , and up to 20 proteins per seed ( about twice the average size of known protein complexes; a similar bound was used in previous works [26] , [27] ) . At this stage , in the case that the query disease has no known gene , but has an interval associated to it , the computed complex is discarded if it contains no member from that interval . After an initial list of putative complexes is formed , a refinement phase takes place where proteins are removed from a putative complex to ensure that not only is the suggested complex disease-related but also its member proteins are densely interacting and , thus , constitute a good candidate for a complex . To this end , we use the following likelihood-based scoring scheme taken from [28]:where is a putative complex and are its edges . Briefly , the score is the log likelihood ratio between a protein complex model ( assuming that every two proteins in a complex should interact with a high probability , independently of all other pairs ) and a random set model ( where connections in the sub-network arise at random , with a probability proportional to the proteins' degrees ) . This score was further enhanced , as in [28] to accommodate for information on the reliability of interactions . In brief , the interaction status of every protein pair was treated as a noisy observation , and its reliability was combined into the likelihood score . The parameter of the scoring scheme was set to 0 . 9 , although results were not sensitive to the actual parameter used as shown in Suppl . Table S3 . At each refinement step , we search for a protein whose removal increases the score the most while maintaining the connectivity of the candidate complex . The refinement is done until no score increase is possible ( while maintaining connectivity ) . We filter candidate complexes with less than four proteins ( to ensure statistical significance ) or with overlap with higher-scoring candidates . For identifying complexes we use the same and values we used for prioritization , which were tuned using cross-validation . An additional parameter , , is used as a threshold that defines the minimal score ( computed using propagation ) needed for a protein to be included in any identified complex . This parameter was tuned separately for the case in which a causal gene for the query disease is known and for the case that no causal gene is unknown . The tuning aimed to obtain a collection of complexes whose average size is similar to that of the manually curated GO complexes ( 8 . 85 after filtering complexes with or ) . The resulting value of is 0 . 1 ( average size of 8 . 3 ) for the case where a causal gene is known , and 0 . 015 ( average size of 8 . 5 ) for the case where no causal gene is known . We compared the protein complexes inferred by PRINCE to three other collections: Following [29] , we evaluated the different collections using three coherency measures: The computational experiments were executed on a single core of an AMD Opteron ( tm ) 2382 processor 2 . 6 Ghz . The average runtime for completing the cross validation iterations or inferring protein complexes was 1–2 minutes . The code and data sets described herein are available upon request .
|
Understanding the genetic background of diseases is crucial to medical research , with implications in diagnosis , treatment and drug development . As molecular approaches to this challenge are time consuming and costly , computational approaches offer an efficient alternative . Such approaches aim at prioritizing genes in a genomic interval of interest according to their predicted strength-of-association with a given disease . State-of-the-art prioritization problems are based on the observation that genes causing similar diseases tend to lie close to one another in a network of protein-protein interactions . Here we develop a novel prioritization approach that uses the network data in a global manner and can tie not only single genes but also whole protein machineries with a given disease . Our method , PRINCE , is shown to outperform previous methods in both the gene prioritization task and the protein complex task . Applying PRINCE to prostate cancer , alzheimer's disease and type 2 diabetes , we are able to infer new causal genes and related protein complexes with high confidence .
|
[
"Abstract",
"Introduction",
"Results/Discussion",
"Methods"
] |
[
"neurological",
"disorders/alzheimer",
"disease",
"computational",
"biology/systems",
"biology",
"oncology/prostate",
"cancer",
"genetics",
"and",
"genomics/disease",
"models",
"genetics",
"and",
"genomics/genetics",
"of",
"disease",
"computational",
"biology/genomics",
"genetics",
"and",
"genomics/bioinformatics",
"diabetes",
"and",
"endocrinology/type",
"2",
"diabetes"
] |
2010
|
Associating Genes and Protein Complexes with Disease via Network Propagation
|
Highly pathogenic avian influenza ( HPAI ) H5N1 was first encountered in 1996 in Guangdong province ( China ) and started spreading throughout Asia and the western Palearctic in 2004–2006 . Compared to several other countries where the HPAI H5N1 distribution has been studied in some detail , little is known about the environmental correlates of the HPAI H5N1 distribution in China . HPAI H5N1 clinical disease outbreaks , and HPAI virus ( HPAIV ) H5N1 isolated from active risk-based surveillance sampling of domestic poultry ( referred to as HPAIV H5N1 surveillance positives in this manuscript ) were modeled separately using seven risk variables: chicken , domestic waterfowl population density , proportion of land covered by rice or surface water , cropping intensity , elevation , and human population density . We used bootstrapped logistic regression and boosted regression trees ( BRT ) with cross-validation to identify the weight of each variable , to assess the predictive power of the models , and to map the distribution of HPAI H5N1 risk . HPAI H5N1 clinical disease outbreak occurrence in domestic poultry was mainly associated with chicken density , human population density , and elevation . In contrast , HPAIV H5N1 infection identified by risk-based surveillance was associated with domestic waterfowl density , human population density , and the proportion of land covered by surface water . Both models had a high explanatory power ( mean AUC ranging from 0 . 864 to 0 . 967 ) . The map of HPAIV H5N1 risk distribution based on active surveillance data emphasized areas south of the Yangtze River , while the distribution of reported outbreak risk extended further North , where the density of poultry and humans is higher . We quantified the statistical association between HPAI H5N1 outbreak , HPAIV distribution and post-vaccination levels of seropositivity ( percentage of effective post-vaccination seroconversion in vaccinated birds ) and found that provinces with either outbreaks or HPAIV H5N1 surveillance positives in 2007–2009 appeared to have had lower antibody response to vaccination . The distribution of HPAI H5N1 risk in China appears more limited geographically than previously assessed , offering prospects for better targeted surveillance and control interventions .
HPAI H5N1 virus infection was first encountered in China in 1996 in the southern part of the country with the discovery of a virus that killed geese in Guangdong province ( Goose/GD/96 ) [1] . In 1997 , Hong-Kong experienced the first major outbreak of HPAI H5N1 associated with several human deaths , alerting the international community to the potential threat caused by this new strain of HPAI virus ( HPAIV ) . Between 1999 and 2003 , the virus underwent a series of evolutionary changes and multiple genotypes of HPAIV H5N1 detected through routine live bird market surveillance in southern China emerged , indicating that the virus was still active and widely circulating [2] . However , the first major outbreak of HPAI H5N1 in mainland China started in January 2004 in Guangxi autonomous region , in southern China , bordering Vietnam . As the outbreak unfolded , the disease was detected widely throughout the country , causing over 110 outbreaks in 23 provinces since the onset of the epidemic and leading to the culling of more than 35 million poultry to curb the spread of the disease . To answer the challenge of controlling HPAI H5N1 across such a vast territory characterized by a diversity of agricultural production systems and economic development , China has taken several important steps to confront and control outbreaks and deal with the occurrence of human cases . These steps include measures such as stamping out , movement controls , cleaning and disinfection of infected premises , and the adoption of a nationwide massive vaccination campaign combined with intensive post-vaccination surveillance efforts . Effective vaccines have been developed and disease outbreaks have been responded to in a timely and effective manner , improving China's capacity to contain the disease and drastically reducing the number of outbreaks over the past years , with no outbreak detected in 2010 . The year 2005 represented a turning point in the control strategy with the enforcement of a so-called universal vaccination campaign , when vaccination became compulsory for all poultry using the H5N1 Re-1 ( A/Goose/Guangdong/1/96-PR8 ) vaccine strain . In parallel , a large amount of surveillance testing has been conducted both at provincial and national levels with the collection of an average of 4 . 7 million samples every year during the period 2007–2009 for the detection of silent viral circulation , the possible emergence of new strains and ultimately the identification of potential vaccination failure . Through the national surveillance program for the monitoring of HPAIV H5N1 circulation , the virus has been regularly detected , generating essential information for understanding the infection distribution pattern . More specifically , it has provided evidence that , despite a reduction in reported HPAI H5N1 outbreaks , some parts of the country still offer a favorable breeding ground for influenza viruses to circulate and potentially novel strains to emerge , representing a threat for the generation of new influenza pandemic strains . This particularly applies to the southern part of the country , which has historically been referred to as a hypothetical influenza epicenter [3] where agricultural and cultural practices place man and animals in close proximity . However , very few studies have actually attempted to map the potential distribution of avian influenza disease and infection risk across the diversity of ecological , cultural and production systems present in China . This lack of a sound description of HPAIV H5N1 geographical niches makes it difficult to refine control strategies that rely heavily on vaccination and that would greatly benefit from more targeted risk management . Better understanding of the infection dynamic pattern , the environmental and ecological factors associated with persistence of the disease in various poultry production systems will significantly strengthen efforts to achieve disease control and exclude infection from major poultry production centers , thus optimizing resources allocated to controlling the disease and reducing the risk for human infection . This study aimed to analyze the interrelationship of HPAI H5N1 in China with its environment , by exploring the association between selected spatial risk factors and two different indicators of HPAIV H5N1 presence , namely reported clinical outbreak occurrence in poultry and detection of sub-clinical HPAIV infection through risk-based surveillance . The study benefits from several improvements over previous work [4] . First , the analysis is not only based on HPAI H5N1 outbreak data which are of limited value in a context of massive vaccination ( especially after the implementation of the national vaccination campaign which began towards the end of 2005 ) , but also uses the results of HPAI H5N1 monitoring implemented as part of the national active surveillance program in live bird markets from 2007 to 2009 , termed risk-based surveillance in the rest of this manuscript . We also make use of an estimation of vaccine efficacy as measured by the Haemagglutination Inhibition ( HI ) test of serological samples collected monthly from poultry at province level . Second , we use updated poultry census data that differentiates between chicken , ducks and geese . This distinction is important as shown by previous HPAI H5N1 disease mapping efforts [5] , [6] , and is probably associated with differences in susceptibility to HPAI H5N1 virus between these species . Third , we used and compared the outputs of bootstrapped logistic regression and boosted regression trees with cross-validation , so that we could robustly estimate the weight of each tested risk factor , the goodness of fit of our predictions , and to allow us to map both the prediction of risk as well as its uncertainty .
Two types of data relating to HPAI H5N1 presence have been used as dependent variables in this study . First , poultry HPAI H5N1 disease outbreak data were compiled from two main sources: ( 1 ) one being the Official Veterinary bulletin published on the website of the Chinese Ministry of Agriculture ( MoA; http://www . agri . gov . cn/ ) ; and ( 2 ) the other source coming from official reports to the World Organisation for Animal Health ( OIE ) that were compared with MoA's report . Where there was an inconsistency in the outbreak date or location , we obtained accurate data through web research and consultation of local experts . Ninety-five percent of poultry outbreak data had detailed address information which was then geocoded . The remaining 5% for which no accurate location could be obtained were geocoded using the prefecture centroid ( administrative level 4 ) . Second , the Ministry of Agriculture in China routinely coordinates a surveillance program twice a year at the national level and monthly at provincial level in live bird markets consisting of sampling domestic poultry for the detection of HPAIV H5N1 . The selection of markets is based on their characteristics with regard to size , trade and hygiene practices which are assumed to increase the likelihood of detecting the virus . All samples collected at provincial level are tested by polymerase chain reaction ( PCR ) . All AI positive samples are sent to the Harbin National Veterinary Research Institute for confirmation , subtyping and virus isolation . The positive HPAIV H5N1 findings are then reported at the central government level and the data are released by the Veterinary Bureau in the MOA through the monthly Official Veterinary Bulletin , from which we extracted data on positive identification of HPAIV H5N1 between January 2007 and September 2009 . The surveillance data were geocoded using the market location when the market name was available . For 15% of the data , however , this information was missing and positives were geocoded using the prefecture centroid . In addition to HPAI H5N1 positives , the monthly proportion of post-vaccination seropositive samples was also extracted from the Monthly Official Veterinary Bulletin at the province level . Post-vaccination monitoring is performed on a regular basis at provincial level to assess the efficacy of the vaccination . Chickens , ducks and geese are sampled 21 days post-immunisation and an effective immune response is defined as a sero-conversion in bird with titres >4Log2 when measured by HI test , using homologous antigen , similar to the vaccine strain . Similar to the surveillance results , post-vaccination serological results obtained are collected at national level and published in the Official Veterinary Bulletin . The spatial distribution of HPAI H5N1 was investigated using a set of 7 explanatory variables which are known to be important risk factors , based on published scientific evidence and expert opinion . First , we considered the abundance of chickens , and domestic waterfowl separately based on previous work that had demonstrated a weak positive association between HPAI H5N1 presence and chicken density [6] , [7] , but a stronger association with duck density [5] . Second , anthropogenic variables were found to be associated with HPAI H5N1 in a number of studies conducted in countries with very different agro-ecological conditions such as Thailand , Bangladesh , Vietnam , and Romania [6] , [8] , [7] , [9] , [10] , and we therefore chose to include human population density . Third , several studies also identified land use and cropping variables as significant predictors of HPAI H5N1 presence in Asia . For example , Pfeiffer et al . [7] found HPAI H5N1 to be associated with the proportion of land occupied by aquaculture and by rice paddy fields in Vietnam , and Gilbert et al . [6] found a strong association with rice cropping intensity in Thailand . Similarly , statistically significant effects of access to water , or density of waterways were identified by Biswas et al [11] for Bangladesh and Ward et al . [12] for Romania . Therefore , we decided to include three variables: the proportion of land occupied by water ( running water or water bodies ) , the proportion of land occupied by rice paddy fields , and the cropping intensity ( number of crops cultivated in an unit area of cropland over a year ) . Finally , we included elevation in our analysis since several studies have reported an increased HPAI H5N1 risk in lowland and river delta areas [6] , [7] , [13] . The risk factor variables and corresponding data sources are presented in Table 1 . The analysis was carried out at a spatial resolution of 0 . 0833 decimal degrees of latitude and longitude ( approximately 5 . 5 to 8 . 8 km for the study area comprised between 54 and 18 degrees of latitude north ) . The statistical methods used in this study required having risk factor variable values for a large set of locations where HPAIV H5N1 or HPAI H5N1 outbreaks would be considered absent ( negatives ) , so as to contrast the agro-ecological conditions associated with HPAIV H5N1 or HPAI H5N1 outbreaks presence ( positives ) . Negatives were selected randomly from throughout the country based on three conditions: i ) no HPAI H5N1 outbreaks had been reported and no HPAIV H5N1 positive results had been obtained from the active risk-based surveillance; ii ) being at a minimum distance >0 . 0833 decimal degree of any positive; and iii ) being in a location where human population density was >1 person/km2 to exclude desert and high mountain areas from the analysis since the focus of this analysis was on locations with likely relevance for disease maintenance in poultry . Two approaches were used to model the spatial distribution of HPAI H5N1 presence or absence: multiple logistic regression and boosted regression trees ( BRT ) . Logistic regression allows predicting a variable with a binary response , such as the presence or absence of a disease , as a function of a number of variables , or predictors . Logistic regression models have been used in a number of studies trying to identify environmental correlates and risk factors associated with HPAI H5N1 presence [6] , [7] , [10] . However , one limitation of logistic regression is the necessity to perform specific adjustments to accommodate non-linearity of effect of the continuous-scale risk factors on the logit form of the outcome variable . Two approaches have been described to account for this . First , a risk factor can be added to the model as a quadratic term so that predicted probabilities of presence can be maximum ( or minimum ) for intermediate values ( e . g . [6] , [9] ) . Second , each continuous-scale risk factor variable is converted into a nominal-scale variable where each category level represents a particular range of values in the original variable ( e . g . [14] , [7] , [10] ) . However , both methods have their limitations . The first method can only partially model more complex non-linear dependencies , and the second method is sensitive to the range of values represented by each category level . In the presence of spatial autocorrelation , logistic regression requires the use of relatively complex estimation algorithms . BRT has been developed relatively recently for predicting the distribution of organisms [15] . It is very efficient for dealing with non-linear relationships and interactions between variables . It can be considered a disadvantage that it does not have the facility to assess the statistical significance of individual effect variables , though it allows estimating the relative importance of each variable to the predictions . In a comprehensive review of presence/absence distribution modeling methods , Elith et al . [16] found BRT to perform best along with the maximum entropy method . Elith et al . [15] published a detailed description of an analysis approach using BRT which implements a cross-validation procedure allowing identification of model parameters . In this study , we compare logistic regression and BRT in terms of validity and ease of interpretation of the outputs . We also discuss our findings in relation to the spatial patterns of HPAI H5N1 described in previously published work that used logistic regression methods . In the logistic regression method , all variables were forced in the model , and the likelihood ratio test was used to assess the contribution of each variable to the predictions . For the BRT model , we used 10 sets of training and test points for cross-validation , a tree complexity of 4 , a learning rate of 0 . 005 and a bag fraction of 75% . Using those parameters , the cross-validation stepwise function presented by Elith et al . [15] was used to identify the optimal number of trees in the model . The weight of each variable estimated over the identified number of trees was used as an indicator of each variable's importance for predicting HPAI H5N1 presence/absence . One should note that those weights are not absolute metrics , and the weights of all variables of a BRT model sum to 1 . This analysis was conducted using two outcome variables , first HPAI H5N1 outbreak occurrence during the entire study period and second HPAIV H5N1 positive findings between 2007 and 2009 . Assessing the performance of our models directly from the logistic regression and BRT predicted probabilities and observed presence/absence had two main limitations . First , logistic regression performance metrics have been shown to be sensitive to low ( <10% ) and high ( >90% ) frequencies of the binary outcome categories [17] . The proportion of positives in our dataset was extremely low , and it was therefore necessary to address this potential bias ( to our knowledge , the presence of this potential bias has not been thoroughly assessed for BRT models , but see [18] ) . Second , quantifying model performances using the data set used to train the model tends to inflate the performance metrics compared to a situation where an independent data set is used . We developed a bootstrapping procedure aiming to generate a robust estimate of model performance by simultaneously addressing those two limitations . The bootstrapping analysis involved a series of sequential steps: ( i ) selection of a balanced subset of data from the complete dataset: all n points with HPAI H5N1 presence were included and an equivalent number of absence points was randomly selected from all ‘absence’ points; ( ii ) creating a training data set and a test data set: the balanced subset of data was randomly divided into two subsets: one for building the models ( training set , with 75% of the points ) and the other for evaluating the models ( test set , with the remaining 25% of the points ) ; ( iii ) model development: a logistic regression and a BRT model were built using the training set , and parameters of both models were stored; ( iv ) model evaluation: the model equations from the logistic regression and BRT models were used to generate predictions using the test set , which in turn were assessed using ROC curves , areas under the curve and Cohen's kappa statistic; ( v ) risk maps: maps of the predictions produced by each model were stored . Steps ( i ) to ( v ) were repeated 50 times , and the mean and standard deviation of all statistics and predicted spatial distributions were estimated . Due to percentage of post-vaccination seropositivity only being available at province level , using yearly data for the period 2007-2009 as unit of analysis , a separate analysis was conducted to quantify the statistical association between antibody response to vaccination expressed as a percentage and two variables: the presence/absence of HPAI H5N1 outbreak records in the province and detection of H5N1 HPAIV detected through national risk-based surveillance activities conducted by the MoA . The post-vaccination seropositivity was analysed as the response variable of a two-way ANOVA with the presence/absence of HPAI H5N1 outbreak records in the province and the year as two explanatory factors . The same analysis was carried out with the presence/absence of H5N1 HPAIV detected through national risk-based surveillance activities and year as explanatory factors . This allowed separating the effect of HPAI H5N1 status from the possible effect of time .
The distribution of HPAI H5N1 outbreaks and HPAIV H5N1 surveillance positives are shown in Fig . 1 . Overall , the two analysis techniques , logistic regression and BRT , provided consistent results in terms of risk factors being identified . In contrast , the set of risk factors and their effect differed strongly between the outcomes of reported HPAI H5N1 outbreak and risk-based surveillance data . Based on the logistic regression results , HPAI H5N1 outbreaks were found to be positively associated with human population density and negatively with elevation ( Table 2 ) . The BRT models also identified chicken density to be an important variable for discriminating between locations with and without reported HPAI H5N1 outbreaks ( BRT weights , Table 2 ) . The averaged BRT model fitted functions shown in Fig . 2 allow a detailed description of these relationships ( maps of the predictions coefficient of variation are presented in Figure S1 in Text S1 ) . The predicted risk of HPAI H5N1 outbreak occurrence appears to be constant for densities of chickens ranging from 0 to 10 , 000 heads/km2 , then increases to a maximum risk at around 100 , 000 heads/km2 . The predicted risk also increases significantly with human population density , starting from a density of 1 , 000 people/km2 . We also identify a strong negative relationship with elevation , with the predicted risk function showing two levels , a high risk for elevation ranging from 0–100 m , and a low risk for higher elevations . The predicted risk is relatively flat for all ranges of domestic waterfowl density and percentage of land with surface water . In contrast , HPAIV H5N1 surveillance positives were found to be positively associated with the density of domestic waterfowl , with percentage of land occupied by water and to human population density ( though this factor was not important in the BRT models ) , and negatively associated with chicken density ( Table 2 ) . Here again , the predicted risk function of the BRT models allows a detailed description of these relationships ( Fig . 2 ) . The predicted risk of HPAIV H5N1 surveillance positives is constant for waterfowl density ranging between 0 and 10 , 000 heads/km2 , then rises sharply for increasing densities . A similar profile as for the HPAI H5N1 outbreak data is found for the association with human population density , with predicted risk increasing with human population density from a density of 1 , 000 people/km2 . The predicted risk increases with percentage of area covered by surface water up to a value of approximately 7% , and then remains constant for higher values . The profiles of predicted risk as a function of chicken density and elevation are relatively constant . The accuracy metrics of the predictions produced by the logistic regression and BRT models are good to excellent , with mean AUC values estimated using the evaluation dataset ranging from 0 . 864 to 0 . 967 ( Fig . 3 ) . One can note that , as expected , the AUC estimated based on the training data is always better than that estimated using the evaluation dataset , and that this difference is much higher for the BRT models . However , even when assessed using the evaluation dataset , the accuracy of BRT models appears better than that of the logistic regression models . In addition , the accuracy metrics are higher for the models for HPAIV H5N1 risk-based surveillance data than those obtained for the HPAI H5N1 outbreak data . One can note that considering only eastern China in the evaluation of AUC values slightly reduces it's value , but to a marginal extent , showing that the good predictive power does not result from predicting risk over wide desert areas unsuitable to disease spread . The predicted geographical distribution of HPAI H5N1 presence also differs according to the type of training data ( clinical disease outbreaks vs . risk-based surveillance; Fig . 4 ) . Maps generated based on the outbreak data place more emphasis than those based on risk-based surveillance data on north-eastern regions where chicken densities are higher . We also note a marked difference for the outbreak data between the outputs of the logistic regression model and of the BRT , the latter predicting many more clustered areas with high probability of HPAI H5N1 presence . In contrast , the distribution of predicted HPAIV H5N1 presence based on risk-based surveillance data identifies areas at risk much more concentrated in the southern part of the country , with outputs from the logistic and BRT models showing similar patterns . In the analysis at province level , we found that the proportion of seropositivity in post-vaccination surveys was lower in provinces that had reported HPAI H5N1 outbreaks than in those that did not ( Fig . 5 left; two-way ANOVA with both year and HPAI H5N1 outbreak status as factor variables; HPAI H5N1 outbreak status: F1 , 87 = 18 . 53 , p<0 . 001; Year: F1 , 87 = 0 . 51 , n . s . ; interaction term - Year by HPAI H5N1 outbreak status: F1 , 87 = 0 . 0264 , n . s . ) , and in provinces where HPAIV H5N1 had been detected during active surveillance and than in those where this had not been the case ( Fig . 5 right; two-way ANOVA with both year and HPAIV H5N1 risk-based surveillance status as factor variables; HPAIV H5N1 risk-based surveillance status: F1 , 87 = 5 . 09 , p = 0 . 026; Year: F1 , 87 = 2 . 02 , n . s . ; interaction term - year by HPAIV H5N1 risk-based surveillance status: F1 , 87 = 0 . 172 , n . s . ) .
In China , where approximately 15 billion head of poultry are produced annually with a standing population of 5 . 6 billion chickens , 760 million ducks and 300 million geese , major regional differences are apparent in ecological systems , husbandry practices , cultural behaviors and economic development with a consequential impact on the distribution of infectious diseases including HPAI H5N1 , as well as their maintenance and spread and therefore on disease control options . To date , spatial studies aiming at identifying HPAI H5N1 risk factors have been undertaken in many countries where the disease was introduced such as Thailand and Vietnam [5] , [6] , [14] , [8] , [7] , [19] , Korea [20] , India and Bangladesh [9] , [11] , [21] , Romania [22] , [23] or Africa [24] , [25] . Only three studies analyzed the distribution of HPAI H5N1 outbreaks in China [4] , [26] . Of these , only the study by Fang et al . [4] attempts to map the distribution of HPAI H5N1 risk . Whilst highly valuable given that it is the first analysis , the output predicts areas at high risk in ecological areas that would not support the maintenance and transmission of the virus such as in the extremely large desert regions of Inner Mongolia , Tibet and Xinjiang autonomous regions . In our study , we reported different results for the analyses based on outbreak and risk-based surveillance data . The distribution of reported HPAI H5N1 outbreaks was found to be primarily associated with lowland regions with high human population and chicken density . In contrast , HPAIV H5N1 presence detected through risk-based surveillance activities was found to be associated with regions with high waterfowl densities and were covered by high proportions of surface water . This result is very interesting since it may be a reflection of differences in HPAIV H5N1 pathogenicity between chickens and ducks , combined with environmental and host population conditions supporting virus spread and clinical disease outbreak occurrence as distinct from clinically silent virus persistence . HPAI H5N1 is far more pathogenic in chickens than in ducks [27] , [28] , though there is also evidence of significant variability in virulence at the species level [29] . In the absence of control or prevention measures , the spread of HPAIV H5N1 and occurrence of clinical disease outbreaks is facilitated in regions where the density of chickens is particularly high , especially in intensive and industrial conditions where high numbers of animals are together facilitating transmission . Such regions are encountered in the north-eastern part of China , where the low cost of grain feed production and a fast-rising demand for poultry meat has supported the rapid development of intensive chicken production . Those intensive poultry production systems invest significant resources in disease prevention measures , and will apply mass-vaccination of their flocks , thereby preventing HPAIV H5N1 spread within and between farms . However , it is likely given the exceptionally high density of chickens and farms that occasional , albeit rare , lapses in vaccination coverage result in a small number of outbreaks . The human population density risk factor can be interpreted as a proxy of several epidemiological processes that are more likely to occur in highly-populated areas , such as a higher likelihood of outbreak detection and higher possibilities of HPAIV H5N1 transmission through trade and farming-related activities . In contrast , long-term persistence of HPAIV H5N1 can only be possible if the virus can circulate without being detected or reported . Domestic ducks have been shown to be able to excrete large amounts of virus whilst remaining apparently healthy [27] . Regions rich in domestic waterfowl are hence more prone to long-term persistence of HPAIV H5N1 . This can be further exacerbated in geographical areas with an abundance of surface water . Permanent water bodies , rivers , rice paddy fields and canals are the habitat of wild and domestic ducks . One may speculate that water facilitates the transmission between hosts without direct contact through the faecal excretion of the virus , its persistence in the water , and the oral infection of other susceptible hosts sharing the same pond or downstream canal or river . The results indicating that HPAIV H5N1 presence detected through risk-based surveillance is associated with areas that have high waterfowl densities and a high proportion of surface water allows thus a straightforward interpretation . Associations between HPAIV H5N1 and domestic duck density had already been identified in other countries [5] , [7] . However , no difference between outbreaks and clinically-silent infections was made in these earlier studies , which indeed becomes essential when analyzing HPAIV H5N1 distribution in the context of mass-vaccination such as in China . Interestingly , the farming and cultural practices encountered in these regions were already described by Shortridge 28 years ago as an avian influenza breeding ground [30] . Among others , Southern China still hosts a massive duck population raised on ponds and rice fields , facilitating frequent faecal-oral transmission of multiple influenza subtypes leading to a year-round and inter-epidemic occurrence of influenza viruses . Historically , agricultural practices in China have developed from the need to feed the people as efficiently as possible , using all available resources , and with little recourse to modern farming methods . Domestic ducks were first moved from rivers to cultivated rice fields at the start of the Qing dynasty in the middle of the 17th century [31] , [32] to help protect the growing rice from pests . This practice reduces farmers' dependence on chemical insecticides , herbicides , fertilizers and mechanical farming aids and provides a close association between bird , water , rice and people . Ducks raised in ponds are also an important feature in the villages and communities of China , especially in southern China and coastal areas including the waterways of the Pearl River delta which are ideal for rice and fish farming [30] . Furthermore , southern China has always been the focus of influenza experts' attention and often been referred as a hypothetical epicenter of AI pandemic strains . The foundation of this concept was originally raised by Webster et al . [33] and supported by the wide variety of influenza virus subtypes discovered in Southern China during decades [34]–[37] . More specifically , the distribution of HPAIV H5N1 risk of persistence inferred from the risk-based virological surveillance data and using the logistic and BRT models is similar and highlights different levels of risk according to the following ecological regions ( Fig . 4 , bottom; see Figure S2 in Text S1 for a map of the zones ) : Zone I ) In Southern China a large potential zone of virus persistence extends from south of the Yangtze River . This area hosts the vast majority of the Chinese duck population comprising birds for meat or egg production . This area can be subdivided into three areas: I-a ) an area which extends from the provinces of Jiangsu , Anhui , Hubei , Jiangxi , Hunan , Guangxi autonomous region down to Guangdong province . This might be one of the most important ecological zones where key epidemiological drivers for emergence , persistence and spread are present , including a huge reservoir population , traditional farming system , a high animal and human population density , some major wild bird congregation sites such as the Poyang lake located in Jiangxi province and an important North – South gradient of poultry trade which crosses this region . This supports the hypothesis of a wider and slightly displaced epicenter of influenza viruses , not only concentrated around the Pearl River delta in Guangdong province but extending south of the Yangtze River and including provinces such as Jiangxi where internal segments of the 1996 geese HPAI H5N1 virus may have originated [38] . I-b ) A coastal area stretching from Jiangsu to Guangdong provinces with a risk hotspot in Guangdong province along the Pearl River delta . This strip of coastal land also hosts the typical duck pond system where the risk of infection and disease is present . I-c ) Few isolated areas within this geographical zone displaying an increased risk located in Yunnan , Guangxi autonomous region , Guizhou , Sichuan and Chongqing provinces which have experienced either outbreaks in the past ( Guangxi autonomous region and Yunnan ) or only reported viral circulation ( Sichuan and Chongqing provinces ) . Zone II ) A vast geographical area in the West and North , displaying radically different geography , socio-economic and animal production features and characterized by scattered and isolated spots of higher predicted risk . This includes specifically southern Tibet autonomous region and scattered areas in the North and South of Xinjiang autonomous region where sporadic outbreaks have occurred in the past . Zone III ) In the North-East of the Yangtze River , a region where the contribution to disease persistence seems fairly limited while localized areas at higher risk of outbreaks encroach regions of intensive production where the disease could rapidly spread in case of virus introduction and breach in biosecurity . This region extends from Shangdong into Liaoning , Jilin and Helongjiang provinces . These provinces are characterized by denser human population and large-scale commercial poultry production , and were predicted as high risk based on the reported clinical disease outbreak data ( Fig . 4 top ) . In these regions of North-Eastern China , chicken production and marketing systems are intensifying and concentrating in response to economic growth and urbanization . Substantial numbers of poultry are now processed at large-scale slaughterhouses in this region , while the majority of poultry are still sold through live poultry markets in the South of the country . In the colder north-eastern provinces water birds are also housed and kept more intensively . The apparent persistence of HPAIV H5N1 in those regions has two main implications . First , given the possible presence of silent infection involving an extremely high population of domestic waterfowl , eradication of the virus through massive vaccination appears extremely difficult , although it has successfully reduced the number of outbreaks . Vaccination has been one major component of the government policy to curb the spread of the disease and reduce the incidence of outbreaks of clinical disease and of transmission of infection . China uses more vaccine against avian influenza than any other country and Chinese veterinary authorities base much of their control and preventive strategy around vaccination [39]–[41] . More than 13 billion doses of AI vaccine have been used each year since 2007 [39] and the objectives of the national strategy are to reach a 100% vaccination coverage for the national poultry population and to ensure an effective immune response ( defined as sero-conversion in bird with titres >4Log2 when measured by HI test ) in more than 70% of the nationwide poultry population all year round . In this study , we also analyzed the post-vaccination surveillance data collected at provincial level since January 2007 and found that provinces where clinical HPAI H5N1 outbreaks had been reported or HPAIV H5N1 detected had a lower level of post-vaccination seropositivity , confirming that increased protection does indeed result in lower disease outbreak or infection risks but would require an approach better targeted at identified high risk areas to drastically reduce the viral load in the environment . Second , the different regions of China are not independent and are possibly epidemiologically linked through poultry trade and likely also through wild bird migrations . High production-demand discrepancies lead to long-distance trade of poultry products ( e . g . chicken from the north exported to southern provinces , or duck meat exported from the south to the north ) . In addition , areas such as the Poyang lake , where a large population of domestic waterfowl is raised in close proximity to thousands of over-wintering wild waterfowl , could favour the transmission between wild and domestic waterfowl and lead to long-distance transmission of the virus . As a consequence , the persistence of HPAIV H5N1 in some particular regions may influence the chances of introduction into other more distant regions . For instance , the wild bird 2 . 2 clade which was associated with the origin of the Qinghai lake epidemic in 2005 in West China was responsible for a major outbreak during the same year in domestic poultry in Liaoning province , in the north eastern part of the country . Likewise , the 2006 Shanxi strain also grouped into the clade 7 cluster present in North-Central China has been found in Jiangsu province in South Eastern China . There is a complex pattern of links that exists between these different ecological regions that offers hiatuses for viruses to escape their reservoir areas and invade others . Continued efforts pursued by the Ministry of Agriculture , its affiliated research centers and local veterinary authorities to strengthen the HPAI national surveillance program and its control strategy have resulted in a steady decrease in the number of outbreaks reported since 2004 and a better understanding of HPAIV H5N1 infection distribution in space , time and within traditional marketing systems known as live bird markets . However , national surveillance programs have also demonstrated that HPAIV H5N1 continues to circulate in poultry on a regular basis . Since 2007 , an apparent increase in virus detection is believed to represent the result of increased and intensive efforts made by the Ministry of Agriculture to detect the virus through targeted risk-based surveillance activities at live bird markets and high-risk farms in a context of massive vaccination efforts that could potentially mask the clinical expression of the disease within a large population of immunized birds . Although the epidemiology of HPAIV H5N1 in China does not seem to present radically different features compared with neighboring countries also affected by the disease , it remains unique in terms of the abundance of reservoir species both domestic or wild , providing ample opportunities for a sustained and rapid evolution of the virus and requiring intensive virus monitoring for pandemic preparedness matters . While revisiting the concept of epicenter for pandemic strains of avian origin , the results of this study represent major improvements over previous efforts in mapping the risk of HPAI H5N1 in two main aspects . First , it allows identifying several risk factors of animal , environmental and anthropogenic nature , with clear biological and epidemiological interpretation . Second , the bootstrapped statistical modeling allows us to robustly estimate the predictive power of our model , but also to map the uncertainty that goes with our predictions ( Figure S1 in Text S1 ) , which is useful information for an applied use of these maps . Combining innovative modeling techniques with data of improved quality and integrating measures of infection persistence , our results have broad fundamental implications in a country where understanding of the ecology of influenza viruses , although of utmost importance for pandemic preparedness purposes , has remained until now mostly speculative . Finally , the analyses presented in this study may be improved in the future by several complementary approaches . First , the potential transmission through trade patterns and bird migration should be more comprehensively assessed . An increasing amount of data are being collected on both aspects , and this will ultimately contribute to better understanding of how areas of high potential for HPAIV H5N1 persistence may be connected to each others . Second , the results could be further integrated into an Asia-wide improved understanding of HPAIV H5N1 distribution models , benefiting from several studies that have been undertaken in neighboring countries . Third , information on true negatives obtained through the national surveillance programme would reduce the risk of including false-negatives in the analyses , and provide higher resolution estimates of the relative importance of risk factors .
|
The geographical distribution of highly pathogenic avian influenza ( HPAI ) H5N1 and agro-ecological risk factors have been studied in a number of countries in Southeast Asia . However , little is know of its distribution in China where HPAI H5N1 first emerged in 1996 , evolved , and spread throughout Asia and the western Palearctic in 2004–2006 . This study analyzes separately the distribution , in domestic poultry , of HPAI virus ( HPAIV ) H5N1 isolated from active risk-based surveillance sampling and HPAI H5N1 clinical disease outbreaks . These data are analyzed in relation to the distribution of chicken and domestic waterfowl population density , proportion of land covered by rice or surface water , cropping intensity , elevation , and human population density . HPAI H5N1 viruses identified by risk-based surveillance are found to be associated with domestic waterfowl density , human population density , and the proportion of land covered by surface water . In contrast , HPAI H5N1 clinical disease outbreak occurrences were mainly associated with chicken density , human population density , and low elevation . These results show that the distribution of HPAI H5N1 risk in China appears more limited geographically than previously assessed , offering prospects for better targeted surveillance and control interventions .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"public",
"health",
"and",
"epidemiology/epidemiology",
"public",
"health",
"and",
"epidemiology/infectious",
"diseases",
"infectious",
"diseases",
"ecology/spatial",
"and",
"landscape",
"ecology"
] |
2011
|
Spatial Distribution and Risk Factors of Highly Pathogenic Avian Influenza (HPAI) H5N1 in China
|
Measuring the prevalence of transmissible Trypanosoma brucei rhodesiense in tsetse populations is essential for understanding transmission dynamics , assessing human disease risk and monitoring spatio-temporal trends and the impact of control interventions . Although an important epidemiological variable , identifying flies which carry transmissible infections is difficult , with challenges including low prevalence , presence of other trypanosome species in the same fly , and concurrent detection of immature non-transmissible infections . Diagnostic tests to measure the prevalence of T . b . rhodesiense in tsetse are applied and interpreted inconsistently , and discrepancies between studies suggest this value is not consistently estimated even to within an order of magnitude . Three approaches were used to estimate the prevalence of transmissible Trypanosoma brucei s . l . and T . b . rhodesiense in Glossina swynnertoni and G . pallidipes in Serengeti National Park , Tanzania: ( i ) dissection/microscopy; ( ii ) PCR on infected tsetse midguts; and ( iii ) inference from a mathematical model . Using dissection/microscopy the prevalence of transmissible T . brucei s . l . was 0% ( 95% CI 0–0 . 085 ) for G . swynnertoni and 0% ( 0–0 . 18 ) G . pallidipes; using PCR the prevalence of transmissible T . b . rhodesiense was 0 . 010% ( 0–0 . 054 ) and 0 . 0089% ( 0–0 . 059 ) respectively , and by model inference 0 . 0064% and 0 . 00085% respectively . The zero prevalence result by dissection/microscopy ( likely really greater than zero given the results of other approaches ) is not unusual by this technique , often ascribed to poor sensitivity . The application of additional techniques confirmed the very low prevalence of T . brucei suggesting the zero prevalence result was attributable to insufficient sample size ( despite examination of 6000 tsetse ) . Given the prohibitively high sample sizes required to obtain meaningful results by dissection/microscopy , PCR-based approaches offer the current best option for assessing trypanosome prevalence in tsetse but inconsistencies in relating PCR results to transmissibility highlight the need for a consensus approach to generate meaningful and comparable data .
For the vector-borne diseases , pathogen prevalence in a vector population is an indicator of disease risk , and accurate measures of the proportion of vectors carrying infections are needed for ( i ) guiding allocation of resources or targeting intervention programs [1]; ( ii ) monitoring the success of control interventions [2]; and ( iii ) as parameters in models of disease transmission which are increasingly used to predict disease distribution and persistence , and plan control interventions [3] . Approaches for detecting parasite prevalence in vector populations , known as xenomonitoring , have until recently usually relied on dissection of insect vectors and visualisation of parasites by microscopy , which is time consuming and reliant on operator skill . PCR has presented an alternative technique for several parasite-vector systems , e . g . Plasmodium spp [4] , Oncocerca volvulus [5] , [6] , Leishmania spp . [7] , [8] , and the nematodes which cause lymphatic filariasis , Wuchereria bancrofti , Brugia malaya and Brugia timori [9] , [10] , generally having better ability to differentiate between species of similar morphology , increased sensitivity , and hence requiring smaller sample sizes [4] , [6] , [8] . Human African trypanosomiasis ( HAT ) is caused in East Africa by Trypanosoma brucei rhodesiense and transmitted by tsetse flies ( Glossina spp ) . Measuring the prevalence of T . b . rhodesiense in the tsetse vector is of particular importance as HAT occurs in developing countries where resources for surveillance and disease control are limited [11] and knowledge of human disease risk is important for effective targeting of available resources . In addition , HAT is characterised by its focal nature , with human cases continuing over long periods of time in specific geographical areas , but the reasons for this persistence are not clear [12] . The prevalence of infection in tsetse is an important component in understanding transmission dynamics and detecting spatiotemporal trends , which have important implications for disease control . Assessment of the prevalence of trypanosomes within tsetse populations has traditionally comprised dissection and microscopic examination of the mouthparts , midguts and salivary glands of the fly , relying on the differing development and maturation sites of the trypanosome subgenera to identify trypanosome species [13] . Trypanosomes found only in the mouthparts are classified as Duttonella or vivax-like , trypanosomes located in the mouthparts and midguts are classified as Nannamonas or congolense-type , and trypanosomes found in the midgut and salivary glands are Trypanozoon or brucei-like . When trypanosomes are found only in the midgut , the infection is assumed to be immature . This dissection/microscopy technique has several disadvantages for use in field studies: it is not possible to differentiate below the level of subgenus ( for example T . simiae cannot be differentiated from T . congolense , since they share development sites in the fly ) ; mature and immature infections cannot always be differentiated; and mixed infections cannot be identified or discriminated . Dissection and trypanosome identification are highly dependent on operator skill , and there exist variations in protocols , with some authors only examining the midgut and salivary glands if trypanosomes are found within the mouthparts [14] , [15] , whilst others examine all the organs [16] , [17] . A suite of molecular tools has been developed for the trypanosomatids [18] , [19] . PCR and sequence analysis techniques have served to overcome some of the disadvantages of dissection/microscopy and highlighted new information about tsetse-trypanosome interactions . PCR primers with high sensitivity and specificity now permit trypanosomes to be reliably identified to species or subspecies level , for example new strains or potentially even species of trypanosome have been identified [20] , [21] , [22] , and human-infective T . b . rhodesiense and its morphologically-identical subspecies Trypanosoma brucei brucei ( not pathogenic to man ) can be accurately differentiated [23] . Mixed infections are common , with approximately one third of PCR positive flies carrying more than one trypanosome species [20] , [24] , [25] and up to four trypanosome species identified in individual flies [24] , [25] . However , when it comes to assessing the prevalence of trypanosome infections in tsetse it is clear that the results generated by dissection/microscopy do not correlate well with data generated by PCR ( for example only 38% [25] to 51% [24] of Nannomonas or T . congolense-like and Duttonella or T . vivax-like infections are classified as the same species by both techniques ) . For T . brucei sensu lato , with its potential for human infection , this presents a particular problem . In areas where T . b . rhodesiense is known to occur in wildlife and livestock hosts , and human cases are reported , the majority of studies of T . brucei s . l . in tsetse by dissection/microscopy show prevalence of zero , even when thousands of flies are examined [16] , [26] . However when whole tsetse flies have been analysed by PCR surprising amounts of T . brucei s . l . DNA has been found , with 2% of G . palpalis and 18% of G . pallidipes testing positive [27] , [28] . The discrepancy between dissection/microscopy and PCR highlights the issues of assessing the true prevalence of human infective trypanosomes in tsetse populations , particularly as it is not clear how these measures relate to transmissibility . Furthermore , it would be useful if a consensus could be reached as to how best to use molecular data , either alone or in combination with results of dissection/microscopy , to generate prevalence measures . This study presents data from a persistent focus of Rhodesian HAT in the Serengeti National Park ( SNP ) , Tanzania . Whilst cases of HAT have been reported in this area for over one hundred years [29] , recent cases in both the local population and tourists have renewed public health concerns about the disease [30] , [31] . With abundant populations of G . swynnertoni and G . pallidipes , and almost 100 000 tourists visiting the SNP each year in addition to resident staff and local populations [32] , understanding and mitigation of human disease risk is a priority . Previous studies carried out in SNP have relied on dissection/microscopy to determine tsetse prevalence ( Table 1 ) . Large scale studies in 1970 and 1971 failed to identify any salivary gland infections [16] , [26] but a subsequent pooled rodent inoculation study detected nine out of 11000 G . swynnertoni flies ( 0 . 08% ) infected with T . brucei s . l . [33] . These findings contrast with results of a more recent study that reported a prevalence of 3 . 0% for T . brucei s . l . in G . swynnertoni [34] and raise questions as to whether the wide variation in detected prevalence reflects real changes in tsetse infection levels and human exposure risk , or reflect methodological differences . This study assessed the prevalence of T . brucei s . l . and T . b . rhodesiense in the two main tsetse species in SNP , G . swynnertoni and G . pallidipes , using ( i ) dissection/microscopy and ( ii ) PCR analysis of infected midguts and salivary glands . A third approach was applied to infer the prevalence of T . b . rhodesiense in tsetse from a mathematical model of disease transmission , to examine whether previously reported low prevalences were consistent with other parameters that have been estimated for this system .
All field work was conducted in SNP , Tanzania , between October and November 2005 and August and October 2006 . Tsetse sampling was carried out with the Tsetse and Trypanosomiasis Research Institute , Tanga , Tanzania . Seven sites were randomly selected for tsetse trapping in savannah and open woodland areas , within 1 km of roads and within a 40 km radius of park headquarters at Seronera , where tsetse dissection was conducted ( coordinates UTM 36M ( i ) 711676 , 9731432; ( ii ) 706816 , 9733868; ( iii ) 710747 , 9733536; ( iv ) 695691 , 9727934; ( v ) 700825 , 9746320; ( vi ) 693961 , 9733122; ( vii ) 695278 , 9741360 ) . In each study site , three Epsilon traps [35] were installed for between five and eleven days , depending on trap catches . Each trap was situated at least 200 m from the next , and erected in mottled shade to reduce fly mortality . When placing traps , areas with fallen trees were avoided and traps were placed so that the entrances were directed towards gaps in vegetation , measures known to maximise fly catches by following the natural patterns of tsetse flight [36] . The location of each trap was recorded using a handheld global positioning system ( Garmin Ltd , Kansas , USA ) . Traps were baited with 4-methylphenol ( 1 mg/h ) , 3-n-propylphenol ( 0 . 1 mg/ ) , 1-octen-3-ol ( 0 . 5 mg/h ) and acetone ( 100 mg/h ) [37] and emptied twice daily . All live non-teneral flies were dissected and labrum , hypopharynx , salivary glands and midgut examined for trypanosomes under 400× magnification [38] . For each fly , species , sex and the presence or absence of trypanosomes in each organ were recorded . To prevent contamination between flies and between different parts of each fly , dissection instruments were cleaned in 5% sodium hypochlorite , followed by rinsing in distilled water then phosphate buffered saline between each organ . Flies carrying trypanosome infections was categorised according to Lloyd and Johnson [13] . Confidence intervals were calculated using binomial exact 95% limits . All trypanosome-positive midguts and salivary glands were macerated in phosphate buffered saline and applied to FTA Classic cards ( Whatman , Maidstone , UK ) for further analysis . A subset of trypanosome-negative midguts was also preserved on FTA cards . FTA cards were allowed to dry for two hours and stored in foil envelopes with dessicant at ambient temperature prior to processing . For each sample , one disc of diameter 2 mm was cut out from the FTA card using a Harris Micro Punch™ tool . Between cutting of the sample discs , 10 punches were taken from clean FTA paper , to prevent contamination between samples . Discs were washed for two washes of 15 minutes each with FTA purification reagent ( Whatman Biosciences , Cambridge , UK ) , followed by two washes of 15 minutes each with 1X TE buffer ( Sigma Aldrich , Dorset , UK ) . Each disc was dried at room temperature for 90 minutes , and then used to seed a PCR reaction . After every seven sample discs , a negative disc was included and the punch tool and mat cleaned , to reduce the risk of contamination between discs , and ensure that any potential contamination would be detected . No evidence of contamination was seen in the sequence of dissection or PCR results . TBR primers were used to detect a 177 bp satellite repeat sequence common to T . b . brucei , T . b . rhodesiense and T . b . gambiense [39] . PCR was carried out in 25 µl reaction volumes containing 16 . 0 mM ( NH4 ) 2SO4 , 67 mM Tris-HCl , 0 . 01% Tween 20 ( NH4 buffer , Bioline Ltd , London , UK ) 1 . 5 mM MgCl2 , 800 µM total dNTP's , 0 . 4 µM of each primer TBR1 and TBR2 , 0 . 7 Units of BioTaq Red DNA polymerase ( Bioline Ltd , London , UK ) and one washed disc . For samples testing positive for T . brucei s . l . , T . b . rhodesiense was differentiated from T . b . brucei by detection of the serum-resistance associated ( SRA ) gene . Simultaneous amplification of another single copy gene , a phospholipase C ( PLC ) sequence found in T . brucei s . l . , confirmed that there was sufficient T . brucei s . l . material present in the sample to detect the presence of T . b . rhodesiense [40] . SRA PLC PCR was carried out in duplicate in a 25 µl reaction volume containing 3 mM MgCl , 1 . 25 µl of Rediload dye ( Invitrogen , Karlsbad , California ) , 1 . 5 Units Hot StarTaq ( Qiagen , Crawley , UK ) , 0 . 2 µM of each primer and one washed disc . The SRA gives a 669 bp product , with a PLC band at 324 bp . For all PCRs , one negative control ( water ) and one positive control ( genomic DNA ) were run for every 16 samples , in addition to negative control blank discs . PCR products were run on a 1 . 5% ( w/v ) agarose gel at 100 V , stained with ethidium bromide and visualised under an ultraviolet transilluminator . Detection of T . b . rhodesiense in a tsetse midgut does not indicate a mature infection as only a small proportion of midgut infections will develop to mature infections in the salivary glands . The following calculation was used to predict the prevalence of mature transmissible T . b . rhodesiense infections , where Dispos is the proportion of flies with midguts which were positive by dissection/microscopy , PCRpos is the proportion of these which tested positive by PCR , PTbr/Tbb is the proportion of T . brucei s . l . positive flies with sufficient genetic material present ( ie give positive results with PLC PCR ) which test positive for T . b . rhodesiense ( as determined by SRA PCR ) and Pmat is the proportion of immature T . b . rhodesiense infections which develop to maturity in the salivary glands , estimated to be 0 . 12 ( CI 0 . 10–0 . 14 ) , [41] , [42]: ( 1 ) This calculation relies on three assumptions: ( i ) that dissection/microscopy is 100% sensitive for detecting trypanosome infections in tsetse midguts , and that all flies carrying T . brucei s . l . will have midgut infectons; ( ii ) that TBR PCR has 100% sensitivity and specificity for detection of T . brucei s . l . in tsetse midguts; ( iii ) that SRA PCR has 100% sensitivity and specificity for detection of T . b . rhodesiense , if the sample is positive on PLC PCR . The implications of potential assumption violations on the prevalence estimate are addressed in the discussion . Confidence intervals were calculated by repeat sampling from nested distributions of the data . Since the value for Pmat was taken from Milligan et al . ( 1995 ) the distribution of the original data was used , where Y is the number of flies with midgut infections and Pmat is the proportion of these which developed mature salivary gland infections ( Y = 1133 , Pmat = 0 . 12 ) . Potential values were generated by sampling from the following nested distributions with 10 000 iterations , and ninety five percent confidence intervals calculated by taking the 2 . 5% and 97 . 5% quantiles of the values obtained: n1∼binom ( N , Dispos ) , n2∼binom ( n1 , PCRpos ) , n3∼binom ( n2 , PTbr/Tbb ) , p1∼binom ( Y , Pmat ) , n4∼binom ( n3 , p1/1133 ) . Rogers' [43] model of vector-borne trypanosome transmission was adapted for one host population ( wildlife , x ) and two vector populations ( G . swynnertoni , y1 and G . pallidipes , y2 ) . Although occasional cases of human African trypanosomiasis do occur , the rate of human feeding by tsetse is very low [0 . 1% of feeds on blood meal analysis , 16] , so the human population was not included in the model . The model is described by the following equations: ( 2 ) ( 3 ) ( 4 ) that were simultaneously solved using the lsoda function in the package odesolve in R ( http://www . r-project . org/ ) to give equilibrium conditions for the prevalence of T . b . rhodesiense in wildlife hosts , G . swynnertoni and G . pallidipes and which could be compared to empirically derived estimates of prevalence . Parameters were based on those described by Rogers [43] but adjusted to reflect infection in wildlife ( Table 2 ) . Parameters specific to T . b . rhodesiense , and to G . swynnertoni and G . pallidipes , were used where possible . The proportion of tsetse developing salivary gland infection after feeding on an infected cow is 16% for G . morsitans ( closely related to G . swynnertoni ) and 2 . 1% for G . pallidipes [44]; however wildlife exhibit a degree of trypanotolerance and generally show low parasitaemia [45] , which reduces the probability that a feeding tsetse will develop infection , also indicated by very low infection rates in tsetse fed on wildlife experimentally [46] , [47] . A number of wildlife species do not appear to develop infection with T . brucei s . l . , either proving uninfectible in experimental infections eg baboons [46] or rarely observed with natural infection despite being popular hosts for tsetse , eg elephant [16] , [48] , [49] , so the probability that an infected tsetse feeding on a host results in an infection is also lower compared to cattle . The incubation period of 18 days follows that of Dale et al . [50] for laboratory infections of T . b . rhodesiense in G . morsitans flies; no specific data were available for G . pallidipes so the same value was used . Wildlife host parameters have been chosen to represent all wildlife species . Duration of incubation period and duration of infection are therefore estimated mean values from experimental infections of wildlife [46] , [51] , [52] . Although age prevalence patterns suggest the development of some immunity to T . brucei s . l . in lions [53] , experimental infections do not indicate a clear immune period in other species [46] . SNP has high densities of both wildlife [54] and tsetse [34] . All statistical analyses and model solving were carried out using R 2 . 12 . 1 ( The R Foundation for Statistical Computing , http://www . r-project . org ) .
In total , 6455 tsetse were dissected and examined , comprising 4356 G . swynnertoni ( 2759 females , 1597 males ) and 2099 G . pallidipes ( 1472 females , 627 males ) . Overall , trypanosomes were observed ( in mouthparts , midgut , or both ) in 9 . 2% of G . swynnertoni ( females 10 . 2% , males 7 . 5% ) , and 3 . 7% of G . pallidipes ( females 3 . 9% , males 3 . 2% ) examined . No salivary gland infections were observed . Using the classical trypanosome species identification based on the location of parasites within the fly , the prevalence of T . vivax-like , T . congolense-like and T . brucei-like trypanosomes is shown in Table 3 . For 5428 flies ( all those sampled in 2006 ) , all midguts where trypanosomes were observed ( n = 133 ) were analysed by PCR ( Table 4 ) . No flies were found with salivary gland infections . The prevalence of flies with trypanosomes in the midgut on dissection/microscopy , which were also midgut PCR positive ( Dispos×PCRpos , assumed to represent T . brucei s . l . immature infections ) was 0 . 83% in G . swynnertoni and 0 . 71% in G . pallidipes . All midguts that tested positive for T . brucei s . l . were further analysed with SRA PCR , with 10 out of 43 PLC positive and 1 of these SRA positive , therefore the proportion of T . brucei s . l . testing positive for T . b . rhodesiense was 0 . 1 . Using the expression in Eq . 1 , this gives a predicted prevalence of transmissible T . b . rhodesiense infections of 0 . 010% for G . swynnertoni and 0 . 0085% for G . pallidipes ( Table 4 ) . The prevalence was also calculated separately by sex and using sex-specific maturation ratios of 0 . 21 for males and 0 . 044 for females [41] . The predicted prevalence of T . b . rhodesiense mature infections in G . swynnertoni was 0 . 016% for males ( the number of flies testing positive on dissection/microscopy and PCR out of the total number examined was 11/1448 ) and 0 . 0035% for females ( 20/2289 ) , and in G . pallidipes was 0 . 019% for males ( 5/541 ) and 0 . 0024% for females ( 7/1151 ) . Midguts from 78 flies with no trypanosomes observed on microscopy were also analysed by PCR . Of these , 3 . 8% ( n = 3 ) tested positive for T . brucei s . l . . None of these tested positive with PLC or SRA . Assuming equilibrium , the model yielded prevalences of T . b . rhodesiense of 0 . 0064% in G . swynnertoni and 0 . 00085% for G . pallidipes . The model predicted the prevalence of T . b . rhodesiense in wildlife hosts to be 2 . 5% , which is within the range of reported prevalences in wildlife in SNP of 1 . 8% and 4 . 3% [55] , [56] . The results of all three approaches are presented in Table 5 .
In this study we present data obtained from three different approaches to measuring the prevalence of transmissible T . b . rhodesiense infections in tsetse populations in Serengeti National Park . Fundamental difficulties have been identified associated with the detection of trypanosome infections in tsetse , requiring new approaches to move beyond generation of infection prevalence data to make inferences about transmissibility . The three approaches used in this study confirmed the prevalence of T . b . rhodesiense in SNP to be very low . The prevalence of T . brucei s . l . measured by dissection/microscopy was zero , despite confirmation by the other techniques that T . brucei s . l . was circulating in the area , and evidence of infection in wildlife and human hosts , highlighting a common problem with this technique . The results from PCR analysis of tsetse midguts were used to generate a measure of transmissible infections . In addition , a mathematical model of disease transmission used to predict the prevalence of transmissible infections based on other parameters for this system , confirmed the low prevalence gained by other approaches was compatible with the prevalence of T . b . rhodesiense in wildlife hosts reported in SNP . This study highlights specific challenges in measuring transmissible T . b . rhodesiense infections in tsetse , which have important implications for assessing this variable , and interpreting temporal and spatial patterns of infection in affected areas of Africa . These results illustrate the difficulties of dissection/microscopy techniques , which in this study estimated the prevalence of T . brucei s . l . in tsetse populations as zero , despite strong evidence to indicate the presence of infection in tsetse using other techniques , and evidence for circulation of T . b . rhodesiense in vertebrate hosts in the same area [30] , [31] , [55] . The low prevalence commonly obtained through dissection/microscopy is often attributed to low diagnostic sensitivity of this technique , and there is evidence that some infections which would be classed as immature by microscopy may actually be transmissible . For example , inoculation of trypanosomes found in the mouthparts from flies with trypanosomes present in the mouthparts and midgut by dissection did give rise to T . brucei s . l . infections in mice , both in laboratory and field studies [57] , [58] , and PCR of dissection-negative salivary glands revealed additional T . brucei s . l . infected flies in Glossina palpalis palpalis in Cote d'Ivoire [59] . Whilst this may play a part in the low prevalence observed , the use of other techniques in this study confirmed the prevalence to be extremely low , and the prevalence of zero by dissection/microscopy in this study is more likely attributed to insufficient sample size than low sensitivity . With a prevalence of 0 . 01% ( the highest of the estimates in this study ) it would be necessary to examine around 30 000 flies to detect a difference from zero with 95% confidence . Dissection/microscopy has a number of other disadvantages: it is time consuming and requires skilled technicians , and whilst it does not require substantial investment in technology , this may be outweighed by high staff costs . Identification of species , mixed infections and immature infections is unreliable , particularly if other trypanosome species are also of interest . Furthermore dissection/microscopy alone cannot differentiate between T . b . brucei and T . b . rhodesiense . The dissection/microscopy technique was first discussed in detail by Lloyd and Johnson in 1924 as an alternative to cumbersome rodent inoculation studies . However , Lloyd and Johnson relied principally on morphology of the developmental and infective forms , using the location within the fly only as an additional aid . It is clear that in areas where the prevalence is very low , dissection is less than ideal . However , since the majority of historical studies have relied on dissection/microscopy it is important to understand how these data compare to those generated by other techniques if we want to be able to detect temporal trends . PCR-based techniques have the potential to provide a sensitive and specific tool to identify flies carrying T . b . rhodesiense . We found that 30% of microscopy-positive midguts tested positive for T . brucei s . l . by PCR in G . swynnertoni and 41% in G . pallidipes . It is difficult to compare these directly with other studies as protocols vary widely , but between 7 . 9% and 19% of microscopy-positive midguts have been reported testing positive for T . brucei s . l . in these tsetse species [20] , [21] , [25] . However , a PCR positive fly does not indicate a transmissible infection , but only indicates the presence of trypanosomal DNA . Here we have combined PCR data with information on the proportion of immature T . b . rhodesiense infections which mature to the salivary glands to estimate the prevalence of mature transmissible infections . The prevalence was within the confidence limits of dissection/microscopy and similar to the predictions of the model . Prevalence was higher in males than females , reflecting the increased probability of maturation in males [41] . Although in this study , dissection/microscopy were carried out prior to PCR , the increased likelihood of detecting immature T . brucei s . l . in midguts by PCR means the sample size can be lower for the equivalent precision , reducing field costs and time compared to the substantial sample sizes needed for dissection/microscopy only . The calculation used to predict the prevalence of mature T . b . rhodesiense infections by incorporating dissection/microscopy and PCR data relied on assumptions regarding the sensitivity of dissection/microscopy for detecting midgut trypanosome infections , and the diagnostic sensitivity and specificity of TBR and SRA PCRs when used on tsetse midgut samples . Whilst identification of trypanosomes in the midgut is widely used in the laboratory there is little data available on the sensitivity of this technique in the field . There is however no evidence to suggest that flies can carry T . brucei s . l . without trypanosomes being present in the midgut . TBR and SRA PCRs have high specificity [40] , [60] . Whilst the analytical sensitivity of TBR and SRA PCRs is known ( they are both able to detect 0 . 1 pg of trypanosome genetic material or less , equivalent to one trypanosome [39] , [40] ) , there is no quantitative data on the diagnostic sensitivity when used on tsetse samples . The diagnostic sensitivity of TBR on blood samples from livestock is 76% [60]; however the number of parasites in tsetse midgut samples is several fold higher than the parasitaemia in livestock ( which is often <10 trypanosomes/ml [40] ) hence diagnostic sensitivity is likely to be considerably higher for tsetse samples . Imperfect test sensitivity and specificity can significantly affect prevalence estimates , particularly when the prevalence is very low [61] . Ideally the sensitivity and specificity of each technique would have been included in the analysis to produce prevalence estimates and confidence intervals that reflect this information . The paucity of data to examine these assumptions illustrates the importance of more critical assessment of these techniques , but likely reflects the difficulty of assessing sensitivity and specificity in the absence of a gold standard technique . In the absence of quantitative data , the most likely violation of the assumptions is that the sensitivity of each technique is not 100% hence the prevalence may have been underestimated . In this study , 10% T . brucei s . l . infections were identified as T . b . rhodesiense . Whilst this is not outside the range of values found in previous studies [62] , a proportion of one third has been more commonly reported [63] . SRA PCR targets a single copy gene , and therefore requires the presence of a large amount of parasite DNA . Despite an initial sample size of over 6000 flies , only ten infected midguts had sufficient genetic material present to check for T . b . rhodesiense , so our estimate of the proportion of T . brucei s . l . which are T . b . rhodesiense is not very precise ( 10% , CI 0 . 2–44% ) . Using the value of 33% resulted in a prevalence of T . b . rhodesiense in G . swynnertoni of 0 . 03% and in G . pallidipes of 0 . 028% . It is interesting that 3 . 8% of microscopy-negative flies tested positive for T . brucei s . l . by PCR . Previous authors have found high prevalences of T . brucei s . l . by PCR ( for example 18% [27] ) , and there are potential explanations for this high detection rate . Flies that test positive on PCR but were microscopy-negative may result from the presence of trypanosomal DNA ( known to be detectable for over 10 days in the absence of live trypanosomes [64] ) or a very small number of trypanosomes for example in a recent blood meal where trypanosomes are not able to establish an infection . Experimentally it has been established that only around 12–43% of susceptible flies feeding on an infected host will develop an immature infection even in teneral flies [44] , [65] . In older flies , the majority of trypanosomes ingested will not develop further . Simple calculations illustrate that if trypanosomal DNA is detectable for 10 days , flies feed every 3 days and 5% of hosts carry T . brucei s . l . , at any one time , up to 17% of flies may have detectable T . brucei s . l . DNA , in the absence of an immature or mature infection . Given the drawbacks of using other techniques , it is reassuring that a model incorporating independently estimated parameters for this system predicted similar values for the prevalence of T . b . rhodesiense in tsetse . Whilst it might seem questionable whether the very low prevalence found by the other techniques is consistent with the reported prevalence of T . b . rhodesiense in wildlife hosts of 1 . 8–4 . 3% [55] , [56] , a simple equilibrium-based model analysis showed that with T . b . rhodesiense prevalence in wildlife of 2 . 5% , the prevalence in tsetse remains below 0 . 01% , and consistent with field measures . For diseases such as HAT where low prevalence raises diagnostic challenges , broad agreement of prevalence estimates using quite different approaches permits a measure of confidence in each . A constraint to going forwards with making assessments of prevalence is the absence of a gold standard technique for identifying transmissible T . b . rhodesiense infections in tsetse . Dissection/microscopy requires prohibitive samples sizes and potentially may not detect all transmissible infections; PCR techniques based on amplification of DNA from midguts rely on assumptions of factors which are known to vary and tests for which the diagnostic performance is poorly defined; models require accurate knowledge of all other parameters in a system and assumptions regarding equilibrium dynamics . Even rodent inoculation may miss infections as rodents often fail to become infected due to their innate resistance to infection . However , approaches for the future are likely to rely on PCR based techniques so it is important that reliable and comparable protocols are developed . Currently , there are many different approaches reported for using PCR data to look at T . brucei s . l . in tsetse populations , including PCR of any organs found infected [25] ( similar to this study although we did not include mouthparts ) , PCR of all organs in the fly if any organ is found infected on dissection/microscopy [59] , [66] and PCR of whole tsetse flies [for example 27] , [28] . This variety of protocols raises two important issues: To interpret data from PCR analysis it is important to be clear what PCR results do or do not represent . For example , identification of T . brucei s . l . DNA by PCR in whole flies does not indicate a mature and therefore transmissible infection , but only the prevalence of T . brucei s . l . DNA . Is it possible to use this measure as a direct indicator of risk ? This approach has been taken for other pathogens . For example in assessing prevalence of West Nile virus in mosquitoes , most screening programs test the whole mosquito , detecting mosquitoes with any trace of WNV present , rather than testing the salivary glands , which would give the rate of transmissible infections [1] . PCR studies to identify the nematodes which cause lymphatic filariasis in mosquito populations give a prevalence of infected mosquitoes , but cannot differentiate between pre-infective L1 and L2 larvae , and infective L3 larvae [10] . However this approach is more common where detecting pathogen presence or absence is the main aim , so the exact nature of the relationship between presence of pathogen DNA and transmissible infections is less critical . Approaches measuring the prevalence of T . brucei s . l . or T . b . rhodesiense DNA , either in infected midguts , in all midguts or in whole flies , are assuming a constant relationship between this measure , and the prevalence of transmissible infections ( in turn assumed to represent human risk ) . In this study we relied on experimental measures of the proportion of midgut T . b . rhodesiense infections which mature to the salivary glands to estimate the prevalence of transmissible infections . However there are two areas for concern with this assumption: ( i ) laboratory studies may not accurately reflect the situation in the field; and ( ii ) this proportion is known to vary with factors such as sex , levels of certain antioxidants , mating in female flies , and environmental factors such as temperature [41] , [67] . While this approach may be suitable for obtaining an approximate measure of prevalence , the validity of the assumptions would be challenged by comparative studies over different spatial or temporal situations where these factors are likely to vary . Interpretation of PCR results from analysis of whole flies or from midguts without prior dissection/microscopy is more problematic . This study illustrates the high proportion of microscopy-negative midguts which test positive by PCR and similar findings are reported from PCR of whole flies [27] , [28] . It is not known how the proportion of flies testing positive by this technique relates to the prevalence of transmissible infections . Approaches involving PCR of salivary glands may hold most promise . PCR of microscopy-negative salivary glands or salivary drops has been shown to increase the prevalence compared to dissection/microscopy alone both in the field [59] and the laboratory [68] . It is not clear what these discrepancies between microscopic and PCR analysis of salivary glands means with regard to transmission and this is an area where further research is required . The second concern is with respect to comparative data analysis , in that the variety of techniques used means it is difficult to assess trends in prevalence . This is a significant problem – prevalences measured in different ways cannot be compared between different areas or times , making it impossible to detect changing disease dynamics and human disease risk , and hindering our understanding of the complex relationships between trypanosomes , hosts and vectors . Agreement on an optimal protocol for the collection and interpretation of data on trypanosome prevalence in tsetse populations would be helpful in generating more comparable data . This study shows that the prevalence of T . b . rhodesiense in G . swynnertoni and G . pallidipes in SNP can be sustained at very low levels . Both the PCR data and the model suggest that G . pallidipes may play a role , albeit a lesser one , in T . b . rhodesiense transmission as well as G . swynnertoni , which has always been regarded as the important vector species in Serengeti . The two species differ in both feeding preferences and vector competence; while both species include suids and bovids in their diet , G . swynnertoni feeds predominantly on warthog while G . pallidipes feeds predominantly on buffalo [69] , [70] . Although both G . swynnertoni and G . pallidipes are known to avoid feeding on man , this effect is particularly evident for G . pallidipes [71] , which likely decreases the importance of this species in human disease transmission . The prevalence found in this study is consistent with that of previous studies by dissection/microscopy [16] , [26] , [72] so we did not find any evidence of long term trends in disease transmission . However , the prevalence in this study does differ significantly from that reported in 2007 of 3% [34] . Whilst this may reflect temporal or spatial variation in prevalence within SNP , our model suggests that a sustained prevalence this high is very unlikely . The low prevalence of T . b . rhodesiense in tsetse found in this study suggests that the risk of HAT to tourists is low . Odour-baited tsetse traps are known to target older flies [73]; flies which bite people are usually younger and less likely to be carrying a transmissible infection since the prevalence of mature infections increases with age [74] . This is consistent with the low number of cases ( <5 per year ) reported in Serengeti , in comparison to the large number of visitors ( almost 100 , 000 per year [32] ) . However , the risk of encountering an infected fly is higher in those who spend extended periods exposed to tsetse in SNP , so it should be ensured that adequate screening and treatment provision is in place to detect cases in park and lodge staff . In conclusion the prevalence of transmissible human infective trypanosomes in tsetse populations is an important parameter but there is no ideal diagnostic test to measure it . While new molecular diagnostic tools offer great potential for epidemiological studies , many challenges remain in the interpretation of field data generated from these tools , and these need to be recognised and addressed . Development of protocols that directly measure the prevalence of transmissible infections , and the consistent application of such protocols , would aid our knowledge of human disease risk , allow detection of spatial and temporal trends in disease transmission and add to our understanding of complex disease systems .
|
Human African trypanosomiasis is a fatal disease that is carried by a tsetse vector . Assessing the proportion of tsetse which carries human-infective trypanosomes is important in assessing human disease risk and understanding disease transmission dynamics . However , identifying flies which carry transmissible infections is difficult , due to potential presence of other trypanosome species in the same fly , and concurrent detection of immature infections which are not transmissible . We used three methods to estimate the proportion of flies carrying human-infective trypanosomes: dissection and microscopic examination of flies to visualise trypanosomes directly in the fly; PCR of fly midguts in which trypanosomes were observed by microscopy; and theoretical analysis using a mathematical model of disease transmission . All three methods found the prevalence to be extremely low . Given the low prevalence , dissection/microscopy requires prohibitively large sample sizes and therefore PCR-based approaches are likely to be of most value . However , interpretation of PCR data is not straightforward; whilst PCR identifies flies carrying pathogen genetic material it does not directly identify flies with transmissible infections . This study highlights the need for a consensus approach on the analysis and interpretation of PCR data to generate reliable and comparable measures of the proportion of flies which carry transmissible human-infective trypanosomes .
|
[
"Abstract",
"Introduction",
"Methods",
"Results",
"Discussion"
] |
[
"public",
"health",
"and",
"epidemiology",
"epidemiology",
"biology",
"microbiology",
"population",
"biology"
] |
2012
|
Using Molecular Data for Epidemiological Inference: Assessing the Prevalence of Trypanosoma brucei rhodesiense in Tsetse in Serengeti, Tanzania
|
The current Ebola virus outbreak has highlighted the uncertainties surrounding many aspects of Ebola virus virology , including routes of transmission . The scientific community played a leading role during the outbreak—potentially , the largest of its kind—as many of the questions surrounding ebolaviruses have only been interrogated in the laboratory . Scientists provided an invaluable resource for clinicians , public health officials , policy makers , and the lay public in understanding the progress of Ebola virus disease and the continuing outbreak . Not all of the scientific communication , however , was accurate or effective . There were multiple instances of published articles during the height of the outbreak containing potentially misleading scientific language that spurred media overreaction and potentially jeopardized preparedness and policy decisions at critical points . Here , we use articles declaring the potential for airborne transmission of Ebola virus as a case study in the inaccurate reporting of basic science , and we provide recommendations for improving the communication about unknown aspects of disease during public health crises .
The current Ebola virus disease ( EVD ) outbreak due to Ebola virus ( EBOV ) infection continues to affect the West African nations of Liberia , Sierra Leone , and Guinea . This outbreak has highlighted the uncertainties surrounding many aspects of ebolavirus virology , including how the virus is transmitted . High mortality and historical case fatality rates , combined with graphic descriptions of the pathology of EVD caused by Ebola virus , have stoked public panic , and a lack of available clinical and public health expertise in treating EVD has led the media to turn to infectious disease scientists for information . This represents a significant involvement in public health efforts by the virology and infectious disease community , on par with the severe acute respiratory syndrome outbreak of 2002–2003 and the 2009 H1N1 influenza pandemic [1 , 2] . A majority of the information presented to news sources has been accurate and factual , helping to inform the public about EVD . However , a small subset of information discussed by scientists has fanned flames of panic among the public , government officials , and policy makers . As professionals with expertise and knowledge , scientists’ position of power over the general public ( and media ) generates a responsibility to convey accurate information [3] . Particularly in areas of scientific inquiry that can be reasonably expected to create a significant public panic—such as commentary on the spread of EVD—scientists must take extra care to convey the radical uncertainty that often surrounds knowledge of emerging infectious diseases .
A particularly concerning lapse in accurate reporting lies in EBOV’s transmission . In September 2014 , an op-ed piece published in the New York Times ( NYT ) , titled “What We’re Afraid to Say About Ebola , ” [4] asserted that the virology community writ large was “loath to discuss openly but are definitely considering in private: that an Ebola virus could mutate to become transmissible through the air . ” Recent studies have examined the mutation rate of the outbreak and determined that the current strains do not have increased rates of mutation compared to previous outbreaks and that these mutations have no apparent effect on virulence [5–7] . The claims of the op–ed were repeated in February 2015: an article published in the journal mBio [8] provided a thorough review of the questions still remaining for the current EBOV outbreak . Toward the conclusion of the article , the authors state: Both articles were syndicated widely , with the NYT piece being covered internationally [9–11] and the mBio article receiving extensive coverage within the United States [12–14] . While a number of agencies questioned and critiqued the claims [15 , 16] , large media outlets circulated the claims of the op-ed and paper with little to no accompanying criticism . The mBio article , in particular , draws conclusions that eschew distinctions between aerosolized droplets and airborne droplet nuclei , to dramatic effect . Aerosol transmission involves any transmission mediated by aerosol droplets . Droplets are typically large in size ( >5 μm ) and cannot travel beyond the immediate vicinity of an infected individual , and they typically are only present in the later stages of EVD [17] . EBOV aerosol transmission is not a given , however , with studies in animal models providing conflicting evidence about the likelihood of transmission [18–20] . Even less can be said about the airborne transmission of EBOV between humans—that is , transmission that involves droplet nuclei ( <5 μm ) that can remain suspended in the air for prolonged durations . To date , there have been no studies conclusively demonstrating EBOV aerosol transmission . This is an obvious knowledge gap that needs to become an investigative priority for future response efforts . Osterholm et al . note these distinctions in the body of their review , yet their conclusions do not match the nuances of aerosol transmission . Moreover , a key claim of theirs—that there is a series of unexplained transmission events in the epidemiological data collected on Ebola virus—does not entail conclusive evidence of novel transmission . There are a number of other plausible reasons for these unexplained events; the most obvious of these is that a self-report of contact history can be unreliable . We don’t discount the importance of these points in the context of a scientific review; rather , the reasons for questioning EBOV’s transmission mechanisms do not entail the claim that airborne transmission of aerosolized respiratory droplets is likely ( or worth considering yet as a matter of public policy ) . While the science supporting or refuting the aerosol transmission of EBOV remains unclear , the word “aerosol” carries significant weight in public discourse . It is far from clear whether media coverage of these articles has articulated the speculative nature of the claims made by Osterholm et al . Moreover , attempts to push back against these claims after the fact are likely to be met with suspicion; public trust in science is at an all-time low [21 , 22] . Given what “aerosol” means to members of the public without an infectious disease background , merely suggesting aerosol spread as a route of EBOV transmission without further definition is misleading . To date , evaluations of the effectiveness of scientific communication during the current EBOV outbreak have been focused on the larger organizations ( e . g . , WHO , the Centers for Disease Control and Prevention [CDC] , etc . ) and their shortcomings [23] . However , the basic science community has had a large input into the public discussion surrounding EBOV . The lay public , policy makers , and officials all turn to basic scientists at times of uncertainty , and that creates a difficult situation that should be managed with caution .
Matters of uncertainty are particularly fraught for scientific communication . Low probability events—or even logically or biologically possible events that cannot have probabilities assigned to them , such as a change of EBOV’s transmission characteristics—are likely to cause confusion and panic . There is a well-established literature on the psychology of human decision-making , suggesting that humans assign higher subjective probabilities to events of great significance [24] . In these circumstances , possibilities become probabilities , and probabilities become likelihoods . The upshot of this is clear . Simply articulating that we should not assume that an event won’t happen becomes a tacit endorsement that the event could happen , along with all the psychological baggage that entails . The startling panic that has occurred in the context of the current Ebola virus disease outbreak is evidence of what does happen when the mere possibility of something occurring provokes a disproportionate response . The responsibility to communicate well can be derived from a number of sources; for our purposes , the central concern is the vulnerability of others to misunderstandings brought about by an improper communication of risk . Scientists , particularly in infectious disease , generate claims and evidence that are probabilistic in nature . Those scientists , presumably , have gained the skills to interpret scientific evidence and understand what the information does—and doesn’t—tell us about the world . These skills , when applied to a broad public forum , generate a responsibility for scientists to do their part to prevent misinformation from spreading; unfortunately , they can do as much harm as good without responsible communication practices .
When communicating more broadly—and in a world of open-source publications , “broadly” should be taken as a given—different strategies are required to discharge the responsibility to communicate accurately . The first is epistemic humility: probabilistic claims should be couched in terms that emphasize the risks of false-positives and false negatives . This is particularly important when we are unable to determine the likelihood of our claims being right—or wrong . When there are no probabilities to assign , this lack of knowledge should be emphasized . Second , language should be chosen to demonstrate relative , rather than absolute , credence in certain claims . In the context of the mBio paper , it would be more useful to make claims about the likelihood that unexplained transmissions were due to errors in reporting , or recall , rather than a new mode of transmission . It is true that we can’t demonstrate that these transmissions weren’t airborne , but it is likely that we can make a partial ordering of scenarios and identify airborne transmission from within a range of possibilities . Fortunately , the current EBOV outbreak has not threatened the target audience of these opinions ( e . g . , United States , Canada , and Mexico ) , thus limiting the political will to engage in expensive or controversial ( e . g . , enforced quarantine ) public health practices in the name of combating an airborne EBOV . Nonetheless , these commentaries have tremendous potential to impact public health policy and emergency planning moving forward . The potential of EBOV spread via aerosol routes drastically changes health care and public health interventions . The impact of an airborne EBOV on public health and clinical medicine would entail large structural changes in contact tracing , the use of PPE , and patient isolation . It could also deter people from seeking medical treatment due to concerns about increased transmissibility , which would disrupt efforts to diagnose and treat infected individuals early . Our best response to an Ebola virus outbreak involves rapid contact tracing; a decrease in help-seeking behavior could fuel an outbreak if transmission chains are hidden from public health authorities . Given the immense costs and the possibility of public panic , it is imperative that more research is done before decisions are made . Until then , the life sciences must ensure that when public statements are made to the media , policy makers , and the general public , the most accurate depiction of scientific knowledge—including uncertainty—is conveyed . The potential impact of the Ebola virus transmission articles that we’ve discussed is illustrated by the public panic that surrounds the articles and the Ebola virus disease more generally . It is up to the scientific community as a whole to ensure that science drives discussions of emerging infectious disease and outbreak management in the future , and that it does so responsibly .
|
Basic scientific research is now considered an integral component of the fight against emerging infectious diseases like Ebola virus . The recent Ebola outbreak , however , demonstrates how the ineffective communication of basic science can stoke public panic more than it provides helpful tools to responders; basic science trades in probabilities and uncertainty , while public communication tends to favor more categorical claims . Here , we discuss the ethics of communicating scientific results , using , as a case study , the recent controversy over whether basic life sciences research demonstrates that Ebola could become transmissible via airborne respiratory droplet nuclei—popularly known as a virus becoming “airborne . ” We show how the science does not demonstrate this possibility , despite claims made in the popular and scientific press . We then recommend that uncertain scientific results in the context of public health crises ought to be communicated with humility , an emphasis on what is unknown , and a clear outline of the kinds of evidence that would give proof to controversial claims .
|
[
"Abstract",
"Introduction",
"“Airborne”",
"Ebola:",
"Potential",
"Pandemic",
"or",
"Tempest",
"in",
"a",
"Teacup?",
"Managing",
"Uncertainty",
"Moving",
"Forward"
] |
[] |
2015
|
Effectively Communicating the Uncertainties Surrounding Ebola Virus Transmission
|
Restoring natural speech in paralyzed and aphasic people could be achieved using a Brain-Computer Interface ( BCI ) controlling a speech synthesizer in real-time . To reach this goal , a prerequisite is to develop a speech synthesizer producing intelligible speech in real-time with a reasonable number of control parameters . We present here an articulatory-based speech synthesizer that can be controlled in real-time for future BCI applications . This synthesizer converts movements of the main speech articulators ( tongue , jaw , velum , and lips ) into intelligible speech . The articulatory-to-acoustic mapping is performed using a deep neural network ( DNN ) trained on electromagnetic articulography ( EMA ) data recorded on a reference speaker synchronously with the produced speech signal . This DNN is then used in both offline and online modes to map the position of sensors glued on different speech articulators into acoustic parameters that are further converted into an audio signal using a vocoder . In offline mode , highly intelligible speech could be obtained as assessed by perceptual evaluation performed by 12 listeners . Then , to anticipate future BCI applications , we further assessed the real-time control of the synthesizer by both the reference speaker and new speakers , in a closed-loop paradigm using EMA data recorded in real time . A short calibration period was used to compensate for differences in sensor positions and articulatory differences between new speakers and the reference speaker . We found that real-time synthesis of vowels and consonants was possible with good intelligibility . In conclusion , these results open to future speech BCI applications using such articulatory-based speech synthesizer .
In the past decades , brain-computer interfaces ( BCIs ) have been developed in order to restore capabilities of people with severe paralysis , such as tetraplegia or locked-in syndrome . The movements of different effectors , like a computer mouse or a robotic arm , were successfully controlled in several BCI studies , with increasing precision [1–9] . In the case of severe paralysis including aphasia ( e . g . , locked-in syndrome ) , ways of communicating can be provided by BCI approaches , mostly through a letter selection or typing process [1 , 10 , 11] . However , speech remains our most natural and efficient way of communication . BCI approaches could thus also be applied to control a parametric speech synthesizer in real-time in order to restore communication by decoding neural activity from speech processing brain areas [12 , 13] . Such perspective is indeed supported by an increasing number of studies reporting encouraging performances in decoding speech utterances , including phones , words or even full sentences , from brain activity [14–21] . Several brain areas are involved in speech processing , forming a wide cortical network , classically modeled by a ventral and a dorsal stream [22 , 23] . Because this speech network is widely distributed , a choice needs to be made on the cortical areas where to extract and decode signals for a speech BCI . One possibility is to probe signals from auditory areas , which encode the spectro-temporal representation of speech . However , auditory areas are widely involved in the sensory perception and integration of all sounds a person is exposed to , including speech and non-speech environmental sounds . For this reason , probing neural activity in areas more specifically dedicated to speech production might be more relevant for conversational speech production with BCI . It should be noted that although aphasia caused by strokes very often affect the articulatory speech motor cortex or other cortical areas necessary for speech production , this is not the case for other types of aphasia , such as in locked-in patients or patients with amyotrophic lateral sclerosis , for whom cortical speech activity can be intact or largely conserved and thus exploitable in a BCI perspective . In such case , and if successful , continuous speech restoration would be more beneficial than an indirect communication scheme such as a letter selection process . Previous anatomical and functional data indicate that the speech sensorimotor cortex exhibits a somatotopic organization mapping the different articulators involved in speech production [24–28] , the detailed dynamics of which can be highlighted using high-density neural recording [21] . Interestingly , it was further showed recently that during speech production , the activity of the speech sensorimotor cortex is rather tuned to the articulatory properties of the produced sounds than to their acoustic properties [29] . While neural data could be decoded directly into acoustic parameters , these data thus support our hypothesis that a relevant strategy could be to consider a more “indirect” approach accounting for the articulatory activity of the vocal tract under control of the speech sensorimotor cortex to produce sounds . In such approach , cortical signals will be probed and decoded to control in real time a parametric articulatory-based speech synthesizer having enough degrees of freedom to ensure continuous intelligible speech production . Interestingly , these articulatory features are generally considered lower dimensional and varying more slowly in time than acoustic features , thus possibly easier to predict from cortical signals . Articulatory-based speech synthesizers are able to generate an intelligible speech audio signal from the position of the main speech articulators: tongue , lips , velum , jaw , and larynx [30–36] . Articulatory-based synthesizers are mainly divided into two categories: physical or geometrical approaches ( such as in [36–39] ) , which aim to model the geometry of the vocal tract and its physical properties , and machine-learning approaches ( such as [34 , 35] ) , which use large databases to automatically learn the mathematical relationship between articulatory and acoustic data . Here we made the choice to build a synthesizer using a machine-learning approach . Indeed , to synthesize speech , geometrical and physical models must first solve the difficult inverse problem of recovering articulatory parameters from the acoustic of each calibration sequence . By contrast , with machine-learning approaches , a large articulatory-acoustic database must be recorded , thus avoiding this issue for all the sentences of this database . For BCI applications , one will require a parallel dataset of brain signals and control parameters for the synthesizer: here we can use known data from the articulatory-acoustic database . Moreover , geometrical and physical models need high computation power while , once trained , a machine-learning model is very fast to apply for real-time synthesis . Finally , we showed in a previous study that such machine-learning approach is robust to noisy input parameters [35] , which is a non-negligible asset for BCI applications , when the decoding of neural data results in non-perfect signals . Interestingly , articulatory-based synthesizers can be controlled with about 10 continuous parameters [35 , 40] , which is of the order of the number of degrees of freedom controlled simultaneously in recent complex motor BCI paradigms in monkeys [41] and human participants [9 , 42] . However , it remains unknown whether a given articulatory-based speech synthesizer built from articulatory-acoustic data obtained in one particular reference speaker can be controlled in real time by any other speaker to produce intelligible speech . In this context , we present here an articulatory-based speech synthesizer producing intelligible speech that can be controlled in real time for future BCI applications . This synthesizer is based on a machine-learning approach in which the articulatory data recorded by electro-magnetic articulography ( EMA ) is converted into acoustic speech signals using deep neural networks ( DNNs ) . We show that intelligible speech could be obtained in a closed-loop paradigm by different subjects controlling this synthesizer in real time from EMA recordings while articulating silently , i . e . without vocalizing . Such a silent speech condition is as close as possible to a speech BCI paradigm where the synthetic voice replaces the actual subject voice . These results thus pave the way toward a future use of such articulatory-based speech synthesizer controlled by neural activity in a speech BCI paradigm .
All subjects gave their informed consent to participate in the study , which was approved by the local ethical committee of Grenoble for non-interventional research ( CERNI ) under approval No . 2016-01-05-82 . Four French native speakers ( 1 female , 3 males ) participated in the study . One male subject was the reference speaker from whom data the synthesizer was built , and all four subjects then controlled in real time the synthesizer . In a first step , we designed an intelligible articulatory-based speech synthesizer converting the trajectories of the main speech articulators ( tongue , lips , jaw , and velum ) into speech ( see Fig 1 ) . For this purpose , we first built a large articulatory-acoustic database , in which articulatory data from a native French male speaker was recorded synchronously with the produced audio speech signal . Then computational models based on DNNs were trained on these data to transform articulatory signals into acoustic speech signals ( i . e . articulatory-to-acoustic mapping ) . When considering articulatory synthesis using physical or geometrical models , the articulatory data obtained by EMA can be mapped to the geometrical parameters of the model [39] . Here we consider a machine-learning approach in which the articulatory data obtained by EMA is directly mapped to the acoustic parameters of a vocoder . In a second step , four speakers controlled the synthesizer in real time . As built , the synthesizer could only be used on the reference data and could not be directly controlled by another speaker or even by the same speaker in a different session . Indeed , from one session to another , sensors might not be placed at the exact same positions with the exact same orientation , or the number of sensors could change , or the speaker could be a new subject with a different vocal tract geometry and different ways of articulating the same sounds . In order to take into account these differences , it was necessary to calibrate a mapping from the articulatory space of each new speaker ( or the same reference speaker in a new session ) to the articulatory space of the reference speaker , that is , an articulatory-to-articulatory mapping ( Fig 4A and 4B , left blue part ) . To achieve this calibration , we acquired articulatory data from the new speaker that corresponded to known reference articulatory data . This calibration model was then applied in real time to incoming articulatory trajectories of each silent speaker to produce continuous input to the speech synthesizer . Since the subjects were in silent speech and thus no glottal activity was available , we chose to perform the synthesis using the fixed-pitch template-based excitation , and in order to reduce the number of control parameters , we chose the synthesis model using 14 articulatory parameters since results showed that it was able to produce fully intelligible speech ( see first part of the Results section ) . Fig 4C summarizes the whole experimental protocol , which is detailed in the following sections . The quality of open and closed-loop speech synthesis was assessed in two ways . We first carried out a qualitative evaluation , in which the acoustic signals obtained by offline synthesis from the reference database corpus ( reference offline synthesis ) or during closed-loop experiments ( closed-loop synthesis ) were compared with the original signals processed through analysis-synthesis . Analysis-synthesis was performed by converting the audio signals into mel-cepstrum ( MEL ) coefficients , which were then directly converted back into audio signals using the MLSA filter and template-based excitation . Such signal is referred here to as the ‘anasynth signal’ . This conversion is not lossless , though it represents what would be the best achievable quality for the synthetic speech signal in the present context . Then we carried out a quantitative evaluation of our system through an intelligibility test . Twelve subjects participated to this test . All participants were French native speakers with no hearing impairment . For the offline reference synthesis , the evaluated stimuli consisted of the 10 vowels /a/ , /i/ , /u/ , /o/ , /œ/ , /e/ , /y/ , /ã/ , /ɛ˜/ , and /ɔ˜/ , and the 48 VCVs made of /p/ , /t/ , /k/ , /f/ , /s/ , /ʃ/ , /b/ , /d/ , /g/ , /v/ , /z/ , /ʒ/ , /m/ , /n/ , /r/ , and /l/ , in /a/ , /i/ and /u/ contexts ( i . e . , ‘apa’ , ‘iti’ , ‘uku’ , and so on ) . Each stimuli was synthesized and thus evaluated in 5 different conditions: 4 times using a pulse train excitation generated using the original pitch for each different number of articulatory parameters ( 27 , 14 , 10 and 7 ) , and one time using the artificial template-based excitation signal ( corresponding to a constantly voiced sound ) with all 27 articulatory parameters . In the following , these 5 conditions are respectively denoted as Pitch_27 , Pitch_14 , Pitch_10 , Pitch_7 and FixedPitch_27 . This allowed us to evaluate both the influence of the number of articulatory parameters on the intelligibility , and the effect of using or not using glottal activity information ( here , the pitch ) . Indeed , while in the real-time closed-loop experiment presented here no glottal activity is recorded , this glottal activity could be obtained by decoding the neural activity in future BCI application . An additional evaluation was performed for the two conditions Pitch_27 and Pitch_14 , which consisted in directly transcribing 30 sentences ( see S1 Appendix for the list of these sentences ) . For each listener , half of the sentences were randomly picked from the first condition and the other half from the other , ensuring that each listener never evaluated the same sentence twice , and that all sentences were evaluated in both conditions . For the real-time closed-loop synthesis , the evaluated stimuli consisted of the 7 vowels /a/ , /e/ , /i/ , /o/ , /u/ , /œ/ , and /y/ , and the 21 VCVs made of /b/ , /d/ , /g/ , /l/ , /v/ , /z/ and /ʒ/ , in /a/ , /i/ and /u/ contexts . Each listener evaluated 3 repetitions ( randomly picked for each listener ) of each of these 28 items for each of the 4 new speakers . Remind that for the real-time closed-loop synthesis , the stimuli were generated using only the fixed-pitch template-based excitation . In total , each listener had thus to identify 626 sounds ( 10 vowels + 48 VCVs for each of the 5 different offline synthesis conditions , 7 vowels + 21 VCVs , three times for each of the 4 speakers ) and transcribe 30 sentences . The sounds were all normalized using automatic gain control , and played in random order at the same sound level through Beyerdynamic DT-770 Pro 80 Ohms headphones , while the listener was seated in a quiet environment . No performance feedback was provided during the test . For the VCVs and vowels evaluation , participants were instructed to select in a list what they thought was the corresponding vowel in the case of an isolated vowel , or the middle consonant in the case of a VCV sequence . Graphical user interface buttons were randomly shuffled for each subject in order to avoid systematic default choice ( e . g . , always choosing the left button when unable to identify a sound ) . The subjects were told that some of the sounds were difficult to identify , and thus to choose the closest sound among the offered possibilities . The recognition accuracy was defined as Acc = R/N with R the number of correct answers for the N presented sounds of the test . Since each item had exactly the same number of repetitions , the chance level was estimated by AccChance = 1/C , with C the number of different item categories . For the offline reference synthesis , the chance level was thus 1/10 = 10% for vowels , and 1/16≈6% for VCVs , while for the real-time closed-loop synthesis , the chance level was 1/7≈14% in both cases . For the sentences , the subjects were asked to transcribe directly the sentences they were listening to . Results were evaluated using the word accuracy WAcc = ( N—S—D—I ) /N ( with N the total number of words , S the number of word substitutions , D the number of deletions and I the number of insertions ) , which is a commonly used metric in the field of automatic speech recognition .
First , we evaluated the proposed DNN-based articulatory synthesizer described in the Methods section . Fig 5 shows the spectrogram of the original sound for an example sentence ( the sentence was “Le fermier est parti pour la foire” , meaning “The farmer went to the fair” ) , together with the 5 different synthesis . Note that there is speech present in the synthesized sample before the actual beginning of the reference sentence , since no assumption can be made on the presence of the air flow when considering only articulatory data . The corresponding synthesized sounds are provided in S1–S6 Audio Files , further illustrating the good intelligibility of the synthesized sounds when using at least 10 articulatory parameters . Note however that , in the following , the quality of the articulatory-to-acoustic mapping was evaluated subjectively by naive listeners mainly on isolated vowels and VCVs in order to avoid the influence of the linguistic context that tends to over-estimate evaluation results . Fig 6 summarizes the result of the subjective listening test . The recognition accuracy was better for vowels than for consonants for FixedPitch_27 , Pitch_27 and Pitch_7 ( P < 0 . 01 ) , while this difference was only a trend for Pitch_14 ( P = 0 . 0983 ) and Pitch_10 ( P > 0 . 99 ) . For vowels , the recognition accuracy was far above chance ( chance level = 10% ) for all conditions ( P < 0 . 01 , Fig 6A ) and decreasing when decreasing the number of articulatory parameters , ranging from 89% for Pitch_27 to 61% for Pitch_7 . Taking Pitch_27 as reference , this decrease was not found significant for Pitch_14 ( P = 0 . 7116 ) , and significant for Pitch_10 and Pitch_7 ( P < 0 . 01 in both cases ) . No statistically significant difference was observed when not using the glottal activity versus when using the glottal activity ( FixedPitch_27 = 87% , Pitch_27 = 89% , P > 0 . 99 ) . For the consonants , the recognition accuracy was also far above chance ( chance level = 6 . 25% ) for all conditions ( P < 0 . 01 , Fig 6A ) . A decrease in recognition accuracy was also observed when decreasing the number of articulatory parameters , ranging from 70% for Pitch_27 to 42% for Pitch_7 . However , taking Pitch_27 as reference , this decrease was not significant for Pitch 14 ( P > 0 . 99 ) and Pitch 10 ( P = 0 . 6328 ) , and only significant for Pitch_7 ( P < 0 . 01 ) . A significant difference was observed when not using the glottal activity ( FixedPitch_27 vs Pitch_27 , P < 0 . 01 ) . The differences in recognition accuracy for each condition were studied regarding the vowel of the VCV ( Fig 6B ) and the consonant ( Fig 6C ) . Overall the intelligibility was higher when the consonant was in /a/ context ( /a/ being the most represented phone in the corpus , see Fig 3A ) than when in /i/ and /u/ context ( P < 0 . 01 ) , and no significant difference was observed between /i/ and /u/ contexts ( P > 0 . 99 ) : for instance , for Pitch_27 , accuracy decreased from 80% for /a/ context , to 63% and 67% for /i/ and /u/ contexts respectively . Regarding consonants ( Fig 6C ) , no clear differences were observed between the three synthesis Pitch_27 , Pitch_14 and Pitch_10 except for /p/ , /l/ , /d/ , /g/ and /ʒ/ . Clear differences between these three conditions and Pitch_7 were observed for consonants /p/ , /f/ , /b/ , /v/ , /ʒ/ , /m/ , /n/ , /r/ and /l/ . Clear differences were also observed between FixedPitch_27 and Pitch_27 for the unvoiced consonants /p/ , /t/ , /k/ , /f/ , /s/ , and /ʃ/ . Conversely , no significant differences between FixedPitch_27 and Pitch_27 were found for all the voiced consonants , which includes all the consonants chosen for the real-time closed loop synthesis that does not use the glottal activity ( i . e . it is similar to FixedPitch_27 ) . All conditions taken together , best results ( at least one condition above 90% ) were achieved for the fricative consonants /f/ , /s/ , /ʃ/ , /z/ , and /ʒ/ , the nasal consonants /m/ and /n/ , and /l/ . Worst results ( all conditions below 50% ) were achieved for the plosive consonants /p/ , /t/ , /k/ , /b/ and /d/ . Note that there is no clear correlation with the number of occurrences of each phone in the training corpus , since for instance the corpus contained few instances of /ʃ/ , and a large number of /t/ ( Fig 3A ) . Analysis of the confusion matrices can enlighten the sources of synthesis errors ( Fig 7 ) . Each row i of a confusion matrix M corresponds to the ground truth phone pi , while column j corresponds to the phone pj recognized by the listeners , so that a diagonal value Mi , i corresponds to the proportion of occurrences of the phone pi that were correctly recognized , and a value Mi , j outside the diagonal corresponds to the proportion of occurrences of the phone pi that were recognized as the phone pj . The order of the rows and columns of the confusion matrices were automatically sorted in order to emphasize the main confusions by forming high value blocks near the diagonal . The confusion matrices of the perceptual listening test for the condition Pitch_27 ( Fig 7 , middle row ) reflect the global good quality of this synthesis ( indicated by the fact that they are near-diagonal matrices ) . For vowels , six out of the ten vowels were always correctly recognized ( /o/ , /u/ , /a/ , /œ/ , /y/ and /e/ ) . Main errors come from confusions between /ɛ˜/ and /a/ ( 67% of / ɛ˜ / were recognized as /a/ ) , and other errors come from confusions between /ɑ˜/ and /a/ ( 17% of /ɑ˜/ were recognized as /a/ ) , and between /ɔ˜/ and /œ/ ( 17% of /ɔ˜/ were recognized as /œ/ . For consonants , main confusions came from /b/ being recognized as /v/ ( 75% ) , /d/ being recognized as /z/ ( 58% ) , /p/ being recognized as /f/ ( 56% ) and /d/ being recognized as /z/ ( 58% ) . Other more minor errors come from /g/ being recognized as /v/ ( 11% ) , and /k/ being recognized as /r/ ( 19% ) and /t/ ( 19% ) . By comparing confusion matrices of Pitch_27 with those of FixedPitch_27 , we can observe that not using the glottal activity resulted in increased confusions mainly for the vowel /ɑ˜/ ( accuracy going from 83% for Pitch_27 to 58% for FixedPitch_27 ) while no clear difference can be observed for the other vowels . Note that between the two conditions Pitch_27 and FixedPitch_27 , the articulatory-to-acoustic model remains the same , the only change being the excitation signal that is used for the final synthesis with the MLSA filter . Importantly , for the consonants , not using the glottal activity resulted in a drastic decrease in the recognition accuracy of all the unvoiced consonants /p/ , /t/ , /k/ , /f/ , /s/ and /ʃ/ , while all the voiced consonants remained recognized with similar accuracy . Indeed , /p/ was mainly recognized as /v/ ( 72% ) , /t/ as /z/ ( 58% ) , /f/ as /v/ ( 64% ) , /s/ as /z/ ( 86% ) , and /ʃ/ as /ʒ/ ( 86% ) . Note that /v/ is the voiced counterpart of /f/ , /z/ of /s/ and /ʒ/ of /ʃ/ . Hence , the use of the template-based excitation naturally leads to a predictable shift of the unvoiced consonants to their more or less corresponding ( in terms of place of articulation ) voiced counterparts . By comparing the confusion matrices of Pitch_27 with those of Pitch_14 , we can observe that there is no clear pattern of increased confusions . This confirms the results previously obtained from Fig 6 , where no significant differences between Pitch_27 and Pitch_14 were found for both vowels and consonants . Finally , the results of the subjective evaluation on sentences are presented in Fig 8 . While the recognition accuracy for Pitch_27 and Pitch_14 was below 90% for vowels and below 70% for consonants , the word recognition accuracy for the sentences is above 90% for both conditions ( 96% for Pitch_27 and 92% for Pitch_14 ) . Note that the difference in recognition accuracy for Pitch_27 and for Pitch_14 is here significant ( P = 0 . 015 ) .
In this paper we first presented an articulatory-based speech synthesizer built from articulatory-acoustic data from one reference speaker using deep neural networks , and then showed that this synthesizer could be controlled in real-time closed-loop situation by several speakers using motion capture data ( electromagnetic articulography ) as input parameters . Experiments included the same reference speaker in a different session , as well as other speakers . All speakers were silently articulating and were given the synthesized acoustic feedback through headphones . A calibration method was used to take into account articulatory differences across speakers ( and across sessions for the reference speaker ) , such as sensor positioning and ways of articulating the different sounds . Subjective listening tests were conducted to assess the quality of the synthesizer and in particular its performance during real-time closed-loop control by new speakers . We first assessed the intelligibility of the synthesizer itself . Several versions of the synthesizer were built to assess the effect of the number of articulatory parameters ( 27 , 14 , 10 and 7 ) , and the effect of using or not using glottal activity ( by comparing synthesis using a constant artificial pitch , and the original pitch ) . The phone recognition accuracy for offline reference synthesis was far above chance level for all five tested parameterizations , and fully intelligible speech sentences could be produced ( see Fig 8 and S1–S5 Audio Files ) . Most errors on vowels were made between vowels that had close articulatory positions ( e . g . /i/ and /e/ , see Fig 3B ) . Regarding consonants , most errors were made on the plosive consonants , and main confusions were observed within pairs of consonants corresponding to relatively similar articulatory movements in terms of place of articulation: for instance , /b/ is a labial consonant and /v/ is a labio-dental , and /d/ is a dental or an alveolar and /z/ is an alveolar . For /b/-/v/ confusion , this could be explained by a positioning of the EMA coils too far from the lip edges , resulting in a tracking of the lips by the EMA system that did not allow to capture sharp differences between /b/ and /v/ lip movements . A similar interpretation can be given for /d/-/z/ confusions , since in practice the coil had to be attached more than 5 mm back from the tongue tip ( see Fig 2A ) . Moreover , results showed that the accuracy on the VCVs was correlated to the vocalic context , with consonant in /a/ context having a better recognition accuracy . This could be explained by the fact that the phone /a/ is more largely present in the training corpus than the phones /i/ and /u/ ( see Fig 3A ) . However , this is not consistent with the fact that some phones that are less represented in the corpus , like /ʃ/ , have high recognition accuracy , while other phone that are largely represented , like /d/ , have low recognition accuracy . Another possible explanation is that /a/ is the most opened vowel and thus VCVs in /a/ context are performed by movements of higher amplitude , which could be more discriminant . By removing the glottal activity information ( here by using a constant pitch ) , we found that the recognition accuracy significantly decreased for all unvoiced consonants , while remaining roughly the same for all voiced consonants and vowels ( see Fig 6 and top and middle rows of Fig 7 ) . The unvoiced consonants were thus confused with their voiced counterpart ( e . g . /ʃ/ with /ʒ/ ) , or with the voiced counterpart of the consonant they were already confused with ( for instance , /p/ was originally confused with /s/ in the Pitch_27 condition and was then confused with /ʒ/ when using a constant pitch , in the FixedPitch_27 condition ) . This supports the choice we made to keep only 7 consonants for the real-time closed-loop synthesis since no glottal activity was available in silent speech condition . It should be noted that the quality of the synthesis could still be improved by estimating an optimal MLSA excitation source using inverse glottal filtering and that it could be envisioned that the parameters of such supplementary signal be predicted from brain signals in a BCI situation . Regarding the number of articulatory parameters , results showed that using 14 articulatory parameters yields intelligibility scores that are close to the best scores achieved with 27 parameters . This supports the choice that was made to use a synthesizer with 14 articulatory parameters for the real-time closed-loop synthesis . Interestingly , using 10 parameters did not significantly impact the intelligibility of consonants , but started to affect that of vowels , although the accuracy remained at the high level of 67% . Decreasing further the number of parameters down to 7 , significantly impacted the intelligibility of both vowels and consonants . Finally , although the accuracy on consonants was inferior to 70% for 27 and 14 articulatory parameters , this was enough to produce very intelligible sentences , with word recognition accuracy superior to 90% ( see Fig 9 ) . This can be explained by the fact that most confusions were made with similar consonants , thus ensuring a good intelligibility when constrained with closed vocabulary and syntactic rules . Thus , overall , the number of parameters required to achieve a sufficient intelligibility is of the order of the 10 degrees of freedoms that could be controlled successfully in recent state of the art BCI experiments ( Wodlinger et al . 2015 ) . It should be noted that the reduction in the number of parameters was done here in a drastic way either by dropping parameters or by PCA , while more efficient dimensionality reduction techniques could be envisioned such as autoencoders that we previously started to investigate [35] . Next , we assessed the intelligibility of the real-time closed-loop synthesis . In this case , the phone recognition accuracy was again far above chance level , both for vowels and consonants ( Fig 11 ) . Interestingly , this good intelligibility was obtained despite significant trajectory errors made on input control parameters obtained by the articulatory-to-articulatory mapping ( about 2 . 5 mm on average , see Fig 9B ) . This confirms our previous results indicating that DNN-based articulatory synthesis is robust to fluctuations of the input parameters [35] . As expected , the closed-loop synthesis intelligibility was lower than for the reference offline synthesis . However , it was relatively limited . Confusions were similarly distributed in both cases , indicating that using the synthesizer in a closed-loop paradigm mainly emphasized the already existing confusions . The fact that most errors were consistent between offline and closed-loop synthesis suggests that real-time closed-loop articulatory synthesis could still benefit from improving the articulatory-to-acoustic mapping . This could be achieved by efficiently detecting specific constrictions from the articulatory data in order to improve the synthesis of plosive consonants , which are the major source of errors . The presence of additional minor confusions suggests that other aspects might also be improved , such as the articulatory-to-articulatory mapping with a better calibration approach . Indeed , to remain in a situation as close as possible to future BCI paradigms with aphasic participants , the articulatory-to-articulatory calibration step was performed under a silent speech condition . This was also consistent with the fact that the closed-loop condition was also performed in a silent speech condition so that the speaker received only the synthesized feedback , not superimposed on his/her own produced speech . Thus the articulatory-to-articulatory mapping converted articulatory trajectories recorded under a silent speech condition ( for each speaker ) into articulatory trajectories recorded under overt speech condition ( of the reference speaker ) . Previous studies have shown that articulatory movements differ between silent and overt speech , and especially that silent speakers tend to hypo-articulate [57 , 58] . Such phenomenon may thus leads to smaller discrimination of articulatory trajectories during silent speech . Improving the articulatory-to-articulatory and the articulatory-to-acoustic mappings might however not be the sole possibility to improve the intelligibility of closed-loop speech synthesis . Indeed , while results from the evaluation of the articulatory-to-articulatory mapping showed that for most sensors the mean prediction error was lower for Speaker 1 ( the reference speaker ) , the results obtained during the real-time experiment showed that other speakers could achieve a control of the articulatory synthesizer similar to Speaker 1 , in particular for consonants ( see Fig 11A ) . For example , episodes of spontaneous conversation could be achieved not only with Speaker 1 but also with Speaker 2 ( see S1 and S2 Video Files ) . This suggests that other factors come into play for the control of the synthesizer . One possibility is that subjects may adapt differently to the articulatory-to-articulatory mapping errors and find behavioral strategies to compensate for these errors . Here , each subject had about 20 minutes of free closed-loop control of the synthesizer between the two productions of test items , but we could not see any significant improvement over this short period of time . Finding behavioral strategies might thus need a more significant amount of training time . Finally , and according to the results from the offline reference synthesis , all unvoiced consonants were excluded since no glottal activity can be recorded in silent speech condition . In a BCI application for speech rehabilitation , such glottal activity could be predicted from the neural activity , thus allowing the synthesis of all the French phones . To our knowledge these results are the first indication that an intelligible articulatory-based speech synthesizer can be controlled in real-time by different speakers to produce not only vowels , but also intelligible consonants and some sentences ( some spontaneous conversations , while not reported here , could be achieved with 2 of the 4 subjects using only the synthesized audio i . e . the interlocutor could not see the subject articulating ) . These results thus go beyond previous preliminary achievements of speech synthesis from EMA data where discrete sentences could be successfully classified in a closed vocabulary context with training and testing performed in the same subjects [59] . Indeed , here the speech synthesis was performed in real time on a frame-by-frame basis to provide an online audio feedback delivered in real time with a very short time delay ( less than 30 ms ) . Moreover , we showed here that a synthesizer built from a reference speaker data in an overt speech condition could be controlled to produce free speech in real time in a silent speech condition by other speakers with a different vocal tract anatomy and a different articulatory strategy using a simple linear calibration stage . This result is of particular interest for the emerging research field on ‘silent speech interfaces’ , which are lightweight devices able to capture silent articulation using non-invasive sensors and convert it into audible speech [60–63] . Indeed , although the presented EMA-based interface is not strictly a silent-speech interface , the present results indicate that it is possible to synthesize intelligible speech in real time from articulatory data acquired in silent speech condition . Further studies could extend these results using less invasive techniques to obtain articulatory signals , such as EMG [61 , 62] and/or ultrasound signals [63 , 64] . Finally , this study is also a first step toward future speech BCI applications . Here we indeed showed that closed-loop speech synthesis was possible by subjects that had different speech production constrains ( e . g . , different anatomy of the vocal tract , different manner of articulation ) than those of the reference speaker from whom the speech synthesizer was built . This means that differences in anatomical constrains could be compensated by the articulatory-to-articulatory mapping . In the context of a speech BCI paradigm , a similar situation will be encountered , where the synthesizer will be built from a subject different that the BCI participants . In this case , the question will be whether differences in neuronal constrains between individuals can also be compensated by a proper neural signal decoding strategy . Here , the DNN-based mapping approach was robust to trajectory errors of several millimeters that were present in the input signals of the synthesizer resulting from imperfections in the articulatory-to-articulatory mapping . This is encouraging given that decoding neural signal into input signals of the synthesizer will also be imperfect , and suggests that an articulatory-based speech synthesizer such as the one developed and tested here is a good candidate for being used in a speech BCI paradigm . The choice we made here to envision articulatory parameters as an intermediate representation for decoding speech from neural activity recorded from the speech motor cortex . This hypothesis will need to be tested in future BCI experiment and compared to a direct decoding of cortical activity into acoustic speech parameters .
|
Speech Brain-Computer-Interfaces may restore communication for people with severe paralysis by decoding cortical speech activity to control in real time a speech synthesizer . Recent data indicate that the speech motor cortex houses a detailed topographical representation of the different speech articulators of the vocal tract . Its activity could thus be used in a speech BCI paradigm to control an articulatory-based speech synthesizer . To date , there has been no report of an articulatory-based speech synthesizer producing fully intelligible speech , as well as no evidence that such device could be controlled in real time by a naive subject . Here , we present an articulatory-based speech synthesizer and show that it could be controlled in real time by several subjects articulating silently to produce intelligible speech .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"acoustics",
"linguistics",
"medicine",
"and",
"health",
"sciences",
"engineering",
"and",
"technology",
"audio",
"signal",
"processing",
"signal",
"processing",
"social",
"sciences",
"signal",
"filtering",
"tongue",
"audio",
"equipment",
"digestive",
"system",
"vowels",
"speech",
"phonetics",
"physics",
"speech",
"signal",
"processing",
"mouth",
"anatomy",
"equipment",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"acoustic",
"signals"
] |
2016
|
Real-Time Control of an Articulatory-Based Speech Synthesizer for Brain Computer Interfaces
|
Intestinal schistosomiasis is widely distributed around Lake Victoria in Kenya where about 16 million people in 56 districts are at risk of the infection with over 9 . 1 million infected . Its existence in rural settings has been extensively studied compared to urban settings where there is limited information about the disease coupled with low level of awareness . This study therefore assessed community awareness on existence , signs and symptoms , causes , transmission , control and risk factors for contracting schistosomiasis as well as attitudes , health seeking behaviour and environmental antecedents that affect its control so as to identify knowledge gaps that need to be addressed in order to strengthen schistosomiasis control interventions in informal urban settings . The study was carried out in an informal urban settlement where the prevalence of intestinal schistosomiasis was previously reported to be the highest ( 36% ) among the eight informal settlements of Kisumu city . The study adopted cross-sectional design and purposive sampling technique . Eight focus group discussions were conducted with adult community members and eight key informant interviews with opinion leaders . Data was audio recorded transcribed , coded and thematically analyzed using ATLAS . ti version 6 software . Most respondents stated having heard about schistosomiasis but very few had the correct knowledge of signs and symptoms , causes , transmission and control of schistosomiasis . However , there was moderate knowledge of risk factors and at high risk groups . Their attitudes towards schistosomiasis and its control were generally indifferent with a general belief that they had no control over their environmental circumstances to reduce transmission . Although schistosomiasis was prevalent in the study area , majority of the people in the community had low awareness . This study , therefore , stresses the need for health education to raise community's awareness on schistosomiasis in such settings in order to augment prevention , control and elimination efforts .
Schistosomiasis is one of the neglected tropical diseases , found in tropical and subtropical areas of Africa , Asia and Latin America [1] where it is found in both urban and rural settings where sanitation and control efforts are inadequate and populations are impoverished , thereby considered a “poor man's disease” [2] . Worldwide , about 207 million people in 74 countries are infected with schistosomiasis , with sub-Saharan Africa ( SSA ) accounting for 93% of cases [3] . In Kenya , prevalence of the disease ranges from 5% to over 65% in different communities and contributes to significant morbidity [4] , [5] , [6] , [7] . It is also estimated that 16 million people in 56 out of 158 districts are at risk of the disease [8] and over 9 . 1 million are infected in the country [9] . Schistosomiasis is majorly reported for rural settings , while urban settings are often not considered as transmission foci , resulting in limited evidence on the disease and subsequent neglect of urban areas [10] . Schistosomiasis transmission in urban settings may result from contamination by infected migrants in search of employment , and the poor sanitation around freshwater harboring snail intermediate hosts . In Kenya , rapid urbanization amid economic downturn has resulted in increased proportions of people living in absolute poverty in the urban areas [11] , including informal settlements of cities such as Kisumu where overcrowding and inadequate water and sanitation are a major challenge [12] . Moreover , in informal settlements , the low quality of housing and the general lack of basic infrastructure especially sanitation drainage , access to energy and clean water supply result in poor social and environmental conditions like poor living conditions and low level of education [13] which have significant impact on the spread of infectious diseases , including promotion of transmission of schistosomiasis . Schistosomiasis can be controlled using three key approaches which include improved sanitation , health education and mass treatment with praziquantel . The government of Kenya is now providing free treatment of schistosomiasis and soil transmitted helminths to school age children but the adults in the community who are equally vulnerable can barely obtain treatment which is not only inaccessible but also unavailable to them . Promotions of hygiene , access to safe water , and sanitation improvement as well as environmental management are additional interventions for infection prevention . Transmission of infection can also be controlled by targeting snail vectors and avoiding contact with infected waters [1] . However , the success of control initiatives involving the community depend on level of the communities' uptake of the program , which is hinged upon understanding the community knowledge and practices towards the disease and recommended preventive and/or treatment regimens . Indeed , the Kenyan Ministry of Public Health and Sanitation ( MOPHS ) through its national multi-year strategic plan for control of Neglected Tropical Diseases 2011–2015 also recommends that more research is needed on the knowledge , attitudes and practices towards schistosomiasis control [8] . This study aimed at assessing community knowledge , attitudes and practices on schistosomiasis , ( its existence , signs and symptoms , causes , transmission , control and risk factors ) , the findings of which will inform bridging of identified gaps to enhance a successful control programmes in Nyalenda B , an informal settlement in Kisumu city , western Kenya .
This study was approved by the Scientific Steering and the National Ethics Review Committee of the Kenya Medical Research Institute ( KEMRI , SSC # 1841 ) . Permission was obtained from the Provincial Commissioner , Nyanza , the Director of Public Health and Sanitation Nyanza Province , the Town Clerk of Kisumu City and the Municipal Public Health Officer of Kisumu City . Permission was then sought from the area assistant chief and village elders were notified about the study . All participants gave written informed consent . All information given by the study participants were kept confidential and anonymity was highly observed . No personal identifiers were used during data entry and analysis . The study was conducted in Nyalenda ‘B’ sub-location which is an informal settlement located in Kisumu city , western Kenya . The sub-location covers an area of 6 . 1 Km2 with a total population of 32 , 430 people , 16 , 189 of whom are male and 16 , 241 female [14] . The sub-location is headed by an Assistant Chief . There are five units which are further sub-divided into nine sub-units within the sub-location . These sub-units are headed by one or two village elders depending on size . The informal settlement has four municipal-run primary schools ( Joel Omino , St . Vitalis Nanga , Pandpieri and Dunga ) that have been greatly overwhelmed by increased enrolment since introduction of universal free primary education in the country [12] . The main economic activities in the area include fishing , car washing and small business enterprises . Intestinal schistosomiasis is a public health problem in the area and its prevalence of 36% is highest compared to all the other informal settlements in the city [10] . Presence of infected intermediate host snails has also been reported , suggesting active transmission [15] . Poor sanitation prevails with high use of bush or fly toilets [12] and domestic water sources ranging from taps , springs , boreholes , water vendors to Lake Victoria . . The area is served by only one frontline government health facility; but there exist a few private clinics and a private hospital . The study was a cross-sectional , descriptive assessment that employed qualitative methods , including focus group discussions ( FGDs ) and key informant interviews ( KIIs ) . FGDs were conducted with community members to gauge community knowledge , attitudes and practices towards schistosomiasis control . KIIs were conducted on opinion leaders on schistosomiasis its control in terms of water , sanitation and treatment . The KIIs were conducted before the FGDs in order to get views of opinion leaders about schistosomiasis to facilitate focused questions to understand the feelings of the general community members , thus both were used to complement each other . After the tools were developed , they were piloted in an informal setting with similar characteristics as Nyalenda B where we carried out the study . Since generally 4 focus groups are considered adequate for a given research question with a given group of target study participants [16] , we conducted eight FGDs so that each category is represented by two so as to minimize bias of chance . The data was collected by trained KEMRI ( Kenya Medical Research Institute ) personnel using both audio recorders and field notes . Eight FGDs , each comprising 10–12 participants , and 8 KIIs were conducted on purposively selected respondents . The total number of participants was 88 . Four of the 8 FGDs comprised of unmarried youth adults ( 18–24 years of age ) while the remaining four comprised of ever married persons ( >24 years of age ) . The FGDs were conducted separately for different sexes and age groups ( Table 1 ) . KII participants were either married or widowed ( Table 2 ) . Permission for the study was obtained from the local administration following briefing on the study . The study team was then introduced to the village elders who assisted in mobilizing adult household members from the various sub-units ensuring representation of at least one participant in each FGD group . The participants were screened for eligibility before the discussion commenced to avoid selection bias . The criteria was that one must have lived in the study area for more than six months , be an adult and able to articulate their speech bearing in mind the representation from all the subunits . The KII were conducted with one opinion leader from each sub unit . The FGDs lasted one and half hours and the KII took about one hour . Socio-demographic profile questionnaire was administered to all the participants . All the qualitative data collected was transcribed verbatim and the text typed into the computer . The data cleaning and analysis was done using ATLAS . ti version 6 . A code sheet was created following the focus group and the key informant guides after which , the textual data was coded into selected themes and a master sheet analysis was carried out , giving all the responses from the FGDs and KIIs a theme . Thematic analysis [17] was used where responses were categorized into themes and then ideas formulated by looking at the patterns of responses . Analyzed data were presented in text form . Quantitative data from the socio-demographic profiles was analyzed using excel spreadsheets .
Participants perceived schistosomiasis as a very serious disease which would cause even death . Some of them had negative attitudes towards bilharzia treatment , its prevention and also towards those suffering from the infection ( Text Box 3 ) . Community members highlighted that whenever they got sick they would seek some intervention . Going to hospitals , buying and taking pain killers were among remedies most mentioned . Perceived cost of medication appeared to influence behavior . Community members believed that medication for treating schistosomiasis was too expensive for them to access it . On the other hand perceived severity prompted people to seek medical attention; however , home management of schistosomiasis and other infections was majorly practiced . The participants suggested that they would like to learn more about schistosomiasis in the following areas: signs and symptoms , prevention , cure , drugs , how long it takes before it kills someone , how it is spread , who are at risk and which hospital to go to for treatment . They suggested various means that information about schistosomiasis could be passed to them . These were: media , chief barazas ( community meetings ) , church , funerals , posters , door-to-door campaigns , hospitals and over the radio .
Our study found out that the overall level of awareness and knowledge about schistosomiasis amongst community members was low . This was similar to findings of a study done in Ethiopia [18] . In our study , drinking contaminated water was perceived to be the major cause of the infection thus avoiding it was considered a powerful prevention tool . This is in line with other studies [19] , [20] . Poor sanitation , overcrowding , contact with contaminated water , community's level of knowledge , their attitudes and practices are factors that promote the transmission of schistosomiasis [21] . This study observed that community members still defecated in the open and had to go to the contaminated waters of Lake Victoria for various reasons even if they understood the dangers . Acka and others also reported a similar practice of poor hygiene , where , villagers tended to defecate where convenient – still rarely using latrines where available [22] . According to Rey [23] , the habit of bathing after urination or defecation near picnic sites as a religious ritual contributed to maintenance of schistosomiasis transmission . This behavior was also observed by Farroq and Nallat [24] . This study revealed some of the misconceptions that community members have about schistosomiasis . People believe that those who had the infection were cursed because of the distended stomach , or were promiscuous because of blood in urine . However , this finding was not unique to our study . In Cameroon , residents of rural areas related hematuria to excessive exposure to sunlight and sexual intercourse thereby dismissing medical treatment in local hospitals as a result of such beliefs [25] . Stigma is known to increase feelings of fear , shame and reduces people's capabilities to successfully obtain appropriate treatment [26] . In our study those infected were stigmatized , people considered them “to be on their way to death” and if they were female they would be labeled promiscuous . However , they also appreciated that schistosomiasis could have other causes like drinking contaminated water which is a misconception . Another study reported that it was shameful to have blood in urine [27] , again highlighting social stigma that is associated with schistosomiasis . A study conducted in Kenya and Tanzania earlier found that schistosomiasis is considered a minor disease by many communities [28] . The disease was perceived not serious since it does not harm or prevent the victim from eating [29] . Our study found a contrary opinion . The community members perceived it to be a severe infection which would lead to death . We also found out that people would only resort to medical care if they realized that their condition was getting worse . Our other discoveries were similar to findings of another study which inferred that lack of money and disease not being serious enough made people not to go to hospital [30] . Moreover , in Ghana it was also established that perceived severity of the disease was the most important determinant of seeking health care or visiting a health facility [31] . In this study personal susceptibility was considerably high . Everyone deemed themselves at risk of getting schistosomiasis mainly because of the lake . This corresponds to a study in Uganda where the risk of infection was highest amongst those who lived near lakes or rivers [32] . However , children and fishermen were considered to be at the highest risk . Gender was not considered a major risk factor since some people thought that the women were more exposed because they do their domestic work at the lake or with water from the lake and others thought that it was men since they go fishing in the lake . Therefore the main risk factor was occupation and age . In this regard , it is worth pointing out that most community members are aware of the health risks of contact with Lake water , yet they lack the capacity to change their behavior , since they depend on the lake for subsistence . In our study , religious believes influenced health seeking behaviour . There were community members who would neither go to the hospital nor take medication and only believed in spiritual healing . Cost of medication also made community members stay away from seeking care since the perceived cost of treatment was considered too high by participants in our study . For this reason , the few who were perceived to be economically able were the ones who would afford health care according to our study . Likewise , in Cote d'Ivoire , people with higher socio-economic status more frequently sought health care and visited health facilities than people with low socio-economic status [33] . In many cases the disease was left untreated because people felt that they could not afford the treatment or that the drug was not available . Some people managed only the symptoms of the infection by using over-the-counter painkillers . The finding that high cost of praziquantel ( PZQ ) and its unavailability in the hospital impacting negatively on the control of schistosomiasis was not unique to our study . This observation is comparable to a study in Nigeria where the progress in the control of schistosomiasis in the country had been insignificant due to high cost of PZQ [34] . Lack of PZQ in most peripheral health facilities in endemic areas of Ghana [35] also affected health seeking behaviour . Furthermore , of the health care facilities that would prescribe PZQ , only 60% had it in stock . The other reason that our participants gave for not going to the health facility to seek treatment , was that health facilities did not have diagnostic kits and the attitude of the healthcare workers was also discouraging . Other studies showed that health professionals in peripheral facilities referred patients for diagnostic test and/or treatment [30] , [36] . There are people who tend to favor home-based treatment using various herbal remedies to treat cases of schistosomiasis and thus do not resort to medical treatment [37] . We also found a number of participants who only believed in herbal therapy . Whereas earlier studies showed a clear impact of distance on the utilization of health care facilities in rural Nigeria , where utilization declined exponentially with distance [38] , our study showed that participants were not aware of any facility in their close proximity that offered schistosomiasis medication . They had to go to the referral hospitals where getting the drugs for the infection was not obvious and they are located far from their place of residence and therefore they must factor in transport costs . For this reason , and consistent with a study in Niger [39] , people preferred traditional methods of treatment . Another study revealed that over 90% of those that self-medicated or visited chemical shops for treatment did not receive PZQ [30] , consistent with our study where the community appreciated finding painkillers in the medicine shops . In the environment there exists predisposing and enabling factors that influence people to behave the way they do . Just as observed in this study , cultural , socio-economic , geographical access and organizational - perceived quality ( standard of drugs , standard of equipment , competence of staff , attitudes of staff , interpersonal process ) are factors in the environment that can influence treatment seeking behaviour [40] . . Surveys performed in Egypt with young students and adults showed they had good information on schistosomiasis . Despite the fact that they knew they had to avoid being exposed to contaminated water , exposure was occurring for lack of other alternatives [28] . The same situation occurred in our study where people knew that to avoid contact with contaminated water was a healthy behaviour but could not uphold it due to dependency on the water for domestic and economic use including fishing , sand-harvesting and car washing . Participants indicated that there was nothing they could do about their situation because even if they put up latrines they are swept away by floods or they sink after a very short time and so instead of sinking in a latrine they would rather go to the bush or use “flying toilets” ( rap and throw ) . This is in contrary to another study where people depended on the schistosomiasis-infected river for all the domestic needs even where there are alternative sources of water arguing that the river/stream gave them purer water than the hand dug well [29] . The research instruments in this study were administered in English and Kiswahili which made the selection criteria biased towards only those who would speak in those two languages . However , this study area being cosmopolitan , these were the appropriate languages to use . Outcomes of this KAP study can be used to design appropriate education messages , for which one needs to first understand what the gaps are in the knowledge that already exists . Health communications/education programs which form one of the social aspects of control should involve knowing the gap and intervening for it through understanding of perceptions , attitudes , and practices [41] . Beliefs about the threat of the disease ( i . e . perceived severity and perceived susceptibility ) would directly influence the likelihood of taking a recommended action ( i . e . steps to control schistosomiasis ) . These perceptions could be affected by socio-demographic factors and knowledge . Perceived benefits and barriers may determine intended action , while cues to action could be information from others [42] . When disease control interventions are built upon lay knowledge and perceptions they become very effective [27] . This study showed that the level of awareness of schistosomiasis was low , people had negative attitudes towards control interventions and the practices documented only promoted the existence of the disease whereas environmental factors catalyzed its transmission . Although it may not be possible to avoid water contact activities and exposure to the parasite in the absence of other alternatives [28] especially for this community where their main economic activities revolve around the lake , it is believed that to raise community awareness of schistosomiasis as well as treatment-seeking behavior in such endemic areas , provision of health education is a useful strategy . This study , therefore , stresses the need for health education to raise community's awareness on schistosomiasis in order to strengthen the impact of control interventions . Such an education campaign should focus on causes , transmission , treatment and prevention of schistosomiasis . Governments are also urged to equip frontline health facilities with schistosomiasis diagnostic kits and the drugs ( PZQ ) in such endemic areas .
|
Bilharzia also known as schistosomiasis is one of the neglected tropical diseases found in western part of Kenya . The major source of infection is Lake Victoria; however , there is evidence of inland transmission especially within the informal settlements of Kisumu city . Schistosomiasis can be controlled using three key approaches which include improved sanitation , health education and mass treatment with praziquantel . Additional interventions for infection prevention include: promotion of hygiene , access to safe water , and sanitation improvement and environmental management . However , the success of control initiatives involving the community depend on the level of the communities' uptake of the program , which is hinged upon understanding the community knowledge and practices towards the disease . This study therefore collected information from the community to assess level of awareness of schistosomiasis . The findings revealed a low level of awareness in spite of a high prevalence of schistosomiasis . These findings are invaluable in the designing of appropriate education messages targeted at raising community awareness on schistosomiasis and relevant behavioural change required for a successful control programme .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"Methods",
"Results",
"Discussion"
] |
[
"social",
"sciences"
] |
2014
|
Low Levels of Awareness Despite High Prevalence of Schistosomiasis among Communities in Nyalenda Informal Settlement, Kisumu City, Western Kenya
|
To understand the genetic mechanisms leading to phenotypic differentiation , it is important to identify genomic regions under selection . We scanned the genome of two chicken lines from a single trait selection experiment , where 50 generations of selection have resulted in a 9-fold difference in body weight . Analyses of nearly 60 , 000 SNP markers showed that the effects of selection on the genome are dramatic . The lines were fixed for alternative alleles in more than 50 regions as a result of selection . Another 10 regions displayed strong evidence for ongoing differentiation during the last 10 generations . Many more regions across the genome showed large differences in allele frequency between the lines , indicating that the phenotypic evolution in the lines in 50 generations is the result of an exploitation of standing genetic variation at 100s of loci across the genome .
Evolution is the process by which populations adapt genetically in response to selection . Understanding the genetic mechanisms leading to phenotypic differentiation requires identification of the regions in a genome that are , or have been , under selection . Maynard Smith and Haigh [1] proposed to find these loci by searching for genetic hitch-hiking ( now also called “selective-sweeps”[2] ) . Most reported selective-sweeps surround novel , major effect mutations that appeared on a single haplotype before sweeping through a population . A potentially more common type of sweep starts from standing genetic variation present at the onset of selection - the “soft sweep” [3]–[6] . Domestic animals and plants have been used as models to study both simple monogenic and complex polygenic traits . One of the unique features of these populations is that their reproduction has been under human control for a long time and planned selection of individuals have led to an exceptionally wide range of phenotypes within species . Here , we report the results of a genome wide scan , using a 60 k SNP chip , in two chicken lines from a long-term , bi-directional , single trait selection experiment . In the Virginia chicken lines used in this study , 40 generations of selection resulted in a nine-fold difference in 56-day body weight ( the selected trait ) between the lines [7] . Long-term selection experiments , where animal and plant breeders have subjected populations to very strong and meticulously documented directional selection for generations , provide a valuable resource for studying the effects of selection [8] , [9] . The resulting populations are examples of accelerated evolution , where the genetic and phenotypic changes that resulted correspond to changes that would most likely take centuries to achieve with the selection pressures in natural populations . The Virginia lines are a chicken resource population for studying the genetic , genomic and phenotypic effects of long-term , single trait , divergent artificial selection [10] . In 1957 , founders for one high- and one low- body weight line were selected from a 7-way cross between partially inbred White Plymouth Rock chickens . Once a year , with some restrictions imposed to minimise inbreeding , the birds with the highest and lowest 8-week body weight within each respective line were selected as parents for the next generation . After more than 40 generations of selection , there was a 9-fold difference in body weight between the lines [7] and a significant selection response continues through 50 generations of selection . Sublines , where selection was relaxed , were established periodically within both the high and low body weight lines to serve as unselected controls . After some generations , the relaxed lines originating from the high line had lower body weights than the line continuously selected for high body weight , and the relaxed lines originating from the low line had heavier body weights than the selected low line [10] . This pattern reinforces the notion that the observed change in phenotype is indeed due to the continuous selection process . The Virginia lines are a valuable resource for studying the effects of selection on the genome . Of particular importance is that the experiment involved bi-directional selection and that the population history , including population sizes , selection intensities as well as expected and observed selection responses each generation are known . This information allows a better separation of the genomic effects of selection and drift than would otherwise be the case . Together with the advent of a new high-density chicken SNP chip the Virginia lines allows a detailed investigation of the effect of selection on the genome that was not previously possible . One current paradigm for identifying selective sweeps ( hitch-hiking ) is to scan the genome of a selected population for regions of homozygosity ( e . g . Sabeti and co-workers [11] ) . In these analyses , it is assumed that the selected allele was present on a single haplotype at the beginning of selection , which is the case when selection acts on a novel mutation . When the beneficial allele is present on multiple haplotypes , effects of selection will not be detected using this approach . If there is standing ( or cryptic ) genetic variation in a population , which is likely when selecting on mutations that have existed in a population for some time before the onset of selection , the expected pattern of fixation is different [3]–[6] , [12] . Although little is known on how common it is that selection starts from standing variation , initial studies with soft sweeps based on limited marker sets and partial genome coverage [13] , [14] indicate that they might be common . In the Virginia lines , selection started from a mixed population where , at each selected locus , the selected allele might be present on haplotypes from any of the founder lines of the base-population . The selected allele might thus be in high linkage disequilibrium , LD , with some marker alleles ( i . e . SNPs ) and lower LD with other marker alleles that are physically close on a chromosome . Therefore we would not necessarily expect to observe regions with complete fixation of all SNPs around the selected loci , but instead regions where some SNPs display large frequency differences between lines ( in the extreme case fixed for different alleles ) and other adjacent SNPs with little frequency differences between lines . Because evidence for selection is strong in these lines , as shown by the selection response and results from the relaxed lines , our aim was to identify the genetic elements that are the most likely to have been under intense selection by identifying the regions in the genome with the most extreme allele frequency differences between the lines . Here , we report on a genome-wide scan for soft-sweeps designed to identify those SNPs that are in LD with regions in the chicken genome that have been under selection during the breeding of the Virginia lines . Analysing 57 , 636 SNPs in individuals from both the high- and low body weight lines after 40 and 50 generations of selection provides a detailed analysis of both past and present genomic effects of selection as well as insights into how selection has acted on the genome in order to achieve the considerable response to selection .
Genotypes from both the high and the low lines were studied at two time-points , namely after 40 and 50 generations of selection . 57 , 636 SNPs were genotyped in 20 individuals from each line after 40 generations of selection and in 10 individuals from the low line and 49 from the high line after 50 generations of selection . The 60 K SNP chip provides a marker density of approximately 1 marker/15 kb . The extent of LD for the SNPs on this chip in the population is not known , but estimates from genome re-sequencing of the lines suggests an LD block size in these populations of 30 kb ( micro chromosomes ) - 60 kb ( macro chromosomes ) . The extent of LD is expected to be relatively large due to three relatively recent bottle-necks in these populations from breed-formation , inbreeding of lines used to create the base population as well as limited size of the base-population . It is , however , unlikely that any of the SNPs on the chip is causative , but most causative mutations are likely to be linked with at least one marker . 56 , 586 SNPs had genotypes in both lines after 40 generations and 56 , 561 after 50 generations of selection . Of the 32 , 846 SNPs that were polymorphic in generation 40 , 13 , 579 were polymorphic in both lines , 10 , 237 only in the low line , 8 , 032 only in the high line and 998 were fixed for alternative alleles in the two lines . There were more fixed SNPs in the sample from the high line , which was expected based on the empirical observation that the phenotypic response to selection ceased in the low-line about generation 30 ( Figure 1 ) . In generation 50 , an additional 748 SNPs were fixed for different alleles in the two lines – an increase by 75% – most of which were already fixed in one line at generation 40 ( Tables S1 and S2 ) . Figure 2 illustrates the different samples included in the study and the two types of comparisons made using these data . First , allele frequencies at all SNPs were compared across time within each line ( arrows labelled A in Figure 2 ) . This comparison identifies the regions within each line with the largest changes in allele frequencies between generations 40 and 50 . Then , allele frequencies for all SNPs were compared between the high and low lines at two different time points: generations 40 and 50 ( arrows labelled B in Figure 2 ) to identify where in the genome the SNPs indicate the strongest divergence between the lines . To evaluate the significance of observed differences in allele frequencies between lines and sample points within a line , association analyses using PLINK [15] were performed . Within line comparisons of frequencies at generation 40 and 50 ( comparisons A in Figure 2 ) are performed to reveal the effects of recent and ongoing selection . The analyses identified significant differences in many regions dispersed over the entire genome . In the high line , there are highly significant changes in allele frequencies ( p<0 . 001 ) on 10 chromosomes and significant changes ( p<0 . 05 ) on 6 additional chromosomes . For example on chromosome 1 ( Figure 3 ) there were six regions with significant differences between generations 40 and 50 in the high line and those regions are thus the most likely to have been under intense recent selection within this line . The low line only shows significant differences ( p<0 . 05 ) on two chromosomes ( for details see ) . This lower number of currently affected regions is expected given the low response to selection since about generation 30 . Comparisons between the high and low lines at generations 40 and 50 ( comparisons B in Figure 2 ) revealed many highly significant differences between them across the genome at both time points ( Figure S2 ) . For example , there were at least ten regions with highly significant allele frequency differences between the lines on chromosome 4 both at generation 40 and 50 . These regions were likely to have been under intense selection earlier in the selection process . An example of a region with recent divergence between the lines was between 60 Mb and 80 Mb on chromosome 4 ( Figure 4 ) . This could be an interesting region to study further as the different selection response in the lines could be caused by the region containing one or several genes that display genetic background dependent effects ( i . e . epistasis ) . It is noteworthy that despite the relatively low number of individuals , a test for allele frequency differences yields a χ2 value of 80 for a SNP fixed for different alleles in the two lines , which is highly significant even with full Bonferroni correction for multiple testing . For comparisons with other studies it is also useful to realize that χ2 and p values from the allelic χ2-test is the same as a χ2-test of Fst , i . e Fst was also highly significant at all the identified regions across the entire genome . To measure the dynamics within the genomes of the low and high lines , allele frequency changes resulting from 10 generations of selection ( from generation 40 to 50 ) were studied . The loci with the highest rates of allele frequency changes are the most likely regions to contain genes under current selection . In total , there are 24 regions with significant allele frequency changes in at least one line , spread across the genome . Only one region , the beginning of chromosome 7 , was significantly affected in both lines . This lack of correspondence is not entirely unexpected because the lines have undergone a large number of independent fixation events , which makes it unlikely that the same regions are concurrently under selection after 40 generations of divergent selection . Figure 5 shows the results for chromosome 1 . The complete results for all chromosomes are provided in Figure S3 . A complicating factor when attempting to identify regions under selection , especially with small effective population sizes , is to discriminate between the effects of selection and drift . Because the full population history of these lines is known , we could use simulations to evaluate how selection and drift were expected to affect the genome . Previous studies to identify QTLs [7] , [16] , [17] indicate that selection has been strong on many loci in the genome . Using the estimated effects of the QTLs to calculate the selection coefficient ( s ) [18] , [19] , yields values of s in the range 0 . 19–0 . 93 ( Table S3 ) . The simulations show that selection on these loci was sufficiently strong to lead to high probability of fixation after only 10–15 generations for the loci with larger effects and well before generation 40 for many other loci ( Table S4 and S5 ) . After 40 generations , the loci with the largest selection coefficients ( i . e . those representing the effects of significant QTL for the selected trait ) always reaches fixation for the selected allele during the simulations with additive alleles . This is illustrated in Figure S4A , S4B , S4C , where selection is applied on the loci Growth4 ( selection coefficient for males , sM = 0 . 56 , and selection coefficient for females , sF = 0 . 34 ) , Growth6 ( sM = 0 . 93 , sF = 0 . 56 ) and Growth9 ( sM = 0 . 79 , sF = 0 . 48 ) in the high line . Even for the QTL with the smallest effect , Growth12 ( sM = 0 . 31 , sF = 0 . 19 ) , fixation occurred in 85% of the replicates at generation 40 ( Figure S4D ) . Using a selection coefficient half the size of the smallest QTL ( i . e sM = 0 . 15 , sF = 0 . 10 ) and otherwise the same parameters , gives fixation in 45% of the replicates . Keep in mind that these values are for fixation within a single line , they should be squared to obtain the probability of concurrent fixation in both lines . The effective population size , Ne , for the selected lines estimated from the number of parents each generation is ∼35 ( See Table S6 for details ) . Calculations of Ne from the actual pedigrees up until generation 48 show higher values ( 44 . 5 for the high line and 49 . 3 for the low line ) [20] . This demonstrates that the breeding scheme to limit inbreeding has been successful . Using Ne = 35 , the Nes for the previously identified QTL with the smallest effect is , 35×0 . 19 = 6 . 6 , which is greater than 1 implicating that selection is the predominant force at this locus [21] . The simulations support this , as the selected allele is always the one that becomes fixed even for the QTL with the smallest effect . It should , however , be noted that the simulations use effects estimated for statistically significant QTL for the selected trait in a line-cross experiment . As these might include multiple genes affecting the trait and there will be a large number of additional loci with smaller effect on the trait , there will also be a large number of loci for which a balance between selection of drift will have determined which allele has been fixed at the end of the experiment . Our results do , however , show that the population size has been sufficiently large to prevent genetic drift from overriding the effect of selection for the loci with the largest s-values in the selected lines . The simulations also show that for a locus with no selection ( i . e . where there is only genetic drift ) , fixation at this locus in one of the lines only occurs in 10–20% of the replicates when the allele frequencies are intermediate in the base population ( 3/7 and 4/7 ) and in approximately 50% of the replicates when the initial frequencies are more uneven ( 1/7 and 6/7 ) ( Figure S5 ) . The probability of observing fixation of one of the alleles in one line or the same allele in both lines is thus rather high , which is what we observe in the data . Approximately 30% of the SNPs were fixed in one line and not in the other , while at another 45% , they were fixed for the same allele . It should , however , be noted that the group of markers displaying fixation for the same allele in both lines contain both those SNPs that have drifted to fixation and those that were monomorphic in the common base-population . The simulations showed that the probability of fixation of one allele in one line and the other allele in the other line by drift is very low . If the initial allele frequencies in the base-generation are 3/7 and 4/7 ( the base population is a mixture of 7 lines ) the probability of fixation of different alleles is: 2 * ( fixation probability for A ) * ( fixation probability for a ) ) = 2*0 . 038*0 . 094 = 0 . 0072≈0 . 7% , for 2/7 and 5/7 it is 0 . 4% and for A = 1/7 and a = 6/7 it is 0 . 2% . The corresponding numbers for fixation of the same allele are 1% , 6% and 27% , respectively . If we assume a uniform distribution of initial frequencies , the expected proportion of loci fixed for the same allele in the two lines would be 11% and the proportion fixed for different alleles in the two lines 0 , 44% . Since an unknown , but likely substantial , fraction of the SNPs were fixed in the base population , this value cannot easily be compared to the observed data . However , we can compare the observed fixation rate between generation 40 and 50 with the corresponding value from the simulations . In the simulations , the ratio of fixation of the same allele divided by fixation of different alleles is 3 . 98 , again assuming a uniform allele frequency ( an assumption that closely matches the true distribution of segregating SNPs in the data [data not shown] ) , whereas the observed ratio is 2 . 12 . This indicates that about 50% of the fixations for different alleles are due to selection rather than drift . Given the decreased selection response in the low line during this period , it is likely that this figure is lower than the average for the entire selection process . We can also look at the raw number of expected fixations of different alleles to estimate the proportion of SNPs fixed by drift . In the worst-case scenario , where all 56 , 000 SNPs would have segregated at intermediate frequencies ( we used 3/7 and 4/7 as the founder population was a mixture of 7 partially inbred lines ) in the original population , at least 60% of the observed fixations for different alleles at 40 generations would be due to selection . If instead we assume a uniform distribution of allele-frequencies in the base population , the proportion of the markers fixed for alternative alleles due to selection would be 70% . These two alternative ways of separating the effects of drift to selection are in reasonably good agreement , and indicate that the proportion of fixed SNPs due to selection is in the range of 50% to 70% . The observed mean heterozygosity , Ho , was calculated at all autosomal loci in each line at both time points . Ho at 40 generations was 0 . 146 and 0 . 156 in the high and low lines , respectively . After 50 generations , Ho had decreased to 0 . 130 and 0 . 142 . This decrease in heterozygosity was significantly ( p = 0 . 0003 ) larger in the high line , and because the population structure is the same in both lines , it is logical that this excess is primarily a function of selection . We also observed a greater loss of genetic variance in the high line during the last generations of selection when the response had weakened in the low line . All this is consistent with the greater response to selection in the high line during those ten generations of the selection experiment . Selection , however , continues in the low line and thus the difference in heterozygosity loss only provides a minimal estimate of the effect of selection . Several theoretical methods exist for estimating the number of genetic factors ( loci ) that determine a complex trait in an experimental intercross between divergent lines [22] , [23] , [24] , [25] . The procedure of Otto and Jones [25] , which takes information about the difference in mean between the parental lines and the effects of known QTL as input to predict the distribution of remaining additive effects , was used to estimate the number of loci affecting body weight in the intercross . When employing the most recent estimates of QTL effects in the lines [17] , this method predicted that the selected trait - body weight at 56 days of age - was determined by 121 loci ( Table 1 ) . This is consistent with our result from comparison on allele frequencies between the two lines , indicating that the selected trait is determined by a large number of loci . These estimates are , however , only an indication of the true number of data . But it is interesting to note that all data indicate that the number of loci involved is more likely to be large ( in the order of 100s ) rather than small . The genome-wide QTL profile from the scan for loci affecting body weight at 56 days of age in an F2 intercross between the selected lines [7] reveals about 30 discrete peaks , where there is a significant ( nominal p<0 . 05 ) additive genetic effect . We expect the distribution of the estimated genetic effects of these loci , even though they do not reach the experiment-wide significance threshold , to have a distribution that resembles that of the genetic effects of the true loci that determine the line difference . The observed distribution is approximately exponential ( Figure S6 ) , and as a consequence of this , the relative differences in genetic effects between the ordered loci are more or less constant . The s-values for the loci are not dependent on the absolute size of the genetic effects - they are determined by the distribution of the genetic effects for the segregating loci , where in the ordered distribution the locus is and how many loci contribute to the trait . When the distribution of genetic effects is exponential , there is a gradient in the strength of selection on individual loci . The locus with the largest effect will be under more intense selection than the second largest locus and the difference in selection intensity is proportional to the relative difference in their genetic effects . Thus , even though all loci that affect the selected trait will technically be under selection at all times , there will always be a subset of loci under more pronounced selection in the population . In our simulations we show that the loci with the largest effects reach fixation in approximately 10–15 generations in this population . Fixation of these loci will affect the s-values for other loci via , at least , two mechanisms . Firstly , fixation of the strongest loci will increase the relative importance of all other loci . This is because ( for additive genes ) the selection differential scales with the allelic effect in standard deviations . As major genes are fixed , the genetic variance decreases and , as a consequence , so does the standard deviation , which results in an increase of the strength of selection . In the selection experiment , the standard deviations for 8 week weights for males from generations 20 , 40 and 50 were 111 , 139 , and 179 g . The increase in standard deviation makes sense as we are seeing large phenotypic changes . Decreasing coefficients of variation do , however , indicate a decrease in the genetic variance due to selection . Respective values for the LW line , where there is a plateau at the phenotypic level were 63 , 54 , and 60 g . The changes in the relative strength of selection for the loci will depend on how their allelic effects scale – will weight increase with a constant amount over time or scale with increasing mean body weights in the population . This is not known , and cannot be estimated , but it is reasonable to expect a scaling with the mean and if so the relative strength of selection will increase over time for these loci . Secondly , earlier studies have shown that extensive capacitating epistasis in important in this population [16 , Besnier , Pettersson and Carlborg , in preparation] . Due to genetic interactions , the genetic effects of some loci will increase with the changes in genetic background due to selection . In addition , new mutations that occur during the selection process might create entirely new selected alleles with larger selective advantage . In either case , it is unlikely that the current selection profile across the genome is different from what it was at onset of selection . When studying the effect of 10 generations of selection ( from S40 to S50 ) , we observe strong sweep signals in approximately 10 loci , which seems reasonable given the expected distribution of genetic effects . Using a clustering criterion that required a maximum of 1 Mb between subsequent fixed SNPs , there were 116 clusters of at least two SNPs that included 96 . 1% of the 998 SNPs fixed for different alleles and covered 10 . 2% of the genome . This indicates highly non-random spatial distribution of fixed SNPs , which is not what we expect to observe when drift is responsible for a majority of the fixations . Using a more stringent criterion of at least 5 SNPs per cluster , there were 65 clusters including 82 . 3% of the SNPs and covering 8 . 6% of the genome ( Figure 6 ) . In generation 50 , there were 1746 SNPs fixed for different alleles in 163 clusters of at least 2 SNPs or , using the more stringent criterion , 102 clusters with at least 5 SNPs . The number of clusters and proportion of the genome covered is relatively stable to variation in the required number of SNPs in clusters and distance between markers ( Table S7 ) . Both in generations 40 and 50 , more than half of the clusters with at least 5 SNPs were longer than 1 Mb and about a quarter was larger than 2 Mb ( Table S8 ) . The results for clusters with at least 2 SNPs are shown in Table S9 . The size in Mb and cM of the 23 clusters longer than 2 Mb at generation 50 can be seen in Table 2 . The largest physical cluster was 5 . 4 Mb long and located on chromosome 2 . The largest cluster with respect to recombination distance was 23 . 3 cM and located on chromosome 24 . Nine of the largest clusters overlapped with previously identified QTLs . Depending on the criteria used for clustering , we thus observe between 102 and 163 clusters fixed for alternative alleles in the two lines at generation 50 . Irrespective of the criteria used , these clusters contain more than 85% of the SNPs fixed for alternative alleles in the lines . Based on the calculations above , we expect that between 50–70% of the SNPs that are fixed for alternative alleles to be due to selection . If we conservatively assume that the fixed SNPs are distributed randomly inside and outside of clusters , we would then expect between 51 and 114 of the observed clusters to be fixed due to selection , This observation fits well with the expectation of 121 major factors contributing to selection response based on the quantitative genetic theory presented above . As can be seen in Table 2 , the size of the 23 largest clusters , in terms of recombination distances , ranges between 5 . 0 and 23 . 3 cM . Since the probability of recombination occurring in a given region increases exponentially with each generation , these regions were most likely fixed rapidly . As expected from population genetics theory ( see e . g . [21] ) , our simulations show that fixation in a single line for a neutral locus takes considerably longer time than for a locus with s-values similar to those in our data . E . g . in 1000 simulated replicates , the first fixation for a neutral locus occurred after 12 generations and it took 35 generation before fixation was reached in 10% of the replicates . This should be compared with the 4 generations it took to reach the first fixation and the 9 generations it took for 10% to be fixed for the locus with the largest effect ( Table 3 ) . The probability that a region of 5 cM will remain un-altered by recombination during the sweep to fixation in this population is 0 . 078 in 3 generations , 0 . 014 in 5 generations and 2 . 1*10−4 in 10 generations for allele frequencies of 1/7 and 6/7 and 6 . 0*10−3 in 3 generations , 2 . 0*10−4 in 5 generations and 3 . 8*10−8 in 10 generations for allele frequencies of 3/7 and 4/7 . This example illustrates how rapidly the probability of un-altered haplotypes decreases with increasing number of generations to fixation . Our results indicate that it is not that probable that 8 regions larger than 10 cM and an additional 10 regions 5–10 cM would have swept through the selected population in the time required for neutral loci to become fixed , and that selection is a more likely explanation for the fixation of these large clusters . Of the 116 clusters identified after 40 generations of selection , 63% contained at least two consecutive fixed SNPs and could therefore be considered as traditional hard sweeps . However , almost two thirds of them had only two consecutive fixed SNPs , and would not be detected under more stringent clustering criteria . The largest stretch of consecutive markers fixed for different alleles is located on chromosome 2 and contains 8 SNPs . In generation 50 those clusters with at least 5 SNPs overlapped to a large extent with clusters that contained at least 2 SNPs in generation 40 . There were , however , 17 new clusters ( Figure 6 ) , which indicate that there were responses to selection at new loci during the last ten generations . Even though some of these new clusters might be due to drift , a number of them are likely to contain genetic elements that have recently come under effective selection . These could be alleles present already at the beginning , but which were not strongly selected due to a relatively small effect size compared to other loci , that have become more important as the scaled phenotypic variance decreases in response to selection [10] or they could be epistatic loci , the effect of which have increased due to changed genetic background [16] . Some of the loci may also be new favourable mutations , although the present data does not allow us to estimate how frequent these are . Moreover , all significant QTLs identified in the Virginia lines by Wahlberg et al . [17] contained one or several clusters of fixed SNPs ( Figure 7 ) .
Improving our understanding of the dynamic changes in allele frequencies that occur across the genome in response to selection is a challenge in genetics . The selective coefficients of loci will not remain constant throughout the time span of a long-term selection experiment . Loci with the largest effects are most likely to be fixed rapidly , resulting in an increase in the proportion of the total variance contributed by loci with smaller effects . Very little , however , is known about how many loci contribute to a complex trait and how many loci are under most intense selection , i . e . undergoing the most rapid allele-frequency changes , at a given point in time . Several recent studies indicate that the number of loci contributing to complex traits is considerable ( Maize [26] , Illinois corn selection lines [27] , height in humans [28] ) . These insights were , however , gained from studies of the association between phenotypes and genotypes , which implicitly means that there will be limits on the power to detect loci due to sample size . Population history and selection for multiple traits also complicates the picture . Here , we study the genomic effects of intense selection on a single complex trait , which facilitates more precise insights on basic genetic regulation and dynamic changes that occur during selection . Earlier genetic studies of the Virginia lines have shown that more than 20 genome regions ( QTL ) are involved in the genetic regulation of the trait under selection , body weight [7] , [16] , [17] , as well as correlated responses including body composition and metabolic traits [29] . Our estimates of the expected number of loci contributing to the trait indicate that there are many loci that remain unidentified . The probability of fixation for alleles with small effects is higher when selection acts on standing genetic variation than on a new mutation , due to the high likelihood of losing a weakly selected new mutation from the gene-pool in the population . Thus , we would expect our approach to identify a larger number of loci than previous QTL mapping experiments that were based on these data because only loci with rather large genetic effect would have reached the detection threshold in those experiments . This is also what we observed . Both the quantitative genetic and molecular assays used to estimate the number of selected genetic elements are in agreement that we have evidence for there being from 50 up to over 100 regions in the genome that have been under strong selection over the first 50 generations of the selection experiment . This study demonstrates that selection on a complex trait will influence more regions than can be identified even in a comprehensive genetic mapping study , and that the genetic regulation of these traits is complex . Our criterion to require fixation for alternative alleles was very stringent and therefore it is likely that additional regions than those reported were actually under selection . This becomes apparent when examining data from generation 50 , where 1776 SNPs were fixed for alternative alleles in our samples , including 17 new clusters of at least 5 fixed SNPs that were formed during the 10 last generations of the selection experiment . Some of these new clusters may have been selected already earlier but not strongly enough to reach fixation before 40 generations , while some might be due to new mutations that have occurred recently . The footprints of selection include regions spread throughout the genome , including previously identified QTLs as well as those hitherto not implicated to affect body weight in chickens . As regions of fixation , of which many certainly contain selected regions , are identified with very high resolution ( in many cases the clusters cover <1 Mb ) , this information can be useful for identifying candidate genes and mutations involved in the phenotypic response to selection . Assigning the functional effects to the identified regions , however , remains a future challenge . Selection coefficients for the genomic regions ( QTL ) identified in previous studies of these lines ranged from 0 . 93 to 0 . 31 and 0 . 56 to 0 . 19 for high line males and females , respectively , with very similar values for the low line ( Table S3 ) . Even if some of these selection coefficients are overestimates , they are , as a group , very high and illustrate the massive selective pressure on the genome in these lines . The intensity of selection is the most likely explanation for the remarkable differences in allele frequencies observed across the whole genome . Selective sweep analyses are powerful in identifying loci that display directional changes in allele frequencies that correlate with the phenotypic responses to selection . With the advent of more affordable methods for high-density genotyping and genome re-sequencing , it is a cost effective approach to identify loci determining complex traits because small samples from existing , divergent populations can be used [30] . The resolution often allows identification of individual genes and thus provides useful insights to the genes and plausible mechanisms involved in the regulation of the traits for which studied populations differ . A major drawback with the sweep analyses is , however , that they do not provide causal evidence for the involvement of particular genetic polymorphisms in phenotypic expression . The divergent populations studied often differ for multiple traits and it is not possible to identify which of these traits that is affected by the polymorphisms . Furthermore , there are no additional insights to the potential genetic mechanisms involved , i . e . whether genes act independently or through interactions in complex gene-networks . This information is , however , provided in e . g . linkage or association studies . Therefore it is necessary to realise that the selective sweep analyses are not a stand-alone method , but rather an addition to the complete set of tools used for understanding the inheritance of complex traits . An example of how sweep and linkage analyses complement each other is obvious in this population . We have earlier used linkage analysis to identify a network of loci that through strong interactions have a major influence on body weight at 56 days of age [16] . Subsequently we replicated the effects and refined their location in an independent advanced intercross line population ( Besnier et al , in preparation ) . The epistatic network contains four loci on chromosomes 3 , 4 , 7 and 20 and there is a clear overlap between one or several sweeps in each of these regions with the QTL ( Figure 7 ) . Combining this information will be a highly useful strategy for identifying the causal mutations underlying the observed genetic interactions . To conclusively rule out drift as the cause of any given fixation event or other observed change in allele frequencies is not possible . However , all available results indicate that the large phenotypic difference in body weight between the Virginia lines is the result of directional selection acting on a large number regions spread across the genome . The number of loci involved in long-term selection response are likely to be in the 100s for a complex trait and that at any point in time selection is likely to simultaneously act on 10s of loci even in populations of limited size . The identified loci are located with high resolution , which makes them obvious candidate regions for attempts to identify causal mutations . The two lines were from the same founder population and were subjected to 50 generations of artificial selection that have led to changes in trait expression and genetics that may resemble those observed from 1000s of years of natural selection . What we observed is genome wide changes that occurred in an accelerated and directed evolution process . In a broader perspective , the results provide not only insights to the effects of artificial selection , but also what may be expected from natural selection when populations adapt to a new environment . This study shows the inherent power and efficiency in combining data from classic long-term selection experiments with modern genomics tools .
Genotyping was performed on 20 low and 20 high line chickens from generation S40 ( the generation of the parents from the F2 cross described in Jacobsson et al . [7] ) , and 10 low and 10 high line chickens from generation S50 . At the later time point we chose to genotype an additional 39 individuals from the high line because this line still exhibited a good response to selection , whereas the low line appeared to have phenotypically plateaued . The genotyping was performed by the company DNA Landmarks with the 60 k chicken chip produced by Illumina Inc for the GWMAS Consortium . The animal husbandry for the later generations were the same as described for the previous generations [10] . All procedures involving animals used in this experiment were carried out in accordance with the Virginia Tech Animal Care Committee animal use protocols . Individual based simulations with parameters chosen to mimic the Virginia lines were performed with a code written in R [31] , in order to evaluate the probability of fixation for selected and neutral loci . The number of selected males and females , calculated proportion of selected and selection intensity , i , is given in Table S6 . For simplicity , the parameters for generation 5–25 in the selection experiment were used for simulation of selection during all 50 generations , because the effective population size for these generations were close to the effective population size for all generations ( 34 . 55 ) ( Table S6 ) . The number of females per male was thus 48/12 = 4 and the number of offspring per female was six , which is the number that gives a population size ( 6×48 = 288 ) close to the mean population sizes in the selected lines . The selected lines originated from a founder population formed by crossing seven partially inbred ( ∼36% ) lines . We assume that the inbred lines were fixed for all loci , i . e . the starting haplotype frequencies were multiples of 1/7 . Simulations were performed with two linked loci , A and B with alleles A/a and B/b , were selection acts on locus A . The fitness of genotypes AA , Aa and , aa were modelled as 1 , 1-hs and , 1-s , respectively , where s is the selection coefficient and h is used to model dominance . Note that since the selection intensity is different for males and females , there is one selection coefficient for males , sM , and another for females , sF , for each locus . Alleles with additive effects ( h = 0 . 5 ) were assumed for the simulations in this paper . The selection coefficient , s , for a given QTL was estimated as s = i2a/σ [18] , [19] . The selection coefficients for the 11 QTLs with significant additive effects in Jacobsson et al [7] in the low and high line are given in Table S3 . The additive effect , a , and the phenotypic standard deviation , σ , for the QTLs are as described in Jacobsson et al [7] . Simulations were performed for the QTL with the largest effect ( Growth6 on chromosome 4 ) , the smallest effect ( Growth12 on chromosome 20 ) and two additional loci ( Growth4 and Growth9 , on chromosomes 3 and 7 respectively ) . Fixation in the simulations was defined as all individuals in the simulated population being homozygous for the same allele . This should be kept in mind when comparing with the observed results , where fixation is measured in a genotyped sample form the selected poplation . Association mapping was performed using the software package PLINK v1 . 07 [15] . The results in the manuscript are based on asymptotic p-values from the χ2-test ( the assoc option in PLINK ) . As the number of expected in some cells in the χ2-test might be small for some SNPs , we have also computed p-values using a Fisher exact test ( the fisher option in PLINK ) to see that the results did not change due to this . A comparison of the results from using asymptotic p-values with those using a Fisher exact test reveals that even though p-values for individual SNPs are slightly different using the two tests , the overall conclusion does not change . URL: http://pngu . mgh . harvard . edu/purcell/plink/ Calculations of fixation , observed heterozygosity and clusters were performed in R [31] . The significance of the difference in the decrease in heterozygosity at each locus between generations 40 and 50 in the high and low lines was tested by a two-sided t-test in R ( the function t . test ) . The length of the clusters in cM was calculated using the chromosome specific ratios of cM/Mb given in Table 2 in [17] . The length of the clusters in cM was then transformed to recombination frequency using Haldanes map function . The clusters will contain different alleles in the two lines if no recombination occurred during the fixation process or if recombination occurred only in homozygous individuals ( = non-informative ) . The probability for this was calculated as ( ( 1−r ) +r ( p2+q2 ) ) 2Ng , where r is the recombination frequency between first and last position in the cluster , p and q are the haplotype frequencies , N is the effective population size and g is the number of generations until fixation of the cluster . Allele frequencies of p = 1/7 , q = 6/7 and p = 3/7 , q = 4/7 was used in the calculations and 3 , 5 and 10 generations was compared . Changes in allele frequencies between generations 40 and 50 , and also the average over blocks of 5 SNPs were calculated . The mean allele frequency change in each block is compared to the distribution of all blocks across the genome , and if it lies in the 95:th percentile , it is identified as a potential locus under selection . Thus , the number of selected loci per set of 20 blocks is Poission-distributed with average 1 , given the assumption that the blocks are independent . The total number of loci affecting a trait was estimated using equations 6 ( n = D/ ( M−T ) and 12 ( T≈ ( aminnd −M ) / ( nd−1 ) ) in [25] . The estimated number of loci is n , D is half the phenotypic difference between the parental lines ( here 670 . 5 ) , M is the average additive effect of the detected loci , T is the detection threshold , amin is the smallest additive effect among the detected loci , and nd is the number of detected loci . Data on additive effects from previously identified QTLs were from Table 3 in [17] . The estimation was done for the body weight traits with at least 3 identified QTLs .
|
Evolution is the process of change in response to selection . Typically , this results in more or less obvious changes to the appearance and physical properties—the phenotype—of an organism . However , these changes reflect underlying changes to the genome of that organism—the genotype . We examine the genomes of two lines of chickens that share a very recent ancestry but have been subjected to 50 generations of selection for high or low body weight , respectively . The effect of selection on the phenotype was dramatic , where on average high line birds are nine times heavier than low line ones . The effect on the genotype was equally dramatic , with a large number of changes distributed all across the genome . We observed more than 100 regions where different genetic variants were established in the two lines of chickens , which is considerable given the number of generations involved .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"computational",
"biology/population",
"genetics",
"genetics",
"and",
"genomics/animal",
"genetics",
"computational",
"biology/molecular",
"genetics",
"genetics",
"and",
"genomics/complex",
"traits",
"genetics",
"and",
"genomics/physiogenomics",
"computational",
"biology/genomics",
"genetics",
"and",
"genomics/bioinformatics",
"genetics",
"and",
"genomics/population",
"genetics"
] |
2010
|
Genome-Wide Effects of Long-Term Divergent Selection
|
Most current methods for detecting natural selection from DNA sequence data are limited in that they are either based on summary statistics or a composite likelihood , and as a consequence , do not make full use of the information available in DNA sequence data . We here present a new importance sampling approach for approximating the full likelihood function for the selection coefficient . Our method CLUES treats the ancestral recombination graph ( ARG ) as a latent variable that is integrated out using previously published Markov Chain Monte Carlo ( MCMC ) methods . The method can be used for detecting selection , estimating selection coefficients , testing models of changes in the strength of selection , estimating the time of the start of a selective sweep , and for inferring the allele frequency trajectory of a selected or neutral allele . We perform extensive simulations to evaluate the method and show that it uniformly improves power to detect selection compared to current popular methods such as nSL and SDS , and can provide reliable inferences of allele frequency trajectories under many conditions . We also explore the potential of our method to detect extremely recent changes in the strength of selection . We use the method to infer the past allele frequency trajectory for a lactase persistence SNP ( MCM6 ) in Europeans . We also infer the trajectory of a SNP ( EDAR ) in Han Chinese , finding evidence that this allele’s age is much older than previously claimed . We also study a set of 11 pigmentation-associated variants . Several genes show evidence of strong selection particularly within the last 5 , 000 years , including ASIP , KITLG , and TYR . However , selection on OCA2/HERC2 seems to be much older and , in contrast to previous claims , we find no evidence of selection on TYRP1 .
Direct observation of the change in allele frequency over time ( the allele frequency trajectory ) allows one to make powerful inferences regarding whether selection acted on the allele [1 , 2] . However , outside of certain contexts such as experimental evolution of viruses or bacteria [3–6] or analyses of ancient DNA samples [7 , 8] , in most cases such direct observations of allele frequencies at multiple points in the history of a population are unavailable . Instead , selection must be inferred from contemporary , modern data . A wide variety of methods have been developed to detect selection based on patterns observed from modern DNA sequences ( e . g . [9–11] ) . The hitch-hiking effect provides a key signature of selection in modern datasets [12 , 13] . Hitch-hiking causes aberrations in the spatial pattern of genetic diversity , including the site frequency spectrum ( SFS ) [14 , 15] and the pattern of haplotype homozygosity [9] . Methods designed to detect these aberrations are particularly useful in the setting where a single population is surveyed , and the only information available is variation within this single population . The most familiar methods for detecting selection are based on linear functions of the SFS , such as Tajima’s D , Fu and Li’s D , or Fay and Wu’s H [14 , 16 , 17] . An advantage of SFS-based methods is that they do not require the data to be phased . However , these methods have several limitations: they tend to confound selection with other non-equilibrium conditions , such as a fluctuating population size [10 , 18]; they are not suitable for estimating parameters such as the value of the selection coefficient s; significance can usually only be established using an empirical null distribution; and crucially , these methods do not incorporate any features of the haplotype structure . To make fuller use of information provided by phased sequence data , a number of methods have incorporated summary statistics based on haplotype structure . In a broad sense , these methods are based on calculations of haplotype similarity in a window around some core site of interest [9] . Several methods have adapted this general concept to specifically detect ongoing selection [11 , 19 , 20] . More recently , [21] showed that the density of singletons surrounding a focal SNP can be a powerful signal of extremely recent selection in large cohorts . In addition to recent and ongoing selection , it has been demonstrated that these methods have compelling advantages to detecting selection from standing variation [20–22] . However , these methods share the major limitation of SFS-based method in that they are not suitable for parametric inference and it is unclear how to establish significance without use of an empirical null model . Recently , supervised machine learning methods have been proposed as an alternative to traditional summary-statistic based methods ( see e . g . , [23] ) . Standard machine learning techniques applied to population genomic data afford some major advantages over methods based on summary-statistics: standard techniques can produce accurate classifiers based on summaries of the data that live in much higher-dimensional space than the aforementioned summary statistics , and these techniques often encompass a wide space of classification functions that are often non-linear ( see e . g . [24 , 25] ) . Some studies have demonstrated these methods can have improved robustness to demographic model mis-specification [22 , 26] . Although these methods can potentially detect complex patterns left by selection , they demand training on large data sets which typically are simulated using models that may not accurately correspond to the empirical data . In contrast to the aforementioned methods , one might aim to develop a full likelihood method which would take into account the full data set , rather than merely summary statistics . A common strategy for obtaining the full likelihood has been to find the distribution of the genealogy under selection . For example , Krone and Neuhauser described the distribution of the coalescence tree of a locus under weak selection and no recombination [27] . Alternatively , one can describe how the genealogy depends on the trajectory of the derived allele ( first described by [28] ) , and in turn how the trajectory depends on selection . To this end , Coop and Griffiths [29] developed a sampling method for approximating the full likelihood of the selection coefficient . Their method uses sampling to marginalize out two layers of latent variables: the allele frequency trajectory and genealogy of the locus . To estimate the likelihood function , they perform random sampling of both the trajectory , and the genealogy conditioned on the trajectory . Unfortunately , selection likelihood methods that consider the both coalescence and recombination are generally considered computationally intractable . Composite likelihood methods ( see e . g . [10 , 30] ) are able to approximate the likelihood function using tractable expressions for the frequency distribution of a neutral site linked to the selected site [15 , 31] . These methods approximate the joint distribution of frequencies observed at linked sites as the product of their marginals . These approaches can be applied to test for selection , and estimate the strength of selection . The approximations made by composite likelihood methods are more accurate under strong selection ( arguably beyond the strength of most recent selection in humans ) , and thus have less power to detect weak selection—although to some extent low power to detect weak selection is a natural outcome of any selection method . Approximate Bayesian computation ( ABC ) and rejection sampling methods approximate the likelihood function by simulation . One advantage over the composite likelihood approach is that ABC can capture dependencies between linked neutral sites . For example , methods have been used to jointly infer the strength and timing of selection acting on a locus and determine whether a sweep occurred from a de novo vs standing variant [32–35] . However , a major disadvantage of such approaches is that the amount of simulation necessary to obtain an accurate estimate grows dramatically with the dimensionality of the model parameters . There is an additional tradeoff between information utilized from the data and computational burden; as the number of the summary statistics used increases , the number of simulations required to approximate the likelihood at a fixed parameter value also increases ( for a discussion , see e . g . [36] ) . The method we present in this paper draws inspiration from the Coop & Griffiths method [29] , and has several key similarities: our method produces a likelihood; involves integrating out the allele frequency trajectory and genealogy , i . e . , the aforementioned two hidden layers; and both methods account for selection by modeling how allele frequency changes depend on selection . However , there are several key differences between this method and our approach: while Coop and Griffiths assume no recombination of the locus , our method is based on the coalescent with recombination ( i . e . the ancestral recombination graph or ARG ) [37] . Also , whereas Coop & Griffiths simulate random trajectories , we use dynamic programming algorithms similar to those used in Hidden Markov Models ( HMMs ) to completely marginalize the latent trajectory . The hidden states represent allele frequencies and the emission probabilities are coalescence probabilities . While the framework is not a traditional HMM because the process is time-inhomogenous and the emission space changes with time , the similarities with traditional HMMs are , nonetheless , so significant that we will refer to this as an HMM . Lastly , our method uses a novel importance sampling scheme that allows us to sample ARGs assuming a neutral prior , and find the likelihood function at arbitrary values of s; this drastically reduces the amount of ARG sampling necessary . Furthermore , the new method is , to our knowledge , the first that is capable of inferring the allele frequency trajectories for models with recombination and selection using only modern data . We are able to accomplish this task using the aforementioned Markovian structure of both coalescence and the trajectory , forming a HMM over these two hidden states and solving for the posterior marginals of each hidden allele frequency state over time . Recently , Edge & Coop proposed a method to reconstruct changes to polygenic scores over time via such estimates of the local trees , but their method is not suitable for estimating allele frequency changes or selection at individual loci [38] .
We begin with an overview of our method for jointly inferring selection and the allele frequency trajectory , which we summarize in Fig 1 . Our method begins with input in the form of phased SNP data from a linked genomic region ( Fig 1A ) , although technically , it is also possible to use unphased data , and sample possible phasings . While the method generalizes to arbitrary sample size , we recommend using n = 25 − 100 diploid individuals when using ARGweaver as done in this study , as ARGweaver runtime increases quadratically with sample size . The method also generalizes to arbitrarily long regions , although we recommend using regions of 102 − 103 kb , roughly the size of many LD blocks in the human genome [39] . Next , we sample the genealogy at the selected site from its posterior distribution , assuming a neutral model ( Fig 1B ) . By sampling over this distribution , we marginalize out the hidden coalescence events , the first of two latent variables or “hidden layers” in our model . Specifically , we sample the full ancestral recombination graph ( ARG ) of the input haplotypes . The ARG is a graph that summarizes all of the common ancestry and recombination events that have occurred within the sample . We sample ARGs rather than gene trees in order to account for recombination , and to incorporate information from sites in long-range linkage disequilibrium with the selected site . Then we extract the genealogy at the site of interest ( the “local tree” ) and from here on , this is the only component of the ARG that goes into our subsequent calculations . To perform ARG sampling , we choose to use ARGweaver [37] , which is the only currently available method to sample the posterior ARG . In practice , it is possible and straightforward to adapt this method to other ARG inference methods designed for larger samples , but sampling the posterior yields beneficial statistical properties ( see “Importance Sampling” under Materials and methods ) . Then , for each local tree we have sampled , we form a hidden Markov model ( HMM , Fig 1C ) indexed in time according to the discretization chosen for ARGweaver; in this HMM , observed states are coalescence times in this local tree and hidden states are the selected allele’s frequency trajectory over time ( i . e . , the second hidden layer of our overall model ) . We use a discrete-time model of the coalescent process to match the model used by ARGweaver , so that the length of the HMM is of manageable , finite length . Emission probabilities ( i . e . , coalescence probabilities ) depend both on the allele frequency and the current coalescent state . Hence , the model is time-inhomogenous as the coalescent state changes through time . However , the dependence structure is otherwise identical to traditional HMMs and all the usual dynamic programming algorithms apply . The transition probabilities of allele frequencies depend on the selection coefficient s , the parameter we are ultimately interested in estimating . Marginalizing out the allele frequency trajectory from the HMM yields the probability of the sample local tree as a function of s . To obtain the likelihood function of s , we perform importance sampling over all sample trees , reweighting their coalescent probabilities and summing them up . This approach allows us to use trees sampled exclusively under a prior of selective neutrality ( s = 0 ) to calculate the likelihood function at arbitrary values of s . In other words , this approach allows us to minimize the amount of ARG sampling necessary to estimate the likelihood function , which is notable because ARG sampling is generally the most computationally intensive step of our method . Finally , we can analyze the results to test for selection or estimate the selection coefficient ( Fig 1D ) . Additionally , we show that we can decode the HMMs depicted in Fig 1C and use them to obtain a posterior estimate of the allele frequency trajectory ( Fig 1E ) . A glossary to accompany the following derivations and description of the method is available in S1 Text . First , let us consider how the distribution of the local tree T at a site under selection depends on the frequency trajectory of an allele at that site . We assume that the tree is labeled , i . e . we know which branches subtend each allele . We also assume the tree to be compatible with the infinite sites assumption , i . e . that there is at most one mutation event that has occurred at the focal site , and thus the site is bi-allelic . We model the likelihood of the tree using a structured coalescent; moving backwards in time from the time of sampling until the time of the mutation , lineages can only coalesce with other lineages that subtend the same allele , and the coalescence rate within the derived and ancestral classes depends on both the derived allele frequency X ( t ) and the effective population size N ( t ) , both indexed by the time t ≥ 0 in coalescent units before the present day . Proceeding back in time , lineages coalesce freely after the time of mutation , and the coalescence rate depends only on N ( t ) . In the rest of this section we treat the trajectory X ( t ) as known , but in practice the trajectory is hidden and highly stochastic; in a later section we develop a hidden Markov model to efficiently integrate out X ( t ) . We use a discrete-time model of the coalescent employed also by ARGweaver [37] . That is , we only observe the coalescent process at a set of K discrete timepoints {t1 , … , tK} , and also make the additional assumption that all lineages must coalesce by tK . ( Typically tK is set to ∼100 × Ne , implying coalescence would be extremely unlikely to occur after tK , and hence this assumption is very reasonable ) . Henceforth , using this discretization we also discretize X and N; we assume X ( t ) = Xi for t ∈ ( ti , ti+1] , and N ( t ) = Ni for t ∈ ( ti , ti+1] . We use C to track the number of lineages remaining at these timepoints leading back into the past; as long as we keep track of the number of lineages belonging to each of the allelic classes , by exchangeability of lineages within an allelic class , we can model the likelihood function in the usual way , as independent of the topology given the waiting times . Hence , we define three simultaneous , related processes C = ( Cder , Canc , Cmix ) . The processes Cder and Canc refer to coalescence within the derived and ancestral classes during the time going back from the time of sampling to the time of the mutation . The mixed process Cmix refers to coalescence going backwards from the time of the mutation . We call it the mixed process because it includes un-coalesced lineages from Canc , as well as the lineage ancestral to all derived lineages . Assuming the infinite sites model , Cmix will have one additional lineage relative to Canc at the time of the mutation , and will eventually reach Canc = Cmix once that lineage coalesces with one of the other lineages in the ancestral class . In Fig 2 and Table 1 , we illustrate the lines-of-descent process in the these three classes . We model the probability of transitioning from Ci → Ci+1 lineages during some time interval [ti , ti+1] using a simple variation of Tavare’s formula for the exact distribution of the number of lines of descent remaining after t generations [40] . We use Tavare’s formula in order to model the coalescent at discrete timepoints , allowing multiple coalescences at each epoch . We write the probability of C given the trajectory X ( note this is distinct from the full likelihood of s ) as P ( C ∣ X , N ) = ∏ i = 0 K - 1 P ( C i + 1 ∣ C i , X i , N i ) ( 1 ) More precisely , in terms of the derived , ancestral , and mixed processes , P ( C ∣ X , N ) = ∏ i = 0 i * - 1 P ( C i + 1 der ∣ C i der , X i , N i ) P ( C i + 1 anc ∣ C i anc , X i , N i ) × ∏ i = i * K - 1 P ( C i + 1 mix ∣ C i mix , N i ) ( 2 ) where i* ≔ max{i: Xi > 0} denotes the index of the epoch during which the allele arose via mutation . Naturally , the mixed process—which we only keep track of while the derived allele is nonexistent—does not depend on X . We can write the transition probabilities using a variation of Tavare’s formula for the transition probabilities of the number of lines of descent [40]; in place of the effective population size , we substitute the size of a allelic class zclass in order to reflect the coalescence rate within an allelic class: P ( C i + 1 class = b ∣ C i class = a , Z i = z i class ) = ∑ k = b a { exp ( - ( k2 ) 2 z i ( t i + 1 - t i ) ) ( 2 k - 1 ) ( - 1 ) k - b b ! ( k - b ) ! ( k + b - 1 ) ∏ l = 0 k - 1 ( b + l ) ( a - l ) ( a + l ) } ( 3 ) where z i class = { N i X i : class = der N i ( 1 - X i ) : class = anc N i : class = mix ( 4 ) We note that this formula is known to be computationally unstable for large values of C , large values of N , and/or small values of Δti = ti+1 − ti; under such conditions , the asymptotic distribution of C i + 1 class ∣ C i class = a ( where a is , e . g . , the number of derived lines of descent present at ti ) takes on a normal distribution [41]: C i + 1 class ∣ C i class = a ∼ N ( μ ( Δ t ) , σ 2 ( Δ t ) ) ( 5 ) where μ ( Δ t ) = 2 η Δ τ ( 6 ) and σ 2 ( Δ t ) = 2 η / Δ τ ( η + β ) 2 ( 1 + η / ( η + β ) - η / α - η / ( α + β ) - 2 η ) β - 2 ( 7 ) and Δ τ = Δ t / ( 2 z i class ) ( 8 ) where α = aΔτ/2 , β = −Δτ/2 , and η = αβ/[α ( eβ − 1 ) + βeb] [41] . In practice , for samples of n = 50 haplotypes under constant Ne = 104 , we find this approximation is unnecessary; however , for the same sample size under a European demographic model , which exhibits very large recent Ne , we find it necessary to use this approximation during the roughly 103 generations preceding the present day , prior to which Δt is sufficiently large that we change over to Tavare’s exact formula [42] . In the previous sections we showed how we obtain P ( C ∣ X ) and P ( X ∣ s ) . Here we illustrate how to model coalescence in the two allelic classes when the trajectory X is as a random latent variable . The probability of C given s is thus P ( C ∣ s ) = ∑ x ∈ X P ( C ∣ X = x ) P ( X = x ∣ s ) . ( 9 ) Naively , this involves a prohibitively large sum over dK−1 terms in X , the space of possible trajectories , where d represents the number of possible values that the allele frequency can take . But due to the conditional independence of the likelihood , we can calculate the likelihood much faster using a recursion similar to the forward and backward algorithms commonly deployed on HMMs; we show the derivations of these recursions in S1 Text . In essence , this algorithm works by iteratively marginalizing out the allele frequency in each epoch; the backward recursion marginalizes frequency during earlier epochs first and later epochs last , and the forward recursions marginalizes in reverse order . At each epoch i the recursions yield bi ( xi ) and fi ( xi ) ; these correspond to P ( C 1 : i ∣ X i , N i - 1 ) and P ( C i + 1 : K - 1 , X i ∣ X i , N i ) , where Ca:b = Ca , Ca+1 , … , Cb . Consequently , the two recursions can be used together to obtain the the posterior probability of the allele frequency during the ith epoch Xi , P ( X i = x i ∣ C , s ) = b i ( x i ) f i ( x i ) ∑ x i ′ b i ( x i ′ ) f i ( x i ′ ) ( 10 ) which gives the posterior marginal of Xi using the familiar forward-backward algorithm . The above formulas pertain immediately only to the case in which the local tree is observed directly and without noise . In practical settings , the local tree is hidden to us and we must integrate over the space of possible local trees using sampling methods . Here we describe a novel importance sampling method to reweight posterior samples of the ARG to approximate the likelihood function of selection . Although we use s to express the argument of the likelihood function , we use this as shorthand for estimating the likelihood function of arbitrarily complex parameters; for example , one could estimate the selection coefficient s , as well as the time of selection’s onset , ts , before which the allele behaved neutrally . We are given haplotype data D representing n haplotypes with l segregating sites . We wish to use D to infer the maximum-likelihood value of s for some site k ∈ {1 , 2 , … , l} , where l is the number of sites in the region , assuming that all other sites are selectively neutral ( i . e . sj = 0 ∀j ∈ {1 , 2 , … , k − 1 , k + 1 , … , l} ) . In other words , we restrict ourselves to testing simple hypotheses of the form “site k has selection coefficient sk and all of its flanking sites are selectively neutral . ” The likelihood of s under the data can be expressed as the expected value of the likelihood of the ARG G given the data D , with respect to the distribution of G given s: L ( s ) = E G ∣ s [ P ( D ∣ G , s ) ] ( 11 ) At this stage , we introduce G , the discrete-time approximation of G ( discussed in more detail by [37] ) , and we assume L ( s ) = E G ∣ s [ P ( D ∣ G , s ) ] ( 12 ) By importance sampling , we are able to express the expectation over an alternative distribution q ( G ) , as long as P ( G , D ∣ s ) > 0 ⇒ q ( G ) > 0 . Notice that this implies we can conduct sampling under q ( G ) once , and reweight these samples for arbitrary values of s without having to conduct additional sampling . In other words , approximating L ( s ) using importance sampling does not require sampling under each value of s at which you want to approximate L ( s ) . In this paper we specifically consider the importance sampling proposal distribution q ( G ) = P ( G ∣ D , s = 0 ) , which corresponds to the posterior ARG assuming a neutral model . Later , we evaluate the performance of the estimator using the Markov chain Monte Carlo method ARGweaver , which samples from the posterior [37] . One can obtain the importance sampling estimate of the full likelihood L ( s ) by expressing Eq 12 as an expectation over a different distribution , i . e . the posterior distribution of the ARG ( assuming selective neutrality ) : L ( s ) = E G ∣ s [ P ( D ∣ G , s ) ] = E G ∣ D , s = 0 [ P ( D ∣ G , s ) P ( G ∣ s ) P ( G ∣ D , s = 0 ) ] ( 13 ) We can express Eq 13 using the Monte Carlo approximation L ^ ( s ) = 1 M ∑ m = 1 M P ( D ∣ G ( m ) , s ) P ( G ( m ) ∣ s ) P ( G ( m ) ∣ D , s = 0 ) → L ( s ) ( 14 ) where G ( m ) ∼ P ( G ∣ D , s = 0 ) , m = 1 , 2 , … , M , and “→” , here and in the following , means that the left-hand side converges almost surely to the right-hand side as M goes to infinity , assuming that a Law of Large Numbers for ergodic processes holds ( the Birkhoff–Khinchin theorem ) . Hence , if we sample genealogies from the posterior under selective neutrality , that is , G ( m ) ∼ P ( G ∣ D , s = 0 ) , m = 1 , 2 , … , M ( where M is the number of ARGs sampled ) , then the right-hand side of Eq 14 can be used as a Monte Carlo estimator of the likelihood function . However , in practice this estimator is highly unstable . However , a more stable estimator of the likelihood ratio L ( s ) L ( s = 0 ) can be derived . We can divide through Eq 13 by L ( s = 0 ) = P ( D ∣ s = 0 ) to get L ( s ) L ( s = 0 ) = E G ∣ D , s = 0 [ P ( D , G ∣ s ) P ( D , G ∣ s = 0 ) ] ( 15 ) Because we assume the data are conditionally independent of selection given the full ARG , we can simplify this as L ( s ) L ( s = 0 ) = E G ∣ D , s = 0 [ P ( D ∣ G ) P ( G ∣ s ) P ( D ∣ G ) P ( G ∣ s = 0 ) ] = E G ∣ D , s = 0 [ P ( G ∣ s ) P ( G ∣ s = 0 ) ] ( 16 ) A key development in our method is that although we sample the ARG of the entire sequence , we only calculate likelihoods using the marginal tree at the selected site , which we will call Gk . First , let us define G\k as the rest of the ARG omitting the local tree at site k , Gk . Consequently , G is equivalent to ( Gk , G\k ) . We make a key assumption that , for differing sweep parameters s and s′ P ( G \ k ∣ G k , s ) ≈ P ( G \ k ∣ G k , s ′ ) ( 17 ) That is , we assume that G\k is approximately conditionally independent of s given the marginal tree at the selected site , Gk . Thus , we can reduce Eq 16 to L ( s ) L ( s = 0 ) = E G ∣ D , s = 0 [ P ( G ∣ s ) P ( G ∣ s = 0 ) ] = E G ∣ D , s = 0 [ P ( G \ k ∣ G k , s ) P ( G \ k ∣ G k , s = 0 ) P ( G k ∣ s ) P ( G k ∣ s = 0 ) ] ≈ E G ∣ D , s = 0 [ P ( G k ∣ s ) P ( G k ∣ s = 0 ) ] which suggests the following importance sampling estimator using genealogies sampled from ARGweaver will converge almost surely to a close approximation to the likelihood ratio: LR ^ ( s ) = 1 M ∑ m = 1 M P ( G k ( m ) ∣ s ) P ( G k ( m ) ∣ s = 0 ) → E G ∣ D , s = 0 [ P ( G k ∣ s ) P ( G k ∣ s = 0 ) ] ≈ L ( s ) L ( s = 0 ) ( 18 ) where G ( m ) ∼ P ( G | D , s = 0 ) for m = 1 , 2 , … , M . Finally , due to exchangeability of lineages within the derived and ancestral allelic classes , we can assume P ( G k ∣ s ) ∝ P ( C k ∣ s ) ⇒ LR ^ ( s ) = 1 M ∑ m = 1 M Ω ( m ) ( s ) ( 19 ) where Ω ( m ) ( s ) ≔ P ( C k ( m ) ∣ s ) P ( C k ( m ) ∣ s = 0 ) ( 20 ) denotes the summand of the importance sampling estimator . We can maximize the likelihood ratio over different values of s to obtain the maximum-likelihood estimate of s s ^ = argmax s LR ^ ( s ) ( 21 ) Finally , we show in S1 Text that an importance sampling estimate of π ( xi∣D , s ) , the posterior marginal of the allele frequency at timepoint i , Xi , is given by π ^ ( x i ∣ D , s ) ≔ ∑ m = 1 M P ( X i ∣ C k ( m ) , s ) Ω ( m ) ( s ) ∑ m = 1 M Ω ( m ) ( s ) ( 22 ) where in the summand we use the posterior marginal established in Eq 10 . In practice , we fix s = s ^ . A concern is , therefore , that this estimator does not take uncertainty in the estimate of s into account . This problem can be addressed by using a Bayesian approach , which we demonstrate briefly in S1 Text . The method is implemented in package named CLUES , available for download at https://github . com/35ajstern/clues , with accompanying documentation currently provided in S1 Text . In this paper and the currect software release we assume positive directional selection with an additive effect on fitness , our method can be easily extended to general dominance relationships as well as negative selection . To evaluate the power of CLUES to determine whether a site has been subject to selection , we simulated a dataset of n = 25 diploid individuals under two different demographic models; ( 1 ) a model of constant effective population size ( N = 104 ) , and ( 2 ) a model of European ( CEU ) demography [43] . We performed both sets of simulations using the program discoal [44] . We set μ = 2r = 2 . 5 × 10−8 mut/bp/gen , L = 1 × 105 bp or 2 × 105 bp for the constant-size and CEU models , respectively , and simulated conditional on a variety of present-day frequencies and selection coefficients , the latter of which we ranged from weak to strong values . Under each condition , we simulated 100 independent iterations . We also sampled 1 ancient haplotype; because ARGweaver , which we used subsequently to sample the posterior ARG , does not incorporate any information about ancestral/derived states , it is best practice to add an ancient individual or outgroup to help polarize the the alleles . In practical settings where the ancestral state is unknown , ARGWeaver accomodates specification of missing data on the ancient haplotype . For the constant-size and CEU models , we used ancient sampling dates of 2 × 104 and 1 . 6 × 104 generations before present , respectively . Because discoal can only simulate piecewise-constant population sizes , we specified population sizes to take on the value of their harmonic mean over the epoch , calculated from the original CEU model . Commands to run simulations of trajectories , local trees , and haplotypes are described in S1 Text . Importantly , we conditioned simulations on the site of interest segregating at a particular frequency in the present day . Hence , when we considered the power to discriminate between neutral and selected alleles , we controlled the present-day frequency to be equal in both of these cases . Avoiding this step would otherwise upwardly bias estimates of the statistical power , due simply to the tendency for selected alleles to segregate at higher frequencies than neutral alleles [45] . ( If the allele frequency in itself is also of interest , this part of the likelihood could trivially be added at a later stage , by simply using the stationary distribution of the allele frequency; see “Allele frequency transition probabilities” in S1 Text ) . We then simulate the allele frequency backwards in time , from the present-day frequency , until the allele reaches a frequency of 0 . Simulators such as discoal achieve this by using the conditional Wright-Fisher diffusion ( see e . g . [46] ) . In the case where effective population size changes over time , running conditional simulations requires additional considerations because the probability of a mutation entering the population scales approximately linearly with population size . Naively sampling the trajectory backwards in time will therefore produce a bias , unless trajectories where the mutation occurs while Ne is low are somehow penalized . Thus , approaches such as reweighting sample trajectories using importance sampling have been used to correct this bias [47] . The program discoal implements a similar bias-correcting scheme using rejection sampling that rejects trajectories where the mutation occurs while Ne is low with higher probability than trajectories where the mutation occurs while Ne is high . Next , we inferred the posterior ARG given the sequence data we simulated using ARGweaver [37] . This method works by proposing adjustments to an initial ARG , and randomly accepting or rejecting these proposals based on calculations of the prior probability of the proposed ARG , as well as its likelihood given the sequence data . Because the prior probability is based on the effective population size , we specified the same effective population size in the prior as we used to generate the sequence data . We found it important to adjust the proposal mechanism of ARGweaver; specifically , we adjusted resample window size and the number of resamples per window to achieve an acceptance rate of about 30-70% . In total , we sampled 3 × 103 ARGs for each simulation , discarding the first 1 × 103 as a burn-in period , and subsequently thinning the remaining samples to reduce the computational burden of downstream analyses; we used a thinning rate of 100 samples , resulting in M = 20 approximately independent samples . Reducing the thinning rate would increase accuracy and convergence of the inference at the cost of additional computation to calculate the likelihood of each additional sample tree . Commands to conduct ARG-sampling and local tree extraction are described in S1 Text . Using utilities in the ARGweaver package , we extracted local trees at the selected site ( at the center of the locus ) from these sample ARGs . We then analyze this final set of trees using CLUES . We also analyzed the same sequence data using nSL , H12 , and Tajima’s D [14 , 19 , 48] . The nSL method is essentially equivalent to iHS [11] , except nSL does not require specifying a genetic map; despite this , these methods have been shown to have very similar statistical power with a slight advantage of nSL under some conditions . H12 is a method to calculate haplotype homozygosity merging the two most common haplogroups; thus , it is a test for selection that is robust to the origin of a sweep , i . e . whether it is hard or soft . Tajima’s D is a site frequency spectrum-based statistic which is sensitive to skews in the frequency distribution of linked alleles caused by hitchhiking on the partially swept selected allele . We used scripts provided by [22] to calculate D and H12 , using a window size of 100kb centered on the selected site . We compare testing for selection under these methods by comparing their power curves under both the constant Ne and CEU demography models ( Figs 3 and 4 ) . We also conducted a similar simulation study for detecting recent selection starting 100 generations ago . We simulated under the same CEU demographic model as previously described , but instead sample n = 50 diploids . We conducted ARG sampling and thinning as previously described , but in our analysis of the sample trees using CLUES , we calculated the likelihood for models of selection where s = 0 up until 100 generations ago , and s ≥ 0 from that point until the present day . This sweep from standing variation ( SSV ) model differs from the hard sweep model we used previously , which assumes s is constant throughout history . Instead of optimizing the likelihood function only with respect to s , we optimized with respect to two parameters , s and ts , jointly; here ts represents the time of the onset of selection .
We found that across all scenarios , CLUES matches or exceeds the statistical power of the other methods evaluated ( Figs 3 and 4 ) . As expected , all methods had highest power under large values of both the selection coefficient and the derived allele frequency ( Fig 3I ) . Under these conditions , CLUES had 100% power at the 1% significance threshhold; the next most powerful method , nSL , had 68% power at the same significance level . CLUES also demonstrated improvement in power under weak selection; as the selection coefficient was decreased , nSL retained about 20% power when s = 0 . 003 and <5% power when s = 0 . 001 , and Tajima’s D and H12 retained <5% power under both s = 0 . 001 , 0 . 003 ( Fig 3G and 3H ) . By contrast , CLUES retained approximately 45% and 90% power under s = 0 . 001 , 0 . 003 , respectively . We conclude that CLUES has high power across a wide regime of selection strengths , and has notably improved power over standard methods under weaker values of s . We also considered the effect of present-day allele frequency on statistical power . Previous studies have shown a strong dependence of power on current allele frequency , with methods such as nSl and iHs having highest power at allele frequencies in the 70-90% range ( see e . g . [11] ) . We tested for selection at alleles ranging in present day frequency from 25% to 75% , and while CLUES showed the expected pattern of increasing power with frequency , it also improved on the performance of other methods at lower frequencies . For example , under strong selection ( s = 0 . 01 ) , the power of CLUES changed from 100% to 90% to 85% as the frequency is decreased from 75% to 50% to 25% ( Fig 3C , 3F and 3I ) . By contrast , the power of the next most powerful method , H12 , dropped from approximately 65% to 45% to 15% ( Fig 3C , 3F and 3I ) . Under moderate selection ( s = 0 . 003 ) , these effects were even more drastic , with the power of CLUES and nSL ( the next most powerful method in this regime ) changing from 90% to 60% to 50% and 20% to 5% to <5% , respectively . We conclude that CLUES has high power compared to standard methods across a wide range of allele frequencies , with the most major improvements in performance occurring when the derived allele is at lower frequencies ( <50% ) . We found that using the approximation due to Griffiths ( Eq 5 , [41] ) decreased power of CLUES by increasing variability of the null distribution of the likelihood ratios . Hence , for testing under nonequilibrium demography we used the exact lines-of-descent probabilities ( Eq 3 ) . By contrast , as we will later show , we found the approximation given by Eq 5 for t ∈ [0 , 1000] to improve estimation of allele frequency trajectories under this demographic model . We also considered the same testing procedure under non-equilibrium demography , simulating under the previously described model of CEU demography ( Figs 3 and 4 ) . We found in general reduced power to detect selection under this regime relative to the constant population size regime ( Fig 4I , cf . Fig 3I ) , consistent with the well-known confounding of expanding population size with selection [10] . Nonetheless , CLUES demonstrated improved power relative to the competing methods across a wide range of selection coefficients ( Fig 4C , 4F and 4I ) , as well as across a wide range of derived allele frequencies ( Fig 4G , 4H and 4I ) . Using the simulations from the previous section to study statistical power in testing for selection , we used our estimate of the likelihood surface for s to estimate the value of the selection coefficient via maximum likelihood ( see Eq 21 ) , restricted to 0 ≤ s ≤ 0 . 5 , as we only calculate transition probabilities for this range of s . We obtained selection coefficient estimates under importance sampling using ARGweaver ( Fig 5 ) , as well as selection coefficient estimates based on the true local tree observed directly ( S1 Fig ) . Generally , the estimates are approximately unbiased . For example , the mean estimates of s = 0 , 1 × 10−3 , 3 × 10−3 , 1 × 10−2 were approximately s ^ ¯ = 1 . 9 × 10 - 4 , 9 . 6 × 10 - 4 , 3 . 2 × 10 - 3 , 1 . 3 × 10 - 2 when the present day frequency was fixed to 75% ( Fig 5A ) . Relative to inference when the true tree is observed , we found that the importance sampling estimates had increased variance , reflecting uncertainty in the tree . For example , we saw increased variability in the importance sampling vs . true tree estimates under constant population size ( Fig 5A vs . S1A Fig ) , as well as under CEU demography ( Fig 5B vs . S1B Fig ) . This pattern is consistent with the additional uncertainty in s when the local tree is not observed directly . Notably , we found that importance sampling under a model of CEU demography yields estimates with a slight bias towards lower values of s , especially under strong selection ( e . g . s = 0 . 01 ) . Using the same simulations and importance sampling estimates we obtained in the previous sections , we decoded the hidden Markov model ( HMM ) described in the section Materials & Methods . Specifically , we take s ^ , the maximum likelihood estimate of s , and plug it into the posterior marginal ( Eq 10 ) to obtain a probabilistic estimate of the allele frequency during a particular epoch; we do this independently for each epoch in our discrete-time model . To get a point estimate , we choose to use the posterior marginal mean; i . e . , for each epoch , we choose the mean of the posterior marginal distribution . We illustrate the accuracy of these allele frequency trajectory estimates assuming the true local tree is observed and under importance sampling when the true tree is unknown in Fig 6 . We find that estimates of the allele frequency trajectory are generally unbiased for both true trees ( Fig 6A and 6B ) and importance sampling ( Fig 6C and 6D ) , with increased variance in the trajectory estimates in the importance sampling setting . We also illustrated variability in true vs . inferred trajectories controlling for s ( S6 Fig , here setting s = 0 ) . Whereas inference tended to be relatively accurate for high-frequency alleles ( Fig 6B and 6D ) , when the derived allele was simulated conditioned on lower frequencies ( e . g . 25% , Fig 6C ) , estimates tend to be downwardly biased . We tracked this bias to a lack of convergence in ARGweaver; specifically , we found that across different demographic scenarios and selection coefficients , ARGweaver can drastically overestimate the occurrence of very recent coalescences ( in our case , in the last 100 generations; see S5 Fig ) . Under constant population size , we see a nearly 7-fold excess in the number of recent coalescences inferred by ARGweaver . Naturally , this bias will affect estimates for low-frequency alleles more strongly , as fewer lineages subtend the derived allele , and thus a larger proportion of them are susceptible to this bias . Because recombination rates vary substantially throughout the genomes of humans and other organisms , we also evaluated the accuracy of the estimates assuming μ = r , larger than the μ = 2ρ setting we used in the other simulations , and estimation accuracy to be robust to this increase in recombination rate ( S2 Fig ) . We also examined trajectory inference under non-equilibrium demography; i . e . , the aforementioned model of CEU demography ( S3 Fig ) . Under the CEU model , we found trajectory estimates to have increased variance under importance sampling vs . true trees , but also a slight downward bias in estimating the selection coefficient under strong selection ( i . e . s = 0 . 01; see Fig 5B , S3D Fig ) . As this bias does not occur under the true trees ( S1B and S3B Figs ) , we inspected the posterior trees sampled by ARGweaver for patterns consistent with this bias . We found that under this demographic model in particular , ARGweaver tends to under-sample trees with short times to most recent common ancestor ( TMRCAs; see S4 Fig ) . For reference , nearly 60% of runs under constant Ne contained even a single sample tree that had a TMRCA less than or equal to that of the true TMRCA ( S4A Fig ) . By comparison , under s = 0 . 01 and CEU demography , only 11% of ARGweaver runs met this criterion ( S4B Fig ) . Some bias is to be expected , as trees were sampled under a posterior distribution that assumes selective neutrality; however , these results suggest that , if ARGweaver is sampling from the true posterior assuming selective neutrality , then importance sampling estimates ( of the selection coefficient , for example ) will at least have much higher variance under the CEU model than under constant population size . We further investigated whether uncertainty in s due to importance sampling variance drove the downward bias when estimating strong selection ( Fig 5B and S3D Fig ) . First , we obtained importance sampling estimates of the trajectory fixing s to its true value ( S7A Fig ) . If uncertainty in s were the cause of the bias , then fixing the true value of s ought to correct for bias due to uncertainty . While we observe less bias in the estimates when fixing the true value of s , the bias is not totally eliminated . We observe a similar reduction in the bias of estimates under neutrality when we fix s = 0 ( see S6B , S6E and S6H Fig vs . S6C , S6F and S6I Fig ) . Thus , we conclude the bias is due to a lack of convergence in ARGweaver , which appears to be exacerbated in settings where strong selection is combined with non-equilibrium demography . We also investigated whether incorporating uncertainty in the estimate of s , rather than fixing s = s ^ , would improve the accuracy of trajectory inference . One strategy for modeling uncertainty in s is to apply a prior distribution to s . We found that marginalizing out s with respect to its posterior distribution ( assuming a uniform prior on s ) did not have a noticeable effect on inference for large values of s ( S7B Fig ) . This result is concordant with our observation that for large values of s , the likelihood surface peaks so strongly that the posterior remains tightly concentrated around the MLE s ^ . Hence , applying a prior distribution to s does not appear to be an adequate strategy to model uncertainty in s . We applied our likelihood model of a sweep from a standing variant ( SSV ) to two types of datasets: selection from a standing variant starting 100 generations ago and selection with constant s ( including s = 0 ) , both described in ‘Simulations’ under Materials and Methods . We inferred trajectories under the best case scenario where the true trees are observed ( Fig 7A and 7B ) . We found that overall the method inferred the trajectory , as well as the strength and timing of selection , with highest accuracy when selection is strong ( e . g . s = 0 . 03 in Fig 7A and 7B ) . However , we found that as s took on smaller values ( s = 0 . 01 ) , many combinations of s and ts had very similar likelihood ( Fig 7B ) , and thus estimates of s , ts , and the allele frequency trajectory tended to be noisier than under very strong selection ( Fig 7A and 7B ) . Adding the extra parameter ts did not cause overfitting when inferring the trajectories of hard sweeps ( Fig 7A ) . We also found good power to distinguish between hard vs . soft sweeps ( i . e . sweeps from a standing variant ) , as apparent in the trajectories inferred in Fig 7A . We calculated statistical power to test for a hard sweep using the statistic max s , t s { L ( s , t s ) } / max s { L ( s , t s = ∞ ) }; intuitively , this statistic is the ratio of the highest likelihood under any model with a SSV ( ts ≠ ∞ ) to the highest likelihood of any hard sweep ( ts = ∞ ) . At the 1% significance level we found 60% and 100% power to distinguish soft vs . hard sweeps with s = 0 . 01 , 0 . 03 , respectively . We also performed importance sampling using ARGweaver and evaluated the power of the importance sampling estimates to detect recent selection vs . neutrality ( Fig 7C ) . Instead of comparing our method to nSL , which is not designed to detect signals of extremely recent selection , we compared to Singleton Density Score ( SDS; [21] ) , as well as H12 and Tajima’s D . We found that for lower values of s , all methods had generally low power . Although CLUES exhibited fairly high power ( 44% ) to detect very strong recent selection ( s = 0 . 03 ) —even outperforming SDS—we found that H12 has about the same power ( 45% ) in this particular case . The lower power ( <5% ) of SDS is consistent with the fact that the method was explicitly designed to have high power for large datasets ( n > 1000 for selection coefficients of this magnitude ) . Although we demonstrate that CLUES has substantial power to detect extremely recent selection , we found that importance sampling point estimates of s , ts , and the trajectory were highly vulnerable to biases in the distribution sampled by ARGweaver ( S5 Fig ) . Specifically , we found that across various demographic and selection conditions , ARGweaver samples trees with substantially more recent coalescent events than in the true trees . Specifically , under the European demographic model with the settings used here to study recent selection , we find ARGweaver samples about a 4-fold excess of recent coalescent events ( S5B Fig ) . Clearly , this bias would produce a false signature of recent selection under neutral conditions . Thus , we did not further explore importance sampling estimates of s and the trajectory under the recent selection model . We conclude that potential ARG-sampling methods that avoid this bias will improve upon power to detect recent selection , as well as point estimates of the strength , timing of selection , and the allele frequency trajectory . Strongly selected alleles tend to have high levels of LD to linked neutral sites , and thus many methods to detect selection are limited in their ability to determine the exact site under selection . To assess whether the likelihood ratio statistic produced by CLUES can be used to fine-map the selected site , we ran CLUES at linked neutral sites in a locus centered on a site under positive selection . Simulations were identical to those used in the simulation study under constant Ne = 104 , with a present-day selected allele frequency of 75% , s = 0 . 01 , and 100 independent simulations . We chose sites with the maximal squared correlation coefficient r2 to the selected allele , such that r2 did not exceed a threshold value , and vary that threshold from 0 . 50 to 0 . 99 ( Fig 8 ) . We found that when the true tree is observed ( or sampled with high accuracy ) , the likelihood ratio statistic identifies the selected site correctly in a head-to-head test with the neutral linked site with 85% accuracy even when r2 ≤ 0 . 99; this quantity reaches 100% for r2 ≤ 0 . 50 ( Fig 8A ) . When the likelihood ratio is estimated using importance sampling via ARGweaver , the accuracy declines to about 50% and 85% , respectively ( Fig 8A ) . Because the exact causal site may not be known in many studies , we also investigated how the estimate of the selection coefficient , s , depends on r2 between the site analyzed and the site under selection . We estimate s given the true tree , and find that , on average , estimates of s decline with r2 , such that for r2 ≤ 0 . 50 , the mean estimate of s at these neutral sites is less than 20% of the true value of s at the causal site ( Fig 8B ) . We also assessed the effects of linked selection on inference of selection at a particular site . In particular , we consider the effects of background selection ( BGS ) on inference of positive selection on a linked site ( S8 Fig ) . We performed forward simulations in SLiM 3 [49] , assuming a model of constant Ne = 103 , a locus of length 1Mb , n = 25 diploid individuals and μ = 2 . 5 × 10−8 mut/bp/gen , r = 1 . 25 × 10−8 recombinations/bp/gen . We let mutations be neutral with probability 90% and ( negatively ) selected with probability 10% . We simulated under deleterious effects of s = 0 , −0 . 001 , −0 . 003 and -0 . 01 , performing 100 independent replicates under each case . To simulate selection at a focal allele , 100 generations prior to the present day , we choose a random neutral allele conditional on its frequency falling in the interval [0 . 005 , 0 . 015] and its position falling in the interval [4 × 105 , 6 × 105]bp ( to ensure it is somewhat centered in the 1Mb region ) , and then endow it with a selection coefficient of s = 0 . 15 . Prior to this timepoint , we perform a burn-in phase of 19900 generations with only neutral and deleterious mutation . From the sampled haplotypes , we perform importance sampling using ARGweaver to estimate the likelihood of the selection coefficient . We find that our method is quite robust to BGS; e . g . , the median estimate of the selection coefficient is approximately unbiased ( mean s ^ = 0 . 13 ) as the strength of BGS is increased ( S8 Fig ) . Also , regardless of BGS strength there is nearly perfect power to detect positive selection when comparing to neutral simulations ( S8 Fig ) . We note that the strength of selection on the beneficial allele investigated here is somewhat strong; for weaker selection on a beneficial allele , inference with sites under BGS may be a more significant determinant of the beneficial allele’s trajectory ( see e . g . [50] ) . We also note that our simulations assume a model of equilibrium demography , but under non-equilbrium conditions ( e . g . , rapid population size expansion ) BGS has a magnified effect on neutral diversity , which may further bias estimates of selection [51] . To explore the effects of demographic model misspecification on inference of selection , we ran CLUES on datasets simulated under a model of European demography ( described earlier in Methods ) , using a mismatched model of constant Ne = 104 to calculate the likelihood , and compare them to calculations under the true demographic model ( S9 Fig ) . We report likelihood ratios ( S9A and S9C Fig ) as well as estimates of the selection coefficient s ( S9B and S9D Fig ) both given the true tree ( S9A and S9B Fig ) and approximated via importance sampling using ARGweaver ( S9C and S9D Fig ) . We find that statistical power to detect selection , especially strong selection , is not substantially impeded by model misspecification ( S9A and S9C Fig ) . We do , however , find upward bias in estimates of the selection coefficient when s is 0 or close to 0 ( S9B and S9D Fig ) . To assess performance of CLUES on empirical data , we applied our method to study selection acting on the SNP rs4988235 in the MCM6 gene , known to regulate the neighboring LCT gene and affect the lactase persistence trait . The derived allele ( A ) current segregates at approximately 72% in the 1000 Genomes Phase 3 reference panel ( British in England and Scotland , henceforth GBR; see S10A Fig ) . We conducted sampling in ARGweaver assuming a model of European demography [43] , using a 300kbp region centered around the focal SNP and polarizing alleles using the genomes of three ancient individuals ( Altai Neandertal , Denisova , and Vindija Neandertal [52–54] ) . We sampled M = 200 ARGs , extracted local trees using tools in the ARGweaver package , and conducted importance sampling to estimate likelihood surfaces and trajectories using CLUES . We found very strong evidence for selection on rs4988235 ( s = 0 . 0161 , logLR = 131 . 82 ) . The trajectory as well as the value of the selection coefficient inferred by CLUES are consistent with previous estimates of the trajectory and s = 0 . 018 due to Mathieson and Mathieson ( 2018 ) , illustrated in Fig 9 [55] . Their method incorporates genomic times series spanning thousands of generations using an HMM-based approach , where hidden states are population-wide allele frequencies , observed states are genotypes of sampled ancient individuals , and transition probabilities are governed by the selection coefficient . Our approach , by contrast , does not utilize any ancient/timecourse data except for the 3 aforementioned ancient individuals , which we use to simply polarize the derived and ancestral states of each allele . Another canonical example of a common SNP under selection in humans in rs3827760 ( see e . g . [56] ) . This SNP is a non-synonymous mutation that is present at high frequency in East Asian populations ( e . g . 94% in Han Chinese [CHB] in the 1000 Genomes database ) , intermediate-high frequency in Central and South America , and low frequency in other geographical regions S10 Fig . This variant is associated with a number of traits , including tooth shape and hair straightness [57 , 58] . To estimate selection on this SNP , we conducted sampling in ARGweaver assuming a model of East Asian demography [59] , using a 300kbp region centered around the focal SNP and polarizing alleles using the genomes of three ancient individuals ( Altai Neandertal , Denisova , and Vindija Neandertal [52–54] ) . We sampled M = 200 ARGs , extracted local trees using tools in the ARGweaver package , and conducted importance sampling to estimate likelihood surfaces and trajectories using CLUES . We estimate that rs3827760 has undergone selection with s = 0 . 0047 , corresponding to an allele age of roughly 45kya ( S11 Fig ) . Our estimates are in stark contrast to some previous estimates obtained using ABC methods , which estimate > 30-fold stronger selection on this SNP , and an allele age of 1 . 4-6 . 9kya [32] . Our results are consistent with ancient DNA evidence , which suggest the derived allele to have originated prior to 7 . 5kya [8] . Using the same GBR panel from 1000 Genomes Phase 3 , we analyzed a set of SNPs associated with pigmentation-related traits , some of which were previously identified as likely targets of recent selection [21] . We conducted sampling in ARGweaver assuming a model of European demography , using a 300kbp region centered around the focal SNP and sampling M = 200 approximately iid ARGs . We ran CLUES and estimated likelihood surfaces and allele frequency trajectories for these SNPs ( Fig 10 ) . We found significant concordance between the SDS values and our likelihood ratio statistics paired for each SNP ( p = 1 . 7 × 10−3 , Spearman one-sided ) [21] . We also illustrated the geographical distribution of these SNPs among diverse populations ( S12 Fig ) using GGV [60] . We found several signals of very strong selection acting on rs619865 ( ASIP , s ≈ 0 . 10 , Fig 10I ) , rs12821256 ( KITLG , s ≈ 0 . 016 , Fig 10H ) , and rs1393350 ( TYR , s ≈ 0 . 011 , Fig 10J ) ; these SNPs are significantly associated with freckling , blonde hair color , and freckling and blue/green eye color , respectively [61–63] . Interestingly , these SNPs all demonstrated a signal of selection mostly concentrated in the last ∼5 kya . The geographical distribution of the frequency of these SNPs shows that the derived version of these variants are mostly concentrated in European populations , with minimal sharing with populations located in Africa and Asia ( S12I , S12H and S12J Fig ) . For example , TYR and KITLG segregate at a frequency ∼20% in several European populations and have a frequency close to 0% in African and East Asian populations ( S12J Fig ) . These three SNPs are the only ones in this set of SNPs which have a frequency of nearly 0% across the African populations surveyed , with the exception of OCA2/HERC2 ( S12A , S12H , S12I and S12J Fig ) , consistent with our evidence for recent selection at these loci . The frequencies of these variants in GBR ranges from ∼10-20%; by contrast , the only other variant in this set with comparable frequency in GBR ( 13% ) , rs35264875 ( TPCN2 ) , we find inconclusive evidence of selection ( Fig 10F ) , consistent with its comparably even geographical distribution relative to the aforementioned SNPs at ASIP , KITLG , and TYR ( S12F Fig ) . At rs12896399 ( SLC24A4 , Fig 10B ) , a SNP identified to be significantly associated with hair color [62] , we found strong evidence for moderate selection ( s ≈ 0 . 005 ) . This result is consistent with a previous analysis that suggested positive selection acted on this allele in Out-of-Africa ( OoA ) populations , based on its high allele differentiation relative to a YRI panel , and low haplotype diversity within CEU individuals [63] . Our results , paired with the apparent low levels of differentiation between European and Asian populations relative to differentiation between OoA populations and African populations at this locus ( S12B Fig ) are consistent with our estimate that selection acted on SLC24A4 as early as ∼30 kya , during the OoA bottleneck as inferred by [43 , 59] . Notably , we find moderate evidence for selection on rs12913832 ( OCA2/HERC2 , Fig 10A , S13 Fig ) , a SNP previously shown to be causal for blue-brown eye color [64] and significantly associated with hair color [62] . This gene exhibits abberantly high differentiation across populations [65] , consistent with a model of local adaptation of eye color . Compared to previous estimates based on ancient DNA samples [66] , we estimate substantially weaker selection acting on this gene ( s ≈ 0 . 002 vs . s ≈ 0 . 04 ) , and we find no evidence to support a recent increase in selection acting on this SNP ( i . e . , our method found a hard sweep to have higher likelihood than a SSV ) . Our estimate of moderate selection and lack of a recent change in the selection coefficient imply that selection on OCA2/HERC2 began at least ∼50 kya , roughly the time of the start of the OoA bottleneck estimated by [43 , 59] . Our analysis suggests that selection on OCA2/HERC2 may have begun much earlier than previously suggested [66] . We also note that the aforementioned rs12913832 ( OCA2/HERC2 ) —as well as rs2153271 ( BNC2 ) , a SNP which is significantly associated with freckling ( Fig 10G ) —occur in high-frequency archaic haplotypes [67] . While our method is not explicitly designed to control for population structure between archaic and modern human lineages , we do find moderate evidence for selection on both of these SNPs . One surprising result is that we found no signal of selection acting at rs13289810 ( TYRP1 , s ≈ 0 , Fig 10E ) . In Europeans , TYRP1 is associated with hair and eye pigmentation [68–71] . Some analyses of European populations have indicated evidence for positive selection on TYRP1 [56 , 63 , 69] . Our results temper these claims , and appear consistent with the fairly even geographical distribution of rs13289810 frequency across European , African , and Middle Eastern populations ( S12E Fig ) .
We have developed an approach to use modern population genomic data to approximate the full likelihood of selection acting on a locus . We use this approach to test for and estimate the strength and timing of selection , as well as estimate the full allele frequency trajectory . The method is effective across a span of selection coefficients ( s = 0 − 0 . 01 ) , derived allele frequencies ( f = 25% − 75% ) , and under multiple demographic models . Our method draws on previously published methods to estimate the ancestral recombination graph ( ARG ) . We chose to use ARGweaver because it is the only currently available method for sampling the posterior of the ARG; as shown in our derivation of the importance sampling estimates , we rely on sampling from the posterior in order to make rigorous guarantees regarding convergence and consistency of our estimators . Intuitively , it is important to model the uncertainty in the local tree in order to marginalize out this latent variable . We showed that estimates of the selection coefficient and the trajectory are generally accurate , barring scenarios where importance sampling is inefficient , or ARGweaver produces a bias in the inferred trees . In light of these biases , under certain conditions—primarily when the derived allele is at low frequencies ( ≤25% ) —importance sampling using ARGweaver trees has limited power to detect selection . Another important limitation of ARGweaver is its computational cost; in order to study selection on short timescales , large sample sizes are necessary , often on the order of thousands of individuals [21] . The runtime of ARGweaver grows dramatically with increasing sample size; not only does the cost of the individual sampling steps increase with sample size , but also so does the size of the state space , necessitating more samples be taken in order to achieve convergence to the stationary distribution . However , we see potential to make use of recent advances in inference of local trees in order to further advance approximate full-likelihood methods to infer selection ( see e . g . , [72–75]; it is worth noting that some of these methods , such as [75] , do not infer the ARG in a strict sense , but rather the sequence of local trees along a recombining locus ) . A major benefit of these methods is that they are far more scalable than ARGweaver , and hence offer more potential to study selection on short , punctuated timescales . However , they also possess several limitations: firstly , several of these methods only infer topologies , rather than branch lengths [73 , 74] . While it is possible to infer branch lengths condition on topology estimates , it is unclear how accurate these estimates would be . In contrast , methods that infer branch lengths along with topology entail a slight tradeoff in their scalability [72 , 75] . Another limitation of these methods is that they only yield a point estimate of the local tree , rather than estimating uncertainty in the tree . Nonetheless , it may be feasible to quantify uncertainty in the local tree using a jackknife approach where the local tree is inferred over random subsets of the individuals . It may also be possible to make use of recent advances in inferring pairwise coalescence times ( e . g . , [76] ) to build an approximation to the full likelihood . Recently , Albers & McVean proposed a composite likelihood method to estimate allele age by “sandwiching” the age using identity-by-descent tracts at the site of interest [77] . However , their method does not extend to inferring how the allele frequency changed over time , and does not explicitly model selection . Currently our method assumes correct knowledge of the demographic history . The effects of latent or mis-specified population structure on inference of selection are well known ( e . g . , [78] ) , but in future work one might try to determine the exact effects of mis-specification of effective population size on both inferring the local tree , and inferring selection conditional on the local tree . One approach to dealing with this is to extend the importance sampling approach we use to correct for selection to additionally correct for demography , when ARG sampling is performed under a mis-specified demographic model . Furthermore , many aspects of our model of selection ( e . g . coalescence , allele frequency transitions ) assume a panmictic population . To extend our model to more complex demographic models and/or linked selection ( i . e . , allowing multiple sites to be subject to selection ) would entail drastically increased computational cost ( e . g . , marginalizing allele frequencies corresponding to each population , rather than the allele frequency in a single population ) . Using a deterministic approximation of the allele frequency trajectory would circumvent this issue , but it would also raise new issues , such as how to model allele frequencies when s ≈ 0 . Despite its limitations , the method presented here provides the first close approximation to a full likelihood function for the selection coefficient under simple models . As demonstrated by our simulations , full likelihood methods have the potential to greatly improve power to detect selection and estimate the strength of selection under a variety of conditions . It also provides a rigorous and accurate method for estimating allele frequency trajectories , and is the first to achieve so using modern data . As methods for inferring ARGs improve in the future , so too will the derived methods for detecting and quantifying selection and inferring allele frequency changes .
|
Current methods to study natural selection using modern population genomic data are limited in their power and flexibility . Here , we present a new method to infer natural selection that builds on recent methodological advances in estimating genome-wide genealogies . By using importance sampling we are able to efficiently estimate the likelihood function of the selection coefficient . We show our method improves power to test for selection over competing methods across a diverse range of scenarios , and also accurately infers the selection coefficient . We also demonstrate a novel capability of our model , using it to infer the allele’s frequency over time . We validate these results with a study of a lactase persistence SNP in Europeans , and also study a SNP at EDAR , as well as a set of 11 pigmentation-associated variants .
|
[
"Abstract",
"Introduction",
"Materials",
"and",
"methods",
"Results",
"Discussion"
] |
[
"taxonomy",
"markov",
"models",
"population",
"dynamics",
"geographical",
"locations",
"genetic",
"mapping",
"simulation",
"and",
"modeling",
"phylogenetics",
"data",
"management",
"mathematics",
"phylogenetic",
"analysis",
"molecular",
"genetics",
"population",
"biology",
"research",
"and",
"analysis",
"methods",
"europe",
"computer",
"and",
"information",
"sciences",
"hidden",
"markov",
"models",
"evolutionary",
"systematics",
"molecular",
"biology",
"probability",
"theory",
"people",
"and",
"places",
"haplotypes",
"natural",
"selection",
"heredity",
"genetics",
"biology",
"and",
"life",
"sciences",
"physical",
"sciences",
"evolutionary",
"biology",
"evolutionary",
"processes",
"geographic",
"distribution"
] |
2019
|
An approximate full-likelihood method for inferring selection and allele frequency trajectories from DNA sequence data
|
Descent of testes from a position near the kidneys into the lower abdomen or into the scrotum is an important developmental process that occurs in all placental mammals , with the exception of five afrotherian lineages . Since soft-tissue structures like testes are not preserved in the fossil record and since key parts of the placental mammal phylogeny remain controversial , it has been debated whether testicular descent is the ancestral or derived condition in placental mammals . To resolve this debate , we used genomic data of 71 mammalian species and analyzed the evolution of two key genes ( relaxin/insulin-like family peptide receptor 2 [RXFP2] and insulin-like 3 [INSL3] ) that induce the development of the gubernaculum , the ligament that is crucial for testicular descent . We show that both RXFP2 and INSL3 are lost or nonfunctional exclusively in four afrotherians ( tenrec , cape elephant shrew , cape golden mole , and manatee ) that completely lack testicular descent . The presence of remnants of once functional orthologs of both genes in these afrotherian species shows that these gene losses happened after the split from the placental mammal ancestor . These “molecular vestiges” provide strong evidence that testicular descent is the ancestral condition , irrespective of persisting phylogenetic discrepancies . Furthermore , the absence of shared gene-inactivating mutations and our estimates that the loss of RXFP2 happened at different time points strongly suggest that testicular descent was lost independently in Afrotheria . Our results provide a molecular mechanism that explains the loss of testicular descent in afrotherians and , more generally , highlight how molecular vestiges can provide insights into the evolution of soft-tissue characters .
In placental mammals—the eutherian crown group consisting of the clades Afrotheria , Xenarthra , and Boreoeutheria [1]—optimal testicular function requires a temperature that is lower than the body temperature . To achieve this , the testes are located outside of the abdominal cavity in a scrotum in many species such as primates , most rodents , lagomorphs , most carnivores , and most terrestrial artiodactyls [2 , 3] . Alternatively , testes are located in the lower abdomen in dolphins , true seals , pangolins , and other mammals . In these species , testicular cooling is achieved by vascular countercurrent heat exchanger systems , as observed in dolphin [4]; direct cooling with blood from the hind limbs , as observed in seals [5]; or testicular cooling may not be necessary , as these species have lower body temperatures [2 , 6 , 7] . The position of the testes in the lower abdomen or in the scrotum is the result of a developmental descent process ( S1 Fig ) . During mammalian development , testes initially form at a position near the kidneys in the embryo . Testicular descent into the scrotum occurs in two phases: first from the abdomen to the inguinal canal and second through the inguinal canal into the scrotum [8 , 9] . The first transabdominal phase is governed by the growth and reorganization of the gubernaculum , a ligament that connects the lower pole of the testes and inner ring of the future inguinal canal [8–10] . Migration of the testes is caused by the swelling of the distal gubernaculum , which anchors the testis to the inguinal canal , while the abdominal cavity enlarges . The second inguinoscrotal phase is dependent on androgen signaling and requires the elongation of the gubernaculum , which migrates into the scrotum [8–10] . The involved signaling and mechanics make testicular descent a difficult and complex developmental process . Failure in any of the descent phases results in a pathological condition called cryptorchidism ( absence of testes from the scrotum ) , which is a congenital birth defect observed at an appreciable frequency in human males ( 2%–4% at birth [11] ) and other animals ( up to 10% in male dogs [12] , 2% in male cats [13] , 2%–8% in male horses [14] ) . Almost all placental mammals exhibit either partial descent ( only the transabdominal phase ) , which results in ascrotal testes located in the lower abdomen , or complete descent ( transabdominal and inguinoscrotal phase ) , which results in scrotal testes [2 , 3] . A notable exception is Afrotheria , in which five of the six main lineages ( represented here by the lesser hedgehog tenrec , cape golden mole , cape elephant shrew , manatee , elephant , and rock hyrax ) do not show any testicular descent and have testes positioned at their initial abdominal position near the kidneys [2 , 3 , 15–17] . This lack of any testicular descent is termed testicondy . The aardvark is the only afrotherian exhibiting descended but ascrotal testes [2 , 3 , 18] . A schematic illustration of the different position of testes in mammals is shown in S1 Fig . Since Afrotheria represent one of the three main clades of placental mammals ( together with Xenarthra and Boreoeutheria ) , two different evolutionary scenarios could explain testicondy in several afrotherian lineages . First , if testicondy is the ancestral condition in placental mammals , then testicular descent was gained two or three times ( depending on the phylogeny ) in Xenarthra , Boreoeutheria , and the aardvark lineage . Second , if testicular descent is the ancestral condition in placental mammals , then testicular descent was lost once or more often ( again depending on the phylogeny ) in five of the six afrotherian lineages . Since soft-tissue structures like testes and the transient gubernaculum ligament are typically not preserved in the fossil record , the evolution of such soft-tissue structures can only be inferred by analytical methods such as parsimony , extant phylogenetic bracketing , or maximum likelihood [19–22] , all of which rely on the given phylogenetic tree . Consequently , resolving whether testicondy or testicular descent is the ancestral condition in placental mammals requires accurate knowledge of the underlying phylogeny . Unfortunately , although integrative approaches using both morphological and molecular characters have brought major advances in our understanding of mammalian phylogeny [23–26] , there is still no final consensus on the relationships between ( and sometimes within ) the main clades of placental mammals . In particular , the placental root and branching pattern of the clades Afrotheria , Xenarthra , and Boreoeutheria are still debated [26–28] ( S2 Fig ) , and an analysis of rare genomic events raised the concrete possibility of a near-simultaneous split [29] . Furthermore , the phylogeny within Afrotheria is not well resolved , because of conflicting evidence for the position of the aardvark ( the only nontesticond afrotherian lineage ) and the relationships between manatees , elephants , and hyraxes [30–34] ( S3 Fig ) . Given these phylogenetic uncertainties , it is probably not surprising that two different studies reached opposite conclusions about whether testicondy is the ancestral or derived state for placental mammals and for Afrotheria . Werdelin and Nilsonne [2] inferred that testicular descent in placental mammals and Afrotheria is the ancestral condition ( testicular descent was subsequently lost ) . However , their results were based on a phylogeny in which Afrotheria were nested within Boreoeutheria , which is not supported by current phylogenies . More recently , Kleisner and colleagues [3] reexamined the evolution of testicular descent in the context of current phylogenies and came to the opposite conclusion that testicondy in placental mammals and Afrotheria is the ancestral phenotypic character . Here , we sought to resolve this debate whether testicondy or testicular descent is the ancestral condition in placental mammals and in Afrotheria by using molecular evidence . First , we reasoned that if testicular descent is ancestral , then testicond afrotherian lineages may have lost key genetic information that is necessary for testicular descent . Such a loss of genetic information may be detectable by comparative genomics analysis . Second , we reasoned that if testicular descent is ancestral and if aardvarks are nested within Afrotheria , then testicondy would have evolved independently several times . This is expected to leave a signature of independent loss of the genetic information that is necessary for testicular descent . By analyzing the evolution of two key genes ( relaxin/insulin-like family peptide receptor 2 [RXFP2] and insulin-like 3 [INSL3] ) that are required for gubernaculum development and function in 71 placental mammals , we found that both genes have loss-of-function mutations only in several testicond afrotherian species . The absence of shared inactivating mutations and our age estimates for the loss of RXFP2 further suggest that testicondy evolved independently in afrotherian lineages at different time points . Together , these results provide not only a molecular mechanism that explains the loss of testicular descent in afrotherian lineages but also shows that testicular descent is the ancestral state for placental mammals and Afrotheria .
To determine if testicondy is the ancestral or derived condition for placental mammals and for Afrotheria , we examined two key genes that are necessary and sufficient for the development of the gubernaculum: INSL3 and RXFP2 . INSL3 encodes a relaxin-like hormone that is secreted by Leydig cells of the testes and binds specifically to the transmembrane receptor encoded by RXFP2 , which is highly expressed in gubernacular cells [35–39] . The INSL3-RXFP2 ligand-receptor pair promotes gubernacular cell proliferation and stimulates the swelling reaction [40–42] . Both genes are necessary for gubernacular function , as knockout of RXFP2 [37 , 39] or INSL3 [40 , 43 , 44] in mice results in the absence of the gubernaculum and no testicular descent , which in turn leads to spermatogenesis defects and male infertility . Despite the fact that RXFP2 is also expressed in postmeiotic spermatogenic cells , surgically correcting the position of undescended testes in global INSL3 knockout mice or a knockout of RXFP2 that is restricted to male sperm cells results in normal spermatogenesis and fertility [39 , 43] , suggesting that both genes are dispensable for spermatogenesis and germ cell survival in adult male mice . To investigate the evolution of RXFP2 and INSL3 in placental mammals , we made use of existing genome alignments between humans and 68 other mammals [45] . In addition , we further computed a genome alignment between human and the most recent genome assemblies of the rock hyrax and Hoffmann’s two-toed sloth ( Materials and methods ) . Inspecting the genomic loci that correspond to human RXFP2 and INSL3 allowed us to examine both genes in all 70 mammals , even in the absence of gene annotations for most of these species . To investigate if testicond Afrotheria lost the genetic information necessary for testicular descent , we first examined the coding region of RXFP2 and INSL3 in seven afrotherians with available genomes ( aardvark , lesser hedgehog tenrec , cape golden mole , cape elephant shrew , manatee , elephant , and hyrax ) . Our genome alignments revealed that four testicond lineages ( tenrec , cape golden mole , cape elephant shrew , and manatee ) have several mutations in RXFP2 that inactivate its reading frame . These gene-inactivating mutations create premature stop codons , shift the reading frame , disrupt the splice site dinucleotides , and delete entire exons ( Fig 1A ) . Furthermore , three out of these four species ( tenrec , cape elephant shrew , manatee ) also have inactivating mutations in the INSL3 gene ( Fig 2A ) . Since these mutations affect several exons and destroy functional protein domains in INSL3 ( A- and B-chain , Fig 2A ) , it is highly unlikely that the remnants of these genes encode a functional protein . Importantly , reciprocal-best BLAST hits and conserved gene order clearly show that these remnants are “molecular vestiges” that correspond to the RXFP2 and INSL3 genes ( S4 Fig ) . In analogy to vestigial organs , these molecular vestiges imply the presence of once functional RXFP2 and INSL3 orthologs that were subsequently lost in several afrotherians during evolution . To confirm that these inactivating mutations are real and do not represent genome assembly or alignment errors , we used a multistep validation approach . Since genome alignments do not take reading frame and splice site information into account , we first sought to rule out the possibility that inactivating mutations are a consequence of alignment ambiguities . To this end , we realigned all coding exons with the Coding Exon-Structure Aware Realigner ( CESAR ) , an exon alignment method that produces an alignment with consensus splice sites and an intact reading frame whenever possible [49 , 50] . CESAR confirmed that all affected exons exhibit inactivating mutations ( Figs 1 and 2A ) . Second , to validate that these mutations are not sequencing or assembly errors , we investigated raw sequencing reads from the Sequence Read Archive ( SRA ) [51] . For both RXFP2 and INSL3 , we found that all genomic loci containing an inactivating mutation are supported by at least 10 sequencing reads , while not a single read aligns to a putative sequence , in which the inactivating mutation was reversed to its ancestral state . We further confirmed the presence of two frameshifting mutations in RXFP2 in the lesser hedgehog tenrec by PCR and Sanger sequencing ( S5A and S5B Fig ) . Overall , this shows that the inactivating mutations shown in Figs 1 and 2A are not sequencing errors or artefacts arising from genome assembly or alignment issues . Finally , we investigated whether hitherto undetected functional copies of RXFP2 or INSL3 exist in afrotherians that may have arisen by lineage-specific duplications . By performing ultra-sensitive genome alignments , we only detected a single orthologous locus for RXFP2 and INSL3 . In addition , we found alignments to RXFP1 , a paralog of RXFP2 that exists in all placental mammals ( S6 Fig ) , showing that these alignment parameters are sufficiently sensitive to even detect more ancient gene duplications . Together , this excludes the possibility that afrotherians possess another functional duplicated copy of RXFP2 or INSL3 . If RXFP2 and INSL3 are truly lost , we further expect that they evolve neutrally in the lineages with inactivating mutations . Indeed , using RELAX [52] , we found that RXFP2 evolves under relaxed selection in all four gene-loss species ( adjusted P values < 3 . 2e−5 , S1 Table ) . For INSL3 , no significant evidence for relaxed selection was found , likely because large deletions in this short 131-residue protein in tenrec , cape elephant shrew , and manatee ( Fig 2A ) severely reduced alignment length . Therefore , we inspected the two protein domains that are necessary for the function of the mature INSL3 hormone . Similar to insulin , the preprohormone INSL3 is processed into an A- and B-chain peptide . The A- and B-chain then forms a heterodimer that is stabilized by two disulfide bonds between the A- and B-chains and one disulfide bond within the A-chain [47] . We found that tenrec , cape elephant shrew , and manatee have deletions that overlap the A- and B-chain and affect residues that are critical for INSL3 structure and function ( Fig 2B ) . Together , our results conclusively show that the remnants of RXFP2 and INSL3 cannot encode functional proteins in several testicond afrotherian lineages . Since INSL3 lacks clear inactivating mutations in the cape golden mole , we examined the residues that are important for INSL3 structure and function . We found that the Cys at position 10 in the A-chain that forms a disulfide bond with Cys at position 15 [47] is mutated to a Tyr in the cape golden mole ( Fig 2B ) . Furthermore , the Lys at position 8 in the B-chain ( Fig 2B ) , a residue that is important for receptor activation [48] , is deleted in this species . This suggests that , while INSL3 still has an intact reading frame in the cape golden mole , it accumulated mutations that most likely render the encoded protein nonfunctional . Interestingly , both RXFP2 and INSL3 lack any gene-inactivating mutations in the elephant and the rock hyrax , two afrotherians that are also testicond [15 , 17] . While the elephant has a 2-bp deletion in the last exon of RXFP2 , this merely truncates the C-terminus by 25 residues and is not an indication of loss ( see section RXFP2 and INSL3 are intact in all nontesticond placental mammals and S7 Fig ) . Furthermore , RELAX estimates a Ka/Ks value of 0 . 31 and 0 . 33 for elephant and rock hyrax , respectively , which is slightly but not significantly higher than the Ka/Ks value of 0 . 27 observed for other mammals . Thus , there is no significant evidence for relaxed selection in these two lineages . We also scanned both genes for amino acid mutations that were only observed in human cryptorchidism patients ( V18M , P49S , W69R , P93L , R102C , R102H , R105H , N110K in INSL3 and T222P in RXFP2 [8] ) . Whereas elephant INSL3 exhibits the R102H mutation , this mutation is observed in many other nontesticond mammals , and cell line experiments have shown that this mutation does not affect INSL3 activity [38] . Similarly , elephant RXFP2 has a T222A ( Thr to Ala ) mutation at a position where a mutation from Thr to Pro renders RXFP2 nonfunctional [37 , 53] . However , the T222A mutation that is present in elephant is also observed in the nontesticond aardvark and pangolin , and experiments have shown that mutating this Thr to Ala does not affect RXFP2 function [53] . The rock hyrax does not exhibit any of the mutations observed in human cryptorchidism patients . Based on these evidences , elephant and rock hyrax RXFP2 and INSL3 may encode functional proteins . So far , our analysis suggests that the function of RXFP2 and INSL3 is only compromised in several testicond afrotherians . Therefore , we examined both genes in the aardvark , the only afrotherian exhibiting partial testicular descent [2 , 18] , and found that RXFP2 and INSL3 are intact and evolve under selection ( S1 Table ) and that INSL3 lacks any mutations of critical amino acids ( Fig 2B ) . To further investigate the relation between loss of RXFP2 and INSL3 and testicondy , we examined both genes in 64 other nontesticond mammals . While the genome alignment showed a few putative inactivating mutations in some species , a detailed manual inspection revealed that these are assembly errors ( S8 Fig ) , assembly gaps ( S9 Fig ) , and alignment ambiguities ( S10 Fig ) . For RXFP2 , we further found that the N-terminus is 17 amino acids longer in human , chimpanzee , bonobo and gorilla ( S11 Fig ) and that several species have small truncations and elongations of the C-terminus without affecting the transmembrane domains of the receptor ( S7 Fig ) . Since RXFP2 does not evolve under relaxed selection in any of these 64 mammals ( S1 Table ) , these length variations are not an indication of gene loss but support previous observations that N- and C-termini of proteins are evolutionarily less constrained [49 , 54] . Together , this shows that both genes are intact and under selection in all other nontesticond placental mammals . Interestingly , we observed no inactivating mutations in RXFP2 and INSL3 that are shared among any testicond afrotherian species , suggesting that the loss of these genes happened independently , after these species split from their common ancestors . Since RXFP2 and INSL3 are expected to evolve neutrally after the loss of testicular descent , an estimate of how long these genes have been evolving neutrally provides an estimate for when testicondy occurred . To this end , for the four branches in the phylogenetic tree leading to the four gene-loss species , we estimated the portion of the branch where the gene evolved under selection and the portion where it evolved neutrally , as described in [55 , 56] . Since large parts of INSL3 are deleted in tenrec , cape elephant shrew and manatee ( Fig 2A ) and since exon 1 overlaps assembly gaps in several other species ( Fig 3 ) , we focused on RXFP2 , for which each gene-loss species provides at least 594 bp in the codon alignment , to obtain robust estimates . As shown in Fig 4 and S2 Table , we estimate that each lineage lost the RXFP2 gene at different time points . Consistent with the large number of inactivating mutations , RXFP2 appears to be lost first in the cape elephant shrew around 66–83 million years ago ( Mya ) . For the lesser hedgehog tenrec , we estimate that RXFP2 loss happened around 50–59 Mya . Consistent with this estimate , we found that the same gene-inactivating mutation in RXFP2 exon 17 is shared with its sister species greater hedgehog tenrec ( S5B Fig ) , suggesting that RXFP2 loss already occurred in the ancestor of both tenrec species that lived 7–14 Mya ( Fig 4 ) . The loss of RXFP2 in manatee is estimated to have happened around 43–51 Mya and thus likely predates the split of the manatee and dugong lineage 26–53 Mya . To test this , we used PCR and sequencing experiments and found that the dugong shares two stop codon mutations in different exons with the manatee ( S5C and S5D Fig ) , confirming that RXFP2 loss predates the split of manatees and dugongs . The loss in the cape golden mole likely happened more recently ( 23–28 Mya ) , consistent with our observations that INSL3 did not yet accumulate a gene-inactivating mutation . While the absolute estimates of gene-loss times are tentative ( as fossil-based time calibrations of the species divergence times are lacking ) , these time intervals are substantially different from each other ( Fig 4 ) . Together with the absence of shared inactivating mutations , this strongly suggests that testicondy evolved independently in Afrotheria .
The evolution of testicondy and whether testicular descent is the ancestral [2] or derived state [3] for placental mammals and for Afrotheria has been controversial , despite agreement in the phenotypic character assignment . The different conclusions are mainly due to persisting differences in the phylogeny , which affect ancestral character reconstruction . To resolve this debate , we investigated the evolution of RXFP2 and INSL3 , two genes encoding a hormone receptor pair that is required for the development of the testes-descending gubernaculum ligament . We found remnants of once functional orthologs of RXFP2 and INSL3 as molecular vestiges in four testicond afrotherian lineages . Together with the presence of orthologs of both genes in other Afrotheria and other placental mammals , this shows that these genes were lost after the testicond lineages split from the afrotherian ancestor . This allows us to conclude that testicular descent is the ancestral condition in placental mammals and was subsequently lost in different afrotherian lineages . Importantly , our conclusion holds regardless of persisting phylogenetic discrepancies that involve the branching pattern of Afrotheria , Xenarthra , and Boreoeutheria at the placental root , and the phylogeny within Afrotheria ( S2 and S3 Figs ) . Our study also provides three lines of evidence that testicondy evolved independently in Afrotheria . First , if testicondy evolved in the common ancestor of any two testicond afrotherians , we would expect inactivating mutations in RXFP2 and INSL3 that are shared among species . However , both genes do not exhibit any shared inactivating mutations . Second , by estimating how long these genes have been evolving neutrally , we found different time intervals in which gene loss and likely testicondy evolved . These estimated time intervals may be helpful to better understand the ecological conditions under which testicondy evolved repeatedly . Third , independent evolution of testicondy is further supported by our finding that the testicond elephant and rock hyrax have intact RXFP2 and INSL3 genes that still appear to evolve under purifying selection . Since recent shifts from purifying selection to neutral evolution will not leave a detectable signature of significantly increased Ka/Ks ratios , our results suggest that elephant and rock hyrax RXFP2 and INSL3 may have come under relaxed or no selection in recent evolutionary time . In agreement with a more recent evolution of testicondy , the rock hyrax still exhibits rudiments of gubernacular structures [15] . It is possible that amino acid mutations or cis-regulatory mutations that affect the expression of RXFP2 or INSL3 are responsible for testicondy in elephant and rock hyrax . Indeed , the RXFP2 promoter region exhibits more sequence divergence in the rock hyrax than in all other nontesticond species , and the elephant sequence also shows elevated divergence compared to most but not all mammals ( S12 Fig ) . However , the precise genomic basis of testicondy in elephant and rock hyrax remains to be elucidated . Spermatogenesis and sperm storage require temperatures below 37 °C . Body temperatures below 35 °C , as observed in the heterothermic tenrecs and golden moles [7] , may abolish the need for testicular cooling , which would explain why testicondy is tolerated in these species . However , other afrotherian species have body temperatures similar to many scrotal mammals . For example , the body temperature of elephants and elephant shrews is 36 . 8 °C and 37 . 2 °C , respectively [7 , 58] . This raises the questions of how these testicond afrotherian lineages maintain normal testicular function at a body temperature of approximately 37 °C and whether these species have evolved novel cooling mechanisms . Previous anatomical studies did not reveal conclusive evidence of such cooling systems . In particular , testicond afrotherians do not possess a pampiniform plexus or a vascular countercurrent heat exchanger that could act as a comparable cooling system [17 , 59 , 60] . In the dugong , the epididymis has a large , folded surface and is surrounded by highly vascularized tissue [60]; however , whether this arrangement acts as a cooling system is not clear . In rock hyrax and elephant , it was observed that the part of the testicular duct that stores spermatozoa lies close to the body surface [59] , but it remains unknown whether this anatomical feature has a cooling function during sperm storage . Hence , detailed comparative and functional studies of the testes , epididymis , and their associated anatomical structures are required to understand whether Afrotheria evolved novel mechanisms for sperm cooling during spermatogenesis and storage . Furthermore , it would be of great interest to learn why several afrotherian but not any other placental lineages have completely lost testicular descent . For example , it is unknown whether the loss of testicular descent is beneficial for these species , whether anatomical constraints associated with body plan or life history explain the loss of descent , or whether this phenotypic reversal is an evolutionary tradeoff for another advantageous trait . More anatomical studies that in particular compare the development of Afrotheria are required to address these questions . More generally , our results highlight how molecular evolution can shed light on the evolution of phenotypes . While the fossil record has contributed substantially to our understanding of how hard-tissue characters such as bones , teeth , or shells evolved , the evolution of soft-tissue structures that do not fossilize can often only be inferred with analytical methods . Moreover , conclusions about the ancestry of such characters are dependent on the underlying phylogeny . Molecular vestiges offer an alternative strategy to investigate character ancestry . This strategy may be broadly applicable , since molecular vestiges are also known for other phenotypes whose ancestry is not controversial based on fossil evidence and accurate knowledge of the underlying phylogeny . For example , remnants of enamel-related genes in toothless mammals [61–63] , remnants of hair development genes in hairless cetaceans [63] , remnants of gastric genes in vertebrates without a stomach [64] , or remnants of eye-related genes in subterranean mammals with degenerated eyes [65–68] show that these species descended from ancestors with teeth , hair , a stomach , and functional eyes , respectively . Thus , instead of investigating a soft-tissue structure directly , one can trace the evolution of genes that are crucial for the development of this structure . Molecular vestiges of such genes can then provide insights into the ancestry of soft-tissue structures , even if the phylogenetic positions of the respective species remain controversial .
To investigate the coding region of RXFP2 and INSL3 in placental mammals , we used an alignment between the human hg38 genome assembly and the genomes of 68 nonhuman placental mammals [45] . Since this alignment does not provide the rock hyrax ( Procavia capensis ) and Hoffmann’s two-toed sloth ( Choloepus hoffmanni ) , two mammals for which improved genome assemblies have recently become available , we used the pipeline of lastz ( version 1 . 03 . 54 , parameters K = 2 , 400 , L = 3 , 000 , Y = 9 , 400 , H = 2 , 000 , and the lastz default scoring matrix ) [69] , chaining ( parameters chainMinScore 1 , 000 , chainLinearGap loose ) , and netting ( default parameters ) [70] to compute new genome alignments . Since several species have assembly gaps overlapping INSL3 exon 1 , likely because this GC-rich region gets poor coverage in Illumina sequencing , we further computed genome alignments using updated assemblies of bonobo , domestic goat , camel , and dolphin . We also inspected the genome alignment to the recent gorGor5 assembly of the gorilla , provided by UCSC [71] . For gorilla , goat , and dolphin , the new assemblies closed these assembly gaps and revealed an intact INSL3 exon 1 . For bonobo and camel , the new assemblies still have assembly gaps overlapping INSL3 exon 1 . S3 Table lists all species and their assemblies that were analyzed . We examined the colinear alignment chains ( loaded in the UCSC genome browser ) in order to assess if gene order around the RXFP2 and the INSL3 genes is conserved in Afrotheria . This showed that both genes and several neighboring genes occur in a conserved order in afrotherian genomes . Furthermore , existing gene annotations ( Ensembl ) and reciprocal-best BLAST of the human proteins confirmed that the aligning loci contain orthologous RXFP2 and INSL3 genes or their remnants . To perform ultra-sensitive genome alignments that could reveal any hitherto undetected functional copies of RXFP2 or INSL3 , we first computed genome alignments between the human genome ( hg38 assembly ) and each of the seven afrotherians , using highly-sensitive lastz parameters by setting K = 2 , 000 and L = 2 , 500 . Subsequently , we used even more sensitive parameters to find additional local alignments by running lastz with parameters K = 1 , 500 , L = 2 , 200 , and W = 5 on all chain gaps ( nonaligning regions flanked by aligning blocks ) that are at least 30 bp and at most 500 Kb long , as described in [45 , 72] . We built alignment chains from all local alignments [70] and visualized them in the UCSC genome browser [71] . We used the genome alignments to systematically search each exon for frameshifting insertions and deletions , mutations that create in-frame stop codons , and mutations that disrupt the conserved splice site dinucleotides ( donor GT/GC , acceptor AG ) . We validated each putative mutation by the following steps: First , we realigned the exonic sequence of each exon with an inactivating mutation , using CESAR [49 , 50] with default parameters . CESAR is a Hidden-Markov-Model based aligner that takes the reading frame and the splice sites into account and tries to produce an intact exon alignment ( no inactivating mutations , consensus splice sites ) between human and a query species , wherever possible . All exons for which CESAR confirmed the existence of at least one inactivating mutation are shown in Fig 3 . S10 Fig shows an example for which CESAR found an alternative alignment that lacks inactivating mutations . Second , we validated the remaining inactivating mutations by searching SRA for unassembled sequencing reads that have a 100% match to the genomic context comprising at least 50 bp around the mutation , as described in [73] . Furthermore , we searched SRA for a putative sequence , in which the inactivating mutation was reversed to its ancestral state . Every mutation that is supported by at least 10 sequencing reads without any hit to the putative “ancestralized” sequence was considered as real . Third , we examined all exons that do not align in the query genome , since these are either truly deleted or an artifact arising from incomplete genome assemblies [68 , 74] . We used the nearest up- and downstream aligning blocks in the alignment chain to determine the genomic locus that corresponds to this exon in the query genome . If this locus does not overlap an assembly gap , we concluded that the exon is lost ( either deleted or accumulated so many mutations that it does not align anymore ) . If this locus overlaps an assembly gap , we conservatively considered this exon as missing sequence , as described in [63] . Indeed , as shown in S9 Fig , exons that overlapped assembly gaps in a previous assembly readily aligned with an intact reading frame in an improved genome assembly , showing that assembly gaps should not be mistaken for exon loss . To determine if the RXFP2 coding sequence evolves neutrally in different species , we used RELAX version 2 [52] . Briefly , given an alignment , a tree topology , and branches labeled as either test or reference , RELAX determines if the gene evolves under relaxed or intensified selection in the test branches relative to the reference branches . We used RELAX’s partition descriptive method that fits three Ka/Ks classes to the test and the reference branches and also estimates an overall Ka/Ks value . To determine if RXFP2 evolves under relaxed selection in any mammal , we iteratively applied RELAX , labeling each of the 71 species as the test branch . The resulting P values were corrected for multiple testing using the Benjamini and Hochberg method ( S1 Table ) . To exclude any bias caused by relaxed selection in the RXFP2-loss species , which might inflate the reference Ka/Ks values , we further applied RELAX , labeling only one afrotherian species as test and excluding all other afrotherian species from the reference branches . These tests confirmed that RXFP2 evolves under relaxed selection only in the four afrotherians with an inactivated RXFP2 gene but not in the three other afrotherian species ( S1 Table ) . To date the loss of RXFP2 , we followed the procedure described in [55 , 56] . For a branch along which the gene was inactivated , this method assumes that a gene evolves under a selective pressure similar to that in other species until it is inactivated . Afterward , the gene is assumed to accumulate both synonymous and nonsynonymous mutations at a neutral rate . The Ka/Ks value ( K ) estimated for this entire branch is then the average of the Ka/Ks value for the part of the branch where the gene was under selection ( Ks ) and the Ka/Ks value for the part of the branch where the gene evolved neutrally ( Kn = 1 ) , weighted by the proportion of time for which the gene was evolving under selection ( Ts / T ) and neutrally ( Tn / T ) : K=Ks×Ts/T+Kn×Tn/T , where T represents the time since the split from the last common ancestor . Using the lower and upper bound of the confidence interval for the species divergence time T obtained from TimeTree [57] ( S4 Table ) and using the Ka/Ks value for mammals with a functional RXFP2 ( Ks = 0 . 2491 ) , one can estimate a lower and upper bound for Tn as Tn=T× ( K−Ks ) / ( 1−Ks ) , which provides an estimate of how long RXFP2 has been evolving neutrally ( S2 Table ) . DNA of the greater hedgehog tenrec ( Setifer setosus ) and the lesser hedgehog tenrec ( Echinops telfairi ) was kindly provided by Athanasia Tzika ( University of Geneva ) . Dugong tissue was kindly provided by Joerns Fickel ( Leibniz-Institute of Zoo and Wildlife Research , Berlin ) . DNA was extracted using the innuPrep DNA Mini Kit ( Analytik Jena , Jena , Germany ) following the manufacturer’s instructions ( protocol for DNA isolation from tissue samples or rodent tails ) but extending tissue lysis overnight and including RNA digestion . Standard PCR reactions were performed in a total volume of 10 μl containing 5–20 ng DNA , 5 μl Phusion Flash High-Fidelity PCR Master Mix ( Thermo Scientific , Waltham , United States of America; #F-548S ) , 0 . 25% DMSO ( New England Biolabs , Frankfurt Main , Germany ) , and 0 . 5 μM of each primer ( S5 Table ) . Cycling conditions were as follows: 35 cycles were used with denaturation at 98 °C ( 15 s but 2 min for the first cycle ) , annealing at a primer pair–specific temperature ( S5 Table , 20s ) , and extension at 72 °C ( 10 s–1 min but 5 min after the last cycle ) . PCR products were Sanger-sequenced after enzymatic cleanup with CleanDTR ( GC biotech , Waddinxveen , Netherlands ) followed by cycle sequencing with the PCR primers and a mix of the BigDye Terminator v3 . 1 Cycle Sequencing Kit and dGTP BigDye Terminator v 3 . 0 Kit ( Applied Biosystems , Foster City , CA , USA ) . Analyses were performed on an ABI 3730xl DNA Analyser ( Applied Biosystems ) . We analyzed per-species sequence divergence of the promoter region of RXFP2 and INSL3 . Using the human hg38 assembly coordinates chr13:31739464–31739544 and chr19:17821524–17821716 , we extracted the sequence of all placental mammals , used PRANK [75] with parameters “-keep -showtree -showanc -prunetree” to align these sequences and to reconstruct the most likely sequence of the placental mammal ancestor and measured the percent sequence identity between the ancestral sequence and the sequence of every extant mammal , as described in [68 , 74] . The result is visualized in S12 Fig .
|
While fossils of whales with legs demonstrate that these species evolved from legged ancestors , the ancestral state of nonfossilizing soft-tissue structures can only be indirectly inferred . This difficulty is also confounded by uncertainties in the phylogenetic relationships between the animals concerned . A prime example is the case of testicular descent , a developmental process that determines the final position of testes , which occurs in most placental mammals but is absent from several afrotherian lineages . Here , we discovered that afrotherians possess remnants of genes known to be required for testicular descent . These “molecular vestiges” show that testicular descent was already present in the placental ancestor and was subsequently lost in Afrotheria . Our study highlights the potential of molecular vestiges in resolving contradictory ancestral states of soft-tissue characters .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"methods"
] |
[
"taxonomy",
"hyraxes",
"vertebrates",
"animals",
"mammals",
"animal",
"phylogenetics",
"phylogenetics",
"data",
"management",
"mammalian",
"genomics",
"zoology",
"research",
"and",
"analysis",
"methods",
"sequence",
"analysis",
"computer",
"and",
"information",
"sciences",
"sequence",
"alignment",
"bioinformatics",
"afrotheria",
"short",
"reports",
"evolutionary",
"systematics",
"animal",
"genomics",
"elephants",
"eutheria",
"eukaryota",
"database",
"and",
"informatics",
"methods",
"genetics",
"biology",
"and",
"life",
"sciences",
"genomics",
"evolutionary",
"biology",
"amniotes",
"organisms"
] |
2018
|
Loss of RXFP2 and INSL3 genes in Afrotheria shows that testicular descent is the ancestral condition in placental mammals
|
Human α-defensins are potent anti-microbial peptides with the ability to neutralize bacterial and viral targets . Single alanine mutagenesis has been used to identify determinants of anti-bacterial activity and binding to bacterial proteins such as anthrax lethal factor . Similar analyses of α-defensin interactions with non-enveloped viruses are limited . We used a comprehensive set of human α-defensin 5 ( HD5 ) and human neutrophil peptide 1 ( HNP1 ) alanine scan mutants in a combination of binding and neutralization assays with human adenovirus ( AdV ) and human papillomavirus ( HPV ) . We have identified a core of critical hydrophobic residues that are common determinants for all of the virus-defensin interactions that were analyzed , while specificity in viral recognition is conferred by specific surface-exposed charged residues . The hydrophobic residues serve multiple roles in maintaining the tertiary and quaternary structure of the defensins as well as forming an interface for virus binding . Many of the important solvent-exposed residues of HD5 group together to form a critical surface . However , a single discrete binding face was not identified for HNP1 . In lieu of whole AdV , we used a recombinant capsid subunit comprised of penton base and fiber in quantitative binding studies and determined that the anti-viral potency of HD5 was a function of stoichiometry rather than affinity . Our studies support a mechanism in which α-defensins depend on hydrophobic and charge-charge interactions to bind at high copy number to these non-enveloped viruses to neutralize infection and provide insight into properties that guide α-defensin anti-viral activity .
Human α- and β-defensins are small ( 3–5 kDa ) , cationic peptides of the innate immune system with broad anti-microbial activity [1] . The six human α-defensins can be further divided by expression pattern and gene structure into myeloid [human neutrophil peptides ( HNPs ) 1–4] or enteric [human α-defensins ( HDs ) 5 and 6] classes [2] , [3] . Despite their variable sequences , α-defensins share common structural features including a triple-stranded β-sheet fold , three intramolecular disulfide bonds , and a salt bridge [4] . The activity of α-defensins against both gram-negative and gram-positive bacterial pathogens has been well characterized , while their anti-viral properties are less well understood [4] , [5] . Their capacity to neutralize enveloped viruses can be explained in part through properties identified in anti-bacterial studies , including lipid perturbation and their ability to function as lectins , although other mechanisms have been proposed [5] , [6] . In contrast , these properties are insufficient to explain their ability to inhibit multiple non-enveloped viruses . In this regard , we have shown that HD5 neutralizes human AdV by binding to fiber and penton base proteins at the vertices of the icosahedral capsid , thereby stabilizing the capsid and preventing uncoating and subsequent genome exposure [7]–[10] . Similarly , recent studies have identified post-entry blocks of HPV and JC polyomavirus infection by HD5 [11] , [12] , suggesting that common mechanisms may govern α-defensin neutralization of non-enveloped viruses . Extensive structure-function studies of multiple α-defensins have identified features that dictate their anti-bacterial activity , including a prominent role for dimerization and higher order multimerization [13]–[15] . Dimerization also contributes to α-defensin binding to glycoproteins and bacterial toxins [13] , [15] . Equivalent structure-function studies of α-defensin anti-viral activity , particularly for non-enveloped viruses , are lacking . In a recent study , we identified certain arginine residues and the need for stable dimer formation as crucial for HD5 inhibition of AdV and HPV [16] . To more globally assess HD5 function and define a viral binding interface , we tested HD5 analogues from a comprehensive alanine scan library for their ability to neutralize AdV and HPV and for their binding kinetics to AdV capsid proteins . We define a critical patch on the surface of HD5 important for both HPV and AdV inhibition . We also show that the stoichiometry rather than affinity of HD5 binding to the AdV vertex correlates with anti-viral activity . Additionally , we identify regions important for HNP1 anti-AdV activity . Comparison of similarities and dissimilarities between these two α-defensins may inform general rules for α-defensins in innate anti-viral immunity .
We have shown that specific arginine residues and the hydrophobicity of residue 29 are important for HD5 to function as an anti-viral molecule against human AdV and HPV [16] . Although human AdVs cause a broad spectrum of diseases , many serotypes are transmitted by the fecal-oral route , where they may encounter intestinal HD5 [9] , [17] , [18] . Similarly , HPV16 and related mucosal serotypes encounter HD5 in the female reproductive tract [19] . To systematically identify additional critical residues , we measured the effect of alanine substitutions in HD5 on its ability to neutralize a human AdV-5-based vector ( AdV5 . eGFP ) and HPV16 pseudoviruses ( PsVs ) . The anti-bacterial activity of these HD5 analogues has been reported [13] . Together with the single arginine substitutions from our previous study , they complete a collection of alanine substitutions for all non-alanine residues in HD5 except for those involved in three conserved features: a glycine ( G18 ) required to form a β-bulge , a salt bridge ( R6 and E14 ) , and the six disulfide-bonded cysteine residues [13] , [20] , [21] . Briefly , defensins were incubated with AdV5 . eGFP or HPV16 PsV , the mixtures were added to cells , and expression of GFP was quantified relative to control cells infected without defensin . HD5 was used at low micromolar concentrations , which are within the physiologic range of HD5 expression in the small intestine and female reproductive tract [5] . As expected , alanine substitutions for most of the residues in HD5 had less than a 2-fold effect on anti-viral activity ( Figure 1 ) . Substitution for hydrophobic residues or a bulky aromatic residue ( Y27 ) had the greatest effect . The only exceptions were I22 for AdV and Y4 for both AdV and HPV . All of the residues that impacted AdV inhibition were also important for HPV , although in general there was less attenuation of HPV inhibition . When ranked by relative effect , the overall order of the alanine substitutions was consistent between the two viruses with the exception of E21 , which was more important for HPV than AdV inhibition . We used a heat map to collate our data from both studies and to compare the importance of HD5 residues for anti-viral and anti-bacterial activity ( Figure 2A ) . Several residues critical for anti-viral activity are grouped towards the C-terminus of the peptide and are primarily hydrophobic , with the exception of the positively charged R28 [16]; however , L16 and V19 in the middle of the peptide sequence are also important for both AdV and HPV16 inhibition . With the exception of L29 , residues critical for neutralization of viral infection differ from those required to kill bacteria [13] , suggesting a distinctive mode of interaction between HD5 and these non-enveloped viruses . As the critical anti-viral residues were separated in the primary structure , we visualized their organization on a space-filling model of the HD5 dimer ( Figures 2B and C ) . The residues needed for neutralization of both AdV and HPV16 localize to one face of the dimer; whereas , mutation of residues on the opposite face had little or no effect . Key residues for inhibition of both AdV and HPV ( L16 , V19 , and L26 ) are clustered together and surface exposed , while the side chains of L29 and Y27 are buried . For AdV , this hydrophobic surface is divided into two discrete patches and extended by the surface-exposed side chain of the positively charged R28 . For HPV , inclusion of E21 forms a contiguous surface across the dimer interface . The surface exposed residues might directly contribute to the interaction of HD5 with the viral capsid . To test this hypothesis , we required a sensitive assay to measure binding . Although not precisely defined , previous studies indicated that the determinants for HD5 binding on AdV are within the vertex proteins , fiber and penton base [9] , [22] . Defensin binding to the vertex stabilizes the capsid and blocks viral uncoating during cell entry [7]–[9] , [23] . Accordingly , we reasoned that surface plasmon resonance ( SPR ) analysis of HD5 binding to penton capsomeres comprised of only fiber and penton base might circumvent a prohibitive mass difference between the intact virus ( ∼150 MDa ) and HD5 ( ∼3 . 6 kDa ) . The baculovirus expression system was used to generate full-length human AdV-5 fiber that was purified by ion exchange chromatography . Human AdV-5 penton base ( PB ) with an N-terminal 6×His tag was created in bacterial cells and purified by cobalt affinity chromatography . Size exclusion chromatography was used to confirm the trimerization of fiber and pentamerization of PB ( data not shown ) . To form the rPenton complex , we co-incubated fiber and PB overnight at a 2∶1 molar ratio to minimize the number of uncomplexed PB subunits in the sample . rPenton was then purified by cobalt affinity chromatography through the 6×His tag on PB . We made multiple independent rPenton preparations . In each , the presence of both PB and fiber in the final rPenton product was confirmed by immunoblot using an anti-fiber monoclonal antibody and PB anti-sera ( Figure 3A ) . Analysis of total protein by SDS-PAGE and Coomassie stain indicated that complex formation contributed substantially to purification ( Figure 3B ) . And , EM analysis of purified rPenton ( Figure 3C ) revealed features identical to intact pentons liberated from mature AdV particles by incubation under hypotonic conditions at low pH [24] . The fiber shaft , fiber knob , and PB are all clearly apparent . A few large aggregates were observed , which may be composed of PB and an unidentified non-viral contaminant ( * in Figure 3B ) . Based on these analyses , a single rPenton preparation with the least amount of free fiber visible by EM was used for binding studies . To evaluate the functionality of rPenton , we individually coupled the parent fiber and PB proteins as well as rPenton to the dextran matrix of serial flow cells of a CM5 chip and employed SPR to measure binding of purified CAR-Ig , a soluble form of the AdV receptor fused to the constant region of rabbit Ig . As expected , CAR-Ig was unable to bind to PB but was able to bind to fiber and rPenton ( Figure 3D ) . Differences in the degree of binding to fiber compared to rPenton likely reflect the amount of each protein immobilized on the chip and the relative purities of the protein preparations . Taken together , these studies indicate that immobilized rPenton is a functional subunit of the AdV capsid and is a suitable substrate for binding analysis by SPR . We first established conditions to measure the affinity of wild type HD5 for rPenton ( Figure 4A ) . Immobilization of 2115 RU of rPenton yielded a response much greater than expected for 1∶1 binding . Despite prolonged injection of analyte ( 30 min ) at 500 nM , we never achieved saturation , suggesting non-specific interaction with the dextran matrix ( data not shown ) . This phenomenon has been previously reported and was most apparent at the highest HD5 concentration ( 9 µM ) [25] . Nonetheless , the system was not mass transport limited in flow rate analyses ( data not shown ) . Thus , we limited our association time to 300 sec and used a low flow rate to conserve analyte . We then measured binding of two HD5 analogues , HD5 Abu and HD5 E21me , that were previously studied in semi-quantitative whole virus binding assays [9] , [16] . HD5Abu is a linear analogue of HD5 that contains substitutions of α-aminobutyric acid ( Abu ) for the six cysteine residues , precluding disulfide bond formation . HD5Abu lacks anti-viral activity and does not bind to whole virus [9] . HD5 E21me cannot dimerize due to disruption of stabilizing hydrogen bonds by methylation of the peptide bond between C20 and E21 [13] . It has significantly reduced anti-viral activity , which correlates with decreased capsid binding [16] . By SPR analysis , E21me has greatly reduced binding to rPenton ( Figure 4B ) , and HD5Abu does not bind ( Figure 4C ) . Thus , rPenton is a reasonable proxy for whole virus . We selected HD5 analogues from the alanine scan with a range of inhibitory activity and analyzed their binding kinetics to rPenton by SPR . Given our results with wild type HD5 , we studied a 3-fold dilution series of each analogue from 3 µM to 111 nM . The sensorgrams of 333 nM of each defensin all had similar shapes , suggesting that the on- and off-rates were comparable ( Figure 5A–C ) . One exception was L29A ( dotted black line in Figure 5C ) , which appeared to have a much slower on-rate . Because the sensorgrams were not well fitted to the built in analysis equations of the Biacore software and did not reflect 1∶1 binding , we derived steady-state binding curves of the data at 80 sec of association time to quantify affinity ( KD ) and maximal binding at saturation ( Bmax ) ( Figure 5D ) . Wild type HD5 bound to rPenton with a KD of 1 . 81±0 . 22 µM ( Figure 5E ) . The majority of the mutants bound with similar affinity ranging from 1 . 26–1 . 92 µM . Two notable exceptions were Y27A ( KD = 3 . 39±0 . 27 µM ) and L29A ( KD = 3 . 82±0 . 30 µM ) , which bound with much lower affinity . When ordered by their relative anti-viral activity , there was no correlation between affinity and neutralization ( p = 0 . 57 , when Y27A and L29A are excluded ) . In contrast , there was a positive correlation between Bmax and anti-viral activity ( p = 0 . 0027 , when L29A is excluded ) . The value of Bmax for wild type HD5 was 3050±130 RU ( Figure 5F ) , corresponding to a stoichiometry of 210 ( Equation 1 ) . The corresponding Bmax value for the most attenuated mutant ( Y27A ) was 640±25 RU . As was the case for KD , the behavior of L29A was unique in that it had a Bmax higher than that of wild type HD5 ( 5445±230 RU ) . Taken together , these studies suggest that the number of HD5 molecules bound to the capsid rather than their absolute affinity dictate anti-viral activity . Nonetheless , a minimal affinity is required , since the stoichiometry of L29A exceeds that of wild type HD5 yet L29A is not capable of neutralizing infection , likely due to its lower affinity . To determine whether the properties of HD5 that are required for potent anti-viral activity might extend to a second human α-defensin , we analyzed a set of alanine scan mutants of the myeloid α-defensin HNP1 for their capacity to inhibit AdV infection . Compared to HD5 , HNP1 has reduced ability to neutralize AdV infection [7] , [9] . In addition , we observed that HNP1 was a more potent inhibitor of AdV when added to virus already pre-bound to the cell rather than when the defensin was pre-incubated with the virus prior to addition to cells ( data not shown ) . Consequently , we restricted our analysis to AdV , as the kinetics of HPV binding and entry preclude parallel analysis of HPV under these conditions . Approximately half of the HNP1 alanine mutants had a greater than 2-fold effect on IC50 ( Figure 6A ) . Two of the most deleterious mutations were of C-terminal hydrophobic residues , W26 and F28 , which align with Y27 and L29 of HD5 in linear sequence ( Figure 6C ) . Similar to our prior studies of the contribution of hydrophobicity at position 29 of HD5 to anti-AdV activity [16] , substitution of W26 with non-natural amino acids of increasing hydrophobicity partially restored the anti-viral activity of HNP1 ( Figure 6B ) . HNP1 W26Nva had an IC50 of almost 20 µM , while W26Nle and W26Ahp had IC50s between 10 and 20 µM . Together , these data emphasize the role of hydrophobicity in α-defensin neutralization of AdV . Analysis of HNP1 surface exposed residues ( Figure 6D ) revealed a more complex pattern than we observed for HD5 ( Figure 2B ) . Rather than a discrete interface located on one side of the molecule , we observed two patches on opposite faces of each monomer for a total of four potential interacting surfaces on each dimer . These surfaces were mostly formed by residues that when mutated had less than a 3-fold effect on IC50 , while for HD5 several of the prominent surface-exposed residues were the most critical . A direct comparison of the structures of HD5 and HNP1 reveals some similarities in the relative geometry of the side chains of key residues ( Figure 7 ) . For both α-defensins , the guanidinium group of an important arginine residue ( R15 for HNP1 and R28 for HD5 ) occupies the center of one aspect of the β-sheet that comprises the α-defensin fold . Buried aromatic residues ( W26 and F28 in HNP1 and Y27 and L29 in HD5 ) are located on the opposite side of the β-sheet and stabilize the hydrophobic core of the defensin . These residues affect the configuration of the surface exposed residues , the integrity of the dimer interface , and the ability of the defensins to self-associate and to neutralize virus infection [15] , [26] . The importance of V19 in HD5 but not T18 in HNP1 underscores the functional importance of hydrophobicity; however , a surprising inconsistency was the importance of L26 in HD5 but not L25 in HNP1 , which might be due to differences in their orientations . Hence , despite a lack of complete overlap in the surface-exposed residues of the two α-defensins that interact with AdV , which was expected due to their modest sequence conservation , there are common features that dictate their capacity to neutralize infection . These properties may be general to other α-defensins binding to AdV and to these α-defensins binding to other non-enveloped viruses .
Our previous structure-function analysis of HD5 focused on the requirement for arginine residues at specific locations in the structure as well as a crucial role for dimerization [16] . The inability of lysine to functionally substitute for arginine implied a more prominent role for properties of arginine other than mere positive charge . In addition , we identified a requirement for hydrophobicity at a key location in the defensin structure ( L29 ) to mediate defensin-defensin interactions . Our current studies have allowed us to explore these ideas more fully and demonstrate a pronounced role for hydrophobicity in the anti-viral capacity of α-defensins . In this context , hydrophobicity can play four distinct but not mutually exclusive roles: 1 ) to mediate direct contacts between the virus and the defensin; 2 ) to stabilize the defensin dimer interface; 3 ) to mediate higher order defensin self-association; and 4 ) to serve a structural role in the hydrophobic core of the defensin that may indirectly affect the functionality of surface-exposed residues . A subset of hydrophobic residues ( L16 , V19 , and L26 ) that is important for AdV/HPV inhibition form a discrete patch on one face of the HD5 dimer . Interestingly , we found previously that a positively charged residue ( R28 ) contiguous with these hydrophobic residues was critical for AdV neutralization but dispensable for anti-HPV activity [16] . In contrast , the negatively charged residue E21 was more important for anti-HPV activity . Unlike R28 , which is more distal , E21 is centrally located at the dimer interface . These requirements likely reflect differences in the nature of the α-defensin binding site on each virus . Our understanding of these determinants is rudimentary for AdV: both fiber and penton base are necessary , yet the molecular features that dictate defensin binding have not been precisely defined [9] , [22] . In contrast , corresponding HPV capsid determinants have not yet been investigated . Our studies are consistent with a model in which a central hydrophobic patch may mediate contacts with both viruses , although roles for these residues in addition to mediating direct contact cannot be excluded , while charge-charge interactions confer specificity for each virus and are dictated by residues with different charges and orientations relative to the dimer interface . Several of the residues identified in both the HPV and AdV screens are not surface-exposed and likely do not contribute directly to virus binding . These hydrophobic residues might fulfill alternative roles such as stabilizing the dimer interface . The HD5 dimer interface is formed by interactions between S17 , L29 , and the C3–C31 disulfide bond of one monomer with I22 and the C5–C20 disulfide bond of the other monomer as well as reciprocal hydrogen bonds between the backbones of V19 and E21 on both monomers [13] . We previously demonstrated that mutation of L29 or disruption of backbone hydrogen bonds mediated by E21 attenuates HD5 inhibition of both AdV and HPV by disrupting dimerization [13] , [16] . Whereas mutation of S17 or I22 had no effect on AdV inhibition in our current study , HPV inhibition was attenuated by mutation of I22 . Thus , the HPV interaction may be even more sensitive than that of AdV to the stability of the dimer interface . This interpretation assumes that the HD5 dimer is the functional unit . I22 is surface exposed in the HD5 monomer and could be involved in recognizing determinants unique to HPV . A similar argument could be made for L29 , which rather than stabilizing the dimer interface could mediate direct contact of the HD5 monomer with either virus . Unlike I22 or L29 , the side chain of Y27 is buried and only minimally exposed on the surface of either the monomeric or dimeric forms of HD5 . Mutation of Y27 is deleterious for both AdV and HPV inhibition . The bulky hydrophobic side chain of W26 in HNP1 adds structural rigidity to the surrounding residues [26] . Y27 may play a similar supportive role in HD5 , maintaining proper orientation and functionality of the surface exposed residues . In summary , the anti-viral mechanism of HD5 against non-enveloped viruses is multifaceted and involves interplay between hydrophobicity and charge on the defensin surface to recognize viral targets as well as the need for a hydrophobic core to support dimerization and the correct orientation of the viral interface . To test whether properties governing the potency of HD5 are general to another α-defensin , we analyzed the anti-viral activity of comparable alanine mutants of HNP1 against AdV . Like for HD5 , stabilization of the dimer interface through the C-terminal hydrophobic residues W26 and F28 of HNP1 was critical . Although in linear sequence these residues are directly analogous to Y27 and L29 of HD5 , respectively , their precise mode of action differs due to the distinct geometries of the dimer interface of each defensin: W26 is directly involved in HNP1 dimerization , and L29 but not Y27 mediates HD5 dimerization [13] . Accordingly , we saw some restoration of activity by substituting non-natural amino acids with increasingly hydrophobic side chains for both HNP1 W26A and HD5 L29A [16] . Mutation of W26 may have an additional effect on self-association of HNP1 independent of canonical dimer contacts [15] , [26] . In further agreement with our studies of HD5 , surface exposed arginine residues of HNP1 were critical . However , unlike in HD5 where there was one key arginine [16] , mutation of R14 , R15 , and R24 attenuated HNP1 activity . Moreover , a combination of surface-exposed hydrophobic and arginine residues did not form a discrete patch on one face of HNP1 . Rather , they are dispersed on multiple faces of the defensin dimer . Interestingly , disruption of the salt bridge in the R5A mutant of HNP1 led to a reduction in anti-viral activity . We had previously shown that an HD5 mutant disrupting the salt bridge through a more conservative mutation ( E14Q ) had no effect [9] . Since E13A , which also disrupts the salt bridge , was not attenuating in the current study , the effect of the R5A mutation must extend beyond its role in forming the salt bridge , perhaps involving the hydrophobic component of the side chain . Alternatively , a non-neutralized positive charge ( in E13A ) is better tolerated and may enhance function unlike a non-neutralized negative charge ( in R5A ) . Since the equivalent residue in HD5 ( R6 ) was not tested , we do not know if a less conservative mutation in HD5 would be similarly attenuating . Thus , some of the principles dictating HD5 anti-viral activity are predictive of qualities critical for HNP1; however , failure to identify a discrete binding interface on HNP1 may reflect its relatively lower anti-viral activity compared to HD5 The alanine scan mutagenesis also provided a means to gain insight into the mechanism of virus neutralization . An outstanding question from our previous work was the parameters of virus binding that dictate defensin potency . To address this issue , we utilized a subunit of the AdV capsid , rPenton , to correlate anti-viral activity and binding . The rationale for the use of rPenton as a surrogate for the complete capsid was based on previous studies of chimeric viruses in which substitutions of penton base and fiber dictated defensin sensitivity [9] . We found , in general , that the anti-viral efficacy of the HD5 mutants correlated most closely with their stoichiometry at saturation rather than their affinity for rPenton . For wild type HD5 and mutants with near wild type activity , there are ∼210 HD5 molecules/rPenton . Given that there are 12 pentons/virion , ∼2520 HD5 molecules would bind to the whole virus . This finding is in line with our previous semi-quantitative binding data in which we found that ∼2750 HD5 molecules bound to each complete AdV particle [9] . This high stoichiometry supports the anti-viral mechanism suggested by previous cryoEM analysis , which indicated that HD5 was binding to and coating the fiber and penton base proteins of the vertex , thereby preventing fiber dissociation [9] . These data are consistent with two models: either mutation of critical residues restricts binding of the defensins to fewer sites on the capsid or limits defensin-defensin self-association that occurs subsequent to initial defensin-capsid binding . Two exceptions to this trend were Y27A and L29A . Y27A exhibited both weaker affinity and reduced stoichiometry relative to wild type , whereas the affinity of L29A was lower but its stoichiometry was much higher than wild type . Thus , as evidenced by L29 , a minimal affinity must be important for neutralization despite high stoichiometry . L29 overall exhibited a unique binding kinetic , possibly indicative of greater interaction with the dextran matrix of the chip . Alternatively , this may reflect that L29A exhibits reduced self-association [13] , [25]; however , the fact that the kinetics of E21me binding had a similar shape to that of wild type argues against this unless a major contribution to binding of E21me is mediated by L29 itself . Overall , mutations that reduced anthrax lethal factor binding vs . rPenton binding differed in their relative effect , highlighting the specificity of each interaction [13] . Although SPR was a vast improvement over our semi-quantitative whole virus binding assays , some limitations include heterogeneous orientation of rPenton on the SPR chip , exposure of surfaces of rPenton that would be inaccessible to defensin binding in the intact capsid , and the presence of an impurity in the rPenton preparation that may have an influence on quantification of binding that we cannot formally exclude . In addition , a portion of the fiber protein in AdV-5 and AdV-2 is glycosylated , which is likely not recapitulated in the baculovirus-derived proteins [27] . Although glycosylation does affect the reactivity of fiber with antibodies , it has no effect on the trimerization of fiber or its incorporation into particles [28] . Moreover , most AdV serotypes that are neutralized by α-defensins are not glycosylated [9] , [29] . Nonetheless , the contribution of fiber glycosylation to defensin-mediated neutralization of AdV could not be assessed in our SPR analysis . This analysis was also limited to AdV , as we have not yet identified a capsid subunit of HPV analogous to rPenton that would serve as a suitable binding partner . In summary , this study maps for the first time a precise region on the surface of an α-defensin dimer crucial for interaction with and inhibition of non-enveloped viruses . Multiple residues involved in this interaction are distinct from those implicated in α-defensin anti-bacterial function [13] . Similar defensin interfaces have only been previously identified for HNP1 binding to Lipid II and for interaction of human β-defensin 1 with the chemokine receptor CCR6 and with E . coli [30] , [31] . Although our comparison of HD5 and HNP1 inhibition of AdV identified some commonalities ( e . g . , important C-terminal hydrophobic residues ) between the α-defensins , the lack of a discrete surface patch on HNP1 as well as the increased dependence on arginine residues suggests that different sets of rules dictate the anti-viral activity of each α-defensin . Further studies of α- and β-defensins are needed to define a core set of rules that govern the broad , yet selective anti-viral activity of both α- and β-defensins , which may allow for the development of novel therapeutics based on defensin mechanisms .
Tissue culture reagents were obtained from Mediatech ( Manassas , VA ) or Invitrogen ( Carlsbad , CA ) . Human A549 , HeLa , 293β5 , and 293TT cells were cultured in DMEM supplemented with 10% fetal bovine serum ( Sigma-Aldrich , St . Louis , MO ) , 4 mM L-glutamine , 100 units/ml penicillin , 100 µg/ml streptomycin , and 0 . 1 mM nonessential amino acids ( complete DMEM ) as previously described [16] . The replication-defective human AdV-5 vector used in these studies ( AdV5 . eGFP ) is E1/E3-deleted and contains an enhanced green fluorescent protein ( eGFP ) reporter gene cassette driven by a CMV promoter . AdV5 . eGFP was propagated in 293β5 cells , purified by CsCl gradient centrifugation , stored and used as previously described [8] , [16] . HPV16 pseudoviruses ( PsVs ) containing L1 and L2 were produced in 293TT cells and purified according to established protocols [16] , [32] , [33] . Synthetic HD5 and HNP1 were obtained from Peptides International , Inc . ( Louisville , KY ) . Alternatively , folded HD5 was generated from a synthesized 80% pure linear peptide ( CPC Scientific , Sunnyvale , CA ) by thiol-disulfide reshuffling and purified to homogeneity by reverse-phase high-pressure liquid chromatography [16] , [34] . The synthesis , refolding , purification , and structural validation of the HD5 and HNP1 alanine scan mutants have been described [13] , [26] . All α-defensins were quantified by UV absorbance at 280 nm using calculated molar extinction coefficients [35] . A549 cell monolayers were infected with serial dilutions of AdV5 . eGFP in black wall , clear bottom 96-well plates ( Perkin-Elmer , San Jose , CA ) . Total monolayer fluorescence was quantified with a Typhoon 9400 variable mode imager ( GE Healthcare , Piscataway , NJ ) 24–30 h post-infection . A virus concentration producing 50–80% maximal signal was chosen for inhibition studies . Serum-free DMEM ( SFM ) alone or containing increasing concentrations of wild type or mutant HD5 was incubated with purified AdV5 . eGFP for 45 min on ice . The mixture ( 35 µl/well ) was added to a confluent monolayer of A549 cells that had been washed twice in SFM . Cells were then incubated at 37°C for 2 h with rocking , washed , and cultured with complete DMEM for 24–30 h . Plates were scanned for eGFP signal as above , and background-subtracted total well fluorescence was quantified using ImageJ software [36] . Experiments with HPV PsVs were performed as above with the following exceptions: 1 ) infection was measured on HeLa cells , 2 ) the PsVs were incubated at 37°C for 4 h prior to washing and removal of the inoculum , and 3 ) GFP was measured 48 h post-infection . For assays with HNP1 , virus alone in SFM was pre-incubated on a confluent monolayer of A549 cells with rocking for 45 min at 4°C . Cells were then washed twice with chilled SFM , and SFM alone or containing increasing concentrations of wild type or mutant HNP1 was added . Cells were incubated for 45 min at 4°C and then shifted to 37°C . After 2 h , cells were washed twice with complete media and incubated a further 24–30 h at 37°C until eGFP signal was quantified as above . An N-terminal 6×His tag followed by a Tev protease recognition site was cloned 5′ of the full-length human AdV5 fiber gene into pFastBAC . Recombinant baculovirus was made in Sf21 cells using the Bac-to-Bac system ( Invitrogen ) following the manufacturer's instructions . Protein expression was in HighFive cells in shaker flasks ( 135 rpm , 27°C , 3 days ) at a multiplicity of infection between 1 and 3 . Cell pellets were frozen at −80°C in lysis buffer ( 50 mM Na2HPO4/NaH2PO4 , 50 mM NaCl , 0 . 1% Triton X-100 , pH 8 . 0 ) , thawed , and lysed by sonication in the presence of Halt EDTA-free protease inhibitors ( Thermo Fisher Scientific ) . Initial purification of clarified lysate with TALON ( Clontech ) resin indicated that the 6×His tag was lost from the majority of the recombinant fiber . Thus , the flow through from the TALON column was purified by FPLC on a Q Sepharose Fast Flow column ( GE Healthcare ) using a linear gradient from 0 to 300 mM NaCl in 25 mM Tris pH 7 . 4 . Fractions containing fiber were pooled and concentrated . Fiber trimerization was confirmed by immunoblot of samples heated or not for 5 min to 95°C using the 4D2 anti-fiber mAb ( Thermo Fisher Scientific ) and by analytical size exclusion chromatography using a Superdex 200 10/300 GL column ( GE Healthcare ) ( data not shown ) . A bacterial expression plasmid ( pRSET-A ) encoding an N-terminal 6×His tag followed by an enterokinase recognition site 5′ of the full-length human AdV5 penton base ( PB ) gene was a gift from Lali Medina-Kauwe [37] . PB was expressed in BL21-CodonPlus ( DE3 ) -RIPL cells ( Stratagene ) upon induction at an OD600 of 0 . 6–0 . 8 using 0 . 4 mM IPTG for 4 hr at 37°C . Cell pellets were frozen at −80°C in lysis buffer . Thawed cells were lysed by sonication in the presence of 1 mM PMSF , treated with 0 . 01 mg/ml DNase I for 10 min at RT , adjusted to 300 mM NaCl and 10 mM imidazole , and clarified by centrifugation ( 18 , 000× g for 45 min ) . Clarified lysate was applied to a TALON column and washed sequentially with 10 column volumes each of MCAC-10 ( 50 mM Na2HPO4/NaH2PO4 , 300 mM NaCl , 0 . 1% Triton X-100 , 10 mM imidazole , pH 8 . 0 ) and MCAC-20 ( 50 mM Na2HPO4/NaH2PO4 , 500 mM NaCl , 20 mM imidazole , 10% glycerol , pH 8 . 0 ) . Bound protein was eluted with MCAC-250 ( 50 mM Na2HPO4/NaH2PO4 , 300 mM NaCl , 250 mM imidazole , 10% glycerol , pH 8 . 0 ) . Fractions containing the highest concentrations of eluted protein were pooled and concentrated/desalted into desalting buffer ( 50 mM Na2HPO4/NaH2PO4 , 130 mM NaCl , 10% glycerol , pH 8 . 0 ) . PB oligomerization was confirmed by analytical size exclusion chromatography as for fiber above ( data not shown ) . Concentrated PB ( ∼2 . 9 µM ) and fiber ( ∼5 . 8 µM ) were mixed in the presence of Halt EDTA-free protease inhibitors and incubated overnight at 4°C in a total volume of 800 µl . rPenton was then purified on TALON beads as for PB above except that the wash and elution buffers contained 150 mM NaCl . Peak fractions were pooled , concentrated , and stored in desalting buffer . Fractions from purification were resolved by SDS-PAGE . Immunoblots were probed for fiber ( 4D2 ) and PB ( rabbit antiserum , a kind gift from Glen R . Nemerow ) and visualized with chemiluminescence . Purified rPenton was diluted to 13 µg/ml in PBS . Samples were negatively stained with 2% uranyl acetate on 400 mesh carbon coated grids ( Ted Pella , Inc . ) that were glow discharged for 30 s at 15 microamps and imaged on a FEI Tecnai TF20 Transmission electron microscope at 200 kV and a nominal magnification of 60 , 000× at the Cleveland Center for Membrane and Structural Biology . A gene encoding residues 1 to 235 of the human coxsackie and adenovirus receptor ( CAR ) was cloned in frame with the Fc portion of the heavy chain of rabbit IgG in the expression plasmid pCB6 [38] . This construct was transfected into 293 cells , and a polycolonal population of cells ( 293-CAR-Ig ) was selected with G418 ( 0 . 5 mg/ml , Cellgro ) . Bovine Ig was depleted from FBS using the Affi-Gel Protein A MAPS II Kit ( Bio-rad ) . 293-CAR-Ig cells were cultured in DMEM/4% Ig-depleted FBS for 48 h , and CAR-Ig was purified from the culture supernatant using the Affi-Gel Protein A MAPS II Kit following the manufacturer's instructions , concentrated , and stored in 25 mM Tris , 150 mM NaCl , pH 7 . 0 . Experiments were performed on a Biacore T200 system ( Biacore , GE Healthcare ) at 25°C in HBS EP+ running buffer ( 9 . 5 mM HEPES , 142 . 5 mM NaCl , 2 . 85 mM EDTA , 0 . 05% surfactant P20 , pH 7 . 4 ) . The first flow cell on a CM5 sensor chip was reserved to measure background binding , while additional cells were used to immobilize fiber , PB , and rPenton by amine-coupling chemistry . A target density of 2000 resonance units ( RU ) was selected for rPenton , and for Figure 3D , an equal number of fiber ( 708 RU ) and PB ( 1292 RU ) were also targeted for immobilization . The final RUs immobilized were: 2115 RU rPenton , 889 RU fiber , and 1061 RU PB . Analyte was injected to the flow cells at a rate of 11 µl/min and binding was measured for 3 min , followed by measuring dissociation for 5 min in analyte-free running buffer . The flow cells were regenerated with two 30 s pulses of 10 mM HCl at 11 µl/min separated by 30 s in running buffer . After regeneration , the flow cells were stabilized in running buffer for 2 min before injection of the next analyte . Data analysis was performed using Biacore T200 evaluation software and Prism ( version 5 . 0d ) . Affinity ( KD ) and Bmax were derived by fitting RU values at 80 s post-injection using the Steady State Surface Bound Affinity model in Biacore T200 Evaluation software . Stoichiometry ( Sm ) of HD5 bound to rPenton at saturation was calculated using the following equation: ( 1 ) Ligand response ( LR ) is the amount of immobilized ligand in RU . Experiments were analyzed using Prism 5 . 0d . For Figures 1A , 1B , and 6A , data were analyzed by two-way ANOVA with Bonferroni post-tests to compare each mutant to wild type HD5 or HNP1 at each concentration . For Figure 6B the post-tests were used to compare each mutant or wild type to HPN1 W26A . Pearson correlation analyses between KD and IC50 or between Bmax and IC50 were performed using IC50 values calculated from the data in Figure 1 and binding data from Figure 5 . Values for Y27 and L29 ( KD ) or only for L29 ( Bmax ) were excluded from the correlation analyses . For all tests , p<0 . 05 was considered significant .
|
Human α-defensins are an important component of the innate immune response and provide an initial block against a broad number of infectious agents , including viruses and bacteria . Characteristics of α-defensins that are necessary for their anti-bacterial activity have been identified , but our understanding of determinants required for activity against non-enveloped viruses is limited . In this work , we utilized alanine scan mutagenesis to systematically and comprehensively investigate the role of hydrophobic and charged residues of two α-defensins in binding to and/or neutralization of human adenovirus and human papillomavirus . Our results implicate common core hydrophobic residues as critical for inhibition of these non-enveloped viruses by the two α-defensins , with specificity provided by charged residues unique to each interaction . We also found that the number of α-defensin molecules bound to the virus was a stronger correlate of the anti-viral potency of the α-defensin mutants than their absolute affinity for the viral capsid . Understanding common characteristics of α-defensins important for non-enveloped virus binding will inform rules that govern the function of these abundant and multifaceted peptides in host defense .
|
[
"Abstract",
"Introduction",
"Results",
"Discussion",
"Materials",
"and",
"Methods"
] |
[
"biochemistry",
"immune",
"system",
"proteins",
"viral",
"entry",
"defensins",
"proteins",
"viral",
"transmission",
"and",
"infection",
"virology",
"biology",
"and",
"life",
"sciences",
"immunology",
"microbiology"
] |
2014
|
Delineation of Interfaces on Human Alpha-Defensins Critical for Human Adenovirus and Human Papillomavirus Inhibition
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.